energy management handbook, sixth edition

ENERGY MANAGEMENTHANDBOOKSIXTH EDITION

EDITORIAL BOARDEDITORWayne C. TurnerSchool of Industrial Engineering and ManagementOklahoma State UniversityStillwater, OklahomaASSOCIATE EDITORSteve DotyColorado Springs UtilitiesColorado Springs, ColoradoCONTRIBUTORSEric AngevineSchool of ArchitectureOklahoma State UniversityStillwater, OKBradley BracherOklahoma City, OKBarney BurroughsIndoor Air Quality ConsultantAlpharetta, GABarney L. CapehartIndustrial EngineeringUniversity of FloridaGainesville, FLClint ChristensonIndustrial EngineeringOklahoma State UniversityStillwater, OKDavid E. ClaridgeMechanical Engineering DepartmentTexas A&M UniversityCollege Station, TexasWilliam E. CrattyVentana CorporationBethal, CTCharles CulpEnergy Systems LaboratoryTexas A&M UniversityCollege Station, TexasSteve DotyColorado Springs UtilitiesColorado Springs, COKeith ElderCoffman Engineers, Inc.Seattle, WAJohn L. Fetters, CEM, CLEPEffective Lighting Solutions, Inc.Columbus, OhioCarol Freedenthal, CEOJofree Corporation,Houston, TXGSA Energy ConsultantsArlington, VARichard WakefieldLynda WhiteJairo GutiemezDale A. GustavsonConsultantOrange, CAJeff HaberlEnergy Systems LaboratoryTexas A&M UniversityCollege Station, TexasMichael R. Harrison, ManagerEngineering & Technical ServicesJohns-Mansfield CorporationDenver, CORussell L. HeisermanSchool of TechnologyOklahoma State UniversityStillwater, OKWilliam J. Kennedy, Jr.Industrial EngineeringClemson UniversityClemson, SCJohn M. Kovacik, RetiredGE Industrial & Power System SalesSchenectady, NYMingsheng LiuArchitectural EngineeringUniversity of NebraskaLincoln, NBKonstantin LobodovskyMotor ManagerPenn Valley, CATom LunnebergCTG Energetics, Inc.Irvine, CAWilliam MashburnVirginia Polytechnic Institute andState UniversityBlacksburg, VAJavier MontJohnson ControlsChesterfield, MOGeorge OwensEnergy and Engineering SolutionsColumbia, MDLes PaceLektron LightingTulsa, OKJerald D. Parker, RetiredMechanical & Aerospace EngineeringOklahoma State UniversityStillwater, OKS.A. ParkerPacific Northwest National LaboratoryRichland, WADavid PrattIndustrial Enginneering and ManagementOklahoma State UniversityStillwater, OKWesley M. RohrerMechanical EngineeringUniversity of PittsburghPittsburgh, PAPhilip S. SchmidtDepartment of Mechanical EngineeringUniversity of TexasAustin, TXR. B. ScollonManager, Energy ConservationAllied Chemical CorporationMorristown, NJR. D. SmithManager, Energy Generation & Feed StocksAllied Chemical CorporationMorristown, NJMark B. SpillerGainesville Regional UtilitiesGainesville, FLNick SteckyNJS Associates, LLCAlbert ThumannAssociation of Energy EngineersAtlanta, GAW.D. TurnerMechanical Engineering DepartmentTexas A&M UniversityCollege Station, TexasAlfred R. WilliamsVentana CorporationBethel, CTLarry C. WitteDepartment of Mechanical EngineeringUniversity of HoustonHouston, TXJorge Wong KcomtGeneral Electric, Evansville, INEric WoodroofJohnson Controls, Santa Barbara, CA

ENERGY MANAGEMENTHANDBOOKSIXTH EDITIONBYWAYNE C. TURNERSCHOOL OF INDUSTRIAL ENGINEERING AND MANAGEMENTOKLAHOMA STATE UNIVERSITYANDSTEVE DOTYCOLORADO SPRINGS UTILITIESCOLORADO SPRINGS, COLORADO

5 Boilers and Fired Systems ....................................................................................................... 87Introduction ....................................................................................................................... 87Analysis of Boilers and Fired Systems ........................................................................... 87Key Elements for Maximum Efficiency ......................................................................... 89Fuel Considerations ........................................................................................................ 116Direct Contact Technology for Hot Water Production .............................................. 1226 Steam and Condensate Systems ........................................................................................... 125Introduction ..................................................................................................................... 125Thermal Properties of Steam ......................................................................................... 126Estimating Steam Usage and its Value ........................................................................ 133Steam Traps and Their Application .............................................................................. 139Condensate Recovery ..................................................................................................... 1477 Cogeneration ............................................................................................................................ 155Introduction ..................................................................................................................... 155Cogeneration System Design and Analysis ................................................................ 157Computer Programs ....................................................................................................... 174U.S. Cogeneration Legislation: PURPA ....................................................................... 176Evaluating Cogeneration Opportunities: Case Examples ........................................ 1788 Waste-Heat Recovery .............................................................................................................. 193Introduction ..................................................................................................................... 193Waste-Heat Survey ......................................................................................................... 201Waste-Heat Exchangers .................................................................................................. 207Commercial Options in Waste-Heat-Recovery Equipment ...................................... 211Economics of Waste-Heat Recovery ............................................................................. 2189 Building Envelope ................................................................................................................... 221Introduction ..................................................................................................................... 221Principles of Envelope Analysis ................................................................................... 223Metal Elements in Envelope Components .................................................................. 225Roofs ................................................................................................................................. 230Floors ................................................................................................................................ 233Fenestration ..................................................................................................................... 234Infiltration ........................................................................................................................ 237Summarizing Envelope Performance with the Building Load Coefficient ......... ...239Thermal “Weight” ........................................................................................................... 240Envelope Analysis for Existing Buildings ................................................................... 240Envelope Analysis for New Buildings ......................................................................... 245Updated Envelope Standards for New and Existing Construction ........................ 245Additional Reading ........................................................................................................ 24610 HVAC Systems ......................................................................................................................... 247Introduction ..................................................................................................................... 247Surveying Existing Conditions ..................................................................................... 247Human Thermal Comfort .............................................................................................. 248HVAC System Types ...................................................................................................... 249Energy Conservation Opportunities ............................................................................ 259Cooling Equipment ........................................................................................................ 269Domestic Hot Water ....................................................................................................... 271Estimating HVAC Energy Consumption .................................................................... 272vi

Insulation Materials ....................................................................................................... .439Insulation Selection ........................................................................................................ .443Insulation Thickness Determination ............................................................................ 448Insulation Economics ..................................................................................................... 46116 Use of Alternative Energy ...................................................................................................... 471Introduction ..................................................................................................................... 471Solar Energy ..................................................................................................................... 471Wind Energy .................................................................................................................... 484Refuse-Derived Fuel ...................................................................................................... .489Fuel Cells .................................................................................................................. ........49317 Indoor Air Quality .................................................................................................................. 497Introduction and Background ............................................................................... ........497What is the Current Situation ....................................................................................... 499Solutions and Prevention of IAQ Problems ............................................................... .50018 Electric and Gas Utility Rates for Commercial and Industrial Consumers ................. 507Introduction ..................................................................................................................... 507Utility Costs .................................................................................................................... .507Rate Structures ................................................................................................................ 508Innovative Rate Type ...................................................................................................... 509Calculation of a Monthly Bill ........................................................................................ 510Conducting a Load Study .............................................................................................. 513Effects of Deregulation on Customer Rates ................................................................ 51619 Thermal Energy Storage ......................................................................................................... 519Introduction ..................................................................................................................... 519Storage Systems ............................................................................................................... 521Storage Mediums ............................................................................................................ 523System Capacity .............................................................................................................. 526Economic Summary ........................................................................................................ 53220 Codes Standards & Legislation ............................................................................................ 539The Energy Policy Act of 1992 ...................................................................................... 539State Codes ....................................................................................................................... 540Model Energy Code ........................................................................................................ 541Federal Energy Efficiency Requirements .................................................................... 541Indoor Air Quality Standards ....................................................................................... 542Regulations & Standards Impacting CFCs .................................................................. 543Regulatory and Legislative Issues Impacting Air Quality ........................................ 544Regulatory and Legislative Issues Impacting Cogeneration & Power ................... 545Opportunities in the Spot Market ................................................................................ 546The Climatic Change Action Plan ................................................................................ 54721 Natural Gas Purchasing ......................................................................................................... 549Preface .............................................................................................................................. 549Introduction ..................................................................................................................... 550Natural Gas as a Fuel ..................................................................................................... 553Buying Natural Gas ........................................................................................................ 566New Frontiers for the Gas Industry ............................................................................. 575viii

22 Control Systems ....................................................................................................................... 577Introduction ..................................................................................................................... 577Why Automatic Control? ............................................................................................... 577Why Optimization? ........................................................................................................ 578Technology Classifications ............................................................................................ 578Control Modes ................................................................................................................. 580Input/Output Devices ................................................................................................... 584Valves and Dampers ....................................................................................................... 586Instrument Accuracy, Repeatability, and Drift ........................................................... 588Basic Control Block Diagrams ....................................................................................... 589Key Fundamentals of Successfully Applied Automataic Controls ......................... 590Operations and Maintenance ........................................................................................ 592Expected Life of Control Equipment ........................................................................... 592Basic Energy-saving Control Applications .................................................................. 594Advanced Energy-saving Control Applications ........................................................ 594Facilities Operations Control Applications ................................................................. 594Control System Application Pitfalls to Avoid ............................................................. 601Costs and Benefits of Automataic Control .................................................................. 601Estimating Savings from Applied Automatic Control Systems ............................... 601Conclusion and Further Study ...................................................................................... 605Glossary of Terms ........................................................................................................... 61623 Energy Security and Reliability ........................................................................................... 621Introduction ..................................................................................................................... 621Risk Analysis Methods ................................................................................................... 624Countermeasures ............................................................................................................ 630Economics of Energy Security and Reliability ............................................................ 632Links to Energy Management ....................................................................................... 633Impact of Utility Deregulation ...................................................................................... 63424 Utility Deregulation and Energy System Outsourcing .................................................... 637Introduction ..................................................................................................................... 637An Historical Perspective of the Electric Power Industry ........................................ 637The Transmission System and The Federal Regulatory Commission's(FERC) Role in Promoting Competition in Wholesale Power ........................ 638Stranded Costs ................................................................................................................ 639Status of State Electric Industry Restructuring Activity ........................................... 640Trading Energy—Marketers and Brokers ................................................................... 640The Impact of Retail Wheeling ..................................................................................... 640The Ten-Step Program to Successful Utility Deregulation ....................................... 641Aggregation ..................................................................................................................... 643In-house vs. Outsourcing Energy Services .................................................................. 64325 Financing Energy Management Projects ............................................................................ 649Introduction ..................................................................................................................... 649Financial Arrangements: A Simple Example .............................................................. 649Financial Arrangements: Details and Terminology ................................................... 652Applying Financial Arrangements: A Case Study ..................................................... 653"Pros" & "Cons" of Each Financial Arrangement ........................................................ 664Characteristics that Influence which Financial Arrangement is Best ...................... 665Incorporating Strategic Issues when Selecting Financial Arrangements ............... 666ix

Glossary ............................................................................................................................ 66626 Commissioning for Energy Management ........................................................................... 671Introduction to Commissioning for Energy Management ....................................... 671Commissioning Definitions ........................................................................................... 671The Commissioning Process in Existing Buildings ................................................... 672Commissioning Measures ............................................................................................. 680Ensuring Optimum Building Performance ................................................................. 695Commissioning New Buildings for Energy Management ........................................ 702Additional Information .................................................................................................. 70427 Measurement and Verification of Energy Savings ............................................................ 707Introduction ..................................................................................................................... 707Overview of Measurement and Verification Methods .............................................. 71128 Ground-source Heat Pumps Applied to Commercial Buildings ................................... 755Abstract ............................................................................................................................ 755Background ...................................................................................................................... 755Introduction to Ground-source Heat Pumps .............................................................. 756About the Technology .................................................................................................... 757Application ...................................................................................................................... 767Technology Performance ............................................................................................... 771Hypothetical Case Studies ............................................................................................. 774The Technology in Perspective ..................................................................................... 781Manufacturers ................................................................................................................. 783For Further Information ................................................................................................. 78529 Sustainability and High Performance Green Buildings ................................................. 793Beginnings ....................................................................................................................... 793Sustainability Gives Rise to the Green Building Movement .................................... 794Introducing the LEED NC Rating System: A Technical Review .............................. 798LEED for Existing Building Rating System (LEED-EB) Adopted in 2004 .............. 801Summary Discussion of Two New LEED Programs ................................................. 804The LEED Process ........................................................................................................... 805ASHRAE Guides Developed to Support LEED ......................................................... 808Appendix I—Thermal Sciences Review ....................................................................................... 815Appendix II—Conversion Factors and Property Tables ............................................................ 837Appendix III—Review of Electrical Science ............................................................................... 887Index .................................................................................................................................................... 901x

FOREWORD TO THE SIXTH EDITIONSince its first edition was published more than two decades ago, Energy ManagementHandbook has remained the leading reference of choice used by thousands of energy managementprofessionals for one fundamental reason. With this new edition, Dr. Turner and Mr.Doty continue to bring readers both the cutting-edge developments they need to know about,as well as the broad scope of practical information they must have to accomplish real andsignificant energy cost reduction goals. No other single publication has been as influential indefining and guiding the energy management profession.This new sixth edition builds upon and is no less essential than its predecessors. Comprehensivein scope, it provides today’s energy managers with the tools they will require to meetthe challenges of a new era of predicted rising energy costs and supply uncertainties—ongoingdevelopments which seem certain to impact virtually every aspect of the cost of doingbusiness in the decades ahead. The new edition also examines the impact of the passage andimplementation of the Energy Policy Act of 2005, which puts in place new energy efficiencyrequirements for government facilities, as well as energy-efficiency-related tax incentives forcommercial buildings.As evidence continues to lend credence to the reality of global climate change, a growingnumber of businesses are seeing the “good business sense” of reducing greenhouse emissionsand developing sustainable, green facilities. The sixth edition of Energy Management Handbookincludes substantial new material on sustainability, high performance facilities and relatedtechnologies.In many ways the evolution of Energy Management Handbook has paralleled that of theAssociation of Energy Engineers (AEE) in meeting the needs of and setting the standards forthe modern energy management profession which has emerged since the 1970s. Therefore,it seems very appropriate that the publication of this important new sixth edition officiallykicks off AEE’s 30 th anniversary celebration. As our organization completes its third decadeof serving more than 8,000 members in 77 countries, it would be nearly impossible to overstatethe impact that Energy Management Handbook has had for those we serve. The book is anofficial reference and preparatory text for AEE’s Certified Energy Manager (CEM) program,the most widely recognized professional credential in the energy management field, havingcertified more than 6,000 professionals since its inception in 1981. In addition numerous largecorporations have selected Energy Management Handbook as their official corporate energymanagement reference.There is no doubt that Energy Management Handbook will continue its role as the indispensablereference for all energy managers who must meet the daunting energy supply andcost control challenges which lie ahead.Albert Thumann, P.E., C.E.M.Executive Director, The Association of Energy EngineersApril 2006xi

This page intentionally left blank

PREFACE TO THE SIXTH EDITIONWhen the first introduction to Energy Management Handbook was written in 1982, Iwas in college, worried more about how to repay student loans than anything else. Butthis is now.This book lives to serve its readers. In helping to edit the book, it has been my goalto keep the material fresh, pertinent and useful. My approach has been to view it fromthe reader’s perspective, and to assure that the book provides good value. I look forwardto further improvements, over time, as technologies continue to change and refine, sinceit will keep me current in the process.As I see it, the core intentions of this book are these:• To address an audience of practicing energy managers and persons entering thistrade. It is a tool for energy managers to get their questions answered, and get thingsaccomplished.• To provide a resource of current, accurate, useful information in a readable format.• To emphasize applications, and include practical examples to clarify key topics,especially savings calculations. Background development and derivations shouldbe limited to just what is needed to support the application messages.For this edition, there are some significant changes, including a rewrite of the automaticcontrols chapter, and all-new chapters on ground source heat pumps, and greenbuilding design.As I have learned, editors check for errors and adjust grammar or format, but do notchange the author’s message. The primary strategy for quality in this book is to choosethe very best authors, so I’ll extend my personal thanks to each of them for their contributions.And I owe Wayne Turner a debt of thanks for the opportunity to contribute. I canonly hope to do as well.I am excited about contributing to the long-running success of this book, becauseof its potential to influence other energy professionals, and therefore the public at large.Writing is just writing, until it changes a behavior or helps someone get something accomplished—thenit becomes golden!Steve DotyColorado Springs, COApril 2006xiii


CHAPTER 1INTRODUCTIONDR. WAYNE C. TURNER,REGENTS PROFESSOROklahoma State UniversityStillwater, Ok.DR. BARNEY L. CAPEHART, PROFESSORUniversity of FloridaGainesville, Fla.STEPHEN A. PARKERPacific Northwest National LaboratoryRichland, WASTEVE DOTYColorado Springs UtilitiesColorado Springs, CO1.1 BACKGROUNDMr. Al Thumann, Executive Director of the Associationof Energy Engineers, said it well in the Foreword.“The energy ‘roller coaster’ never ceases with new turnsand spirals which make for a challenging ride.” Thoseprofessionals who boarded the ride in the late 70’s andstayed on board have experienced several ups anddowns. First, being an energy manager was like beinga mother, John Wayne, and a slice of apple pie all inone. Everyone supported the concept and success wasaround every bend. Then, the mid-80’s plunge in energyprices caused some to wonder “Do we really need tocontinue energy management?”Sometime in the late 80’s, the decision was made.Energy management is good business but it needs tobe run by professionals. The Certified Energy ManagerProgram of the Association of Energy Engineers becamepopular and started a very steep growth curve. AEEcontinues to grow in membership and stature.About the same time (late 80’s), the impact of theNatural Gas Policy Act began to be felt. Now, energymanagers found they could sometimes save significantamounts of money by buying “ spot market” natural gasand arranging transportation. About the only thing thatcould be done in purchasing electricity was to choosethe appropriate rate schedule and optimize parameters(power factor, demand, ratchet clauses, time of use,etc.—see Chapter 18 on energy rate schedules).With the arrival of the Energy Policy Act of 1992,electricity deregulation moved closer to reality, creatingthe opportunity of purchasing electricity from whereverthe best deal could be found and to wheel the electricenergy through the grid. Several states moved towardelectrical deregulation, with some successes. But therewere also some failures that made the energy industrypause and reflect. The prospect of electric deregulationand sharing grid infrastructure caused utilities tochange their business view of their portion of the grid.Investment in expanding or upgrading this infrastructurebecame risky business for individual utilities, andso most chose to maintain the existing grid systemsthey owned, with a wait-and-see approach. Throughelectricity trading that manipulated pricing, problemswith implementation changed the electric deregulationmovement trend from slow to stop. Since good businessrelationships are good for all, some revisions tothe EPACT-92 deregulation provisions may be necessaryto see greater acceptance, and to sustain the concept inpractice. To regain the confidence of the consumers, agreater degree of oversight of the business practices andthe sharing of the vital US grid infrastructure may benecessary. This need is further accentuated by concernsof security and reliability of our nation's electrical grid,spurred by national events in September 2001 (9-11) andAugust 2003 (Blackout). Even with the bumps as electricityderegulation was first tried, wider scale electricderegulation remains an exciting concept and energymanagers are watching with anticipation. As new skillsare learned and beneficial industry relationships arecreated, the prospects of larger scale deregulation willimprove.However, EPACT-1992's impact is further reaching.If utilities must compete with other producers ofelectricity, then they must be “lean and mean.” As Mr.Thumann mentions in the Foreword, this means manyof the Demand Side Management (DSM) and otherconservation activities of the utilities are being cut oreliminated. The roller coaster ride goes on.In 2005, the Bush Administration enacted theEnergy Policy Act of 2005. This Act provides new opportunitiesand incentives for energy improvements inthe country, including strong incentives for renewableenergy sources and net metering. It is hoped that the taxincentives provided under this Act will become tools forthe private sector to spur change with the free enterprisesystem. Similar in style to individual utility incentiveprograms, the Act's success will depend largely on theability of private firms, such as consultants, ESCOs and1

2 ENERGY MANAGEMENT HANDBOOKPerformance Contractors, to find partnering solutionsto connect the program funding mechanism and thecustomer points of use. EPACT-2005 also updates thefederal energy improvement mandates with a newer,stricter, baseline year (2003) and a new timeline forenergy reduction requirements. The federal buildingsegment remains an excellent target for large-scaleimprovement, as well as setting the all-important highvisibility example for private industry to follow.The Presidential Executive Orders mentioned inChapter 20 created the Federal Energy ManagementProgram (FEMP) to aid the federal sector in meetingfederal energy management goals. The potential FEMPsavings are mammoth and new professionals affiliatedwith federal, as well as state and local governmentshave joined the energy manager ranks. However, asCongress changes complexion, the FEMP and even DOEitself may face uncertain futures. The roller coaster ridecontinues.FEMP efforts are showing results. Figure 1.3 outlinesthe goals that have been established for FEMP andreports show that the savings are apparently on scheduleto meet all these goals. As with all such programs,reporting and measuring is difficult and critical. However,that energy and money is being saved is undeniable.More important, however, to most of this book'sreaders are the Technology Demonstration Programsand Technology Alerts being published by the PacificNorthwest Laboratories of Battelle in cooperation withthe US DOE. Both of these programs are dramaticallyspeeding the incorporation of new technology and theAlerts are a great source of information for all energymanagers. (Information is available on the WEB).As utility DSM programs shrink, while privatesector businesses and the federal government expandtheir needs for energy management programs, the dooris opening for the ESCOs (Energy Service Companies),Shared Savings Providers, Performance Contractors,and other similar organizations. These groups are providingthe auditing, energy and economic analyses,capital and monitoring to help other organizationsreduce their energy consumption and reduce theirexpenditures for energy services. By guaranteeing andsharing the savings from improved energy efficiencyand improved productivity, both groups benefit andprosper.Throughout it all, energy managers have proventime and time again, that energy management is costeffective. Furthermore, energy management is vital toour national security, environmental welfare, and economicproductivity. This will be discussed in the nextsection.1.2 THE VALUE OF ENERGY MANAGEMENTBusiness, industry and government organizationshave all been under tremendous economic andenvironmental pressures in the last few years. Beingeconomically competitive in the global marketplace andmeeting increasing environmental standards to reduceair and water pollution have been the major drivingfactors in most of the recent operational cost and capitalcost investment decisions for all organizations. Energymanagement has been an important tool to help organizationsmeet these critical objectives for their short termsurvival and long-term success.The problems that organizations face from boththeir individual and national perspectives include:• Meeting more stringent environmental qualitystandards, primarily related to reducing globalwarming and reducing acid rain.Energy management helps improve environmentalquality. For example, the primary culprit in globalwarming is carbon dioxide, CO 2 . Equation 1.1, a balancedchemistry equation involving the combustion ofmethane (natural gas is mostly methane), shows that2.75 pounds of carbon dioxide is produced for everypound of methane combusted. Thus, energy management,by reducing the combustion of methane candramatically reduce the amount of carbon dioxide inthe atmosphere and help reduce global warming. Commercialand industrial energy use accounts for about 45percent of the carbon dioxide released from the burningof fossil fuels, and about 70 percent of the sulfur dioxideemissions from stationary sources.CH 4 + 2 O 2 = CO 2 + 2 H 2 O(12 + 4*1) +2(2*16) =(12 + 2*16) + 2(2*1 +16) (1.1)Thus, 16 pounds of methane produces 44 poundsof carbon dioxide; or 2.75 pounds of carbon dioxideis produced for each pound of methaneburned.Energy management reduces the load on powerplants as fewer kilowatt hours of electricity are needed.If a plant burns coal or fuel oil, then a significant amountof acid rain is produced from the sulphur dioxideemitted by the power plant. Acid rain problems thenare reduced through energy management, as are NO xproblems.Less energy consumption means less petroleumfield development and subsequent on-site pollution.

INTRODUCTION 3Less energy consumption means less thermal pollutionat power plants and less cooling water discharge. Reducedcooling requirements or more efficient satisfactionof those needs means less CFC usage and reduced ozonedepletion in the stratosphere. The list could go on almostindefinitely, but the bottom line is that energy managementhelps improve environmental quality.• Becoming—or continuing to be—economicallycompetitive in the global marketplace, which requiresreducing the cost of production or services,reducing industrial energy intensiveness, andmeeting customer service needs for quality anddelivery times.Significant energy and dollar savings are availablethrough energy management. Most facilities (manufacturingplants, schools, hospitals, office buildings, etc)can save according to the profile shown in Figure 1.1.Even more savings have been accomplished by someprograms.√√√Low cost activities first year or two: 5 to15%Moderate cost, significant effort, three to fiveyears: 15 to 30%Long-term potential, higher cost, more engineering:30 to 50%Figure 1.1Typical SavingsThrough Energy ManagementThus, large savings can be accomplished often withhigh returns on investments and rapid paybacks. Energymanagement can make the difference between profit andloss and can establish real competitive enhancements formost companies.Energy management in the form of implementingnew energy efficiency technologies, new materials andnew manufacturing processes and the use of new technologiesin equipment and materials for business andindustry is also helping companies improve their productivityand increase their product or service quality.Often, the energy savings is not the main driving factorwhen companies decide to purchase new equipment,use new processes, and use new high-tech materials.However, the combination of increased productivity,increased quality, reduced environmental emissions, andreduced energy costs provides a powerful incentive forcompanies and organizations to implement these newtechnologies.Total Quality Management (TQM) is another emphasisthat many businesses and other organizationshave developed over the last decade. TQM is an integratedapproach to operating a facility, and energy costcontrol should be included in the overall TQM program.TQM is based on the principle that front-line employeesshould have the authority to make changes and otherdecisions at the lowest operating levels of a facility. Ifemployees have energy management training, they canmake informed decisions and recommendations aboutenergy operating costs.• Maintaining energy supplies that are:— Available without significantinterruption, and— Available at costs that do notfluctuate too rapidly.Once again, the country is becoming dependent onimported oil. During the time of the 1979 oil price crisis,the U.S. was importing almost 50% of our total oil consumption.By 1995, the U.S. was again importing 50% ofour consumption. Today (2003) we are importing evenmore (approximately 54%), and the price has dramaticallyincreased. Thus, the U.S. is once again vulnerableto an oil embargo or other disruption of supply. Themajor difference is that there is a better balance of oilsupply among countries friendly to the U.S. Nonetheless,much of the oil used in this country is not producedin this country. The trade balance would be much morefavorable if we imported less oil.• Helping solve other national concerns which include:— Need to create new jobs— Need to improve the balance of payments byreducing costs of imported energy— Need to minimize the effects of a potentiallimited energy supply interruptionNone of these concerns can be satisfactorily metwithout having an energy efficient economy. Energymanagement plays a key role in helping move towardthis energy efficient economy.1.3 THE ENERGY MANAGEMENT PROFESSIONEnergy management skills are important to peoplein many organizations, and certainly to people who

4 ENERGY MANAGEMENT HANDBOOKperform duties such as energy auditing, facility orbuilding management, energy and economic analysis,and maintenance. The number of companies employingprofessionally trained energy managers is large andgrowing. A partial list of job titles is given in Figure 1.2.Even though this is only a partial list, the breadth showsthe robustness of the profession.For some of these people, energy management willbe their primary duty, and they will need to acquirein-depth skills in energy analysis as well as knowledgeabout existing and new energy using equipment andtechnologies. For others—such as maintenance managers—energymanagement skills are simply one morearea to cover in an already full plate of duties and expectations.The authors are writing this Energy ManagementHandbook for both of these groups of readers andusers.Twenty years ago, few university faculty memberswould have stated their primary interest wasenergy management, yet today there are numerous facultywho prominently list energy management as theirprincipal specialty. In 2003, there were 26 universitiesthroughout the country listed by DOE as IndustrialAssessment Centers or Energy Analysis and DiagnosticCenters. Other Universities offer coursework and/ordo research in energy management but do not haveone of the above centers. Finally, several professionalJournals and Magazines now publish exclusively forenergy managers while we know of none that existed15 years ago.The need for energy management in federal facilitiespredates the U.S. Department of Energy. Since 1973,the President and Congress have called on federal agenciesto lead by example in energy conservation and managementin its own facilities, vehicles and operations.Both the President and the Congress have addressed theissue of improving energy efficiency in federal facilitiesseveral times since the mid- 1970’s. Each new piece oflegislation and executive order has combined past experienceswith new approaches in 12 an effort to promotefurther efficiency gains in federal agencies . The FederalEnergy Management Program (FEMP) was establishedin the early 1970’s to coordinate federal agency reporting,analysis of energy use and to encourage energyconservation and still leads that effort today. ExecutiveOrder 13123, Greening the Government Through EfficientEnergy Management, signed by President Clinton in June1999, is the most recent directive for federal agencies.A brief summary of the goals of that executive order isgiven in Figure 1.3. In addition to the goals, ExecutiveOrder 13123 outlined several other requirements forfederal agencies aimed at improving energy efficiency,reducing greenhouse gases and other emissions, increasingthe use of renewable energy, and promoting federalleadership in energy management.Like energy management itself, utility DSM programshave had their ups and downs. DSM effortspeaked in the late 80s and early 90s, and have sinceretrenched significantly as utility deregulation and themovement to retail wheeling have caused utilities toreduce staff and cut costs as much as possible. Thisshort-term cost cutting is seen by many utilities as theironly way to become a competitive low-cost supplierof electric power. Once their large customers have thechoice of their power supplier, they want to be able tohold on to these customers by offering rates that arecompetitive with other producers around the country. Inthe meantime, the other energy services provided by theutility are being reduced or eliminated in this corporatedownsizing effort.This reduction in electric utility incentive and rebateprograms, as well as the reduction in customer support,has produced a gap in energy service assistance that is beingmet by a growing sector of equipment supply companiesand energy service consulting firms that are willingand able to provide the technical and financial assistancethat many organizations previously got from their localelectric utility. New business opportunities and manynew jobs are being created in this shift away from utilitysupport to energy service company support. Energy managementskills are extremely important in this rapidlyexpanding field, and even critical to those companies thatare in the business of identifying energy savings and providinga guarantee of the savings results.• Plant Energy Manager• Utility Energy Auditor• State Agency Energy Analyst• Consulting Energy Manager• DSM Auditor/Manager• Building/Facility Energy Manager• Utility Energy Analyst• Federal Energy Analyst• Consulting Energy EngineerFigure 1.2 Typical Energy Management Job Titles

INTRODUCTION 5√√√√√√√Sec. 201. Greenhouse Gases Reduction Goal.Reduce greenhouse gas emissions attributed tofacility energy use by 30% by 2010 compared to1990.Sec. 202. Energy Efficiency Improvement Goals.Reduce energy consumption per gross square footof facilities by 30% by 2005 and by 35% by 2010relative to 1985.Sec. 203. Industrial and Laboratory Facilities. Reduceenergy consumption per square foot, per unitof production, or per other unit as applicable by20% by 2005 and 25% by 2010 relative to 1990.Sec. 204. Renewable Energy. Strive to expand useof renewable energy. The federal government shallstrive to install 2,000 solar energy systems at federalfacilities by the end of 2000, and 20,000 solarenergy systems at federal facilities by 2010.Sec. 205. Petroleum. Each agency shall reduce theuse of petroleum within its facilities. [Althoughno specific goal is identified.]Sec. 206. Source Energy. The federal governmentshall strive to reduce total energy use as measuredat the source. [Although agency reportingrequirements for energy consumption are basedon site energy, this section allows for an agencyto receive a credit for activities where source energydecreases but site energy increase, such asin cogeneration systems.]Sec. 207. Water Conservation. Reduce waterconsumption and associated energy use in theirfacilities to reach the goals (subsequently) set bythe Secretary of Energy. [The Secretary of Energy,through the DOE Federal Energy ManagementProgram, issued guidance to establish water efficiencyimprovement goal for federal agenciesin May 2000. See www.eere.energy.gov/femp/resources/waterguide.html for details.Figure 1.3. Federal Agency Goals as Established by ExecutiveOrder 13123.Thus, the future for energy management is extremelypromising. It is cost effective, it improves environmentalquality, it helps reduce the trade deficit, andit helps reduce dependence on foreign fuel supplies.Energy management will continue to grow in size andimportance.1.4 SOME SUGGESTED PRINCIPLES OFENERGY MANAGEMENT(The material in this section is repeated verbatimfrom the first and second editions of this handbook.Mr. Roger Sant who was then director of the EnergyProductivity Center of the Carnegie-Mellon Institute ofResearch in Arlington, VA, wrote this section for the firstedition. It was unchanged for the second edition. Now,the fourth edition is being printed. The principles developedin this section are still sound. Some of the numberquoted may now be a little old; but the principles arestill sound. Amazing, but what was right 18 years agofor energy management is still right today. The game haschanged, the playing field has moved; but the principlesstay the same).If energy productivity is an important opportunityfor the nation as a whole, it is a necessity for the individualcompany. It represents a real chance for creativemanagement to reduce that component of product costthat has risen the most since 1973.Those who have taken advantage of these opportunitieshave done so because of the clear intent and commitmentof the top executive. Once that commitment isunderstood, managers at all levels of the organizationcan and do respond seriously to the opportunities athand. Without that leadership, the best designed energymanagement programs produce few results. In addition,we would like to suggest four basic principles which,if adopted, may expand the effectiveness of existingenergy management programs or provide the startingpoint of new efforts.The first of these is to control the costs of the energyfunction or service provided, but not the Btu of energy. Asmost operating people have noticed, energy is just ameans of providing some service or benefit. With thepossible exception of feedstocks for petrochemical production,energy is not consumed directly. It is alwaysconverted into some useful function. The existing dataare not as complete as one would like, but they doindicate some surprises. In 1978, for instance, the aggregateindustrial expenditure for energy was $55 billion.Thirty-five percent of that was spent for machinedrive from electric motors, 29% for feedstocks, 27% forprocess heat, 7% for electrolytic functions, and 2% forspace conditioning and light. As shown in Table 1.1,this is in blunt contrast to measuring these functions inBtu. Machine drive, for example, instead of 35% of thedollars, required only 12% of the Btu.In most organizations it will pay to be even morespecific about the function provided. For instance, evaporation,distillation, drying, and reheat are all typical of

6 ENERGY MANAGEMENT HANDBOOKTable 1.1 Industrial Energy Functions by Expenditureand Btu, 1978—————————————————————————DollarExpenditure Percent of Percent ofFunction (billions) Expenditure Total Btu—————————————————————————Machine drive 19 35 12Feedstocks 16 29 35Process steam 7 13 23Direct heat 4 7 13Indirect heat 4 7 13Electrolysis 4 7 3Space conditioningand lighting–––1–––1–––1Total 55 100 100—————————————————————————Source: Technical Appendix, The Least-Cost Energy Strategy, Carnegie-MellonUniversity Press, Pittsburgh, Pa., 1979, Tables 1.2.1 and 11.3.2.the uses to which process heat is put. In some cases ithas also been useful to break down the heat in terms oftemperature so that the opportunities for matching theheat source to the work requirement can be utilized.In addition to energy costs, it is useful to measurethe depreciation, maintenance, labor, and other operatingcosts involved in providing the conversion equipmentnecessary to deliver required services. These costsadd as much as 50% to the fuel cost.It is the total cost of these functions that must bemanaged and controlled, not the Btu of energy. The largedifference in cost of the various Btu of energy can makethe commonly used Btu measure extremely misleading.In November 1979, the cost of 1 Btu of electricity wasnine times that of 1 Btu of steam coal. Table 1.2 showshow these values and ratios compare in 2005.One of the most desirable and least reliable skillsfor an energy manager is to predict the future cost ofenergy. To the extent that energy costs escalate in pricebeyond the rate of general inflation, investment paybackswill be shortened, but of course the reverse is alsotrue. A quick glance at Table 1.2 shows the inconsistencyin overall energy price changes over this period in time.Even the popular conception that energy prices alwaysgo up was not true for this period, when normalized toconstant dollars. This volatility in energy pricing mayaccount for some business decisions that appear overlyconservative in establishing rate of return or paybackperiod hurdles.Availabilities also differ and the cost of maintainingfuel flexibility can affect the cost of the product.And as shown before, the average annual price increaseof natural gas has been almost three times that of electricity.Therefore, an energy management system thatcontrols Btu per unit of product may completely missthe effect of the changing economics and availabilitiesof energy alternatives and the major differences in usabilityof each fuel. Controlling the total cost of energyfunctions is much more closely attuned to one of theprincipal interests of the executives of an organization—controllingcosts.NOTE: The recommendation to control energy dollarsand not Btus does not always apply. For example,tracking building energy use per year for comparison toprior years is best done with Btus since doing so negatesthe effect of energy price volatility. Similarly, comparingthe heating use of a commercial facility against an industrysegment benchmark using cost alone can yield wildresults if, for example, one building uses natural gas toheat while another uses electric resistance; this is anothercase where using Btus yields more meaningful results.Table 1.2 Cost of Industrial Energy per Million Btu, 1979 and 2005

INTRODUCTION 7A second principle of energy management is tocontrol energy functions as a product cost, not as a part ofmanufacturing or general overhead. It is surprising howmany companies still lump all energy costs into onegeneral or manufacturing overhead account without identifyingthose products with the highest energy functioncost. In most cases, energy functions must become partof the standard cost system so that each function can beassessed as to its specific impact on the product cost.The minimum theoretical energy expenditure toproduce a given product can usually be determineden route to establishing a standard energy cost for thatproduct. The seconds of 25-hp motor drive, the minutesnecessary in a 2200°F furnace to heat a steel part for fabrication,or the minutes of 5-V electricity needed to makean electrolytic separation, for example, can be determinedas theoretical minimums and compared with the actualfigures. As in all production cost functions, the minimumstandard is often difficult to meet, but it can serve as anindicator of the size of the opportunity.In comparing actual values with minimum values,four possible approaches can be taken to reduce thevariance, usually in this order:1. An hourly or daily control system can be installedto keep the function cost at the desired level.2. Fuel requirements can be switched to a cheaperand more available form.3. A change can be made to the process methodologyto reduce the need for the function.4. New equipment can be installed to reduce the costof the function.The starting point for reducing costs should bein achieving the minimum cost possible with the presentequipment and processes. Installing managementcontrol systems can indicate what the lowest possibleenergy use is in a well-controlled situation. It is only atthat point when a change in process or equipment configurationshould be considered. An equipment changeprior to actually minimizing the expenditure under thepresent system may lead to oversizing new equipmentor replacing equipment for unnecessary functions.The third principle is to control and meter only themain energy functions—the roughly 20% that make up80% of the costs. As Peter Drucker pointed out sometime ago, a few functions usually account for a majorityof the costs. It is important to focus controls on thosethat represent the meaningful costs and aggregate theremaining items in a general category. Many manufacturingplants in the United States have only one meter,that leading from the gas main or electric main into theplant from the outside source. Regardless of the reasonablenessof the standard cost established, the inability tomeasure actual consumption against that standard willrender such a system useless. Submetering the mainfunctions can provide the information not only to measurebut to control costs in a short time interval. The costof metering and submetering is usually incidental to thepotential for realizing significant cost improvements inthe main energy functions of a production system.The fourth principle is to put the major effort ofan e nergy management program into installing controls andachieving results. It is common to find general knowledgeabout how large amounts of energy could be saved in aplant. The missing ingredient is the discipline necessaryto achieve these potential savings. Each step in savingenergy needs to be monitored frequently enough by themanager or first-line supervisor to see noticeable changes.Logging of important fuel usage or behavioral observationsare almost always necessary before any particularsavings results can be realized. Therefore, it is critical thatan energy director or committee have the authority fromthe chief executive to install controls, not just advise linemanagement. Those energy managers who have achievedthe largest cost reductions actually install systems andcontrols; they do not just provide good advice.As suggested earlier, the overall potential for increasingenergy productivity and reducing the cost of energyservices is substantial. The 20% or so improvementin industrial energy productivity since 1972 is just thebeginning. To quote the energy director of a large chemicalcompany: “Long-term results will be much greater.”Although no one knows exactly how much we canimprove productivity in practice, the American PhysicalSociety indicated in their 1974 energy conservation studythat it is theoretically possible to achieve an eightfoldimprovement of the 1972 energy/production ratio. 9 Mostcertainly, we are a long way from an economic saturationof the opportunities (see, e.g., Ref. 10). The commonargument that not much can be done after a 15 or 20%improvement has been realized ought to be dismissedas baseless. Energy productivity provides an expandingopportunity, not a last resort. The chapters in this bookprovide the information that is necessary to make themost of that opportunity in each organization.References1. Statistical Abstract of the United States, U.S. Government PrintingOffice, Washington, D.C., 1999.2. Energy User News, Jan. 14, 1980.3. JOHN G. WINGER et al., Outlook for Energy in the United States

8 ENERGY MANAGEMENT HANDBOOKto 1985, The Chase Manhattan Bank, New York, 1972, p 52.4. DONELLA H. MEADOWS et al., The Limits to Growth, UniverseBooks, New York, 1972, pp. 153-154.5. JIMMY E. CARTER, July 15, 1979, “Address to the Nation,” WashingtonPost, July 16, 1979, p. A14.6. Monthly Energy Review, Jan. 1980, U.S. Department of Energy,Washington, D.C., p. 16.7. Monthly Energy Review, Jan. 1980, U.S. Department of Energy,Washington D.C., p. 8; Statistical Abstract of the United States, U.S.Government Printing Office, Washington, D.C., 1979, Table 1409;Energy User News, Jan. 20, 1980, p. 14.8. American Association for the Advancement of Science, “U.S. EnergyDemand: Some Low Energy Futures,” Science, Apr. 14, 1978,p. 143.9. American Physical Society Summer Study on Technical Aspects ofEfficient Energy Utilization, 1974. Available as W.H. CARNAHANet al., Efficient Use of Energy, a Physics Perspective, from NTIS PB-242-773, or in Efficient Energy Use, Vol. 25 of the American Instituteof Physics Conference Proceedings.10. R.W. SANT, The Least-Cost Energy Strategy, Carnegie-Mellon UniversityPress, Pittsburgh, Pa., 197911. U.S. Congress Office of Technology Assessment (OTA). Energy Efficiencyin the Federal Government: Government by Good Example?OTA-E-492, U.S. Government Printing Office, Washington D.C., May1991.12. U.S. Air Force. DOD Energy Manager’s Handbook Volume 1: InstallationEnergy Management. Washington D.C., April 1993.

CHAPTER 2EFFECTIVE ENERGY MANAGEMENTWILLIAM H. MASHBURN, P.E., CEMProfessor EmeritusMechanical Engineering DepartmentVirginia Polytechnic Institute & State UniversityBlacksburg, Virginia2.1 INTRODUCTIONSome years ago, a newspaper headline stated,“Lower energy use leaves experts pleased but puzzled.”The article went on to state “Although the data arepreliminary, experts are baffled that the country appearsto have broken the decades-old link between economicgrowth and energy consumption.”For those involved in energy management, thiscomes as no surprise. We have seen companies becomingmore efficient in their use of energy, and that’s showingin the data. Those that have extracted all possiblesavings from downsizing, are now looking for otherways to become more competitive. Better managementof energy is a viable way, so there is an upward trend inthe number of companies that are establishing an energymanagement program. Management is now beginningto realize they are leaving a lot of money on the tablewhen they do not instigate a good energy managementplan.With the new technologies and alternative energysources now available, this country could possibly reduceits energy consumption by 50%—if there wereno barriers to the implementation. But of course, thereare barriers, mostly economic. Therefore, we mightconclude that managing energy is not a just technicalchallenge, but one of how to best implement thosetechnical changes within economic limits, and with aminimum of disruption.Unlike other management fads that have comeand gone, such as value analysis and quality circles, theneed to manage energy will be permanent within oursociety.There are several reasons for this:• There is a direct economic return. Most opportunitiesfound in an energy survey have less than a twoyear payback. Some are immediate, such as loadshifting or going to a new electric rate schedule.9• Most manufacturing companies are looking for acompetitive edge. A reduction in energy costs tomanufacture the product can be immediate andpermanent. In addition, products that use energy,such as motor driven machinery, are beingevaluated to make them more energy efficient, andtherefore more marketable. Many foreign countrieswhere energy is more critical, now want to knowthe maximum power required to operate a pieceof equipment.• Energy technology is changing so rapidly thatstate-of-the-art techniques have a half life of tenyears at the most. Someone in the organizationmust be in a position to constantly evaluate andupdate this technology.• Energy security is a part of energy management.Without a contingency plan for temporary shortagesor outages, and a strategic plan for long rangeplans, organizations run a risk of major problemswithout immediate solutions.• Future price shocks will occur. When world energymarkets swing wildly with only a five percent decreasein supply, as they did in 1979, it is reasonableto expect that such occurrences will happenagain.Those people then who choose—or in many casesare drafted—to manage energy will do well to recognizethis continuing need, and exert the extra effort to becomeskilled in this emerging and dynamic profession.The purpose of this chapter is to provide the fundamentalsof an energy management program that can be,and have been, adapted to organizations large and small.Developing a working organizational structure may bethe most important thing an energy manager can do.2.2 ENERGY MANAGEMENT PROGRAMAll the components of a comprehensive energymanagement program are depicted in Figure 2-1. Thesecomponents are the organizational structure, a policy, andplans for audits, education, reporting, and strategy. It ishoped that by understanding the fundamentals of managingenergy, the energy manager can then adapt a good

10 ENERGY MANAGEMENT HANDBOOKworking program to the existing organizational structure.Each component is discussed in detail below.2.3 ORGANIZATIONAL STRUCTUREThe organizational chart for energy managementshown in Figure 2-1 is generic. It must be adapted tofit into an existing structure for each organization. Forexample, the presidential block may be the general manager,and VP blocks may be division managers, but thefundamental principles are the same. The main featureof the chart is the location of the energy manager. Thisposition should be high enough in the organizationalstructure to have access to key players in management,and to have a knowledge of current events within thecompany. For example, the timing for presenting energyprojects can be critical. Funding availability and othermanagement priorities should be known and understood.The organizational level of the energy manageris also indicative of the support management is willingto give to the position.2.3.1 Energy ManagerOne very important part of an energy managementprogram is to have top management support. More important,however, is the selection of the energy manager,who can among other things secure this support. Theperson selected for this position should be one with avision of what managing energy can do for the company.Every successful program has had this one thingin common—one person who is a shaker and mover thatmakes things happen. The program is then built aroundthis person.There is a great tendency for the energy managerto become an energy engineer, or a prima donna, and attemptto conduct the whole effort alone. Much has beenaccomplished in the past with such individuals workingalone, but for the long haul, managing the program byinvolving everyone at the facility is much more productiveand permanent. Developing a working organizationalstructure may be the most important thing anenergy manager can do.The role and qualifications of the energy managerhave changed substantially in the past few years, causedmostly by EPACT-1992 requiring certification of federalenergy managers, deregulation of the electric utility industrybringing both opportunity and uncertainty, andby performance contracting requiring more businessskills than engineering. In her book titled “PerformanceContracting: Expanded Horizons,” Shirley Hansen givethe following requirements for an energy management:• Set up an Energy Management Plan• Establish energy records• Identify outside assistance• Assess future energy needs• Identify financing sources• Make energy recommendations• Implement recommendationsENERGY MANAGEMENT PROGRAMPresidentPolicyVPVPVPAudit PlanEnergy ManagerEducationalPlanCoordinatorCoordinatorCoordinatorReportingSystemEmployeesStrategicPlanFigure 2.1

EFFECTIVE ENERGY MANAGEMENT 11• Provide liaison for the energy committee• Plan communication strategies• Evaluate program effectivenessEnergy management programs can, and have,originated within one division of a large corporation.The division, by example and savings, motivates peopleat corporate level to pick up on the program and makeenergy management corporate wide. Many also originateat corporate level with people who have facilitiesresponsibility, and have implemented a good corporatefacilities program. They then see the importance andpotential of an energy management program, and takea leadership role in implementing one. In every caseobserved by the author, good programs have beeninstigated by one individual who has recognized thepotential, is willing to put forth the effort—in additionto regular duties—will take the risk of pushing newconcepts, and is motivated by a seemingly higher callingto save energy.If initiated at corporate level, there are some advantagesand some precautions. Some advantages are:• More resources are available to implement theprogram, such as budget, staff, and facilities.• If top management support is secured at corporatelevel, getting management support at divisionlevel is easier.• Total personnel expertise throughout the corporationis better known and can be identified andmade known to division energy managers.• Expensive test equipment can be purchased andmaintained at corporate level for use by divisionsas needed.• A unified reporting system can be put in place.• Creative financing may be the most needed andthe most important assistance to be provided fromcorporate level.• Impacts of energy and environmental legislationcan best be determined at corporate level.• Electrical utility rates and structures, as well aseffects of unbundling of electric utilities, can beevaluated at corporate level.Some precautions are:• Many people at division level may have alreadydone a good job of saving energy, and are cautiousabout corporate level staff coming in and takingcredit for their work.• All divisions don’t progress at the same speed.Work with those who are most interested first,then through the reporting system to top managementgive them credit. Others will then requestassistance.2.3.2 Energy TeamThe coordinators shown in Figure 2-1 representthe energy management team within one given organizationalstructure, such as one company within acorporation. This group is the core of the program. Themain criteria for membership should be an indicationof interest. There should be a representative from theadministrative group such as accounting or purchasing,someone from facilities and/or maintenance, and arepresentative from each major department.This energy team of coordinators should be appointedfor a specific time period, such as one year.Rotation can then bring new people with new ideas,can provide a mechanism for tactfully removing nonperformers,and involve greater numbers of people inthe program in a meaningful way.Coordinators should be selected to supplementskills lacking in the energy manager since, as pointedout above, it is unrealistic to think one energy managercan have all the qualifications outlined. So, total skillsneeded for the team, including the energy manager maybe defined as follows:• Have enough technical knowledge within thegroup to either understand the technology usedby the organization, or be trainable in that technology.• Have a knowledge of potential new technologythat may be applicable to the program.• Have planning skills that will help establish theorganizational structure, plan energy surveys, determineeducational needs, and develop a strategicenergy management plan.• Understand the economic evaluation system usedby the organization, particularly payback and lifecycle cost analysis.• Have good communication and motivationalskills since energy management involves everyonewithin the organization.

12 ENERGY MANAGEMENT HANDBOOKThe strengths of each team member should beevaluated in light of the above desired skills, and theirassignments made accordingly.2.3.3 EmployeesEmployees are shown as a part of the organizationalstructure, and are perhaps the greatest untappedresource in an energy management program. A structuredmethod of soliciting their ideas for more efficientuse of energy will prove to be the most productive effortof the energy management program. A good energymanager will devote 20% of total time working withemployees. Too many times employee involvement islimited to posters that say “Save Energy.”Employees in manufacturing plants generallyknow more about the equipment than anyone else inthe facility because they operate it. They know how tomake it run more efficiently, but because there is nomechanism in place for them to have an input, theirideas go unsolicited.An understanding of the psychology of motivationis necessary before an employee involvement programcan be successfully conducted. Motivation may be definedas the amount of physical and mental energy thata worker is willing to invest in his or her job. Three keyfactors of motivation are listed below:• Motivation is already within people. The task ofthe supervisor is not to provide motivation, but toknow how to release it.• The amount of energy and enthusiasm people arewilling to invest in their work varies with the individual.Not all are over-achievers, but not all arelazy either.• The amount of personal satisfaction to be deriveddetermines the amount of energy an employee willinvest in the job.Achieving personal satisfaction has been thesubject of much research by industrial psychologists,and they have emerged with some revealing facts. Forexample. They have learned that most actions taken bypeople are done to satisfy a physical need—such as theneed for food—or an emotional need—such as the needfor acceptance, recognition, or achievement.Research has also shown that many efforts to motivateemployees deal almost exclusively with trying tosatisfy physical needs, such as raises, bonuses, or fringebenefits. These methods are effective only for the shortterm, so we must look beyond these to other needs thatmay be sources of releasing motivation,A study done by Heresy and Blanchard [1] in 1977asked workers to rank job related factors listed below.The results were as follows:1. Full appreciation for work done2. Feeling “in” on things3. Understanding of personal problems4. Job security5. Good wages6. Interesting work7. Promoting and growth in the company8. Management loyalty to workers9. Good working conditions10. Tactful discipline of workersThis priority list would no doubt change with timeand with individual companies, but the rankings of whatsupervisors thought employees wanted were almost diametricallyopposed. They ranked good wages as first.It becomes obvious from this that job enrichmentis a key to motivation. Knowing this, the energy managercan plan a program involving employees that canprovide job enrichment by some simple and inexpensiverecognitions.Some things to consider in employee motivationare as follows:• There appears to be a positive relationship betweenfear arousal and persuasion if the fear appealsdeal with topics primarily of significance to theindividual; e.g., personal well being.• The success of persuasive communication is directlyrelated to the credibility of the source of communicationand may be reduced if recommendedchanges deviate too far from existing beliefs andpractices.• When directing attention to conservation, displaythe reminder at the point of action at the appropriatetime for action, and specify who is responsiblefor taking the action and when it should occur.Generic posters located in the work area are noteffective.• Studies have shown that pro-conservation attitudesand actions will be enhanced through associationswith others with similar attitudes, such as beingpart of an energy committee.• Positive effects are achieved with financial incentivesif the reward is in proportion to the savings,

EFFECTIVE ENERGY MANAGEMENT 13and represents respectable increments of spendableincome.• Consumers place considerable importance on thepotential discomfort in reducing their consumptionof energy. Changing thermostat settings from thecomfort zone should be the last desperate act foran energy manager.• Social recognition and approval is important, andcan occur through such things as the award ofmedals, designation of employee of the month,and selection to membership in elite sub-groups.Note that the dollar cost of such recognitions isminimal.• The potentially most powerful source of socialincentives for conservation behavior—but the leastused—is the commitment to others that occurs inthe course of group decisions.Before entering seriously into a program involvingemployees, be prepared to give a heavy commitment oftime and resources. In particular, have the resources torespond quickly to their suggestions.manager, coordinators, and any committees or taskgroups.• Reporting—Without authority from top management,it is often difficult for the energy managerto require others within the organization to complywith reporting requirements necessary to properlymanage energy. The policy is the place to establishthis. It also provides a legitimate reason forrequesting funds for instrumentation to measureenergy usage.• Training—If training requirements are establishedin the policy, it is again easier to include this inbudgets. It should include training at all levelswithin the organization.Many companies, rather that a comprehensivepolicy encompassing all the features described above,choose to go with a simpler policy statement.Appendices A and B give two sample energypolicies. Appendix A is generic and covers the itemsdiscussed above. Appendix B is a policy statement of amultinational corporation.2.4. ENERGY POLICYA well written energy policy that has been authorizedby management is as good as the proverbiallicense to steal. It provides the energy manager with theauthority to be involved in business planning, new facilitylocation and planning, the selection of productionequipment, purchase of measuring equipment, energyreporting, and training—things that are sometimes difficultto do.If you already have an energy policy, chances arethat it is too long and cumbersome. To be effective,the policy should be short—two pages at most. Manypeople confuse the policy with a procedures manual. Itshould be bare bones, but contain the following itemsas a minimum:• Objectives—this can contain the standard motherhoodand flag statements about energy, butthe most important is that the organization willincorporate energy efficiency into facilities andnew equipment, with emphasis on life cycle costanalysis rather than lowest initial cost.• Accountability—This should establish the organizationalstructure and the authority for the energy2.5 PLANNINGPlanning is one of the most important parts of theenergy management program, and for most technicalpeople is the least desirable. It has two major functionsin the program. First, a good plan can be a shield fromdisruptions. Second, by scheduling events throughoutthe year, continuous emphasis can be applied to theenergy management program, and will play a major rolein keeping the program active.Almost everyone from top management to the custodiallevel will be happy to give an opinion on what canbe done to save energy. Most suggestions are worthless. Itis not always wise from a job security standpoint to saythis to top management. However, if you inform people—especially top management—that you will evaluate theirsuggestion, and assign a priority to it in your plan, notonly will you not be disrupted, but may be consideredeffective because you do have a plan.Many programs were started when the fear ofenergy shortages was greater, but they have declinedinto oblivion. By planning to have events periodicallythrough the year, a continued emphasis will be placedon energy management. Such events can be trainingprograms, audits, planning sessions, demonstrations,research projects, lectures, etc.

14 ENERGY MANAGEMENT HANDBOOKThe secret to a workable plan is to have peoplewho are required to implement the plan involved inthe planning process. People feel a commitment to makingthings work if they have been a part of the design.This is fundamental to any management planning, butmore often that not is overlooked. However, in order toprevent the most outspoken members of a committeefrom dominating with their ideas, and rejecting ideasfrom less outspoken members, a technique for managingcommittees must be used. A favorite of the author is theNominal Group Technique developed at the Universityof Wisconsin in the late 1980’s by Andre Delbecq andAndrea Van de Ven [2]. This technique consists of thefollowing basic steps:1. Problem definition—The problem is clearly definedto members of the group.2. Grouping—Divide large groups into smallergroups of seven to ten, then have the group electa recording secretary.3. Silent generation of ideas—Each person silentlyand independently writes as many answers to theproblem as can be generated within a specifiedtime.4. Round-robin listing—Secretary lists each ideaindividually on an easel until all have been recorded.5. Discussion—Ideas are discussed for clarification,elaboration, evaluation and combining.6. Ranking—Each person ranks the five most importantitems. The total number of points receivedfor each idea will determine the first choice ofthe group.2.6 AUDIT PLANNINGThe details of conducting audits are discussed ina comprehensive manner in Chapter 4, but planningshould be conducted prior to the actual audits. The planningshould include types of audits to be performed,team makeup, and dates.By making the audits specific rather than generalin nature, much more energy can be saved. Examples ofsome types of audits that might be considered are:• Tuning-Operation-Maintenance (TOM)• Compressed air• Motors• Lighting• Steam system• Water• Controls• HVAC• Employee suggestionsBy defining individual audits in this manner, itis easy to identify the proper team for the audit. Don’tneglect to bring in outside people such as electric utilityand natural gas representatives to be team members.Scheduling the audits, then, can contribute to the eventsthat will keep the program active.With the maturing of performance contracting,energy managers have two choices for the energy auditprocess. They may go through the contracting process toselect and define the work of a performance contractor,or they can set up their own team and conduct audits,or in some cases such as a corporate energy manager,performance contracting may be selected for one facility,and energy auditing for another. Each has advantagesand disadvantages.Advantages of performance contracting are:• No investment is required of the company—otherthan that involved in the contracting process,which can be very time consuming.• A minimum of in-house people are involved,namely the energy manager and financial people.Disadvantages are:• Technical resources are generally limited to thecontracting organization.• Performance contracting is still maturing, andmany firms underestimate the work required• The contractor may not have the full spectrum ofskills needed.• The contractor may not have an interest in low/cost no/cost projects.Advantages of setting up an audit team are:• The team can be selected to match equipment to beaudited, and can be made up of in-house personnel,outside specialists, or best, a combination ofboth.

EFFECTIVE ENERGY MANAGEMENT 15• They can identify all potential energy conservationprojects, both low-cost/no-cost as well as largecapital investments.• The audit can be an excellent training tool byinvolving others in the process, and by adding atraining component as a part of the audit.Disadvantages of an audit team approach:• Financing identified projects becomes a separateissue for the energy manager.• It takes a well organized energy managementstructure to take full advantage of the work of theaudit team.2.7 EDUCATIONAL PLANNINGA major part of the energy manager’s job is toprovide some energy education to persons within theorganization. In spite of the fact that we have been concernedwith it for the past two decades, there is still asea of ignorance concerning energy.Raising the energy education level throughout theorganization can have big dividends. The program willoperate much more effectively if management understandsthe complexities of energy, and particularly thepotential for economic benefit; the coordinators will bemore effective is they are able to prioritize energy conservationmeasures, and are aware of the latest technology;the quality and quantity of employee suggestionswill improve significantly with training.Educational training should be considered forthree distinct groups—management, the energy team,and employees.2.7.1 Management TrainingIt is difficult to gain much of management’s time,so subtle ways must be developed to get them up tospeed. Getting time on a regular meeting to provide updateson the program is one way. When the momentumof the program gets going, it may be advantageous tohave a half or one day presentation for management.A good concise report periodically can be a toolto educate management. Short articles that are pertinentto your educational goals, taken from magazinesand newspapers can be attached to reports and sentselectively. Having management be a part of a trainingprogram for either the energy team or employees, orboth, can be an educational experience since we learnbest when we have to make a presentation.Ultimately, the energy manager should aspire tobe a part of business planning for the organization. Astrategic plan for energy should be a part of every businessplan. This puts the energy manager into a positionfor more contact with management people, and thus theopportunity to inform and teach.2.7.2 Energy Team TrainingSince the energy team is the core group of the energymanagement program, proper and thorough trainingfor them should have the highest priority. Trainingis available from many sources and in many forms.• Self study—this necessitates having a good libraryof energy related materials from which coordinatorscan select.• In-house training—may be done by a qualifiedmember of the team—usually the energy manager,or someone from outside.• Short courses offered by associations such as theAssociation of Energy Engineers [3], by individualconsultants, by corporations, and by colleges anduniversities.• Comprehensive courses of one to four weeksduration offered by universities, such the one atthe University of Wisconsin, and the one beingrun cooperatively by Virginia Tech and N.C. StateUniversity.For large decentralized organizations with perhapsten or more regional energy managers, an annual twoor three-day seminar can be the base for the educationalprogram. Such a program should be planned carefully.The following suggestions should be incorporated intosuch a program:• Select quality speakers from both inside and outsidethe organization.• This is an opportunity to get top managementsupport. Invite a top level executive from the organizationto give opening remarks. It may be wiseto offer to write the remarks, or at least to providesome material for inclusion.• Involve the participants in workshop activities sothey have an opportunity to have an input intothe program. Also, provide some practical tips

16 ENERGY MANAGEMENT HANDBOOKon energy savings that they might go back andimplement immediately. One or two good ideascan sometimes pay for their time in the seminar.• Make the seminar first class with professionalspeakers; a banquet with an entertaining—nottechnical—after dinner speaker; a manual thatincludes a schedule of events, biosketches of speakers,list of attendees, information on each topicpresented, and other things that will help pull thewhole seminar together. Vendors will contributethings for door prizes.• You may wish to develop a logo for the program,and include it on small favors such as cups, carryingcases, etc.2.7.3 Employee TrainingA systematic approach for involving employeesshould start with some basic training in energy. Thiswill produce a much higher quality of ideas fromthem. Employees place a high value on training, so aside benefit is that morale goes up. Simply teachingthe difference between electrical demand and kilowatthours of energy, and that compressed air is very expensiveis a start. Short training sessions on energy canbe injected into other ongoing training for employees,such as safety. A more comprehensive training programshould include:• Energy conservation in the home• Fundamentals of electric energy• Fundamentals of energy systems• How energy surveys are conducted and what tolook for2.8 STRATEGIC PLANNINGDeveloping an objective, strategies, programs, andaction items constitutes strategic planning for the energymanagement program. It is the last but perhaps themost important step in the process of developing theprogram, and unfortunately is where many stop. Thevery name “Strategic Planning” has an ominous soundfor those who are more technically inclined. However,by using a simplified approach and involving the energymanagement team in the process, a plan can be developedusing a flow chart that will define the program forthe next five years.If the team is involved in developing each of thecomponents of objective, strategies, programs, and actionitems—using the Nominal Group Technique—theresult will be a simplified flow chart that can be usedfor many purposes. First, it is a protective plan thatdiscourages intrusion into the program, once it is establishedand approved. It provides the basis for resourcessuch as funding and personnel for implementation. Itprojects strategic planning into overall planning by theorganization, and hence legitimizes the program at topmanagement level. By involving the implementers in theplanning process, there is a strong commitment to makeit work.Appendix C contains flow charts depicting a strategicplan developed in a workshop conducted by theauthor by a large defense organization. It is a modelplan in that it deals not only with the technical aspectsof energy management but also the funding, communications,education, and behavior modification.2.9 REPORTINGThere is no generic form to that can be usedfor reporting. There are too many variables such asorganization size, product, project requirements, andprocedures already in existence. The ultimate reportingsystem is one used by a chemical company making atextile product. The Btu/lb of product is calculated ona computer system that gives an instantaneous reading.This is not only a reporting system, but one that detectsmaintenance problems. Very few companies are set upto do this, but many do have some type of energy indexfor monthly reporting.When energy prices fluctuate wildly, the best energyindex is usually based on Btus; but, when energyprices are stable, the best index is dollars. However,there are still many factors that will influence any index,such as weather, production, expansion or contraction offacilities, new technologies, etc.The bottom line is that any reporting system hasto be customized to suit individual circumstances. And,while reporting is not always the most glamorous partof managing energy, it can make a contribution to theprogram by providing the bottom line on its effectiveness.It is also a straight pipeline into management, andcan be a tool for promoting the program.The report is probably of most value to the onewho prepares it. It is a forcing function that requires allinformation to be pulled together in a coherent manner.This requires much thought and analysis that might nototherwise take place.By making reporting a requirement of the energy

EFFECTIVE ENERGY MANAGEMENT 17policy, getting the necessary support can be easier. Inmany cases, the data may already be collected on aperiodic basis and put into a computer. It may simplyrequire combining production data and energy data todevelop an energy index.Keep the reporting requirements as simple as possible.The monthly report could be something as simpleas adding to an ongoing graph that compares presentusage to some baseline year. Any narrative should beshort, with data kept in a file that can be provided forany supporting in-depth information.With all the above considered, the best way toreport is to do it against an audit that has been performedat the facility. One large corporation has itsfacilities report in this manner, and then has an awardfor those that complete all energy conservation measureslisted on the audit.2.10 OWNERSHIPThe key to a successful energy management programis within this one word— ownership. This extendsto everyone within the organization. Employees thatoperate a machine “own” that machine. Any attemptto modify their “baby” without their participation willnot succeed. They have the knowledge to make or breakthe attempt. Members of the energy team are not goingto be interested in seeing one person—the energymanger—get all the fame and glory for their efforts.Management people that invest in energy projects wantto share in the recognition for their risk taking. A corporateenergy team that goes into a division for an energyaudit must help put a person from the division in theenergy management position, then make sure the auditbelongs to the division. Below are more tips for successthat have been compiled from observing successful energymanagement programs.• Have a plan. A plan dealing with organization,surveys, training, and strategic planning—withevents scheduled—has two advantages. It preventsdisruptions by non-productive ideas, and it sets upscheduled events that keeps the program active.• Give away—or at least share—ideas for savingenergy. The surest way to kill a project is to bepossessive. If others have a vested interest theywill help make it work.• Be aggressive. The energy team—after some training—willbe the most energy knowledgeable groupwithin the company. Too many management decisionsare made with a meager knowledge of theeffects on energy.• Use proven technology. Many programs getbogged down trying to make a new technologywork, and lose sight of the easy projects with goodpayback. Don’t buy serial number one. In spite ofprice breaks and promise of vendor support, it canbe all consuming to make the system work.• Go with the winners. Not every department withina company will be enthused about the energy program.Make those who are look good through thereporting system to top management, and all willfollow.• A final major tip—ask the machine operator whatshould be done to reduce energy. Then make surethey get proper recognition for ideas.2.11 SUMMARYLet’s now summarize by assuming you have justbeen appointed energy manager of a fairly large company.What are the steps you might consider in settingup an energy management program? Here is a suggestedprocedure.2.11.1 Situation AnalysisDetermine what has been done before. Was therea previous attempt to establish an energy managementprogram? What were the results of this effort? Next,plot the energy usage for all fuels for the past two—ormore—years, then project the usage, and cost, for thenext five years at the present rate. This will not onlyhelp you sell your program, but will identify areas ofconcentration for reducing energy.2.11.2 PolicyDevelop some kind of acceptable policy that givesauthority to the program. This will help later on withsuch things as reporting requirements, and need formeasurement instrumentation.2.11.3 OrganizationSet up the energy committee and/or coordinators.2.11.4 TrainingWith the committee involvement, develop a trainingplan for the first year.

18 ENERGY MANAGEMENT HANDBOOK2.11.5 AuditsAgain with the committee involvement, developan auditing plan for the first year.2.11.6 ReportingDevelop a simple reporting system.2.11.7 ScheduleFrom the above information develop a scheduleof events for the next year, timing them so as to giveperiodic actions from the program, which will help keepthe program active and visible.2.11.8 Implement the program2.12 CONCLUSIONEnergy management has now matured to the pointthat it offers outstanding opportunities for those willingto invest time and effort to learn the fundamentals. Itrequires technical and management skills which broadenseducational needs for both technical and managementpeople desiring to enter this field. Because of theeconomic return of energy management, it is attractiveto top management, so exposure of the energy managerat this level brings added opportunity for recognitionand advancement. Managing energy will be a continuousneed, so persons with this skill will have personaljob security as we are caught up in the down sizing fadnow permeating our society.References1. Hersey, Paul and Kenneth H. Blanchard, Management of OrganizationalBehavior: Utilizing Human Resources, Harper and Row, 19702. Delbecq, Andre L., Andrew H. Van de Ven, and David H. Gustafson,Group Techniques for Program Planning, Green Briar Press,1986.3. Mashburn, William H., Managing Energy Resources in Times of DynamicChange, Fairmont Press, 19924. Turner, Wayne, Energy Management Handbook, 2nd edition, Chapter2, Fairmont Press, 1993.Appendix AENERGY POLICYAcme Manufacturing CompanyPolicy and Procedures ManualSubject: Energy Management ProgramI. PolicyEnergy Management shall be practiced in all areasof the Company’s operation.II. Energy Management Program ObjectivesIt is the Company’s objective to use energy efficientlyand provide energy security for the organizationfor both immediate and long range by:• Utilizing energy efficiently throughout the Company’soperations.• Incorporating energy efficiency into existing equipmentand facilities, and in the selection and purchaseof new equipment.• Complying with government regulations—federal,state, and local.• Putting in place an Energy Management Programto accomplish the above objectives.III. ImplementationA. OrganizationThe Company’s Energy Management Program shall beadministered through the Facilities Department.1. Energy ManagerThe Energy Manager shall report directly to theVice President of Facilities, and shall have overall responsibilityfor carrying out the Energy ManagementProgram.2. Energy CommitteeThe Energy Manager may appoint and EnergyCommittee to be comprised of representatives fromvarious departments. Members will serve for a specifiedperiod of time. The purpose of the Energy Committee isto advise the Energy Manager on the operation of theEnergy Management Program, and to provide assistanceon specific tasks when needed.3. Energy CoordinatorsEnergy Coordinators shall be appointed to representa specific department or division. The Energy Managershall establish minimum qualification standards forCoordinators, and shall have joint approval authority foreach Coordinator appointed.Coordinators shall be responsible for maintainingan ongoing awareness of energy consumption andexpenditures in their assigned areas. They shall recommendand implement energy conservation projects andenergy management practices.

EFFECTIVE ENERGY MANAGEMENT 19Coordinators shall provide necessary informationfor reporting from their specific areas.They may be assigned on a full-time or part-timebasis; as required to implement programs in their areas.B. ReportingThe energy Coordinator shall keep the Energy Officeadvised of all efforts to increase energy efficiency intheir areas. A summary of energy cost savings shall besubmitted each quarter to the Energy Office.The Energy Manager shall be responsible for consolidatingthese reports for top management.cost-effective programs that will maintain or improvethe quality of the work environment, optimize servicereliability, increase productivity, and enhance the safetyof our workplace.Appendix CC. TrainingThe Energy Manager shall provide energy trainingat all levels of the Company.IV. Policy UpdatingThe Energy Manager and the Energy AdvisoryCommittee shall review this policy annually and makerecommendations for updating or changes.Appendix BPOLICY STATEMENTAcme International Corporation is committed tothe efficient, cost effective, and environmentally responsibleuse of energy throughout its worldwide operations.Acme will promote energy efficiency by implementing Figure 2.2

20 ENERGY MANAGEMENT HANDBOOKFigure 2.3 Figure 2.4

EFFECTIVE ENERGY MANAGEMENT 21Figure 2.5 Figure 2.6

22 ENERGY MANAGEMENT HANDBOOKFigure 2.7

CHAPTER 3ENERGY AUDITINGBARNEY L. CAPEHART AND MARK B. SPILLERUniversity of Florida — Gainesville Regional UtilitiesGainesville, FLSCOTT FRAZIEROklahoma State University3.1 INTRODUCTIONSaving money on energy bills is attractive to businesses,industries, and individuals alike. Customers whoseenergy bills use up a large part of their income, andespecially those customers whose energy bills representa substantial fraction of their company’s operating costs,have a strong motivation to initiate and continue an ongoingenergy cost-control program. No-cost or very lowcostoperational changes can often save a customer oran industry 10-20% on utility bills; capital cost programswith payback times of two years or less can often savean additional 20-30%. In many cases these energy costcontrol programs will also result in both reduced energyconsumption and reduced emissions of environmentalpollutants.The energy audit is one of the first tasks to beperformed in the accomplishment of an effective energycost control program. An energy audit consists of a detailedexamination of how a facility uses energy, whatthe facility pays for that energy, and finally, a recommendedprogram for changes in operating practices orenergy-consuming equipment that will cost-effectivelysave dollars on energy bills. The energy audit is sometimescalled an energy survey or an energy analysis, sothat it is not hampered with the negative connotation ofan audit in the sense of an IRS audit. The energy auditis a positive experience with significant benefits to thebusiness or individual, and the term “audit” should beavoided if it clearly produces a negative image in themind of a particular business or individual.3.2 ENERGY AUDITING SERVICESEnergy audits are performed by several differentgroups. Electric and gas utilities throughout the countryoffer free residential energy audits. A utility’s residentialenergy auditors analyze the monthly bills, inspect theconstruction of the dwelling unit, and inspect all of theenergy-consuming appliances in a house or an apartment.Ceiling and wall insulation is measured, ducts areinspected, appliances such as heaters, air conditioners,water heaters, refrigerators, and freezers are examined,and the lighting system is checked.Some utilities also perform audits for their industrialand commercial customers. They have professionalengineers on their staff to perform the detailed auditsneeded by companies with complex process equipmentand operations. When utilities offer free or low-costenergy audits for commercial customers, they usuallyonly provide walk-through audits rather than detailedaudits. Even so, they generally consider lighting, HVACsystems, water heating, insulation and some motors.Large commercial or industrial customers may hirean engineering consulting firm to perform a completeenergy audit. Other companies may elect to hire anenergy manager or set up an energy management teamwhose job is to conduct periodic audits and to keep upwith the available energy efficiency technology.The U.S. Department of Energy (U.S. DOE) funds aprogram where universities around the country operateIndustrial Assessment Centers which perform free energyaudits for small and medium sized manufacturingcompanies. There are currently 30 IACs funded by theIndustrial Division of the U.S. DOE.The Institutional Conservation Program (ICP) isanother energy audit service funded by the U.S. Departmentof Energy. It is usually administered through stateenergy offices. This program pays for audits of schools,hospitals, and other institutions, and has some fundingassistance for energy conservation improvements.3.3 BASIC COMPONENTS OF AN ENERGY AUDITAn initial summary of the basic steps involved inconducting a successful energy audit is provided here,and these steps are explained more fully in the sectionsthat follow. This audit description primarily addressesthe steps in an industrial or large-scale commercialaudit, and not all of the procedures described in thissection are required for every type of audit.The audit process starts by collecting informationabout a facility’s operation and about its past record ofutility bills. This data is then analyzed to get a picture ofhow the facility uses—and possibly wastes—energy, aswell as to help the auditor learn what areas to examine23

24 ENERGY MANAGEMENT HANDBOOKto reduce energy costs. Specific changes—called EnergyConservation Opportunities (ECOs)—are identified andevaluated to determine their benefits and their cost-effectiveness.These ECOs are assessed in terms of theircosts and benefits, and an economic comparison is madeto rank the various ECOs. Finally, an Action Plan is createdwhere certain ECOs are selected for implementation,and the actual process of saving energy and savingmoney begins.3.3.1 The Auditor’s ToolboxTo obtain the best information for a successfulenergy cost control program, the auditor must makesome measurements during the audit visit. The amountof equipment needed depends on the type of energyconsumingequipment used at the facility, and on therange of potential ECOs that might be considered. Forexample, if waste heat recovery is being considered,then the auditor must take substantial temperaturemeasurement data from potential heat sources. Toolscommonly needed for energy audits are listed below:Tape MeasuresThe most basic measuring device needed is thetape measure. A 25-foot tape measure l" wide and a 100-foot tape measure are used to check the dimensions ofwalls, ceilings, windows and distances between piecesof equipment for purposes such as determining thelength of a pipe for transferring waste heat from onepiece of equipment to the other.LightmeterOne simple and useful instrument is the lightmeterwhich is used to measure illumination levels infacilities. A lightmeter that reads in footcandles allowsdirect analysis of lighting systems and comparison withrecommended light levels specified by the IlluminatingEngineering Society. A small lightmeter that is portableand can fit into a pocket is the most useful. Many areasin buildings and plants are still significantly overlighted,and measuring this excess illumination thenallows the auditor to recommend a reduction in lightinglevels through lamp removal programs or by replacinginefficient lamps with high efficiency lamps that maynot supply the same amount of illumination as the oldinefficient lamps.ThermometersSeveral thermometers are generally needed tomeasure temperatures in offices and other worker areas,and to measure the temperature of operating equipment.Knowing process temperatures allows the auditor todetermine process equipment efficiencies, and also toidentify waste heat sources for potential heat recoveryprograms. Inexpensive electronic thermometers withinterchangeable probes are now available to measuretemperatures in both these areas. Some common typesinclude an immersion probe, a surface temperatureprobe, and a radiation shielded probe for measuringtrue air temperature. Other types of infra-red thermometersand thermographic equipment are also available.An infra-red “gun” is valuable for measuring temperaturesof steam lines that are not readily reached withouta ladder.Infrared CamerasInfrared cameras are still rather expensive pieces ofequipment. An investment of at least $25,000 is neededto have a quality infrared camera. However, these arevery versatile pieces of equipment and can be used tofind overheated electrical wires, connections, neutrals,circuit breakers, transformers, motors and other piecesof electrical equipment. They can also be used to findwet insulation, missing insulation, roof leaks, and coldspots. Thus, infrared cameras are excellent tools for bothsafety related diagnostics and energy savings diagnostics.A good rule of thumb is that if one safety hazard isfound during an infrared scan of a facility, then that haspaid for the cost of the scan for the entire facility. Manyinsurers require infrared scans of buildings for facilitiesonce a year.VoltmeterAn inexpensive voltmeter is useful for determiningoperating voltages on electrical equipment, and especiallyuseful when the nameplate has worn off of a pieceof equipment or is otherwise unreadable or missing. Themost versatile instrument is a combined volt-ohm-ammeterwith a clamp-on feature for measuring currents inconductors that are easily accessible. This type of multimeteris convenient and relatively inexpensive. Anynewly purchased voltmeter, or multimeter, should be atrue RMS meter for greatest accuracy where harmonicsmight be involved.Clamp On AmmeterThese are very useful instruments for measuringcurrent in a wire without having to make any liveelectrical connections. The clamp is opened up and putaround one insulated conductor, and the meter readsthe current in that conductor. New clamp on ammeterscan be purchased rather inexpensively that read trueRMS values. This is important because of the level ofharmonics in many of our facilities. An idea of the level

ENERGY AUDITING 25of harmonics in a load can be estimated from using anold non-RMS ammeter, and then a true RMS ammeterto measure the current. If there is more than a five toten percent difference between the two readings, thereis a significant harmonic content to that load.Wattmeter/Power Factor MeterA portable hand-held wattmeter and power factormeter is very handy for determining the powerconsumption and power factor of individual motorsand other inductive devices. This meter typically has aclamp-on feature which allows an easy connection to thecurrent-carrying conductor, and has probes for voltageconnections. Any newly purchased wattmeter or powerfactor meter, should be a true RMS meter for greatestaccuracy where harmonics might be involvedCombustion AnalyzerCombustion analyzers are portable devices capableof estimating the combustion efficiency of furnaces, boilers,or other fossil fuel burning machines. Two typesare available: digital analyzers and manual combustionanalysis kits. Digital combustion analysis equipmentperforms the measurements and reads out in percentcombustion efficiency. These instruments are fairly complexand expensive.The manual combustion analysis kits typicallyrequire multiple measurements including exhauststack temperature, oxygen content, and carbon dioxidecontent. The efficiency of the combustion process canbe calculated after determining these parameters. Themanual process is lengthy and is frequently subject tohuman error.Airflow Measurement DevicesMeasuring air flow from heating, air conditioningor ventilating ducts, or from other sources of air flowis one of the energy auditor’s tasks. Airflow measurementdevices can be used to identify problems with airflows, such as whether the combustion air flow into agas heater is correct. Typical airflow measuring devicesinclude a velometer, an anemometer, or an airflow hood.See section 3.4.3 for more detail on airflow measurementdevices.Blower Door AttachmentBuilding or structure tightness can be measuredwith a blower door attachment. This device is frequentlyused in residences and in office buildings to determinethe air leakage rate or the number of air changesper hour in the facility. This is often helps determinewhether the facility has substantial structural or ductleaks that need to be found and sealed. See section 3.4.2for additional information on blower doors.Smoke GeneratorA simple smoke generator can also be used inresidences, offices and other buildings to find air infiltrationand leakage around doors, windows, ducts andother structural features. Care must be taken in usingthis device, since the chemical “smoke” produced maybe hazardous, and breathing protection masks may beneeded. See section 3.4.1 for additional information onthe smoke generation process, and use of smoke generators.Safety EquipmentThe use of safety equipment is a vital precautionfor any energy auditor. A good pair of safety glasses isan absolute necessity for almost any audit visit. Hearingprotectors may also be required on audit visits to noisyplants or areas with high horsepower motors drivingfans and pumps. Electrical insulated gloves should beused if electrical measurements will be taken, and asbestosgloves should be used for working around boilersand heaters. Breathing masks may also be needed whenhazardous fumes are present from processes or materialsused. Steel-toe and steel-shank safety shoes may beneeded on audits of plants where heavy materials, hotor sharp materials or hazardous materials are beingused. (See section 3.3.3 for an additional discussion ofsafety procedures.)Miniature Data LoggersMiniature—or mini—data loggers have appearedin low cost models in the last five years. These are oftendevices that can be held in the palm of the hand, andare electronic instruments that record measurements oftemperature, relative humidity, light intensity, light on/off, and motor on/off. If they have an external sensorinput jack, these little boxes are actually general purposedata loggers. With external sensors they can record measurementsof current, voltage, apparent power (kVA),pressure, and CO 2 .These data loggers have a microcomputer controlchip and a memory chip, so they can be initialized andthen can record data for periods of time from days toweeks. They can record data on a 24 hour a day basis,without any attention or intervention on the part ofthe energy auditor. Most of these data loggers interfacewith a digital computer PC, and can transfer data into aspreadsheet of the user’s choice, or can use the softwareprovided by the suppliers of the loggers.Collecting audit data with these small data loggers

26 ENERGY MANAGEMENT HANDBOOKgives a more complete and accurate picture of an energysystem’s overall performance because some conditionsmay change over long periods of time, or when no oneis present.Vibration Analysis GearRelatively new in the energy manager’s tool box isvibration analysis equipment. The correlation betweenmachine condition (bearings, pulley alignment, etc.)and energy consumption is related and this equipmentmonitors such machine health. This equipment comes invarious levels of sophistication and price. At the lowerend of the spectrum are vibration pens (or probes) thatsimply give real-time amplitude readings of vibratingequipment in in/sec or mm/sec. This type of equipmentcan cost under $1,000. The engineer comparesthe measured vibration amplitude to a list of vibrationlevels (ISO2372) and is able to determine if the vibrationis excessive for that particular piece of equipment.The more typical type of vibration equipment willmeasure and log the vibration into a database (on-boardand downloadable). In addition to simply measuringvibration amplitude, the machine vibration can be displayedin time or frequency domains. The graphs ofvibration in the frequency domain will normally exhibitspikes at certain frequencies. These spikes can be interpretedby a trained individual to determine the relativehealth of the machine monitored.The more sophisticated machines are capable oftrend analysis so that facility equipment can be monitoredon a schedule and changes in vibration (amplitudesand frequencies) can be noted. Such trending canbe used to schedule maintenance based on observationsof change. This type of equipment starts at about $3,000and goes up depending on features desired.3.3.2 Preparing for the Audit VisitSome preliminary work must be done before theauditor makes the actual energy audit visit to a facility.Data should be collected on the facility’s use of energythrough examination of utility bills, and some preliminaryinformation should be compiled on the physicaldescription and operation of the facility. This datashould then be analyzed so that the auditor can do themost complete job of identifying Energy ConservationOpportunities during the actual site visit to the facility.Energy Use DataThe energy auditor should start by collecting dataon energy use, power demand and cost for at least theprevious 12 months. Twenty-four months of data mightbe necessary to adequately understand some types ofbilling methods. Bills for gas, oil, coal, electricity, etc.should be compiled and examined to determine boththe amount of energy used and the cost of that energy.This data should then be put into tabular and graphicform to see what kind of patterns or problems appearfrom the tables or graphs. Any anomaly in the patternof energy use raises the possibility for some significantenergy or cost savings by identifying and controllingthat anomalous behavior. Sometimes an anomaly onthe graph or in the table reflects an error in billing, butgenerally the deviation shows that some activity is goingon that has not been noticed, or is not completelyunderstood by the customer.Rate StructuresTo fully understand the cost of energy, the auditormust determine the rate structure under which thatenergy use is billed. Energy rate structures may go fromthe extremely simple ones—for example, $1.00 per gallonof Number 2 fuel oil, to very complex ones—forexample, electricity consumption which may have acustomer charge, energy charge, demand charge, powerfactor charge, and other miscellaneous charges that varyfrom month to month. Few customers or businesses reallyunderstand the various rate structures that controlthe cost of the energy they consume. The auditor canhelp here because the customer must know the basis forthe costs in order to control them successfully.• Electrical Demand Charges: The demand charge isbased on a reading of the maximum power in kWthat a customer demands in one month. Power isthe rate at which energy is used, and it varies quiterapidly for many facilities. Electric utilities averagethe power reading over intervals from fifteen minutesto one hour, so that very short fluctuations donot adversely affect customers. Thus, a customermight be billed for demand for a month based ona maximum value of a fifteen minute integratedaverage of their power use.• Ratchet Clauses: Some utilities have a ratchetclause in their rate structure which stipulates thatthe minimum power demand charge will be thehighest demand recorded in the last billing periodor some percentage (i.e., typically 70%) of thehighest power demand recorded in the last year.The ratchet clause can increase utility charges forfacilities during periods of low activity or wherepower demand is tied to extreme weather.• Discounts/Penalties: Utilities generally provide

ENERGY AUDITING 27discounts on their energy and power rates forcustomers who accept power at high voltage andprovide transformers on site. They also commonlyassess penalties when a customer has a powerfactor less than 0.9. Inductive loads (e.g., lightlyloaded electric motors, old fluorescent lighting ballasts,etc.) reduce the power factor. Improvementcan be made by adding capacitance to correct forlagging power factor, and variable capacitor banksare most useful for improving the power factor atthe service drop. Capacitance added near the loadscan effectively increase the electrical system capacity.Turning off idling or lightly loaded motors canalso help.• Wastewater charges: The energy auditor also frequentlylooks at water and wastewater use andcosts as part of the audit visit. These costs are oftenrelated to the energy costs at a facility. Wastewatercharges are usually based on some proportion ofthe metered water use since the solids are difficultto meter. This can needlessly result in substantialincreases in the utility bill for processes whichdo not contribute to the wastewater stream (e.g.,makeup water for cooling towers and other evaporativedevices, irrigation, etc.). A water meter canbe installed at the service main to supply the loadsnot returning water to the sewer system. This canreduce the charges by up to two-thirds.Energy bills should be broken down into the componentsthat can be controlled by the facility. These costcomponents can be listed individually in tables and thenplotted. For example, electricity bills should be brokendown into power demand costs per kW per month, andenergy costs per kWh. The following example illustratesthe parts of a rate structure for an industry in Florida.Example: A company that fabricates metal productsgets electricity from its electric utility at the followinggeneral service demand rate structure.Rate structure:Customer cost = $21.00 per monthEnergy cost = $0.051 per kWhDemand cost = $6.50 per kW per monthTaxes = Total of 8%Fuel adjustment = A variable amount perkWh each monthThe energy use and costs for that company for a yearare summarized below:Summary of Energy Usage and CostsDemand TotalkWh Used kWh Cost Demand Cost CostMonth (kWh) ($) (kW) ($) ($)Mar 44960 1581.35 213 1495.26 3076.61Apr 47920 1859.68 213 1495.26 3354.94May 56000 2318.11 231 1621.62 3939.73Jun 56320 2423.28 222 1558.44 3981.72Jul 45120 1908.16 222 1558.44 3466.60Aug 54240 2410.49 231 1621.62 4032.11Sept 50720 2260.88 222 1558.44 3819.32Oct 52080 2312.19 231 1621.62 3933.81Nov 44480 1954.01 213 1495.26 3449.27Dec 38640 1715.60 213 1495.26 3210.86Jan 36000 1591.01 204 1432.08 3023.09Feb 42880 1908.37 204 1432.08 3340.45Totals 569,360 24,243.13 2,619 18,385.38 42,628.51Monthly 47,447 2,020.26 218 1,532.12 3,552.38Averages

28 ENERGY MANAGEMENT HANDBOOKThe auditor must be sure to account for all thetaxes, the fuel adjustment costs, the fixed charges, andany other costs so that the true cost of the controllableenergy cost components can be determined. In the electricrate structure described above, the quoted costs fora kW of demand and a kWh of energy are not completeuntil all these additional costs are added. Although therate structure says that there is a basic charge of $6.50per kW per month, the actual cost including all taxes is$7.02 per kW per month. The average cost per kWh ismost easily obtained by taking the data for the twelvemonth period and calculating the cost over this periodof time. Using the numbers from the table, one can seethat this company has an average energy cost of $0.075per kWh.These data are used initially to analyze potentialECOs and will ultimately influence which ECOs arerecommended. For example, an ECO that reduces peakdemand during a month would save $7.02 per kW permonth. Therefore, the auditor should consider ECOsthat would involve using certain equipment during thenight shift when the peak load is significantly less thanthe first shift peak load. ECOs that save both energy anddemand on the first shift would save costs at a rate of$0.075 per kWh. Finally, ECOs that save electrical energyduring the off-peak shift should be examined too, butthey may not be as advantageous; they would only saveat the rate of $0.043 per kWh because they are alreadyused off-peak and there would not be any additionaldemand cost savings.Physical and Operational Data for the FacilityThe auditor must gather information on factors likelyto affect energy use in the facility. Geographic location,weather data, facility layout and construction, operatinghours, and equipment can all influence energy use.• Geographic Location/Weather Data: The geographiclocation of the facility should be noted,together with the weather data for that location.Contact the local weather station, the local utilityor the state energy office to obtain the averagedegree days for heating and cooling for that locationfor the past twelve months. This degree-daydata will be very useful in analyzing the needfor energy for heating or cooling the facility. Binweather data would also be useful if a thermalenvelope simulation of the facility were going tobe performed as part of the audit.• Facility Layout: Next the facility layout or planshould be obtained, and reviewed to determinethe facility size, floor plan, and construction featuressuch as wall and roof material and insulationlevels, as well as door and window sizes and construction.A set of building plans could supply thisinformation in sufficient detail. It is important tomake sure the plans reflect the “as-built” featuresof the facility, since many original building plansdo not get updated after building alterations.• Operating Hours: Operating hours for the facilityshould also be obtained. Is there only a single shift?Are there two shifts? Three? Knowing the operatinghours in advance allows some determination asto whether some loads could be shifted to off-peaktimes. Adding a second shift can often be cost effectivefrom an energy cost view, since the demandcharge can then be spread over a greater amountof kWh.• Equipment List: Finally, the auditor should get anequipment list for the facility and review it beforeconducting the audit. All large pieces of energy-consumingequipment such as heaters, air conditioners,water heaters, and specific process-related equipmentshould be identified. This list, together withdata on operational uses of the equipment allows agood understanding of the major energy-consumingtasks or equipment at the facility. As a general rule,the largest energy and cost activities should be examinedfirst to see what savings could be achieved.The greatest effort should be devoted to the ECOswhich show the greatest savings, and the least effortto those with the smallest savings potential.The equipment found at an audit location willdepend greatly on the type of facility involved. Residentialaudits for single-family dwellings generallyinvolve smaller-sized lighting, heating, air conditioningand refrigeration systems. Commercial operations suchas grocery stores, office buildings and shopping centersusually have equipment similar to residences, butmuch larger in size and in energy use. However, largeresidential structures such as apartment buildings haveheating, air conditioning and lighting that is very similarto many commercial facilities. Business operationsis the area where commercial audits begin to involveequipment substantially different from that found inresidences.Industrial auditors encounter the most complexequipment. Commercial-scale lighting, heating, air conditioningand refrigeration, as well as office businessequipment, is generally used at most industrial facilities.The major difference is in the highly specialized equip-

ENERGY AUDITING 29ment used for the industrial production processes.This can include equipment for chemical mixing andblending, metal plating and treatment, welding, plasticinjection molding, paper making and printing, metalrefining, electronic assembly, and making glass, for example.3.3.3 Safety ConsiderationsSafety is a critical part of any energy audit. Theaudit person or team should be thoroughly briefed onsafety equipment and procedures, and should neverplace themselves in a position where they could injurethemselves or other people at the facility. Adequatesafety equipment should be worn at all appropriatetimes. Auditors should be extremely careful makingany measurements on electrical systems, or on hightemperature devices such as boilers, heaters, cookers,etc. Electrical gloves or asbestos gloves should be wornas appropriate.The auditor should be careful when examining anyoperating piece of equipment, especially those with opendrive shafts, belts or gears, or any form of rotating machinery.The equipment operator or supervisor shouldbe notified that the auditor is going to look at that pieceof equipment and might need to get information fromsome part of the device. If necessary, the auditor mayneed to come back when the machine or device is idlein order to safely get the data. The auditor should neverapproach a piece of equipment and inspect it withoutthe operator or supervisor being notified first.Safety Checklist1. Electrical:a. Avoid working on live circuits, if possible.b. Securely lock off circuits and switches beforeworking on a piece of equipment.c. Always keep one hand in your pocket whilemaking measurements on live circuits to helpprevent cardiac arrest.2. Respiratory:a. When necessary, wear a full face respiratormask with adequate filtration particle size.b. Use activated carbon cartridges in the maskwhen working around low concentrations ofnoxious gases. Change the cartridges on aregular basis.c. Use a self-contained breathing apparatus forwork in toxic environments.3. Hearing:a. Use foam insert plugs while working aroundloud machinery to reduce sound levels up to30 decibels.3.3.4 Conducting the Audit VisitOnce the information on energy bills, facilityequipment and facility operation has been obtained, theaudit equipment can be gathered up, and the actual visitto the facility can be made.Introductory MeetingThe audit person—or team—should meet with thefacility manager and the maintenance supervisor andbriefly discuss the purpose of the audit and indicate thekind of information that is to be obtained during the visitto the facility. If possible, a facility employee who is ina position to authorize expenditures or make operatingpolicy decisions should also be at this initial meeting.Audit InterviewsGetting the correct information on facility equipmentand operation is important if the audit is going tobe most successful in identifying ways to save money onenergy bills. The company philosophy towards investments,the impetus behind requesting the audit, and theexpectations from the audit can be determined by interviewingthe general manager, chief operating officer, orother executives. The facility manager or plant manageris one person that should have access to much of theoperational data on the facility, and a file of data onfacility equipment. The finance officer can provide anynecessary financial records (e.g., utility bills for electric,gas, oil, other fuels, water and wastewater, expendituresfor maintenance and repair, etc.).The auditor must also interview the floor supervisorsand equipment operators to understand the buildingand process problems. Line or area supervisors usuallyhave the best information on times their equipmentis used. The maintenance supervisor is often the primaryperson to talk to about types of lighting and lamps, sizesof motors, sizes of air conditioners and space heaters,and electrical loads of specialized process equipment.Finally, the maintenance staff must be interviewed tofind the equipment and performance problems.The auditor should write down these people’snames, job functions and telephone numbers, since it isfrequently necessary to get additional information afterthe initial audit visit.Walk-through TourA walk-through tour of the facility or plant tourshould be conducted by the facility/plant manager, andshould be arranged so the auditor or audit team cansee the major operational and equipment features of thefacility. The main purpose of the walk-through tour isto obtain general information. More specific information

30 ENERGY MANAGEMENT HANDBOOKshould be obtained from the maintenance and operationalpeople after the tour.Getting Detailed DataFollowing the facility or plant tour, the auditor oraudit team should acquire the detailed data on facilityequipment and operation that will lead to identifyingthe significant Energy Conservation Opportunities(ECOs) that may be appropriate for this facility. Thisincludes data on lighting, HVAC equipment, motors,water heating, and specialized equipment such as refrigerators,ovens, mixers, boilers, heaters, etc. This data ismost easily recorded on individualized data sheets thathave been prepared in advance.What to Look for• Lighting: Making a detailed inventory of all lightingis important. Data should be recorded on numbersof each type of light fixtures and lamps, wattagesof lamps, and hours of operation of groupsof lights. A lighting inventory data sheet shouldbe used to record this data. Using a lightmeter,the auditor should also record light intensity readingsfor each area. Taking notes on types of tasksperformed in each area will help the auditor selectalternative lighting technologies that might bemore energy efficient. Other items to note are theareas that may be infrequently used and may becandidates for occupancy sensor controls of lighting,or areas where daylighting may be feasible.• HVAC Equipment: All heating, air conditioningand ventilating equipment should be inventoried.Prepared data sheets can be used to record type,size, model numbers, age, electrical specificationsor fuel use specifications, and estimated hours ofoperation. The equipment should be inspected todetermine the condition of the evaporator andcondenser coils, the air filters, and the insulationon the refrigerant lines. Air velocity measurementmay also be made and recorded to assess operatingefficiencies or to discover conditioned air leaks.This data will allow later analysis to examinealternative equipment and operations that wouldreduce energy costs for heating, ventilating, andair conditioning.• Electric Motors: An inventory of all electric motorsover 1 horsepower should also be taken. Prepareddata sheets can be used to record motor size, use,age, model number, estimated hours of operation,other electrical characteristics, and possibly theoperating power factor. Measurement of voltages,currents, and power factors may be appropriatefor some motors. Notes should be taken on theuse of motors, particularly recording those thatare infrequently used and might be candidates forpeak load control or shifting use to off-peak times.All motors over 1 hp and with times of use of 2000hours per year or greater, are likely candidates forreplacement by high efficiency motors—at leastwhen they fail and must be replaced.• Water Heaters: All water heaters should be examined,and data recorded on their type, size, age,model number, electrical characteristics or fuel use.What the hot water is used for, how much is used,and what time it is used should all be noted. Temperatureof the hot water should be measured.• Waste Heat Sources: Most facilities have manysources of waste heat, providing possible opportunitiesfor waste heat recovery to be used as thesubstantial or total source of needed hot water.Waste heat sources are air conditioners, air compressors,heaters and boilers, process cooling systems,ovens, furnaces, cookers, and many others.Temperature measurements for these waste heatsources are necessary to analyze them for replacingthe operation of the existing water heaters.• Peak Equipment Loads: The auditor should particularlylook for any piece of electrically poweredequipment that is used infrequently or whose usecould be controlled and shifted to off-peak times.Examples of infrequently used equipment includetrash compactors, fire sprinkler system pumps(testing), certain types of welders, drying ovens,or any type of back-up machine. Some productionmachines might be able to be scheduled for offpeak.Water heating could be done off-peak if astorage system is available, and off-peak thermalstorage can be accomplished for on-peak heatingor cooling of buildings. Electrical measurementsof voltages, currents, and wattages may be helpful.Any information which leads to a piece ofequipment being used off-peak is valuable, andcould result in substantial savings on electric bills.The auditor should be especially alert for thoseinfrequent on-peak uses that might help explainanomalies on the energy demand bills.• Other Energy-Consuming Equipment: Finally, aninventory of all other equipment that consumesa substantial amount of energy should be taken.Commercial facilities may have extensive computer

ENERGY AUDITING 31and copying equipment, refrigeration and coolingequipment, cooking devices, printing equipment,water heaters, etc. Industrial facilities will havemany highly specialized process and productionoperations and machines. Data on types, sizes,capacities, fuel use, electrical characteristics, age,and operating hours should be recorded for all ofthis equipment.Preliminary Identification of ECOsAs the audit is being conducted, the auditorshould take notes on potential ECOs that are evident.Identifying ECOs requires a good knowledge of theavailable energy efficiency technologies that can accomplishthe same job with less energy and less cost.For example, overlighting indicates a potential lampremoval or lamp change ECO, and inefficient lampsindicates a potential lamp technology change. Motorswith high use times are potential ECOs for high efficiencyreplacements. Notes on waste heat sourcesshould indicate what other heating sources they mightreplace, and how far away they are from the enduse point. Identifying any potential ECOs during thewalk-through will make it easier later on to analyzethe data and to determine the final ECO recommendations.3.3.5 Post-Audit AnalysisFollowing the audit visit to the facility, the datacollected should be examined, organized and reviewedfor completeness. Any missing data items should beobtained from the facility personnel or from a re-visitto the facility. The preliminary ECOs identified duringthe audit visit should now be reviewed, and the actualanalysis of the equipment or operational change shouldbe conducted. This involves determining the costs andthe benefits of the potential ECO, and making a judgmenton the cost-effectiveness of that potential ECO.Cost-effectiveness involves a judgment decisionthat is viewed differently by different people anddifferent companies. Often, Simple Payback Period(SPP) is used to measure cost-effectiveness, and mostfacilities want a SPP of two years or less. The SPP foran ECO is found by taking the initial cost and dividingit by the annual savings. This results in findinga period of time for the savings to repay the initialinvestment, without using the time value of money.One other common measure of cost-effectiveness is thediscounted benefit-cost ratio. In this method, the annualsavings are discounted when they occur in futureyears, and are added together to find the present valueof the annual savings over a specified period of time.The benefit-cost ratio is then calculated by dividing thepresent value of the savings by the initial cost. A ratiogreater than one means that the investment will morethan repay itself, even when the discounted future savingsare taken into account.Several ECO examples are given here in orderto illustrate the relationship between the audit informationobtained and the technology and operationalchanges recommended to save on energy bills.Lighting ECOFirst, an ECO technology is selected—such asreplacing an existing 400 watt mercury vapor lampwith a 325 watt multi-vapor lamp when it burns out.The cost of the replacement lamp must be determined.Product catalogs can be used to get typical prices forthe new lamp—about $10 more than the 400 watt mercuryvapor lamp. The new lamp is a direct screw-inreplacement, and no change is needed in the fixtureor ballast. Labor cost is assumed to be the same to installeither lamp. The benefits—or cost savings—mustbe calculated next. The power savings is 400-325 = 75watts. If the lamp operates for 4000 hours per yearand electric energy costs $0.075/kWh, then the savingsis (.075 kW)(4000 hr/year)($0.075/kWh) = $22.50/year.This gives a SPP = $10/$22.50/yr =.4 years, or about 5months. This would be considered an extremely costeffectiveECO. (For illustration purposes, ballast wattagehas been ignored.)Motor ECOA ventilating fan at a fiberglass boat manufacturingcompany has a standard efficiency 5 hp motorthat runs at full load two shifts a day, or 4160 hoursper year. When this motor wears out, the companywill have an ECO of using a high efficiency motor.A high efficiency 5 hp motor costs around $80 moreto purchase than the standard efficiency motor. Thestandard motor is 83% efficient and the high efficiencymodel is 88.5% efficient. The cost savings is found bycalculating (5 hp)(4160 hr/yr)(.746 kW/hp)[(1/.83) –(1/.885)]($.075/kWh) = (1162 kWh)*($0.075) = $87.15/year. The SPP = $80/$87.15/yr =.9 years, or about 11months. This is also a very attractive ECO when evaluatedby this economic measure.The discounted benefit-cost ratio can be foundonce a motor life is determined, and a discount rate isselected. Companies generally have a corporate standardfor the discount rate used in determining theirmeasures used to make investment decisions. For a 10year assumed life, and a 10% discount rate, the presentworth factor is found as 6.144 (see Chapter 4, Ap-

32 ENERGY MANAGEMENT HANDBOOKpendix 4A). The benefit-cost ratio is found as B/C =($87.15)(6.144)/$80 = 6.7. This is an extremely attractivebenefit-cost ratio.Peak Load Control ECOA metals fabrication plant has a large shot-blastcleaner that is used to remove the rust from heavysteel blocks before they are machined and welded. Thecleaner shoots out a stream of small metal balls—likeshotgun pellets—to clean the metal blocks. A 150hp motor provides the primary motive force for thiscleaner. If turned on during the first shift, this machinerequires a total electrical load of about 180 kW whichadds directly to the peak load billed by the electricutility. At $7.02/kW/month, this costs (180 kW)*($7.02/kW/month) = $1263.60/month. Discussions with lineoperating people resulted in the information that theneed for the metal blocks was known well in advance,and that the cleaning could easily be done on the eveningshift before the blocks were needed. Based on thisinformation, the recommended ECO is to restrict theshot-blast cleaner use to the evening shift, saving thecompany $15,163.20 per year. Since there is no cost toimplement this ECO, the SPP = O; that is, the paybackis immediate.3.3.6 The Energy Audit ReportThe next step in the energy audit process is toprepare a report which details the final results andrecommendations. The length and detail of this reportwill vary depending on the type of facility audited. Aresidential audit may result in a computer printout fromthe utility. An industrial audit is more likely to have adetailed explanation of the ECOs and benefit-cost analyses.The following discussion covers the more detailedaudit reports.The report should begin with an executive summarythat provides the owners/managers of the auditedfacility with a brief synopsis of the total savings availableand the highlights of each ECO. The report shouldthen describe the facility that has been audited, andprovide information on the operation of the facility thatrelates to its energy costs. The energy bills should bepresented, with tables and plots showing the costs andconsumption. Following the energy cost analysis, therecommended ECOs should be presented, along withthe calculations for the costs and benefits, and the costeffectivenesscriterion.Regardless of the audience for the audit report, itshould be written in a clear, concise and easy-to understandformat and style. The executive summary shouldbe tailored to non-technical personnel, and technicaljargon should be minimized. A client who understandsthe report is more likely to implement the recommendedECOs. An outline for a complete energy audit report isshown below.Energy Audit Report FormatExecutive SummaryA brief summary of the recommendations and costsavingsTable of ContentsIntroductionPurpose of the energy auditNeed for a continuing energy cost control programFacility DescriptionProduct or service, and materials flowSize, construction, facility layout, and hours ofoperationEquipment list, with specificationsEnergy Bill AnalysisUtility rate structuresTables and graphs of energy consumptions andcostsDiscussion of energy costs and energy billsEnergy Conservation OpportunitiesListing of potential ECOsCost and savings analysisEconomic evaluationAction PlanRecommended ECOs and an implementationscheduleDesignation of an energy monitor and ongoingprogramConclusionAdditional comments not otherwise covered3.3.7 The Energy Action PlanThe last step in the energy audit process is torecommend an action plan for the facility. Some companieswill have an energy audit conducted by theirelectric utility or by an independent consulting firm,and will then make changes to reduce their energybills. They may not spend any further effort in the energycost control area until several years in the futurewhen another energy audit is conducted. In contrast tothis is the company which establishes a permanent energycost control program, and assigns one person—ora team of people—to continually monitor and improvethe energy efficiency and energy productivity of thecompany. Similar to a Total Quality Management programwhere a company seeks to continually improve

ENERGY AUDITING 33the quality of its products, services and operation, anenergy cost control program seeks continual improvementin the amount of product produced for a givenexpenditure for energy.The energy action plan lists the ECOs whichshould be implemented first, and suggests an overallimplementation schedule. Often, one or more of therecommended ECOs provides an immediate or veryshort payback period, so savings from that ECO—orthose ECOs can be used to generate capital to pay forimplementing the other ECOs. In addition, the actionplan also suggests that a company designate one personas the energy monitor for the facility. This person canlook at the monthly energy bills and see whether anyunusual costs are occurring, and can verify that the energysavings from ECOs is really being seen. Finally, thisperson can continue to look for other ways the companycan save on energy costs, and can be seen as evidencethat the company is interested in a future program ofenergy cost control.3.4 SPECIALIZED AUDIT TOOLS3.4.1 Smoke SourcesSmoke is useful in determining airflow characteristicsin buildings, air distribution systems, exhaust hoodsand systems, cooling towers, and air intakes. Thereare several ways to produce smoke. Ideally, the smokeshould be neutrally buoyant with the air mass aroundit so that no motion will be detected unless a force isapplied. Cigarette and incense stick smoke, althoughinexpensive, do not meet this requirement.Smoke generators using titanium tetrachloride(TiCl 4 ) provide an inexpensive and convenient way toproduce and apply smoke. The smoke is a combinationof hydrochloric acid (HCl) fumes and titanium oxidesproduced by the reaction of TiCl 4 and atmospheric watervapor. This smoke is both corrosive and toxic so theuse of a respirator mask utilizing activated carbon isstrongly recommended. Commercial units typically useeither glass or plastic cases. Glass has excellent longevitybut is subject to breakage since smoke generatorsare often used in difficult-to-reach areas. Most types ofplastic containers will quickly degrade from the actionof hydrochloric acid.Small Teflon ® squeeze bottles (i.e., 30 ml) with attachedcaps designed for laboratory reagent use resistdegradation and are easy to use. The bottle should bestuffed with 2-3 real cotton balls then filled with about0.15 fluid ounces of liquid TiCl 4 . Synthetic cotton ballstypically disintegrate if used with titanium tetrachloride.This bottle should yield over a year of service withregular use. The neck will clog with debris but can becleaned with a paper clip.Some smoke generators are designed for short timeuse. These bottles are inexpensive and useful for a dayof smoke generation, but will quickly degrade. Smokebombs are incendiary devices designed to emit a largevolume of smoke over a short period of time. The smokeis available in various colors to provide good visibility.These are useful in determining airflow capabilities ofexhaust air systems and large-scale ventilation systems.A crude smoke bomb can be constructed by placing astick of elemental phosphorus in a metal pan and ignitingit. A large volume of white smoke will be released.This is an inexpensive way of testing laboratory exhausthoods since many labs have phosphorus in stock.More accurate results can be obtained by measuringthe chemical composition of the airstream after injectinga known quantity of tracer gas such as sulphurhexafluoride into an area. The efficiency of an exhaustsystem can be determined by measuring the rate oftracer gas removal. Building infiltration/exfiltrationrates can also be estimated with tracer gas.3.4.2 Blower DoorThe blower door is a device containing a fan, controller,several pressure gauges, and a frame which fitsin the doorway of a building. It is used to study thepressurization and leakage rates of a building and its airdistribution system under varying pressure conditions.The units currently available are designed for use in residencesalthough they can be used in small commercialbuildings as well. The large quantities of ventilation airlimit blower door use in large commercial and industrialbuildings.An air leakage/pressure curve can be developedfor the building by measuring the fan flow rate necessaryto achieve a pressure differential between the buildinginterior and the ambient atmospheric pressure overa range of values. The natural air infiltration rate of thebuilding under the prevailing pressure conditions canbe estimated from the leakage/pressure curve and localair pressure data. Measurements made before and aftersealing identified leaks can indicate the effectiveness ofthe work.The blower door can help to locate the source of airleaks in the building by depressurizing to 30 Pascals andsearching potential leakage areas with a smoke source.The air distribution system typically leaks on both thesupply and return air sides. If the duct system is locatedoutside the conditioned space (e.g., attic, underfloor, etc.), supply leaks will depressurize the building

34 ENERGY MANAGEMENT HANDBOOKand increase the air infiltration rate; return air leakswill pressurize the building, causing air to exfiltrate. Acombination of supply and return air leaks is difficultto detect without sealing off the duct system at theregisters and measuring the leakage rate of the buildingcompared to that of the unsealed duct system. Thedifference between the two conditions is a measure ofthe leakage attributable to the air distribution system.3.4.3 Airflow Measurement DevicesTwo types of anemometers are available for measuringairflow: vane and hot-wire. The volume of airmoving through an orifice can be determined by estimatingthe free area of the opening (e.g., supply airregister, exhaust hood face, etc.) and multiplying by theair speed. This result is approximate due to the difficultyin determining the average air speed and the free ventarea. Regular calibrations are necessary to assure theaccuracy of the instrument. The anemometer can alsobe used to optimize the face velocity of exhaust hoodsby adjusting the door opening until the anemometerindicates the desired airspeed.Airflow hoods also measure airflow. They containan airspeed integrating manifold which averages thevelocity across the opening and reads out the airflowvolume. The hoods are typically made of nylon fabricsupported by an aluminum frame. The instrument islightweight and easy to hold up against an air vent. Thelip of the hood must fit snugly around the opening toassure that all the air volume is measured. Both supplyand exhaust airflow can be measured. The result must beadjusted if test conditions fall outside the design range.3.5 INDUSTRIAL AUDITS3.5.1 IntroductionIndustrial audits are some of the most complexand most interesting audits because of the tremendousvariety of equipment found in these facilities. Muchof the industrial equipment can be found during commercialaudits too. Large chillers, boilers, ventilatingfans, water heaters, coolers and freezers, and extensivelighting systems are often the same in most industrialoperations as those found in large office buildings orshopping centers. Small cogeneration systems are oftenfound in both commercial and industrial facilities.The highly specialized equipment that is used inindustrial processes is what differentiates these facilitiesfrom large commercial operations. The challenge for theauditor and energy management specialist is to learnhow this complex—and often unique—industrial equipmentoperates, and to come up with improvements tothe processes and the equipment that can save energyand money. The sheer scope of the problem is so greatthat industrial firms often hire specialized consultingengineers to examine their processes and recommendoperational and equipment changes that result in greaterenergy productivity.3.5.2 Audit ServicesA few electric and gas utilities are large enough,and well-enough staffed, that they can offer industrialaudits to their customers. These utilities have a trainedstaff of engineers and process specialists with extensiveexperience who can recommend operational changes ornew equipment to reduce the energy costs in a particularproduction environment. Many gas and electric utilities,even if they do not offer audits, do offer financialincentives for facilities to install high efficiency lighting,motors, chillers, and other equipment. These incentivescan make many ECOs very attractive.Small and medium-sized industries that fall intothe Manufacturing Sector—SIC 2000 to 3999, and arein the service area of one of the Industrial AssessmentCenters funded by the U.S. Department of Energy, canreceive free energy audits throughout this program.There are presently 30 IACs operating primarily in theeastern and mid-western areas of the U.S. These IACsare administered by the U.S. Department of Energy. Asearch vehicle using "Industrial Assessment Centers"will yield updated locations.3.5.3 Industrial Energy Rate StructuresExcept for the smallest industries, facilities will bebilled for energy services through a large commercial orindustrial rate category. It is important to get this ratestructure information for all sources of energy—electricity,gas, oil, coal, steam, etc. Gas, oil and coal are usuallybilled on a straight cost per unit basis—e.g. $0.90per gallon of #2 fuel oil. Electricity and steam mostoften have complex rate structures with componentsfor a fixed customer charge, a demand charge, and anenergy charge. Gas, steam, and electric energy are oftenavailable with a time of day rate, or an interruptiblerate that provides much cheaper energy service withthe understanding that the customer may have his supplyinterrupted (stopped) for periods of several hoursat a time. Advance notice of the interruption is almostalways given, and the number of times a customer canbe interrupted in a given period of time is limited.3.5.4 Process and Technology Data SourcesFor the industrial audit, it is critical to get in ad-

ENERGY AUDITING 35vance as much information as possible on the specializedprocess equipment so that study and research can beperformed to understand the particular processes beingused, and what improvements in operation or technologyare available. Data sources are extremely valuablehere; auditors should maintain a library of informationon processes and technology and should know where tofind additional information from research organizations,government facilities, equipment suppliers and otherorganizations.EPRI/GRIThe Electric Power Research Institute (EPRI) andthe Gas Research Institute (GRI) are both excellentsources of information on the latest technologies of usingelectric energy or gas. EPRI has a large number ofon-going projects to show the cost-effectiveness of electro-technologiesusing new processes for heating, drying,cooling, etc. GRI also has a large number of projectsunderway to help promote the use of new cost-effectivegas technologies for heating, drying, cooling, etc. Bothof these organizations provide extensive documentationof their processes and technologies; they also have computerdata bases to aid customer inquiries.U.S. DOE Industrial DivisionThe U.S. Department of Energy has an IndustrialDivision that provides a rich source of information onnew technologies and new processes. This divisionfunds research into new processes and technologies,and also funds many demonstration projects to helpinsure that promising improvements get implementedin appropriate industries. The Industrial Division ofUSDOE also maintains a wide network of contacts withgovernment-related research laboratories such as OakRidge National Laboratory, Brookhaven National Laboratory,Lawrence Berkeley National Laboratory, SandiaNational Laboratory, and Battelle National Laboratory.These laboratories have many of their own research, developmentand demonstration programs for improvedindustrial and commercial technologies.State Energy OfficesState energy offices are also good sources of information,as well as good contacts to see what kindof incentive programs might be available in the state.Many states offer programs of free boiler tune-ups, freeair conditioning system checks, seminars on energy efficiencyfor various facilities, and other services. Moststate energy offices have well-stocked energy libraries,and are also tied into other state energy research organizations,and to national laboratories and the USDOE.Equipment SuppliersEquipment suppliers provide additional sourcesfor data on energy efficiency improvements to processes.Marketing new, cost-effective processes and technologiesprovides sales for the companies as well as helpingindustries to be more productive and more economicallycompetitive. The energy auditor should compare theinformation from all of the sources described above.3.5.5 Conducting the AuditSafety ConsiderationsSafety is the primary consideration in any industrialaudit. The possibility of injury from hot objects,hazardous materials, slippery surfaces, drive belts, andelectric shocks is far greater than when conductingresidential and commercial audits. Safety glasses, safetyshoes, durable clothing and possibly a safety hat andbreathing mask might be needed during some audits.Gloves should be worn while making any electrical measurements,and also while making any measurementsaround boilers, heaters, furnaces, steam lines, or othervery hot pieces of equipment. In all cases, adequate attentionto personal safety is a significant feature of anyindustrial audit.LightingLighting is not as great a percent of total industrialuse as it is in the commercial sector on the average, butlighting is still a big energy use and cost area for manyindustrial facilities. A complete inventory of all lightingshould be taken during the audit visit. Hours of operationof lights are also necessary, since lights are commonly lefton when they are not needed. Timers, Energy ManagementSystems, and occupancy sensors are all valuableapproaches to insuring that lights that are not needed arenot turned on. It is also important to look at the facility’soutside lighting for parking and for storage areas.During the lighting inventory, types of tasks beingperformed should also be noted, since light replacementwith more efficient lamps often involves changing thecolor of the resultant light. For example, high pressuresodium lamps are much more efficient than mercury vaporlamps or even metal halide lamps, but they produce ayellowish light that makes fine color distinction difficult.However, many assembly tasks can still be performedadequately under high pressure sodium lighting. Thesetypically include metal fabrication, wood product fabrication,plastic extrusion, and many others.Electric MotorsA common characteristic of many industries istheir extensive use of electric motors. A complete inven-

36 ENERGY MANAGEMENT HANDBOOKtory of all motors over 1 hp should be taken, as well asrecording data on how long each motor operates duringa day. For motors with substantial usage times, replacementwith high-efficiency models is almost always costeffective. In addition, consideration should be given toreplacement of standard drive belts with synchronousbelts which transmit the motor energy more efficiently.For motors which are used infrequently, it may be possibleto shift the use to off-peak times, and to achievea kW demand reduction which would reduce energycost.HVAC SystemsAn inventory of all space heaters and air conditionersshould be taken. Btu per hour ratings and efficienciesof all units should be recorded, as well as usage patterns.Although many industries do not heat or air conditionthe production floor area, they almost always have officeareas, cafeterias, and other areas that are normallyheated and air conditioned. For these conditioned areas,the construction of the facility should be noted—howmuch insulation, what are the walls and ceilings madeof, how high are the ceilings. Adding additional insulationmight be a cost effective ECO.Production floors that are not air conditioned oftenhave large numbers of ventilating fans that operate anywherefrom one shift per day to 24 hours a day. Plantswith high heat loads and plants in the mild climate areasoften leave these ventilating fans running all year long.These are good candidates for high efficiency motorreplacements. Timers or an Energy Management Systemmight be used to turn off these ventilating fans whenthe plant is shut down.BoilersAll boilers should be checked for efficient operationusing a stack gas combustion analyzer. Boilerspecifications on Btu per hour ratings, pressures andtemperatures should be recorded. The boiler should bevaried between low-fire, normal-fire, and high-fire, withcombustion gas and temperature readings taken at eachlevel. Boiler tune-up is one of the most common, andmost energy-saving operations available to many facilities.The auditor should check to see whether any wasteheat from the boiler is being recovered for use in a heatrecuperator or for some other use such as water heating.If not, this should be noted as a potential ECO.Specialized EquipmentMost of the remaining equipment encounteredduring the industrial audit will be the highly specializedprocess production equipment and machines. Thisequipment should all be examined and operationaldata taken, as well as noting hours and periods of use.All heat sources should be considered carefully as towhether they could be replaced with sources usingwaste heat, or whether a particular heat source couldserve as a provider of waste heat to another application.Operations where both heating and cooling occur periodically—suchas a plastic extrusion machine—are goodcandidates for reclaiming waste heat, or in sharing heatfrom a machine needing cooling with another machineneeding heat.Air CompressorsAir compressors should be examined for size, operatingpressures, and type (reciprocating or screw), andwhether they use outside cool air for intake. Large aircompressors are typically operated at night when muchsmaller units are sufficient. Also, screw-type air compressorsuse a large fraction of their rated power whenthey are idling, so control valves should be installedto prevent this loss. Efficiency is improved with intakeair that is cool, so outside air should be used in mostcases—except in extremely cold temperature areas.The auditor should determine whether there aresignificant air leaks in air hoses, fittings, and in machines.Air leaks are a major source of energy loss inmany facilities, and should be corrected by maintenanceaction. Finally, air compressors are a good source ofwaste heat. Nearly 90% of the energy used by an aircompressor shows up as waste heat, so this is a largesource of low temperature waste heat for heating inputair to a heater or boiler, or for heating hot water forprocess use.3.6 COMMERCIAL AUDITS3.6.1 IntroductionCommercial audits span the range from very simpleaudits for small offices to very complex audits formulti-story office buildings or large shopping centers.Complex commercial audits are performed in substantiallythe same manner as industrial audits. The followingdiscussion highlights those areas where commercialaudits are likely to differ from industrial audits.Commercial audits generally involve substantialconsideration of the structural envelope features of thefacility, as well as significant amounts of large or specializedequipment at the facility. Office buildings, shoppingcenters and malls all have complex building envelopesthat should be examined and evaluated. Building materials,insulation levels, door and window construction,

ENERGY AUDITING 37skylights, and many other envelope features must beconsidered in order to identify candidate ECOs.Commercial facilities also have large capacityequipment, such as chillers, space heaters, water heaters,refrigerators, heaters, cookers, and office equipmentsuch as computers and copy machines. Small cogenerationsystems are also commonly found in commercialfacilities and institutions such as schools and hospitals.Much of the equipment in commercial facilities is thesame type and size as that found in manufacturing orindustrial facilities. Potential ECOs would look at moreefficient equipment, use of waste heat, or operationalchanges to use less expensive energy.3.6.2 Commercial Audit ServicesElectric and gas utilities, as well as many engineeringconsulting firms, perform audits for commercialfacilities. Some utilities offer free walk-through auditsfor commercial customers, and also offer financial incentivesfor customers who change to more energy efficientequipment. Schools, hospitals and some other governmentinstitutions can qualify for free audits under theICP program described in the first part of this chapter.Whoever conducts the commercial audit must initiatethe ICP process by collecting information on the rateenergy rate structures, the equipment in use at the facility,and the operational procedures used there.3.6.3 Commercial Energy Rate StructuresSmall commercial customers are usually billed forenergy on a per energy unit basis, while large commercialcustomers are billed under complex rate structurescontaining components related to energy, rate of energyuse (power), time of day or season of year, power factor,and numerous other elements. One of the first steps ina commercial audit is to obtain the rate structures forall sources of energy, and to analyze at least one to twoyear’s worth of energy bills. This information should beput into a table and also plotted.3.6.4 Conducting the AuditA significant difference in industrial and commercialaudits arises in the area of lighting. Lightingin commercial facilities is one of the largest energycosts—sometimes accounting for half or more of the entireelectric bill. Lighting levels and lighting quality areextremely important to many commercial operations.Retail sales operations, in particular, want light levelsthat are far in excess of standard office values. Qualityof light in terms of color is also a big concern in retailsales, so finding acceptable ECOs for reducing lightingcosts is much more difficult for retail facilities than foroffice buildings. The challenge is to find new lightingtechnologies that allow high light levels and warm colorwhile reducing the wattage required. New T8 and T10fluorescent lamps, and metal halide lamp replacementsfor mercury vapor lamps offer these features, and usuallyrepresent cost-effective ECOs for retail sales andother facilities.3.7 RESIDENTIAL AUDITSAudits for large, multi-story apartment buildingscan be very similar to commercial audits. (See section3.6.) Audits of single-family residences, however, aregenerally fairly simple. For single-family structures, theenergy audit focuses on the thermal envelope and theappliances such as the heater, air conditioner, waterheater, and “plug loads.”The residential auditor should start by obtainingpast energy bills and analyzing them to determine anypatterns or anomalies. During the audit visit, the structureis examined to determine the levels of insulation,the conditions of and seals for windows and doors, andthe integrity of the ducts. The space heater and/or airconditioner is inspected, along with the water heater.Equipment model numbers, age, size, and efficienciesare recorded. The post-audit analysis then evaluates potentialECOs such as adding insulation, adding doublepanewindows, window shading or insulated doors, andchanging to higher efficiency heaters, air conditioners,and water heaters. The auditor calculates costs, benefits,and Simple Payback Periods and presents them to theowner or occupant. A simple audit report—often in theform of a computer printout is given to the owner oroccupant.3.8 INDOOR AIR QUALITY3.8.1 IntroductionImplementation of new energy-related standardsand practices has contributed to a degradation of indoorair quality. In fact, the quality of indoor air has beenfound to exceed the Environmental Protection Agency(EPA) standards for outdoor air in many homes, businesses,and factories. Thus, testing for air quality problemsis done in some energy audits both to prevent exacerbatingany existing problems and to recommend ECOsthat might improve air quality. Air quality standards forthe industrial environment have been published by theAmerican Council of Governmental Industrial Hygienists(ACGIH) in their booklet “Threshold Limit Values.”

38 ENERGY MANAGEMENT HANDBOOKNo such standards currently exist for the residential andcommercial environments although the ACGIH standardsare typically and perhaps inappropriately used.The EPA has been working to develop residential andcommercial standards for quite some time.3.8.2 Symptoms of Air Quality ProblemsSymptoms of poor indoor air quality include, butare not limited to: headaches; irritation of mucous membranessuch as the nose, mouth, throat, lungs; tearing,redness and irritation of the eyes; numbness of the lips,mouth, throat; mood swings; fatigue; allergies; coughing;nasal and throat discharge; and irritability. Chronicexposure to some compounds can lead to damage tointernal organs such as the liver, kidney, lungs, andbrain; cancer; and death.3.8.3 TestingTesting is required to determine if the air qualityis acceptable. Many dangerous compounds, like carbonmonoxide and methane without odorant added, areodorless and colorless. Some dangerous particulatessuch as asbestos fibers do not give any indication of aproblem for up to twenty years after inhalation. Testingmust be conducted in conjunction with pollution-producingprocesses to ensure capture of the contaminants.Testing is usually performed by a Certified IndustrialHygienist (CIH).3.8.4 Types of PollutantsAirstreams have three types of contaminants: particulateslike dust and asbestos; gases like carbon monoxide,ozone, carbon dioxide, volatile organic compounds,anhydrous ammonia, radon, outgassing from urea-formaldehydeinsulation, low oxygen levels; and biologicalslike mold, mildew, fungus, bacteria, and viruses.3.8.5 Pollutant Control MeasuresParticulatesParticulates are controlled with adequate filtrationnear the source and in the air handling system. Mechanicalfilters are frequently used in return air streams, andbaghouses are used for particulate capture. The coarsefilters used in most residential air conditioners typicallyhave filtration efficiencies below twenty percent. Mechanicalfilters called high efficiency particulate apparatus(HEPA) are capable of filtering particles as small as 0.3microns at up to 99% efficiency. Electrostatic precipitatorsremove particulates by placing a positive charge on thewalls of collection plates and allowing negatively chargedparticulates to attach to the surface. Periodic cleaning ofthe plates is necessary to maintain high filtration efficiency.Loose or friable asbestos fibers should be removedfrom the building or permanently encapsulated to prevententry into the respirable airstream. While conductingan audit, it is important to determine exactly what typeof insulation is in use before disturbing an area to maketemperature measurements.Problem GasesProblem gases are typically removed by ventilatingwith outside air. Dilution with outside air is effective,but tempering the temperature and relative humidityof the outdoor air mass can be expensive in extremeconditions. Heat exchangers such as heat wheels, heatpipes, or other devices can accomplish this task with reducedenergy use. Many gases can be removed from theairstream by using absorbent/adsorbent media such asactivated carbon or zeolite. This strategy works well forspaces with limited ventilation or where contaminantsare present in low concentrations. The media must bechecked and periodically replaced to maintain effectiveness.Radon gas—Ra 222—cannot be effectively filtereddue to its short half life and the tendency for its Poloniumdaughters to plate out on surfaces. Low oxygenlevels are a sign of inadequate outside ventilation air.A high level of carbon dioxide (e.g., 1000-10,000 ppm)is not a problem in itself but levels above 1000 ppmindicate concentrated human or combustion activity ora lack of ventilation air. Carbon dioxide is useful as anindicator compound because it is easy and inexpensiveto measure.Microbiological ContaminantsMicrobiological contaminants generally requireparticular conditions of temperature and relative humidityon a suitable substrate to grow. Mold and mildew areinhibited by relative humidity levels less than 50%. Airdistribution systems often harbor colonies of microbialgrowth. Many people are allergic to microscopic dustmites. Cooling towers without properly adjusted automatedchemical feed systems are an excellent breedingground for all types of microbial growth.Ventilation RatesRecommended ventilation quantities are publishedby the American Society of Heating, Refrigerating, andAir-Conditioning Engineers (ASHRAE) in Standard 62,“Ventilation for Acceptable Air Quality.” These ventilationrates are for effective systems. Many existingsystems fail in entraining the air mass efficiently. Thedensity of the contaminants relative to air must be con-

ENERGY AUDITING 39sidered in locating the exhaust air intakes and ventilationsupply air registers.LiabilityLiability related to indoor air problems appearsto be a growing but uncertain issue because few caseshave made it through the court system. However, inretrospect, the asbestos and ureaformaldehyde pollutionproblems discovered in the last two decadessuggest proceeding with caution and a proactive approach.3.9 CONCLUSIONEnergy audits are an important first step in theoverall process of reducing energy costs for any building,company, or industry. A thorough audit identifiesand analyzes the changes in equipment and operationsthat will result in cost-effective energy cost reduction.The energy auditor plays a key role in the successfulconduct of an audit, and also in the implementation ofthe audit recommendations.BibliographyInstructions For Energy Auditors, Volumes I and II, U.S. Department ofEnergy, DOE/CS-0041/12&13, September, 1978. Available throughNational Technical Information Service, Springfield, VA.Energy Conservation Guide for Industry and Commerce, National Bureau ofStandards Handbook 115 and Supplement, 1978. Available throughU.S. Government Printing Office, Washington, DC.Guide to Energy Management, Third Edition, Capehart, B.L., Turner, W.C.,and Kennedy, W.J., The Fairmont Press, Atlanta, GA, 2000.Illuminating Engineering Society, IES Lighting Handbook, Ninth Edition,New York, NY, 2000.Total Energy Management, A Handbook prepared by the National ElectricalContractors Association and the National Electrical ManufacturersAssociation, Washington, DC.Handbook of Energy Audits, Thumann, Albert, Fifth Edition, The FairmontPress, Atlanta, GA.Industrial Energy Management and Utilization, Witte, Larry C., Schmidt,Philip S., and Brown, David R., Hemisphere Publishing Corporation,Washington, DC, 1988.Threshold Limit Values for Chemical Substances and Physical Agentsand Biological Exposure In dices, 1990-91 American Conference ofGovernmental Industrial Hygienists.Ventilation for Acceptable Indoor Air Quality, ASHRAE 62-1999, AmericanSociety of Heating, Refrigerating and Air-Conditioning Engineers,Inc., 1999.Facility Design and Planning Engineering Weather Data, Departmentsof the Air Force, the Army, and the Navy, 1978.Handbook of Energy Engineering, Fourth Edition, Thumann, A., andMehta, D.P., The Fairmont Press, Atlanta, GA.


CHAPTER 4ECONOMIC ANALYSISDR. DAVID PRATTIndustrial Engineering and ManagementOklahoma State UniversityStillwater, OK4.1 OBJECTIVEThe objective of this chapter is to present a coherent,consistent approach to economic analysis of capital investments(energy related or other). Adherence to theconcepts and methods presented will lead to sound investmentdecisions with respect to time value of moneyprinciples. The chapter opens with material designed tomotivate the importance of life cycle cost concepts in theeconomic analysis of projects. The next three sectionsprovide foundational material necessary to fully developtime value of money concepts and techniques. Thesesections present general characteristics of capital investments,sources of funds for capital investment, and abrief summary of tax considerations which are importantfor economic analysis. The next two sections introducetime value of money calculations and several approachesfor calculating project measures of worth basedon time value of money concepts. Next the measures ofworth are applied to the process of making decisionswhen a set of potential projects are to be evaluated. Thefinal concept and technique section of the chapter presentsmaterial to address several special problems thatmay be encountered in economic analysis. This materialincludes, among other things, discussions of inflation,non-annual compounding of interest, and sensitivityanalysis. The chapter closes with a brief summary and alist of references which can provide additional depth inmany of the areas covered in the chapter.4.2 INTRODUCTIONCapital investment decisions arise in many circumstances.The circumstances range from evaluatingbusiness opportunities to personal retirement planning.Regardless of circumstances, the basic criterion for evaluatingany investment decision is that the revenues (savings)generated by the investment must be greater than41the costs incurred. The number of years over which therevenues accumulate and the comparative importance offuture dollars (revenues or costs) relative to present dollarsare important factors in making sound investmentdecisions. This consideration of costs over the entire lifecycle of the investments gives rise to the name life cyclecost analysis which is commonly used to refer to the economicanalysis approach presented in this chapter. Anexample of the importance of life cycle costs is shown inFigure 4.1 which depicts the estimated costs of owningand operating an oil-fired furnace to heat a 2,000-squarefoothouse in the northeast United States. Of particularnote is that the initial costs represent only 23% of thetotal costs incurred over the life of the furnace. The lifecycle cost approach provides a significantly better evaluationof long term implications of an investment thanmethods which focus on first cost or near term results.Figure 4.1 15-Year life cycle costs of a heating systemLife cycle cost analysis methods can be applied tovirtually any public or private business sector investmentdecision as well as to personal financial planningdecisions. Energy related decisions provide excellentexamples for the application of this approach. Such decisionsinclude: evaluation of alternative building designswhich have different initial costs, operating and maintenancecosts, and perhaps different lives; evaluation ofinvestments to improve the thermal performance of anexisting building (wall or roof insulation, window glazing);or evaluation of alternative heating, ventilating, orair conditioning systems. For federal buildings, Congressand the President have mandated, through legislationand executive order, energy conservation goals thatmust be met using cost-effective measures. The life cyclecost approach is mandated as the means of evaluatingcost effectiveness.

42 ENERGY MANAGEMENT HANDBOOK4.3 GENERAL CHARACTERISTICSOF CAPITAL INVESTMENTS4.3.1 Capital Investment CharacteristicsWhen companies spend money, the outlay of cashcan be broadly categorized into one of two classifications;expenses or capital investments. Expenses aregenerally those cash expenditures that are routine, ongoing,and necessary for the ordinary operation of thebusiness. Capital investments, on the other hand, aregenerally more strategic and have long term effects. Decisionsmade regarding capital investments are usuallymade at higher levels within the organizational hierarchyand carry with them additional tax consequences ascompared to expenses.Three characteristics of capital investments are ofconcern when performing life cycle cost analysis. First,capital investments usually require a relatively large initialcost. “Relatively large” may mean several hundreddollars to a small company or many millions of dollarsto a large company. The initial cost may occur as a singleexpenditure such as purchasing a new heating system oroccur over a period of several years such as designingand constructing a new building. It is not uncommonthat the funds available for capital investments projectsare limited. In other words, the sum of the initial costsof all the viable and attractive projects exceeds the totalavailable funds. This creates a situation known as capitalrationing which imposes special requirements on the investmentanalysis. This topic will be discussed in Section4.8.3.The second important characteristic of a capitalinvestment is that the benefits (revenues or savings) resultingfrom the initial cost occur in the future, normallyover a period of years. The period between the initialcost and the last future cash flow is the life cycle or lifeof the investment. It is the fact that cash flows occurover the investment’s life that requires the introductionof time value of money concepts to properly evaluateinvestments. If multiple investments are being evaluatedand if the lives of the investments are not equal, specialconsideration must be given to the issue of selecting anappropriate planning horizon for the analysis. Planninghorizon issues are introduced in Section 4.8.5.The last important characteristic of capital investmentsis that they are relatively irreversible. Frequently,after the initial investment has been made, terminatingor significantly altering the nature of a capital investmenthas substantial (usually negative) cost consequences. Thisis one of the reasons that capital investment decisions areusually evaluated at higher levels of the organizationalhierarchy than operating expense decisions.4.3.2 Capital Investment Cost CategoriesIn almost every case, the costs which occur overthe life of a capital investment can be classified into oneof the following categories:• Initial Cost,• Annual Expenses and Revenues,• Periodic Replacement and Maintenance, or• Salvage Value.As a simplifying assumption, the cash flows whichoccur during a year are generally summed and regardedas a single end-of-year cash flow. While this approachdoes introduce some inaccuracy in the evaluation, itis generally not regarded as significant relative to thelevel of estimation associated with projecting future cashflows.Initial costs include all costs associated with preparingthe investment for service. This includes purchasecost as well as installation and preparation costs.Initial costs are usually nonrecurring during the life ofan investment. Annual expenses and revenues are therecurring costs and benefits generated throughout thelife of the investment. Periodic replacement and maintenancecosts are similar to annual expenses and revenuesexcept that they do not (or are not expected to) occur annually.The salvage (or residual) value of an investmentis the revenue (or expense) attributed to disposing of theinvestment at the end of its useful life.4.3.3 Cash Flow DiagramsA convenient way to display the revenues (savings)and costs associated with an investment is a cashflow diagram. By using a cash flow diagram, the timingof the cash flows are more apparent and the chancesof properly applying time value of money concepts areincreased. With practice, different cash flow patterns canbe recognized and they, in turn, may suggest the mostdirect approach for analysis.It is usually advantageous to determine the timeframe over which the cash flows occur first. This establishesthe horizontal scale of the cash flow diagram. Thisscale is divided into time periods which are frequently,but not always, years. Receipts and disbursements arethen located on the time scale in accordance with theproblem specifications. Individual outlays or receiptsare indicated by drawing vertical lines appropriatelyplaced along the time scale. The relative magnitudes canbe suggested by the heights, but exact scaling generallydoes not enhance the meaningfulness of the diagram.Upward directed lines indicate cash inflow (revenues orsavings) while downward directed lines indicate cash

ECONOMIC ANALYSIS 43outflow (costs).Figure 4.2 illustrates a cash flow diagram. Thecash flows depicted represent an economic evaluationof whether to choose a baseboard heating and windowair conditioning system or a heat pump for a ranger’shouse in a national park [Fuller and Petersen, 1994]. Thedifferential costs associated with the decision are:• The heat pump costs (cash outflow) $1500 morethan the baseboard system,• The heat pump saves (cash inflow) $380 annuallyin electricity costs,• The heat pump has a $50 higher annual maintenancecosts (cash outflow),• The heat pump has a $150 higher salvage value(cash inflow) at the end of 15 years,• The heat pump requires $200 more in replacementmaintenance (cash outflow) at the end of year 8.Although cash flow diagrams are simply graphicalrepresentations of income and outlay, they shouldexhibit as much information as possible. During theanalysis phase, it is useful to show the Minimum AttractiveRate of Return (an interest rate used to account forthe time value of money within the problem) on the cashflow diagram, although this has been omitted in Figure4.2. The requirements for a good cash flow diagram arecompleteness, accuracy, and legibility. The measure of asuccessful diagram is that someone else can understandthe problem fully from it.4.4 SOURCES OF FUNDSCapital investing requires a source of funds. Forlarge companies multiple sources may be employed.The process of obtaining funds for capital investment iscalled financing. There are two broad sources of financialfunding; debt financing and equity financing. Debt financinginvolves borrowing and utilizing money whichis to be repaid at a later point in time. Interest is paid tothe lending party for the privilege of using the money.Debt financing does not create an ownership position forthe lender within the borrowing organization. The borroweris simply obligated to repay the borrowed fundsplus accrued interest according to a repayment schedule.Car loans and mortgage loans are two examples ofthis type of financing. The two primary sources of debtcapital are loans and bonds. The cost of capital associatedwith debt financing is relatively easy to calculatesince interest rates and repayment schedules are usuallyclearly documented in the legal instruments controllingthe financing arrangements. An added benefit to debtfinancing under current U.S. tax law (as of April 2000)is that the interest payments made by corporations ondebt capital are tax deductible. This effectively lowersthe cost of debt financing since for debt financing withdeductible interest payments, the after-tax cost of capitalis given by:Cost of Capital AFTERTAX =Cost of Capital BEFORETAX * (1 - TaxRate)where the tax rate is determine by applicable tax law.The second broad source of funding is equity financing.Under equity financing the lender acquires anownership (or equity) position within the borrower’sorganization. As a result of this ownership position, thelender has the right to participate in the financial successof the organization as a whole. The two primary sourcesof equity financing are stocks and retained earnings. Thecost of capital associated with shares of stock is muchdebated within the financial community. A detailedpresentation of the issues and approaches is beyond theFigure 4.2. Heat pump and baseboard system differential life cycle costs

44 ENERGY MANAGEMENT HANDBOOKscope of this chapter. Additional reference material canbe found in Park and Sharp-Bette [1990]. One issue overwhich there is general agreement is that the cost of capitalfor stocks is higher than the cost of capital for debtfinancing. This is at least partially attributable to the factthat interest payments are tax deductible while stockdividend payments are not.If any subject is more widely debated in the financialcommunity than the cost of capital for stocks, it isthe cost of capital for retained earnings. Retained earningsare the accumulation of annual earnings surplusesthat a company retains within the company’s coffersrather than pays out to the stockholders as dividends.Although these earnings are held by the company, theytruly belong to the stockholders. In essence the companyis establishing the position that by retaining theearnings and investing them in capital projects, stockholderswill achieve at least as high a return throughfuture financial successes as they would have earnedif the earnings had been paid out as dividends. Hence,one common approach to valuing the cost of capital forretained earnings is to apply the same cost of capital asfor stock. This, therefore, leads to the same generallyagreed result. The cost of capital for financing throughretained earnings generally exceeds the cost of capitalfor debt financing.In many cases the financing for a set of capitalinvestments is obtained by packaging a combination ofthe above sources to achieve a desired level of availablefunds. When this approach is taken, the overall cost ofcapital is generally taken to be the weighted averagecost of capital across all sources. The cost of each individualsource’s funds is weighted by the source’s fractionof the total dollar amount available. By summingacross all sources, a weighted average cost of capital iscalculated.Example 1Determine the weighted average cost of capital forfinancing which is composed of:25% loans with a before tax cost of capital of 12%/yr and75% retained earnings with a cost of capital of10%/yr.The company’s effective tax rate is 34%.Cost of Capital LOANS = 12% * (1 - 0.34) = 7.92%Cost of Capital RETAINED EARNINGS = 10%Weighted Average Cost of Capital = (0.25)*7.92% +(0.75)*10.00% = 9.48%4.5 TAX CONSIDERATIONS4.5.1 After Tax Cash FlowsTaxes are a fact of life in both personal and businessdecision making. Taxes occur in many forms andare primarily designed to generate revenues for governmentalentities ranging from local authorities to theFederal government. A few of the most common formsof taxes are income taxes, ad valorem taxes, sales taxes,and excise taxes. Cash flows used for economic analysisshould always be adjusted for the combined impact ofall relevant taxes. To do otherwise, ignores the significantimpact that taxes have on economic decision making.Tax laws and regulations are complex and intricate.A detailed treatment of tax considerations as they applyto economic analysis is beyond the scope of this chapterand generally requires the assistance of a professionalwith specialized training in the subject. A high levelsummary of concepts and techniques that concentrate onFederal income taxes are presented in the material whichfollows. The focus is on Federal income taxes since theyimpact most decisions and have relatively wide and generalapplication.The amount of Federal taxes due are determinedbased on a tax rate multiplied by a taxable income. Therates (as of April 2000) are determined based on tables ofrates published under the Omnibus Reconciliation Act of1993 as shown in Table 4.1. Depending on income range,the marginal tax rates vary from 15% of taxable incometo 39% of taxable income. Taxable income is calculated bysubtracting allowable deductions from gross income. Grossincome is generated when a company sells its productor service. Allowable deductions include salaries andwages, materials, interest payments, and depreciation aswell as other costs of doing business as detailed in thetax regulations.The calculation of taxes owed and after tax cashflows (ATCF) requires knowledge of:• Before Tax Cash Flows (BTCF), the net project cashflows before the consideration of taxes due, loanpayments, and bond payments;• Total loan payments attributable to the project,including a breakdown of principal and interestcomponents of the payments;• Total bond payments attributable to the project,including a breakdown of the redemption and interestcomponents of the payments; and• Depreciation allowances attributable to the project.

ECONOMIC ANALYSIS 45Table 4.1 Federal tax rates based on the Omnibus Reconciliation Act of 1993Taxable Income Taxes Marginal(TI) Due Tax Rate$0 < TI ≤ $50,000 0.15*TI 0.15$50,000 < TI ≤ $75,000 $7,500+0.25(TI-$50,000) 0.25$75,000 < TI ≤ $100,000 $13,750+0.34(TI-$75,000) 0.34$100,000 < TI ≤ $335,000 $22,250+0.39(TI-$100,000) 0.39$335,000 < TI ≤ $10,000,000 $113,900+0.34(TI-$335,000) 0.34$10,000,000 < TI ≤ $15,000,000 $3,400,000+0.35(TI-$10,000,000) 0.35$15,000,000 < TI ≤ $18,333,333 $5,150,000+0.38(TI-$15,000,000) 0.38$18,333,333 < TI $6,416,667+0.35(TI-$18,333,333) 0.35Given the availability of the above information, the procedureto determine the ATCF on a year-by-year basisproceeds using the following calculation for each year:• Taxable Income = BTCF - Loan Interest - Bond Interest- Deprecation• Taxes = Taxable Income * Tax Rate• ATCF = BTCF - Total Loan Payments - Total BondPayments - TaxesAn important observation is that Depreciationreduces Taxable Income (hence, taxes) but does not directlyenter into the calculation of ATCF since it is not atrue cash flow. It is not a true cash flow because no cashchanges hands. Depreciation is an accounting conceptdesign to stimulate business by reducing taxes over thelife of an asset. The next section provides additional informationabout depreciation.4.5.2 DepreciationMost assets used in the course of a business decreasein value over time. U.S. Federal income tax lawpermits reasonable deductions from taxable income toallow for this. These deductions are called depreciationallowances. To be depreciable, an asset must meet threeprimary conditions: (1) it must be held by the businessfor the purpose of producing income, (2) it must wearout or be consumed in the course of its use, and (3) itmust have a life longer than a year.Many methods of depreciation have been allowedunder U.S. tax law over the years. Among these methodsare straight line, sum-of-the-years digits, de clining balance,and the accelerated cost recovery system. Descriptionsof these methods can be found in many referencesincluding economic analysis text books [White, et al.,1998]. The method currently used for depreciation ofassets placed in service after 1986 is the Modified AcceleratedCost Recovery System (MACRS). Determinationof the allowable MACRS depreciation deduction for anasset is a function of (1) the asset’s property class, (2) theasset’s basis, and (3) the year within the asset’s recoveryperiod for which the deduction is calculated.Eight property classes are defined for assets whichare depreciable under MACRS. The property classes andseveral examples of property that fall into each class areshown in Table 4.2. Professional tax guidance is recommendedto determine the MACRS property class for aspecific asset.The basis of an asset is the cost of placing the assetin service. In most cases, the basis includes the purchasecost of the asset plus the costs necessary to place the assetin service (e.g., installation charges).Given an asset’s property class and its depreciablebasis the depreciation allowance for each year of theasset’s life can be determined from tabled values ofMACRS percentages. The MACRS percentages specifythe percentage of an asset’s basis that are allowable asdeductions during each year of an asset’s recovery period.The MACRS percentages by recovery year (age ofthe asset) and property class are shown in Table 4.3.Example 2Determine depreciation allowances during eachrecovery year for a MACRS 5-year property with a basisof $10,000.Year 1 deduction: $10,000 * 20.00% = $2,000Year 2 deduction: $10,000 * 32.00% = $3,200

46 ENERGY MANAGEMENT HANDBOOKTable 4.2 MACRS property classesProperty ClassExample Assets———————————————————————————————————————————3-Year Propertyspecial handling devices for foodspecial tools for motor vehicle manufacturing———————————————————————————————————————————5-Year Propertycomputers and office machinesgeneral purpose trucks———————————————————————————————————————————7-Year Propertyoffice furnituremost manufacturing machine tools———————————————————————————————————————————10-Year Propertytugs & water transport equipmentpetroleum refining assets———————————————————————————————————————————15-Year Propertyfencing and landscapingcement manufacturing assets———————————————————————————————————————————20-Year Propertyfarm buildingsutility transmission lines and poles———————————————————————————————————————————27.5-Year Residential Rental Propertyrental houses and apartments———————————————————————————————————————————31.5-Year Nonresidential Real Propertybusiness buildingsYear 3 deduction: $10,000 * 19.20% = $1,920Year 4 deduction: $10,000 * 11.52% = $1,152Year 5 deduction: $10,000 * 11.52% = $1,152Year 6 deduction: $10,000 * 5.76% = $576The sum of the deductions calculated in Example2 is $10,000 which means that the asset is “fully depreciated”after six years. Though not shown here, tablessimilar to Table 4.3 are available for the 27.5-Year and31.5-Year property classes. Their usage is similar to thatoutlined above except that depreciation is calculatedmonthly rather than annually.4.6 TIME VALUE OF MONEY CONCEPTS4.6.1 IntroductionMost people have an intuitive sense of the timevalue of money. Given a choice between $100 today and$100 one year from today, almost everyone would preferthe $100 today. Why is this the case? Two primary factorslead to this time preference associated with money;interest and inflation. Interest is the ability to earn areturn on money which is loaned rather than consumed.By taking the $100 today and placing it in an interestbearing bank account (i.e., loaning it to the bank), oneyear from today an amount greater than $100 would beavailable for withdrawal. Thus, taking the $100 todayand loaning it to earn interest, generates a sum greaterthan $100 one year from today and thus is preferred.The amount in excess of $100 that would be availabledepends upon the interest rate being paid by the bank.The next section develops the mathematics of the relationshipbetween interest rates and the timing of cashflows.The second factor which leads to the time preferenceassociated with money is inflation. Inflation is acomplex subject but in general can be described as adecrease in the purchasing power of money. The impactof inflation is that the “basket of goods” a consumer canbuy today with $100 contains more than the “basket” theconsumer could buy one year from today. This decreasein purchasing power is the result of inflation. The subjectof inflation is addressed in Section 4.9.4.4.6.2 The Mathematics of InterestThe mathematics of interest must account for theamount and timing of cash flows. The basic formula forstudying and understanding interest calculations is:where:F n = P + I nF n = a future amount of money at the end ofthe nth year;P =a present amount of money at the beginningof the year which is n years prior toF n ;

ECONOMIC ANALYSIS 47Table 4.3 MACRS percentages by recovery year and property classRecovery 3-Year 5-Year 7-Year 10-Year 15-Year 20-YearYear Property Property Property Property Property Property————————————————————————————————————1 33.33% 20.00% 14.29% 10.00% 5.00% 3.750%————————————————————————————————————2 44.45% 32.00% 24.49% 18.00% 9.50% 7.219%————————————————————————————————————3 14.81% 19.20% 17.49% 14.40% 8.55% 6.677%————————————————————————————————————4 7.41% 11.52% 12.49% 11.52% 7.70% 6.177%————————————————————————————————————5 11.52% 8.93% 9.22% 6.93% 5.713%————————————————————————————————————6 5.76% 8.92% 7.37% 6.23% 5.285%————————————————————————————————————7 8.93% 6.55% 5.90% 4.888%————————————————————————————————————8 4.46% 6.55% 5.90% 4.522%————————————————————————————————————9 6.56% 5.91% 4.462%————————————————————————————————————10 6.55% 5.90% 4.461%————————————————————————————————————11 3.28% 5.91% 4.462%————————————————————————————————————12 5.90% 4.461%————————————————————————————————————13 5.91% 4.462%————————————————————————————————————14 5.90% 4.461%————————————————————————————————————15 5.91% 4.462%————————————————————————————————————16 2.95% 4.461%————————————————————————————————————17 4.462%————————————————————————————————————18 4.461%————————————————————————————————————19 4.462%————————————————————————————————————20 4.461%————————————————————————————————————21 2.231%I n =the amount of accumulated interest overn years; andn = the number of years between P and F.The goal of studying the mathematics of interest isto develop a formula for F n which is expressed only interms of the present amount P, the annual interest ratei, and the number of years n. There are two major approachesfor determining the value of I n ; simple interestand compound interest. Under simple interest, interestis earned (charged) only on the original amount loaned(borrowed). Under compound interest, interest is earned(charged) on the original amount loaned (borrowed)plus any interest accumulated from previous periods.4.6.3 Simple InterestFor simple interest, interest is earned (charged)only on the original principal amount at the rate of i%per year (expressed as i%/yr). Table 4.4 illustrates theannual calculation of simple interest. In Table 4.4 andthe formulas which follow, the interest rate i is to beexpressed as a decimal amount (e.g., 8% interest is expressedas 0.08).At the beginning of year 1 (end of year 0), Pdollars (e.g., $100) are deposited in an account earningi%/yr (e.g., 8%/yr or 0.08) simple interest. Undersimple compounding, during year 1 the P dollars ($100)earn P*i dollars ($100*0.08 = $8) of interest. At the endof the year 1 the balance in the account is obtained byadding P dollars (the original principal, $100) plus P*i(the interest earned during year 1, $8) to obtain P+P*i($100+$8=$108). Through algebraic manipulation, theend of year 1 balance can be expressed mathematicallyas P*(1+i) dollars ($100*1.08=$108).The beginning of year 2 is the same point in timeas the end of year 1 so the balance in the account isP*(1+i) dollars ($108). During year 2 the account again

48 ENERGY MANAGEMENT HANDBOOKTable 4.4 The mathematics of simple interestYear Amount At Interest Earned Amount At EndBeginning Of During Year Of Year(t) Year (F t )—————————————————————————————————0 - - P1 P Pi P + Pi= P (1 + i)2 P (1 + i) Pi P (1+ i) + Pi= P (1 + 2i)3 P (1 + 2i) Pi P (1+ 2i) + Pi= P (1 + 3i)n P (1 + (n-1)i) Pi P (1+ (n-1)i) + Pi= P (1 + ni)earns P*i dollars ($8) of interest since under simplecompounding, interest is paid only on the original principalamount P ($100). Thus at the end of year 2, thebalance in the account is obtained by adding P dollars(the original principal) plus P*i (the interest from year1) plus P*i (the interest from year 2) to obtain P+P*i+P*i($100+$8+$8=$116). After some algebraic manipulation,this can be written conveniently mathematically asP*(1+2*i) dollars ($100*1.16=$116).Table 4.4 extends the above logic to year 3 and thengeneralizes the approach for year n. If we return ourattention to our original goal of developing a formulafor F n which is expressed only in terms of the presentamount P, the annual interest rate i, and the number ofyears n, the above development and Table 4.4 results canbe summarized as follows:For Simple InterestF n = P (1+n*i)Example 3Determine the balance which will accumulate atthe end of year 4 in an account which pays 10%/yrsimple interest if a deposit of $500 is made today.F n = P * (1 + n*i)F 4 = 500 * (1 + 4*0.10)F 4 = 500 * (1 + 0.40)F 4 = 500 * (1.40)F 4 = $7004.6.4 Compound InterestFor compound interest, interest is earned (charged)on the original principal amount plus any accumulatedinterest from previous years at the rate of i% per year (i%/yr). Table 4.5 illustrates the annual calculation of compoundinterest. In the Table 4.5 and the formulas whichfollow, i is expressed as a decimal amount (i.e., 8% interestis expressed as 0.08).At the beginning of year 1 (end of year 0), P dollars(e.g., $100) are deposited in an account earningi%/yr (e.g., 8%/yr or 0.08) compound interest. Undercompound interest, during year 1 the P dollars ($100)earn P*i dollars ($100*0.08 = $8) of interest. Noticethat this the same as the amount earned under simplecompounding. This result is expected since the interestearned in previous years is zero for year 1. At the endof the year 1 the balance in the account is obtain byadding P dollars (the original principal, $100) plus P*i(the interest earned during year 1, $8) to obtain P+P*i($100+$8=$108). Through algebraic manipulation, theend of year 1 balance can be expressed mathematicallyas P*(1+i) dollars ($100*1.08=$108).During year 2 and subsequent years, we begin tosee the power (if you are a lender) or penalty (if you area borrower) of compound interest over simple interest.The beginning of year 2 is the same point in time as theend of year 1 so the balance in the account is P*(1+i) dollars($108). During year 2 the account earns i% intereston the original principal, P dollars ($100), and it earns i%interest on the accumulated interest from year 1, P*i dollars($8). Thus the interest earned in year 2 is [P+P*i]*idollars ([$100+$8]*0.08=$8.64). The balance at the end ofyear 2 is obtained by adding P dollars (the original principal)plus P*i (the interest from year 1) plus [P+P*i]*i(the interest from year 2) to obtain P+P*i+[P+P*i]*i dollars($100+$8+$8.64=$116.64). After some algebraic ma-

ECONOMIC ANALYSIS 49Table 4.5 The Mathematics of Compound Interest———————————————————————————————————————Year Amount At Interest Earned Amount At EndBeginning Of During Year Of Year(t) Year (F t )———————————————————————————————————————0 - - P———————————————————————————————————————1 P Pi P + Pi= P (1 + i)———————————————————————————————————————2 P (1 + i) P (1 + i) i P (1+ i) + P (1 + i) i= P (1 + i) (1 + i)= P (1+i) 2———————————————————————————————————————3 P (1+i) 2 P (1+i) 2 i P (1+ i) 2 + P (1 + i) 2 i= P (1 + i) 2 (1 + i)= P (1+i) 3———————————————————————————————————————n P (1+i) n-1 P (1+i) n-1 i P (1+ i) n-1 + P (1 + i) n-1 i= P (1 + i) n-1 (1 + i)= P (1+i) n———————————————————————————————————————nipulation, this can be written conveniently mathematicallyas P*(1+i) n dollars ($100*1.082 =$116.64).Table 4.5 extends the above logic to year 3 and thengeneralizes the approach for year n. If we return ourattention to our original goal of developing a formulafor F n which is expressed only in terms of the presentamount P, the annual interest rate i, and the number ofyears n, the above development and Table 4.5 results canbe summarized as follows:For Compound InterestF n = P (1+i) nExample 4Repeat Example 3 using compound interest ratherthan simple interest.F n = P * (1 + i) nF 4 = 500 * (1 + 0.10) 4F 4 = 500 * (1.10) 4F 4 = 500 * (1.4641)F 4 = $732.05Notice that the balance available for withdrawalis higher under compound interest ($732.05 > $700.00).This is due to earning interest on principal plus interestrather than earning interest on just original principal.Since compound interest is by far more common in practicethan simple interest, the remainder of this chapteris based on compound interest unless explicitly statedotherwise.4.6.5 Single Sum Cash FlowsTime value of money problems involving compoundinterest are common. Because of this frequentneed, tables of compound interest time value of moneyfactors can be found in most books and reference manualsthat deal with economic analysis. The factor (1+i) n isknown as the single sum, future worth factor or the singlepayment, compound amount factor. This factor is denoted(F|P,i,n) where F denotes a future amount, P denotesa present amount, i is an interest rate (expressed as apercentage amount), and n denotes a number of years.The factor (F|P,i,n) is read “to find F given P at i% forn years.” Tables of values of (F|P,i,n) for selected valuesof i and n are provided in Appendix 4A. The tables ofvalues in Appendix 4A are organized such that the annualinterest rate (i) determines the appropriate page,the time value of money factor (F|P) determines theappropriate column, and the number of years (n) determinesthe appropriate row.Example 5Repeat Example 4 using the single sum, futureworth factor.

50 ENERGY MANAGEMENT HANDBOOKF n = P * (1 + i) nF n = P * (F|P,i,n)F 4 = 500 * (F|P,10%,4)F 4 = 500 * (1.4641)F 4 = 732.05The above formulas for compound interest allowus to solve for an unknown F given P, i, and n. What ifwe want to determine P with known values of F, i, andn? We can derive this relationship from the compoundinterest formula above:F n = P (1+i) ndividing both sides by (1+i) n yieldsF nP = ————(1 + i) nwhich can be rewritten asP = F n (1+i) -nThe factor (1+i) -n is known as the single sum, presentworth factor or the single payment, present worth factor. Thisfactor is denoted (P|F,i,n) and is read “to find P givenF at i% for n years.” Tables of (P|F,i,n) are provided inAppendix 4A.Example 6To accumulate $1000 five years from today in anaccount earning 8%/yr compound interest, how muchmust be deposited today?P = F n * (1 + i) -nP = F 5 * (P|F,i,n)P = 1000 * (P|F,8%,5)P = 1000 * (0.6806)P = 680.60To verify your solution, try multiplying 680.60 *(F|P,8%,5). What would expect for a result? (Answer:$1000) If you're still not convinced, try building a tablelike Table 4.5 to calculate the year end balances each yearfor five years.4.6.6 Series Cash FlowsHaving considered the transformation of a singlesum to a future worth when given a present amount andvice versa, let us generalize to a series of cash flows. Thefuture worth of a series of cash flows is simply the sumof the future worths of each individual cash flow. Similarly,the present worth of a series of cash flows is thesum of the present worths of the individual cash flows.Example 7Determine the future worth (accumulated total) atthe end of seven years in an account that earns 5%/yrif a $600 deposit is made today and a $1000 deposit ismade at the end of year two?for the $600 deposit, n=7 (years between today and endof year 7)for the $1000 deposit, n=5 (years between end of year 2and end of year 7)F7 = 600 * (F|P,5%,7) + 1000 * (F|P,5%,5)F7 = 600 * (1.4071) + 1000 * (1.2763)F7 = 844.26 + 1276.30 = $2120.56Example 8Determine the amount that would have to be depositedtoday (present worth) in an account paying 6%/yr interest if you want to withdraw $500 four years fromtoday and $600 eight years from today (leaving zero inthe account after the $600 withdrawal).for the $500 deposit n=4, for the $600 deposit n=8P = 500 * (P|F,6%,4) + 600 * (P|F,6%,8)P = 500 * (0.7921) + 600 * (0.6274)P = 396.05 + 376.44 = $772.494.6.7 Uniform Series Cash FlowsA uniform series of cash flows exists when thecash flows in a series occur every year and are all equalin value. Figure 4.3 shows the cash flow diagram of auniform series of withdrawals. The uniform series haslength 4 and amount 2000. If we want to determine theamount of money that would have to be deposited todayto support this series of withdrawals starting one yearfrom today, we could use the approach illustrated in Example8 above to determine a present worth componentfor each individual cash flow. This approach would requireus to sum the following series of factors (assumingthe interest rate is 9%/yr):P = 2000*(P|F,9%,1) + 2000*(P|F,9%,2) +2000*(P|F,9%,3) + 2000*(P|F,9%,4)After some algebraic manipulation, this expression canbe restated as:

ECONOMIC ANALYSIS 51P = 2000*[(P|F,9%,1) + (P|F,9%,2) +(P|F,9%,3) + (P|F,9%,4)]P = 2000*[(0.9174) + (0.8417) + (0.7722) + (0.7084)]P = 2000*[3.2397] = $6479.402000 2000 2000 20000 1 2 3 4Figure 4.3. Uniform series cash flowFortunately, uniform series occur frequentlyenough in practice to justify tabulating values to eliminatethe need to repeatedly sum a series of (P|F,i,n)factors. To accommodate uniform series factors, weneed to add a new symbol to our time value of moneyterminology in addition to the single sum symbols Pand F. The symbol “A” is used to designate a uniformseries of cash flows. When dealing with uniform seriescash flows, the symbol A represents the amount of eachannual cash flow and the n represents the number ofcash flows in the series. The factor (P|A,i,n) is knownas the uniform series, present worth factor and is read “tofind P given A at i% for n years.” Tables of (P|A,i,n) areprovided in Appendix 4A. An algebraic expression canalso be derived for the (P|A,i,n) factor which expressesP in terms of A, i, and n. The derivation of this formulais omitted here, but the resulting expression is shown inthe summary table (Table 4.6) at the end of this section.An important observation when using a (P|A,i,n)factor is that the “P” resulting from the calculation occursone period prior to the first “A” cash flow. In ourexample the first withdrawal (the first “A”) occurred oneyear after the deposit (the “P”). Restating the exampleproblem above using a (P|A,i,n) factor, it becomes:provided in Appendix 4A and the algebraic expressionfor (A|P,i,n) is shown in Table 4.6 at the end of this section.This factor enables us to determine the amount ofthe equal annual withdrawals “A” (starting one year afterthe deposit) that can be made from an initial depositof “P.”Example 9Determine the equal annual withdrawals that canbe made for 8 years from an initial deposit of $9000 inan account that pays 12%/yr. The first withdrawal is tobe made one year after the initial deposit.A = P * (A|P,12%,8)A = 9000 * (0.2013)A = $1811.70Factors are also available for the relationships betweena future worth (accumulated amount) and a uniformseries. The factor (F|A,i,n) is known as the uniformseries future worth factor and is read “to find F given A ati% for n years.” The reciprocal factor, (A|F,i,n), is knownas the uniform series sinking fund factor and is read “tofind A given F at i% for n years.” An important observationwhen using an (F|A,i,n) factor or an (A|F,i,n) factoris that the “F” resulting from the calculation occursat the same point in time as to the last “A” cash flow.The algebraic expressions for (A|F,i,n) and (F|A,i,n) areshown in Table 4.6 at the end of this section.Example 10If you deposit $2000 per year into an individualretirement account starting on your 24th birthday, howmuch will have accumulated in the account at the timeof your deposit on your 65th birthday? The account pays6%/yr.P = A * (P|A,i,n)P = 2000 * (P|A,9%,4)P = 2000 * (3.2397) = $6479.40This result is identical (as expected) to the resultusing the (P|F,i,n) factors. In both cases the interpretationof the result is as follows: if we deposit $6479.40in an account paying 9%/yr interest, we could makewithdrawals of $2000 per year for four years starting oneyear after the initial deposit to deplete the account at theend of 4 years.The reciprocal relationship between P and A issymbolized by the factor (A|P,i,n) and is called the uniformseries, capital recovery factor. Tables of (A|P,i,n) aren = 42 (birthdays between 24th and 65th, inclusive)F = A * (F|A,6%,42)F = 2000 * (175.9505) = $351,901Example 11If you want to be a millionaire on your 65th birthday,what equal annual deposits must be made in an accountstarting on your 24th birthday? The account pays10%/yr.n = 42 (birthdays between 24th and 65th, inclusive)A = F * (A|F,10%,42)A = 1000000 * (0.001860) = $1860

52 ENERGY MANAGEMENT HANDBOOK4.6.8 Gradient SeriesA gradient series of cash flows occurs when thevalue of a given cash flow is greater than the value of theprevious period’s cash flow by a constant amount. Thesymbol used to represent the constant increment is G.The factor (P|G,i,n) is known as the gradient series, presentworth factor. Tables of (P|G,i,n) are provided in Appendix4A. An algebraic expression can also be derivedfor the (P|G,i,n) factor which expresses P in terms of G,i, and n. The derivation of this formula is omitted here,but the resulting expression is shown in the summarytable (Table 4.6) at the end of this section.It is not uncommon to encounter a cash flow seriesthat is the sum of a uniform series and a gradient series.Figure 4.4 illustrates such a series. The uniform componentof this series has a value of 1000 and the gradientseries has a value of 500. By convention the first elementof a gradient series has a zero value. Therefore, in Figure4.4, both the uniform series and the gradient series havelength four (n=4). Like the uniform series factor, the “P”calculated by a (P|G,i,n) factor is located one periodbefore the first element of the series (which is the zeroelement for a gradient series).Example 12Assume you wish to make the series of withdrawalsillustrated in Figure 4.4 from an account which pays15%/yr. How much money would you have to deposittoday such that the account is depleted at the time of thelast withdrawal?This problem is best solved by recognizing that thecash flows are a combination of a uniform series ofvalue 1000 and length 4 (starting at time=1) plus agradient series of size 500 and length 4 (starting attime=1).P = A * (P|A,15%,4) + G * (P|G,15%,4)P = 1000 * (2.8550) + 500 * (3.7864)P = 2855.00 + 1893.20 = $4748.20Occasionally it is useful to convert a gradient seriesto an equivalent uniform series of the same length.Equivalence in this context means that the present value(P) calculated from the gradient series is numericallyequal to the present value (P) calculated from the uniformseries. One way to accomplish this task with thetime value of money factors we have already consideredis to convert the gradient series to a present value usinga (P|G,i,n) factor and then convert this present valueto a uniform series using an (A|P,i,n) factor. In otherwords:2000 2000 2000 20001 2 3 4Figure 4.4. Combined uniform series and gradient seriescash flowA = [G * (P|G,i,n)] * (A|P,i,n)An alternative approach is to use a factor known asthe gradient-to-uniform series conversion factor, symbolizedby (A|G,i,n). Tables of (A|G,i,n) are provided in Appendix4A. An algebraic expression can also be derived forthe (A|G,i,n) factor which expresses A in terms of G, i,and n. The derivation of this formula is omitted here, butthe resulting expression is shown in the summary table(Table 4.6) at the end of this section.4.6.9 Summary of Time Value of Money FactorsTable 4.6 summarizes the time value of moneyfactors introduced in this section. Time value of moneyfactors are useful in economic analysis because they providea mechanism to accomplish two primary functions:(1) they allow us to replace a cash flow at one point intime with an equivalent cash flow (in a time value ofmoney sense) at a different point in time and (2) they allowus to convert one cash flow pattern to another (e.g.,convert a single sum of money to an equivalent cashflow series or convert a cash flow series to an equivalentsingle sum). The usefulness of these two functionswhen performing economic analysis of alternatives willbecome apparent in Sections 4.7 and 4.8 which follow.4.6.10 The Concepts of Equivalence and IndifferenceUp to this point the term “equivalence” has beenused several times but never fully defined. It is appropriateat this point to formally define equivalence as well asa related term, indifference.In economic analysis, “ equivalence” means “thestate of being equal in value.” The concept is primarilyapplied to the comparison of two or more cash flowprofiles. Specifically, two (or more) cash flow profiles areequivalent if their time value of money worths at a commonpoint in time are equal.Question: Are the following two cash flows equivalentat 15%/yr?Cash Flow 1: Receive $1,322.50 two years from todayCash Flow 2: Receive $1,000.00 today

ECONOMIC ANALYSIS 53Table 4.6 Summary of discrete compounding time value of money factorsTo Find Given Factor Symbol NameP F (1+i) -n (P|F,i,n) Single Payment, Present Worth FactorF P (1+i) n (F|P,i,n) Single Payment, Compound Amount Factor(1+i) n – 1P A ———— (P|A,i,n) Uniform Series, Present Worth Factori(1 + i) ni(1+i) nA P ———— (A|P,i,n) Uniform Series, Capital Recovery Factor(1+i) n – 1(1+i) n – 1F A ———— (F|A,i,n) Uniform Series, Compound Amount FactoriAFi1+i n –1 (A|F,i,n) Uniform Series, Sinking Fund FactorP G 1– 1+ni 1+i –n(P|G,i,n) Gradient Series, Present Worth Factori 2A G 1+i –n – 1 + ni (A|G,i,n) Gradient Series, Uniform Series Factori 1 + i –n –1Analysis Approach 1: Compare worths at t=0 (presentworth)PW(1) = 1,322.50*(P|F,15,2) = 1322.50*0.756147 = 1,000PW(2) = 1,000Answer: Cash Flow 1 and Cash Flow 2 are equivalentAnalysis Approach 2: Compare worths at t=2 (futureworth)FW(1) = 1,322.50FW(2) = 1,000*(F|P,15,2) = 1,000*1.3225 = 1,322.50Answer: Cash Flow 1 and Cash Flow 2 are equivalentGenerally the comparison (hence the determinationof equivalence) for the two cash flow series in thisexample would be made as present worths (t=0) or futureworths (t=2), but the equivalence definition holdsregardless of the point in time chosen. For example:Analysis Approach 3: Compare worths at t=1W1(1) = 1,322.50*(P|F,15,1)= 1,322.50*0.869565 = 1,150.00W1(2) = 1,000*(F|P,15,1) = 1,000*1.15 = 1,150.00Answer: Cash Flow 1 and Cash Flow 2 are equivalentThus, the selection of the point in time, t, at whichto make the comparison is completely arbitrary. Clearlyhowever, some choices are more intuitively appealingthan others (t= 0 and t=2 in the above example).In economic analysis, “ indifference” means “tohave no preference” The concept is primarily applied inthe comparison of two or more cash flow profiles. Specifically,a potential investor is indifferent between two(or more) cash flow profiles if they are equivalent.Question: Given the following two cash flows at 15%/yrwhich do you prefer?Cash Flow 1: Receive $1,322.50 two years from todayCash Flow 2: Receive $1,000.00 todayAnswer: Based on the equivalence calculations above,given these two choices, an investor is indifferent.The concept of equivalence can be used to breaka large, complex problem into a series of smaller moremanageable ones. This is done by taking advantage of

54 ENERGY MANAGEMENT HANDBOOKthe fact that, in calculating the economic worth of a cashflow profile, any part of the profile can be replaced by anequivalent representation without altering the worth ofthe profile at an arbitrary point in time.Question: You are given a choice between (1) receiving Pdollars today or (2) receiving the cash flow series illustratedin Figure 4.5. What must the value of P be for youto be indifferent between the two choices if i=12%/yr?2000 2000 2000 2000 20000 1 2 3 4 5 6 7Figure 4.5 A cash flow seriesAnalysis Approach: To be indifferent between the choices,P must have a value such that the two alternativesare equivalent at 12%/yr. If we select t=0 as the commonpoint in time upon which to base the analysis(present worth approach), then the analysis proceedsas follows.PW(Alt 1) = PBecause P is already at t=0 (today), no time value ofmoney factors are involved.PW(Alt 2)Step 1 - Replace the uniform series (t=3 to 7) with anequivalent single sum, V 2 , at t=2 (one period beforethe first element of the series).V 2 = 2,000 * (P|A,12%,5) = 2,000 * 3.6048 = 7,209.60Step 2 - Replace the single sum V 2, with an equivalentvalue V 0 at t=0:PW(Alt 2) = V 0 =V 2 * (P|F,12,2) =7,209.60 * 0.7972 = 5,747.49Answer: To be indifferent between the two alternatives,they must be equivalent at t=0. To be equivalent, Pmust have a value of $5,747.494.7 PROJECT MEASURES OF WORTH4.7.1 IntroductionIn this section measures of worth for investmentprojects are introduced. The measures are used to evaluatethe attractiveness of a single investment opportunity.The measures to be presented are (1) present worth,(2) annual worth, (3) internal rate of return, (4) savingsinvestment ratio, and (5) payback period. All but oneof these measures of worth require an interest rate tocalculate the worth of an investment. This interest rate iscommonly referred to as the Minimum Attractive Rateof Return (MARR). There are many ways to determine avalue of MARR for investment analysis and no one wayis proper for all applications. One principle is, however,generally accepted. MARR should always exceed thecost of capital as described in Section 4.4, Sources ofFunds, presented earlier in this chapter.In all of the measures of worth below, the followingconventions are used for defining cash flows. At anygiven point in time (t = 0, 1, 2,..., n), there may exist bothrevenue (positive) cash flows, R t , and cost (negative)cash flows, C t . The net cash flow at t, A t , is defined as R t- C t .4.7.2 Present WorthConsider again the cash flow series illustrated inFigure 4.5. If you were given the opportunity to “buy”that cash flow series for $5,747.49, would you be interestedin purchasing it? If you expected to earn a 12%/yrreturn on your money (MARR=12%), based on the analysisin the previous section, your conclusion would be(should be) that you are indifferent between (1) retainingyour $5,747.49 and (2) giving up your $5,747.49 in favorof the cash flow series. Figure 4.6 illustrates the net cashflows of this second investment opportunity.What value would you expect if we calculated thepresent worth (equivalent value of all cash flows at t=0)of Figure 4.6? We must be careful with the signs (directions)of the cash flows in this analysis since some representcash outflows (downward) and some represent cashinflows (upward).PW = -5747.49 + 2000*(P|A,12%,5)*(P|F,12%,2)PW = -5747.49 + 2000*(3.6048)*(0.7972)PW = -5747.49 + 5747.49 = $0.00The value of zero for present worth indicates indifferenceregarding the investment opportunity. We wouldjust as soon do nothing (i.e., retain our $5747.49) as investin the opportunity.What if the same returns (future cash inflows)where offered for a $5000 investment (t=0 outflow),Figure 4.6 An investment opportunity

ECONOMIC ANALYSIS 55would this be more or less attractive? Hopefully, after alittle reflection, it is apparent that this would be a moreattractive investment because you are getting the samereturns but paying less than the indifference amount forthem. What happens if calculate the present worth ofthis new opportunity?PW = -5000 + 2000*(P|A,12%,5)*(P|F,12%,2)PW = -5000 + 2000*(3.6048)*(0.7972)PW = -5000.00 + 5747.49 = $747.49The positive value of present worth indicates anattractive investment. If we repeat the process with aninitial cost greater than $5747.49, it should come as nosurprise that the present worth will be negative indicatingan unattractive investment.The concept of present worth as a measure of investmentworth can be generalized as follows:Measure of Worth: Present WorthDescription: All cash flows are converted to a single sumequivalent at time zero using i=MARR.Calculation Approach:nΣPW = A t P|F,i,tt=0Decision Rule: If PW≥0, then the investment is attractive.Example 13Installing thermal windows on a small officebuilding is estimated to cost $10,000. The windows areexpected to last six years and have no salvage value atthat time. The energy savings from the windows are expectedto be $2525 each year for the first three years and$3840 for each of the remaining three years. If MARR is15%/yr and the present worth measure of worth is to beused, is this an attractive investment?Figure 4.7 Thermal windows investmentThe cash flow diagram for the thermal windows isshown in Figure 4.7.PW =PW =PW =–10000+2525*(P|F,15%,1)+2525*(P|F,15%,2)+2525*(P|F,15%,3)+3840*(P|F,15%,4)+3840*(P|F,15%,5)+3840*(P|F,15%,6)–10000+2525*(0.8696)+2525*(0.7561)+2525*(0.6575)+3840*(0.5718)+3840*(0.4972)+3840*(0.4323)–10000+2195.74+1909.15+1660.19+2195.71+1909.25+1660.03PW = $1530.07Decision: PW≥0 ($1530.07≥0.0), therefore the windowinvestment is attractive.An alternative (and simpler) approach to calculatingPW is obtained by recognizing that the savingscash flows are two uniform series; one of value $2525and length 3 starting at t=1 and one of value $3840 andlength 3 starting at t=4.PW =-10000+2525*(P|A,15%,3)+3840*(P|A,15%,3)*(P|F,15%,3)PW = -10000+2525*(2.2832)+3840*(2.2832)*(0.6575) = $1529.70Decision: PW≥0 ($1529.70>0.0), therefore the windowinvestment is attractive.The slight difference in the PW values is causedby the accumulation of round off errors as the variousfactors are rounded to four places to the right of thedecimal point.4.7.3 Annual WorthAn alternative to present worth is annual worth(AW). The annual worth measure converts all cashflows to an equivalent uniform annual series of cashflows over the investment life using i=MARR. The annualworth measure is generally calculated by first calculatingthe present worth measure and then multiplyingthis by the appropriate (A|P,i,n) factor. A thoroughreview of the tables in Appendix 4A or the equations inTable 4.6 leads to the conclusion that for all values of i(i>0) and n (n>0), the value of (A|P,i,n) is greater thanzero. Hence,

56 ENERGY MANAGEMENT HANDBOOKif PW>0, then AW>0;if PW MARR, the project is attractive,if IRR < MARR, the project is unattractive,if IRR = MARR, indifferent.Thus, if MARR changes, no new calculations arerequired. We simply compare the calculated IRR for theproject to the new value of MARR and we have our decision.The value of IRR is typically determined througha trial and error process. An expression for the presentworth of an investment is written without specifying avalue for i in the time value of money factors. Then, variousvalues of i are substituted until a value is found thatsets the present worth (PW) equal to zero. The value of ifound in this way is the IRR.As appealing as the flexibility of this approachis, their are two major drawbacks. First, the iterationsrequired to solve using the trial and error approach tosolution can be time consuming. This factor is mitigatedby the fact that most spreadsheets and financial calculatorsare pre-programmed to solve for an IRR valuegiven a cash flow series. The second, and more serious,drawback to the IRR approach is that some cash flowseries have more than one value of IRR (i.e., more thanone value of i sets the PW expression to zero). A detaileddiscussion of this multiple solution issue is beyond thescope of this chapter, but can be found in White, et al.[1998], as well as most other economic analysis references.However, it can be shown that, if a cash flow seriesconsists of an initial investment (negative cash flowat t=0) followed by a series of future returns (positiveor zero cash flows for all t>0) then a unique IRR exists.If these conditions are not satisfied a unique IRR is notguaranteed and caution should be exercised in makingdecisions based on IRR.The concept of internal rate of return as a measureof investment worth can be generalized as follows:Measure of Worth: Internal Rate of ReturnDescription: An interest rate, IRR, is determined whichyields a present worth of zero. IRR implicitly assumesthe reinvestment of recovered funds at IRR.Calculation Approach:Σfind IRR such that PW = A t P|F,IRR,t = 0nt=0Important Note: Depending upon the cash flow series,multiple IRRs may exist! If the cash flow series consistsof an initial investment (net negative cash flow) followedby a series of future returns (net non-negative cashflows), then a unique IRR exists.

ECONOMIC ANALYSIS 57Decision Rule: If IRR is unique and IRR ≥MARR, thenthe investment is attractive.Example 15Reconsider the thermal window data of Example13. If the internal rate of return measure of worth is tobe used, is this an attractive investment?First we note that the cash flow series has a singlenegative investment followed by all positive returns,therefore, it has a unique value for IRR. Forsuch a cash flow series it can also be shown that asi increases PW decreases.From example 11, we know that for i=15%:PW =-10000+2525*(P|A,15%,3)+3840*(P|A,15%,3)*(P|F,15%,3)PW = -10000+2525*(2.2832)+3840*(2.2832)*(0.6575) = $1529.70Because PW>0, we must increase i to decrease PW towardzero for i=18%:PW =-10000+2525*(P|A,18%,3)+3840*(P|A,18%,3)*(P|F,18%,3)PW = -10000+2525*(2.1743)+3840*(2.1743)*(0.6086) = $571.50Since PW>0, we must increase i to decrease PW towardzero for i=20%:PW =-10000+2525*(P|A,20%,3)+3840*(P|A,20%,3)*(P|F,20%,3)PW = -10000+2525*(2.1065)+3840*(2.1065)*(0.5787) = -$0.01Although we could interpolate for a value of i for whichPW=0 (rather than -0.01), for practical purposes PW=0 ati=20%, therefore IRR=20%.Decision: IRR≥MARR (20%>15%), therefore the windowinvestment is attractive.4.7.5 Saving Investment RatioMany companies are accustomed to working withbenefit cost ratios. An investment measure of worthwhich is consistent with the present worth measure andhas the form of a benefit cost ratio is the savings investmentratio (SIR). The SIR decision rule can be derivedfrom the present worth decision rule as follows:Starting with the PW decision rulePW ≥0replacing PW with its calculation expressionnΣ A t P|F,i,t ≤ 0t=0which, using the relationship A t = R t - C t ,can be restatednΣt=0R t –C t P|F,i,t ≥ 0which can be algebraically separated intonΣt=0R t P|F,i,t –nΣ C tt=0P|F,i,t ≥ 0adding the second term to both sides of the inequalitynΣt=0nΣR t P|F,i,t ≥ C 1 P|F,i,t ≥ 0t=0dividing both sides of the inequalityby the right side termnΣ R 1 P|F,i,tt=0nΣ C 1 P|F,i,tt=0≥ 1which is the decision rule for SIR.The SIR represents the ratio of the present worthof the revenues to the present worth of the costs. If thisratio exceeds one, the investment is attractive.The concept of savings investment ratio as ameasure of investment worth can be generalized asfollows:Measure of Worth: Savings Investment RatioDescription: The ratio of the present worth of positivecash flows to the present worth of (the absolute value of)negative cash flows is formed using i=MARR.Calculation Approach: SIR =nΣ R t P|F,i,tt=0nΣ C t P|F,i,tt=0

58 ENERGY MANAGEMENT HANDBOOKDecision Rule: If SIR ≥1, then the investment is attractive.Example 16Reconsider the thermal window data of Example13. If the savings investment ratio measure of worth isto be used, is this an attractive investment?From example 13, we know that for i=15%:SIR =nΣ R t P|F,i,tt=0nΣ C t P|F,i,tt=02525*(P|A,15%,3)+3840*(P|A,15%,3)*(P|F,15%,3)SIR = —————————————————————100011529.70SIR = ————— = 1.1529710000.00Decision: SIR≥1.0 (1.15297>1.0), therefore the windowinvestment is attractive.An important observation regarding the four measuresof worth presented to this point (PW, AW, IRR,and SIR) is that they are all consistent and equivalent.In other words, an investment that is attractive underone measure of worth will be attractive under each ofthe other measures of worth. A review of the decisionsdetermined in Examples 13 through 16 will confirm theobservation. Because of their consistency, it is not necessaryto calculate more than one measure of investmentworth to determine the attractiveness of a project. Therationale for presenting multiple measures which areessentially identical for decision making is that variousindividuals and companies may have a preference forone approach over another.4.7.6 Payback PeriodThe payback period of an investment is generallytaken to mean the number of years required to recoverthe initial investment through net project returns. Thepayback period is a popular measure of investmentworth and appears in many forms in economic analysisliterature and company procedure manuals. Unfortunately,all too frequently, payback period is used inappropriatelyand leads to decisions which focus exclusivelyon short term results and ignore time value of moneyconcepts. After presenting a common form of paybackperiod these shortcomings will be discussed.Measure of Worth: Payback PeriodDescription: The number of years required to recover theinitial investment by accumulating net project returns isdetermined.Calculation Approach:ΣPBP = the smallest m such that A t ≥ C 0mt=1Decision Rule: If PBP is less than or equal to a predeterminedlimit (often called a hurdle rate), then the investmentis attractive.Important Note: This form of payback period ignoresthe time value of money and ignores returns beyond thepredetermined limit.The fact that this approach ignores time value ofmoney concepts is apparent by the fact that no timevalue of money factors are included in the determinationof m. This implicitly assumes that the applicable interestrate to convert future amounts to present amountsis zero. This implies that people are indifferent between$100 today and $100 one year from today, which is animplication that is highly inconsistent with observablebehavior.The short-term focus of the payback period measureof worth can be illustrated using the cash flow diagramsof Figure 4.8. Applying the PBP approach aboveyields a payback period for investment (a) of PBP=2(1200>1000 @ t=2) and a payback period for investment(b) of PBP=4 (1000300>1000) @ t=4). If the decision hurdlerate is 3 years (a very common rate), then investment(a) is attractive but investment (b) is not. Hopefully, it isobvious that judging (b) unattractive is not good decisionmaking since a $1,000,000 return four years after a$1,000 investment is attractive under almost any valueof MARR. In point of fact, the IRR for (b) is 465% so forany value of MARR less than 465%, investment (b) is attractive.4.8 ECONOMIC ANALYSIS4.8.1 IntroductionThe general scenario for economic analysis is thata set of investment alternatives are available and a decisionmust be made regarding which ones (if any) to acceptand which ones (if any) to reject. If the analysis isdeterministic, then an assumption is made that cash flowamounts, cash flow timing, and MARR are known with

ECONOMIC ANALYSIS 59600 600 600 6000 1 2 3 41000(a)1,000,000100 100 1000 1 2 3 4the others. Contingency creates dependence since priorto accepting a project, all projects on which it is contingentmust be accepted.If none of the above dependency situations arepresent and the projects are otherwise independent, thenthe evaluation of the set of projects is done by evaluatingeach individual project in turn and accepting the setof projects which were individually judged acceptable.This accept or reject judgment can be made using eitherthe PW, AW, IRR, or SIR measure of worth. The unconstraineddecision rules for each or these measures ofworth are restated below for convenience.Unconstrained PW Decision Rule: If PW ≥0, then theproject is attractive.1,000(b)Unconstrained AW Decision Rule: If AW ≥0, then theproject is attractive.Figure 4.8 Two investments evaluated using paybackperiodcertainty. Frequently, although this assumption does nothold exactly, it is not considered restrictive in terms ofpotential investment decisions. If however the lack ofcertainty is a significant issue then the analysis is stochasticand the assumptions of certainty are relaxed usingprobability distributions and statistical techniques toconduct the analysis. The remainder of this section dealswith deterministic economic analysis so the assumptionof certainty will be assumed to hold. Stochastic techniquesare introduced in Section 4.9.5.4.8.2 Deterministic Unconstrained AnalysisDeterministic economic analysis can be furtherclassified into unconstrained deterministic analysis andconstrained deterministic analysis. Under unconstrainedanalysis, all projects within the set available are assumedto be independent. The practical implication of thisindependence assumption is that an accept/reject decisioncan be made on each project without regard to thedecisions made on other projects. In general this requiresthat (1) there are sufficient funds available to undertakeall proposed projects, (2) there are no mutually exclusiveprojects, and (3) there are no contingent projects.A funds restriction creates dependency since, beforedeciding on a project being evaluated, the evaluatorwould have to know what decisions had been madeon other projects to determine whether sufficient fundswere available to undertake the current project. Mutualexclusion creates dependency since acceptance of one ofthe mutually exclusive projects precludes acceptance ofUnconstrained IRR Decision Rule: If IRR is unique andIRR ≥MARR, then the project is attractive.Unconstrained SIR Decision Rule: If SIR ≥1, then theproject is attractive.Example 17Consider the set of four investment projects whosecash flow diagrams are illustrated in Figure 4.9. If MARRis 12%/yr and the analysis is unconstrained, which projectsshould be accepted?Using present worth as the measure of worth:PW A = -1000+600*(P|A,12%,4) = -1000+600(3.0373) =$822.38 ⇒ Accept APW B = -1300+800*(P|A,12%,4) = -1300+800(3.0373) =$1129.88 ⇒ Accept BPW C = -400+120*(P|A,12%,4) = -400+120(3.0373) =-$35.52 ⇒ Reject CPW D = -500+290*(P|A,12%,4) = -500+290(3.0373) =$380.83 ⇒ Accept DTherefore,Accept Projects A, B, and D and Reject ProjectC4.8.3 Deterministic Constrained AnalysisConstrained analysis is required any time a dependencyrelationship exists between any of the projects

60 ENERGY MANAGEMENT HANDBOOKwithin the set to be analyzed. In general dependencyexists any time (1) there are insufficient funds availableto undertake all proposed projects (this is commonlyreferred to as capital rationing), (2) there are mutuallyexclusive projects, or (3) there are contingent projects.Several approaches have been proposed for selectingthe best set of projects from a set of potential projectsunder constraints. Many of these approaches will selectthe optimal set of acceptable projects under some conditionsor will select a set that is near optimal. However,only a few approaches are guaranteed to select the optimalset of projects under all conditions. One of theseapproaches is presented below by way of a continuationof Example 17.The first steps in the selection process are to specifythe cash flow amounts and cash flow timings for eachproject in the potential project set. Additionally, a valueof MARR to be used in the analysis must be specified.These issues have been addressed in previous sections sofurther discussion will be omitted here. The next step isto form the set of all possible decision alternatives fromthe projects. A single decision alternative is a collection ofzero, one, or more projects which could be accepted (allothers not specified are to be rejected). As an illustration,the possible decision alternatives for the set of projects illustratedin Figure 4.9 are listed in Table 4.7. As a generalrule, there will be 2 n possible decision alternatives generatedfrom a set of n projects. Thus, for the projects ofFigure 4.9, there are 2 4 = 16 possible decision alternatives.Since this set represents all possible decisions that couldbe made, one, and only one, will be selected as the best(optimal) decision. The set of decision alternatives developedin this way has the properties of being collectivelyexhaustive (all possible choices are listed) and mutuallyexclusive (only one will be selected).The next step in the process is to eliminate decisionsfrom the collectively exhaustive, mutually exclusiveset that represent choices which would violate one(or more) of the constraints on the projects. For the projectsof Figure 4.9, assume the following two constraintsexist:Project B is contingent on Project C, andA budget limit of $1500 exists on capital expendituresat t=0.Based on these constraints the following dec isionalternatives must be removed from the collectively exhaustive,mutually exclusive set: any combination thatincludes B but not C (B only; A&B; B&D; A&B&D),any combination not already eliminated whose t=0costs exceed $1500 (B&C, A&B&C, A&C&D, B&C&D,A&B&C&D). Thus, from the original set of 16 possibledecision alternatives, 9 have been eliminated and neednot be evaluated. These results are illustrated in Table4.8. It is frequently the case in practice that a significantpercentage of the original collectively exhaustive, mutuallyexclusive set will be eliminated before measures ofworth are calculated.The next step is to create the cash flow series forthe remaining (feasible) decision alternatives. This is astraight forward process and is accomplished by settinga decision alternative’s annual cash flow equal to thesum of the annual cash flows (on a year by year basis)of all projects contained in the decision alternative. Table4.9 illustrates the results of this process for the feasibledecision alternatives from Table 4.8.The next step is to calculate a measure of worthfor each decision alternative. Any of the four consistentmeasures of worth presented above (PW, AW, IRR,600 600 600 600800 800 800 8000 1 2 3 40 1 2 3 41000Project A1,300Project B120 120 120 120290 290 290 2900 1 2 3 40 1 2 3 4400500Project CProject DFigure 4.9 Four investments projects

ECONOMIC ANALYSIS 61or SIR but NOT PBP) can be used. The measures areentirely consistent and will lead to the same decisionalternative being selected. For illustrative purposes, PWwill be calculated for the decision alternatives of Table4.9 assuming MARR=12%.PW A = -1000 + 600*(P|A,12%,4) = -1000 + 600 (3.0373) = $822.38PW C = -400 + 120*(P|A,12%,4) = -400 + 120 (3.0373) = -$35.52PW D = -500 + 290*(P|A,12%,4) = -500 + 290 (3.0373) = $380.83PW A&C = -1400 + 720*(P|A,12%,4) = -1400 + 720 (3.0373) = $786.86PW A&D = -1500 + 890*(P|A,12%,4)= -1500 + 890 (3.0373) =$1203.21PW C&D = -900 + 410*(P|A,12%,4) = -900 + 410 (3.0373) = $345.31PW null = -0 + 0*(P|A,12%,4) = -0 + 0(3.0373) = $0.00Table 4.7 The decision alternatives from four projects————————————————————————Accept A onlyAccept B onlyAccept C onlyAccept D only————————————————————————Accept A and B onlyAccept A and C onlyAccept A and D onlyAccept B and C onlyAccept B and D onlyAccept C and D only————————————————————————Accept A, B, and C onlyAccept A, B, and D onlyAccept A, C, and D onlyAccept B, C, and D only————————————————————————Accept A, B, C, and D(frequently called the do everything alternative)————————————————————————Accept none(frequently called the do nothing or null alternative)————————————————————————Table 4.8 The decision alternatives with constraints imposed————————————————————————————————————Accept A onlyOKAccept B onlyinfeasible, B contingent on CAccept C onlyOKAccept D onlyOK————————————————————————————————————Accept A and B onlyinfeasible, B contingent on CAccept A and C onlyOKAccept A and D onlyOKAccept B and C onlyinfeasible, capital rationingAccept B and D onlyinfeasible, B contingent on CAccept C and D onlyOK————————————————————————————————————Accept A, B, and C onlyinfeasible, capital rationingAccept A, B, and D onlyinfeasible, B contingent on CAccept A, C, and D onlyinfeasible, capital rationingAccept B, C, and D onlyinfeasible, capital rationing————————————————————————————————————Do Everythinginfeasible, capital rationing————————————————————————————————————nullOK————————————————————————————————————

62 ENERGY MANAGEMENT HANDBOOKThe decision rules for the various measures ofworth under constrained analysis are list below.Constrained PW Decision Rule: Accept the decision alternativewith the highest PW.Constrained AW Decision Rule: Accept the decision alternativewith the highest AW.Constrained IRR Decision Rule: Accept the decision alternativewith the highest IRR.Constrained SIR Decision Rule: Accept the decision alternativewith the highest SIR.For the example problem, the highest presentworth ($1203.21) is associated with accepting projects Aand D (rejecting all others). This decision is guaranteedto be optimal (i.e., no feasible combination of projectshas a higher PW, AW, IRR, or SIR).4.8.4 Some Interesting ObservationsRegarding Constrained AnalysisSeveral interesting observations can be made regardingthe approach, measures of worth, and decisionsassociated with constrained analysis. Detailed developmentof these observations is omitted here but maybe found in many engineering economic analysis texts[White, et al., 1998].• The present worth of a decision alternative is thesum of the present worths of the projects containedwithin the alternative. (From above PW A&D = PW A+ PW D ).• The annual worth of a decision alternative is thesum of the annual worths of the projects containedwithin the alternative.• The internal rate of return of a decision alternativeis NOT the sum of internal rates of returns of theprojects contained within the alternative. The IRRfor the decision alternative must be calculated bythe trial and error process of finding the value ofi that sets the PW of the decision alternative tozero.• The savings investment ratio of a decision alternativeis NOT the sum of the savings investmentratios of the projects contained within the alternative.The SIR for the decision alternative mustbe calculated from the cash flows of the decisionalternative.• A common, but flawed, procedure for selecting theprojects to accept from the set of potential projectsinvolves ranking the projects (not decision alternatives)in preferred order based on a measure ofworth calculated for the project (e.g., decreasingproject PW) and then accepting projects as fardown the list as funds allow. While this procedurewill select the optimal set under some conditions(e.g., it works well if the initial investments of allprojects are small relative to the capital budgetlimit), it is not guaranteed to select the optimal setunder all conditions. The procedure outlined abovewill select the optimal set under all conditions.• Table 4.10 illustrates that the number of decisionalternatives in the collectively exhaustive, mutuallyexclusive set can grow prohibitively large as thenumber of potential projects increases. The mitigatingfactor in this combinatorial growth problem isthat in most practical situations a high percentageof the possible decision alternatives are infeasibleand do not require evaluation.Table 4.9 The decision alternatives cash flowsyr \ Alt A only C only D only A&C A&D C&D null——————————————————————————————————————————0 -1000 -400 -500 -1400 -1500 -900 0——————————————————————————————————————————1 600 120 290 720 890 410 0——————————————————————————————————————————2 600 120 290 720 890 410 0——————————————————————————————————————————3 600 120 290 720 890 410 0——————————————————————————————————————————4 600 120 290 720 890 410 0

ECONOMIC ANALYSIS 63Table 4.10 The number of decision alternativesas a function of the number of projectsNumber ofNumber of Projects Decision Alternatives1 22 43 84 165 326 647 1288 2569 51210 1,02415 32,76820 1,048,57625 33,554,4324.8.5 The Planning Horizon IssueWhen comparing projects, it is important to comparethe costs and benefits over a common period oftime. The intuitive sense of fairness here is based uponthe recognition that most consumers expect an investmentthat generates savings over a longer period of timeto cost more than an investment that generates savingsover a shorter period of time. To facilitate a fair, comparableevaluation a common period of time over which toconduct the evaluation is required. This period of time isreferred to as the planning horizon. The planning horizonissue arises when at least one project has cash flowsdefined over a life which is greater than or less than thelife of at least one other project. This situation did notoccur in Example 17 of the previous section since allprojects had 4 year lives.There are four common approaches to establishinga planning horizon for evaluating decision alternatives.These are (1) shortest life, (2) longest life, (3) least commonmultiple of lives, and (4) standard. The shortest lifeplanning horizon is established by selecting the projectwith the shortest life and setting this life as the planninghorizon. A significant issue in this approach is how tovalue the remaining cash flows for projects whose livesare truncated. The typical approach to this valuation isto estimate the value of the remaining cash flows as thesalvage value (market value) of the investment at thatpoint in its life.Example 18Determine the shortest life planning horizon forprojects A, B, C with lives 3, 5, and 6 years, respectively.The shortest life planning horizon is 3 years basedon Project A. A salvage value must be establishedat t=3 for B’s cash flows in years 4 and 5. A salvagevalue must be established at t=3 for C’s cash flowsin years 4, 5, and 6.The longest life planning horizon is establishedby selecting the project with the longest life and settingthis life as the planning horizon. The significant issuein this approach is how to handle projects whose cashflows don’t extend this long. The typical resolution forthis problem is to assume that shorter projects are repeatedconsecutively (end-to-end) until one of the repetitionsextends at least as far as the planning horizon.The assumption of project repeatability deserves carefulconsideration since in some cases it is reasonable and inothers it may be quite unreasonable. The reasonablenessof the assumption is largely a function of the type ofinvestment and the rate of innovation occurring withinthe investment’s field (e.g., assuming repeatability of investmentsin high technology equipment is frequently illadvised since the field is advancing rapidly). If in repeatinga project’s cash flows, the last repetition’s cash flowsextend beyond the planning horizon, then the truncatedcash flows (those that extend beyond the planning horizon)must be assigned a salvage value as above.Example 19Determine the longest life planning horizon forprojects A, B, C with lives 3, 5, and 6 years, respectively.The longest life planning horizon is 6 years basedon Project C. Project A must be repeated twice, thesecond repetition ends at year 6, so no terminationof cash flows is required. Project B’s second repetitionextends to year 10, therefore, a salvage value att=6 must be established for B’s repeated cash flowsin years 7, 8, 9, and 10.An approach that eliminates the truncation salvagevalue issue from the planning horizon question is theleast common multiple approach. The least commonmultiple planning horizon is set by determining thesmallest number of years at which repetitions of all projectswould terminate simultaneously. The least commonmultiple for a set of numbers (lives) can be determinedmathematically using algebra. Discussion of this ap-

64 ENERGY MANAGEMENT HANDBOOKproach is beyond the scope of this chapter. For a smallnumber of projects, the value can be determined by trialand error by examining multiples of the longest life project.Example 20Determine the least common multiple planninghorizon for projects A, B, C with lives 3, 5, and 6 years,respectively.The least common multiple of 3, 5, and 6 is 30. This canbe obtained by trial and error starting with the longestproject life (6) as follows:1st trial: 6*1=6; 6 is a multiple of 3 but not 5; reject 6 andproceed2nd trial: 6*2=12; 12 is a multiple of 3 but not 5; reject 12and proceed3rd trial: 6*3=18; 18 is a multiple of 3 but not 5; reject 18and proceed4th trial: 6*4=24; 24 is a multiple of 3 but not 5; reject 24and proceed5th trial: 6*5=30; 30 is a multiple of 3 and 5; accept 30and stopUnder a 30-year planning horizon, A’s cash flows arerepeated 10 times, B’s 6 times, and C’s 5 times. Notruncation is required.The standard planning horizon approach uses aplanning horizon which is independent of the projectsbeing evaluated. Typically, this type of planning horizonis based on company policies or practices. The standardhorizon may require repetition and/or truncation dependingupon the set of projects being evaluated.Example 21Determine the impact of a 5 year standard planninghorizon on projects A, B, C with lives 3, 5, and 6years, respectively.With a 5-year planning horizon:Project A must be repeated one time with the secondrepetition truncated by one year.Project B is a 5 year project and does not require repetitionor truncation.Project C must be truncated by one year.There is no single accepted approach to resolvingthe planning horizon issue. Companies and individualsgenerally use one of the approaches outlined above. Thedecision of which to use in a particular analysis is generallya function of company practice and consideration ofthe reasonableness of the project repeatability assumptionand the availability of salvage value estimates attruncation points.4.9 SPECIAL PROBLEMS4.9.1 IntroductionThe preceding sections of this chapter outline anapproach for conducting deterministic economic analysisof investment opportunities. Adherence to the conceptsand methods presented will lead to sound investmentdecisions with respect to time value of money principles.This section addresses several topics that are of specialinterest in some analysis situations.4.9.2 Interpolating Interest TablesAll of the examples previously presented in thischapter conveniently used interest rates whose timevalue of money factors were tabulated in Appendix4A. How does one proceed if non-tabulated time valueof money factors are needed? There are two viable approaches;calculation of the exact values and interpolation.The best and theoretically correct approach is tocalculate the exact values of needed factors based on theformulas in Table 4.6.Example 22Determine the exact value for (F|P,13%,7).From Table 4.6,(F|P,i,n) = (1+i) n = (1+.13) 7 = 2.3526Interpolation is often used instead of calculation ofexact values because, with practice, interpolated valuescan be calculated quickly. Interpolated values are not“exact” but for most practical problems they are “closeenough,” particularly if the range of interpolation is keptas narrow as possible. Interpolation of some factors, forinstance (P|A,i,n), also tends to be less error prone thanthe exact calculation due simpler mathematical operations.Interpolation involves determining an unknowntime value of money factor using two known valueswhich bracket the value of interest. An assumption ismade that the values of the time value of money factorvary linearly between the known values. Ratios are then

ECONOMIC ANALYSIS 65used to estimate the unknown value. The example belowillustrates the process.Example 23Determine an interpolated value for (F|P,13%,7).The narrowest range of interest rates which bracket13% and for which time value of money factor tablesare provided in Appendix 4A is 12% to 15%.The values necessary for this interpolation arei values(F|P,i%,7)12% 2.210713% (F|P,13%,7)15% 2.6600The interpolation proceeds by setting up ratiosand solving for the unknown value, (F|P,13%,7),as follows:change between rows 2 & 1 of left column——————————————————— =change between rows 3 & 1 of left column0.13 – 0.12————— =(F|P,13%,7) – 2.2107—————————0.15 – 0.12 2.6600 – 2.21070.01 (F|P,13%,7) – 2.2107—— = —————————0.03 0.44930.1498 = (F|P,13%,7) – 2.2107(F|P,13%,7) = 2.3605The interpolated value for (F|P,13%,7), 2.3605, differsfrom the exact value, 2.3526, by 0.0079. Thiswould imply a $7.90 difference in present worth forevery thousand dollars of return at t=7. The relativeimportance of this interpolation error can bejudged only in the context of a specific problem.4.9.3 Non-Annual Interest CompoundingMany practical economic analysis problems involveinterest that is not compounded annually. It is commonpractice to express a non-annually compounded interestrate as follows:12% per year compounded monthly or 12%/yr/mo.When expressed in this form, 12%/yr/mo isknown as the nominal annual interest rate The techniquescovered in this chapter up to this point can notbe used directly to solve an economic analysis problemof this type because the interest period (per year) andcompounding period (monthly) are not the same. Twoapproaches can be used to solve problems of this type.One approach involves determining a period interestrate, the other involves determining an effective interestrate.To solve this type of problem using a period interestrate approach, we must define the period interestrate:Nominal Annual Interest RatePeriod Interest Rate = ———————————————Number of Interest Periods per YearIn our example,12%/yr/moPeriod Interest Rate = ——————— = 1%/mo/mo12 mo/yrBecause the interest period and the compoundingperiod are now the same, the time value of moneyfactors in Appendix 4A can be applied directly. Notehowever, that the number of interest periods (n) must beadjusted to match the new frequency.Example 24$2,000 is invested in an account which pays 12%per year compounded monthly. What is the balance inthe account after 3 years?Nominal Annual Interest Rate = 12%/yr/mo12%/yr/moPeriod Interest Rate = ——————— = 1%/mo/mo12 mo/yrNumber of Interest Periods = 3 years × 12 mo/yr =36 interest periods (months)F = P (F|P,i,n) = $2,000 (F|P,1,36) = $2,000 (1.4308)= $2,861.60Example 25What are the monthly payments on a 5-year carloan of $12,500 at 6% per year compounded monthly?Nominal Annual Interest Rate = 6%/yr/mo6%/yr/moPeriod Interest Rate = —————12 mo/yr= 0.5%/mo/mo

66 ENERGY MANAGEMENT HANDBOOKNumber of Interest Periods = 5 years × 12 mo/yr =60 interest periodsA = P (A|P,i,n) = $12,500 (A|P,0.5,60) = $12,500(0.0193) = $241.25To solve this type of problem using an effectiveinterest rate approach, we must define the effectiveinterest rate. The effective annual interest rate is the annualizedinterest rate that would yield results equivalentto the period interest rate as previously calculated. Notehowever that the effective annual interest rate approachshould not be used if the cash flows are more frequentthan annual (e.g., monthly). In general, the interestrate for time value of money factors should match thefrequency of the cash flows (e.g., if the cash flows aremonthly, use the period interest rate approach withmonthly periods).As an example of the calculation of an effective interestrate, assume that the nominal interest rate is 12%/yr/qtr, therefore the period interest rate is 3%/qtr/qtr.One dollar invested for 1 year at 3%/qtr/qtr would havea future worth of:F = P (F|P,i,n) = $1 (F|P,3,4) = $1 (1.03) 4= $1 (1.1255) = $1.1255To get this same value in 1 year with an annual ratethe annual rate would have to be of 12.55%/yr/yr. Thisvalue is called the effective annual interest rate. The effectiveannual interest rate is given by (1.03) 4 - 1 = 0.1255or 12.55%.The general equation for the Effective Annual InterestRate is:Effective Annual Interest Rate = (1 + (r/m)) m –1where: r = nominal annual interest ratem = number of interest periods per yearExample 26What is the effective annual interest rate if thenominal rate is 12%/yr compounded monthly?nominal annual interest rate = 12%/yr/moperiod interest rate = 1%/mo/moeffective annual interest rate = (1+0.12/12) 12 -1 =0.1268 or 12.68%4.9.4 Economic Analysis Under InflationInflation is characterized by a decrease in thepurchasing power of money caused by an increase ingeneral price levels of goods and services without anaccompanying increase the value of the goods and services.Inflationary pressure is created when more dollarsare put into an economy without an accompanyingincrease in goods and services. In other words, printingmore money without an increase in economic outputgenerates inflation. A complete treatment of inflation isbeyond the scope of this chapter. A good summary canbe found in Sullivan and Bontadelli [1980].When consideration of inflation is introduced intoeconomic analysis, future cash flows can be stated interms of either constant-worth dollars or then-currentdollars. Then-current cash flows are expressed in termsof the face amount of dollars (actual number of dollars)that will change hands when the cash flow occurs. Alternatively,constant-worth cash flows are expressed interms of the purchasing power of dollars relative to afixed point in time known as the base period.Example 27For the next 4 years, a family anticipates buying$1000 worth of groceries each year. If inflation is expectedto be 3%/yr what are the then-current cash flowsrequired to purchase the groceries.To buy the groceries, the family will need to takethe following face amount of dollars to the store.We will somewhat artificially assume that the familyonly shops once per year, buys the same set ofitems each year, and that the first trip to the storewill be one year from today.Year 1: dollars required $1000.00*(1.03)=$1030.00Year 2: dollars required $1030.00*(1.03)=$1060.90Year 3: dollars required $1060.90*(1.03)=$1092.73Year 4: dollars required $1092.73*(1.03)=$1125.51What are the constant-worth cash flows, if today’sdollars are used as the base year.The constant worth dollars are inflation free dollars,therefore the $1000 of groceries costs $1000each year.Year 1: $1000.00Year 2: $1000.00Year 3: $1000.00Year 4: $1000.00The key to proper economic analysis under inflationis to base the value of MARR on the types ofcash flows. If the cash flows contain inflation, then the

ECONOMIC ANALYSIS 67value of MARR should also be adjusted for inflation.Alternatively, if the cash flows do not contain inflation,then the value of MARR should be inflation free. WhenMARR does not contain an adjustment for inflation, itis referred to as a real value for MARR. If it contains aninflation adjustment, it is referred to as a combined valuefor MARR. The relationship between inflation rate, thereal value of MARR, and the combined value of MARRis given by:1 + MARR COMBINED =(1 + inflation rate) * (1 + MARR REAL )Example 28If the inflation rate is 3%/yr and the real valueof MARR is 15%/yr, what is the combined value ofMARR?1 + MARR COMBINED= (1 + inflation rate) * (1 + MARR REAL )1 + MARR COMBINED = (1 + 0.03) * (1 + 0.15)1 + MARR COMBINED = (1.03) * (1.15)1 + MARR COMBINED = 1.1845MARR COMBINED = 1.1845 - 1 = 0.1845 = 18.45%If the cash flows of a project are stated in terms ofthen-current dollars, the appropriate value of MARR isthe combined value of MARR. Analysis done in this wayis referred to as then-current analysis. If the cash flowsof a project are stated in terms of constant-worth dollars,the appropriate value of MARR is the real value ofMARR. Analysis done in this way is referred to as thenconstant worth analysis.Example 29Using the cash flows of Examples 27 and interestrates of Example 28, determine the present worth of thegrocery purchases using a constant worth analysis.Constant worth analysis requires constant worthcash flows and the real value of MARR.PW = 1000 * (P|A,15%,4)= 1000 * (2.8550) = $2855.00Example 30Using the cash flows of Examples 27 and interestrates of Example 28, determine the present worth of thegrocery purchases using a then-current analysis.Then-current analysis requires then current cashflows and the combined value of MARR.PW = 1030.00 * (P|F,18.45%,1) + 1060.90 *(P|F,18.45%,2) + 1092.73 * (P|F,18.45%,3)+1125.51 * (P|F,18.45%,4)PW = 1030.00 * (0.8442) + 1060.90 * (0.7127) +1092.73 * (0.6017) +1125.51 * (0.5080)PW = 869.53 + 756.10 + 657.50 + 571.76 = 2854.89The notable result of Examples 29 and 30 is thatthe present worths determined by the constant-worthapproach ($2855.00) and the then-current approach($2854.89) are equal (the $0.11 difference is due torounding). This result is often unexpected but it ismathematically sound. The important conclusion isthat if care is taken to appropriately match the cashflows and value of MARR, the level of general priceinflation is not a determining factor in the acceptabilityof projects. To make this important result hold, inflationmust either (1) be included in both the cash flows andMARR (the then-current approach) or (2) be includedin neither the cash flows nor MARR (the constantworthapproach).4.9.5 Sensitivity Analysis and Risk AnalysisOften times the certainty assumptions associatedwith deterministic analysis are questionable. These certaintyassumptions include certain knowledge regardingamounts and timing of cash flows as well as certainknowledge of MARR. Relaxing these assumptions requiresthe use of sensitivity analysis and risk analysistechniques.Initial sensitivity analyses are usually conducted onthe optimal decision alternative (or top two or three) ona single factor basis. Single factor sensitivity analysis involvesholding all cost factors except one constant whilevarying the remaining cost factor through a range ofpercentage changes. The effect of cost factor changes onthe measure of worth is observed to determine whetherthe alternative remains attractive under the evaluatedchanges and to determine which cost factor affects themeasure of worth the most.Example 31Conduct a sensitivity analysis of the optimal decisionresulting from the constrained analysis of the datain Example 17. The sensitivity analysis should explorethe sensitivity of present worth to changes in annualrevenue over the range -10% to +10%.The PW of the optimal decision (Accept A & Donly) was determined in Section 4.8.3 to be:

68 ENERGY MANAGEMENT HANDBOOKPW A&D = -1500 + 890*(P|A,12%,4) = -1500 + 890(3.0373) = $1203.21If annual revenue decreases 10%, it becomes 890- 0.10*890 = 801 and PW becomesPW A&D = -1500 + 801*(P|A,12%,4) = -1500 + 801(3.0373) = $932.88If annual revenue increases 10%, it becomes 890 +0.10*890 = 979 and PW becomesPW A&D = -1500 + 979*(P|A,12%,4) = -1500 + 979(3.0373) = $1473.52The sensitivity of PW to changes in annual revenueover the range -10% to +10% is +$540.64 from$932.88 to $1473.52.Example 32Repeat Example 31 exploring of the sensitivity ofpresent worth to changes in initial cost over the range-10% to +10%.The PW of the optimal decision (Accept A & Donly) was determined in Section 4.8.3 to be:PW A&D = -1500 + 890*(P|A,12%,4) = -1500 + 890(3.0373) = $1203.21If initial cost decreases 10% it becomes 1500 -0.10*1500 = 1350 and PW becomesPW A&D = -1350 + 890*(P|A,12%,4) = -1350 + 890(3.0373) = $1353.20If initial cost increases 10% it becomes 1500 +0.10*1500 = 1650 and PW becomesPW A&D = -1650 + 890*(P|A,12%,4) = -1500 + 890(3.0373) = $1053.20The sensitivity of PW to changes in initial cost overthe range -10% to +10% is -$300.00 from $1353.20to $1053.20.Example 33Repeat Example 31 exploring the sensitivity of thepresent worth to changes in MARR over the range -10%to +10%.The PW of the optimal decision (Accept A & Donly) was determined in Section 4.8.3 to be:PW A&D = -1500 + 890*(P|A,12%,4) = -1500 + 890(3.0373) = $1203.21If MARR decreases 10% it becomes 12% - 0.10*12%= 10.8% and PW becomesPW A&D = -1500 + 890*(P|A,10.8%,4) = -1500 + 890(3.1157) = $1272.97If MARR increases 10% it becomes 12% + 0.10*12%= 13.2% and PW becomesPW A&D = -1500 + 890*(P|A,13.2%,4) = -1500 + 890(2.9622) = $1136.36The sensitivity of PW to changes in MARR overthe range -10% to +10% is -$136.61 from $1272.97to $1136.36.The sensitivity data from Examples 31, 32, and33 are summarized in Table 4.11. A review of the tablereveals that the decision alternative A&D remains attractive(PW ≥0) within the range of 10% changes inannual revenues, initial cost, and MARR. An appealingway to summarize single factor sensitivity data is usinga “spider” graph. A spider graph plots the PW valuesdetermined in the examples and connects them withlines, one line for each factor evaluated. Figure 4.10 illustratesthe spider graph for the data of Table 4.11. Onthis graph, lines with large positive or negative slopes(angle relative to horizontal regardless of whether it isincreasing or decreasing) indicate factors to which thepresent value measure of worth is sensitive. Figure 4.10shows that PW is least sensitive to changes in MARR(the MARR line is the most nearly horizontal) and mostsensitivity to changes in annual revenue (the annual revenueline has the steepest slope). Additional sensitivitiescould be explored in a similar manner.Table 4.11 Sensitivity analysis data tableFactor \Percent Change - 10% Base + 10%1st Cost 1353.20 1203.21 1053.20Annual Revenue 932.88 1203.21 1473.52MARR 1272.97 1203.21 1136.36When single factor sensitivity analysis is inadequateto assess the questions which surround thecertainty assumptions of a deterministic analysis, risk

ECONOMIC ANALYSIS 69Figure 4.10. Sensitivity analysis “spider” graphanalysis techniques can be employed. One approach torisk analysis is the application of probabilistic and statisticalconcepts to economic analysis. These techniquesrequire information regarding the possible values thatuncertain quantities may take on as well as estimates ofthe probability that the various values will occur. A detailedtreatment of this topic is beyond the scope of thischapter. A good discussion of this subject can be foundin Park and Sharp-Bette [1990].A second approach to risk analysis in economicanalysis is through the use of simulation techniquesand simulation software. Simulation involves using acomputer simulation program to sample possible valuesfor the uncertain quantities in an economic analysisand calculating the measure of worth. This process isrepeated many times using different samples each time.After many samples have been taken, probability statementsregarding the measure of worth may be made. Agood discussion of this subject can be found in Park andSharp-Bette [1990].4.10 SUMMARY AND ADDITIONALEXAMPLE APPLICATIONSIn this chapter a coherent, consistent approach toeconomic analysis of capital investments (energy relatedor other) has been presented. To conclude the chapter,this section provides several additional examples to illustratethe use of time value of money concepts for energyrelated problems. Additional example applicationsas well as a more in depth presentation of conceptualdetails can be found in the references listed at the endof the chapter. These references are by no means exclusive;many other excellent presentations of the subjectmatter are also available. Adherence to the concepts andmethods presented here and in the references will leadto sound investment decisions with respect to time valueof money principles.Example 34In Section 4.3.3 an example involving the evaluationof a baseboard heating and window air conditionerversus a heat pump was introduced to illustrate cashflow diagramming (Figure 4.2). A summary of the differentialcosts is repeat here for convenience.• The heat pump costs $1500 more than the baseboardsystem,• The heat pump saves $380 annually in electricitycosts,• The heat pump has a $50 higher annual maintenancecosts,• The heat pump has a $150 higher salvage value atthe end of 15 years,• The heat pump requires $200 more in replacementmaintenance at the end of year 8.If MARR is 18%, is the additional investment in theheat pump attractive?

70 ENERGY MANAGEMENT HANDBOOKUsing present worth as the measure of worth:PW =-1500 + 380*(P|A,18%,15) - 50*(P|A,18%,15)+ 150*(P|F,18%,15) - 200*(P|F,18%,8)PW = -1500 + 380*(5.0916) - 50*(5.0916) +150*(0.0835) - 200*(0.2660)PW = -1500.00 + 1934.81 - 254.58 + 12.53 - 53.20 =$139.56Decision: PW≥0 ($139.56>0.0), therefore the additionalinvestment for the heat pump is attractive.Example 35A homeowner needs to decide whether to installR-11 or R-19 insulation in the attic of her home. The R-19insulation costs $150 more to install and will save approximately400 kWh per year. If the planning horizon is20 years and electricity costs $0.08/kWh is the additionalinvestment attractive at MARR of 10%?At $0.08/kWh, the annual savings are: 400 kWh *$0.08/kWh = $32.00Using present worth as the measure of worth:PW = -150 + 32*(P|A,10%,20)PW = -150 + 32*(8.5136) = -150 + 272.44 = $122.44Decision: PW≥0 ($122.44>0.0), therefore the R-19insulation is attractive.Example 36The homeowner from Example 35 can install R-30insulation in the attic of her home for $200 more than theR-19 insulation. The R-30 will save approximately 250kWh per year over the R-19 insulation. Is the additionalinvestment attractive?Assuming the same MARR, electricity cost, andplanning horizon, the additional annual savingsare: 250 kWh * $0.08/kWh = $20.00Using present worth as the measure of worth:PW = -200 + 20*(P|A,10%,20)PW = -200 + 20*(8.5136) = -200 + 170.27 = -$29.73Decision: PW

ECONOMIC ANALYSIS 71AppendixTime Value of Money Factors—Discrete Compoundingi = 1%

72 ENERGY MANAGEMENT HANDBOOKTime Value of Money Factors—Discrete Compoundingi = 2%

ECONOMIC ANALYSIS 73Time Value of Money Factors—Discrete Compoundingi = 3%














CHAPTER 5BOILERS AND FIRED SYSTEMSS.A. PARKERSenior Research Engineer, Energy DivisionPacific Northwest National LaboratoryRichland, WashingtonR.B. SCOLLONCorporate Manager, Energy ConservationR.D. SMITHManager, Energy Generation and FeedstocksAllied CorporationMorristown, New Jerseyoperations, maintenance, and troubleshooting. Considerationsrelative to fuel comparison and selection are alsodiscussed. Developing technologies relative to alternativefuels and types of combustion equipment are alsodiscussed. Some of the technologies discussed hold thepotential for significant cost reductions while alleviatingenvironmental problems.The chapter concludes with a brief discussion ofsome of the major regulations impacting the operation ofboilers and fired systems. It is important to emphasize theneed to carefully assess the potential impact of federal,state, and local regulations.5. 1 INTRODUCTIONBoilers and other fired systems are the most significantenergy consumers. Almost two-thirds of the fossil-fuelenergy consumed in the United States involves the use ofa boiler, furnace, or other fired system. Even most electricenergy is produced using fuel-fired boilers. Over 68% ofthe electricity generated in the United States is producedthrough the combustion of coal, fuel oil, and natural gas.(The remainder is produced through nuclear, 22%; hydroelectric,10%; and geothermal and others,

88 ENERGY MANAGEMENT HANDBOOKload varies as a function of the process being supported.The efficiency varies as a function of the load and otherfunctions, such as time or weather. In addition, the fuelcost may also vary as a function of time (such as in seasonal,time-of-use, or spot market rates) or as a functionof load (such as declining block or spot market rates.)Therefore, solving the equation for the energy consumptionor energy cost may not always be simplistic.5.2.2 Balance EquationsBalance equations are used in an analysis of a processwhich determines inputs and outputs to a system.There are several types of balance equations which mayprove useful in the analysis of a boiler or fired-system.These include a heat balance and mass balance.Heat BalanceA heat balance is used to determine where all theheat energy enters and leaves a system. Assuming thatenergy can neither be created or destroyed, all energy canbe accounted for in a system analysis. Energy in equalsenergy out. Whether through measurement or analysis,all energy entering or leaving a system can be determined.In a simple furnace system, energy enters through thecombustion air, fuel, and mixed-air duct. Energy leavesthe furnace system through the supply-air duct and theexhaust gases.In a boiler system, the analysis can become morecomplex. Energy input comes from the following: condensatereturn, make-up water, combustion air, fuel, andmaybe a few others depending on the complexity of thesystem. Energy output departs as the following: steam,blowdown, exhaust gases, shell/surface losses, possiblyash, and other discharges depending on the complexityof the system.Mass BalanceA mass balance is used to determine where all massenters and leaves a system. There are several methods inwhich a mass balance can be performed that can be usefulin the analysis of a boiler or other fired system. In thecase of a steam boiler, a mass balance can be used in theform of a water balance (steam, condensate return, makeupwater, blowdown, and feedwater.) A mass balance canalso be used for water quality or chemical balance (totaldissolved solids, or other impurity.) The mass balance canalso be used in the form of a combustion analysis (firesidemass balance consisting of air and fuel in and combustiongasses and excess air out.) This type of analysisis the foundation for determining combustion efficiencyand determining the optimum air-to-fuel ratio.For analyzing complex systems, the mass and energybalance equations may be used simultaneously suchas in solving multiple equations with multiple unknowns.This type of analysis is particularly useful in determiningblowdown losses, waste heat recovery potential, andother interdependent opportunities.5.2.3 EfficiencyThere are several different measures of efficiencyused in boilers and fired systems. While this may lead tosome confusion, the different measures are used to conveydifferent information. Therefore, it is important tounderstand what is being implied by a given efficiencymeasure.The basis for testing boilers is the American Societyof Mechanical Engineers (ASME) Power Test Code 4.1(PTC-4.1-1964.) This procedure defines and establishedtwo primary methods of determining efficiency: the input-outputmethod and the heat-loss method. Both ofthese methods result in what is commonly referred to asthe gross thermal efficiency. The efficiencies determinedby these methods are “gross” efficiencies as apposed to“net” efficiencies which would include the additional energyinput of auxiliary equipment such as combustion airfans, fuel pumps, stoker drives, etc. For more informationon these methods, see the ASME PTC-4.1-1964 or Taplin1991.Another efficiency term commonly used for boilersand other fired systems is combustion efficiency. Combustionefficiency is similar to the heat loss method, but onlythe heat losses due to the exhaust gases are considered.Combustion efficiency can be measured in the field byanalyzing the products of combustion the exhaust gases.Typically measuring either carbon dioxide (CO 2 ) oroxygen (O 2 ) in the exhaust gas can be used to determinethe combustion efficiency as long as there is excess air. Excessair is defined as air in excess of the amount requiredfor stoichiometric conditions. In other words, excess airis the amount of air above that which is theoretically requiredfor complete combustion. In the real world, however,it is not possible to get perfect mixture of air and fuelto achieve complete combustion without some amount ofexcess air. As excess air is reduced toward the fuel richside, incomplete combustion begins to occur resulting inthe formation of carbon monoxide, carbon, smoke, andin extreme cases, raw unburned fuel. Incomplete combustionis inefficient, expensive, and frequently unsafe.Therefore, some amount of excess air is required to ensurecomplete and safe combustion.However, excess air is also inefficient as it results inthe excess air being heated from ambient air temperaturesto exhaust gas temperatures resulting in a form of heatloss. Therefore while some excess air is required it is also

BOILERS AND FIRED SYSTEMS 89desirable to minimize the amount of excess air.As illustrated in Figure 5.1, the amount of carbondioxide, percent by volume, in the exhaust gas reaches amaximum with no excess air stoichiometric conditions.While carbon dioxide can be used as a measure of completecombustion, it can not be used to optimally controlthe air-to-fuel ratio in a fired system. A drop in the levelof carbon dioxide would not be sufficient to inform thecontrol system if it were operating in a condition of excessair or insufficient air. However, measuring oxygen in theexhaust gases is a direct measure of the amount of excessair. Therefore, measuring oxygen in the exhaust gas is amore common and preferred method of controlling theair-to-fuel ratio in a fired system.5.2.4 Energy Conservation MeasuresAs noted above, energy cost reduction opportunitiescan generally be placed into one of the following categories:reducing load, increasing efficiency, and reducingunit energy cost. As with most energy conservation andcost reducing measures there are also a few additionalopportunities which are not so easily categorized. Table5.1 lists several energy conservation measures that havebeen found to be very cost effective in various boilers andfired-systems.5.3 KEY ELEMENTS FOR MAXIMUM EFFICIENCYThere are several opportunities for maximizing efficiencyand reducing operating costs in a boiler or otherfired-system as noted earlier in Table 5.1. This section examinesin more detail several key opportunities for energyand cost reduction, including excess air, stack temperature,load balancing, boiler blowdown, and condensatereturn.5.3.1 Excess AirIn combustion processes, excess air is generally definedas air introduced above the stoichiometric or theoreticalrequirements to effect complete and efficient combustionof the fuel.There is an optimum level of excess-air operationfor each type of burner or furnace design and fuel type.Only enough air should be supplied to ensure completecombustion of the fuel, since more than this amount in-%AIRFigure 5.1 Theoretical flue gas analysis versus air percentage for natural gas.

90 ENERGY MANAGEMENT HANDBOOKTable 5.1 Energy Conservation measuresfor boilers and fired systems aLoad ReductionInsulation—steam lines and distribution system—condensate lines and return system—heat exchangers—boiler or furnaceRepair steam leaksRepair failed steam strapsReturn condensate to boilerReduce boiler blowdownImprove feedwater treatmentImprove make-up water treatmentRepair condensate leaksShut off steam tracers during the summerShut off boilers during long periods of no useEliminate hot standbyReduce flash steam lossInstall stack dampers or heat traps in natural draft boilersReplace continuous pilots with electronic ignition pilotsWaste Heat Recovery (a form of load reduction)Utilize flash steamPreheat feedwater with an economizerPreheat make-up water with an economizerPreheat combustion air with a recuperatorRecover flue gas heat to supplement other heating system, such as domestic or service hot water, or unit space heaterRecover waste heat from some other system to preheat boiler make-up or feedwaterInstall a heat recovery system on incinerator or furnaceInstall condensation heat recovery system—indirect contact heat exchanger—direct contact heat exchangerEfficiency ImprovementReduce excess airProvide sufficient air for complete combustionInstall combustion efficiency control system—Constant excess air control—Minimum excess air control—Optimum excess air and CO controlOptimize loading of multiple boilersShut off unnecessary boilersInstall smaller system for part-load operation—Install small boiler for summer loads—Install satellite boiler for remote loadsInstall low excess air burnersRepair or replace faulty burnersReplace natural draft burners with forced draft burnersInstall turbulators in firetube boilersInstall more efficient boiler or furnace system—high-efficiency, pulse combustion, or condensing boiler or furnace systemClean heat transfer surfaces to reduce fouling and scaleImprove feedwater treatment to reduce scalingImprove make-up water treatment to reduce scalingFuel Cost ReductionSwitch to alternate utility rate schedule—interruptible rate schedulePurchase natural gas from alternate source, self procurement of natural gasFuel switching—switch between alternate fuel sources—install multiple fuel burning capability—replace electric boiler with a fuel-fired boiler

BOILERS AND FIRED SYSTEMS 91Switch to a heat pump—use heat pump for supplemental heat requirements—use heat pump for baseline heat requirementsOther OpportunitiesInstall variable speed drives on feedwater pumpsInstall variable speed drives on combustion air fanReplace boiler with alternative heating systemReplace furnace with alternative heating systemInstall more efficient combustion air fanInstall more efficient combustion air fan motorInstall more efficient feedwater pumpInstall more efficient feedwater pump motorInstall more efficient condensate pumpInstall more efficient condensate pump motora Reference: F.W. Payne, Efficient Boiler Operations Sourcebook, 3rd ed., Fairmont Press, Lilburn, GA, 1991.creases the heat rejected to the stack, resulting in greaterfuel consumption for a given process output.To identify the point of minimum excess-air operationfor a particular fired system, curves of combustiblesas a function of excess O 2 should be constructed similarto that illustrated in Figure 5.2. In the case of a gas-fueledsystem, the combustible monitored would be carbonmonoxide (CO), whereas, in the case of a liquid- or solidfueledsystem, the combustible monitored would be theSmoke Spot Number (SSN). The curves should be developedfor various firing rates as the minimal excess-air operatingpoint will also vary as a function of the firing rate(percent load). Figure 5.2 illustrates two potential curves,one for high-fire and the other for low-fire. The optimalexcess-air-control set point should be set at some margin(generally 0.5 to 1%) above the minimum O 2 point to allowfor response and control variances. It is important tonote that some burners may exhibit a gradual or steepCO-O 2 behavior and this behavior may even change withvarious firing rates. It is also important to note that someburners may experience potentially unstable operationwith small changes in O 2 (steep CO-O 2 curve behavior).Figure 5.2 Hypothetical CO-O2 characteristic curve for a gas-fired industrial boiler.

92 ENERGY MANAGEMENT HANDBOOKUpper control limits for carbon monoxide vary dependingon the referenced source. Points referenced for gasfiredsystems are typically 400 ppm, 200 ppm, or 100ppm. Today, local environmental regulations may dictateacceptable upper limits. Maximum desirable SSN for liquidfuels is typically SSN=1 for No. 2 fuel oil and SSN=4for No. 6 fuel oil. Again, local environmental regulationsmay dictate lower acceptable upper limits.Typical optimum levels of excess air normally attainablefor maximum operating efficiency are indicatedin Table 5.2 and classified according to fuel type and firingmethod.The amount of excess air (or O 2 ) in the flue gas,unburned combustibles, and the stack temperature riseabove the inlet air temperature are significant in definingthe efficiency of the combustion process. Excess oxygen(O 2 ) measured in the exhaust stack is the most typicalmethod of controlling the air-to-fuel ratio. However, formore precise control, carbon monoxide (CO) measurementsmay also be used to control air flow rates in combinationwith O 2 monitoring. Careful attention to furnaceoperation is required to ensure an optimum level of performance.Figures 5.3, 5.4, and 5.5 can be used to determinethe combustion efficiency of a boiler or other fired systemburning natural gas, No. 2 fuel oil, or No. 6 fuel oil respectivelyso long as the level of unburned combustibles isconsidered negligible. These figures were derived from H.R. Taplin, Jr., Combustion Efficiency Tables, Fairmont Press,Lilburn, GA, 1991. For more information on combustionefficiency including combustion efficiencies using otherfuels, see Taplin 1991.Where to Look for Conservation OpportunitiesFossil-fuel-fired steam generators, process firedheaters/furnaces, duct heaters, and separately fired superheatersmay benefit from an excess-air-control program.Specialized process equipment, such as rotarykilns, fired calciners, and so on, can also benefit from anair control program.How to Test for Relative EfficiencyTo determine relative operating efficiency and to establishenergy conservation benefits for an excess-air-controlprogram, you must determine: (1) percent oxygen (byvolume) in the flue gas (typically dry), (2) stack temperaturerise (the difference between the flue gas temperatureand the combustion air inlet temperature), and (3) fueltype.To accomplish optimal control over avoidable losses,continuous measurement of the excess air is a necessity.There are two types of equipment available to measureTable 5.2 Typical Optimum Excess Air aOptimum Equivalent O 2Fuel Type Firing Method Excess Air (%) (by Volume)————— ——————— ——————— ——————Natural gas Natural draft 20-30 4-5Natural gas Forced draft 5-10 1-2Natural gas Low excess air .04-0.2 0.1-0.5Propane — 5-10 1-2Coke oven gas — 5-10 1-2No. 2 oil Rotary cup 15-20 3-4No. 2 oil Air-atomized 10-15 2-3No. 2 oil Steam-atomized 10-15 2-3No. 6 oil Steam-atomized 10-15 2-3Coal Pulverized 15-20 3-3.5Coal Stoker 20-30 3.5-5Coal Cyclone 7-15 1.5-3a To maintain safe unit output conditions, excess-air requirements may be greater than theoptimum levels indicated. This condition may arise when operating loads are substantiallyless than the design rating. Where possible, check vendors’ predicted performancecurves. If unavailable, reduce excess-air operation to minimum levels consistent withsatisfactory output.

BOILERS AND FIRED SYSTEMS 93Figure 5.3 Combustion efficiency chart for natural gas.Figure 5.4 Combustion efficiency chart for number 2fuel oil.flue-gas oxygen and corresponding “excess air”: (1) portableequipment such as an Orsat flue-gas analyzer, heatprover, electronic gas analyzer, or equivalent analyzingdevice; and (2) permanent-type installations probe-typecontinuous oxygen analyzers (available from variousmanufacturers), which do not require external gas samplingsystems.The major advantage of permanently mountedequipment is that the on-line indication or recording allowsremedial action to be taken frequently to ensurecontinuous operation at optimum levels. Computerizedsystems which allow safe control of excess air over theboiler load range have proven economic for many installations.Even carbon monoxide-based monitoring andcontrol systems, which are notably more expensive thansimple oxygen-based monitoring and control systems,prove to be cost effective for larger industrial-and utilitysizedboiler systems.Portable equipment only allows performance checkingon an intermittent or spot-check basis. Periodic monitoringmay be sufficient for smaller boilers or boilers whichdo not undergo significant change in operating conditions.However, continuous monitoring and control systems havethe ability to respond more rapidly to changing conditions,such as load and inlet air conditions.The stack temperature rise may be obtained withportable thermocouple probes in conjunction with apotentiometer or by installing permanent temperatureprobes within the exhaust stack and combustion air inletand providing continuous indication or recording. Eachtype of equipment provides satisfactory results for theplanning and operational results desired.An analysis to establish performance can be madewith the two measurements, percent oxygen and thestack temperature rise, in addition to the particular fuelfired. As an illustration, consider the following example.Example: Determine the potential energy savings associatedwith reducing the amount of excess air to an optimumlevel for a natural gas-fired steam boiler.

94 ENERGY MANAGEMENT HANDBOOKB) Enter the chart with an oxygen level of 9% andfollowing a line to the curve, read the percentexcess air to be approximately 66%.C) Continue the line to the curve for a stack temperaturerise of 500°F and read the current combustionefficiency to be 76.4%.STEP 2: Determine the proposed boiler combustionefficiency using the same figure.D) Repeat steps A through C for the proposed combustionefficiency assuming the same stack temperatureconditions. Read the proposed combustionefficiency to be 81.4%.Note that in many cases reducing the amount ofexcess air will tend to reduce the exhaust stacktemperature, resulting in an even more efficientoperating condition. Unfortunately, it is difficultto predict the extent of this added benefit.Figure 5.5 Combustion efficiency chart for number 6fuel oil.Operating Data.Current energy consumption 1,100,000 therms/yrBoiler rated capacity600 boiler horsepowerOperating hours8,500 hr/yrCurrent stack gas analysis9% Oxygen (by volume, dry)Minimal CO readingCombustion air inlet temperature 80°FExhaust gas stack temperature 580°FProposed operating condition 2% Oxygen (by volume, dry)Calculation and Analysis.STEP 1: Determine current boiler combustion efficiencyusing Figure 5.6 for natural gas. Note thatthis is the same figure as Figure 5.3.A) Determine the current stack temperature rise.STR = (exhaust stack temperature)– (combustion air temperature)STR = 580°F - 80°F = 500°FFigure 5.6. Combustion efficiency curve for reducing excessair example.

BOILERS AND FIRED SYSTEMS 95STEP 3: Determine the fuel savings.E) Percent fuel savings = [(new efficiency)– (old efficiency)]/(new efficiency)Percent fuel savings = [(81.4%)– (76.4%)]/(81.4%)Percent fuel savings = 6.14%F) Fuel savings =(current fuel consumption)× (percent fuel savings)Fuel savings = (1,100,000 therms/yr) × (6.14%)Fuel savings = 67,540 therms/yrConclusionsThis example assumes that the results of the combustionanalysis and boiler load are constant. Obviouslythis is an oversimplification of the issue. Because theair-to-fuel ratio (excess air level) is different for differentboiler loads, a more thorough analysis should take thisinto account. One method to accomplish this would be toperform the analysis at various firing rates, such as highfireand low-fire. For modulating type boilers which canvary between high- and low-firing rates, a modified binanalysis approach or other bin-type methodology couldbe employed.Requirements to Effect Maximum EconomyTo obtain the maximum benefits of an excess-aircontrolprogram, the following modifications, additions,checks, or procedures should be considered:Key Elements for Maximum Efficiency1. Ensure that the furnace boundary walls and fluework are airtight and not a source of air infiltrationor exfiltration.a. Recognized leakage problem areas include (1)test connection for oxygen analyzer or portableOrsat connection; (2) access doors and ash-pitdoors; (3) penetration points passing throughfurnace setting; (4) air seals on soot-blower elementsor sight glasses; (5) seals around boilerdrums and header expansion joints; (6) cracks orbreaks in brick settings or refractory; (7) operationof the furnace at too negative a pressure; (8)burner penetration points; and (9) deteriorationof air preheater radial seals or tube-sheet expansionand cracks on tubular air heater applications.b. Tests to locate leakage problems: (1) a light testwhereby a strong spotlight is placed in the furnaceand the unit inspected externally; (2) theuse of a pyrometer to obtain a temperatureprofile on the outer casing. This test generallyindicates points where refractory or insulationhas deteriorated; (3) a soap-bubble test on suspectedpenetration points or seal welds; (4) asmoke-bomb test and an external examinationfor traces of smoke; (5) holding a lighted candlealong the casing seams has pinpointed leakageproblems on induced- or natural-draft units; (6)operating the forced draft fan on high capacitywith the fire out, plus use of liquid chemicalsmoke producers has helped identify seal leaks;and (7) use of a thermographic device to locate“hot spots” which may indicate faulty insulationor flue-gas leakage.2. Ensure optimum burner performance.a. Table 5.3 lists common burner difficulties thatcan be rectified through observation and maintenance.b. Ascertain integrity of air volume control: (1)the physical condition of fan vanes, dampers,and operators should be in optimum workingcondition; and (2) positioning air volume controlsshould be checked for responsiveness andadequacy to maintain optimum air/fuel ratios.Consult operating manual or control manufacturerfor test and calibration.c. Maintain or purchase high-quality gas analyzingsystems: calibrate instrument against a knownflue-gas sample.d. Purchase or update existing combustion controlsto reflect the present state of the art.e. Consider adapting “oxygen trim” feature to existingcombustion control system.3. Establish a maintenance program.a. Table 5.4 presents a summary of frequent boilersystem problems and possible causes.b. Perform period maintenance as recommendedby the manufacturer.c. Keep a boiler operator’s log and monitor keyparameters.d. Perform periodic inspections.Guidelines for Day-to-Day OperationThe following steps must be taken to assure peakboiler efficiency and minimum permissible excess-air operation.1. Check the calibration of the combustion gas analyzerfrequently and check the zero point daily.2. If a sampling system is employed, check to assureproper operation of the sampling system.

96 ENERGY MANAGEMENT HANDBOOK3. The forced-draft damper should be checked for itsphysical condition to ensure that it is not broken ordamaged.4. Casing leakage must be detected and stopped.5. Routinely check control drives and instruments.6. If the combustion gas analyzer is used for monitoringpurposes, the excess air must be checked daily.The control may be manually altered to reduce excessair, without shortcutting the safety of operation.7. The fuel flow and air flow charts should be carefullychecked to ensure that the fuel follows the air onincreasing load with proper safety margin and alsothat the fuel leads the air on decreasing load. Thisshould be compared on a daily shift basis to ensureconsistency of safe and efficient operation.8. Check the burner flame configuration frequentlyduring each shift and note burner register changesin the operator’s log.9. Periodically check flue-gas CO levels to ensure completecombustion. If more than a trace amount of COis present in the flue gas, investigate burner conditionsidentified on Table 5.3 or fuel supply qualitylimits such as fuel-oil viscosity/temperature or coalfineness and temperature.5.3.2 Exhaust Stack TemperatureAnother primary factor affecting unit efficiency andultimately fuel consumption is the temperature of combustiongases rejected to the stack. Increased operating efficiencywith a corresponding reduction in fuel input canbe achieved by rejecting stack gases at the lowest practicaltemperature consistent with basic design principles.In general, the application of additional heat recoveryequipment can realize this energy conservation objectivewhen the measured flue-gas temperature exceedsapproximately 250°F. For a more extensive coverage ofwaste-heat recovery, see Chapter 8.Where to LookSteam boilers, process fired heaters, and other combustionor heat-transfer furnaces can benefit from a heatrecoveryprogram.The adaptation of heat-recovery equipment to existingunits as discussed in this section will be limited toflue gas/liquid and/or flue gas/air preheat exchangers.Specifically, economizers and air preheaters come underthis category. Economizers are used to extract heat energyfrom the flue gas to heat the incoming liquid process feedstreamto the furnace. Flue gas/air preheaters lower theflue-gas temperature by exchanging heat to the incomingcombustion air stream.Planning-quality guidelines will be presented to determinethe final sink temperature, as well as comparativeeconomic benefits to be derived by the installation ofheat-recovery equipment. Costs to implement this energyconservation opportunity can then be compared againstthe potential benefits.Table 5.3 Malfunctions in Fired SystemsFuelMalfunction Coal Oil Gas Detection ActionUneven air distribution x x x Observe flame patterns Adjust registersto burners(trial and error)Uneven fuel distribution x x x Observe fuel pressure Consult manufacturerto burnersgages, or take coalsample and analyzeImproperly positioned x x Observe flame patterns Adjust guns (trialguns or impellersand error)Plugged or worn burners x x Visual inspection Increase frequencyof cleaning; installstrainers (oil)Damaged burner throats x x x Visual inspection Repair

BOILERS AND FIRED SYSTEMS 97Table 5.4 Boiler Performance TroubleshootingSystem Problem Possible CauseHeat transfer related High stack gas temperature Buildup of gas- or water-side depositsImproper water treatment procedureImproper soot blower operationCombustion related High excess air Improper control system operationLow fuel supply pressureChange in fuel heating valueChange in oil viscosityDecrease in inlet air temperatureLow excess airImproper control system operationFan limitationsIncrease in inlet air temperatureHigh carbon monoxide andPlugged gas burnerscombustible emissionsUnbalanced fuel and air distribution inmultiburner furnacesImproper air register settingsDeterioration of burner throat refractoryStoker grate conditionStoker fuel distribution orientationLow fineness on pulverized systemsMiscellaneous Casing leakage Damaged casing and insulationAir heater leakageWorn or improper adjusted seals on rotaryheatersTube corrosionCoal pulverizer powerPulverizer in poor repairToo low classifier settingExcessive blowdownSteam leaksMissing or loose insulationExcessive soot blower operationImproper operationHoles in waterwall tubeValve packingOverheatingWeatheringArbitrary operation schedule that is inexcess of requirements

98 ENERGY MANAGEMENT HANDBOOKHow to Test for Heat-Recovery PotentialIn assessing overall efficiency and potential for heatrecovery, the parameters of significant importance aretemperature and fuel type/sulfur content. To obtain ameaningful operating flue-gas temperature measurementand a basis for heat-recovery selection, the unit underconsideration should be operating at, or very closeto, design and optimum excess-air values as defined onTable 5.2.Temperature measurements may be made bymercury or bimetallic element thermometers, opticalpyrometers, or an appropriate thermocouple probe.The most adaptable device is the thermocouple probein which an iron or chromel constantan thermocoupleis used. Temperature readout is accomplished by connectingthe thermocouple leads to a potentiometer. Theoutput of the potentiometer is a voltage reading whichmay be correlated with the measured temperature forthe particular thermocouple element employed.To obtain a proper and accurate temperature measurement,the following guidelines should be followed:1. Locate the probe in an unobstructed flow path andsufficient distance, approximately five diametersdownstream or upstream, of any major change ofdirection in the flow path.2. Ensure that the probe entrance connection is relativelyleak free.3. Take multiple readings by traversing the cross-sectionalarea of the flue to obtain an average and representativeflue-gas temperature.Modifications or Additions for Maximum EconomyThe installation of economizers and/or flue-gas airpreheaters on units not presently equipped with heat-recoverydevices and those with minimum heat-recoveryequipment are practical ways of reducing stack temperaturewhile recouping flue-gas sensible heat normallyrejected to the stack.There are no “firm” exit-temperature guidelinesthat cover all fuel types and process designs. However,certain guiding principles will provide direction to thelowest practical temperature level of heat rejection. Theelements that must be considered to make this judgmentinclude (1) fuel type, (2) flue-gas dew-point considerations,(3) heat-transfer criteria, (4) type of heat-recoverysurface, and (5) relative economics of heat-recoveryequipment.Tables 5.5 and 5.6 may be used for selecting the lowestpractical exit-gas temperature achievable with installationof economizers and/or flue-gas air preheaters.As an illustration of the potential and methodologyfor recouping flue-gas sensible heat by the additionof heat-recovery equipment, consider the following example.Example: Determine the energy savings associated withinstalling an economizer or flue-gas air preheater onthe boiler from the previous example. Assume that theexcess-air control system from the previous example hasalready been implemented.Available DataCurrent energy consumptionBoiler rated capacityOperating hoursExhaust stack gas analysis1,032,460 therms/yr600 boiler horsepower8,500 hr/yr2% Oxygen (by volume, dry)Minimal CO readingCurrent operating conditions:Combustion air inlet temperature 80°FExhaust gas stack temperature 580°FFeedwater temperature 180°FOperating steam pressure110 psiaOperating steam temperature 335°FProposed operating condition:Combustion air inlet temperature 80°FExhaust gas stack temperature 380°FCalculation and AnalysisSTEP 1: Compare proposed stack temperatureagainst minimum desired stack temperature.A) Heat transfer criteria:T g = T1 + 100°F (minimum)T g = 180 + 100°F (minimum)T g = 280°F (minimum)B) Flue-gas dew point:T g = 120°F (from Figure 5.8)C) Proposed stack temperatureT g = 380°F is acceptableSTEP 2: Determine current boiler combustion efficiencyusing Figure 5.7 for natural gas. Note thatthis is the same figure as Figure 5.3.A) Determine the stack temperature rise.STR = (exhaust stack temperature)– (combustion air temperature)STR = 580°F - 80°F = 500°FB) Enter the chart with an oxygen level of 2% andfollowing a line to the curve, read the percentexcess air to be approximately 9.3%.

BOILERS AND FIRED SYSTEMS 99Table 5.5 EconomizersFuel TypeGaseous fuel(minimum percent sulphur)Test for Determination of ExitFlue-Gas TemperaturesHeat-transfer criteria:T g = T 1 + 100°F (minimum): typically the higherof (a) or (b) below.Fuel oils and coal(a) Heat-transfer criteria:T g = T 1 + 100°F (min.)(b) Flue-gas dew point(from Figure 5.8 for a particular fuel andpercent sulphur by weightWhere:T g = Final stack flue temperatureT 1 = Process liquid feed temperatureTable 5.6 Flue-Gas/Air PreheatersFuel TypeGaseous fuelTest for Determination of ExitFlue-Gas TemperaturesHistoric economic breakpoint:T g (min.) = approximately 250°FFuel oils and coalAverage cold-end considerations;see Figure 5.9 for determination of T ce;the exit-gas temperature relationship is T g = 2T ce – T aWhere:T g = Final stack flue temperatureT ce = Flue gas air preheater recommended average cold end temperatureT a= Ambient air temperatureC) Continue the line to the curve for a stack temperaturerise of 500°F and read the current combustionefficiency to be 81.4%.STEP 3: Determine the proposed boiler combustionefficiency using the same figure.D) Repeat steps A through C for the proposed combustionefficiency assuming the new exhauststack temperature conditions. Read the proposedcombustion efficiency to be 85.0%.STEP 4: Determine the fuel savings.E) Percent fuel savings = [(new efficiency)– (old efficiency)]/(new efficiency)Percent fuel savings = [(85.0%) - (81.4%)]/(85.0%)Percent fuel savings = 4.24%F) Fuel savings =(current fuel consumption)× (percent fuel savings)Fuel savings = (1,032,460 therms/yr) × (4.24%)Fuel savings = 43,776 therms/yrConclusionAs with the earlier example, this analysis methodologyassumes that the results of the combustion analysisand boiler load are constant. Obviously this is an oversimplificationof the issue. Because the air-to-fuel ratio(excess air level) is different for different boiler loads, amore thorough analysis should take this into account.

100 ENERGY MANAGEMENT HANDBOOKFigure 5.8 Flue-gas dew point. Based on unit operationat or close to “optimal” excess-air.Figure 5.7 Combustion efficiency curve for stack temperaturereduction example.Additional considerations in flue-gas heat recovery include:1. Space availability to accommodate additionalheating surface within furnace boundary walls oradjacent area to stack.2. Adequacy of forced-draft and/or induced-draftfan capacity to overcome increased resistance ofheat-recovery equipment.3. Adaptability of soot blowers for maintenance ofheat-transfer-surface cleanliness when firing ashandsoot-forming fuels.4. Design considerations to maintain average coldendtemperatures for flue gas/air preheater applicationsin cold ambient surroundings.5. Modifications required of flue and duct work andadditional insulation needs.Figure 5.9 Guide for selecting flue-gas air preheaters.6. The addition of structural steel supports.7. Adequate pumping head to overcome increasedfluid pressure drop for economizer applications.8. The need for bypass arrangements around economizersor air preheaters.

BOILERS AND FIRED SYSTEMS 1019. Corrosive properties of gas, which would requirespecial materials.10. Direct flame impingement on recovery equipment.Guidelines for Day-to-Day Operation1. Maintain operation at goal excess air levels andstack temperature to obtain maximum efficiencyand unit thermal performance.2. Log percent O 2 or equivalent excess air, inlet airtemperature, and stack temperatures, once pershift or more frequent, noting the unit load andfuel fired.3. Use oxygen analyzers with recorders for unitslarger than about 35 × 10 6 Btu/hr output.4. Maintain surface cleanliness by soot blowing atleast once per shift for ash- and soot-forming fuels.5. Establish a more frequent cleaning schedule whenheat-exchange performance deteriorates due tofiring particularly troublesome fuels.6. External fouling can also cause high excess airoperation and higher stack temperatures thannormal to achieve desired unit outputs. Externalfouling can be detected by use of draft loss gaugesor water manometers and periodically (once aweek) logging the results.7. For flue gas/air preheaters, oxygen checks shouldbe taken once a month before and after the heatingsurface to assess condition of circumferentialand radial seals. If O 2 between the two readingsvaries in excess of 1% O 2 , air heater leakage isexcessive to the detriment of operating efficiencyand fan horsepower.8. Check fan damper operation weekly. Adjust fandamper or operator to correspond to desired excessair levels.9. Institute daily checks on continuous monitoringequipment measuring flue-gas conditions. Checkcalibration every other week.10. Establish an experience guideline on optimumtime for cleaning and changing oil guns and tips.11. Receive the “as-fired” fuel analysis on a monthlybasis from the supplier. The fuel base may havechanged, dictating a different operating regimen.12. Analyze boiler blowdown every two months foriron. Internal surface cleanliness is as importantto maintaining heat-transfer characteristics andperformance as external surface cleanliness.13. When possible, a sample of coal, both raw andpulverized, should be analyzed to determine ifoperating changes are warranted and if the designcoal fineness is being obtained.5.3.3 Waste-Heat-Steam GenerationPlants that have fired heaters and/or low-residencetimeprocess furnaces of the type designed during the eraof cheap energy may have potentially significant energysavingopportunities. This section explores an approachto maximize energy efficiency and provide an analysis todetermine overall project viability.The major problem on older units is to determine apractical and economical approach to utilize the sensibleheat in the exhaust flue gas. Typically, many vintage unitshave exhaust-flue-gas temperatures in the range 1050 to1600°F. In this temperature range, a conventional flue-gasair preheater normally is not a practical approach becauseof materials of construction requirements and significantburner front modifications. Additionally, equippingthese units with an air preheater could materially alterthe inherent radiant characteristics of the furnace, thusadversely affecting process heat transfer. An alternativeapproach to utilizing the available flue-gas sensible heatand maximizing overall plant energy efficiency is to consider:(1) waste-heat-steam generation: (2) installing anunfired or supplementary fired recirculating hot-oil loopor ethylene glycol loop to effectively utilize transferredheat to a remote location: and (3) installing a process feedeconomizer.Because most industrial process industries have aneed for steam, the example is for the application of anunfired waste-heat-steam generator.The hypothetical plant situation is a reformer furnaceinstalled in the plant in 1963 at a time when it wasnot considered economical to install a waste-heat-steamgenerator. As a result, the furnace currently vents hot fluegas (1562°F) to the atmosphere after inspiriting ambientair to reduce the exhaust temperature so that standardmaterials of construction could be utilized.The flue-gas temperature of 1562°F is predicatedon a measured value by thermocouple and is basedon a typical average daily process load on the furnace.This induced-draft furnace fires a No. 2 fuel oil and hasbeen optimized for 20% excess air operation. Flue-gasflow is calculated at 32,800 lb/hr. The plant utilizes approximately180,000 lb/hr of 300-psig saturated steam

102 ENERGY MANAGEMENT HANDBOOKfrom three boilers each having a nameplate capacity of75,000 lb/hr. The plant steam load is shared equally bythe three operating boilers, each supplying 60,000 lb/hr.Feedwater to the units is supplied at 220°F from a commonwater-treating facility. The boilers are fired withlow-sulfur (0.1% sulphur by weight) No. 2 fuel oil. Boilerefficiency averages 85% at load. Present fuel costs are$0.76/gal or $5.48/10 6 Btu basis of No. 2 fuel oil havinga heating value of 138,800 Btu/gal. The basic approachto enhancing plant energy efficiency and minimizingcost is to generate maximum quantities of “waste” heatsteam by recouping the sensible heat from the furnaceexhaust flue gas.Certain guidelines would provide a “fix” on theamount of steam that could be reasonably generated. Theflue-gas temperature drop could practically be reducedto 65 to 100°F above the boiler feedwater temperatureof 220°F. Using an approach temperature of 65°F yieldsan exit-flue gas temperature of 220 + 65 = 285°F. This assumesthat an economizer would be furnished integralwith the waste-heat-steam generator.A heat balance on the flue-gas side (basis of fluegastemperature drop) would provide the total heat dutyavailable for steam generation. The sensible heat contentof the flue gas is derived from Figures 5.10a and 5.10bbased on the flue-gas temperature and percent moisturein the flue gas.Percentage moisture (by weight) in the flue gas is afunction of the type of fuel fired and percentage excessairoperation. Typical values of percentage moisture areindicated in Table 5.7 for various fuels and excess air. ForNo. 2 fuel oil firing at 20% excess air, percent moisture byweight in flue gas is approximately 6.8%.Therefore, a flue-gas heat balance becomesFlue-Gas TemperatureDrop (°F)Sensible Heat in FlueGas (Btu/lb W.G.)1562 412 (Fig. 5.15)285 52 (Fig. 5.14)1277 360Figure 5.10a Heat in flue gases vs. percent moisture by weight. (Derived from Keenan and Kayes 1948.)

BOILERS AND FIRED SYSTEMS 103Table 5.7 Percent Moisture by Weight in Flue GasPercent Excess Air———————————————Fuel Type 10 15 20 25Natural gas 12.1 11.7 11.2 10.8No. 2 fuel oil 7.3 7.0 6.8 6.6Coal (varies) 6.7-5.1 6.4-4.9 6.3-4.7 6.1-4.6Propane 10.1 9.7 9.4 9.1The total heat available from the flue gas for steamgeneration becomes(32,800 lb.W.G.) × (360 Btu/lb.W.G.) = (11.8 × 10 6 Btu/h)The amount of steam that may be generated is determinedby a thermodynamic heat balance on the steamcircuit.Enthalpy of steam at 300 psig saturatedh 3 = 1203 Btu/lbEnthalpy of saturated liquid at drum pressure of 300psigh f = 400 Btu/lbEnthalpy corresponding to feedwater temperatureof 200°Fh 1 = 188 Btu/lbFor this example, assume that boiler blowdown is10% of steam flow. Therefore, feedwater flow throughthe economizer to the boiler drum will be 1.10 times thesteam outflow from the boiler drum. Let the steam outflowbe designated as x. Equating heat absorbed by thewaste-heat-steam generator to the heat available fromreducing the flue-gas temperature from 1562°F to 285°Fyields the following steam flow:(1.10)(x)(h f –h 1 ) + (x)(h 3 –h f ) = 11.8 × 10 6 Btu/hrTherefore,steam flow, xfeedwater flowboiler blowdown= 11,388 lb/hr= 1.10(x)= 1.10(11,388)= 12,527 lb/hr= 12,527 – 11,388 = 1,139 lb/hrFigure 5.10a Heat in flue gases vs. percent moisture by weight. (Derived from Keenan and Kayes 1948.)

104 ENERGY MANAGEMENT HANDBOOKDetermine the equivalent fuel input in conventionalfuel-fired boilers corresponding to the waste heat-steamgenerator capability. This would be defined as follows:Fuel input to conventional boilers= (output)/(boiler efficiency)Therefore,Fuel input= (11.8 × 10 6 Btu/h)/(0.85)= 13.88 × 10 6 Btu/hThis suggests that with the installation of the wasteheat-steamgenerator utilizing the sensible heat of thereformer furnace flue gas, the equivalent of 13.88 × 10 6Btu/hr of fossil-fuel input energy could be saved in thefiring of the conventional boilers while still satisfying theoverall plant steam demand.As with other capital projects, the waste-heat-steamgenerator must compete for capital, and to be viable, itmust be profitable. Therefore, the decision to proceed becomesan economical one. For a project to be consideredlife-cycle cost effective it must have a net-present valuegreater than or equal to zero, or an internal rate of returngreater than the company’s hurdle rate. For a thoroughcoverage of economic analysis, see Chapter 4.5.3.4 Load BalancingEnergy Conservation OpportunitiesThere is an inherent variation in the energy conversionefficiencies of boilers and their auxiliaries with theoperating load imposed on this equipment. It is desirable,therefore, to operate each piece of equipment at the capacitythat corresponds to its highest efficiency.Process plants generally contain multiple boilerunits served by common feedwater and condensate returnfacilities. The constraints imposed by load variationsand the requirement of having excess capacity on line toprovide reliability seldom permit operation of each pieceof equipment at optimum conditions. The energy conservationopportunities therefore lie in the establishmentof an operating regimen which comes closest to attainingthis goal for the overall system in light of operationalconstraints.How to Test for Energy Conservation PotentialInformation needed to determine energy conservationopportunities through load-balancing techniques requiresa plant survey to determine (1) total steam demandand duration at various process throughputs (profile ofsteam load versus runtime), and (2) equipment efficiencycharacteristics (profile of efficiency versus load).Steam DemandChart recorders are the best source for this information.Individual boiler steam flowmeters can be totalizedfor plant output. Demands causing peaks and valleysshould be identified and their frequency estimated.Equipment Efficiency CharacteristicsThe efficiency of each boiler should be documentedat a minimum of four load points between half and maximumload. A fairly accurate method of obtaining unitefficiencies is by measuring stack temperature rise andpercent O 2 (or excess air) in the flue gas or by the input/output method defined in the ASME power test codes.Unit efficiencies can be determined with the aid of Figure5.3, 5.4, or 5.5 for the particular fuel fired. For pump(s)and fan(s) efficiencies, the reader should consult manufacturers’performance curves.An example of the technique for optimizing boilerloading follows.Example: A plant has a total installed steam-generatingcapacity of 500,000 lb/hr, and is served by three boilershaving a maximum continuous rating of 200,000, 200,000,and 100,000 lb/hr, respectively. Each unit can deliver superheatedsteam at 620 psig and 700°F with feedwatersupplied at 250°F. The fuel fired is natural gas priced intothe operation at $3.50/10 6 Btu. Total plant steam averages345,000 lb/hr and is relatively constant.The boilers are normally operated according to thefollowing loading (top of following page).Analysis. Determine the savings obtainable with optimumsteam plant load-balancing conditions.STEP 1. Begin with approach (a) or (b).a) Establish the characteristics of the boiler(s) overthe load range suggested through the use of aconsultant and translate the results graphicallyas in Figures 5.11 and 5.12.b) The plant determines boiler efficiencies for eachunit at four load points by measuring unit stacktemperature rise and percent O 2 in the flue gas.With these parameters known, efficiencies areobtained from Figures 5.3, 5.4, or 5.5. Tabulatethe results and graphically plot unit efficienciesand unit heat inputs as a function of steam load.The results of such an analysis are shown in thetabulation and graphically illustrated in Figures5.11 and 5.12.(Unit input) = (unit output)/(efficiency)

BOILERS AND FIRED SYSTEMS 105Normal——Measured——Boiler Size Boiler Boiler Load Stack Temp. O2 Unit Eff.No. (103 lb/hr) (103 lb/hr) (°F) (%) (%)————————————————————————————————————————1 200 140 290 5 85.02 200 140 540 6 77.43 100 65 540 7 76.5Plant steam demand 345Figure 5.11 Unit efficiency vs. steam load.Figure 5.12 Unit input vs. steam load.

106 ENERGY MANAGEMENT HANDBOOKSTEP 2. Sum up the total unit(s) heat input at thepresent normal operating steam plant load conditions.From Figure 5.12:Boiler Steam Load Heat InputNo.———(10 3 lb/hr)——————(10 6 Btu/hr)——————1 140 1862 140 2043 65——96——Plant totals 345 486STEP 3. Optimum steam plant load-balancing conditionsare satisfied when the total plant steam demandis met according to Table 5.8.(Boiler No. 1 input) + (Boiler No. 2 input) + (Boiler No. 3input) +... = minimumBy trial and error and with the use of Figure 5.12, optimumplant heat input is:Boiler Steam Load Heat InputNo.———(10 3 lb/hr)——————(10 6 Btu/hr)——————1 173 2262 172 2503 (Banked standby)——————————476Plant totals 345STEP 4. The annual fuel savings realized from optimumload balancing is the difference between theexisting boiler input and the optimum boiler input.Steam plant energy savings= (existing input) – (optimum input)= 486 - 476 × 10 6 Btu/hr= 10 × 10 6 Btu/hror annually:= (10 × 10 6 Btu/hr) × (8500 hr/yr)× ($3.50/10 6 Btu)= $297,500/yrCosts that were not considered in the preceding exampleare the additional energy savings due to more efficientfan operation and the cost of maintaining the thirdboiler in banked standby.The cost savings were possible in this example becausethe plant had been maintaining a high ratio of totalcapacity in service to actual steam demand. This results inlow-load inefficient operation of the boilers. Other operatingmodes which generally result in inefficient energyusage are:1. Base-loading boilers at full capacity. This can resultin operation of the base-loaded boilers and theswing boilers at less than optimum efficiency unnecessarily.2. Operation of high-pressure boilers to supply lowpressuresteam demands directly via letdownsteam.Table 5.8 Unit Efficiency and Input TabulationStack Measured CombustionBoiler Steam Load Temperature Oxygen Efficiency Output Fuel InputNo. (10 3 lb/hr) (°F) (%) (%) (10 6 Btu/hr) (10 6 Btu/hr)1 200 305 2 85.0 226.2 266.1170 280 2 86.0 192.3 223.6130 300 7 84.0 147.0 175.0100 280 12 81.5 113.1 138.82 200 625 2 77.5 226.2 291.0170 570 4 78.0 192.3 246.5130 520 7 77.0 147.0 190.9100 490 11 74.0 113.1 152.83 100 600 2 78.0 113.1 145.085 570 2 78.5 96.1 122.565 540 7 76.5 73.5 96.150 500 11 73.5 56.6 76.9

BOILERS AND FIRED SYSTEMS 1073. Operation of an excessive number of auxiliarypumps. This results in throttled, inefficient operation.Requirements for Maximum EconomyEstablish a Boiler Loading Schedule. An optimizedloading schedule will allow any plant steam demand tobe met with the minimum energy input. Some generalpoints to consider when establishing such a schedule areas follows:1. Boilers generally operate most efficiently at 65 to85% full-load rating; centrifugal fans at 80 to 90%design rating. Equipment efficiencies fall off athigher or lower load points, with the decrease mostpronounced at low-load conditions.2. It is usually more efficient to operate a lesser numberof boilers at higher loads than a larger numberat low loads.3. Boilers should be put into service in order of decreasingefficiency starting with the most efficientunit.4. Newer units and units with higher capacity are generallymore efficient than are older, smaller units.5. Generally, steam plant load swings should be takenin the smallest and least efficient unit.Optimize the Use of High-Pressure Boilers. Theboilers in a plant that operate at the highest pressure areusually the most efficient. It is, therefore, desirable to supplyas much of the plant demand as possible with theseunits provided that the high-grade energy in the steamcan be effectively used. This is most efficiently done byinstallation of back-pressure turbines providing usefulwork output, while providing the exhaust steam for lowpressureconsumers.Degrading high-pressure steam through a pressurereducing and desuperheating station is the least efficientmethod of supplying low-pressure steam demands. Directgeneration at the required pressure is usually moreefficient by comparison.Establish an Auxiliary Loading Schedule. A schedulefor cutting plant auxiliaries common to all boilers inand out of service with rising or falling plant load shouldbe established.Establish Procedures for Maintaining Boilers inStandby Mode. It is generally more economical to runfewer boilers at a higher rating. On the other hand, theintegrity of the steam supply must be maintained in theface of forced outage of one of the operating boilers. Bothconditions can sometimes be satisfied by maintaining astandby boiler in a “live bank” mode. In this mode theboiler is isolated from the steam system at no load butkept at system operating pressure. The boiler is kept ata pressure by intermittent firing of either the ignitors ora main burner to replace ambient heat losses. Guidelinesfor live banking of boilers are as follows:1. Shut all dampers and registers to minimize heatlosses from the unit.2. Establish and follow strict safety procedures for ignitor/burnerlight-off.3. For units supplying turbines, take measures to ensurethat any condensate which has been formedduring banking is not carried through to the turbines.Units with pendant-type superheaters willgenerally form condensate in these elements.Operators should familiarize themselves with emergencystartup procedures and it should be ascertainedthat the system pressure decay which will be experiencedwhile bringing the banked boiler(s) up to load can be tolerated.Guidelines for Day-to-Day Operation1. Monitor all boiler efficiencies continuously and immediatelycorrect items that detract from performance.Computerized load balancing may provebeneficial.2. Ensure that load-balancing schedules are followed.3. Reassess the boiler loading schedule whenever amajor change in the system occurs, such as an increaseor decrease in steam demand, derating ofboilers, addition/decommissioning of boilers, oraddition/removal of heat-recovery equipment.4. Recheck parameters and validity of established operatingmode.5. Measure and record fuel usage and correlate tosteam production and flue-gas analysis for determinationof the unit heat input relationship.6. Keep all monitoring instrumentation calibrated andfunctioning properly.7. Optimize excess air operation and minimize boilerblowdown.

108 ENERGY MANAGEMENT HANDBOOKComputerized Systems AvailableThere are commercially available direct digital controlsystems and proprietary sensor devices which accomplishoptimal steam/power plant operation, includingtie-line purchased power control. These systems controlindividual boilers to minimum excess air, SO 2 , NO x , CO(and opacity if desired), and control boiler and cogenerationcomplexes to reduce and optimize fuel input.Boiler plant optimization is realized by boiler controlswhich ensure that the plant’s steam demands aremet in the most cost-effective manner, continuously recognizingboiler efficiencies that differ with time, load, andfuel quality. Similarly, computer control of cogenerationequipment can be cost effective in satisfying plant electricaland process steam demands.As with power boiler systems, the efficiencies forelectrical generation and extraction steam generation canbe determined continuously and, as demand changesoccur, loading for optimum overall efficiency is determined.Fully integrated computer systems can also provideelectric tie-line control, whereby the utility tie-line load iscontinuously monitored and controlled within the electricalcontract’s limits. For example, loads above the peakdemand can automatically be avoided by increasing inplantpower generation, or in the event that the turbinesare at full capacity, shedding loads based on previouslyestablished priorities.5.3.5 Boiler BlowdownIn the generation of steam, most water impurities arenot evaporated with the steam and thus concentrate in theboiler water. The concentration of the impurities is usuallyregulated by the adjustment of the continuous blowdownvalve, which controls the amount of water (and concentratedimpurities) purged from the steam drum.When the amount of blowdown is not properly establishedand/or maintained, either of the following mayhappen:1. If too little blowdown, sludge deposits and carryoverwill result.2. If too much blowdown, excessive hot water is removed,resulting in increased boiler fuel requirements,boiler feedwater requirements, and boilerchemical requirements.Significant energy savings may be realized by utilizingthe guides presented in this section for (1) establishingoptimum blowdown levels to maintain acceptable boilerwaterquality and to minimize hot-water losses, and (2)the recovery of heat from the hot-water blowdown.Where to Look For Energy-Saving OpportunitiesThe continuous blowdown from any steam-generatingequipment has the potential for energy savingswhether it is a fired boiler or waste-heat-steam generator.The following items should be carefully considered tomaximize savings:1. Reduce blowdown (BD) by adjustment of the blowdownvalve such that the controlling water impurityis held at the maximum allowable level.2. Maintain blowdown continuously at the minimumacceptable level. This may be achieved by frequentmanual adjustments or by the installation of automaticblowdown controls. At current fuel costs, automaticblowdown controls often prove to be economical.3. Minimize the amount of blowdown required by:a. Recovering more clean condensate, which reducesthe concentration of impurities cominginto the boiler.b. Establishing a higher allowable drum solidslevel than is currently recommended by ABMAstandards (see below). This must be done onlyon recommendation from a reputable watertreatment consultant and must be followed upwith lab tests for steam purity.c. Selecting the raw-water treatment system whichhas the largest effect on reducing makeup waterimpurities. This is generally considered applicableonly to grass-roots or revamp projects.4. Recover heat from the hot blowdown water. Thisis typically accomplished by flashing the water toa low pressure. This produces low-pressure steam(for utilization in an existing steam header) and hotwater which may be used to preheat boiler makeupwater.Tests and EvaluationsSTEP 1: Determine Actual Blowdown. Obtain the followingdata:T = ppm of impurities in the makeupwater to the deaerator from thetreatment plant; obtain averagevalue through lab testsB = ppm of concentrated impurities inthe boiler drum water (blowdownwater); obtain average valuethrough lab tests

BOILERS AND FIRED SYSTEMS 109lb/hr MU = lb/hr of makeup water to thedeaerator from the water treatmentplant; obtain from flow indicatorlb/hr BFW = lb/hr of boiler feedwater to eachlb/hr STM = lb/hr of steam output from eachboiler; obtain from flow indicatorlb/hr CR = lb/hr of condensate returnNote: percentages for BFW, MU, and CR are determinedas a percentage of STM.Calculate the following:%MU= lb/hr MU × 100%/(total lb/hr BFW)= lb/hr MU × 100%/[(boiler no. 1 lb/hrBFW) + (boiler no. 2 lb/hr BFW) +...]%MU = 100% – %CR (5.3)A = ppm of impurity in BFW = T × %MU (5.4)Now actual blowdown (BD) may be calculated as afunction (percentage) of steam output:%BD = (A × 100%)/(B-A) (5.5)Converting to lb/hr BD yieldslb/hr BD = % BD × lb/hr STM (5.6)Note: In using all curves presented in this section.Blowdown must be based on steam output from theboiler as calculated above. Boiler blowdown basedon boiler feedwater rate (percent BD BFW) to theboiler should not be used. If blowdown is reportedas a percent of the boiler feedwater rate, it may beconverted to a percent of steam output using%BD = %BD BFW × (1)/(1 - %BD BFW ) (5.7)STEP 2: Determine Required Blowdown. The amountof blowdown required for satisfactory boiler operationis normally based on allowable limits for waterimpurities as established by the American BoilerManufacturers Association (ABMA).These limits are presented in Table 5.9. Modificationsto these limits are possible as discussedbelow. The required blowdown may be calculatedusing the equations presented above by substitutingthe ABMA limit for B (concentration of impurity inboiler).% BD required = (A)/(B required - A) × 100% (5.8)lb/hr BD required = % BD required × lb/hr STM (5.9)STEP 3: Evaluate the Cost of Excess Blowdown. Theamount of actual boiler blowdown (as calculatedin equation 5.4) that is in excess of the amount ofrequired blowdown (as calculated in equation 5.6)is considered as wasting energy since this water hasalready been heated to the saturation temperaturecorresponding to the boiler drum pressure. Thecurves presented in Figure 5.13 provide an easymethod of evaluating the cost of excess blowdownas a function of various fuel costs and boiler efficiencies.As an illustration of the cost of boiler blowdown,consider the following example.Table 5.9 Recommended Limits for Boiler-Water ConcentrationsDrum Total Solids Alkalinity Suspended Solids SilicaPressure(psig) ABMA Possible ABMA Possible ABMA Possible ABMA0 to 300 3500 6000 700 1000 300 250 125301 to 450 3000 5000 600 900 250 200 90451 to 600 2500 4000 500 500 150 100 50601 to 750 2000 2500 400 400 100 50 35751 to 900 1500 — 300 300 60 — 20901 to 1000 1250 — 250 250 40 — 81001 to 1500 1000 — 200 200 20 — 2

110 ENERGY MANAGEMENT HANDBOOKExample: Determine the potential energy savingsassociated with reducing boiler blowdown from12% to 10% using Figure 5.13.Operating DataAverage boiler load75,000 lb/hrSteam pressure150 psigMake up water temperature 60°FOperating hours8,200 hr/yrBoiler efficiency 80%Average fuel cost$2.00/10 6 BtuCalculation and AnalysisUsing the curves in Figure 5.13, enter Chart Aat 10% blowdown to the curve for 150 psig boilerdrum pressure. Follow the line over to chart B andthe curve for a unit efficiency of 80%. Then followthe line down to Chart C and the curve for a fuel costof $2.00/10 6 Btu. Read the scale for the equivalentfuel value in blowdown. The cost of the blowdownis estimated at $8.00/hr per 100,000 lb/hr of steamgenerated. Repeat the procedure for the blowdownrate of 12% and find the cost of the blowdown is$10.00/hr per 100,000 lb/hr of steam generated.Potential energy savings then is estimated to be= ($10.00 - 8.00/hr/100,000 lb/hr)× (75,000 lb/hr) × (8,200 hr/yr)= $12,300/yrEnergy Conservation Methods1. Minimize Blowdown by Manual Adjustment. Thisis accomplished by establishing an operating procedurerequiring frequent water quality testing andreadjustment of blowdown valves so that water impuritiesin the boiler are held at the allowable limit.Continuous indicating/recording analyzers may beemployed allowing the operator to establish quicklythe actual level of water impurity and manually readjustblowdown valves.2. Minimize Blowdown by Automatic Adjustment.The adjustment of blowdown may be automated bythe installation of automatic analyzing equipmentand the replacement of manual blowdown valveswith control valves (see Figure 5.14). The cost ofthis equipment is frequently justifiable, particularlyFigure 5.13 Hourly cost of blowdown.

BOILERS AND FIRED SYSTEMS 111when there are frequent load changes on the steamgeneratingequipment since the automation allowscontinuous maintenance of the highest allowablelevel of water contaminants. Literature has approximatedthat the average boiler plant can save about20% blowdown by changing from manual control toautomatic adjustment.3. Decrease Blowdown by Recovering More Condensate.Since clean condensate may be assumed tobe essentially free of water impurities, addition ofcondensate to the makeup water serves to dilute theconcentration of impurities. The change in requiredblowdown may be calculated using equations 5.3and 5.5.Example: Determine the effect on boiler blowdown of increasingthe rate of condensate return from 50 to 75%:Operating DataMU water impurities = 10 ppm = TMaximum allowable limit in drum= 100 ppm = BCalculate:For 50% return condensate:MU = 100% – 50% = 50%A = 10 × 0.5 = 5 ppm in BFW% BD = A/(B-A) = 5/(100-5) = 5.3%For 75% return condensate:MU = 100 – 75% = 25%A = 10 × 0.25 = 2.5 ppm% BD = 2.5/(100-2.5) = 2.6%ConclusionThese values may then be used with the curves onFigure 5.13 to approximate the potential energy savings.4. Increase Allowable Drum Solids Level. In someinstances it may be possible to increase the maximumallowable impurity limit without adverselyaffecting the operation of the steam system. However,it must be emphasized that a water treatmentconsultant should be contacted for recommendationon changes in the limits as given in Table 5.9. Thechanges must also be followed by lab tests for steampurity to verify that the system is operating as anticipated.The energy savings may be evaluated by usingthe foregoing equations for blowdown and thegraphs in Figures 5.13 and 5.15. Consider the followingexample.Example: Determine the blowdown rates as a percentageof steam flow required to maintain boiler drum waterimpurity concentrations at an average of 3000 ppm and6,000 ppm.Operating DataAverage makeup water impurity(measurement) ..........................................350 ppmCondensate return (percent of steam flow) ........25%Assume condensate return free from impuritiesCalculation and AnalysisCalculate the impurity concentration in the boilerfeedwater (BFW):A = MU impurity × (1.00 - % CR)A = 350 ppm × (1.00 - 0.25)A = 262 ppmMathematical solution.For drum water impurity level of 3000 ppm:% BD = A/(B - A)% BD = 262/(3000 - 262)% BD = 9.6%Figure 5.14 Automated continuous blowdown system.For drum water impurity level of 6000 ppm:

112 ENERGY MANAGEMENT HANDBOOKFigure 5.15 Required percent blowdown. Based on equation 5.5.% BD = A/(B - A)% BD = 262/(6000 - 262)% BD = 4.6%Graphical Solution. Referring to Figure 5.15Enter the graph at feedwater impurity level of 262ppm and follow the line to the curves for 3000 ppmand 6000 ppm boiler drum water impurity level.Then read down to the associated boiler blowdownpercentage.ConclusionThe blowdown percentages may not be used in conjunctionwith Figure 5.13 to determine the annualcost of blowdown and the potential energy cost savingsassociated with reducing boiler blowdown.5. Select Raw-Water Treatment System for LargestReduction in Raw-Water Impurities. Since a largeinvestment would be associated with the installationof new equipment, this energy conservation methodis usually applicable to new plants or revamps only.A water treatment consultant should be retained torecommend the type of treatment applicable. Anexample of how water treatment affects blowdownfollows.Example: Determine the effects on blowdown of using asodium zeolite softener producing a water quality of 350ppm solids and of using a demineralization unit producinga water quality of 5 ppm solids. The makeup waterrate is 30% and the allowable drum solids level is 3000ppm.Solution:For sodium zeolite:% BD = (350 × 0.3 × 100%)/[3000- (350 × 0.3)] = 3.6%For demineralization unit:% BD = (5 × 0.3 × 100%)/[3000 - (5 × 0.3)]= 0.6%Therefore, the percentage blowdown would be reducedby 3.0%.A secondary benefit derived from increasing feedwaterquality is the reduced probability of scale formationin the boiler. Internal scale reduces the effectivenessof heat-transfer surfaces and can result in a reduction ofas much as 1 to 2% in boiler efficiency in severe cases.6. Heat Recovery from Blowdown. Since a certainamount of continuous blowdown must be main-

BOILERS AND FIRED SYSTEMS 113tained for satisfactory boiler performance,a significant quantity of heat is removedfrom the boiler. A large amount of theheat in the blowdown is recoverable byusing a two-stage heat-recovery system asshown in Figure 5.16 before dischargingto the sewer. In this system, blowdownlines from each boiler discharge into acommon flash tank. The flashed steammay be tied into an existing header, useddirectly by process, or used in the deaerator.The remaining hot water may be usedto preheat makeup water to the deaeratoror preheat other process streams.The following procedure may be usedto calculate the total amount of heat thatis recoverable using this system and theassociated cost savings.STEP 1. Determine the annual cost ofblowdown using the percent blowdown,steam flow rate (lb/hr), unit efficiency,and fuel cost. This can be accomplishedin conjunction with Figure 5.13.STEP 2. Determine:Flash % = percent of blowdown that is flashed tosteam (using Figure 5.17, curve B, atthe flash tank pressure or using equation5.10a or 5.10b)COND % = 100% - Flash %h tk = enthalpy of liquid leaving the flashtank (using Figure 5.17, curve A, at theflash tank pressure)h ex = enthalpy of liquid leaving the heat exchanger[using Figure 5.17, curve C;for planning purposes, a 30 to 40°F approachtemperature (condensate dischargeto makeup water temperature)may be used]STEP 3. Calculate the amount of heat recoverablefrom the condensate (% QC) usingFigure 5.16 Typical two-stage blowdown heat-recovery system.STEP 5. The annual savings from heat recovery maythen be determined by using this percent (% Q) withthe annual cost of blowdown found in step 1:annual savings = (% QC/100) × BD costTo further illustrate this technique consider the followingexample.Example: Determine the percent of heat recoverable(%Q) from a 150 psig boiler blowdown wastestream, if the stream is sent to a 20 psig flashtank and heat exchanger.Available DataBoiler drum pressure150 psigFlash tank pressure20 psigMakeup water temperature 70°FAssume a 30°F approach temperature between condensatedischarge and makeup water temperature%QC = [(h tk - h ex )/h tk ] × COND %STEP 4. Since all of the heat in the flashed steam isrecoverable, the total percent of heat recoverable (%Q) from the flash tank and heat-exchanger systemis% Q = % QC + Flash %Calculation and AnalysisReferring to Figure 5.17.Determine Flash % using Chart B:Entering chart B with a boiler drum pressure of150 psig and following a live to the curve for aflash tank pressure of 20 psig, read the Steampercentage (Flash %) to be 12.5%.

114 ENERGY MANAGEMENT HANDBOOKDetermine COND %:COND % = 100 - Flash %COND % = 100 - 12.5 %COND % = 87.5%Determine h tk using Chart A:Entering chart A with a flash tank pressure of 20psig and following a line to the curve for saturatedliquid, read the enthalpy of the drum water(h tk ) to be 226 Btu/lb.Determine h ex using Chart C:Assuming a 30°F approach temperature betweencondensate discharge and makeup watertemperature, the temperature of the blowdowndischarge is equal to the makeup water temperatureplus the approach temperature whichequals 100°F (70°F + 30°F).Entering chart C with a blowdown heat exchangerrejection temperature of 100°F and followinga line to the curve, read the enthalpy ofthe blowdown discharge water to be 68 Btu/lb.Determine the % QC:% QC = [(h tk - h ex )/h tk ] × COND %% QC = [(226 Btu/lb - 68 Btu/lb)/226 Btu/lb]× 87.5%% QC = 61.2%Determine the % Q:% Q = % QC + Flash %% Q = 61.2% + 12.5%% Q = 73.7%ConclusionTherefore, approximately 73.7% of the heat energycan be recovered using this blowdown heat recoverytechnique.More on Flash SteamTo determine the amount of flash steam that is generatedby high-pressure, high-temperature condensatebeing reduced to a lower pressure you can use the followingequation:Flash % = (h HPl - h LPl ) × 100%/(h LPv – h LPl )orFlash % = (h HPl - h LPl ) × 100%/(h LPevp )(5.10a)(5.10b)Figure 5.17 Percent of heat recoverable from blowdown.

BOILERS AND FIRED SYSTEMS 115where: Flash % = amount of flash steam as a percent oftotal massh HPl = enthalpy of the high pressure liquidh LPl = enthalpy of the low pressure liquidh LPv = enthalpy of the low pressure vaporh LPevap = evaporation enthalpy of the low pressureliquid = (h LPv – h LPl )Guidelines for Day-to-Day Operation1. Maintain concentration of impurities in the boilerdrum at the highest allowable level. Frequent checksshould be made on water quality and blowdownvalves adjusted accordingly.2. Continuous records of impurity concentration inmakeup water and boiler drum water will indicatetrends in deteriorating water quality so that earlycorrective actions may be taken.3. Control instruments should be calibrated on a weeklybasis.5.3.6 Condensate ReturnIn today’s environment of ever-increasing fuel costs,the return and utilization of the heat available in cleansteam condensate streams can be a practical and economicalenergy conservation opportunity. Refer to Chapter 6for a comprehensive discussion of condensate return. Theinformation below is presented to summarize briefly andemphasize the benefits and major considerations pertinentto optimum steam generator operations. Recognizedbenefits of return condensate include:Reduction in steam power plant raw-water makeup andassociated treatment costs.Reduction in boiler blowdown requirements resulting indirect fuel savings. Refer to section on boiler blowdown.Reduced steam required for boiler feedwater deaeration.Raw-water and boiler-water chemical cost reduction.Opportunities for increased useful work output withoutadditional energy input.Reduces objectionable environmental discharges fromcontaminated streams.Where to LookExamine and survey all steam-consuming unitswithin a plant to determine the present disposition of anycondensate produced or where process modifications canbe made to produce “clean” condensate. Address the following:1. Is the condensate clean and being sewered?2. Is the stream essentially clean but on occasion becomescontaminated?3. If contaminated, can return to the steam system bejustified by polishing the condensate?4. Can raw makeup or treated water be substituted forcondensate presently consumed?5. Is condensate dumped for operating convenience orlack of chemical purity?Results from chemical purity tests, establishing batterylimit conditions and analysis of these factors, providethe basis of obtaining maximum economy.Modifications Required for Maximum EconomyOften, the only requirement to gain the benefits ofreturn condensate is to install the necessary piping and/or pumping facilities. Other solutions are more complexand accordingly, require a more in-depth analysis. Chemically“clean” or “contaminated” condensate can be effectivelyutilized by:Providing single- or multistage flashing for contaminatedstreams, and recouping the energy of the flashedsteam. Recovering additional heat from the flash drumcondensate by indirect heat exchange is also a possibility.Collecting condensate from an atmospheric flashdrum with “automatic” provision to dump on indicationof stream contamination. This concept, when conditionswarrant, allows “normally clean” condensate to be usedwithin the system.Installing ion-exchange polishing units for condensatestreams which may be contaminated but are significantin quantity and heat value.Providing a centrally located collection tank andpump to return the condensate to the steam system. Thisavoids a massive and complex network of individual returnlines.Using raw water in lieu of condensate and returningthe condensate to the system. An example is the use ofcondensate to regenerate water treatment units.Changing barometric condensers or other directcontactheat exchangers to surface type or indirect exchangers,respectively, and returning the clean condensateto the system.Collecting condensate from sources normally overlooked,such as space heating, steam tracing, and steamtraps.Providing flexibility to isolate and sewer individualreturn streams to maintain system integrity. Providing

116 ENERGY MANAGEMENT HANDBOOK“knockout” or disengaging drum(s) to ensure clean condensatereturn to the system.Recovering the heat content from contaminatedcondensate by indirect heat exchangers. An example isusing an exchanger to heat the boiler makeup water.Returning the contaminated condensate stream toa clarifier or hot lime unit to cleanup for boiler makeuprather than sewering the stream.Allowing provision for manual water testing of thecondensate stream suspected of becoming contaminated.Using the contaminated condensate for noncriticalapplications, such as space heating, tank heating, and soon.Guidelines for Day-to-Day Operation1. Maintain the system, including leak detection andinsulation repair.2. Periodically test the return water at its source of entrywithin the steam system for (a) contamination,(b) corrosion, and (c) acceptable purity.3. Maintain and calibrate monitoring and analyzingequipment.4. Ensure that the proper operating regimen is followed;that is, the condensate is returned and notsewered.5.4 FUEL CONSIDERATIONSThe selection and application of fuels to variouscombustors are becoming increasingly complex. Most existingunits have limited flexibility in their ability to firealternative fuels and new units must be carefully plannedto assure the lowest first costs without jeopardizing thefuture capability to switch to a different fuel. This sectionpresents an overview of the important considerations inboiler and fuel selection. Also refer to Section 5.5.5.4.1 Natural GasNatural-gas firing in combustors has traditionallybeen the most attractive fuel type, because:1. Gas costs were low until about 2000 when theystarted to rise dramatically.2. Only limited fuel-handling equipment typicallyconsisting of pipelines, metering, a liquid knockoutdrum, and appropriate controls is required.3. Boiler costs are minimized due to smaller boiler sizes;which result from highly radiant flame characteristicsand higher velocities, resulting in enhancedheat transfer and less heating surface.4. Freedom from capital and operating costs associatedwith pollution control equipment.Natural gas, being the cleanest readily availableconventional form of fuel, also makes gas-fired units theeasiest to operate and maintain.However, as discussed elsewhere, the continued useof natural gas as fuel to most combustors will probablybe limited in the future by government regulations, risingfuel costs, and inadequate supplies. One further disadvantage,which often seems to be overlooked, is the lowerboiler efficiency that results from firing gas, particularlywhen compared to oil or coal.5.4.2 Fuel OilClassificationsInfluential in the storage, handling, and combustionefficiency of a liquid fuel are its physical and chemicalcharacteristics.Fuel oils are graded as No. 1, No. 2, No. 4, No. 5(light), No. 5 (heavy), and No. 6. Distillates are Nos. 1 and2 and residual oils are Nos. 4, 5, and 6. Oils are classifiedaccording to physical characteristics by the American Societyfor Testing and Materials (ASTM) according to StandardD-396.No. 1 oil is used as domestic heating oil and as alight grade of diesel fuel. Kerosene is generally in a lighterclass; however, often both are classified the same. No.2 oil is suitable for industrial use and home heating. Theprimary advantage of using a distillate oil rather than aresidual oil is that it is easier to handle, requiring no heatingto transport and no temperature control to lower theviscosity for proper atomization and combustion. However,there are substantial purchase cost penalties betweenresidual and distillate.It is worth noting that distillates can be divided intotwo classes: straight-run and cracked. A straight-run distillateis produced from crude oil by heating it and thencondensing the vapors. Refining by cracking involveshigher temperatures and pressures or catalysts to producethe required oil from heavier crudes. The differencebetween these two methods is that cracked oils containsubstantially more aromatic and olivinic hydrocarbonswhich are more difficult to burn than the paraffinic andnaphthenic hydrocarbons from the straight-run process.Sometimes a cracked distillate, called industrial No. 2, isused in fuel-burning installations of medium size (smallpackage boiler or ceramic kilns for example) with suitableequipment.

BOILERS AND FIRED SYSTEMS 117Because of the viscosity range permitted by ASTM,No. 4 and No. 5 oil can be produced in a variety of ways:blending of No. 2 and No. 6, mixture of refinery by-products,through utilization of off-specification products, andso on. Because of the potential variations in characteristics,it is important to monitor combustion performanceroutinely to obtain optimum results. Burner modificationsmay be required to switch from, say, a No. 4 that is ablend and a No. 4 that is a distillate.Light (or cold) No. 5 fuel oil and heavy (or hot) aredistinguished primarily by their viscosity ranges: 150 to300 SUS (Saybolt Universal Seconds) at 100°F and 350 to750 SUS at 100°F respectively. The classes normally delineatethe need for preheating with heavy No. 5 requiringsome heating for proper atomization.No. 6 fuel oil is also referred to as residual, BunkerC, reduced bottoms, or vacuum bottoms. It is a veryheavy oil or residue left after most of the light volatileshave been distilled from crude. Because of its high viscosity,900 to 9000 SUS at 100°F, it can only be used insystems designed with heated storage and sufficient temperature/viscosityat the burner for atomization.Heating ValueFuel oil heating content can be expressed as higher(or gross) heating value and low (or net) heating value.The higher heating value (HHV) includes the water contentof the fuel, whereas the lower heating value (LHV)does not. For each gallon of oil burned, approximately 7to 9 lb of water vapor is produced. This vapor, when condensedto 60°F, releases 1058 Btu. Thus the HHV is about1000 Btu/lb or 8500 Btu/gal higher than the LHV. Whilethe LHV is representative of the heat produced duringcombustion, it is seldom used in the United States exceptfor exact combustion calculations.ViscosityViscosity is a measure of the relative flow characteristicsof an oil an important factor in the design andoperation of oil-handling and -burning equipment, theefficiency of pumps, temperature requirements, and pipesizing. Distillates typically have low viscosities and canbe handled and burned with relative ease. However, No.5 and No. 6 oils may have a wide range of viscosities,making design and operation more difficult.Viscosity indicates the time required in secondsfor 60 cm 3 of oil to flow through a standard-size orificeat a specific temperature. Viscosity in the United Statesis normally determined with a Saybolt viscosimeter. TheSaybolt viscosimeter has two variations (Universal andFurol) with the only difference being the size of orificeand sample temperature. The Universal has the smallestopening and is used for lighter oils. When stating an oil’sviscosity, the type of instrument and temperature mustalso be stated.Flash PointFlash point is the temperature at which oil vaporsflash when ignited by an external flame. As heating continuesabove this point, sufficient vapors are driven offto produce continuous combustion. Since flash point isan indication of volatility, it indicates the maximum temperaturefor safe handling. Distillate oils normally haveflash points from 145 to 200°F, whereas the flash point forheavier oils may be up to 250°F. Thus under normal ambientconditions, fuel oils are relatively safe to handle (unlesscontaminated).Pour PointPour point is the lowest temperature at which an oilflows under standard conditions. It is 5°F above the oil’ssolidification temperature. The wax content of the oil significantlyinfluences the pour point (the more wax, thehigher the pour point). Knowledge of an oil’s pour pointwill help determine the need for heated storage, storagetemperature, and the need for pour-point depressant.Also, since the oil may cool while being transferred, burnerpreheat temperatures will be influenced and should bewatched.Sulfur ContentThe sulfur content of an oil is dependent upon thesource of crude oil. Typically, 70 to 80% of the sulfur in acrude oil is concentrated in the fuel product, unless expensivedesulfurization equipment is added to the refiningprocess. Fuel oils normally have sulfur contents offrom 0.3 to 3.0% with distillates at the lower end of therange unless processed from a very high sulfur crude. Often,desulfurized light distillates are blended with highsulfurresidual oil to reduce the residual’s sulfur content.Sulfur content is an important consideration primarily inmeeting environmental regulations.AshDuring combustion, impurities in oil produce a metallicoxide ash in the furnace. Over 25 different metalscan be found in oil ash, the predominant being nickel,iron, calcium, sodium, aluminum, and vanadium. Theseimpurities are concentrated from the source crude oilduring refining and are difficult to remove since they arechemically bound in the oil. Ash contents vary widely:distillates have about 0 to 0.1% ash and heavier oil 0.2 to1.5%. Although percentages are small, continuous boileroperations can result in considerable accumulations of

118 ENERGY MANAGEMENT HANDBOOKash in the firebox.Problems associated with ash include reduction inheat-transfer rates through boiler tubes, fouling of superheaters,accelerated corrosion of boiler tubes, and deteriorationof refractories. Ashes containing sodium, vanadium,and/or nickel are especially troublesome.Other ContaminantsOther fuel-oil contaminants include water, sediment,and sludge. Water in fuel oil comes from condensation,leaks in storage equipment, and/or leaking heatingcoils. Small amounts of water should not cause problems.However, if large concentrations (such as at tank bottoms)are picked up, erratic and inefficient combustion may result.Sediment comes from dirt carried through with thecrude during processing and impurities picked up instorage and transportation. Sediment can cause line andstrainer plugging, control problems, and burned/nozzleplugging. More frequent filter cleaning may be required.Sludge is a mixture of organic compounds that haveprecipitated after different heavy oils are blended. Theseare normally in the form of waxes or asphaltenese, whichcan cause plugging problems.AdditivesFuel-oil additives may be used in boilers to improvecombustion efficiency, inhibit high-temperature corrosion,and minimize cold-end corrosion. In addition, additivesmay be useful in controlling plugging, corrosion,and the formation of deposits in fuel-handling systems.However, caution should be used in establishing the needfor and application of any additive program. Before selectingan additive, clearly identify the problem requiringcorrection and the cause of the problem. In many cases,solutions may be found which would obviate the needand expense of additives. Also, be sure to understandclearly both the benefits and the potential debits of theadditive under consideration.Additives to fuel-handling systems may be warrantedif corrosion problems persist due to water whichcannot be removed mechanically. Additives are alsoavailable which help prevent sludge and/or other depositsfrom accumulating in equipment, which could resultin increased loading due to increased pressure drops onpumps and losses in heat-transfer-equipment efficiencies.Additive vendors claim that excess air can be controlledat lower values when catalysts are used. Althoughthese claims appear to be verifiable, consideration shouldbe given to mechanically controlling O 2 to the lowest possiblelevels. Accurate O 2 measurement and control shouldfirst be implemented and then modifications to burnerassemblies considered. Catalysts, consisting of metallicoxides (typically manganese and barium), have demonstratedthe capability of reducing carbon carryover in theflue gas and thus would permit lower O 2 levels withoutsmoking. Under steady load conditions, savings can beachieved. However, savings may be negligible undervarying loads, which necessitate prevention of fuel-richmixtures by maintaining air levels higher than optimal.Other types of combustion additives are availablewhich may be beneficial to specific boiler operating problems.However, these are not discussed here since theyare specific in nature and are not necessarily related toimproved boiler efficiency. Generally, additives are usedwhen a specific problem exists and when other conventionalsolutions have been exhausted.AtomizationOil-fired burners, kilns, heat-treating furnaces, ovens,process reactors, and process heaters will realize increasedefficiency when fuel oil is effectively atomized.The finer the oil is atomized, the more complete combustionand higher overall combustion efficiency. Obtainingthe optimum degree of atomization depends on maintaininga precise differential between the pressure of theoil and pressure of the atomizing agent normally steamor air. The problem usually encountered is that the steam(or air pressure) remains constant while the oil pressurecan vary substantially. One solution is the addition of adifferential pressure regulator which controls the steampressure so that the differential pressure to the oil is maintained.Other solutions, including similar arrangementsfor air-atomized systems, should be reviewed with equipmentvendors.Fuel-Oil EmulsionsIn general, fuel-oil emulsifier systems are designedto produce an oil/water emulsion that can be combustedin a furnace or boiler.The theory of operation is that micro-size dropletsof water are injected and evenly dispersed throughoutthe oil. As combustion takes place, micro explosions ofthe water droplets take place which produce very fine oildroplets. Thus more surface area of the fuel is exposedwhich then allows for a reduction in excess air level andimproved efficiency. Unburned particles are also reduced.Several types of systems are available. One uses aresonant chamber in which shock waves are started in thefluid, causing the water to cavitate and breakdown intosmall bubbles. Another system produces an emulsion byinjecting water into oil. The primary technical differenceamong the various emulsifier systems currently marketed

BOILERS AND FIRED SYSTEMS 119is the water particle-size distribution that is formed. Themean particle size as well as the particle-size distributionare very important to fuel combustion performance, andthese parameters are the primary reasons for differingoptimal water content for various systems (from 3% toover 200%). Manufacturers typically claim improvementsin boiler efficiency of from 1 to 6% and further savingsdue to reduced particulate emissions which allow burningof higher-sulfur (lower-cost) fuels. However, recentindependent testing seems to indicate that while savingsare achieved, these are often at the low end of the range,particularly for units that are already operating near optimumconditions. Greater savings are more probable fromemulsifiers that are installed on small, inefficient, andpoorly instrumented boilers. Tests have confirmed reductionsin particulate and NO x emissions and the ability tokeep boiler internals cleaner. Flames exhibit characteristicsof more transparency and higher radiance (similar togas flames).Before considering installation of an emulsifier system,existing equipment should be tuned up and put inoptimum condition for maximum efficiency. A qualifiedservice technician may be helpful in obtaining best results.After performance tests are completed, meaningfultests with an emulsified oil can be run. It is suggested thatcase histories of a manufacturer’s equipment be reviewedand an agreement be based on satisfaction within a specifictime frame.5.4.3 CoalThe following factors have a negative impact on theselection of coal as a fuel:1. Environmental limitations which necessitate theinstallation of expensive equipment to control particulates,SO 2 , and NO x . These requirements, whencombined with the low price of oil and gas in thelate 1960s and early 1970s, forced many existing industrialand utility coal-fired units to convert to oilor gas. And new units typically were designed withno coal capability at all.2. Significantly higher capital investments, not onlyfor pollution abatement but also for coal-receivingequipment; raw-coal storage; coal preparation(crushing, conveying, pulverizing, etc.): preparedcoalstorage; and ash handling.3. Space requirements for equipment and coal storage.4. Higher maintenance costs associated with the installationof more equipment.5. Concern over uninterrupted availability of coal resultingfrom strikes.6. Increasing transportation costs.The use of coal is likely to escalate even in light ofthe foregoing factors, owing primarily to its relatively secureavailability both on a short-term and long-term basis,and its lower cost. The substantial operating cost savingsat current fuel prices of coal over oil or gas can economicallyjustify a great portion (if not all) of the significantlyhigher capital investments required for coal. In addition,most predictions of future fuel costs indicate that oil andgas costs will escalate much faster than coal.Enhancing Coal-Firing EfficiencyA potentially significant loss from the combustionof coal fuels is the unburned carbon loss. All coal-firedsteam generators and coal-fired vessels inherently sufferan efficiency debit attributable to unburned carbon. At agiven process output, quantities in excess of acceptablevalues for the particular unit design and coal rank detractgreatly from the unit efficiency, resulting in increased fuelconsumption.It is probable that pulverized coal-fired installationssuffer from an inordinately high unburned carbon losswhen any of the following conditions are experienced:A change in the raw-fuel quality from the original designbasis.Deterioration of the fuel burners, burner throats, or burnerswirl plates or impellers.Increased frequency of soot blowing to maintain heattransfer-surfacecleanliness.Noted increase in stack gas opacity.Sluggish operation of combustion controls of antiquateddesign.Uneven flame patterns characterized by a particularlybright spot in one sector of the flame and a quitedark spot in another.CO formation as determined from a flue-gas analysis.Frequent smoking observed in the combustion zone.Increases in refuse quantities in collection devices.Lack of continued maintenance and/or replacement ofcritical pulverized internals and classifier assembly.High incidence of coal “hang-up” in the distribution pipingto the burners.Frequent manipulation of the air/coal primary and secondaryair registers.How to Test for Relative Operating EfficiencyThe aforementioned items are general symptoms

120 ENERGY MANAGEMENT HANDBOOKthat are suspect in causing a high unburned carbon lossfrom the combustion of coal. However, the magnitude ofthe problem often goes undetected and remedial action isnever taken because of the difficulty in establishing andquantifying this loss while relating it to the overall unitoperating efficiency.In light of this, it is recommended that the boilermanufacturer be consulted to establish the magnitude ofthis operating loss as well as to review the system equipmentand operating methods.The general test procedure to determine the unburnedcarbon loss requires manual sampling of the ash(refuse) in the ash pit, boiler hopper(s), air heater hopper,and dust collector hopper and performing a laboratoryanalysis of the samples. For reference, detailed methods,test procedures, and results are outlined in the ASMEPower Test Codes, publications PTC 3.2 (Solid Fuels) andPTC 4.1 (Steam Generating Units), respectively. In additionto the sampling of the ash, a laboratory analysisshould be performed on a representative raw coal sampleand pulverized coal sample.Results and analysis of such a testing program will:1. Quantify the unburned carbon loss.2. Determine if the unburned carbon loss is high forthe type of coal fired, unit design, and operatingmethods.Figure 5.18 can be used to determine the approximateenergy savings resulting from reducing unburnedcoal fuel loss. The unburned fuel is generally collectedwith the ash in either the boiler ash hopper(s) or in variouscollection devices. The quantity of ash collected atvarious locations is dependent and unique to the systemdesign. The boiler manufacturer will generally specify theproportion of ash normally collected in various ash hoppersfurnished on the boiler proper. The balance of ashand unburned fuel is either collected in flue-gas cleanupdevices or discharged to the atmosphere. A weighted averageof total percentage combustibles in the ash must becomputed to use Figure 5.18. To further illustrate the potentialsavings from reducing combustible losses in coalfiredsystems using Figure 5.18, consider the followingexample.Example: Determine the benefits of reducing the combustiblelosses for a coal-fired steam generator having amaximum continuous rating of 145,000 lbs/hr at an operatingpressure of 210 psig and a temperature of 475°F atthe desuperheater outlet, feedwater is supplied at 250°F.Available DataAverage boiler load125,000 lb/hrSuperheater pressure210 psigSuperheater temperature 475°FFuel typeCoalMeasured flue gas oxygen 3.5%Operating combustibles in ash 40%Obtainable combustibles in ash 5%Yearly operating time8500 hr/yrDesign unit heat output 150 × 10 6 Btu/hrAverage unit heat output 129 × 10 6 Btu/hrAverage fuel cost$1.50/10 6 BtuAnalysis. Referring to Figure 5.18:Chart A.Operating combustibles 40%Obtainable combustibles 5%3. Reveal a reasonable and attainable value for unburnedcarbon loss.Chart B.Design Unit output150 × 10 6 Btu/hr4. Provide guidance for any corrective action.5. Allow the plant, with the aid of Figure 5.18, to assessthe annual loss in dollars by operating with a highunburned carbon loss.6. Suggest an operating mode to reduce the excess airrequired for combustion.Chart C.Average fuel cost$1.50/10 6 BtuAnnual fuel savings $210,000Annual fuel savings corrected to actual operating conditions:Savings $= (curve value × [operating heat output)/(design heat output)] × [actual annual operatinghours)/(8,760 hr/yr)]= ($210,000/yr) × [(129 × 10 6 Btu/hr)/(150 ×10 6 Btu/hr)] × [(8,500 hr/yr)/(8,760 hr/yr)]= $175,200/yrNotes: If the unit heat output or average fuel cost exceedsthe limit of Figure 5.18, use half the particular value anddouble the savings obtained from Chart C. The annualfuel savings are based on 8,760 hr/yr. For an operatingfactor other that the curve basis, apply a correction factorto curve results. The application of the charts assumesoperating at, or close to, optimum excess air values.

BOILERS AND FIRED SYSTEMS 121Modifications forMaximum EconomyIt is very difficult to pinpointand rectify the majorproblem detracting from efficiency,as there are many interdependentvariables andnumerous pieces of equipmentrequired for the combustionof coal. Further testing and/oroperating manipulations maybe required to “zero in” on asolution. Modification(s) thathave been instituted with a fairdegree of success to reduce highunburned carbon losses and/orhigh-excess-air operation are:Modifying or changingthe pulverizer internals to increasethe coal fineness andthereby enhance combustioncharacteristics.Rerouting or modifyingair/coal distribution piping toavoid coal hang-up and slugflow going to the burners.Installing additional or new combustion controls tosmooth out and maintain consistent performance.Purchasing new coal feeders compatible with andresponsive to unit demand fluctuations.Calibrating air flow and metering devices to ensurecorrect air/coal mixtures and velocities at the burnerthroats.Installing new classifiers to ensure that proper coalfines reach the burners for combustion. Optimally settingair register positions for proper air/fuel mixing and combustion.Replacing worn and abraded burner impellerplates.Increasing the air/coal mixture temperature exitingthe pulverizers to ensure good ignition without coking.Cleaning the burning throat areas of deposits.Installing turning vanes or air foils in the secondaryair supply duct or air plenum to ensure even distributionand proper air/fuel mixing at each burner.Purchasing new, or updating existing combustioncontrols to reflect the present state of the art.As can be seen, the solutions can be varied, simpleor complex, relatively cheap or quite expensive.The incentives to correct the problem(s) are offeredin Figure 5.18. Compare the expected benefits to the boilermanufacturer’s solution and cost to implement and judgethe merits.Figure 5.18 Coal annual savings from reducing combustible losses.Guidelines for Day-to-Day OperationTo a great extent, the items listed which detract fromoperating performance should be those checked on a dayto-daybasis. A checklist should be initiated and developedby the plant to correct those conditions that do notrequire a shutdown, such as:1. Maintaining integrity of distribution dampers andair registers with defined positions for operatingload.2. Ensuring pulverizer exit temperature by maintainingair calibration devices and air-moving equipment.3. Checking feeder speeds and coal hopper valves, ensuringan even and steady coal feed.4. Frequent blending of the raw coal pile to providesome measure of uniformity.Coal ConversionConverting existing boilers that originally were designedto fire oil and/or natural gas to coal capability ispossible for certain types of units provided that significantderatings in steam rate capacities are acceptable toexisting operations. Since boiler installations are some-

122 ENERGY MANAGEMENT HANDBOOKwhat custom in design, fuel changes should be addressedon an individual basis. In general, however, conversionof field-erected units to coal firing is technically possible,whereas conversion of shop-assembled boilers is not.Modifications required to convert shop-assembledunits would include installation of ash-handling facilitiesin the furnace and convection pass, soot-blowing equipment,and modifications to tube spacing each of whichis almost physically impossible to do and prohibitive incost. In addition, even if these modifications could bemade, derating of the unit by about two-thirds wouldbe necessary. Alternatives for consideration rather thanconverting existing shop-assembled units would be to replacewith new coal-fired units, or to carefully assess theapplication of one of an alternative fuel, such as a coal/oilmixture or ultra-fine coal.Many field-erected boilers can be converted to coalfiring without serious compromises being made to gooddesign. The most serious drawback to converting theseboilers is the necessity to downrate by 40 to 50%. Althoughin some instances downrating may be acceptable,most operating locations cannot tolerate losses in steamsupply of this magnitude.Why is downrating necessary? Generally, there aredifferences between boilers designed for oil and gas andthose designed for coal. Burning of coal requires significantlymore combustion volume than that needed for oilor gas due to flame characteristics and the need to avoidexcessive slagging and fouling of heat-transfer surfaces.This means that a boiler originally designed for oil or gasmust be downrated to maintain heat-release rates andfirebox exit temperatures within acceptable limits forcoal. Other factors influencing downrating are acceptableflue-gas velocities to minimize erosion and tube spacing.A suggested approach for assuring the viability ofcoal conversion is:1. Contact the original boiler manufacturer and determinevia a complete inspection the actual modificationsthat would be required for the boiler.2. Determine if the site can accommodate coal storageand other associated equipment, including unloadingfacilities, conveyors, dust abatement, ash disposal,and so on.3. Select the type of coal to burn, considering cost,availability, and pollution restrictions. Before finalselection of a coal, it must be analyzed to determineits acceptability and effect on unit rating. For anyinstallation, final selection must be based on thecoal’s heating value, moisture content, mineral mattercontent, ash fusion and chemical characteristics,and/or pulverizers’ grindability.4. Assess over economics.BibliographyAMERICAN SOCIETY OF MECHANICAL ENGINEERS(ASME), ASME Power Test Code (PTC) 4.1-1964;Steam Generating Units, ASME, New York, 1964.KEENAN, J.H., and J. KAYE, The Gas Tables, Wiley, NewYork, 1948.PAYNE, F.W., Efficient Boiler Operations Source Book,3rd ed., Fairmont, Lilburn, GA., 1991.SCOLLON, R.B., and R.D. SMITH, Energy ConservationGuidebook, Allied Corporation, Morristown, N.J.,1976TAPLIN Jr., H.R., Combustion Efficiency Tables, Fairmont,Lilburn, GA., 1991.

BOILERS AND FIRED SYSTEMS 123AppendixDIRECT CONTACT TECHNOLOGYFOR HOT WATER PRODUCTIONB. Keith WalkerQuikWater a Division of Webco IndustriesSand Springs, OK 74063Direct contact water heating is a heat transfer methodin which there are no tubes isolating hot combustiongasses from the fluid to be heated. The exhaust gases areallowed to come into direct contact with water in a totallynon-pressurized environment so that all heating occurs atatmospheric pressure.Direct contact water heaters consist of bodies sometimesmade 100% out of stainless steel using a verticalhollow chamber in which water is sprayed at the topof this vertical chamber. The upper portion of this verticalchamber is filled with stainless steel balls (or othershapes) which provide a large heat transfer surface areafor the heat of the rising exhaust gasses to be conductedinto the water. This “heat transfer zone” is approximately24” to 36” deep and almost all heat transfer in a directcontact water heater occurs in this area.In the same vertical column, below the “heat transferzone” a burner is mounted which provides the energyused to heat the water. Depending on manufacturer, thisburner may be substantially removed from the fallingwater with a dedicated firing chamber or it may residedirectly in the path of the water flow with an overheadmetal shield to protect the bulk of the flame from waterimpingement. The burner is forced draft and is typicallyfired on natural gas or propane although fuel oils can beused some of the time.Because the water loses all pressure there must be apool for water to collect in after passing through a directcontact heater. This pool is either a full sized tank or justa holding area at the base of the vertical heating tower.Water collects in this tank and is then held until the hotwater is required at a remote location.Mechanical pumps must be used to re-pressurizethe water in this tank and move the volume required tothe needed application. These pumps can provide virtuallyany pressure that a customer might require and can beused as a pressure booster over the typical water pressurethat a facility would see off of a regular city water line.ULTRA HIGH EFFICIENCYDirect contact water heaters are very thermal efficient(hot water energy output ÷ fuel energy in). In allcombustion devices water is a by-product of the burningprocess, because oxygen and hydrogen from the air combineto make H 2 O. This water is vaporized immediatelyfrom liquid to gaseous form in the heat of firing. This phasechange of water from liquid to gas is a cooling processand accounts for about a 12% combustion efficiency loss.Direct contact water heaters super cool the combustiongasses below the point where this newly formed watervapor will re-condense back into liquid form. By returningthe water back to a liquid state direct contact waterheaters are able to reclaim all of the thermal energy thatis normally lost out the stack of a traditional boiler. Thephase change that steals energy by changing from liquidto gas returns heat energy when the gas is returned backinto liquid form. By transferring heat in this way directcontact water heaters are capable of achieving efficienciesthat approach 100%.The thermal efficiency of a direct contact water heateris easily predicted by the exhaust gas temperature.This can also be translated into determining efficiencyby knowing what the inlet water temperature is.If the heating system is in reasonable good order then theexhaust gas temperature should mirror the inlet watertemperature by 5 - 10°F. For example if your inlet water is55°F then your exhaust gas temperature should be about60°F and the heater efficiency would be about 99%. Thetypical operating inlet water temperature for a direct contactwater heater is between 45°F - 80°F. Which translatesto 99.7%-97.7%. If you are recirculating hot water back tothe upper portion of a heater then the efficiency of the

124 ENERGY MANAGEMENT HANDBOOKdirect contact water heater will go down for the timeframe in which the recirculation occurs. For example ifthe return water temperature is 160°F then the exhaustgas temperature would be about 165°F making the directcontact heater efficiency about 75% for the time that 160°Fis being reheated. These dynamics make direct contactheater system design different from boiler system design.Because in all cases you want the coldest possible returnwater temperature from a system loop (not the typical20°F temperature drop for heat exchanger return).WATER TEMPERATURE LIMITATIONSDirect contact water heaters are incapable of producingusable steam and have difficulty in achieving outletwater temperatures higher than 188°F. Direct contact waterheaters can achieve water temperatures up to 193°F.However, the efficiency of the heater drops dramaticallyeven with cold water entering the upper portion of theheating tower as large amounts of water are vaporizedcarrying heat energy away from the heater.Direct contact water heaters operate with a systempressure just a couple of inches water column above atmosphericpressure. In theory it should be possible underthese condition to heat water up to 200°F or even higherin a direct contact water heater. The limiting factor is presentedby when the water is heated. The water is heatedas the droplet falls through open air in the heat transfersection of the direct contact. Because the heating occurswhile the water is moving through the air there are windaerodynamics that drop the localized pressure around thewater droplets to a point that is lower than the availableatmospheric pressure. This in turn lowers the temperatureat which the liquid water phase changes to a gaseousform. Because if this effect attempting to achieve exitwater temperatures higher than 188°F becomes inefficientand unstable. The altitude at which a heater is operatingwill affect the maximum temperature output capability ofthe direct contact heater because of the reduction in availableatmospheric pressure.APPLICATIONS FOR DIRECT CONTACTDirect contact water heaters can be used in almostany application that has a need for large (sometimes notso large) quantities of hot water. There are three generalclassifications of water usage. These groups are:Single Pass ApplicationOpen Loop Recirculation SystemsSpecial ApplicationsA short list of applications is given below, thoughthere are many more possible applications.WATER QUALITY AND EMISSIONSWater quality from a direct contact heater is not usuallynegatively affected after having been heated. Thereare differences in the designs from the different manufacturerswhich make the heaters have higher or lower PHdrops. After a single pass through the body of a directcontact heating unit the introduction of 7.4 PH water exitsthe unit between 6.9 and 6.3 PH, but in general the PHchange is minimized in heating units that have a dedicatedspace to complete the combustion of fuels.Other than CO 2 injection into the water stream directcontact heaters do not negatively affect the consistencyof the water stream. There are cases in poor heaterdesign where combustion has been quenched early andacids created from incomplete combustion lower the PHan extra amount than is explainable by CO 2 injection.Exhaust emissions from a direct contact water heateris typically lower than the emission rates of a boiler firingwith the same burner at the same firing rate. This does notmake the units low NO x emitters, but it does allow certainlow NO x technologies to work even more effectively in directcontact heaters than in comparable boilers.OVERALL DIRECTCONTACT EFFICIENCYThermally, direct contact water heaters are up to 99%efficient, however there are pumps and blowers on theunits. An energy calculation of the electricity consumedby these items should be done based on loading, controldesign, and operating time.Single Pass Open Loop SpecialApplications Applications Applications————————————————————————Plant Cleanup Bottle Warming Hydronic Heating————————————————————————Bird Scalding Hospital Systems Caustic Fluids————————————————————————Car Washes Swimming Pools Green Houses————————————————————————Concrete Batches Fruit Cleaning Hotel Showers————————————————————————Dyeing Cleanup Systems Hotel Sinks————————————————————————Parts CleaningDemineralized————————————————————————DA tank feedBio Gas————————————————————————Comm. LaundryMultiple Temp.————————————————————————Indust. LaundryInstitutional————————————————————————

CHAPTER 6STEAM AND CONDENSATE SYSTEMSPHILIP S. SCHMIDTDepartment of Mechanical EngineeringUniversity of Texas at AustinAustin, Texas6. 1 INTRODUCTIONNearly half of the energy used by industry goes into theproduction of process steam, approximately the sametotal energy usage as that required to heat all the homesand commercial buildings in America. Why is so muchof our energy resource expended for the generation ofindustrial steam?Steam is one of the most abundant, least expensive,and most effective heat-transfer media obtainable. Wateris found everywhere, and requires relatively little modificationfrom its raw state to make it directly usable inprocess equipment. During boiling and condensation, ifthe pressure is held constant and both water and steamare present, the temperature also remains constant. Further,the temperature is uniquely fixed by the pressure,and hence by maintaining constant pressure, which is arelatively easy parameter to control, excellent control ofprocess temperature can also be maintained. The conversionof a liquid to a vapor absorbs large quantities of heatin each pound of water. The resulting steam is easy totransport, and because it is so energetic, relatively smallquantities of it can move large amounts of heat. Thismeans that relatively inexpensive pumping and pipingcan be used compared to that needed for other heatingmedia.Finally, the process of heat transfer by condensation,in the jacket of a steam-heated vessel, for example,is extremely efficient. High rates of heat transfer can beobtained with relatively small equipment, saving bothspace and capital. For these reasons, steam is widely usedas the heating medium in thousands of industries.Prior to the mid-1970s, many steam systems inindustry were relatively energy wasteful. This was notnecessarily bad design for the times, because energy wasso cheap that it was logical to save first cost, even at theexpense of a considerable increase in energy requirements.But things have changed, and today it makesgood sense to explore every possibility for improving125the energy efficiency of steam systems.This chapter will introduce some of the languagecommonly used in dealing with steam systems andwill help energy managers define the basic design constraintsand the applicability of some of the vast array ofmanufacturer’s data and literature which appears in themarketplace. It will discuss the various factors that produceinefficiency in steam system operations and someof the measures that can be taken to improve this situation.Simple calculation methods will be introduced toestimate the quantities of energy that may be lost andmay be partially recoverable by the implementation ofenergy conservation measures. Some important energyconservation areas pertinent to steam systems are coveredin other chapters and will not be repeated here. Inparticular the reader is referred to Chapters 5 (on boilers),7 (on cogeneration), and 15 (on industrial insulation).6.1.1 Components of Steam and Condensate SystemsFigure 6.1 shows a schematic of a typical steam systemin an industrial plant. The boiler, or steam generator,produces steam at the highest pressure (and thereforethe highest temperature) required by the variousprocesses. This steam is carried from the boiler housethrough large steam mains to the process equipmentarea. Here, transfer lines distribute the steam to eachpiece of equipment. If some processes require lower temperatures,the steam may be throttled to lower pressurethrough a pressure-regulating valve (designated PRVon the diagram) or through a back-pressure turbine.Steam traps located on the equipment allow condensateto drain back into the condensate return line, where itflows back to the condensate receiver. Steam traps alsoperform other functions, such as venting air from thesystem on startup, which is discussed in more detaillater.The system shown in Figure 6.1 is, of course, highlyidealized. In addition to the components shown, otherelements, such as strainers, check valves, and pumpingtraps may be utilized. In some plants, condensate maybe simply released to a drain and not returned; thepotential for energy conservation through recovery ofcondensate will be discussed later. Also, in the systemshown, the flash steam produced in the process of throt-

126 ENERGY MANAGEMENT HANDBOOKFig. 6.1 Typical steam system components.tling across the steam traps is vented to the atmosphere.Prevention of this loss represents an excellent opportunityto save Btus and dollars.6.1.2 Energy ConservationOpportunities in Steam SystemsMany opportunities for energy savings exist insteam system operations, ranging from simple operatingprocedure modifications to major retrofits requiring significantcapital expenditures. Table 6.1 shows a checklistof energy conservation opportunities applicable to moststeam systems. It is helpful for energy conservationmanagers to maintain such a running list applicable totheir own situations. Ideas are frequently presented inthe technical and trade literature, and plant operatorsoften make valuable contributions, since after all, theyare the people closest to the problem.To “sell” such improvements to plant managementand operating personnel, it is necessary to demonstratethe dollars-and-cents value of a project or operatingchange. The following sections discuss the thermalproperties of steam, how to determine the steam requirementsof plant equipment and to estimate the amountof steam required to make up for system losses, howto assign a dollar value to this steam, and various approachesto alleviating these losses.6.2 THERMAL PROPERTIES OF STEAM6.2.1 Definitions and TerminologyBefore discussing numerical calculations of steamproperties for various applications, it is necessary toestablish an understanding of some terms commonlyused in the operation of steam systems.British Thermal Unit (Btu). One Btu is the amountof heat required to raise 1 pound of water 1 degreeFahrenheit in temperature. To get a perspective on thisquantity, a cubic foot of natural gas at atmospheric pressurewill release about 1000 Btu when burned in a boilerwith no losses. This same 1000 Btu will produce a littleless than 1 pound of steam at atmospheric conditions,starting from tap water.Boiling Point. The boiling point is the temperatureat which water begins to boil at any given pressure.The boiling point of water at sea-level atmosphericpressure is about 212°F. At high altitude, where theatmospheric pressure is lower, the boiling point is alsolower. Conversely, the boiling point of water goes upwith increasing pressure. In steam systems, we usuallyrefer to the boiling point by another term, “saturationtemperature.”Absolute and Gauge Pressure. In steam systemliterature, we frequently see two different pressuresused. The “ absolute pressure,” designated psia, is thetrue force per unit of area (e.g., pounds per square inch)exerted by the steam on the wall of the pipe or vesselcontaining it. We usually measure pressures, however,with sensing devices that are exposed to the atmosphereoutside, and which therefore register an indication, notof the true force inside the vessel, but of the differencebetween that force and the force exerted by the outsideatmosphere. We call this difference the “ gauge pressure,”designated psig. Since atmospheric pressure at

STEAM AND CONDENSATE SYSTEMS 127Table 6.1 Checklist of Energy Conservation Opportunities in Steam and Condensate SystemsGeneral Operations1. Review operation of long steam lines to remote single-service applications. Consider relocation or conversion ofremote equipment, such as steam-heated storage tanks.2. Review operation of steam systems used only for occasional services, such as winter-only steam tracing lines.Consider use of automatic controls, such as temperature-controlled valves, to assure that the systems are used onlywhen needed.3. Implement a regular steam leak survey and repair program.4. Publicize to operators and plant maintenance personnel the annual cost of steam leaks and unnecessary equipmentoperations.5. Establish a regular steam-use monitoring program, normalized to production rate, to track progress in reduction ofsteam consumption. Publicize on a monthly basis the results of this monitoring effort.6. Consider revision of the plant-wide steam balance in multipressure systems to eliminate venting of low-pressuresteam. For example, provide electrical backup for currently steam-driven pumps or compressors to permit shutoff ofturbines when excess low-pressure steam exists.7. Check actual steam usage in various operations against theoretical or design requirement. Where signifi cant disparitiesexist, determine the cause and correct it.8. Review pressure-level requirements of steam-driven mechanical equipment to evaluate feasibility of using lowerpressure levels.9. Review temperature requirements of heated storage vessels and reduce to minimum acceptable temperatures.10. Evaluate production scheduling of batch operations and revise if possible to minimize startups and shutdowns.Steam Trapping1. Check sizing of all steam traps to assure they are adequately rated to provide proper condensate drainage. Alsoreview types of traps in various services to assure that the most effi cient trap is being used for each application.2. Implement a regular steam trap survey and maintenance program. Train maintenance personnel in techniques fordiagnosing trap failure.Condensate Recovery1. Survey condensate sources presently being discharged to waste drains for feasibility of condensate recovery.2. Consider opportunities for fl ash steam utilization in low-temperature processes presently using fi rst-generationsteam.3. Consider pressurizing atmospheric condensate return systems to minimize fl ash losses.Mechanical Drive Turbines1. Review mechanical drive standby turbines presently left in the idling mode and consider the feasibility of shuttingdown standby turbines.2. Implement a steam turbine performance testing program and clean turbines on a regular basis to maximize effi -ciency.3. Evaluate the potential for cogeneration in multipressure steam systems presently using large pressure-reducingvalves.Insulation1. Survey surface temperatures using infrared thermometry or thermography on insulated equipment and piping tolocate areas of insulation deterioration. Maintain insulation on a regular basis.2. Evaluate insulation of all uninsulated lines and fi ttings previously thought to be uneconomic. Recent rises in energycosts have made insulation of valves, fl anges, and small lines desirable in many cases where this was previouslyunattractive.3. Survey the economics of retrofi tting additional insulation on presently insulated lines, and upgrade insulation if economicallyfeasible.

128 ENERGY MANAGEMENT HANDBOOKsea level is usually around 14.7 psi, we can obtain theabsolute pressure by simply adding 14.7 to the gaugepressure reading. In tables of steam properties, it ismore common to see pressures listed in psia, and henceit is necessary to make the appropriate correction to thepressure indicated on a gauge.Saturated and Superheated Steam. If we put coldwater into a boiler and heat it, its temperature willbegin to rise until it reaches the boiling point. If wecontinue to heat the water, rather than continuing to risein temperature, it begins to boil and produce steam. Aslong as the pressure remains constant, the temperaturewill remain at the saturation temperature for the givenpressure, and the more heat we add, the more liquidwill be converted to steam. We call this boiling liquid a“saturated liquid” and refer to the steam so generatedas “saturated vapor.” We can continue to add more andmore heat, and we will simply generate more saturatedvapor (or simply “ saturated steam”) until the water iscompletely boiled off. At this point, if we continue toadd heat, the steam temperature will begin to rise oncemore. We call this “ superheated steam.” This chapterwill concentrate on the behavior of saturated steam,because this is the steam condition most commonlyencountered in industrial process heating applications.Superheated steam is common in power generation andis often produced in industrial systems when cogenerationof power and process heat is used.Sensible and Latent Heat. Heat input that isdirectly registered as a change in temperature of a substanceis called “ sensible heat,” for the simple reasonthat we can, in fact, “sense” it with our sense of touchor with a thermometer. For example, the heating of thewater mentioned above before it reaches the boilingpoint would be sensible heating. When the heat goesinto the conversion of a liquid to a vapor in boiling, orvice versa in the process of condensation, it is termed“ latent heat.” Thus when a pound of steam condenseson a heater surface to produce a pound of saturatedliquid at the same temperature, we say that it has releasedits latent heat. If the condensate cools further, itis releasing sensible heat.Enthalpy. The total energy content of a flowingmedium, usually expressed in Btu/lb, is termed its “enthalpy.”The enthalpy of steam at any given conditiontakes into account both latent and sensible heat, andalso the “mechanical” energy content reflected in itspressure. Hence steam at 500 psia and 600°F will havea higher enthalpy than steam at the same temperaturebut at 300 psia. Also, saturated steam at any temperatureand pressure has a higher enthalpy than condensate atthe same conditions due to the latent heat content of thesteam. Enthalpy, as listed in tables of steam properties,does not include the kinetic energy of motion, but thiscomponent is insignificant in most energy conservationapplications.Specific Volume. The specific volume of a substanceis the amount of space (e.g., cubic feet) occupiedby 1 pound of substance. This term will become importantin some of our later discussions, because steam normallyoccupies a much greater volume for a given massthan water (i.e., it has a much greater specific volume),and this must be taken into account when consideringthe design of condensate return systems.Condensate. Condensate is the liquid producedwhen steam condenses on a heater surface. As shownlater, this condensate still contains a significant fractionof its energy, and can be returned to the boiler to conservefuel.Flash Steam. When hot condensate at its saturationtemperature corresponding to the elevated pressurein a heating vessel rapidly drops in pressure, as,for example, when passing through a steam trap or avalve, it suddenly finds itself at a temperature abovethe saturation temperature for the new pressure. Steamis thus generated which absorbs sufficient energy todrop the temperature of the condensate to the appropriatesaturation level. This is called “ flash steam,” andthe pressure-reduction process is called “flashing.” Inmany condensate return systems, flash steam is simplyreleased to the atmosphere, but it may, in fact, havepractical applications in energy conservation.Boiler Efficiency. The boiler efficiency is the percentageof the energy released in the burning of fuel in aboiler which actually goes into the production of steam.The remaining percentage is lost through radiation fromthe boiler surfaces, blowdown of the boiler water tomaintain satisfactory impurity levels, and loss of thehot flue gas up the stack. Although this chapter doesgo into detail on the subject of boiler efficiency, which isdiscussed in Chapter 5, it is important to recognize thatthis parameter relates the energy savings obtainable byconserving steam to the fuel savings obtainable at theboiler, a relation of obvious economic importance. Thusif we save 100 Btu of steam energy and have a boilerwith an efficiency of 80%, the actual fuel energy savedwould be 100/0.80, or 125 Btu. Because boilers alwayshave an efficiency of less than 100% (and more com-

STEAM AND CONDENSATE SYSTEMS 129Table 6.2 Thermodynamic Properties of Saturated Steam(l) (2) (3) (4) (5) (6) (7)Absolute Enthalpy of Enthalpy of SpecificGauge Pressure Steam Temp. Sat. Liquid Latent Heat Steam VolumePressure (psia) (°F) (Btu/lb) (Btu/lb) (Btu/lb) (ft 3 /lb)In. vacuum29.743 0.08854 32.00 0.00 1075.8 1075.8 3306.0029.515 0.2 53.14 21.21 1063.8 1085.0 1526.0027.886 1.0 101.74 69.70 1036.3 1106.0 333.6019.742 5.0 162.24 130.13 1001.0 1131.1 73.529.562 10.0 193.21 161.17 982.1 1143.3 38.427.536 11.0 197.75 165.73 979.3 1145.0 35.145.490 12.0 201.96 169.96 976.6 1146.6 32.403.454 13.0 205.88 173.91 974.2 1148.1 30.061.418 14.0 209.56 177.61 971.9 1149.5 28.04Psig0.0 14.696 212.00 180.07 970.3 1150.4 26.801.3 16.0 216.32 184.42 967.6 1152.0 24.752.3 17.0 219.44 187.56 965.5 1153.1 23.395.3 20.0 227.96 196.16 960.1 1156.3 20.0910.3 25.0 240.07 208.42 952.1 1160.6 16.3015.3 30.0 250.33 218.82 945.3 1164.1 13.7520.3 35.0 259.28 227.91 939.2 1167.1 11.9025.3 40.0 267.25 236.03 933.7 1169.7 10.5030.3 45.0 274.44 243.36 928.6 1172.0 9.4040.3 55.0 287.07 256.30 919.6 1175.9 7.7950.3 65.0 297.97 267.50 911.6 1179.1 6.6660.3 75.0 307.60 277.43 904.5 1181.9 5.8270.3 85.0 316.25 286.39 897.8 1184.2 5.1780.3 95.0 324.12 294.56 891.7 1186.2 4.6590.3 105.0 331.36 302.10 886.0 1188.1 4.23100.0 114.7 337.90 308.80 880.0 1188.8 3.88110.3 125.0 344.33 315.68 875.4 1191.1 3.59120.3 135.0 350.21 321.85 870.6 1192.4 3.33125.3 140.0 353.02 324.82 868.2 1193.0 3.22130.3 145.0 355.76 327.70 865.8 1193.5 3.11140.3 155.0 360.50 333.24 861.3 1194.6 2.92150.3 165.0 365.99 338.53 857.1 1195.6 2.75160.3 175.0 370.75 343.57 852.8 1196.5 2.60180.3 195.0 379.67 353.10 844.9 1198.0 2.34200.3 215.0 387.89 361.91 837.4 1199.3 2.13225.3 240.0 397.37 372.12 828.5 1200.6 1.92250.3 265.0 406.11 381.60 820.1 1201.7 1.74300.0 417.33 393.84 809.0 1202.8 1.54400.0 444.59 424.00 780.5 1204.5 1.16450.0 456.28 437.20 767.4 1204.6 1.03500.0 467.01 449.40 755.0 1204.4 0.93600.0 486.21 471.60 731.6 1203.2 0.77900.0 531.98 526.60 668.8 1195.4 0.501200.0 567.22 571.70 611.7 1183.4 0.361500.0 596.23 611.60 556.3 1167.9 0.281700.0 613.15 636.30 519.6 1155.9 0.242000.0 635.82 671.70 463.4 1135.1 0.192500.0 668.13 730.60 360.5 1091.1 0.132700.0 679.55 756.20 312.1 1068.3 0.113206.2 705.40 902.70 0.0 902.7 0.05What is the temperature inside the line? Coming down column (1) we find a pressure of 150, and moving over tocolumn (3), we note that the corresponding steam temperature at this pressure is about 366°F.Column (4) lists the enthalpy of the saturated liquid in Btu/lb of water. We can see at the head of the column thatthis enthalpy is designated as 0 at a temperature [column (3)] of 32°F.

130 ENERGY MANAGEMENT HANDBOOKmonly around 75 to 80%) there is a built-in “amplifier”on any energy savings effected in the steam system.6.2.2 Properties of Saturated SteamIn calculating the energy savings obtainablethrough various measures, it is important to understandthe quantitative thermal properties of steam and condensate.Table 6.2 shows a typical compilation of theproperties of saturated steam.Columns 1 and 2 list various pressures, either ingauge (psig) or absolute (psia). Note that these twopressures always differ by about 15 psi (14.7 to be moreprecise). Remember that the former represents the pressureindicated on a normal pressure gauge, while thelatter represents the true pressure inside the line. Column3 shows the saturation temperature correspondingto each of these pressures. Note, for example, that at anabsolute pressure of 14.696 (the normal pressure of theatmosphere at sea level) the saturation temperature is212°F, the figure we are all familiar with. Suppose thatwe have a pressure of 150 psi indicated on the pressuregauge on a steam line. This is an arbitrary referencepoint, and therefore the heat indicated at any other temperaturetells us the amount of heat added to raise thewater from an initial value of 32°F to that temperature.For example, referring back to our 150 psig steam, thewater contains about 338.5 Btu/lb: starting from 32°F,10 lb of water would contain 10 times this number, orabout 3385 Btu. We can also subtract one number fromanother in this column to find the amount of heat necessaryto raise the water from one temperature to another.If the water started at 101.74°F, it would contain a heatof 69.7 Btu/lb, and to raise it from this temperatureto 366°F would require 338.5—69.7, or about 268.8 Btufor each pound of water. Column (5) shows the latentheat content of a pound of steam for each pressure. Forour 150-psig example, we can see that it takes about857 Btu to convert each pound of saturated water intosaturated steam. Note that this is a much larger quantitythan the heat content of the water alone, confirming theearlier observation that steam is a very effective carrierof heat; each pound can give up, in this case, 857 Btuwhen condensed on a surface back to saturated liquid.Column (6), the enthalpy of the saturated steam, representssimply the sum of columns (4) and (5), since eachpound of steam contains both the latent heat requiredto vaporize the water and the sensible heat required toraise the water to the boiling point in the first place.Column (7) shows the specific volume of the saturatedsteam at each pressure. Note that as the pressureincreases, the steam is compressed; that is, it occupiesless space per pound. 150-psig steam occupies only 2.75ft 3 /lb; if released to atmospheric pressure (0 psig) itwould expand to nearly 10 times this volume. By comparison,saturated liquid at atmospheric pressure has aspecific volume of only 0.017 ft 3 /lb (not shown in thetable), and it changes only a few percent over the entirepressure range of interest here. Thus 1 lb of saturatedliquid condensate at 212°F will expand more than 1600times in volume in converting to a vapor. This illustratesthat piping systems for the return of condensate fromsteam-heated equipment must be sized primarily toaccommodate the large volume of flashed vapor, andthat the volume occupied by the condensate itself isrelatively small.The steam tables can be a valuable tool in estimatingenergy savings, as illustrated in the followingexample.Example: A 100-ft run of 6-in. steam piping carriessaturated steam at 95 psig. Tables obtained from aninsulation manufacturer indicate that the heat loss fromthis piping run is presently 110,000 Btu/hr. With properinsulation, the manufacturer’s tables indicate that thisloss could be reduced to 500 Btu/hr. How many poundsper hour of steam savings does this installation represent,and if the boiler is 80% efficient, what would bethe resulting fuel savings?From the insulation manufacturer’s data, we canfind the reduction in heat loss:heat-loss reduction = 110,000 – 500 = 109,500 Btu/hrFrom Table 6.2 at 95 psig (halfway between 90 and100), the total heat of the steam is about 1188.4 Btu/lb.The steam savings, is therefore109,500 Btu/hrsteam savings = ————————1188.4 Btu/lbAssume that condensate is returned to the boiler ataround 212°F; thus the condensate has a heat content ofabout 180 Btu/lb. The heat required to generate 95 psigsteam from this condensate is 1188.4 – 180.0 or 1008.4Btu/lb. If the boiler is 80% efficient, then1008.4 Btu/lb × 92 lb/hrfuel savings = ———————————0.80= approximately 0.116 million Btu/hr6.2.3 Properties of Superheated SteamIf additional heat is added to saturated steam withno liquid remaining, it begins to superheat and the tem-

STEAM AND CONDENSATE SYSTEMS 131Table 6.3 Thermodynamic Properties of Superheated SteamTemperature (°F)Abs. Press(psi) 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 15001 v 0.0161 392.5 452.3 511.9 571.5 631.1 690.7h 68.00 1150.2 1195.7 1241.8 1288.6 1336.1 1384.55 v 0.0161 78.14 90.24 102.24 114.21 126.15 138.08 150.01 161.94 173.86 185.78 197.70 209.62 221.53 233.45h 68.01 1148.6 1194.8 1241.3 1288.2 1335.9 1384.3 1433.6 1483.7 1534.7 1586.7 1639.6 1693.3 1748.0 1803.510 v 0.0161 38.84 44.98 51.03 57.04 63.03 69.00 74.98 80.94 86.91 92.87 98.84 104.80 110.76 116.72h 68.02 1146.6 1193.7 1240.6 1287.8 1335.5 1384.0 1433.4 1483.5 1534.6 1586.6 1639.5 1693.3 1747.9 1803.415 v 0.0161 0.0166 29.899 33.963 37.985 41.986 45.978 49.964 53.946 57.926 61.905 65.882 69.858 73.833 77.807h 68.04 168.09 1192.5 1239.9 1287.3 1335.2 1383.8 1433.2 1483.4 1534.5 1586.5 1639.4 1693.2 1747.8 1803.420 v 0.0161 0.0166 22.356 25.428 28.457 31.466 34.465 37.458 40.447 43.435 46.420 49.405 52.388 55.370 58.352h 68.05 168.11 1191.4 1239.2 1286.9 1334.9 1383.5 1432.9 1483.2 1534.3 1586.3 1639.3 1693.1 1747.8 1803.340 v 0.0161 0.0166 11.036 12.624 14.165 15.685 17.195 18.699 20.199 21.697 23.194 24.689 26.183 27.676 29.168h 68.10 168.15 1186.6 1236.4 1285.0 1333.6 1382.5 1432.1 1482.5 1533.7 1585.8 1638.8 1992.7 1747.5 1803.060 v 0.0161 0.0166 7.257 8.354 9.400 10.425 11.438 12.466 13.450 14.452 15.452 16.450 17.448 18.445 19.441h 68.15 168.20 1181.6 1233.5 1283.2 1332.3 1381.5 1431.3 1481.8 1533.2 1585.3 1638.4 1692.4 1747.1 1802.880 v 0.0161 0.0166 0.0175 6.218 7.018 7.794 8.560 9.319 10.075 10.829 11.581 12.331 13.081 13.829 14.577h 68.21 168.24 269.74 1230.5 1281.3 1330.9 1380.5 1430.5 1481.1 1532.6 1584.9 1638.0 1692.0 1746.8 1802.5100 v 0.0161 0.0166 0.0175 4.935 5.558 6.216 6.833 7.443 8.050 8.655 9.258 9.860 10.460 11.060 11.659h 68.26 168.29 269.77 1227.4 1279.3 1329.6 1379.5 1429.7 1480.4 1532.0 1584.4 1637.6 1691.6 1746.5 1802.2120 v 0.0161 0.0166 0.0175 4.0786 4.6341 5.1637 5.6831 6.1928 6.7006 7.2060 7.7096 8.2119 8.7130 9.2134 9.7130h 68.31 168.33 269.81 1224.1 1277.4 1328.1 1378.4 1428.8 1479.8 1531.4 1583.9 1637.1 1691.3 1746.2 1802.0140 v 0.0161 0.0166 0.0175 3.4661 3.9526 4.4119 4.8585 5.2995 5.7364 6.1709 6.6036 7.0349 7.4652 7.8946 8.3233h 68.37 168.38 269.85 1220.8 1275.3 1326.8 1377.4 1428.0 1479.1 1530.8 1583.4 1636.7 1690.9 1745.9 1801.7160 v 0.0161 0.0166 0.0175 3.0060 3.4413 3.8480 4.2420 4.6295 5.0132 5.3945 5.7741 6.1522 6.5293 6.9055 7.2811h 68.42 168.42 269.89 1217.4 1273.3 1325.4 1376.4 1427.2 1478.4 1530.3 1582.9 1636.3 1690.5 1745.6 1801.4180 v 0.0161 0.0166 0.0174 2.6474 3.0433 3.4093 3.7621 4.1084 4.4505 4.7907 5.1289 5.4657 5.8014 6.1363 6.4704h 68.47 168.47 269.92 1213.8 1271.2 1324.0 1375.3 1426.3 1477.7 1529.7 1582.4 1635.9 1690.2 1745.3 1801.2200 v 0.0161 0.0166 0.0174 2.3598 2.7247 3.0583 3.3783 3.6915 4.0008 4.3077 4.6128 4.9165 5.2191 5.5209 5.8219h 68.52 168.51 269.96 1210.1 1269.0 1322.6 1374.3 1425.5 1477.0 1529.1 1581.9 1635.4 1689.8 1745.0 1800.9250 v 0.0161 0.0166 0.0174 0.0186 2.1504 2.4662 2.6872 2.9410 3.1909 3.4382 3.6837 3.9278 4.1709 4.4131 4.6546h 68.66 168.63 270.05 375.10 1263.5 1319.0 1371.6 1423.4 1475.3 1527.6 1580.6 1634.4 1688.9 1744.2 1800.2300 v 0.0161 0.0166 0.0174 0.0186 1.7665 2.0044 2.2263 2.4407 2.6509 2.8585 3.0643 3.2688 3.4721 3.6746 3.8764h 68.79 168.74 270.14 375.15 1257.7 1315.2 1368.9 1421.3 1473.6 1526.2 1579.4 1633.3 1688.0 1743.4 1799.6350 v 0.0161 0.0166 0.0174 0.0186 1.4913 1.7028 1.8970 2.0832 2.2652 2.4445 2.6219 2.7980 2.9730 3.1471 3.3205h 68.92 168.85 270.24 375.21 1251.5 1311.4 1366.2 1419.2 1471.8 1524.7 1578.2 1632.3 1687.1 1742.6 1798.9400 v 0.0161 0.0166 0.0174 00162 1.2841 1.4763 1.6499 1.8151 1.9759 2.1339 2.2901 2.4450 2.5987 2.7515 2.9037h 69.05 168.97 270.33 375.27 1245.1 1307.4 1363.4 1417.0 1470.1 1523.3 1576.9 1631.2 1686.2 1741.9 1798.2500 v 0.0161 0.0166 0.0174 0.0186 0.9919 1.1584 1.3037 1.4397 1.5708 1.6992 1.8256 1.9507 2.0746 2.1977 2.3200h 69.32 169.19 270.51 375.38 1231.2 1299.1 1357.7 1412.7 1466.6 1520.3 1574.4 1629.1 1684.4 1740.3 1796.9600 v 0.0161 0.0166 0.0174 0.0186 0.7944 0.9456 1.0726 1.1892 1.3008 1.4093 1.5160 1.6211 1.7252 1.8284 1.9309h 69.58 169.42 270.70 375.49 1215.9 1290.3 1351.8 1408.3 1463.0 1517.4 1571.9 1627.0 1682.6 1738.8 1795.6700 v 0.0161 0.0166 0.0174 0.0186 0.0204 0.7928 0.9072 1.010 2 1.1078 1.2023 1.2948 1.3858 1.4757 1.5647 1.6530h 69.84 169.65 270.89 375.61 487.93 1281.0 1345.6 1403.7 1459.4 1514.4 1569.4 1624.8 1680.7 1737.2 1794.3800 v 0.0161 0.0166 0.0174 0.0186 0.0204 0.6774 0.7828 0.8759 0.9631 1.0470 1.1289 1.2093 1.2885 1.3669 1.4446h 70.11 169.88 271.07 375.73 487.88 1271.1 1339.2 1399.1 1455.8 1511.4 1566.9 1622.7 1678.9 1736.0 1792.9900 v 0.0161 0.0166 0.0174 0.0186 0.0204 0.5869 0.6858 0.7713 0.8504 0.9262 0.9998 1.0720 1.1430 1.2131 1.2825h 70.37 170.10 271.26 375.84 487.83 1260.6 1332.7 1394.4 1452.2 1508.5 1564.4 1620.6 1677.1 1734.1 1791.61000 v 0.0161 0.0166 0.0174 0.0186 0.0204 0.5137 0.6080 0.6875 0.7603 0.8295 0.8966 0.9622 1.0266 1.0901 1.1529h 70.63 170.33 271.44 375.96 487.79 1249.3 1325.9 1389.6 1448.5 1504.4 1561.9 1618.4 167~.3 1732.5 1790.3perature will rise. Table 6.3 shows the thermodynamicproperties of superheated steam. Unlike the saturatedsteam of Table 6.2, in which each pressure had only asingle temperature (the saturation temperature) associatedwith it, superheated steam may exist, for a givenpressure, at any temperature above the saturation temperature.Thus the properties must be tabulated as afunction of both temperature and pressure, rather thanpressure alone. With this exception, the values in thesuperheated steam table may be used exactly like thosein Table 6.2.Example: Suppose, in the preceding example, that thesteam line is carrying superheated steam at 250 psia(235 psig) and 500°F (note that both temperature andpressure must be specified for superheated steam). For

132 ENERGY MANAGEMENT HANDBOOKthe same reduction in heat loss (109,500 Btu/hr), howmany pounds per hour of steam is saved?From Table 6.3 the enthalpy of steam at 250 psiaand 500°F is 1263.5 Btu/lb. Thus109,500 Btu/hrsteam savings = ————————1263.5 Btu/lb= 86.6 lb/hrTable 6.4 Orders of Magnitudeof Convective Conductances—————————————————————————Order of Magnitude of hHeating Process (Btu/hr ft 2 • F)—————————————————————————Free convection, air 1Forced convection, air 5-10Forced convection, water 250-1000Condensation, steam 5000-10,000—————————————————————————6.2.4 Heat-Transfer Characteristics of SteamAs mentioned in Section 6.1 steam is one of themost effective heat-transfer media available. The rateof heat transfer from a fluid medium to a solid surface(such as the surface of a heat-exchanger tube or a jacketedheating vessel) can be expressed by Newton’s lawof cooling:q = h(T f – T s )where q is the rate of heat transfer per unit of surface area(e.g., Btu/hr/ft 2 ), h is a proportionality factor called the“convective conductance,” T f is the temperature of thefluid medium, and T s is the temperature of the surface.Table 6.4 shows the order of magnitudes of h for severalheat-transfer media. Condensation of steam can beseveral times as effective as the flow of water over a surfacefor the transfer of heat, and may be 1000 times moreeffective than a gaseous heating medium, such as air.In a heat exchanger, the overall effectiveness musttake into account the fluid resistances on both sides ofthe exchanger and the conduction of heat through thetube wall. These effects are generally lumped into a single“overall conductance,” U, defined by the equationq = U(T f1 – T f2 )where q is defined as before, U is the conductance, andT f1 and T f2 are the temperatures of the two fluids. Inaddition, there is a tendency for fluids to deposit “foulinglayers” of crystalline, particulate, or organic matteron transfer surfaces, which further impede the flowof heat. This impediment is characterized by a “foulingresistance,” which, for design purposes, is usuallyincorporated as an additional factor in determining theoverall conductance.Table 6.5 illustrates typical values of U (not includingfouling) and the fouling resistance for exchangersemploying steam on the shell side versus exchangers usinga light organic liquid (such as a typical heat-transferoil). The 30 to 50% higher U values for steam translatedirectly into a proportionate reduction in required heatexchangerarea for the same fluid temperatures. Furthermore,fouling resistances for the steam-heated exchangersare 50 to 100% lower than for a similar service usingan organic heating medium, since pure steam containsno contaminants to deposit on the exchanger surface.From the design standpoint, this means that the additionalheating surface incorporated to allow for foulingneed not be as great. From the operating viewpoint, ittranslates into energy conservation, since more heat canbe transferred per hour in the exchanger for the samefluid conditions, or the same heating duty can be metwith a lower fluid temperature difference if the foulingresistance is lower.Table 6.5 Comparison of Steam and Light Organics as Heat-Exchange MediaTypical FoulingTypical UResistanceShell-Side Fluid Tube-Side Fluid (Btu/hr ft 2 • °F) (hr ft 2 • °F/Btu)Steam Light organic liquid 135-190 0.001Steam Heavy organic liquid 45-80 0.002Light organic liquid Light organic liquid 100-130 0.002Light organic liquid Heavy organic liquid 35-70 0.003

STEAM AND CONDENSATE SYSTEMS 1336.3 ESTIMATING STEAMUSAGE AND ITS VALUETo properly assess the worth of energy conservationimprovements in steam systems, it is first necessaryto determine how much steam is actually required tocarry out a desired process, how much energy is beingwasted through various system losses, and the dollarvalue of these losses. Such information will be needed todetermine the potential gains achievable with insulation,repair or improvement of steam traps, and condensaterecovery systems.6.3.1 Determining Steam RequirementsSeveral approaches can be used to determine processsteam requirements. In order of increasing reliability,they include the use of steam consumption tables fortypical equipment, detailed system energy balances, anddirect measurement of steam and/or condensate flows.The choice of which method is to be used depends onhow critical the steam-using process is to the plant’soverall energy consumption and how the data are to beused.For applications in which a high degree of accuracyis not required, such as developing rough estimatesof the distribution of energy within a plant, steamconsumption tables have been developed for variouskinds of process equipment. Table 6.6 shows steam consumptiontables for a number of typical industrial andcommercial applications. To illustrate, suppose that wewish to estimate the steam usage in a soft-drink bottlingplant for washing 2000 bottles/min. From the table wesee that, typically, a bottle washer uses about 310 lbof 5-psig steam per hour for each 100 bottles/min ofcapacity. For a washer with a 2000-bottle/min capacitywe would, of course, use about 20 times this value, or6200 lb/hr of steam. Referring to Table 6.2, we see that5-psig steam has a total enthalpy of about 1156 Btu/lb.The hourly heat usage of this machine, therefore, wouldbe approximately 1156 × 6200, or a little over 7 millionBtu/hr. Remember that this is the heat content of thesteam itself, not the fuel heat content required at theboiler, since the boiler efficiency has not yet been takeninto account.Note that most of the entries in Table 6.6 showthe steam consumption “in use,” not the peak steamconsumption during all phases of operation. Thesefigures are fairly reliable for equipment that operateson a more-or-less steady basis; however, they may bequite low for batch-processing operations or operationswhere the load on the equipment fluctuates significantlyduring its operation. For this reason, steam equipmentmanufacturers recommend that estimated steam consumptionvalues be multiplied by a “factor of safety,”typically between 2 and 5, to assure that the equipmentwill operate properly under peak-load conditions. Thiscan be quite important from the standpoint of energyefficiency. For example, if a steam trap is sized for averageload conditions only, during startup or heavy-loadoperations, condensate will tend to back up into theheating vessel, reducing the effective area for condensationand hence reducing its heating capacity. For steamtraps and condensate return lines, the incremental costof slight oversizing is small, and factors of safety areused to assure that the design is adequately conservativeto guarantee rapid removal of the condensate. Grossoversizing of steam traps, however, can also cause excessivesteam loss. This point is discussed in more detail inthe section on steam traps, where appropriate factors ofsafety for specific applications are given.A second, and generally more accurate, approachto estimating steam requirements is by direct energybalancecalculations on the process. A comprehensivediscussion of energy balances is beyond the scope ofthis section: analysis of complex equipment should beundertaken by a specialist. It is, however, possible todetermine simple energy balances on equipment involvingthe heating of a single product.The energy-balance concept simply states that anyenergy put into a system with steam must be absorbedby the product and/or the equipment itself, dissipatedto the environment, carried out with the product, or carriedout in the condensate. Recall that the concepts ofsensible and latent heat were discussed in Section 6.2 inthe context of heat absorption by water in the productionof steam. We can extend this concept to considerheat absorption of any material, such as the heating ofair in a dryer or the evaporation of water in the processof condensing milk.The sensible heat requirement of any process isdefined in terms of the “specific heat” of the materialbeing heated. Table 6.7 gives specific heats for a numberof common substances. The specific heat specifiesthe number of Btu required to raise 1 pound of a substancethrough a temperature rise of 1°F. Remember thatfor water, we stated that 1 Btu was, by definition, theamount of heat required to raise 1 lb by 1°F. The specificheat of water, therefore, is exactly 1 (at least near normalambient conditions). To see how the specific heat can beused to calculate steam requirements for the sensibleheating of products, consider the following example.Example: A paint dryer requires about 3000 cfm of200°F air, which is heated in a steam-coil unit. How

134 ENERGY MANAGEMENT HANDBOOKTable 6.6 Typical Steam Consumption Rates for Industrial and Commercial EquipmentSteamTypical ConsumptionPressure in UseType of Installation Description (psig) (lb/hr)Bakeries Dough-room trough, 8 ft long 10 4Oven, white bread, 120-ft 2 surface 10 29Bottle washing Soft drinks, per 100 bottles/min 5 310Dairies Pasteurizer, per 100 gal heated/20 min 15-75 232 (max)Dishwashers Dishwashing machine 15-20 60-70Hospitals Sterilizers, instrument, per 100 in. 3 , approx. 40-50 3Sterilizers, water, per 10 gal, approx. 40-50 6Disinfecting ovens, double door, 50-100 ft 3 , per 40-50 2110 ft 3 , approx.Laundries Steam irons, each 100 4Starch cooker, per 10 gal capacity 100 7Laundry presses, per 10-in. length, approx. 100 7Tumblers, 40 in., per 10-in. length, approx. 100 38Plastic molding Each 12-15 ft 2 platen surface 125 29Paper manufacture Corrugators, per 1,000 ft 2 175 29Wood pulp paper, per 100 lb of paper 50 372Restaurants Standard steam tables, per ft of length 5-20 36Steamjacketed kettles, 25 gal of stock 5-20 29Steamjacketed kettles, 60 gal of stock 5-20 58Warming ovens, per 20 ft 3 5-20 29Silver mirroring Average steam tables 5 102Tire shops Truck molds, large 100 87Passenger molds 100 29many pounds of 50-psig steam does this unit requireper hour?The density of air at temperatures of several hundreddegrees or below is about 0.075 lb/ft 3 . The numberof pounds of air passing through the dryer is then3000 ft 3 /min × 60 min/hr × 0.075 lb/ft 3 = 13,500 lb/hrSuppose that the air enters the steam-coil unit at70°F. Its temperature will then be raised by 200 – 70 =130°F. From Table 6.7 the specific heat of air is 0.24 Btu/lb °F. The energy required to provide this temperaturerise is, therefore,13,500 lb/hr × 0.24 Btu/lb °F × 130°F = 421,200 Btu/hrEmploying the energy-balance principle, whateverenergy is absorbed by the product (air) must be providedby an equal quantity of steam energy, less theenergy contained in the condensate.From Table 6.2 the total enthalpy per pound of50-psig steam is about 1179.1 Btu, and the condensate(saturated liquid) has an enthalpy of 267.5 Btu/lb. Eachpound of steam therefore gives up 1179.1 – 267.5, or911.6 Btu (i.e., its latent heat) to the air. The steam requiredis, therefore, justheat required by air per hour 421,200——————————————— = ——— = 462 lb/hrheat released per pound of steam 911.6To illustrate how latent heat comes into play whenconsidering the steam requirements of a process, consideranother example, this time involving a steam-heatevaporator.Example: A milk evaporator uses a steamjacketed kettle,in which milk is batch-processed at atmospheric pressure.The kettle has a 1500-lb per batch capacity. Milk isheated from a temperature of 80°F to 212°F, where 25%of its mass is then driven off as vapor. Determine theamount of 15-psig steam required per batch, not includ-

STEAM AND CONDENSATE SYSTEMS 135Table 6.7 Specific Heats of Common MaterialsMaterial Btu/lb • °F Material Btu/lb • °FSolidsAluminum 0.22 Iron, cast 0.49Asbestos 0.20 Lead 0.03Cement, dry 0.37 Magnesium 0.25Clay 0.22 Porcelain 0.26Concrete, stone 0.19 Rubber 0.48Concrete, cinder 0.18 Silver 0.06Copper 0.09 Steel 0.12Glass, common 0.20 Tin 0.05Ice, 32°F 0.49 Wood 0.32-0.48LiquidsAcetone 0.51 Milk 0.90Alcohol, methyl, 60-70°F 0.60 Naphthalene 0.41Ammonia, 104°F 1.16 Petroleum 0.51Ethylene glycol 0.53 Soybean oil 0.47Fuel oil, sp. gr. 86 0.45 Tomato juice 0.95Glycerine 0.58 Water 1.00Gases(Constant-Pressure Specific Heats)Acetone 0.35 Carbon dioxide 0.20Air, dry, 32-392°F 0.24 Methane 0.59Alcohol 0.45 Nitrogen 0.24Ammonia 0.54 Oxygen 0.22ing the heating of the kettle itself.We must first heat the milk from 80°F to 212°F(sensible heating) and then evaporate off 0.25 × 1500 =375 lb of water.From Table 6.7, the specific heat of milk is 0.90Btu/lb °F. The sensible heat requirement is, therefore,0.90 × 1500 lb × (212 – 80)°F = 178,200 Btu/batchIn addition, we must provide the latent heat tovaporize 375 lb of water at 212°F. From Table 6.2, 970.3Btu/lb is required. The total latent heat is, therefore,375 × 970.3 = 363,863 Btu/batchand the total heat input is363,863 + 178,200 = 542,063 Btu/batchThis heat must be supplied as the latent heat of 15-psig steam, which, from Table 6.2, is about 945 Btu/lb.The total steam requirement is, then,542,063 Btu/batch———————— = 574 lb of 15-psig steam per batch945 Btu/lbWe could have also determined the startup requirementto heat the steel kettle, if we could estimate its weight,using the specific heat of 0.12 Btu/lb °F for steel, asshown in Table 6.7.6.3.2 Estimating Surface and Leakage LossesIn addition to the steam required to actually carryout a process, heat is lost through the surfaces of pipes,storage tanks, and jacketed heater surfaces, and steamis lost through malfunctioning steam traps, and leaks inflanges, valves, and other fittings. Estimation of theselosses is important, because fixing them can often be themost cost-effective energy conservation measure available.Figure 6.2 illustrates the annual heat loss, based on24-hr/day, 365-day/yr operation, for bare steam lines at

136 ENERGY MANAGEMENT HANDBOOKFig. 6.2 Heat loss from bare steam lines.various pressures. The figure shows, for example, thata 100-ft run of 6-in. line operating at 100 psig will loseabout 1400 million Btu/yr. The economic return on aninsulation retrofit can easily be determined with pricedata obtained from an insulation contractor.Figure 6.3 can be used to estimate heat losses fromflat surfaces at elevated temperatures, or from alreadyinsulated piping runs for which the outside jacket surfacetemperature is known. The figure shows the heatflow per hour per square foot of exposed surface area asa function of the difference in temperature between thesurface and the surrounding air. It will be noted that thenature of the surface significantly affects the magnitudeof the heat loss. This is because thermal radiation, whichis strongly dependent on the character of the radiatingsurface, plays an important role in heat loss at elevatedtemperature, as does convective heat loss to the air.Another important source of energy loss in steamsystems is leakage from components such as looseflanges or malfunctioning steam traps. Figure 6.4 permitsestimation of this loss of steam at various pressuresleaking through holes. The heat losses are represented inmillion Btu/yr, based on full-time operation. Using thefigure, we can see that a stuck-open steam trap with a1/8-in. orifice would waste about 600 million Btu/yr ofsteam energy when leaking from a 100-psig line. Thisfigure can also be used to estimate magnitudes of leakagefrom other sources of more complicated geometry.It is necessary to first determine an approximate areaof leakage (in square inches) and then calculate theequivalent hole diameter represented by that area. Thefollowing example illustrates this calculation.Fig. 6.3 Heat losses from surfaces at elevatedtemperatures.Example: A flange on a 200-psig steam line has a leakinggasket. The maintenance crew, looking at the gasket,estimates that it is about 0.020 in. thick and that it isleaking from about 1/8-in. of the periphery of the flange.Estimate the annual heat loss in the steam if the line isoperational 8000 hr/yr.The area of the leak is a rectangle 0.020 in. wideand 1/8-in. in length:leak area = 0.020 × 1/8 = 0.0025 in. 2An equivalent circle will have an area of πD 2 /4, soif πD 2 /4 = 0.0025, then D = 0.056 in. From Figure 6.4,this leak, if occurring year-round (8760 hr), would wasteabout 200 million Btu/yr of steam energy. For actualoperation8000heat lost = ——— × 200,000,000 = 182.6 million Btu/hr8760Since the boiler efficiency has not been considered, theactual fuel wastage would be about 25% greater.6.3.3 Measuring Steam and Condensate RatesIn maintaining steam systems at peak efficiency, it isoften desirable to monitor the rate of steam flow throughthe system continuously, particularly at points of major

STEAM AND CONDENSATE SYSTEMS 137Fig. 6.4 Heat loss from steam leaks.Fig. 6.5 Orifice flowmeter.usage, such as steam mains. Figure 6.5 shows one of themost common types of flowmetering device, the calibratedorifice. This is a sharp-edged restriction that causesthe steam flow to “neck down” and then re-expand afterpassing through the orifice. As the steam accelerates topass through the restriction, its pressure drops, and thispressure drop, if measured, can be easily related to theflow rate. The calibrated orifice is one of a class of devicesknown as obstruction flowmeters, all of which work onthe same principle of restricting the flow and producinga measurable pressure drop. Other types of obstructionmeters are the ASME standard nozzle and the venturi.Orifices, although simple to manufacture and relativelyeasy to install, between flanges for example, are also subjectto wear, which causes them, over a period of time, togive unreliable readings. Nozzles and venturis, althoughmore expensive initially, tend to be more resistant to erosionand wear, and also produce less permanent pressuredrop once the steam re-expands to fill the pipe. With allof these devices, care must be exercised in the installation,since turbulence and flow irregularities produced byvalves, elbows, and fittings immediately upstream of theobstruction will produce erroneous readings.Figure 6.6 shows another type of flowmeteringdevice used for steam, called an annular averagingelement. The annular element principle is somewhatdifferent from the devices discussed above; it averagesthe pressure produced when steam impacts on the holesfacing into the flow direction, and subtracts from thisaverage impact pressure a static pressure sensed bya tube facing downstream. As with obstruction-typeflowmeters, the flow must be related to this pressuredifference.A device that does not utilize pressure drop forsteam metering applications is the vortex sheddingflowmeter, illustrated in Figure 6.7. A solid bar extendsthrough the flow, and as steam flows around the bar,vortices are shed alternately from one side to the otherFig. 6.6 Annular averaging element.in its wake, As the vortices shift from side to side, thefrequency of shedding can be detected with a thermalor magnetic detector, and this frequency varies directlywith the rate of flow. The vortex shedding meter is quiterugged, since the only function of the object extendinginto the flow stream is to provide an obstruction togenerate vortices; hence it can be made of heavy-dutystainless steel. Also, vortex shedding meters tend to berelatively insensitive to variations in the steam properties,since they produce a pulsed output rather than ananalog signal.The target flowmeter, not shown, is also suitablefor some steam applications. This type of meter usesa “target,” such as a small cylinder, mounted on theend of a metal strut that extends into the flow line. Thestrut is gauged to measure the force on the target, andif the properties of the fluid are accurately known, thisforce can be related to flow velocity. The target meteris especially useful when only intermittent measurementsare needed, as the unit can be “hot-tapped” in apipe through a ball valve and withdrawn when not inuse. The requirement of accurate property data limitsthe usefulness of this type of meter in situations wheresteam conditions vary considerably, especially wherehigh moisture is present.

138 ENERGY MANAGEMENT HANDBOOKFig. 6.8Weigh bucket technique for condensate measurement.Fig. 6.7 Vortex shedding flowmeter with various methodsfor sensing fluctuations.The devices discussed above are useful in permanentinstallations where it is desired to continuouslyor periodically measure steam flow; there is no simpleway to directly measure steam flow on a spotcheck basiswithout cutting into the system. There is, however, arelatively simple indirect method, illustrated in Figure6.8, for determining the rate of steam usage in systemswith unpressurized condensate return lines, or opensystems in which condensate is dumped to a drain. If adrain line is installed after the trap, condensate may becaught in a barrel and the weight of a sample measuredover a given period of time. Precautions must be takenwhen using this technique to assure that the flash steam,generated when the condensate drops in pressure as itpasses through the trap, does not bubble out of the barrel.This can represent both a safety hazard and an errorin the measurements due to the loss of mass in vaporform. The barrel should be partially filled with coldwater prior to the test so that flash steam will condenseas it bubbles through the water. An energy balance canalso be made on the water at the beginning and end ofthe test by measuring its temperature, and with properapplication of the steam tables, a check can be made toassure that the trap is not blowing through.Figure 6.9 illustrates another instrument which canbe used to monitor condensate flow on a regular basis,the rotameter. A rotameter indicates the flow rate of theliquid by the level of a specially shaped float which risesin a calibrated glass tube, suchthat its weight exactly balancesthe drag force of the flowing condensate.Measurements of this typecan be very useful in monitoringsystem performance, sinceany unusual change in steam orcondensate rate not associatedwith a corresponding change inproduction rate would tend to indicatean equipment malfunctionproducing poor efficiency.6.3.4 Computing the DollarValue of SteamIn analyzing energy conservationmeasures for steamsystems, it is important to establisha steam value in dollars perpound: this value will depend onthe steam pressure, the boiler efficiency,and the price of fuel.Steam may be valued fromtwo points of view. The morecommon approach, termed the“enthalpy method,” takes intoaccount only the heating capabilityof the steam, and is mostFig. 6.9Liquid rotameter.

STEAM AND CONDENSATE SYSTEMS 139appropriate when steam is used primarily for processheating. The enthalpy method can be illustrated withan example.Example: An oil refinery produces 200-psig saturatedsteam in a large boiler, some of which is used directlyin high-temperature processes, and some of which islet down to 30 psig through regulating valves for useat lower temperatures. The feedwater is added to theboiler at about 160°F. The boiler efficiency has beendetermined to be 82%, and boiler fuel is priced at $2.20million Btu. Establish the values of 200-psig and 30-psigsteam ($/lb).Using the enthalpy method, we determine theincreased heat content for each steam pressure requiredfrom the boiler if feedwater enters at 160°F. From Table6.2:total enthalpy of steam at 200 psig= 1199.3 Btu/lbtotal enthalpy of steam at 30 psig (ignoring superheatafter PRV)= 1172 Btu/lbenthalpy of feedwater at 160°F= approx. 130 Btu/lb (about the sameas saturated liquid at 160°F)heat added per lb of 200-psig steam= 1199.3 – 130 = 1069.3 Btu/lbheat added per lb of 30-psig steam= 1042 Btu/lbfuel Btu required per pound at 200 psig= 1069.3/0.82 = 1304 Btufuel Btu required per lb at 30 psig= 1042/0.82 = 1271 BtuWith boiler fuel priced at $2.20/million Btu,or .22 per 1000 Btu:value of 200-psig steam= 0.22 × 1.304 =.29/lb or $2.90/1000 lbvalue of 30-psig steam= 0.22 × 1.271 =.28/lb or $2.80/1000 lbAnother approach to the valuation of steam istermed the “availability” or “entropy” method and takesinto account not only the heat content of the steam, butalso its power-producing potential if it were expandedthrough a steam turbine. This method is most applicablein plants where cogeneration (the sequential generationof power and use of steam for process heat) is practiced.The availability method involves some fairly complexthermodynamic reasoning and will not be covered here.The reader should, however, be aware of its existencein analyzing the economics of cogeneration systems, inwhich there is an interchangeability between purchasedelectric power and in-plant power and steam generation.6.4 STEAM TRAPS AND THEIR APPLICATION6.4.1 Functions of Steam TrapsSteam traps are important elements of steam andcondensate systems, and may represent a major energyconservation opportunity (or problem, as the case maybe). The basic function of a steam trap is to allow condensateformed in the heating process to be drainedfrom the equipment. This must be done speedily toprevent backup of condensate in the system.Inefficient removal of condensate produces twoadverse effects. First, if condensate is allowed to backup in the steam chamber, it cools below the steam temperatureas it gives up sensible heat to the process andreduces the effective potential for heat transfer. Sincecondensing steam is a much more effective heat-transfermedium than stagnant liquid, the area for condensationis reduced, and the efficiency of the heat-transfer processis deteriorated. This results in longer cycle times forbatch processes or lower throughput rates in continuousheating processes. In either case, inefficient condensateremoval almost always increases the amount of energyrequired by the process.A second reason for efficient removal of condensateis the avoidance of “water hammer” in steamsystems. This phenomenon occurs when slugs of liquidbecome trapped between steam packets in a line. Thesteam, which has a much larger specific volume, canaccelerate these slugs to high velocity, and when theyimpact on an obstruction, such as a valve or an elbow,they produce an impact force not unlike hitting the elementwith a hammer (hence the term). Water hammercan be extremely damaging to equipment, and properdesign of trapping systems to avoid it is necessary.The second crucial function of a steam trap is tofacilitate the removal of air from the steam space. Aircan leak into the steam system when it is shut down,and some gas is always liberated from the water in theboiling process and carried through the steam lines. Airmixed with steam occupies some of the volume thatwould otherwise be filled by the steam itself. Each ofthese components, air and steam, contributes its shareto the total pressure exerted in the system; it is a fundamentalthermodynamic principle that, in a mixture ofgases, each component contributes to the pressure in the

140 ENERGY MANAGEMENT HANDBOOKsame proportion as its share of the volume of the space.For example, consider a steam system at 100 psia (notethat in this case it is necessary to use absolute pressures),with 10% of the volume air instead of steam. Therefore,from thermodynamics, 10% of the pressure, or 10 psia, iscontributed by the air, and only 90%, or 90 psia, by thesteam. Referring to Table 6.2, the corresponding steamtemperature is between 316 and 324°F, or approximately320°F. If the air were not present, the steam pressurewould be 100 psia, corresponding to a temperature ofabout 328°F, so the presence of air in the system reducesthe temperature for heat transfer. This means that moresteam must be generated to do a given heating job. Table6.8 indicates the temperature reduction caused by thepresence of air in various quantities at given pressures(shown in psig), and shows that the effective temperaturemay be seriously degraded.In actual operation the situation is usually evenworse than indicated in Table 6.8. We have consideredthe temperature reduction assuming that the air andsteam are uniformly mixed. In fact, on a real heatingsurface, as air and steam move adjacent to the surface,the steam is condensed out into a liquid, while the airstays behind in the form of vapor. In the region verynear the surface, therefore, the air occupies an evenlarger fraction of the volume than in the steam space asa whole, acting effectively as an insulating blanket onthe surface. Suffice it to say that air is an undesirableparasite in steam systems, and its removal is importantfor proper operation.Oxygen and carbon dioxide, in particular, have anotheradverse effect, and this is corrosion in condensateand steam lines. Oxygen in condensate produces pittingor rusting of the surface, which can contaminate thewater, making it undesirable as boiler feed, and CO 2 insolution with water forms carbonic acid, which is highlycorrosive to metallic surfaces. These components mustbe removed from the system, partially by good steamtrapping and partially by proper deaeration of condensate,as is discussed in a subsequent section.6.4.2 Types of Steam Traps and Their SelectionVarious types of steam traps are available on themarket, and the selection of the best trap for a givenapplication is an important one. Many manufacturersproduce several types of traps for specific applications,and manufacturers’ representatives should be consultedin arriving at a choice. This section will give a brief introductionto the subject and comment on its relevanceto improved energy utilization in steam systems.Steam traps may be generally classified into threegroups: mechanical traps, which work on the basis ofthe density difference between condensate and steam orair; thermostatic, which use the difference in temperaturebetween steam, which stays close to its saturationtemperature, and condensate, which cools rapidly; andthermodynamic, which functions on the difference inflow properties between liquids and vapors.Figures 6.10 and 6.11 show two types of mechanicaltraps in common use for industrial applications.Figure 6.10 illustrates the principle of the “bucket trap. “In the trap illustrated, an inverted bucket is placed overthe inlet line, inside an external chamber. The bucket isattached to a lever arm which opens and closes a valveas the bucket rises and falls in the chamber. As long ascondensate flows through the system, the bucket has anegative buoyancy, since liquid is present both insideand outside the bucket. The valve is open and condensateis allowed to drain continuously to the return line.As steam enters the trap it fills the bucket, displacingcondensate, and the bucket rises, closing off the valve.Noncondensable gases, such as air and CO 2 , bubblethrough a small vent hole and collect at the top of thetrap, to be swept out with flash steam the next time thevalve opens. Steam may also leak through the vent, butit is condensed on contact with the cool chamber walls,and collects as condensate in the chamber. The vent holeis quite small, so the rate of steam loss through this leakageaction is not excessive. As condensate again beginsto enter the bucket, it loses buoyancy and begins to dropuntil the valve opens and again discharges condensateand trapped air.Table 6.8 Temperature Reduction Caused by Air inSteam Systems—————————————————————————Temp. of Steam Mixed with VariousPercentages of AirTemp. of(by Volume)Pressure Steam, No ———————————————(psig)—————————————————————————Air Present 10 20 3010 240.1 234.3 228.0 220.925 267.3 261.0 254.1 246.450 298.0 291.0 283.5 275.175 320.3 312.9 204.8 295.9100 338.1 330.3 321.8 312.4—————————————————————————The float-and-thermostatic (F&T) trap, illustratedin Figure 6.11, works on a similar principle. In thiscase, instead of a bucket, a buoyant float rises and fallsin the chamber as condensate enters or is discharged.The float is attached to a valve, similar to the one in

STEAM AND CONDENSATE SYSTEMS 141Fig. 6.10 Inverted bucket trap.Fig. 6.11 Float and thermostatic steam trap.a bucket trap, which opens and closes as the ball risesand falls. Since there is no natural vent in this trapand the ball cannot distinguish between air and steam,which have similar densities, special provision must bemade to remove air and other gases from the system.This is usually done by incorporating a small thermostaticallyactuated valve in the top of the trap. At lowtemperature, the valve bellows contracts, opening thevent and allowing air to be discharged to the returnline. When steam enters the chamber, the bellows expands,sealing the vent. Some float traps are also availablewithout this thermostatic air-vent feature; externalprovision must then be provided to permit proper airremoval from the system. The F&T-type trap permitscontinuous discharge of condensate, unlike the buckettrap, which is intermittent. This can be an advantagein certain applications.Figure 6.12 illustrates a thermostatic steam trap. Inthis trap, a temperature-sensitive bellows expands andFig. 6.12 Thermostatic steam trap.contracts in response to the temperature of the fluid inthe chamber surrounding the bellows. When condensatesurrounds the bellows, it contracts, opening thedrain port. As steam enters the chamber, the elevatedtemperature causes the bellows to expand and seal thedrain. Since air also enters the chamber at a temperaturelower than that of steam, the thermostatic trap isnaturally self-venting, and it also is a continuous-draintypetrap. The bellows in the trap can be partially filledwith a fluid and sealed, such that an internal pressureis produced which counterbalances the external pressureimposed by the steam. This feature makes the bellowstypethermostatic trap somewhat self-compensating forvariations in steam pressure. Another type of thermostatictrap, which uses a bimetallic element, is also available.This type of trap is not well suited for applicationsin which significant variations in steam pressure mightbe expected, since it is responsive only to temperaturechanges in the system.The thermodynamic, or controlled disk steamtrap is shown in Figure 6.13. This type of trap is verysimple in construction and can be made quite compactand resistant to damage from water hammer. In a thermodynamictrap, a small disk covers the inlet orifice.Condensate or air, moving at relatively low velocity, liftsthe disk off its seat and is passed through to the outletdrain. When steam enters the trap, it passes through athigh velocity because of its large volume. As the steampasses through the space between the disk and its seat, itimpacts on the walls of the control chamber to producea rise in pressure. This pressure imbalance between theoutside of the disk and the side facing the seat causes itto snap shut, sealing off the chamber and preventing thefurther passage of steam to the outlet. When condensateagain enters the inlet side, the disk lifts off the seat andpermits its release.

142 ENERGY MANAGEMENT HANDBOOKFig. 6.14 Drain orifice.Fig. 6.13 Disk or thermodynamic steam trap.An alternative to conventional steam traps, thedrain orifice is illustrated in Figure 6.14. This deviceconsists simply of an obstruction to the flow of condensate,similar to the orifice flowmeter described inan earlier section but much smaller. This small holeallows the pressure in the steam system to force condensateto drain continuously into the lower-pressurereturn system. Obviously, if steam enters, rather thancondensate, it will also pass through the orifice and belost. The strategy of using drain orifices is to select anorifice size that permits condensate to drain at such arate that live steam seldom enters the system. Even ifsteam does occasionally pass through, the small size ofthe orifice limits the steam leakage rate to a value muchless than would be lost due to a “stuck-open” malfunctionof one of the types of traps discussed above. Drainorifices can be successfully applied in systems that havea well-defined and relatively constant condensate load.They are not suited for use where condensate load mayvary widely with operating conditions.As mentioned above, a number of operating requirementsmust be taken into consideration in selectingthe appropriate trap for a given application. Table6.9 lists these application considerations and presentsone manufacturer’s ratings on the performance of thevarious traps discussed above. In selecting a trap for agiven application, assistance from manufacturers’ representativesshould be obtained, since a great body ofexperience in actual service has been accumulated overthe years.6.4.3 Considerations in Steam Trap SizingAs mentioned earlier in this section, good energyconservation practice demands the efficient removal ofcondensate from process equipment. It is thus necessaryto assure that traps are properly sized for the givencondensate load. Grossly oversized traps waste steam byexcessive surface heat loss and internal venting, whileundersized traps permit accumulation of condensatewith resultant loss in equipment heat-transfer effectiveness.Steam traps are sized based in two specifications,the condensate load (e.g., in lbs/hr or gal/min) and thepressure differential across the trap (in psig). Section 6.3discussed various methods for estimating condensateloads expected under normal operating conditions.It is good practice to size the capacity of the trapbased on this expected load times a factor of safety toaccount for peaks at startup and fluctuations in normaloperating conditions. It is not unusual for startupcondensate loads to be three to four times higher thansteady operational loads, and in some applications theymay range up to 10 times the steady-state load.Table 6.10 presents typical factors of safety forcondensate capacity recommended by steam trap manufacturers.This indicates typical ranges of factor of safetyto consider in various applications. Although there isconsiderable variation in the recommended values, inboth energy and economic terms the cost of oversizingis ordinarily not prohibitive, and conservative safetyfactors are usually used. The exception to this rule ofthumb is in the sizing of disk-type traps, which may notfunction properly if loaded considerably below design.Drain orifices also must be sized close to normal operatingloads. Again, the advice of the manufacturer shouldbe solicited for the specific application in mind.The other important design specification is thepressure differential over which the trap will operate.Since pressure is the driving force that moves condensatethrough the trap and on to the receiver, trap capacitywill increase, for a given trap size, as the pressureincreases. The trap operating-pressure differential is notsimply the boiler pressure. On the upstream side of thetrap, steam pressure may drop through valves and fittingsand through heat-transfer passages in the processequipment. Thus the appropriate upstream pressureis the pressure at the trap inlet, which to a reasonableapproximation, can usually be considered to be theprocess steam pressure at the equipment. Back pressureon the outlet side of the trap must also be considered.This includes the receiver pressure (if the condensate

STEAM AND CONDENSATE SYSTEMS 143Table 6.9 Comparison of Steam Trap Characteristics———————————————————————————————————————————————————InvertedBellowsCharacteristic Bucket F&T Disk Thermostatic———————————————————————————————————————————————————Method of operation Intermittent Continuous Intermittent Continuous aEnergy conservation (time in Excellent Good Poor Fairservice)Resistance to wear Excellent Good Poor FairCorrosion resistance Excellent Good Excellent GoodResistance to Hydraulic shock Excellent Poor Excellent PoorVents air and CO 2 at steam Yes No No NotemperatureAbility to vent air at very low Poor Excellent NR b Goodpressure (1/4 psig)Ability to handle start-up air loads Fair Excellent Poor ExcellentOperation against back pressure Excellent Excellent Poor ExcellentResistance to damage from freezing c Good Poor Good GoodAbility to purge system Excellent Fair Excellent GoodPerformance on very light loads Excellent Excellent Poor ExcellentResponsiveness to slugs of Immediate Immediate Delayed DelayedcondensateAbility to handle dirt Excellent Poor Poor FairComparative physical size Large d Large Small SmallAbility to handle “flash steam” Fair Poor Poor PoorMechanical failure (open-closed) Open Closed Open e Closed f———————————————————————————————————————————————————a Can be intermittent on low load.b Not recommended for low-pressure operations.c Cast iron traps not recommended.d In welded stainless steel construction—medium.e Can fail closed due to dirt.f Can fail open due to wear.———————————————————————————————return system is pressurized), the pressure drop associatedwith flash steam and condensate flow through thereturn lines, and the head of water associated with risersif the trap is located at a point below the condensatereceiver. Condensate return lines are usually sized fora given capacity to maintain a velocity no greater than5000 ft/min of the flash steam. Table 6.11 shows theexpected pressure drop per 100 ft of return line whichcan be expected under design conditions. Referring tothe table, a 60-psig system, for example, returning condensateto an unpressurized receiver (0 psig) through a2-in. line, would have a return-line pressure drop of justunder 1/2 psi/100-ft run, and the condensate capacityof the line would be about 2600 lb/hr. The pressurehead produced by a vertical column of water is about1 psi/2 ft of rise. These components can be summed toestimate the back pressure on the system, and the appropriatepressure for sizing the trap is then the differencebetween the upstream pressure and the back pressure.Table 6.10 Typical Factors of Safety for Steam Traps(Condensate Flow Basis)—————————————————————————ApplicationFactor of Safety—————————————————————————Autoclaves 3-4Blast coils 3-4Dry cans 2-3Dryers 3-4Dry kilns 3-4Fan system heating service 3-4Greenhouse coils 3-4Hospital equipment 2-3Hot-water heaters 4-6Kitchen equipment 2-3Paper machines 3-4Pipe coils (in still air) 3-4Platen presses 2-3Purifiers 3-4Separators 3-4Steamjacketed kettles 4-5Steam mains 3-4Submerged surfaces 5-6Tracer lines 2-3Unit heaters 3-4

144 ENERGY MANAGEMENT HANDBOOKTable 6.11 Condensate Capacities and Pressure Drops for Return Lines aSupply Press.(psig) 5 15 30 60 100 200Return Press.(Psig) 0 0 5 0 5 10 0 5 10 20 0 5 10 20 30 0 5 10 20 30 50Pipe size (in.), schedule 40 pipe1/2 1,425 590 1,335 360 640 1,065 235 370 535 1,010 180 270 370 615 955 115 165 215 325 450 7604.0 4.0 5.3 4.0 5.3 6.5 4.0 5.3 6.5 8.9 4.0 5.3 6.5 8.9 11.3 4.0 5.3 6.5 8.9 11.3 15.93/4 2,495 1,035 2,340 635 1,125 1,855 415 650 940 1 770 310 470 645 1,085 1,675 200 285 375 570 795 1,3302.35 2.35 3.14 2.35 3.14 3.88 2.35 3.14 3.88 5.32 2.35 3.14 3.88 5.32 6.72 2.35 3.14 3.88 5.32 6.72 9.401 4,045 1,880 3,790 1,030 1,820 3,005 670 1,055 1,520 2,865 505 765 1,045 1,755 2,715 325 465 605 925 1. 285 2,1551.53 1.53 2.04 1.53 2.04 2.51 1.53 2.04 2.51 3.44 1.63 2.04 2.51 3.44 4.36 1.53 2.04 2.51 3.44 4.36 6.151-1/4 7,000 2,905 6,565 1,780 3,150 5,200 1,155 1,830 2,635 4,960 875 1,320 1,810 3,035 4,695 560 800 1,050 1,600 2,225 3.7350.95 0.95 1.26 .95 1.26 1.55 .95 1.26 1.55 2.13 95 1.26 1.55 2.13 2.69 0.95 1.26 1.55 2.13 2.69 3.801-1/2 9,530 3,955 8,935 2,425 4,290 7.080 1,575 2,490 3,585 6,750 2,190 1,795 2,465 4,135 6,395 760 1,090 1,430 2,175 3,025 5,0800.73 0.73 0.97 0.73 0.97 1.20 0.73 0.97 1.20 1.64 0.73 0.97 1.20 1.64 2.07 0.73 0.97 1.20 1.64 2.07 2.932 15,710 6,525 14,725 3,995 7,070 11,670 2,595 4,105 5,910 11,125 1,985 2.960 4,060 6,810 10,540 1,255 1,800 2,355 3,585 4,990 8,3750.48 0.48 0.64 0.48 0.64 0.79 0.48 0.64 0.79 1.08 0.48 0.64 0.79 1.08 1.37 0.48 0.64 0.79 1.08 1.37 1.932-1/2 22,415 9,305 21,005 5,700 10,085 16,650 3,705 5,855 8,430 15,875 2,800 4,225 5,795 9,720 15,035 1,790 2,565 3,380 5,115 7,120 11,9500.36 0.36 0.48 0.36 0.48 0.59 0.36 0.48 0.69 0.81 0.36 0.48 0.59 0.81 1.03 0.36 0.48 0.59 0.91 1.03 1.453 34,610 14,370 32,435 8,800 15,570 25,710 5,720 9,045 13,020 24 515 4,325 6,525 8,950 15,005 23,220 2,765 3,965 5.185 7,900 10,990 18,4500.26 0.26 0.34 0.26 0.34 0.42 0.26 0.34 0.42 0.58 0.26 0.34 0.42 0.58 0.73 0.26 0.34 0.42 0.58 0.73 1.033-1/2 46,285 19,220 43,380 11,765 20,825 34,385 7,650 12,095 17,410 32,785 5,785 6,725 11,970 20,070 31,050 3,695 5,300 6,940 10,565 14,700 24,6750.21 0.21 0.27 0.21 0.27 0.34 0.21 0.27 0.34 0.46 0.21 0.27 0.34 0.46 0.59 0.21 0.27 0.314 0.46 0.59 0.834 59,595 24,745 55,855 15,150 26,815 44,275 9,850 15,575 22,415 42,210 7,450 11,235 15,410 25.840 39,960 4,780 6,825 8,935 13,600 18,925 31,7700.17 0.17 0.23 0.17 0.23 0.28 0.17 0.23 0.28 0.38 0.17 0.23 0.28 0.38 0.49 0.17 0.23 0.28 0.36 0.49 0.255 93,655 38,890 87,780 23,810 42,140 69,580 15,480 24,475 35,230 66,335 11,705 17.660 24.220 40,610 62,830 7,475 10,725 14,040 21,375 29,745 49,9300.12 0.12 0.16 0.12 0.16 0.20 0.12 0.16 0.20 0.05 0.12 0.16 0.20 0.05 0.17 0.12 0.16 0.20 0.05 0.17 0.116 135,245 58,160 126,760 34,385 60,855 100,480 22,350 35,345 50,875 95,795 16,905 25,500 34,975 58,645 90,735 10,800 15,490 20,270 30,865 42,950 72,1050.10 0.10 0.13 0.10 0.13 0.04 0.10 0.13 0.04 0.05 0.10 0.13 0.04 0.05 0.01 0.10 0.13 0.04 0.05 0.01 0.018 234,195 97,245 219,505 59,540 105,380 173,995 38,705 61,205 88,095 165,880 29,270 44,160 60,565 101,650 157,115 18,700 26,820 35,105 53,450 74,175 124,8550.02 0.02 0.02 0.02 0.02 0.01 0.02 0.02 0.01 0.01 0.02 0.02 0.01 0.01 0.01 0.02 0.02 0.01 0.01 0.01 0.01a Return-line Capacity (lb/hr) with Pressure Drop (psi) for 100 ft of Pipe at a Velocity of 5000 ft/min.Table 6.12 shows a typical pressure-capacity tableextracted from a manufacturer’s catalog. The use of sucha table can be illustrated with the following example.Example: A steamjacketed platen press in a plastic laminationoperation uses about 500 lb/hr of 30-psig steamin normal operation. A 100-ft run of 1-in. pipe returnscondensate to a receiver pressurized to 5 psig: the receiveris located 15 ft above the level of the trap. Fromthe capacity differential pressure specifications in Table6.12, select a suitable trap for this application.Using a factor of safety of 3 from Table 6.10, a trapcapable of handling 3 × 500 = 1500 lb/hr of condensatewill be selected.To determine the system back pressure, add thereceiver pressure, the piping pressure drop, and thehydraulic head due to the elevation of the receiver.Entering Table 6.11 at 30-psig supply pressure and5-psig return pressure, the pipe pressure drop for a 1-in. pipe is just slightly over 2 psi at a condensate ratesomewhat higher than our 1500 lb/hr; a 2-psi pressuredrop is a reasonable estimate.The hydraulic head due to the 15-ft riser is 2 psi/ft× 15 = 7.5 psi. The total back pressure is, therefore,Table 6.12 Typical Pressure-Capacity Specifications forSteam TrapsCapacities (lb/hr)Model: A B C DDifferential pressure (psi)5 450 830 1600 290010 560 950 1900 350015 640 1060 2100 390020 680 880 1800 350025 460 950 1900 380030 500 1000 2050 400040 550 770 1700 380050 580 840 1 900 410060 635 900 2000 440070 660 950 2200 380080 690 800 1650 4000100 640 860 1800 3600125 680 950 2000 3900150 570 810 1500 3500200 — 860 1600 3200250 — 760 1300 3500300 — 510 1400 2700400 — 590 1120 3100450 — — 1200 3200

STEAM AND CONDENSATE SYSTEMS 1455 psi (receiver) + 7.5 psi (riser) + 2 psi (pipe) = 14.5 psior the differential pressure driving the condensate flowthrough the trap is 30 – 14.5 = 15.5 psi.From Table 6.12 we see that a Model C trap willhandle 2100 lb/hr at 15-psi differential pressure; thiswould then be the correct choice.6.4.4 Maintaining Steam Traps for Efficient OperationSteam traps can and do malfunction in twoways. They may stick in the closed position, causingcondensate to back up into the steam system, or theymay stick open, allowing live steam to discharge intothe condensate system. The former type of malfunctionis usually quickly detectable, since flooding of aprocess heater with condensate will usually so degradeits performance that the failure is soon evidence by asignificant change in operating conditions. This typeof failure can have disastrous effects on equipmentby producing damaging water hammer and causingprocess streams to back up into other equipment.Because of these potential problems, steam traps areoften designed to fail in the open position; for thisreason, they are among the biggest energy wasters inan industrial plant. Broad experience in large processplants using thousands of steam traps has shown that,typically, from 15 to 60% of the traps in a plant maybe blowing through, wasting enormous amounts ofenergy. Table 6.13 shows the cost of wasted 100-psigsteam (typical of many process plant conditions) forleak diameters characteristic of steam trap orifices. Athigher steam pressures, the leakage would be evengreater; the loss rate does not go down in direct proportionat lower steam pressures but declines at a rateproportional to the square root of the pressure. Forexample, a 1/8-in. leak in a system at 60 psig, insteadof the 100 psig shown in the table, would still wasteover 75% of the steam rate shown (the square root of60/100). The cost of wasted steam far outweighs thecost of proper maintenance to repair the malfunctions,and comprehensive steam trap maintenance programshave proven to be among the most attractive energyconservation investments available in large processplants. Most types of steam traps can be repaired, andsome have inexpensive replaceable elements for rapidturnaround.Table 6.13 Annual Cost of Steam Leaks—————————————————————————SteamWasted perLeak Diame- Month Cost per Cost perter (in.) (lb) a Month b Year—————————————————————————b1/16 13,300 $40 $4801/8 52,200 156 1,8901/4 209,000 626 7,8001/2 833,000 2,500 30,000—————————————————————————a Based on 100-psig differential pressure across the orifice.b Based on steam value of $3/1000 lb. Cost will scale in directproportion for other steam values.A major problem facing the energy conservationmanager is diagnosis of open traps. The fact that a trapis blowing through can often be detected by a rise intemperature at the condensate receiver, and it is quiteeasy to monitor this simple parameter. There are alsoseveral direct methods for checking trap operation. Figure6.15 shows the simplest approach for open condensatesystems where traps drain directly to atmosphericpressure. In proper normal operation, a stream of condensatedrains from the line together with a lazy cloudof flash steam, produced as the condensate throttlesacross the trap. When the trap is blowing through, aTable 6.14 Operating Sounds of Various Types of Steam TrapsTrap Proper Operation MalfunctioningDisk type (impulse of Opening and snap-closing of Rapid chattering of disk asthermodynamic) disk several times per minute steam blows throughMechanical type (bucket) Cycling sound of the bucket as Fails open—sound of steamit opens and closesblowing throughFails closed—no soundThermostatic type Sound of periodic discharge if Fails closed—no soundmedium to high load;possibly no sound if lightload; throttled discharge

146 ENERGY MANAGEMENT HANDBOOKFig. 6.15 Visual observation of steam trap operation in opensystem, (a) Proper operation, (b) Improper operation.well-defined jet of live steam will issue from the linewith either no condensate, or perhaps a condensate mistassociated with steam condensation at the periphery ofthe jet.Visual observation is less convenient in a closedcondensate system, but can be utilized if a test valve isplaced in the return line just downstream of the trap, asshown in Figure 6.16. This system has the added advantagethat the test line may be used to actually measurecondensate discharge rate as a check on equipmentefficiency, as discussed earlier. An alternative in closedcondensate return systems is to install a sight glass justdownstream of the trap. These are relatively inexpensiveand permit quick visual observation of trap operationwithout interfering with normal production.Another approach to steam trap testing is to observethe sound of the trap during operation. Table 6.14describes the sounds made by various types of trapsFig. 6.16 Visual observation of steam trap operation inclosed systems.during normal and abnormal operation. This methodis most effective with disk-type traps, although it canbe used to some extent with the other types as well.An industrial stethoscope can be used to listen to thetrap, although under many conditions, the characteristicsound will be masked by noises transmitted from otherparts of the system. Ultrasonic detectors may be usedeffectively in such cases; these devices are, in effect,electronic stethoscopes with acoustic filtering to makethem sensitive to sound and vibration only in the veryhigh frequency range. Steam blowing through a trapemits a very high-pitched sound, produced by intenseturbulence at the trap orifice, as contrasted with thelower-pitched and lower-intensity sound of liquid flowingthrough. Ultrasonic methods can, therefore, give amore reliable measure of steam trap performance thanconventional “listening” devices.A third approach to steam trap testing makes useof the drop in saturation temperature associated pressuredrop across the trap. Condensate tends to coolrapidly in contact with uninsulated portions of thereturn line, accentuating the temperature difference. Ifthe temperature on each side of the trap is measured,a sharp temperature drop should be evident. Table6.15 shows typical temperatures that can be expectedon the condensate side for various condensate pressures.In practice, the temperature drop method can berather uncertain, because of the range of temperaturesthe condensate may exhibit and because, in blowingthrough a stuck-open trap, live steam will, itself, undergosome temperature drop. For example, 85-psigsaturated steam blowing through an orifice to 15 psigwill drop from 328°F to about 300°F, and may then coolfurther by radiation and convection from uninsulatedsurfaces. From Table 6.15, the expected condensatesidetemperature is about 215 to 238°F for this pressure.Thus although the difference is still substantial,misinterpretation is possible, particularly if accuratemeasurements of the steam and condensate pressureson each side of the trap are not available.The most successful programs of steam trap diagnosisutilize a combination of these methods, coupledwith a regular maintenance program, to assure thattraps are kept in proper operating condition.This section has discussed the reasons why goodsteam trap performance can be crucial to successfulenergy conservation in steam systems. Traps must beproperly selected and installed for the given serviceand appropriately sized to assure efficient removal ofcondensate and gases. Once in service, expendituresfor regular monitoring and maintenance easily pay forthemselves in fuel savings.

STEAM AND CONDENSATE SYSTEMS 147Table 6.15 Typical Pipe Surface Temperatures for VariousOperating Pressures—————————————————————————Typical LineOperating Pressure Temperatures(psig)(°F)—————————————————————————0 190-21015 215-23845 248-278115 295-330135 304-340450 395-437—————————————————————————6.5 CONDENSATE RECOVERYCondensate from steam systems is wasted, or atleast used inefficiently, in many industrial operations.Yet improvements in the condensate system can offerthe greatest savings of any of the measures discussedin this chapter. In this section methods are presentedfor estimating the potential energy and mass savingsachievable through good condensate recovery, and theconsiderations involved in condensate recovery systemdesign are discussed. It is not possible in a brief surveyto provide a comprehensive guide to the detailed designof such systems. Condensate return systems can, in fact,be quite complex, and proper design usually requires acareful engineering analysis. The energy manager can,however, define the type of system best suited to therequirements and determine whether sufficient justificationexists for a comprehensive design study.6.5.1 Estimation of Heat and Mass Lossesin Condensate SystemsThe saturated liquid condensate produced whensteam condenses on a heating surface still retains asignificant fraction of the energy contained in the steamitself. Referring to Table 6.2, for example, it is seen thatat a pressure of about 80 psig, each pound of saturatedliquid contains about 295 Btu, or nearly 25% of the originalenergy contained in the steam at the same pressure.In some plants, this condensate is simply dischargedto a wastewater system, which is wasteful not onlyof energy, but also of water and the expense of boilerfeedwater treatment. Even if condensate is returned toan atmospheric pressure receiver, a considerable fractionof it is lost in the form of flash steam. Table 6.16 showsthe percent of condensate loss due to flashing fromsystems at the given steam pressure to a flash tank ata lower pressure. For example, if in the 80-psig systemdiscussed above, condensate is returned to a ventedreceiver instead of discharging it to a drain, nearly 12%is vented to the atmosphere as flash steam. Thus aboutTable 6.16 Percent of Mass Converted to Flash Steam in a Flash TankSteamFlash Tank Pressure (psig)Pressure(psig) 0 2 5 10 15 20 30 40 60 80 1005 1.7 1.0 010 2.9 2.2 1.4 015 4.0 3.2 2.4 1.1 020 4.9 4.2 3.4 2.1 1.1 030 6.5 5.8 5.0 3.8 2.6 1.7 040 7.8 7.1 6.4 5.1 4.0 3.1 1.3 060 10.0 9.3 8.6 7.3 6.3 5.4 3.6 2.2 080 11.7 11.1 10.3 9.0 8.1 7.1 5.5 4.0 1.9 0100 13.3 12.6 11.8 10.6 9.7 8.8 7.0 5.7 3.5 1.7 0125 14.8 14.2 13.4 12.2 11.3 10.3 8.6 7.4 5.2 3.4 1.8160 16.8 16.2 15.4 14.1 13.2 12.4 10.6 9.5 7.4 5.6 4.0200 18.6 18.0 17.3 16.1 15.2 14.3 12.8 11.5 9.3 7.5 5.9250 20.6 20.0 19.3 18.1 17.2 16.3 14.7 13.6 11.2 9.8 8.2300 22.7 21.8 21.1 19.9 19.0 18.2 16.7 15.4 13.4 11.8 10.1350 24.0 23.3 22.6 21.6 20.5 19.8 18.3 17.2 15.1 13.5 11.9400 25.3 24.7 24.0 22.9 22.0 21.1 19.7 18.5 16.5 15.0 13.4

148 ENERGY MANAGEMENT HANDBOOK3% of the original steam energy (0.12 × 0.25) goes upthe vent pipe. The table shows that this loss could bereduced by half by operating the flash tank at a pressureof 30 psig, providing a low-temperature steam source forpotential use in other parts of the process. Flash steamrecovery is discussed later in more detail.Even when condensate is fully recovered using oneof the methods to be described below, heat losses canstill occur from uninsulated or poorly insulated returnlines. These losses can be recovered very cost effectivelyby the proper application of thermal insulation as discussedin Chapter 15.6.5.2 Methods of Condensate Heat RecoverySeveral options are available for recovery of condensate,ranging in cost and complexity from simpleand inexpensive to elaborate and costly. The choice ofwhich option is best depends on the amount of condensateto be recovered, other uses for its energy, and thepotential cost savings relative to other possible investments.The simplest system, which can be utilized if condensateis presently being discharged, is the installationof a vented flash tank which collects condensate fromvarious points of formation and cools it sufficiently to allowit to be delivered back to the boiler feed tank. Figure6.17 schematically illustrates such a system. It consists ofa series of collection lines tying the points of condensategeneration to the flash tank, which allows the liquid toseparate from the flash steam; the flash steam is ventedto the atmosphere through an open pipe. Condensatemay be gravity-drained through a strainer and a trap.To avoid further generation of flash steam, a cooling legmay be incorporated to cool the liquid below its saturationtemperature.Flash tanks must be sized to produce proper separationof the flash steam from the liquid. As condensateis flashed, steam will be generated rather violently,and as vapor bubbles burst at the surface, liquid maybe entrained and carried out through the vent. Thisrepresents a nuisance, and in some cases a safety hazard,if the vent is located in proximity to personnel orequipment. Table 6.17 permits the estimation of flashtank size required for a given application. Althoughstrictly speaking, flash tanks must be sized on the basisof volume, if a typical length to diameter of about 3:1is assumed, flash tank dimensions can be representedas the product of diameter times length, which has theunits of square feet (area), even though this particularproduct has no direct physical significance. Consider, forexample, the sizing of a vented flash tank for collectionof 80-psig condensate at a rate of about 3000 lb/hr. Fora flash tank pressure of 0 psig (atmospheric pressure),the diameter-length product is about 2.5 per 1000 lb.Therefore, a diameter times length of 7.5 ft 2 would beneeded for this application. A tank 1.5 ft in diameterby 5 ft long would be satisfactory. Of course, for flashtanks as with other condensate equipment, conservativedesign would suggest the use of an appropriate safetyfactor.As noted above, venting of flash steam to the atmosphereis a wasteful process, and if significant amountsof condensate are to be recovered, it may be desirableto attempt to utilize this flash steam. Figure 6.18 showsa modification of the simple flash tank system to accomplishthis. Rather than venting to the atmosphere,the flash tank is pressurized and flash steam is pipedto a low-pressure steam main, where it can be utilizedfor process purposes. From Table 6.17 it will be notedthat the flash tank can be smaller in physical size at elevatedpressure, although, of course, it must be properlydesigned for pressure containment. If in the exampleabove the 80-psig condensate were flashed in a 15-psigtank, only about 2.7 ft 2 of diameter times length wouldbe required. A tank 1 ft in diameter by 3 ft long couldbe utilized. Atmospheric vents are usually provided forautomatic pressure relief and to allow manual ventingif desired.Fig. 6.17 Flash tank vented to atmosphere.

STEAM AND CONDENSATE SYSTEMS 149Fig. 6.18 Pressurized flash tank discharging to low-pressuresteam system.If pressurized flash steam is to be used, the cost ofpiping to set up a low-pressure steam system may besignificant, particularly if the flash tank is remote fromthe potential low-pressure steam applications. Thus it isdesirable to plan such a system to minimize these pipingcosts by generating the flash steam near its point ofuse. Figure 6.19 illustrates such an application. Here anair heater having four sections formerly utilized 100-psisteam in all sections. Because the temperature differen cebetween the steam and the cold incoming air is largerthan the difference at the exit end, the condensate loadwould be unevenly balanced among the four sections,with the heaviest load in the first section; lower-temperaturesteam could be utilized here. In the revised arrangementshown, 100-psi steam is used in the last threesections; condensate is drained to a 5-psi flash tank,where low-pressure steam is generated and piped tothe first section, substantially reducing the overall steamload to the heater. Note that for backup purposes, a pressure-controlledreducing valve has been incorporated tosupplement the low-pressure flash steam at light-loadconditions. This example shows how flash steam canbe used directly without an expensive piping system todistribute it. A similar approach could apply to adjacentpieces of equipment in a multiple-batch operation.Table 6.17 Flash Tank Sizing a———————————————————————————————————————————————————SteamFlash Tank Pressure (psig)Pressure ——————————————————————————————————————————————(psig) 0 2 5 10 15 20 30 40 60 80 100———————————————————————————————————————————————————400 5.41 4.70 3.89 3.01 2.44 2.03 1.49 1.15 0.77 0.56 0.42350 5.14 4.45 3.66 2.84 2.28 1.91 1.38 1.07 0.70 0.51 0.37300 4.86 4.15 3.42 2.62 2.11 1.75 1.26 0.96 0.62 0.44 0.31250 4.41 3.82 3.12 2.39 1.91 1.56 1.11 0.85 0.52 0.37 0.25200 3.98 3.40 2.80 2.12 1.68 1.37 0.97 0.72 0.43 0.28 0.18175 3.75 3.20 2.61 1.95 1.57 1.26 0.87 0.64 0.38 0.23 0.15160 3.60 3.08 2.50 1.86 1.46 1.19 0.80 0.59 0.34 0.21 0.12150 3.48 2.98 2.41 1.80 1.40 1.14 0.77 0.56 0.31 0.19 0.10140 3.36 2.86 2.31 1.72 1.35 1.08 0.72 0.52 0.29 0.16 0.08130 3.24 2.76 2.23 1.65 1.29 1.02 0.67 0.49 0.26 0.14 0.07120 3.12 2.65 2.15 1.57 1.22 0.97 0.64 0.44 0.23 0.12 0.04110 2.99 2.52 2.05 1.50 1.15 0.91 0.58 0.40 0.20 0.09 0.02100 2.85 2.41 1.92 1.40 1.07 0.85 0.53 0.36 0.16 0.0690 2.68 2.26 1.81 1.30 0.99 0.77 0.48 0.31 0.13 0.0580 2.52 2.12 1.67 1.18 0.90 0.68 0.42 0.25 0.0970 2.34 1.95 1.55 1.08 0.81 0.61 0.35 0.20 0.0460 2.14 1.77 1.39 0.96 0.70 0.52 0.27 0.1450 1.94 1.59 1.22 0.81 0.58 0.41 0.20 0.0840 1.68 1.36 1.02 0.67 0.44 0.30 0.1130 1.40 1.10 0.81 0.50 0.29 0.1620 1.06 0.81 0.55 0.28 0.1212 0.75 0.48 0.2810 0.62 0.42 0.23———————————————————————————————————————————————————a Flash tank area (ft2 ) = diameter X length of horizontal tank for discharge of 1000 lb/hr of condensate.

150 ENERGY MANAGEMENT HANDBOOKFig. 6.19 Flash steam utilization within a process unit.Although the utilization offlash steam in a low-pressure systemappears to offer an almost “free”energy source, its practical applicationinvolves a number of problemsthat must be carefully considered.These are all essentially economic innature.As mentioned above, the quantityof condensate and its pressure(thus yielding a given quantity offlash steam) must be sufficiently largeto provide a significant amount ofavailable energy at the desired pressure.System costs do not go up insimple proportion to capacity. Rather,there is a large initial cost for piping,installation of the flash tank,and other system components, andtherefore the overall cost per unit ofheat recovered becomes significantly less as the systembecomes larger. The nature of the condensate-producingsystem itself is also important. For example, if condensateis produced at only two or three points from largesteam users, the cost of the condensate collection systemwill be considerably less than that of a system in whichthere are many small users.Another important consideration is the potentialfor application of the flash steam. The availability of5000 lb/hr of 15-psig steam is meaningless unless thereis a need for a heat source of this magnitude in the250°F temperature range. Thus potential uses must beproperly matched to the available supply. Flash steam ismost effectively utilized when it can supplement an existinglow-pressure steam supply rather than providingthe sole source of heat to equipment. Not only must thetotal average quantity of flash steam match the needsof the process, but the time variations of source anduser must be taken into account, since steam cannot beeconomically stored for use at a later time. Thus flashsteam might not be a suitable heat source for sequentialbatch processes in which the number of operating unitsis small, such that significant fluctuations in steam demandexist.When considering the possible conversion to lowpressuresteam of an existing piece of equipment presentlyoperating on high-pressure steam, it is importantto recognize that steam pressure can have a significanteffect on equipment operation. Since a reduction insteam pressure also means a reduction in temperature,a unit may not have adequate heating-surface area toprovide the necessary heat capacity to the process at reducedpressure. Existing steam distribution piping maynot be adequate, since steam is lower in density at lowpressure than at high pressure. Typically, larger pipingis required to transport the low-pressure vapor at acceptablevelocities. Although one might expect that theheat losses from the pipe surface might be lower withlow-pressure steam because of its lower temperature, infact, this may not be the case if a large pipe (and hencelarger surface area) is needed to handle the lower-pressurevapor. This requirement will also make insulationmore expensive.When flash steam is used in a piece of equipment,the resulting low-pressure condensate must stillbe returned to a receiver for delivery back to the boiler.Flash steam will again be produced if the receiver isvented, although somewhat less than in the flashingof high-pressure condensate. This flash steam and thatproduced from the flash tank condensate draining intothe receiver will be lost unless some additional provisionis made for its recovery, as shown in Figure 6.20. In thissystem, rather than venting to the atmosphere, the steamrises through a cold-water spray, which condenses it.This spray might be boiler makeup water, for example,and hence the energy of the flash steam is used formakeup preheat. Not only is the heat content of the flashsteam saved, but its mass as well, reducing makeupwaterrequirements and saving the incremental costsof makeup-water treatment. This system has the addedadvantage that it produces a deaerating effect on thecondensate and feedwater. If the cold spray is meteredso as to produce a temperature in the tank above about190°F, dissolved gases in the condensate and feedwater,

STEAM AND CONDENSATE SYSTEMS 151Fig. 6.20 Flash steam recovery in spray tank.particularly oxygen and CO 2 , will come out of solution,and since they are not condensed by the cold-waterspray, will be released through the atmospheric vent. Aswith flash steam systems, this system (usually termed a“barometric condenser” or “spray deaerator”) requirescareful consideration to assure its proper application.The system must be compatible with the boiler feedwatersystem, and controls must be provided to coordinateboiler makeup demands with the condensate load.An alternative approach to the barometric condenseris shown in Figure 6.21. In this system, condensateis cooled by passing it through a submerged coil in theflash tank before it is flashed. This reduces the amountof flash steam generated. Cold-water makeup (possiblyboiler feedwater) is regulated by a temperature-controlledvalve.The systems described above have one featurein common. In all cases, the final condensate state isatmospheric pressure, which may be required to permitreturn of the condensate to the existing boiler-feedwatermakeup tank. If condensate can be returned at elevatedpressure, a number of advantages may be realized.Figure 6.22 shows schematics of two pressurizedcondensate return systems. Condensate is returned,in some cases without the need for a steam trap, toa high-pressure receiver, which routes the condensatedirectly back to the boiler. The boiler makeup unitand/or deaerator feeds the boiler in parallel with thecondensate return unit, and appropriate controls mustbe incorporated to coordinate the operation of the twounits. Systems such as the one shown in Figure 6.22aare available for condensate pressures up to about 15psig. For higher pressures, the unit can be used in conjunctionwith a flash tank, as shown in Figure 6.22b.This system would be suitable where an applicationfor 15-psig steam is available. These elevated-pressuresystems represent an attractive option in relatively low-Fig. 6.21 Flash tank with condensate precooling.pressure applications, such as steam-driven absorptionchillers. When considering them, care must be exercisedto assure that dissolved gases in the boiler makeup areat suitable levels to avoid corrosion, since the naturaldeaeration effect of atmospheric venting is lost.One of the key engineering considerations thatmust be accounted for in the design of all the systemsdescribed above is the problem of pumping high-temperaturecondensate. To understand the nature of theproblem, it is necessary to introduce the concept of “netpositive suction head” (NPSH) for a pump. This termmeans the amount of static fluid pressure that must beprovided at the inlet side of the pump to assure thatno vapor will be formed as the liquid passes throughthe pump mechanism, a phenomenon known as cavitation.As liquid moves into the pump inlet from an initiallystatic condition, it accelerates and its pressure dropsrather suddenly. If the liquid is at or near its saturationtemperature in the stationary condition, this sudden dropin pressure will produce boiling and the generation ofvapor bubbles. Vapor can also be generated by air comingout of solution at reduced pressure. These bubblestravel through the pump impeller, where the fluid pressurerises, causing the bubbles to collapse. The inrush ofliquid into the vapor space produces an impact on theimpeller surface which can have an effect comparable to

152 ENERGY MANAGEMENT HANDBOOKFig. 6.22 Pressurized condensate receiver systems, (a)Low-pressure process requirements, (b) Flash tank for usewith high-pressure systems.sandblasting. Clearly, this is deleterious to the impellerand can cause rapid wear. Most equipment operators arefamiliar with the characteristic “grinding” sound of cavitationin pumps when air is advertently allowed to enterthe system, and the same effect can occur due to steamgeneration in high-temperature condensate pumping.To avoid cavitation, manufacturers specify aminimum pressure above saturation which must bemaintained on the inlet side of the pump, such that,even when the pressure drops through the inlet port,saturation or deaeration conditions will not occur. Thisminimum-pressure requirement is termed the net positivesuction head.For condensate applications, special low-NPSHpumps have been designed. Figure 6.23 illustrates thedifference between a conventional pump and a low-NPSH pump. In the conventional pump (Figure 6.23a)fluid on the suction side is drawn directly into the impellerwhere the rapid pressure drop occurs in the entrypassage. In the low-NPSH pump (Figure 6.23b) a smallFig. 6.23 Conventional and low-NPSH pumps, (a) Conventionalcentrifugal pump, (b) Low-NPSH centrifugalpump.“preimpeller” provides an initial pressure boost to theincoming fluid, with relatively little drop in pressure atthe entrance. This extra stage of pumping essentiallyprovides a greater head to the entry passage of the mainimpeller, so that the system pressure at the suction sideof the pump can be much closer to saturation conditionsthan that required for a conventional pump. Low-NPSHpumps are higher in price than conventional centrifugalpumps, but they can greatly simplify the problem ofdesign for high-temperature condensate return, and can,in some cases, actually reduce overall system costs.An alternative device for the pumping of condensate,called a “pumping trap,” utilizes the pressureof the steam itself as the driving medium. Figure 6.24illustrates the mechanism of a pumping trap. Condensateenters the inlet side and rises in the body until itactivates a float-operated valve, which admits steam orcompressed air into the chamber. A check valve preventscondensate from being pushed back through theinlet port, and another check valve allows the steam or

STEAM AND CONDENSATE SYSTEMS 153Fig. 6.24 Pumping trap.6.5.3 Overall Planning Considerations inCondensate Recovery SystemsAs mentioned earlier, condensate recovery systemsrequire careful engineering to assure that they are compatiblewith overall plant operations, that they are safeand reliable, and that they can actually achieve theirenergy-efficiency potential. In this section a few of theoverall planning factors that should be considered areenumerated and discussed.1. Availability of Adequate Condensate Sources.An energy audit should be performed to collect detailedinformation on the quantity of condensate available fromall of the various steam using sources in the plant andthe relevant data associated with these sources. Suchair pressure to drive it out through the exit side. Whenthe condensate level drops to a predetermined position,the steam or air valve is closed, allowing the pumpingcycle to start again. Pumping traps have certaininherent advantages over electrically driven pumps forcondensate return applications. They have no NPSH requirement,and hence can handle condensate at virtuallyany temperature without regard to pressure conditions.They are essentially self-regulating, since the condensatelevel itself determines when the trap pumps; thus noauxiliary electrical controls are required for the system.This has another advantage in environments whereexplosion-proofing is required, such as refineries andchemical plants. Electrical lines need not be run to thesystem, since it utilizes steam as a driving force, andthe steam line is usually close at hand. Pumping trapsoperate more efficiently using compressed air, if available,because when steam is introduced to the chamber,some of it condenses before its pressure can drive thecondensate out. Thus for the same pressure, more steamis required to give the same pumping capacity as compressedair. The disadvantages of pumping traps aretheir mechanical complexity, resulting in a susceptibilityto maintenance problems, and the fact that they areavailable only in limited capacities.The engineering of a complete condensate recoverysystem from scratch can be a rather involved process,requiring the design of tanks, plumbing, controls, andpumping devices. For large systems, there is little alternativeto engineering and fabricating the system tothe specific plant requirements. For small- to moderate-capacityapplications, however, packaged systemsincorporating all of the foregoing components are commerciallyavailable. Figures 6.25a and b show examplesof two such systems, the former using electrically drivenlow-NPSH pumps, and the latter utilizing a pumpingtrap (the lower unit in the figure).Fig. 6.25 Pressurized condensate return systems.

154 ENERGY MANAGEMENT HANDBOOKdata include, for example, condensate pressure, quantity,and source location relative to other steam-using equipmentand to the boiler room. Certain other informationmay also be pertinent. For example, if the condensate iscontaminated by contact with other process streams, itmay be unsuitable for recovery. It is not valid to assumethat steam used in the process automatically results inrecoverable condensate. Stripping steam used in refiningof petroleum and in other separation processes is a goodexample.2. Survey of Possible Flash Steam Applications.The process should be surveyed in detail to assesswhat applications presently using first-generation steammight be adaptable to the use of flash steam or to heatrecovered from condensate. Temperatures and typicalheat loads are necessary but not sufficient. Heat-transfercharacteristics of the equipment itself may be important.A skilled heat-transfer engineer, given the presentoperating characteristics of a process heater can makereasonable estimates of that heater’s capability to operateat a lower pressure.3. Analysis of Condensate and Boiler FeedwaterChemistry. To assure that conditions in the boiler arekept in a satisfactory state to avoid scaling and corrosion,a change in feedwater treatment may be requiredwhen condensate is recovered and recycled. Water samplesfrom the present condensate drain, from the boilerblow-down, from the incoming water source, and fromthe outlet of the present feedwater treatment systemshould be obtained. With this information, a water treatmentspecialist can analyze the overall water chemicalbalance and assure that the treatment system is properlyconfigured to maintain good boiler-water conditions.4. Piping Systems. A layout of the present steamand condensate piping system is needed to assess theneed for new piping and the adequacy of existing runs.This permits the designer to select the best locations forflash tanks to minimize the need for extensive new piping.5. Economic Data. The bottom line on any energyrecovery project, including condensate recovery, is itsprofitability. In order to analyze the profitability of thesystem, it is necessary to estimate the quantity of heatrecovered, converted into its equivalent fuel usage, thecosts and cost savings associated with the water treatmentsystem, and savings in water cost. In addition tothe basic capital and installation costs of the condensaterecovery equipment, there may be additional costs associatedwith modification of existing equipment tomake it suitable for flash steam utilization, and therewill almost certainly be a cost of lost production duringinstallation and checkout of the new system.6.6 SUMMARYThis chapter has discussed a number of considerationsin effecting energy conservation in industrial andcommercial steam systems. Good energy managementbegins by improving the operation of existing systems,and then progresses to evaluation of system modificationsto maximize energy efficiency. Methods and datahave been presented to assist the energy manager inestimating the potential for savings by improving steamsystem operations and by implementing system designchanges. As with all industrial and commercial projects,expenditure of capital must be justified by reductionsin operating expense. With energy costs rising at a ratesubstantially above the rate of increase of most equipment,it is clear that energy conservation will continue toprovide ever more attractive investment opportunities.BIBLIOGRAPHYArmstrong Machine Works, “Steam Conservation Guidelines for CondensateDrainage,” Bulletin M101, Three Rivers, Mich., 1976.Babcock and Wilcox Corporation, Steam: Its Generation and Use, 38thed., New York, 1972.Department of Commerce, National Bureau of Standards, EnergyConservation Program Guide, for Industry and Commerce, NBSHandbook 115, Washington, D.C., 1974.Department of Energy, United Kingdom, Utilization of Steam, for Processand Heating, Fuel Efficiency Booklet 3, London, 1977.Department of Energy, United Kingdom, Flash Steam and Vapour Recovery,Fuel Efficiency Booklet 6, London, 1977.Department of Energy, United Kingdom, How to Make the Best Use ofCondensate, Fuel Efficiency Booklet 9, London, 1978.IDA, MASAHIRO, “Utilization of Low Pressure Steam and CondensateRecovery,” TLV Technical Information Bulletin 5123, TLVCompany Ltd., Tokyo, 1975.ISLES, DAVID L., “Maintaining Steam Traps for Best Efficiency,” HydrocarbonProcessing, Vol. 56, No. 1 (1977).JESIONOWSKI, J.M., W.E. DANEKIND, and W.R. RAGER, “SteamLeak Evaluation Technique: Quick and Easy to Use,” Oil andGas Journal, Vol. 74, No. 2 (1976).KERN, DONALD Q., Process Heat Transfer, McGraw-Hill, New York,1950.MAY, DONALD, “First Steps in Cutting Steam Costs,” Chemical Engineering,Vol. 80, No. 26 (1973).MONROE, E.S., JR., “Select the Right Steam Trap,” Chemical Engineering,Vol. 83, No. 1 (1976).MOWER, JOSEPH, “Conserving Energy in Steam Systems,” PlantEngineering, Vol. 32, No. 8 (1978).Sarco Company, Inc., Hook-Up Designs for Steam and Fluid Systems, 5thed., Allentown, Pa., 1975.

CHAPTER 7COGENERATION AND DISTRIBUTED GENERATIONJORGE B. WONGGeneral Electric CompanyEvansville, IndianaJOHN M. KOVACIKGeneral Electric Co.Retired7.1 INTRODUCTIONCogeneration is broadly defined as the coincident orsimultaneous generation of combined heat and power(CHP). In a true cogeneration system a “significant” portionof the generated or recovered heat must be used ina thermal process as steam, hot air, hot water, etc. Thecogenerated power is typically in the form of mechanicalor electrical energy.The power may be totally used in the industrialplant that serves as the “host” of the cogeneration system,or may be partially/totally exported to a utilitygrid. Figure 7.1 illustrates the potential of saving inprimary energy when separate generation of heat andelectrical energy is substituted by a cogeneration system.This chapter presents an overview of current design,analysis and evaluation procedures.The combined generation of useful heat and poweris not a new concept. The U.S. Department of Energy(1978) reported that in the early 1900s, 58% of the totalpower produced by on-site industrial plants was cogenerated.However, Pohmeros (1981) stated that by 1950,on-site CHP generation accounted for only 15 percentof total U.S. electrical generation; and by 1974, thisfigure had dropped to about 5 percent. In Europe, theexperience has been very different. The US Departmentof Energy, US-DOE (1978), reported that “historically,industrial cogeneration has been five to six times morecommon in some parts of Europe than in the U.S.” In1972, “16% of West Germany’s total power was cogeneratedby industries; in Italy, 18%; in France, 16%; andin the Netherlands, 10%.”Since the promulgation of the Public Utilities RegulatoryPolicies Act of 1978 (PURPA), however, U.S. cogenerationdesign, operation and marketing activities havedramatically increased, and have received a much largerattention from industry, government and academy. As aresult of this incentive, newer technologies such as variouscombined cycles have achieved industrial maturity.Also, new or improved cogeneration-related processesand equipment have been developed and are activelymarketed; e.g. heat recovery steam generator and ductburners, gas-engine driven chillers, direct and indirectfired two-stage absorption chiller-heaters, etc.The impact of cogeneration in the U.S., both asan energy conservation measure and as a means tocontribute to the overall electrical power generationcapacity cannot be overemphasized. SFA Pacific Inc.(1990) estimated that non-utility generators (NUGs)produce 6-7% of the power generated in the US. Andthat NUGs will account for about half of the 90,000MW that will be added during the next decade. Morerecently, Makansi (1991) reported that around 40,000MW of independently produced and/or cogenerated(IPP/COGEN) power was put on line since the establishmentof PURPA. This constitutes a significant partof new capacity. Makansi also reports that another60,000 MW of IPP/COGEN power is in constructionand in development. This trend is likely to increasesince CHP is a supplementary way of increasing the existingU.S. power generation capacity. Thus, distributedgeneration (DG) is arising as a serious CHP alternative(see Section 7.2.2.4).PURPA is considered the foremost regulatoryinstrument in promoting cogeneration, IPPs and/orNUGs. Thus, to promote industrial energy efficiencyand resource conservation through cogeneration, PUR-PA requires electric utilities to purchase cogeneratedpower at fair rates to both, the utility and the generator.PURPA also orders utilities to provide supplementaryand back-up power to qualifying facilities QFs).The Promulgation of the Energy Policy Act of1992 EPAct (1992) have brought a “deregulated” andcompetitive structure to the power industry. EPAct-1992 establishes the framework that allows the possibilityof “Retail Wheeling” (anyone can purchase powerfrom any generator, utility or otherwise). This is inaddition to the existing Wholesale Wheeling or electricitytrade that normally occurs among U.S. utilities.EPAct-1992’s main objective is to stimulate competitionin the electricity generation sector to and to reduceelectricity costs. The impact of EPAct 1992, deregulation,and retailed wheeling on cogeneration can be155

156 ENERGY MANAGEMENT HANDBOOKA) Separate generation of heat and electricity in boiler and conventional condensation power plant.B) Cogeneration of heat and power.C) Integration of a cogeneration system and an industrial process.Figure 7.1 Potential of saving in primary energy when separate generation of heat and electrical energy is substitutedby an industrial cogeneration system.

COGENERATION AND DISTRIBUTED GENERATION 157tremendous, since it creates a new class of generatingfacility called Exempt Wholesale Generators (EWGs).Economies of scale and scope—i.e. the natural higherefficiency of larger CHP plants—could trigger the growof cogeneration based EWGs. In addition, EPAct-1992will open the transmission grid to utilities and NUGsby ordering FERC to allow open transmission access toall approved EWGs. However, specific retail wheelingrules have not been legislated at the time of this writing.So far, only a few retail wheeling cases have beenstarted. Thus, the actual impact of deregulation oncogeneration is considered to be uncertain. Nevertheless,some consider that retail wheeling/deregulationand cogeneration could have a synergistic effect andshould be able to support each other.7.2. COGENERATION SYSTEM DESIGNAND ANALYSISThe process of designing and evaluating a cogenerationsystem has so many factors that it has been comparedto the Rubik’s cube—the ingenious game to arrange amulti-colored cube. The change in one of the cube faceswill likely affect some other face. The most importantfaces are: fuel security, regulations, economics, technology,contract negotiation and financing.7.2.1 General Considerations and DefinitionsKovacik (1982) indicates that although cogenerationshould be evaluated as a part of any energymanagement plan, the main prerequisite is that a plantshows a significant and concurrent demand for heat andpower. Once this scenario is identified, he states that cogenerationsystems can be explored under the followingcircumstances:1. Development of new facilities2. Major expansions to existing facilities which increaseprocess heat demands and/or process energyrejection.3. When old process and/or power plant equipmentis being replaced, offering the opportunity to upgradethe energy supply system.The following terms and definitions are regularlyused in the discussion of CHP systems.Industrial Plant: the facility requiring process heatand electric and/or shaft power. It can be a processplant, a manufacturing facility, a college campus,etc. See Figure 7.1c.Process Heat (PH): the thermal energy requiredin the industrial plant. This energy is supplied assteam, hot water, hot air, etc.Process Returns (PR): the fluid returned from theindustrial plant to the cogeneration system. Forsystems where the process heat is supplied assteam, the process returns are condensate.Net Heat to Process (NHP): the difference betweenthe thermal energy supplied to the industrial plantand the energy returned to the cogeneration system.Thus, NHP = PH – PR. The NHP may or maynot be equal to the actual process heat demand(PH).Plant Power Demand (PPD): the electrical poweror load demanded (kW or MW) by the industrialplant. It includes the power required in for industrialprocesses, air-conditioning, lighting, etc.Heat/Power Ratio (H/P): the heat-to-power ratio ofthe industrial plant (demand), or the rated heatto-powerratio of the cogeneration system or cycle(capacity).Topping Cycles: thermal cycles where power isproduced prior to the delivery of heat to the industrialplant. One example is the case of heat recoveredfrom a diesel-engine generator to producesteam and hot water. Figure 7.2 shows a dieselengine topping cycle.Bottoming Cycle: power production from the recoveryof heat that would “normally” be rejectedto a heat sink. Examples include the generationof power using the heat from various exothermicchemical processes and the heat rejected from kilnsused in various industries. Figure 7.3 illustrates abottoming cycle.Combined Cycle: this is a combination of the twocycles described above. Power is produced in atopping cycle—typically a gas-turbine generator.Then, heat exhausted from the turbine is used toproduce steam; which is subsequently expanded ina steam turbine to generate more electric or shaftpower. Steam can also be extracted from the cycleto be used as process heat. Figure 7.4 depicts acombined cycle.

158 ENERGY MANAGEMENT HANDBOOKFigure 7.2 Diesel engine topping cycle.Figure 7.3 Steam turbine bottoming cycle.Figure 7.4 Combined cycle: gas-turbine/generator set,waste heat recovery boiler and steam-turbine/generator set.

COGENERATION AND DISTRIBUTED GENERATION 159Prime Mover: a unit of the CHP system thatgenerates electric or shaft power. Typically, it is agas turbine generator, a steam turbine drive or adiesel-engine generator.7.2.2 Basic Cogeneration SystemsMost cogeneration systems are based on primemovers such as steam turbines, gas turbines, internalcombustion engines and packaged cogeneration. Table7.1 shows typical performance data for various cogenerationsystems. Figures in this table (and in this chapter)are based on higher heating values, unless statedotherwise.7.2.2.1 Steam Turbine SystemsSteam turbines are currently used as prime moversin topping, bottoming and combined cycles. There aremany types of steam turbines to accommodate variousheat/power ratios and loads. For limited expansion(pressure drop) and smaller loads (

160 ENERGY MANAGEMENT HANDBOOKFigure 7.5 Various multi-stage steam turbine systems. Courtesy of Coppus Engineering/Murray TurbomachineryCorporation.additional conventional units. This discussion assumesthat both heat and power demands remain constantall times. Hence, the design problem becomes one ofspecifying two variables: (1) how much power should becogenerated on-site and (2) how much power should beimported. Thus, given the technological, economical andlegal constraints for a particular plant, and assuming theCHP system must be constructed at a minimum overallcost, it becomes a constrained optimization problem.2. Initial Steam Conditions. Many industrial plants donot have adequate process steam demands to generateall the power required. Thus, it is important for thedesigner to examine those variables over which he hascontrol so he can optimize the amount of power that canbe economically generated. One set of these variablesare the initial steam conditions, i.e. the initial pressureand temperature of the steam generated. In general,an increase in initial pressure and/or temperature willincrease the amount of energy available for power generation.But the prime mover construction and cost, andthe heat demand impose economical limits for the initialsteam conditions. Thus, higher initial steam conditionscan be economically justified in industrial plants havingrelatively large process steam demands.3. Process Steam Pressure. For a given set of initial steamconditions, lowering the exhaust pressure also increasesthe energy available for power generation. However,this pressure is limited—totally, with non-condensing

COGENERATION AND DISTRIBUTED GENERATION 161turbines, or partially, with extraction turbines—by themaximum pressure required in the industrial process.For instance, in paper mills, various pressure levels areneeded to satisfy various process temperature requirements.4. Feedwater Heating. Feedwater heating through useof steam exhausted and/or extracted from a turbineincreases the power that can be generated.Calculation of Steam Turbine PowerGiven the initial steam conditions (psig, °F) andthe exhaust saturated pressure (psig), Theoretical SteamRates (TSR) specify the amount of steam heat input requiredto generate a kWh in an ideal turbine. The TSRis defined by3412 Btu/kWhTSR (lb/kWh) = ————————— (7.1)h i – h o Btu/lbwhere h i - h o is the difference in enthalpy from theinitial steam conditions to the exhaust pressure basedon an isentropic (ideal) expansion. These values can beobtained from steam tables or a Mollier chart*. However,they are conveniently tabulated by the American Societyof Mechanical Engineers.The TSR can be converted to the Actual Steam Rate(ASR)TSRASR (lb/kWh) = ——— (7.2)η tgwhere η tg is the turbine-generator overall efficiency, statedor specified at “design” or full-load conditions. Some ofthe factors that define the overall efficiency of a turbinegeneratorset are: the inlet volume flow, pressure ratio,speed, geometry of turbine staging, throttling losses, frictionlosses, generator losses and kinetic losses associatedwith the turbine exhaust. Most turbine manufacturersprovide charts specifying either ASR or η g values. Oncethe ASR has been established, the net enthalpy of thesteam supplied to process (NEP) can be calculated:NEP (Btu/lb)= Hi – 3500/ASR - Hc(x) – Hm(1-x) (7.3)where Hi = enthalpy at the turbine inlet conditions(Btu/lb)3500 = conversion from heat to power (Btu/kWh), including the effect of 2.6% radiation,mechanical and generator lossesHc = enthalpy of condensate return (Btu/lb)*See Appendix 1 at the end of the book, “Mollier Diagram for Steam.”Hm = enthalpy of make-up water (Btu/lb)x = condensate flow fraction in boiler feedwater(1–x) = make-up water flow fraction in boilerfeed waterHence, assuming a straight flow turbine (See Figure7.5), the net heat to process (Btu/hr) defined in Section7.2.1 can be obtained by multiplying equation 7.3 bythe flow rate in lb/hr. The analysis of the overall cyclewould require the replication of complete heat and massbalance calculations at part-load efficiencies. To expeditethese computations, there are a number of commerciallyavailable software packages, which also produce mass/heat balance tables. See Example 8 for a cogenerationsoftware application.Selection of Smaller Single-stage Steam TurbinesThere exist many applications for smaller units(condensing and noncondensing), especially in mechanicaldrives or auxiliaries (fans, pumps, etc.); but a typicalapplication is the replacement (or bypass) of a pressurereducing valve (PRV) by a single-stage back-pressureturbine.After obtaining the TSR from inlet and outlet steamconditions, Figure 7.6 helps in determining the approximatesteam rate (ASR) for smaller (

162 ENERGY MANAGEMENT HANDBOOKFigure 7.7 Power/speed ranges for single stage turbines.Courtesy Skinner Engine Co.have to be sized according to a commercially availableunit size, e.g. 300 kW.Figure 7.6 Estimation of steam rates for smaller (3000 kW), Figure 7.8 shows achart to determine the approximate turbine efficiencywhen the power range, speed and steam conditions areknown. Figure 7.9 gives steam rate correction factors foroff-design loads and speeds. However, manufacturers ofmultistage turbines advise that stage selection and otherthermodynamic parameters must be evaluated takinginto account other important factors such as speed

COGENERATION AND DISTRIBUTED GENERATION 163limitations, mechanical stresses, leakage and throttlinglosses, windage, bearing friction and reheat. Thus, aftera preliminary evaluation, it is important to comparenotes with the engineers of a turbine manufacturer.Example 2. Steam flow rate must be estimated to designa heat recovery steam generator (a bottoming cycle). Thesteam is to be used for power generation and is to beexpanded in a 5,000 RPM multistage condensing turbineto produce a maximum of 18,500 kW. Steam inlet conditionsis 600 psig/750°F and exhaust pressure is 4” HGA(absolute). Additional data are given below.Data (for a constant entropy steam expansion @ S = 1.61Btu/lb/°R)PGC : 8,500 kWInlet Steam: : 600 psig (615 psia), 750°FEnthalpy, h i : 1378.9 Btu/lb (from Mollierchart or steam tables)Outlet Steam : 4” HG absolute. (2 psia)Enthalpy, h o : 935.0 Btu/lb (from Mollierchart or ASME steam tables)Turbine Speed :5000 RPMa) Calculate TSR using equation 7.1 (TSR can also beobtained from ASME Tables or the Mollier chart):3412 Btu/kWhTSE (lb/kWh = ———————h i – h o Btu/lb3412 Btu/kWh= ——————————1378.9 – 935.0 Btu/lb= 7.68 lb/kWhb) Using the data above, Figure 7.8 givesη tg = 77%.c) Combining equations 7.2 and 7.4 andsolving for Ws, the total steam flow requiredisW s = PCG × TSR/η tg18,500 kW × 7.68 lb/kWh= ————————————0.77= 184,760 lb/hrFigure 7.8 Approximate steam turbine efficiency chart for multistagesteam turbines (>3,000 kW. Courtesy Elliot Co.Figure 7.9 Steam turbine part-load/speed correctioncurves. Courtesy Elliot Co.

164 ENERGY MANAGEMENT HANDBOOKd) Hence, the waste heat recovery steam generatorshould be able to generate about 185,000 lb/hr. Aunit with a 200,000 lb/hr nominal capacity willlikely be specified.Example 3. Find the ASR and steam flow required whenthe turbine of Example 2 is operated at 14,800 kW and4,500 RPM.% Power variation = 14,800 kW/185,500 kW = 80%% Speed variation = 4500 RPM/5000 RPM = 90%From Figure 7.9 the power correction factor is 1.04 andthe speed correction factor is 1.05. Then, the total correctionfactor is:1.04 × 1.05 = 1.09.Therefore, the part-load ASR is7.68 lb/kWh= ——————0.77× 1.09 = 10.9 lb/kWhThe steam flow required (@ 600 psig, 750°F) is= 10.9 lb/kWh × 14,500 kW= 158,050 lb/hr.7.2.2.2 Gas Turbine SystemsGas turbines are extensively applied in industrialplants. Two types of gas turbines are utilized: one is thelighter aircraft derivative turbine and the other is theheavier industrial gas turbine. Both industrial and aircraftengines have demonstrated excellent reliability/availabilityin the base load service. Due to the nature of theunit designs, aircraft derivative units usually have highermaintenance costs ($/kWh) than the industrial units.Since gas turbines can burn a variety of liquidand gas fuels and run long times unattended, they areconsidered to be versatile and reliable. For a fixed capacity,they have the smallest relative foot-print (sq-ft perkW).Gas Turbine Based CHP SystemsExhaust gases from gas turbines (from 600 to1200°F) offer a large heat recovery potential. The exhausthas been used directly, as in drying processes. Toppingcycles have also been developed by using the exhaustgases to generate process steam in heat recovery steamgenerators (HRSGs). Where larger power loads exist,high pressure steam is generated to be subsequentlyexpanded in a steam turbine-generator; this constitutesthe so called combined cycle (See Fig 7.4).If the demand for steam and/or power is evenhigher, the exhaust gases are used (1) as preheatedcombustion air of a combustion process or (2) are additionallyfired by a “duct burner” to increase their heatcontent and temperature.Recent developments include combined cycle systemswith steam injection or STIG—from the HRSG tothe gas turbine (Cheng Cycle)-to augment and modulatethe electrical output of the system. The Cheng Cycle allowsthe cogeneration system to handle a wider rangeof varying heat and power loads.All these options present a greater degree of CHPgeneration flexibility, allowing a gas turbine system tomatch a wider variety of heat-to-power demand ratiosand variable loads.Gas Turbine Ratings and PerformanceThere is a wide range of gas turbine sizes anddrives. Available turbines have ratings that vary in discretesizes from 50 kW to 160,000 kW. Kovacik (1982)and Hay (1988) list the following gas turbine data requiredfor design and off-design conditions:1. Unit Fuel Consumption-Output Characteristics.These depend on the unit design and manufacturer.The actual specific fuel consumption orefficiency and output also depend on (a) ambienttemperature, (b) pressure ratio and (c) part-loadoperation. Figure 7.11 shows performance data ofa gas turbine. Vendors usually provide this kind ofinformation.2. Exhaust Flow Temperature. This data item allowsthe development of the exhaust heat recoverysystem. The most common recovery system areHRSGs which are classified as unfired, supplementaryfired and fired units. The amount of steam thatcan be generated in an unfired or supplementaryfired HRSG can be estimated by the following relationship:W g C p (T i – T 3 ) e L fWs = —————————— (7.5)h sh – h satwhereW s = steam flow rateW g = exhaust flow rate to HRSGC p = specific heat of products of combustionT 1 = gas temperature-after burner, if applicable

COGENERATION AND DISTRIBUTED GENERATION 165T 3 = saturation temperature in steam drumL = a factor to account radiation and other losses,0.985h sh = enthalpy of steam leaving superheaterh sat = saturated liquid enthalpy in the steam drume = HRSG effectiveness = (T 1 –T 2 )/(T 1 –T 3 ), definedby Fig. 7.10.f = fuel factor, 1.0 for fuel oil, 1.015 for gas.3. Parametric Studies for Off-design Conditions. Varyingthe amount of primary or supplementary firing willchange the gas flow rate or temperature and the HRSGsteam output. Thus, according to the varying temperatures,several iterations of equation [7.5] are required toevaluate off-design or part load conditions. When thisevaluation is carried over a range of loads, firing rates,and temperatures, it is called a parametric study. Modelscan be constructed or off-design conditions using gasturbine performance data provided by manufacturers(See Figure 7.11).4. Exhaust Pressure Effects on Output and ExhaustTemperature. Heat recovery systems increase the exhaustbackpressure, reducing the turbine output inrelation to simple operation (without HRSG). Turbinemanufacturers provide test data about inlet and backpressureeffects, as well as elevation effects, on turbineoutput and efficiency (Fig 7.11).Example 4. In a combined cycle (Fig 7.4), the steam forthe turbine of Example 2 must be generated by severalHRSGs (See Fig 7.10). To follow variable CHP loads andto optimize overall system reliability, each HRSG will beconnected to a dedicated 10-MWe gas turbine-generator(GTG) set. The GTG sets burn fuel-oil and each unitexhausts 140,000 kg/hr of gas at 900°F. Estimate thetotal gas flow to the HRSGs, if the system must producea maximum of 160,000 lb/hr of 615 psia/750°F steam.How many HRSG-gas turbine sets are needed?DATA (See notation of Equation 7.5 and Figure 7.10)Gas Wg = 140,000 kg/hr per gas-turbine unitTurbines Cp = 0.26 Btu/lb/°F, average specific heatof gases between T 1 and T 4T 1 = 900°F, hot gas temperaturef = fuel factor, 1 for fuel oil.The inlet steam conditions required in the steam turbineof Example 3 are: (615 psia/750°F) and h i = 1378.9 Btu/lb (from Mollier chart). Thus,HRSGs Ws = 160,000 lb/hr from all HRSGsT3 = 488.8°F (temp of 615 psia sat steam)L = 0.98e = HRSG effectiveness, 0.9Therefore,h Sh = h i = 1378.9 Btu/lb and,h sat = 474.7 Btu/lb (Sat. Water @615 psia)From equation 7.5, the total combustion gas flow requiredisW g (h sh – h sat )W s = ——————————Cp (T 1 – T 3 ) e L f160,000 lb/hr (1381.1 – 474.7 Btu/lb= ——————————————————0.26 Btu/lb/°F (900–488.8°F) 0.9 × 0.98 × 1= 1,537,959.3 lb/hr or 427.2 lb/sec.Temp. °F†T 1T 4¨ApproachT 2††®PinchGasT 3†Steam®WaterEconomizerEvaporatorSuperheaterHeat transferred (UA) Btu/hr-°F®Figure 7.10 Heat recovery steam generator diagram.

166 ENERGY MANAGEMENT HANDBOOKFigure 7.11 Performance ratings and curves of a gas turbine-generator set. Nominal ratings are given at ISO conditions:59°F, sea level, 60% relative humidity and no external pressure losses. Courtesy of Solar Turbines, Inc.

COGENERATION AND DISTRIBUTED GENERATION 167Finally, the number of required gas turbine/HRSG sets is= [(1,537,959 lb/hr) (1kg/2.21b)/(140,000 kg/hr/set))= 4.99, i.e. 5 sets.7.2.2.3 Reciprocating Engine SystemsReciprocating engines include a variety of internallyfired, piston driven engines. Their sizes range from10 bhp to 50,000 bhp. According to Kovacik (1982), thelargest unit supplied by a U.S. manufacturer is rated at13,500 bhp. In larger plants, several units are used toaccommodate part load and to provide redundancy andbetter availability.In these engines, combustion heat rejected throughthe jacket water, lube oil and exhaust gases, can be recoveredthrough heat exchangers to generate hot waterand/or steam. Fig. 7.13 shows an internal combustionengine cogeneration system.Exhaust gases have also been used directly. Reciprocatingengines are classified by:and the jacket water heat are 22% and 33%, respectively.Thus, the flow rate of water heated from 70 to 180°Fis:= (1200 kW × 75%) [(90% × 22%) + (80% × 33%)](3412 Btu/kW) (1 lb.°F/Btu)(1 lb/8.33 gal)(1 hr/60 min)/(180 – 70°F)= 25.8 gal per minute.— the thermodynamic cycle: Diesel or Otto cycle.— the rotation speed: high-speed (1200-1800 rpm)medium-speed (500-900 rpm) low speed (450rpm or less)— the aspiration type: naturally aspirated or turbocharged— the operating cycle: two-cycle or four-cycle— the fuel burned: fuel-oil fired, natural-gasfired.Reciprocating engines are widely used to movevehicles, generators and a variety of shaft loads. Largerengines are associated to lower speeds, increased torque,and heavier duties. The total heat utilization of CHPsystems based on gas-fired or fuel-oil fired engines approach60-75%. Figure 7.12 shows the CHP balance vs.load of a diesel engine.Example 5. Estimate the amount of 180 F water thatcan be produced by recovering heat; first from thejacket water and then from the exhaust of a 1200 kWdiesel generator. In the average, the engine runs at a75% load and the inlet water temperature is 70°F. Theeffectiveness of the jacket water heat exchanger is 90%and exhaust heat exchanger is 80%.From Figure 7.12, at 75% load, the exhaust heatFig. 7.12 Diesel engine heat and power balance.7.2.2.4 Distributed GenerationDistributed Generation (DG) is emerging as thegeneration of heat and power (CHP) through relativelysmall distributed units (from 25 MW to a few kilowatts).Typically, these are smaller, self contained, power generationsystems located close to, adjacent to, or within theboundaries of a CHP user or consumer facility. Typicalfuels include natural gas, liquefied petroleum gas (LPG),kerosene and diesel fuel. DG plants may or may not beinterconnected to a utility grid. Earlier DG systems havealso been known as packaged cogeneration units.This section focuses on current developments ofthe following DG technologies:• Combustion or Gas Turbines(See also section 7.2.2.2)• Reciprocating Engines(See also section 7.2.2.3)• Fuel Cells(See also section 16.5)• Photovoltaics or solar cells(See also section 16.2.6)

168 ENERGY MANAGEMENT HANDBOOK• Microturbines(See also section 7.2.2.2)Previous sections in this chapter discuss some ofthe underlying technologies. Chapter 16, on “AlternativeEnergy,” provides a fundamental and renewable-energyapproach to Photovoltaics and Fuel Cells. Next, Table 7.2shows a comparison of current DG technologies.DG EconomicsIn addition to the economic factors listed in table7.2, i.e. turn key (installed), heat recovery (installed) andoperation and maintenance (O&M) costs, there are otherevaluation considerations which are often site specific.These include local fuel availability and cost, size andweight limitations, emission and noise regulations, andother factors.DG can be used for various purposes and applicationswith varying economic merit:1. As a prime mover to supply base-load electricaldemand with or without heat recovery.2. As a peak shaving generator3. As an uninterruptible power supply, emergency orback-up power unit.4. As a combination emergency power and peakshavinggenerator.Table 7.2 Comparison of DG TechnologiesIn general, DG as a prime mover for continuousoperation does not show feasible economics unlessthere is significant heat recovery. Thus, for 5000 hrs/yror more, with heat recovery (60-70% CHP system efficiency),our experience with several projects in theUS and the Caribbean shows that for the 200 to 800kW- range of gas- or diesel fired engine generators,there is an acceptable payback (3 years or less) onlyif the displaced electricity costs more than $0.10 perkWh and the natural gas fuel is $6.00 per million Btuor less. The evaluation considers a $0.015/ kWh CHPplant operation and maintenance fee, in addition tofuel consumption. See upper part of Table 7.3. Theshadowed cells on the upper-right part of the tableslist paybacks of three years or less.Without heat Recovery, DG makes little economicsense for continuous operation, unless an inexpensiveand plentiful source of fuel is available; e.g. bio-dieselfrom waste cooking oil or bio-gas from obtained fromagricultural waste, sewer treatment plant and landfills.DG is also feasible in the cases where surplus or pipelinenatural or liquefied petroleum gas exists. To have apayback of three years or less, the bottom part of Table7.3 shows the DG fuel must cost $6/million Btu orless and the displaced power must cost $0.11/kWh ormore. This sensitivity analysis assumes $0.010/kWh foroperation and maintenance, in addition to fuel cost.

COGENERATION AND DISTRIBUTED GENERATION 169Table 7.3. DG Economics and Sensitivity AnalysisPeak Shaving Optimization Model and Case StudyWhile DG applications for continuous powergeneration (with no heat recovery) are rarely economicallyfeasible, DG for peak shaving, in conjunction withback-up or emergency power generation, is generallyan attractive business proposition. See Appendix A atthe end of this chapter for a detailed case study whichshows there is generally a good business case for peakshaving using diesel or gas fired generators. Such casestudy illustrates an optimization model to estimate theoptimal peak shaving generator size (in kW or MW).Combustion or Gas Turbines*Combustion turbine (CT) sizes for distributedgeneration vary from 1 to 30 MW. CT’s are used topower aircraft, marine vessels, gas compressors, utilityand industrial generators. In 1998, over 500 CT’s wereshipped from the US to worldwide facilities totaling3,500 MW of power capacity. Most of these were soldoverseas. The North American market represents only11% share of the total. The primary application of CT’sis as prime mover for continuous power particularly, incombined cycle arrangement, or as a peaking unit to*The updates on DG technologies have been obtained from:Fundamentals of Distributed Generation” and “DistributedGeneration: A Primer” from the Gas Research Institute website: http://www.gri.org/pub/solutions/dg/index.html.generate during peak demand periods.Low maintenance, high reliability and high qualityexhaust heat make CT’s an excellent choice for industrialand commercial CHP applications larger than 3 MW.CT’s can burn natural gas, liquid fuels, such as dieseloil, or both gas and liquid (dual-fuel operation). Thus,they contribute to the fuel security of the DG plant. CTemissions can be controlled by using dry low NO x combustors,water or steam injection, or exhaust treatmentssuch as selective catalytic reduction (SCR). Due to theirinherent reliability and remote diagnostic capability, CTstend to have one of the lowest maintenance costs amongDG technologies.Reciprocating EnginesReciprocating internal combustion (IC) enginesare a widespread and well-known technology. NorthAmerican production tops 35 million units per year forautomobiles, trucks, construction and mining equipment,lawn care, marine propulsion, and of course, all types ofpower generation from small portable gen-sets to enginesthe size of a house, powering generators of several megawatts.Spark ignition engines for power generation usenatural gas as the preferred fuel—though they can be setup to run on propane or gasoline. Diesel cycle, compressionignition engines can operate on diesel fuel or heavyoil, or they can be set up in a dual-fuel configuration that

170 ENERGY MANAGEMENT HANDBOOKburns primarily natural gas with a small amount of dieselpilot fuel and can be switched to 100% diesel.Current generation IC engines offer low first cost,easy start-up, proven reliability when properly maintained,good load-following characteristics, and heatrecovery potential. IC engine systems with heat recoveryhave become a popular form of DG in Europe. Emissionsof IC engines have been reduced significantly inthe last several years by exhaust catalysts and throughbetter design and control of the combustion process. ICengines are well suited for standby, peaking, and intermediateapplications and for combined heat and power(CHP) in commercial and light industrial applications ofless than 10 MW.MicroturbinesMicroturbines or turbogenerators are small combustionturbines with outputs of 30-200 kW. Individual unitscan be packaged together to serve larger loads. Turbogeneratortechnology has evolved from automotive andtruck turbochargers auxiliary power units for airplanes,and small jet engines used for pilot military aircraft.Recent development of these microturbines hasbeen focused on this technology as the prime moverfor hybrid electric vehicles and as a stationary powersource for the DG market. In most configurations, theturbine shaft spinning at up to 100,000 rpm drives ahigh speed generator. This high frequency output is firstrectified and then converted to 60 Hz (or 50 Hz). Thesystems are capable of producing power at around 25 to30% efficiency by employing a recuperator that transfersheat energy from the exhaust stream back into the incomingair stream. Like larger turbines, these units arecapable of operating on a variety of fuels. The systemsare air-cooled and some even use air bearings, therebyeliminating both water and oil systems. Low-emissioncombustion systems are being demonstrated that provideemissions performance comparable to larger CTS.Turbogenerators are appropriately size for commercialbuildings or light industrial markets for cogeneration orpower-only applications.Fuel CellsFuel cells (Figure 7.13) produce power electrochemicallylike a battery rather than like a conventional generatingsystem that converts fuel to heat to shaft-power andfinally to electricity. Unlike a storage battery, however,which produces power from stored chemicals, fuel cellsproduce power when hydrogen fuel is delivered to thenegative pole (cathode) of the cell and oxygen in air isdelivered to the positive pole (anode). The hydrogenfuel can come from a variety of sources, but the mosteconomic is steam reforming of natural gas—a chemicalprocess that strips the hydrogen from both the fuel andthe steam. Several different liquid and solid media canbe used to created the fuel cell’s electrochemical reaction—phosphoric acid fuel cell (PAFC), molten carbonatefuel cell (MCFC), solid oxide fuel cell (SOFC), andproton exchange membrane (PEM). Each of these mediacomprises a distinct fuel cell technology with its ownperformance characteristics and development schedule.PAFCs are in early commercial market development nowwith 200 kW units delivered to over 120 customers.The SOFC and MCFC technologies are now in fieldtest or demonstration. PEM units are in early developmentand testing. Direct electrochemical reactions aregenerally more efficient than using fuel to drive a heatengine to produce electricity. Fuel cell efficiencies rangefrom 35-40% for the PAFC up to 60% with MCFC andSOFC systems under development. PEM unit efficienciesare as high as 50%. Fuel cells are inherently quietand extremely clean running. Like a battery, fuel cellsproduce direct current (DC) that must be run throughan inverter to get 60 Hz alternating current (AC). Thesepower electronics components can be integrated withother components as part of a power quality controlstrategy for sensitive customers. Because of current highcosts, fuel cells are best suited to environmentally sensitiveareas and customers with power quality concerns.Some fuel cell technology is modular and capable ofapplication in small commercial and even residentialmarkets; other technology utilizes high temperatures inlarger sized systems that would be well sited to industrialcogeneration applications.PhotovoltaicsPhotovoltaic power cells use solar energy to producepower. Photovoltaic power is modular and canbe sited wherever the sun shines. These systems havebeen commercially demonstrated in extremely sensitiveenvironmental areas and for remote (grid-isolated) applications.Battery banks are needed to store the energyharnessed during daytime. High costs make these systemsa niche technology that is able to compete more onthe basis of environmental benefits than on economics.Isolated facilities with need for limited but critical powerare typical applications.7.2.3 The Cogeneration Design ProcessThe following evaluation steps are suggested tocarry out cogeneration system design.1. Develop the profile of the various process steam(heat) demands at the appropriate steam pressures

COGENERATION AND DISTRIBUTED GENERATION 171Once this initial data bank has been established,the various alternatives that can satisfy plant heat andpower demands can be identified. Subsequently, detailedtechnical analyses are conducted. Thus, energybalances are made, investment cost estimated, and theeconomic merit of each alternative evaluated. Some approachesfor evaluation are discussed next.Figure 7.13 Proton Exchange Membrane (PEM ) fuelcells use platinum catalysts to promote the flow ofanions (positive ions) through the membrane, thuscreating a direct current (DC). An inverter can be usedto convert DC to AC.for the applications being studied. Also, collectdata with regard to condensate returned from theprocess and its temperature. Data must includedaily fluctuations due to normal variations in processneeds, as well as seasonal weather effects; includingthe influence of not-working periods suchas weekends, vacation periods, and holidays.2. A profile for electric power must be developed inthe same manner as the process heat demand profile.These profiles typically include hour-by-hourheat and power demands for “typical” days (orweeks) for each season or month of the year.3. Fuel availability and present-day cost as well asprojected future costs. The study should also factorprocess by-product fuels into the development ofthe energy supply system.4. Purchased power availability and its present andexpected future cost.5. Plant discharge stream data in the same degreeof detail as the process heat demand data.6. Number and rating of major (demand and generation)equipment items. This evaluation usuallyestablish whether spare capacity and/or supplementaryfiring should be installed.7. Plant, process and CHP system economic lives.7.2.4 Economic Feasibility Evaluation MethodsCogeneration feasibility evaluation is an iterativeprocess—further evaluations generally require moredata. There are a number of evaluation methods usingvarious approaches and different levels of technicaldetail. Most of them consider seasonal loads and equipmentperformance characteristics. Some of the mostrepresentative methods are discussed as follows.7.2.4.1 General Approaches ForDesign and EvaluationHay (1988) presents a structured approach forsystem design and evaluation. It is a sequence of evaluationiterations, each greater than the previous and eachproducing information whether the costs of the next stepis warranted. His suggested design process is based onthe following steps.Step 1: Site Walkthrough and Technical ScreeningStep 2: Preliminary Economic ScreeningStep 3: Detailed Engineering Design.Similarly, Butler (1984) considers three steps toperform studies, engineering and construction of cogenerationprojects. These are discussed as follow.Step 1. Preliminary studies and conceptual engineering.This is achieved by performing a technicalfeasibility and economic cost-benefit study to rank andrecommend alternatives. The determination of technicalfeasibility includes a realistic assessment with respect toenvironmental impact, regulatory compliance, and interfacewith a utility. Then, an economic analysis-basedon the simple payback period-serves as a basis for morerefined evaluations.Step 2. Engineering and Construction Planning.Once an alternative has been selected and approved bythe owner, preliminary engineering is started to developthe general design criteria. These include specific siteinformation such as process heat and power requirements,fuel availability and pricing, system type definition,modes of operation, system interface, review ofalternatives under more detailed load and equipment

172 ENERGY MANAGEMENT HANDBOOKdata, confirm selected alternative and finally size theplant equipment and systems to match the application.Step 3. Design Documentation. This includes thepreparation of project flow charts, piping and instrumentdiagrams, general arrangement drawings, equipmentlayouts, process interface layouts, building, structuraland foundation drawings, electrical diagrams, andspecifying an energy management system, if required.Several methodologies and manuals have been developedto carry out Step 1, i.e. screening analysis andpreliminary feasibility studies. Some of them are brieflydiscussed in the next sections. Steps 2 and 3 usually requiread-hoc approaches according to the characteristicsof each particular site. Therefore, a general methodologyis not applicable for such activities.7.2.4.2 Preliminary Feasibility Study ApproachesAGA Manual—GKCO Consultants (1982) developeda cogeneration feasibility (technical and economical)evaluation manual for the American Gas Association,AGA. It contains a “Cogeneration ConceptualDesign Guide” that provides guidelines for the developmentof plant designs. It specifies the following steps toconduct the site feasibility study:a) Select the type of prime mover or cycle (pistonengine, gas turbine or steam turbine);b) Determine the total installed capacity;c) Determine the size and number of prime movers;d) Determine the required standby capacity.According to its authors “the approach taken (inthe manual) is to develop the minimal amount of informationrequired for the feasibility analysis, deferringmore rigorous and comprehensive analyses to the actualconcept study.” The approach includes the discussionof the following “Design Options” or design criteria todetermine (1) the size and (2) the operation mode of theCHP system.Isolated Operation, Electric Load Following—Thefacility is independent of the electric utility grid, andis required to produce all power required on-site andto provide all required reserves for scheduled and unscheduledmaintenance.Baseloaded, Electrically Sized—The facility issized for baseloaded operation based on the minimumhistoric billing demand. Supplemental power is purchasedfrom the utility grid. This facility concept generallyresults in a shorter payback period than that fromthe isolated site.Baseloaded, Thermally Sized—The facility issized to provide most of the site’s required thermalenergy using recovered heat. The engines operated tofollow the thermal demand with supplemental boilerfired as required. The authors point out that: “this optionfrequently results in the production of more powerthan is required on-site and this power is sold to theelectric utility.”In addition, the AGA manual includes a descriptionof sources of information or processes by whichbackground data can be developed for the specific gasdistribution service area. Such information can be usedto adapt the feasibility screening procedures to a specificutility.7.2.4.3 Cogeneration System Selection and Sizing.The selection of a set of “candidate” cogenerationsystems entails to tentatively specify the most appropriateprime mover technology, which will be furtherevaluated in the course of the study. Often, two or morealternative systems that meet the technical requirementsare pre-selected for further evaluation. For instance, aplant’s CHP requirements can be met by either, a reciprocatingengine system or combustion turbine system.Thus, the two system technologies are pre-selected fora more detailed economic analysis.To evaluate specific technologies, there exist a vastnumber of technology-specific manuals and references.A representative sample is listed as follows. Mackay(1983) has developed a manual titled “Gas TurbineCogeneration: Design, Evaluation and Installation.” Kovacik(1984) reviews application considerations for bothsteam turbine and gas turbine cogeneration systems.Limaye (1987) has compiled several case studies on industrialcogeneration applications. Hay (1988) discussestechnical and economic considerations for cogenerationapplication of gas engines, gas turbines, steam enginesand packaged systems. Keklhofer (1991) has written atreatise on technical and economic analysis of combinedcyclegas and steam turbine power plants. Ganapathy(1991) has produced a manual on waste heat boilers.Usually, system selection is assumed to be separatefrom sizing the cogeneration equipment (kWe). However,since performance, reliability and cost are verydependent on equipment size and number, technologyselection and system size are very intertwined evaluationfactors. In addition to the system design criteriagiven by the AGA manual, several approaches for co-

COGENERATION AND DISTRIBUTED GENERATION 173generation system selection and/or sizing are discussedas follows.Heat-to-Power RatioCanton et al (1987) of The Combustion and FuelsResearch Group at Texas A&M University has developeda methodology to select a cogeneration system fora given industrial application using the heat to power ratio(HPR). The methodology includes a series of graphsused 1) to define the load HPR and 2) to compare andmatch the load HPR to the HPRs of existing equipment.Consideration is then given to either, heat or power loadmatching and modulation.Sizing ProceduresHay (1987) considers the use of the load durationcurve to model variable thermal and electrical loads insystem sizing, along with four different scenarios describedin Figure 7.14. Each one of these scenarios definesan operating alternative associated to a system size.Oven (1991) discusses the use of the load durationcurve to model variable thermal and electrical loads insystem sizing in conjunction with required thermal andelectrical load factors. Given the thermal load durationand electrical load duration curves for a particularfacility, different sizing alternatives can be defined forvarious load factors.Eastey et al. (1984) discusses a model (CO-GENOPT) for sizing cogeneration systems. The basicinputs to the model are a set of thermal and electricprofiles, the cost of fuels and electricity, equipment costand performance for a particular technology. The modelcalculates the operating costs and the number of unitsfor different system sizes. Then it estimates the net presentvalue for each one of them. Based on the maximumnet present value, the “optimum” system is selected. Themodel includes cost and load escalation.Wong, Ganesh and Turner (1991) have developedtwo statistical computer models to optimize cogenerationsystem size subject to varying capacities/loads andFigure 7-14. Each operation mode defines a sizing alternative. Source: Hay (1987).

174 ENERGY MANAGEMENT HANDBOOKto meet an availability requirement. One model is forinternal combustion engines and the other for unfired gasturbine cogeneration systems. Once the user defines a requiredavailability, the models determine the system sizeor capacity that meets the required availability and maximizesthe expected annual worth of its life cycle cost.7.3 COMPUTER PROGRAMSThere are several computer programs-mainly PCbased-available for detailed evaluation of cogenerationsystems. In opposition to the rather simple methodsdiscussed above, CHP programs are intended for systemconfiguration or detailed design and analysis. For thesereasons, they require a vast amount of input data. Below,we examine two of the most well known programs.7.3.1 CELCAPLee (1988) reports that the Naval Civil EngineeringLaboratory developed a cogeneration analysis computerprogram known as Civil Engineering LaboratoryCogeneration Program (CELCAP), “for the purpose ofevaluating the performance of cogeneration systems ona lifecycle operating cost basis.” He states that “selectionof a cogeneration energy system for a specific applicationis a complex task.” He points out that the firststep in the selection of cogeneration system is to makea list of potential candidates. These candidates shouldinclude single or multiple combinations of the varioustypes of engine available. The computer program doesnot specify CHP systems; these must be selected by thedesigner. Thus, depending on the training and previousexperience of the designer, different designers may selectdifferent systems of different sizes. After selecting ashort-list of candidates, modes of operations are definedfor the candidates. So, if there are N candidates andM modes of operation, then NxM alternatives must beevaluated. Lee considers three modes of operation:1) Prime movers operating at their full-rated capacity,any excess electricity is sold to the utility and anyexcess heat is rejected to the environment. Anyelectricity shortage is made up with imports. Processsteam shortages are made-up by an auxiliaryboiler.2) Prime movers are specified to always meet theentire electrical load of the user. Steam or heatdemand is met by the prime mover. An auxiliaryboiler is fired to meet any excess heat deficit andexcess heat is rejected to the environment.3) Prime movers are operated to just meet the steamor heat load. In this mode, power deficits are madeup by purchased electricity. Similarly, any excesspower is sold back to the utility.For load analysis, Lee considers that “demand of theuser is continuously changing. This requires that data onthe electrical and thermal demands of the user be availablefor at least one year.” He further states that “electricaland heat demands of a user vary during the year becauseof the changing working and weather conditions.”However, for evaluation purposes, he assumes that theworking conditions of the user-production related CHPload-remain constant and “that the energy-demand patterndoes not change significantly from year to year.”Thus, to consider working condition variations, Lee classifiesthe days of the year as working and non-workingdays. Then, he uses “average” monthly load profiles and“typical” 24-hour load profiles for each class.“Average” load profiles are based on electric andsteam consumption for an average weather condition atthe site. A load profile is developed for each month, thusmonthly weather and consumption data is required. Abest fit of consumption (Btu/month or kWh/month)versus heating and cooling degree days is thus obtained.Then, actual hourly load profiles for working and nonworkingdays for each month of the year are developed.The “best representative” profile is then chosen for the“typical working day” of the month. A similar procedureis done for the non-working days.Next an energy balance or reconciliation is performedto make sure the consumption of the hourlyload profiles agrees with the monthly energy usage. Amultiplying factor K is defined to adjust load profilesthat do not balance.K j = E mj /(AE wj + AE nwj ) (7.9)whereK j = multiplying factor for month jE mj = average consumption (kWh) by the user forthe month j selected from the monthly electricityusage versus degree day plotAE wj = typical working-day electric usage (kWh),i.e. the area under the typical working dayelectric demand profile for the month jAE nwj = typical non-working day usage (kWh), i.e.the area under the typical non-working dayelectric demand profile for the month j.Lee suggests that each hourly load in the loadprofiles be multiplied by the K factor to obtain the “cor-

COGENERATION AND DISTRIBUTED GENERATION 175rect working and non-working day load profiles for themonth.” The procedure is repeated for all months of theyear for both electric and steam demands. Lee states that“the resulting load profiles represent the load demandfor average weather conditions.”Once a number of candidate CHP systems hasbeen selected, equipment performance data and the loadprofiles are fed into CELCAP to produce the requiredoutput. The output can be obtained in a brief or detailedform. In brief form, the output consists of a summary ofinput data and a life cycle cost analysis including fuel,operation and maintenance and purchased power costs.The detailed printout includes all the information of thebrief printout, plus hourly performance data for 2 daysin each month of the year. It also includes the maximumhourly CHP output and fuel consumption. The hourlyelectric demand and supply are plotted, along with thehourly steam demand and supply for each month of theyear.Despite the simplifying assumptions introduced byLee to generate average monthly and typical daily loadprofiles, it is evident that still a large amount of datahandling and preparation is required before CELCAPis run. By recognizing the fact that CHP loads varyover time, he implicitly justifies the amount of effort inrepresenting the input data through hourly profiles fortypical working and non-working days of the month.If a change occurs in the products, process orequipment that constitute the energy consumers withinthe industrial plant, a new set of load profiles must begenerated. Thus, exploring different conditions requiressensitivity analyses or parametric studies for off-designconditions.A problem that becomes evident at this pointis that, to accurately represent varying loads, a largenumber of load data points must be estimated for subsequentuse in the computer program. Conversely, thepreliminary feasibility evaluation methods discussedpreviously, require very few and only “average” loaddata. However, criticism of preliminary methods hasarisen for not being able to truly reflect seasonal variationsin load analysis (and economic analysis) and forlacking the flexibility to represent varying CHP systemperformance at varying loads.7.3.2 COGENMASTERLimaye and Balakrishnan (1989) of Synergic ResourcesCorporation have developed COGENMASTER.It is a computer program to model the technical aspectsof alternative cogeneration systems and options, evaluateeconomic feasibility, and prepare detailed cash flowstatements.COGENMASTER compares the CHP alternativesto a base case system where electricity is purchased fromthe utility and thermal energy is generated at the site.They extend the concept of an option by referring notonly to different technologies and operating strategiesbut also to different ownership structures and financingarrangements. The program has two main sections:a Technology and a Financial Section. The technologySection includes 5 modules:• Technology Database Module• Rates Module• Load Module• Sizing Module• Operating ModuleThe Financial Section includes 3 modules:• Financing Module• Cash Flow Module• Pricing ModuleIn COGENMASTER, facility electric and thermalloads may be entered in one of three ways, dependingon the available data and the detail required for projectevaluation:— A constant average load for every hour of the year.— Hourly data for three typical days of the year— Hourly data for three typical days of each monthThermal loads may be in the form of hot water orsteam; but system outlet conditions must be specifiedby the user. The sizing and operating modules permita variety of alternatives and combinations to be considered.The system may be sized for the base or peak,summer or winter, and electric or thermal load. There isalso an option for the user to define the size the systemin kilowatts. Once the system size is defined, severaloperation modes may be selected. The system may beoperated in the electric following, thermal following orconstantly running modes of operation. Thus, N sizingoptions and M operations modes define a total of NxMcogeneration alternatives, from which the “best” alternativemust be selected. The economic analysis is based onsimple payback estimates for the CHP candidates versusa base case or do-nothing scenario. Next, dependingon the financing options available, different cash flowsmay be defined and further economic analysis-based

176 ENERGY MANAGEMENT HANDBOOKon the Net Present Value of the alternatives—may beperformed.7.4 U.S. COGENERATION LEGISLATION: PURPAIn 1978 the U.S. Congress amended the Federal PowerAct by promulgation of the Public Utilities RegulatoryAct (PURPA). The Act recognized the energy savingpotential of industrial cogeneration and small powerplants, the need for real and significant incentives fordevelopment of these facilities and the private sectorrequirement to remain unregulated.PURPA of 1978 eliminated several obstacles tocogeneration so cogenerators can count on “fair” treatmentby the local electric utility with regard to interconnection,back-up power supplies, and the sale of excesspower. PURPA contains the major federal initiativesregarding cogeneration and small power production.These initiatives are stated as rules and regulationspertaining to PURPA Sections 210 and 201; which wereissued in final form in February and March of 1980,respectively. These rules and regulations are discussedin the following sections.Initially, several utilities—especially those withexcess capacity-were reticent to buy cogenerated powerand have, in the past, contested PURPA. Power (1980)magazine reported several cases in which oppositionpersisted in some utilities to private cogeneration. Butafter the Supreme Court ruling in favor of PURPA, moreand more utilities are finding that PURPA can work totheir advantage. Polsky and Landry (1987) report thatsome utilities are changing attitudes and are even investingin cogeneration projects.7.4.1 PURPA 201*Section 201 of PURPA requires the Federal EnergyRegulatory Commission (FERC) to define the criteriaand procedures by which small power producers (SPPs)and cogeneration facilities can obtain qualifying statusto receive the rate benefits and exemptions set forth inSection 210 of PURPA. Some PURPA 201 definitions arestated below.Small Power Production FacilityA “Small Power Production Facility” is a facilitythat uses biomass, waste, or renewable resources, includingwind, solar and water, to produce electric power and*Most of the following sections have been adapted from CFR18 (1990)and Harkins (1980), unless quoted otherwise.is not greater than 80 megawatts.Facilities less than 30 MW are exempt from thePublic Utility Holding Co. Act and certain state lawand regulation. Plants of 30 to 80 MW which use biomass,may be exempted from the above but may notbe exempted from certain sections of the Federal PowerAct.Cogeneration FacilityA “Cogeneration Facility” is a facility which produceselectric energy and forms of useful thermal energy(such as heat or steam) used for industrial, commercial,heating or cooling purposes, through the sequential useof energy. A Qualifying Facility (QF) must meet certainminimum efficiency standards as described later. Cogenerationfacilities are generally classified as “topping”cycle or “bottoming” cycle facilities.7.4.2 Qualification of a “Cogeneration Facility” or a“Small Power Production Facility” under PURPACogeneration FacilitiesTo distinguish new cogeneration facilities whichwill achieve meaningful energy conservation fromthose which would be “token” facilities producingtrivial amounts of either useful heat or power, the FERCrules establish operating and efficiency standards forboth topping-cycle and bottom-cycle NEW cogenerationfacilities. No efficiency standards are required forEXISTING cogeneration facilities regardless of energysource or type of facility. The following fuel utilizationeffectiveness (FUE) values—based on the lower heatingvalue (LHV) of the fuel—are required from QFs.• For a new topping-cycle facility:— No less than 5% of the total annual energyoutput of the facility must be useful thermalenergy.• For any new topping-cycle facility that uses anynatural gas or oil:— All the useful electric power and half the usefulthermal energy must equal at least 42.5%of the total annual natural gas and oil energyinput; and— If the useful thermal output of a facility is lessthan 15% of the total energy output of the facility,the useful power output plus one-half theuseful thermal energy output must be no lessthan 45% of the total energy input of naturalgas and oil for the calendar.

COGENERATION AND DISTRIBUTED GENERATION 177For a new bottoming-cycle facility:• If supplementary firing (heating of water or steambefore entering the electricity generation cyclefrom the thermal energy cycle) is done with oilor gas, the useful power output of the bottomingcycle must, during any calendar year, be no lessthan 45% of the energy input of natural gas andoil for supplementary firing.Small Power Production FacilitiesTo qualify as a small power production facilityunder PURPA, the facility must have production capacityof under 80 MW and must get more than 50% of itstotal energy input from biomass, waste, or renewableresources. Also, use of oil, coal, or natural gas by thefacility may not exceed 25% of total annual energy inputto the facility.Ownership Rules Applying toCogeneration and Small Power ProducersA qualifying facility may not have more than 50%of the equal interest in the facility held by an electricutility.7.4.3 PURPA 210Section 210 of PURPA directs the Federal EnergyRegulatory Commission (FERC) to establish the rulesand regulations requiring electric utilities to purchaseelectric power from and sell electric power to qualifyingcogeneration and small power production facilities andprovide for the exemption to qualifying facilities (QF)from certain federal and state regulations.Thus, FERC issued in 1980 a series of rules to relaxobstacles to cogeneration. Such rules implement sectionsof the 1978 PURPA and include detailed instructions tostate utility commissions that all utilities must purchaseelectricity from cogenerators and small power producersat the utilities’ “avoided” cost. In a nutshell, this meansthat rates paid by utilities for such electricity must reflectthe cost savings they realize by being able to avoidcapacity additions and fuel usage of their own.Tuttle (1980) states that prior to PURPA 210, cogenerationfacilities wishing to sell their power were facedwith three major obstacles:• Utilities had no obligation to purchase power, andcontended that cogeneration facilities were toosmall and unreliable. As a result, even those cogeneratorsable to sell power had difficulty gettingan equitable price.• Utility rates for backup power were high and oftendiscriminatory• Cogenerators often were subject to the same strictstate and federal regulations as the utility.PURPA was designed to remove these obstacles,by requiring utilities to develop an equitable programof integrating cogenerated power into their loads.Avoided CostsThe costs avoided by a utility when a cogenerationplant displaces generation capacity and/or fuel usageare the basis to set the rates paid by utilities for cogeneratedpower sold back to the utility grid. In somecircumstances, the actual rates may be higher or lowerthan the avoided costs, depending on the need of theutility for additional power and on the outcomes of thenegotiations between the parties involved in the cogenerationdevelopment process.All utilities are now required by PURPA to providedata regarding present and future electricity costs on acent-per-kWh basis during daily, seasonal, peak and offpeakperiods for the next five years. This informationmust also include estimates on planned utility capacityadditions and retirements, and cost of new capacity andenergy costs.Tuttle (1980) points out that utilities may agree topay greater price for power if a cogeneration facilitycan:• Furnish information on demonstrated reliabilityand term of commitment.• Allow the utility to regulate the power productionfor better control of its load and demandchanges.• Schedule maintenance outages for low-demandperiods.• Provide energy during utility-system daily andseasonal peaks and emergencies.• Reduce in-house on-site load usage during emergencies.• Avoid line losses the utility otherwise would haveincurred.In conclusion, a utility is willing to pay better“buyback” rates for cogenerated power if it is short incapacity, if it can exercise a level of control on the CHPplant and load, and if the cogenerator can provide and/or demonstrate a “high” system availability.

178 ENERGY MANAGEMENT HANDBOOKPURPA further states that the utility is not obligatedto purchase electricity from a QF during periods thatwould result in net increases in its operating costs. Thus,low demand periods must be identified by the utilityand the cogenerator must be notified in advance. Duringemergencies (utility outages), the QF is not requiredto provide more power than its contract requires, but autility has the right to discontinue power purchases ifthey contribute to the outage.7.4.4 Other RegulationsSeveral U.S. regulations are related to cogeneration.For example, among environmental regulations,the Clean Air Act may control emissions from a wasteto-energypower plant. Another example is the regulationof underground storage tanks by the ResourceConservation and Recovery Act (RCRA). This applies toall those cogenerators that store liquid fuels in undergroundtanks. Thus, to maximize benefits and to avoidcostly penalties, cogeneration planners and developersshould become savvy in related environmental matters.There are many other issues that affect the developmentand operation of a cogeneration project.For further study, the reader is referred to a variety ofsources such proceedings from the various World EnergyEngineering Congresses organized by the Associationof Energy Engineers (Atlanta, GA). Other sourcesinclude a general compendium of cogeneration planningconsiderations given by Orlando (1990), and a manualdevelopedby Spiewak (1994)—which emphasizes theregulatory, contracting and financing issues of cogeneration.7.5 EVALUATING COGENERATIONOPPORTUNITIES: CASE EXAMPLESThe feasibility evaluation of cogeneration opportunitiesfor both, new construction and facility retrofit, requirethe comparison and ranking of various options using afigure of economic merit. The options are usually combinationsof different CHP technologies, operating modesand equipment sizes.A first step in the evaluation is the determinationof the costs of a base-case (or do-nothing) scenario.For new facilities, buying thermal and electrical energyfrom utility companies is traditionally considered thebase case. For retrofits, the present way to buy and/orgenerate energy is the base case. For many, the base-casescenario is the “actual plant situation” after “basic” energyconservation and management measures have beenimplemented. That is, cogeneration should be evaluatedupon an “efficient” base case plant.Next, suitable cogeneration alternatives are generatedusing the methods discussed in sections 7.2 and7.3. Then, the comparison and ranking of the base caseversus the alternative cases is performed using an economicanalysis.Henceforth, this section addresses a basic approachfor the economic analysis of cogeneration. Specifically,it discusses the development of the cash flows for eachoption including the base case. It also discusses somefigures of merit such as the gross pay out period (simplepayback) and the discounted or internal rate of return.Finally, it describes two case examples of evaluations inindustrial plants. The examples are included for illustrativepurposes and do not necessarily reflect the latestavailable performance levels or capital costs.7.5.1 General ConsiderationsA detailed treatise on engineering economy is presentedin Chapter 4. Even so, since economic evaluationsplay the key role in determining whether cogenerationcan be justified, a brief discussion of economic considerationsand several evaluation techniques follows.The economic evaluations are based on examiningthe incremental increase in the investment cost for thealternative being considered relative to the alternativeto which it is being compared and determining whetherthe savings in annual operating cost justify the increasedinvestment. The parameter used to evaluate the economicmerit may be a relatively simple parameter suchas the “gross payout period.” Or one might use moresophisticated techniques which include the time value ofmoney, such as the “discounted rate of return,” on thediscretionary investment for the cogeneration systemsbeing evaluated.Investment cost and operating cost are the expenditurecategories involved in an economic evaluation.Operating costs result from the operations of equipment,such as (1) purchased fuel, (2) purchased power, (3) purchasedwater, (4) operating labor, (5) chemicals, and (6)maintenance. Investment-associated costs are of primaryimportance when factoring the impact of federal andstate income taxes into the economic evaluation. Thesecosts (or credits) include (1) investment tax credits, (2)depreciation, (3) local property taxes, and (4) insurance.The economic evaluation establishes whether the operatingand investment cost factors result in sufficientafter-tax income to provide the company stockholdersan adequate rate of return after the debt obligations withregard to the investment have been satisfied.When one has many alternatives to evaluate, the

COGENERATION AND DISTRIBUTED GENERATION 179less sophisticated techniques, such as “gross payout,”can provide an easy method for quickly ranking alternativesand eliminating alternatives that may beparticularly unattractive. However, these techniques areapplicable only if annual operating costs do not changesignificantly with time and additional investments donot have to be made during the study period.The techniques that include the time value ofmoney permit evaluations where annual savings canchange significantly each year. Also, these evaluationprocedures permit additional investments at any timeduring the study period. Thus these techniques trulyreflect the profitability of a cogeneration investment orinvestments.Fig. 7.15 Discounted rate of return versus gross payoutperiod. Basis: (1) depreciation period, 28 years; (2) sumof-the-years’-digitsdepreciation; (3) economic life, 28years; (4) constant annual savings with time; (5) localproperty taxes and insurance, 4% of investment cost;(6) state and federal income taxes, 53%; (7) investmenttax credit, 10% of investment cost.7.5.2 Cogeneration Evaluation Case ExamplesThe following examples illustrate evaluation proceduresused for cogeneration studies. Both examples arebased on 1980 investment costs for facilities located inthe U.S. Gulf Coast area.For simplicity, the economic merit of each alternativeexamined is expressed as the “gross payout period”(GPO). The GPO is equal to the incremental investmentfor cogeneration divided by the resulting first-year annualoperating cost savings. The GPO can be convertedto a “discounted rate of return” (DRR) using Figure 7.15.However, this curve is valid only for evaluations involvinga single investment with fixed annual operating costsavings with time. In most instances, the annual savingsdue to cogeneration will increase as fuel costs increaseto both utilities and industries in the years ahead. Theseincreased future savings enhance the economics of cogeneration.For example, if we assume that a project has aGPO of three years based on the first-year operating costsavings, Figure 7.15 shows a DRR of 18.7%. However, ifthe savings due to cogeneration increase 10% annuallyfor the first three operating years of the project and areconstant thereafter, the DRR increases to 21.6%; if the savingsincrease 10% annually for the first six years, the DRRwould be 24.5%; and if the 10% increase was experiencedfor the first 10 years, the DRR would be 26.6%.Example 6: The energy requirements for a large industrialplant are given in Table 7.3. The alternativesconsidered include:Base case. Three half-size coal-fired process boilers areinstalled to supply steam to the plant’s 250-psig steamheader. All 80-psig steam and steam to the 20-psig deaeratingheater is pressure-reduced from the 250-psig steamheader. The powerhouse auxiliary power requirementsare 3.2 MW. Thus the utility tie must provide 33.2 MWto satisfy the average plant electric power needs.Case 1. This alternative is based on installation of anoncondensing steam turbine generator. The unit initialTable 7.3 Plant Energy Supply System Considerations: Example 6———————————————————————————————————————————————————Process steam demandsNet heat to process at 250 psig. 410°F—317 million Btu/hr avg.Net heat to process at 80 psig, 330°F—208 million Btu/hr avg. (peak requirements are 10% greater thanaverage values)Process condensate returns: 50% of steam delivered at 280°FMakeup water at 80°FPlant fuel is 3.5% sulfur coalCoal and limestone for SO 2 scrubbing are available at a total cost of $2/million Btu firedProcess area power requirement is 30 MW avg.Purchased power cost is 3.5 cents/kWh———————————————————————————————————————————————————

180 ENERGY MANAGEMENT HANDBOOKsteam conditions are 1450 psig, 950°F with automaticextraction at 250 psig and 80 psig exhaust pressure.The boiler plant has three half-size units providing thesame reliability of steam supply as the Base Case. Thefeedwater heating system has closed feedwater heatersat 250 psig and 80 psig with a 20 psig deaeratingheater. The 20-psig steam is supplied by noncondensingmechanical drive turbines used as powerhouse auxiliarydrives. These units are supplied throttle steam from the250-psig steam header. For this alternative, the utility tienormally provides 4.95 MW. The simplified schematicand energy balance is given in Figure 7.16.The results of this cogeneration example are tabulatedin Table 7.4. Included are the annual energy requirements,the 1980 investment costs for each case, andthe annual operating cost summary. The investment costdata presented are for fully operational plants, includingoffices, stockrooms, machine shop facilities, lockerrooms, as well as fire protection and plant security. Thecost of land is not included.The incremental investment cost for Case 1 givenin Table 7.4 is $17.2 million. Thus the incremental cost is$609/kW for the 28.25-MW cogeneration system. This illustratesthe favorable per unit cost for cogeneration systemscompared to coal-fired facilities designed to providekilowatts only, which cost in excess of $1000/kW.The impact of fuel and purchased power costsother than Table 7.3 values on the GPO for this exampleis shown in Figure 7.17. Equivalent DRR values basedFig. 7.16 Simplified schematic and energy-balancediagram: Example 6, Case 1. All numbers are flows in10 3 lb/hr; Plant requirements given in Table 7.8, grossgeneration, 30.23 MW; powerhouse auxiliaries, 5.18MW; net generation, 25.05 MW.on first-year annual operating cost savings can be estimatedusing Figure 7.15.Sensitivity analyses often evaluate the impactof uncertainties in the installed cost estimates on theprofitability of a project. If the incremental investmentcost for cogeneration is 10% greater than the Table 7.4estimate, the GPO would increase from 3.2 to 3.5 years.Thus the DRR would decrease from 17.5% to about 16%,as shown in Figure 7.15.Table 7.4 Energy and Economic Summary: Example 6———————————————————————————————————————————————————Alternative Base Case Case 1———————————————————————————————————————————————————Energy summaryBoiler fuel (10 6 Btu/hr HHV) 599 714Purchased power (MW) 33.20 4.95Estimated total installed cost (10 6 $) 57.6 74.8Annual operating costs (10 6 $)Fuel and limestone at $2/10 6 Btu 10.1 12.0Purchased power at 3.5 cents/kWh 9.8 1.5Operating labor 0.8 1.1Maintenance 1.4 1.9Makeup water 0.3 0.5Total 22.4 17.0Annual savings (10 6 $) Base 5.4Gross payout period (yrs) Base 3.2———————————————————————————————————————————————————Basis: (1) boiler efficiency is 87%; (2) operation equivalent to 8400 hr/yr at Table 7-3 conditions; (3) maintenanceis 2.5% of the estimated total installed cost; (4) makeup water cost for case 1 is 80 cents/1000 gal greater than BaseCase water costs; (5) stack gas scrubbing based on limestone system.———————————————————————————————————————————————————

COGENERATION AND DISTRIBUTED GENERATION 181Example 7: The energy requirements for a chemicalplant are presented in Table 7.5. The alternatives consideredinclude:Base case. Three half-size oil-fired packaged process boilersare installed to supply process steam at 150 psig. Eachunit is fuel-oil-fired and includes a particulate removalsystem. The plant has a 60-day fuel-oil-storage capacity.A utility tie provides 30.33 MW average to supply processand boiler plant auxiliary power requirements.Case 1. (Refer to Figure 7.18). This alternative examinesthe merit of adding a noncondensing steam turbinegenerator with 850 psig, 825°F initial steam conditions,150-psig exhaust pressure. Steam is supplied bythree half-size packaged boilers. The feedwater heatingsystem is comprised of a 150-psig closed heater and a20-psig deaerating heater. The steam for the deaeratingheater is the exhaust of a mechanical drive turbine(MDT). The MDT is supplied 150-psig steam and drivesFig. 7.17Effect of differentfueland powercosts oncogenerationprofitability:Example 1.Basis: Conditionsgivenin Tables 7.3and 7.4.some of the plant boiler feed pumps. The net generationof this cogeneration system is 6.32 MW when operatingat the average 150-psig process heat demand. A utilitytie provides the balance of the power required.Case 2. (Refer to Figure 7.19). This alternative is a combinedcycle using the 25,000-kW gas turbine generatorwhose performance is given in Table 7.7. An unfiredHRSG system provides steam at both 850 psig, 825°Fand 150 psig sat. Plant steam requirements in excess ofthat available from the two-pressure level unfired HRSGsystem are generated in an oil fired packaged boiler.The steam supplied to the noncondensing turbine isexpanded to the 150-psig steam header. The net generationfrom the overall system is 26.54 MW. A utility tieprovides power requirements in excess of that suppliedby the cogeneration system. The plant-installed cost estimatesfor Case 2 include two half-size package boilers.Thus full steam output can be realized with any steamgenerator out of service for maintenance.The energy summary, annual operating costs, andeconomic results are presented in Table 7.6. The resultsshow that the combined cycle provides a GPO of 2.5years based on the study fuel and purchased powercosts. The incremental cost for Case 2 relative to theBase Case is $395/kW compared to $655/kW for Case1 relative to the Base Case. This favorable incrementalinvestment cost combined with a FCP of 5510 Btu/kWhcontribute to the low CPO.The influence of fuel and power costs other thanthose given in Table 7.5 on the GPO for cases 1 and 2 isTable 7.5 Plant Energy Supply System Considerations:Example 7—————————————————————————Process steam demandsNet heat to process at 150 psig sat—158.5 millionBtu/hr avg. (peak steam requirements are 10%greater than average values)Process condensate returns: 45% of the steam deliveredat 300°FMakeup water at 80°FPlant fuel is fuel oilFuel cost is $5/million BtuProcess areas require 30 MWPurchased power cost is 5 cents/kWh—————————————————————————Fig. 7.18 Simplified schematic and energy-balancediagram: Example 7, Case 1. All numbers are flowsin 1000 lb/hr; gross generation, 6.82 MW; powerhouseauxiliaries, 0.50 MW; net generation; 6.32 MW.

182 ENERGY MANAGEMENT HANDBOOKFig. 7.19 Simplified schematic andenergy-balance diagram: Example7, Case 2. All numbers are flows in1000 lb/hr; gross generation, 26.77MW, powerhouse auxiliaries, 0.23MW: net generation, 26.54 MW.Table 7.6 Energy and Economic Summary: Example 7———————————————————————————————————————————————————Alternative Base Case Case 1 Case 2———————————————————————————————————————————————————Energy summaryFuel (10 6 Btu/hr HHV)Boiler 183 209 34Gas turbine — 297Total 183 209 331Purchased power (MW) 30.33 23.77 3.48Estimated total installed cost (10 6 $) 8.3 12.6 18.9Annual operating cost (10 6 $)Fuel at $5/M Btu HHV 7.7 8.8 13.9Purchased power at 5 cents/kWh 12.7 10.0 1.5Operating labor 0.6 0.9 0.9Maintenance 0.2 0.3 0.5Makeup water 0.1 0.2 0.2Total 21.3 20.2 17.0Annual savings (10 6 $) Base 1.1 4.3Gross payout period (yr) Base 3.9 2.5———————————————————————————————————————————————————Basis: (1) gas turbine performance per Table 7-7; (2) boiler efficiency, 87%; (3) operation equivalent to 8400 hr/yrat Table 7-5 conditions; (4) maintenance, 2.5% of the estimated total installed costs; (5) incremental makeup watercost for cases 1 and 2 relative to the Base Case. $1 /1000 gal.shown in Figure 7.20. These GPO values can be translatedto DRRs using Figure 7.15.Example 8. A gas-turbine and HRSG cogeneration systemis being considered for a brewery to supply baseloadelectrical power and part of the steam needed forprocess. An overview of the proposed system is shownin Figure 7.21. This example shows the use of computertools in cogeneration design and evaluation.Base Case.: Currently, the plant purchases about3,500,000 kWh per month at $0.06 per kWh. The breweryuses an average of 24,000 lb/hr of 30 psig saturatedsteam. Three 300-BHP gas fired boilers produce steamat 35 psig, to allow for pressure losses. The minimumsteam demand is 10,000 lb/hr. The plant operates continuouslyduring ten months or 7,000 hr/year. The baseor minimum electrical load during production is 3,200kW. The rest of the time (winter) the brewery is downfor maintenance. The gas costs $3.50/MMBtu.Case 1: Consider the gas turbine whose ratings are givenon Figure 7.11. We will evaluate this turbine in conjunctionwith an unfired water-tube HRSG to supply part ofthe brewery’s heat and power loads. First, we obtain theratings and performance data for the selected turbine,which has been sized to meet the electrical base load (3.5MW). An air washer/evaporative cooler will be installed

COGENERATION AND DISTRIBUTED GENERATION 183Table 7.7 Steam Generation and Fuel Chargeable to Power: 25,000-kW ISO Gas Turbine and HRSG (DistillateOil Fuel) a———————————————————————————————————————————————————Type HRSG Unfired Supplementary Fired Fully Fired———————————————————————————————————————————————————Gas TurbineFuel (10 6 Btu/hr HHV) 297®Output (MW) 21.8 21.6 21.4Airflow (10 3 lb/hr) 915®Exhaust temperature (°F) 920 922 925HRSG fuel (10 6 Btu/hr HHV) NA 190 769Steam FCP Steam FCP Steam FCP(10 3 lb/hr) (Btu/kWh HHV) (10 3 lb/hr) (Btu/kWh HHV) (10 3 lb/hr) (Btu/kWh HHV)Steam conditions250 psig sat. 133 6560 317 5620 851 4010400 psig, 650°F 110 7020 279 5630 751600 psig, 750°F 101 7340 268 5660 722850 psig, 825°F 93 7650 261 5700 7031250 psig, 900°F — — 254 5750 6871450 psig, 950°F — — 250 5750 675†———————————————————————————————————————————————————a Basis: (1) gas turbine performance given for 80°F ambient temperature, sea-level site; (2) HRSG performance based on 3%blowdown, 1-1/2% radiation and unaccounted losses, 228°F feedwater; (3) no HRSG bypass stack loss; (4) gas turbine exhaustpressure loss is 10 in. H 2 O with unfired, 14 in. H 2 O with supplementary fired, and 20 in. H 2 O with fully fired HRSG; (5) fullyfired HRSG based on 10% excess air following the firing system and 300°F stack. (6) fuel chargeable to gas turbine powerassumes total fuel credited with equivalent 88% boiler fuel required to generate steam; (7) steam conditions are at utilizationequipment; a 5% AP and 5°F AT have been assumed from the outlet of the HRSG.at the turbine inlet to improve (reduce) the overall heatrate by precooling the inlet air to an average 70°F (80°For less), during the summer production season Additionaloperating data are given below.Operating DataInlet air pressure losses (filter and airpre-cooler): 5” H 2 OExhaust Losses (ducting, by-pass valve,HRSG and Stack): 12” H 2 OLocation Elevation above sea level: 850 ftFig. 7.20 Effect of different fuel and power cost on cogenerationprofitability: Example 2. Basis: Conditionsgiven in Tables 7.4 and 7.5.Thus, on a preliminary basis, we assume the turbinewill constantly run at full capacity, minus the effectof elevation, the inlet air pressure drop and exhaustlosses. Since the plant will be located at 850 ft above sealevel, from Figure 7.11, the elevation correction factoris 0.90. Hence, the corrected continuous power rating(before deducting pressure losses) when firing naturalgas and using 70°F inlet air is:

184 ENERGY MANAGEMENT HANDBOOKFigure 7.21 Gas turbine/HRSG cogeneration application.= (Generator Output @ 70°F)(Elevation correction @ 850 ft)= 4,200 kWe × 0.9= 3,780 kWeNext, by using the Inlet and Exhaust Power Lossgraphs in Figure 7.11, we get the exhaust and inletlosses (@ 3780 kW output): 17 and 7 kW/inch H 2 O,respectively. So, the total power losses due to inlet andexhaust losses are:= (17 in)(5 kW/in) + (12 in)(8 kW/in)= 181 kWConsequently, the net turbine output after elevation andpressure losses is= 3,780 - 181= 3,599 kWeNext, from Figure 7.11 we get the following performancedata for 70°F inlet air:Heat rate: 12,250 Btu/kWh (LHV)Exhaust Temperature : 935°FExhaust Flow : 160,000 lb/hrThese figures have been used as input data forHGPRO—a prototype HRSG software program developedby V. Ganesh, W.C. Turner and J.B. Wong in 1992at Oklahoma State University. The program results areshown in Fig. 7.22.The total installed cost of the complete cogenerationplant including gas turbine, inlet air precooling,HRSG, auxiliary equipment and computer based controlsis $4,500,000. Fuel for cogeneration is available ona long term contract basis (>5 years) at $2.50/MMBtu.The brewery has a 12% cost of capital. Using a 10-yearafter tax cash flow analysis with current depreciationand tax rates, should the brewery invest in this cogenerationoption? For this evaluation, assume: (1) A 1%inflation for power and non-cogen natural gas; (2) anoperation and maintenance (O&M) cost of $0.003/kWhfor the first year after the project is installed. Then, theO&M cost should escalate at 3% per year; (3) the plantsalvage value is neglected.Economic AnalysisNext, we present system operation assumptionsrequired to conduct a preliminary economic analysis.1) The cogeneration system will operate during allthe production season (7,000 hrs/year).2) The cogeneration system will supply an averageof 3.5 MW of electrical power and 24,000 lb of 35psig steam per hour. The HRSG will be providedwith an inlet gas damper control system to modulateand by-pass hot gas flow. This is to allow forvariable steam production or steam load-followingoperation.3) The balance of power will be obtained from theexisting utility at the current cost ($0.06/kWh)4) The existing boilers will remain as back-up units.Any steam deficit (considered to be negligible) willbe produced by the existing boiler plant.5) The cogeneration fuel (natural gas) will be meteredwith a dedicated station and will be available at

COGENERATION AND DISTRIBUTED GENERATION 185Figure 7.22 Results from HGPRO 1.0, aprototype HRSG software.$2.50/MMBtu during the first five years and at$2.75/MMBtu during the next five-year period.Non cogeneration fuel will be available at the currentprice of $3.50/MMBtu.The discounted cash flow analysis was carried outusing an electronic spreadsheet (Table 7.8). The resultsof the spreadsheet show a positive net present value.Therefore, when using the data and assumptions givenin this case, the cogeneration project appears to be costeffective. The brewery should consider this project forfunding and implementation.Note: These numbers ignore breakdowns and possibleratchet clause effects.7.6 CLOSURECogeneration has been used for almost a century tosupply both process heat and power in many largeindustrial plants in the United States. This technologywould have been applied to a greater extent if we didnot experience a period of plentiful low-cost fuel andreliable low-cost electric power in the 25 years followingthe end of World War 11. Thus economic rather thantechnical considerations have limited the application ofthis energy-saving technology.The continued increase in the cost of energy isthe primary factor contributing to the renewed interestin cogeneration and its potential benefits. Thischapter discusses the various prime movers thatmerit consideration when evaluating this technology.Furthermore, approximate performance levels andtechniques for developing effective cogeneration systemsare presented.The cost of all forms of energy is rising sharply.Cogeneration should remain an important factor in effectivelyusing our energy supplies and economicallyproviding goods and services in those base-load applicationsrequiring large quantities of process heat andpower.7.6 REFERENCES1. Butler, C.H., (1984), Cogeneration: Engineering, DesignFinancing, and Regulatory Compliance, McGraw-Hill,Inc., New York, N.Y.2. Caton, J.A., et al., (1987), Cogeneration Systems, TexasA&M University, College Station, TX.3. CFR-18 (1990): Code of Federal Regulations, Part 292Regulations Under Sections 201 and 210 of the Public

186 ENERGY MANAGEMENT HANDBOOK

COGENERATION AND DISTRIBUTED GENERATION 187Utility Regulatory Policies Act of 1978 With Regardto Small Power Production and Cogeneration, (4-1-90Edition).4. Estey P.N., et al., (1984). “A Model for Sizing CogenerationSystems,” Proceedings of the 19th IntersocietyEnergy Conversion Engineering Conference” Vol. 2 of4, August, 1984, San Francisco, CA.5. Ganapathy, V. (1991). Waste Heat Boiler Deskbook, TheFairmont Press, Inc., Lilburn, CA.6. Harkins H.L., (1981), “PURPA New Horizons forElectric Utilities and Industry,” IEEE Transactions, Vol.PAS-100, pp 27842789.7. Hay, N., (1988), Guide to Natural Gas Cogeneration, TheFairmont Press, Lilburn, GA.8. Kehlhofer, R., (1991). Combined-Cycle Gas & SteamTurbine Power Plants, The Fairmont Press, Inc. Lilburn,Ga.9. Kostrzewa, L.J. & Davidson, K.G., (1988). “PackagedCogeneration,” ASHRAE Journal, February 1988.10. Kovacik, J.M., (1982), “Cogeneration,” in Energy ManagementHandbook, ed. by W.C. Turner, Wiley, New York,N.Y.11. Kovacik, J.M., (1985), “Industrial Cogeneration: SystemApplication Consideration,” Planning Cogeneration Systems,The Fairmont Press, Lilburn, Ga.12. Lee, R.T.Y., (1988), “Cogeneration System SelectionUsing the Navy’s CELCAP Code,” Energy Engineering,Vol. 85, No. 5, 1988.13. Limaye, D.R. and Balakrishnan, S., (1989), “Technicaland Economic Assessment of Packaged CogenerationSystems Using Cogenmaster,” The Cogeneration Journal,Vol. 5, No. 1, Winter 1989-90.14. Limaye, D.R., (1985), Planning Cogeneration Systems, TheFairmont Press, Atlanta, CA.15. Limaye, D.R., (1987), Industrial Cogeneration Applications,The Fairmont Press, Atlanta, CA.16. Mackay, R. (1983). “Gas Turbine Cogeneration: Design,Evaluation and Installation.” The Garret Corporation,Los Angeles, CA, The Association Of Energy Engineers,Los Angeles CA, February, 1983.17. Makansi, J., (1991). “Independent Power/Cogeneration,Success Breeds New Obligation-Delivering on Performance,”Power, October 1991.18. Mulloney, et. al., (1988). “Packaged Cogeneration InstallationCost Experience,” Proceedings of The 11thWorld Energy Engineering Congress, October 18-21,1988.19. Orlando, J.A., (1991). Cogeneration Planners Handbook,The Fairmont Press, Atlanta, GA.20. Oven, M., (1991), “Factors Affecting the Financial ViabilityApplications of Cogeneration,” XII SeminarioNacional Sobre El Uso Racional de La Energia,” MexicoCity, November, 1991.21. Polimeros, G., (1981), Energy Cogeneration Handbook,IndustrialPress Inc., New, York.22. Power (1980), “FERC Relaxes Obstacles to Cogeneration,”Power, September 1980, pp 9-10.23. SFA Pacific Inc. (1990). “Independent Power/Cogeneration,Trends and Technology Update,” Power, October1990.24. Somasundaram, S., et al., (1988), A Simplified Self-HelpApproach To Sizing of Small-Scale Cogeneration Systems,Texas A&M University, College Station, TX25. Spiewak, S.A. and Weiss L., (1994) Cogeneration & SmallPower Production Manual, 4th Edition, The FairmontPress, Inc. Lilburn, CA.26. Turner, W.C. (1982). Energy Management Handbook, JohnWiley & Sons, New York, N.Y.27. Tuttle, D.J., (1980), PURPA 210: New Life for Cogenerators,”Power, July, 1980.28. Williams, D. and Good, L., (1994) Guide to the EnergyPolicy Act of 1992, The Fairmont Press, Inc. Lilburn,GA.29. Wong, J.B., Ganesh, V. and Turner, W.C. (1991), “SizingCogeneration Systems Under Variable Loads,” 14thWorld Energy Engineering Congress, Atlanta, GA.30. Wong, J.B. and Turner W.C. (1993), “Linear Optimizationof Combined Heat and Power Systems,” IndustrialEnergy Technology Conference, Houston, March,1993.APPRECIATIONMany thanks to Mr. Lew Gelfand for using andtesting over the years the contents of this chapter inthe evaluation and development of actual cogenerationopportunities, and to Mr. Scott Blaylock for the informationprovided on fuel cells and microturbines. Messrs.Gelfand and Blaylock are with DukeEnergy/DukeSolutions.

188 ENERGY MANAGEMENT HANDBOOKAppendix AStatistical Modeling of Electric Demand andPeak–Shaving Generator Economic OptimizationJorge B. Wong, Ph.D., PE, CEMABSTRACTThis paper shows the development a basic electricdemand statistical model to obtain the optimal kW–sizeand the most cost–effective operating time for an electricalpeak shaving generator set. This model considersthe most general (and simplified) case of a facilitywith an even monthly demand charge and a uniformlydistributed random demand, which corresponds to alinear load–duration curve. A numerical example andcomputer spreadsheet output illustrate the model.INTRODUCTIONThroughout the world, electrical utilities includea hefty charge in a facility’s bill for the peak electricaldemand incurred during the billing period, usually amonth. Such a charge is part of the utility’s cost recoveryor amortization of newly installed capacity and for operatingless efficient power plant capacity during higherload periods.Demand charge is a good portion of a facility’selectrical bill. Typically a demand charge can be asmuch as 50% of the bill, or more. Thus, to reduce thedemand cost, many industrial and commercial facilitiestry to “manage their loads.” One example is by movingsome of the electricity–intense operations to “off–peak”hours”—when a facility’s electrical load is much smallerand the rates ($/kW) are lower. But, when moving electricalloads to “off–peak” hours is not practical or significant,a facility will likely consider a set of engine–drivenor fuel cell generators to run in parallel with the utilitygrid to supply part or all the electrical load demandduring “on–peak” hours. We call these Peak ShavingGenerators or PSGs.While the electric load measurement is instantaneous,the billing demand is typically a 15–to–30– minuteaverage of the instantaneous electrical powerdemand (kW). To obtain the monthly demand charge,utilities multiply the billing demand by a demand rate.Some utilities charge a flat rate ($/kW–peak per month)for all months of the year. Other utilities have seasonalcharges (i.e. different rates for different seasons of theyear). Still, others use ratchet clauses to account for thehighest “on–peak” season demand of the year.Thus, the model presented in this paper focuses onthe development of a method to obtain the optimal PSGsize (g*kW) and PSG operation time (hours per year)for a given facility. This model is for the case of a facilitywith a constant billing demand rate ($/kW/month)throughout the year. The analysis is based on a linearload–duration curve and uses a simplified life–cycle–cost approach. An example illustrates the underlyingapproach and optimization method. In addition, thepaper shows an EXCEL spreadsheet to implement theoptimization model. We call this model PSG–1.ELECTRIC DEMAND STATISTICAL MODELThis section develops the statistical–and–mathmodel for the economical sizing of an electrical peak–shaving generator set (PSG) for a given facility. The fundamentalquestion is: What is the most economical generator–setsize—g* in kW—for a given site demand profile?Figure 1 shows a sample record for a facility’s electricaldemand, which is uniformly distributed between 2000and 5300 kW. Next, Figure 2 shows the correspondingstatistical distributions.The statistical model of electrical demand is expressedgraphically in Figure 2, in terms of two functions:• The load–duration curve D(t), is the demand as afunction of cumulative time t (i.e. the accumulatedannual duration t in hrs/year of a given D(t) loadin kW), and• The load frequency distribution f(D) (rectangularshaded area in Figures 1 and 2) is the “uniform”probability density function.MODEL ASSUMPTIONSThis statistical model is based upon the followingassumptions:1. The electrical demand is represented by a linearload–duration curve, as shown in Figure 2. Thus,for a typical year, the facility has a demand Dthat varies between an upper value D u (annualmaximum) and a lower value D 1 (annual minimum).This implies the electrical load is uniformlydistributed between the maximum and minimumdemands. The facility operates T hours per year.

COGENERATION AND DISTRIBUTED GENERATION 189Figure 1. Sample record for a uniformly distributed random demand.Figure 2. Load—Duration Curve for uniformly distributed demand.2. There is an even energy or consumption rate Ce($/kWh) throughout the year.3. There is an even demand rate Cd ($/kW/month)for every month of the year.4. There is a same demand peak D u for every month.Demand ratchet clauses are not applicable in thiscase.5. The equipment’s annual ownership or amortizationunit installed cost ($/kW/year) is constant forall sizes of PSGs. The unit ownership or rental cost($/kW/year) is considered independent of unitsize. Ownership, rental or lease annualized costsare denoted by A c .6. A PSG set is installed to reduce the peak demandby a maximum of g kW, operating t g hours peryear.BASE CASE ELECTRICITY ANNUALCOST—WITHOUT PEAK–SHAVINGConsider a facility with the load–duration characteristicshown in Figures 1 and 2. For a unit consumptioncost Ce, the annual energy or consumption cost (withoutPSG) for the facility isAEC = T• D 1 • Ce + 1/2 T (D u – D 1 ) CeWhich is equivalent to

190 ENERGY MANAGEMENT HANDBOOKAEC = T/2 • Ce (D u + D 1 ) [1]Next, considering a peak demand D u occurs everymonth, the annual demand cost is defined byADC = 12 D u • Cd [2]Thus, the total annual cost for the facility isTAC = AEC + ADC[3a]Substituting [1] and [2] in equation [3], we have the basecase total annual cost:TAC 1 = T/2 (D u + D 1 ) Ce + 12 D u • CdELECTRICITY ANNUAL COST WITHPEAK–SHAVING[3b]If a peak shaving generator of size g is installed inthe facility to run in parallel with the utility grid duringpeak–load hours, so the maximum load seen by theutility is (D u – g), then the electric bill cost isEBC = T/2 (D u – g + D 1 ) Ce + 12 (D u – g) • CdIn addition, the facility incurs an ownership (amortization)unit cost Ac ($/kW/yr) and operation andmaintenance unit cost O&M ($/kWh). Hence, the totalannual cost with demand peak shaving isTAC 2 = [T• D 1 + (T+ t g )/2 (D u – g – D 1 )]Ce +12 (D u – g) Cd + (Ac + 1/2 O&M • t g )g [4]ANNUAL WORTH OF THE PEAK-SHAVING GENERATORThe annual worth or net savings AW ($/yr) of thePSG set are obtained by subtracting equation [4] fromequation [3]. That is AW = TAC 1 – TAC 2 . So,AW = 1/2 t g • g • Ce + 12 • g• Cd –(Ac + 1/2 • O&M • t g )g [5]From Figure 2 we obtain g: t g = (D u – D 1 ): TSo, the expected PSG operating time ist g = g • T/(D u – D 1 ) [6]Substituting the value of t g in equation [5], we have:AW = g 2 • T/[2(D u – D 1 )] Ce + 12 • g • Cd –{Ac + O&M • g • T/[2(D u – D 1 )]} g [7]OPTIMUM CONDITIONSWe next determine the necessary and sufficientconditions for an optimal PSG size g* and the correspondingmaximum AW to exist.Necessary ConditionBy taking the derivative of AW, Equation [7], withrespect to g and equating it to zero we obtain the necessarycondition for the maximum annual worth or netsaving per year. That is:AW’ = g • T • Ce/(D u – D 1 ) + 12 Cd – Ac –g • T • O&M/(D u – D 1 ) = 0 [8]Sufficient Condition. If the second derivative ofAW with respect to g is negative, i.e. AW”

COGENERATION AND DISTRIBUTED GENERATION 191FOR FURTHER RESEARCHFurther research is underway to develop enhancedmodels which consider:• Demand profile flexibility. Other load–durationshapes with different underlying frequency distributions(e.g. triangular, normal and auto–correlatedloads).• Economies of Scale. The fact that larger units havebetter fuel–to–electricity efficiencies (lower heatrates) and lower per unit installed cost ($/kW).__________________________________________________EXAMPLE. A manufacturing plant operates 7500 hoursper year and has a fairly constant electrical (billing) peakdemand every month (See Figure 1). The actual load,however, varies widely between a minimum of 2000 kWand a maximum of 5300 kW (See Figure 2). The demandcharge is $10/kW/month and the energy charge is$0.05/kWh. The installed cost of a diesel generator set,the auxiliary electrical switch gear and peak–shavingcontrols is about $300 per kW. Alternatively, the plantcan lease a PSG for $50/kW/yr. The operation andmaintenance cost (including diesel fuel) is $0.10/kWh.Assuming the plant leases the PSG, estimate (1) the optimalPSG size, (2) the annual savings and (3) the PSGannual operation time.1) The optimal generator size is calculated using equation[9](12. $10 – $50/kWh) (5300 – 2000 kW)g* = —————————————————7500 h/yr ($0.10/kWh –$0.05/kWh)= 616 kW2) Using a commercially available PSG of size g* = 600kW, the potential annual savings are estimated usingequation [7]AW = g 2 • T • Ce/[2(D u – D 1 )] + 12 • g • Cd– {Ac + O&M • g • T/[2(D u – D 1 )]} g= 600 2 × 7500 x 0.05/(2(5300 – 2000)) + 12 × 600 × 10– ($50 + 0.10 × 600 × 7500/(2 (5300 – 2000))) 600= $20,455 + $72,000 – $70,909= $21,546/year3) The expected annual operating time for the PSG isestimated using Equation [6]Figure 3. PSG-1 Spreadsheet and Chart

192 ENERGY MANAGEMENT HANDBOOKt g = g • T/(D u – D 1 )= 600 × 7500/(5300–2000)= 1,364 hours/yearThe Excel spreadsheet and chart used to solve this caseexample is shown in Figure 3.CONCLUDING REMARKSThe reader should note that the underlying statisticaland optimization model is quite “responsive androbust.” That is, the underlying methodology can beused in, or adapted to, a variety of demand profiles andrates, while the results remain relatively valid. A forthcomingpaper by this author will show how to adaptthe linear load– duration models of Figures 1 and 2 tomore complex demand profiles. Thus, for example, onetypical case is when the electrical load is represented bya Gauss or normal distribution. Also, we will show howto apply equation [9] to more involved industrial caseswith multiple billing seasons and demand rates.Appendix ReferencesBeightler, C.S., Phillips, D.T., and Wilde, D.J., Foundationsof Optimization, Prentice–Hall, EnglewoodCliffs, 1979.Turner, W.C., Energy Management Handbook, 4th Edition,the Fairmont Press, Lilburn, GA, 2001.Hahn & Shapiro, Statistical Models in Engineering, JohnWiley 1967, Wiley Classics Library, reprinted in1994.Witte, L.C., Schmidt, P.S., and Brown, D.R, IndustrialEnergy Management and Utilization, HemispherePublishing Co. and Springer–Verlag, Berlin, 1988.Appendix NomenclatureAc Equipment ownership, lease or rental cost ($/kW/year)ADC Annual Demand Cost ($/year)AEC Annual Energy Cost ($/year)AW Annual Worth ($/year)Cd Electric demand unit cost ($/kW/month)Ce Electric energy unit cost ($/kWh)D Electric demand or load (kW)D 1 Lower bound of a facility’s electric demand orminimum load (kW)D u Upper bound of a facility’s electric demand ormaximum load (kW)EBC Electric bill cost for a facility with PSG, ($/year)f(D) Frequency of occurrence of a demand, (unit less)O&M Operation and Maintenance cost, including fuelcost ($/kWh)g Peak shaving generator size or rated output capacity(kW)g* Optimal peak shaving generator size or outputcapacity (kW)t Time, duration of a given load, (hours/year)t g Expected time of operation for a PSG, hours/yearT Facility operation time using power(hours/year)TAC Total annual electric costTAC 1 Total annual cost, base case w/o PSG ($/year)TAC 2 Total annual cost, with PSG ($/year)Jorge B. Wong, Ph.D., PE, CEM is an energy managementadvisor and instructor. Jorge helps facility managers andengineers. Contact Jorge: jorgebwong@att.net

CHAPTER 8WASTE-HEAT RECOVERYWESLEY M. ROHRER, JR.Emeritus Associate Professor ofMechanical EngineeringUniversity of PittsburghPittsburgh, Pennsylvania8. 1 INTRODUCTION8.1.1 DefinitionsWaste heat, in the most general sense, is the energyassociated with the waste streams of air, exhaust gases,and/or liquids that leave the boundaries of a plant orbuilding and enter the environment. It is implicit thatthese streams eventually mix with the atmospheric air orthe groundwater and that the energy, in these streams,becomes unavailable as useful energy. The absorptionof waste energy by the environment is often termedthermal pollution.In a more restricted definition, and one that willbe used in this chapter, waste heat is that energy whichis rejected from a process at a temperature high enoughabove the ambient temperature to permit the economicrecovery of some fraction of that energy for useful purposes.8.1.2 BenefitsThe principal reason for attempting to recoverwaste heat is economic. All waste heat that is successfullyrecovered directly substitutes for purchased energyand therefore reduces the consumption of and the costof that energy. A second potential benefit is realizedwhen waste-heat substitution results in smaller capacityrequirements for energy conversion equipment. Thusthe use of waste-heat recovery can reduce capital costsin new installations. A good example is when waste heatis recovered from ventilation exhaust air to preheat theoutside air entering a building. The waste-heat recoveryreduces the requirement for space-heating energy. Thispermits a reduction in the capacity of the furnaces orboilers used for heating the plant. The initial cost of theheating equipment will be less and the overhead costswill be reduced. Savings in capital expenditures forthe primary conversion devices can be great enough tocompletely offset the cost of the heat-recovery system.Reduction in capital costs cannot be realized in retrofit193installations unless the associated primary energy conversiondevice has reached the end of their useful livesand are due for replacement.A third benefit may accrue in a very special case.As an example, when an incinerator is installed todecompose solid, liquid, gaseous or vaporous pollutants,the cost of operation may be significantly reducedthrough waste-heat recovery from the incinerator exhaustgases.Finally, in every case of waste-heat recovery, agratuitous benefit is derived: that of reducing thermalpollution of the environment by an amount exactlyequal to the energy recovered, at no direct cost to therecoverer.8.1.3 Potential for Waste-Heat Recovery in IndustryIt had been estimated 1 that of the total energyconsumed by all sectors of the U.S. economy in 1973,that fully 50% was discharged as waste heat to theenvironment. Some of this waste is unavoidable. Thesecond law of thermodynamics prohibits 100% efficiencyin energy conversion except for limiting cases which arepractically and economically unachievable. Ross andWilliams, 2 in reporting the results of their second-lawanalysis of U.S. energy consumption, estimated that in1975, economical waste-heat recovery could have savedour country 7% of the energy consumed by industry, or1.82 × 10 16 Btus (1.82 quads.)Roger Sant 3 estimated that in 1978 industrial heatrecovery could have resulted in a national fuel savingsof 0.3%, or 2.65 × 10 16 Btus (2.65 quads). However, hisstudy included only industrial furnace recuperators.*In terms of individual plants in energy-intensive industries,this percentage can be greater by more than anorder of magnitude.The Annual Energy Review 1991 4 presents data toshow that although U.S. manufacturing energy intensityincreased by an average of 26.7% during the period 1980to 1988, the manufacturing sector’s energy use efficiency,for all manufacturing, increased by an average of 25.1%.In reviewing the Annual Energy Reviews over the years,it becomes quite clear that during periods of rising fuel*Recuperators are heat exchangers that recover waste heat from the stacksof furnaces to preheat the combustion air. Section 8.4.2 subjects this deviceto more detailed scrutiny.

194 ENERGY MANAGEMENT HANDBOOKprices energy efficiency increases, while in periods ofdeclining fuel prices energy efficiency gains are eroded.Although the average gain in energy use efficiency, inthe 7-year period mentioned above, is indeed impressive,several industrial groups accomplished much lessthan the average or made no improvements at all duringthat time. As economic conditions change to favorinvestments in waste-heat recovery there will be furtherlarge gains made in energy use efficiency throughoutindustry.8.1.4 Quantifying Waste HeatThe technical description of waste heat must necessarilyinclude quantification of the following characteristics:(1) quantity, (2) quality, and (3) temporal availability.The quantity of waste heat available is ordinarilyexpressed in terms of the enthalpy flow of the wastestream, orwhereH = mh (.1)H = total enthalpy flow rate of waste stream, Btu ⁄ hrm = mass flow rate of waste stream, lb ⁄ hrh = specific enthalpy of waste stream, Btu ⁄ lbThe mass flow rate, m, can be calculated from the expressionm = ρQ (8.2)where ρ = density of material, lb/ft 3Q = volumetric flow rate, ft 3 /hrThe potential for economic waste-heat recovery, however,does not depend as much on the quantity availableas it does on whether its quality fits the requirements ofthe potential heating load which must be supplied andwhether the waste heat is available at the times whenit is required.The quality of waste heat can be roughly characterizedin terms of the temperature of the wastestream. The higher the temperature, the more availablethe waste heat for substitution for purchased energy.The primary source of energy used in industrial plantsare the combustion of fossil fuels and nuclear reaction,both occurring at temperatures approaching 3000°F.Waste heat, of any quantity, is ordinarily of little useat temperatures approaching ambient, although the useof a heat pump can improve the quality of waste heateconomically over a limited range of temperatures nearand even below ambient. As an example, a waste-heatstream at 70°F cannot be used directly to heat a fluidstream whose temperature is 100°F. However, a heatpump might conceivably be used to raise the temperatureof the waste heat stream to a temperature above100°F so that a portion of the waste-heat could then betransferred to the fluid stream at 100°F. Whether this iseconomically feasible depends upon the final temperaturerequired of the fluid to be heated and the cost ofowning and operating the heat pump.8.1.5 Matching Loads to SourceIt is necessary that the heating load which will absorbthe waste heat be available at the same time as thewaste heat. Otherwise, the waste heat may be useless,regardless of its quantity and quality. Some examples ofsynchrony and non-synchrony of waste-heat sources andloads are illustrated in Figure 8.1. Each of the graphs inthat figure shows the size and time availability of a wasteheatsource and a potential load. In Figure 8.1a the sizeof the source, indicated by the solid line, is an exhauststream from an oven operating at 425°F during the secondproduction shift only. One possible load is a waterheater for supplying a washing and rinsing line at 135°F.As can be seen by the dashed line, this load is availableonly during the first shift. The respective quantities andqualities seem to fit satisfactorily, but the time availabilityof the source could not be worse. If the valuable sourceis to be used, it will be necessary to (1) reschedule eitherof the operations to bring them into time correspondence,(2) generate the hot water during the second shift andstore it until needed at the beginning of the first shiftthe next day, or (3) find another heat load which has anoverall better fit than the one shown.In Figure 8.1b we see a waste-heat source (solidline) consisting of the condenser cooling water of anair-conditioning plant which is poorly matched with itsload (dashed line)—the ventilating air preheater for thebuilding. The discrepancy in availability is not diurnalas before, but seasonal.In Figure 8.1c we see an almost perfect fit forsource and load, but the total availability over a 24 hourperiod is small. The good fit occurs because the source,the hot exhaust gases from a heat-treat furnace, is usedto preheat combustion air for the furnace burner. However,the total time of availability over a 24-hour periodis so small as to cast doubt on the ability to pay off thecapital costs of this project.8.1.6 Classifying Waste-Heat QualityFor convenience, the total range of waste-heattemperatures, 80 to 3000°F, is broken down into three

WASTE-HEAT RECOVERY 195Low range 80 ≤ T < 400Waste heat in the high-temperature range is notonly the highest quality but is the most useful, and costsless per unit to transfer than lower-quality heat. However,the equipment needed in the highest part of therange requires special engineering and special materialsand thus requires a higher level of investment. All of theapplications listed in Table 8.1 result from direct-firedprocesses. The waste heat in the high range is availableto do work through the utilization of steam turbines orgas turbines and thus is a good source of energy forcogeneration plants.*Table 8.2 gives the temperatures of waste gasesprimarily from direct-fired process equipment in themedium-temperature range. This is still in the temperaturerange in which work may be economically extractedusing gas turbines in the range 15 to 30 psig or steamturbines at almost any desired pressure. It is an economicrange for direct substitution of process heat sincerequirements for equipment are reduced from those inthe high-temperature range.The use of waste heat in the low-temperature rangeis more problematic. It is ordinarily not practical toextract work directly from the waste-heat source in thistemperature range. Practical applications are generallyfor preheating liquids or gases. At the higher temperaturesin this range air preheaters or economizers can beTable 8.1 Waste-heat sources in the high-temperaturerange.Figure 8.1 Matching waste-heat sources and loads.subranges: high, medium, and low. These classes aredesigned to match a similar scale which classifies commercialwaste-heat-recovery devices. The two systemsof classes allow matches to be made between industrialprocess waste heat and commercially available recoveryequipment. Subranges are defined in terms of temperaturerange as:High range 1100 ≤ T ≤ 3000Medium range 400 ≤ T < 1100Type of DeviceTemperature (°F)Nickel refining furnace 2500-3000Aluminum refining furnace 1200-1400Zinc refining furnace 1400-2000Copper refining furnace 1400-1500Steel heating furnaces 1700-1900Copper reverberatory furnace 1650-2000Open hearth furnace 1200-1300Cement kiln (dry process) 1150-1350Glass melting furnace 1800-2800Hydrogen plants 1200-1800Solid waste incinerators 1200-1800Fume incinerators 1200-2600*The waste heat generates high-pressure steam in a waste-heat boiler whichis used in a steam turbine generator to generate electricity. The turbineexhaust steam at a lower pressure provides process heat. Alternatively,the high-temperature gases may directly drive a gas turbine generatorwith the exhaust generating low-pressure steam in a waste-heat boilerfor process heating.

196 ENERGY MANAGEMENT HANDBOOKTable 8.2 Waste-heat sources in the medium-temperaturerange.Type of Deviceutilized to preheat combustion air or boiler make-upwater, respectively. At the lower end of the range heatpumps may be required to raise the source temperatureto one that is above the load temperature. An exampleof an application which need not involve heat pumpassistance would be the use of 95°F cooling water froman air compressor to preheat domestic hot water fromits ground temperature of 50°F to some intermediatetemperature less than 95°F. Electric, gas-fired, or steamheaters could then be utilized to heat the water to thetemperature desired. Another application could be theuse of 90°F cooling water from a battery of spot weldersto preheat the ventilating air for winter space heating.Since machinery cooling can’t be interrupted or diminished,the waste-heat recovery system, in this latter case,must be designed to be bypassed or supplemented whenseasonal load requirements disappear. Table 8.3 listssome waste-heat sources in the low-temperature range.8.1.7 Storage of Waste HeatWaste heat can be utilized to adapt otherwisemismatched loads to waste-heat sources. This is possiblebecause of the inherent ability of all materials toabsorb energy while undergoing a temperature increase.The absorbed energy is termed stored heat. The quantitythat can be stored is dependent upon the temperaturerise that can be achieved in the storage material as wellas the intrinsic thermal qualities of the material, and canbe estimated from the equationT 2Q= mC dT =T 1T 1T 2ρ VC dTTemperature (°F)Steam boiler exhausts 450-900Gas turbine exhausts 700-1000Reciprocating engine exhausts 600-1100Reciprocating engine exhausts 450-700(turbocharged)Heat treating furnaces 800-1200Drying and baking ovens 450-1100Catalytic crackers 800-1200Annealing furnace cooling systems 800-1200Selective catalytic reductionsystems for NO x control 525-750= ρ VC (T – T 0 ) for constant specific heat (8.3)where m = mass of storage material, lb mρ = density of storage material, lb/ft 3V = volume of storage material, ft 3C = specific heat of storage material, Btu/lb m °RT = temperature in absolute degrees, °RThe specific heat for solids is a function of temperaturewhich can usually be expressed in the formC 0 = C 0 [1 + α (T – T 0 )] (8.4)where C 0 = specific heat at temperature T 0T 0 = reference temperatureα = temperature coefficient of specific heatIt is seen from equation 8.3 that storage materials shouldhave the properties of high density and high specificheat in order to gain maximum heat storage for a giventemperature rise in a given space. The rate at whichheat can be absorbed or given up by the storage materialdepends upon its thermal conductivity, k, which isdefined by the equationδ Qδ t=–kAdTdx x =0= Q(8.5)Table 8.3 Waste-heat sources in the low-temperaturerange.SourceTemperature (°F)Process steam condensate 130-190Cooling water from:Furnace doors 90-130Bearings 90-190Welding machines 90-190Injection molding machines 90-190Annealing furnaces 150-450Forming dies 80-190Air compressors 80-120Pumps 80- 190Internal combustion engines 150-250Air conditioning and 90-110refrigeration condensersLiquid still condensers 90-190Drying, baking, and curing ovens 200-450Hot-processed liquids 90-450Hot-processed solids 200-450

WASTE-HEAT RECOVERY 197where t = time, hrk = thermal conductivity, Btu-ft/hr ft 2 °FA = surface areadTdx x = C =temperature gradient at the surfaceThus additional desirable properties are high thermalconductivity and large surface area per unit mass (specificarea). This latter property is inversely proportionalto density but can also be manipulated by designingthe shape of the solid particles. Other important propertiesfor storage materials are low cost, high meltingtemperature, and a resistance to spalling and crackingunder conditions of thermal cycling. To summarize: themost desirable properties of thermal storage materialsare (1) high density, (2) high specific heat, (3) high specificarea, (4) high thermal conductivity, (5) high meltingtemperature, (6) low coefficient of thermal expansion,and (7) low cost.Table 8.4 lists the thermophysical properties of anumber of solids suitable for heat-storage materials.The response of a storage system to a waste-heatstream is given approximately by the following expressiondue to Rummel 4 :Q∆T l,m /(θ + θ")— = —————————————————— (8.6)A 1/h"θ + 1/h'θ' + 1/2.5C s ρ s R B /k(θ' + θ")whereT l,m = logarithmic mean temperature differencebased upon the uniform inlettemperature of each stream andthe average outlet temperaturesC s = specific heat of storage material,Btu/lb °Fρ s = density of storage material, lb/ft 3k = conductivity of storage material,Btu/hr ft °FR B = volume per unit surface area forstorage material, fth = coefficient of convective heat transferof gas streams, Btu/hr ft 2 °Fθ = time cycle for gas stream flows, hrThe primed and double-primed values refer, respectively,to the hot and cold entering streams. In cases wherethe fourth term in the denominator is large comparedto the other three terms, this equation should not beTable 8.4 Common refractory materials a,b .Mean Thermal Coefficient ofConductivity Cubical Maximum UseDensity Specific Heat (Btu/ft hr °F) Expansion Temperature Melting PointName Formula (lbm/ft 3 ) (Btu/lb m ) (to 1000°) (per °F) (°F) (°F)Alumina Al 2 O 3 230 0.24 2.0 8 × 10 –6 3300 3700Beryllium oxide BeO 190 0.24 — 9 × 10 –6 4000 4600Calcium oxide CaO 200 0.18 4.5 13 × 10 –6 4200 4700Carbon, graphite C 120 0.36 7 3 × 10 –6 4000 6500 cChrome 40% Cr 2 O 3 200 0.20 1.0 8 × 10 –6 3200 3800Corundum 90% Al 2 O 3 200 0.22 1.5 7 × 10 –6 3200 —Forsterite 2 MgO SiO 2 160 0.23 1.2 10 × 10 –6 3000 3300Magnesia MgO 210 0.25 2.3 11 × 10 –6 4000 5000Magnesium oxide MgO 175 0.25 2.0 10 × 10 –6 3500 5000Mullite 3 Al 2 O 3 SiO 2 160 0.23 1.2 5 × 10 –6 3000 3350Silica SiO 2 110 0.24 1.0 7 × 10 –6 2800 3100Silicon carbide SiC 170 0.23 8 3 × 10 –6 3000 4000 dSpinel MgO Al 2 O 3 220 0.23 5 7 × 10 –6 3300 —Titanium oxide TiO 2 260 0.17 2.2 8 × 10 –6 3000 3300Zircon ZrO 2 SiO 2 220 0.15 1.3 5 × 10 –6 3500 4500Zirconium oxide ZrO 2 360 0.13 1.3 4 × 10 –6 4400 4800a Most of these materials are available commercially as refractory tile, brick, and mortar. Properties will depend on form, purity, and mixture. Temperaturesgiven should be considered as high limits.b For density in kg/m 3 , multiply value in lb/ft 3 by 16.02. For specific heat in J/kg K, multiply value in Btu/lb m °F by 4184. For thermal conductivityin W/m K, multiply value in Btu/ft hr °F by 1.73.c Sublimes.d Dissociates.

198 ENERGY MANAGEMENT HANDBOOKused. This will occur when the cycle times are shortand the thermal resistance to heat transfer is large. Inthose cases there exists insufficient time for the particlesto get heated and cooled. Additional equations for determiningthe rise and fall in temperatures, and graphsgiving temperature histories for the flow streams andthe storage material, may be found in Rohsenow andHartnett. 58.1.8 Enhancing Waste Heat with Heat PumpsHeat pumps offer only limited opportunities forwaste-heat recovery simply because the cost of owningand operating the heat pump may exceed the value ofthe waste heat recovered.A heat pump is a device that operates cyclically sothat energy absorbed at low temperature is transformedthrough the application of external work to energy at ahigher-temperature which can be absorbed by an existingload. The commercial mechanical refrigeration plantcan be utilized as a heat pump with small modifications,as indicated in Figure 8.2. The coefficient of performance(COP) of the heat pump cycle is the simple ratio of heatdelivered to work required:COP HP = Q hQ net= Q hW net (8.7)Since the work requirement must be met by aprime mover that is either an electric motor or a liquidfueledengine, the COP must be considerably greaterthan 3.0 in order to be an economically attractive energysource. That is true because the efficiency of the primemovers used to drive the heat pump, or to generate theelectrical energy for the motor drive, have efficienciesless than 33%. The maximum theoretical COP for anideal heat pump is given byFigure 8.3 Theoretical COP vs. load temperature.COP H = 11–T L ⁄ T Hwhere T L = temperature of energy sourceT H = temperature of energy loadThe ideal cycle, however, uses an ideal turbine as a vaporexpander instead of the usual throttle valve in theexpansion line of the mechanical refrigeration plant.Figure 8.3 is a graph of the theoretical COP versusload temperature for a number of source temperatures.Several factors prevent the actual heat pump from approachingthe ideal:1. The compressor efficiency is not 100%, but is ratherin the range 65 to 85%.2. A turbine expander is too expensive to use in anybut the largest units. Thus the irreversible throttlingprocess is used instead of an ideal expansionthrough a turbine. All of the potential turbine workis lost to the cycle.3. Losses occur from fluid friction in lines, compressors,and valving.4. Higher condenser temperatures and lower evaporatortemperatures than the theoretical are requiredto achieve practical heat flow rates from the sourceand into the load.Figure 8.2 Heat pump.An actual two-stage industrial heat pump installationshowed 7 an annual average COP of 3.3 for an averagesource temperature of 78°F and a load temperatureof 190°F. The theoretical COP is 5.8. Except for verycarefully designed industrial units, one can expect to

WASTE-HEAT RECOVERY 199achieve actual COP values ranging from 50% to 65% ofthe theoretical.An additional constraint on the use of heat pumpsis that high-temperature waste heat above 230°F cannotbe supplied directly to the heat pump because of thelimits imposed by present compressor and refrigeranttechnology. The development of new refrigerants mightraise the limit of heat pump use to 400°F.8.1.9 Dumping Waste HeatIt cannot be emphasized too strongly that the interruptionof a waste heat load, either accidentally orintentionally, may impose severe operating conditionson the source system, and might conceivably causecatastrophic failures of that system.In open system cooling the problem is easier todeal with. Consider the waste-heat recovery from thecooling water from an air compressor. In this case thecooling water is city tap water which flows seriallythrough the water jackets and the intercooler and isthen used as makeup water for several heated treatmentbaths. Should it become necessary to shut off theflow of makeup water to the baths, it would be necessaryto valve the cooling water flow to a drain so thatthe compressor cooling continues with no interruption.Otherwise, the compressor would become overheatedand suffer damage.In a closed cooling system supplying waste heatto a load requires more extensive safeguards and provisionsfor dumping heat rather than fluid flow. Figure8.4 is the schematic of a refrigeration plant condensersupplying waste heat for space heating during the winter.Since the heating load varies hourly and daily, anddisappears in the summer months, it is necessary toprovide an auxiliary heat sink which will accommodatethe entire condenser discharge when the waste-heat loaddisappears. In the installation shown, the auxiliary heatsink is a wet cooling tower which is placed in serieswith the waste-heat exchanger. The series arrangementis preferable to the alternative parallel arrangement forseveral reasons. One is that fewer additional controls areneeded. Using the parallel arrangement would requirethat the flows through the two paths be carefully controlledto maintain required condenser temperature andat the same time optimize the waste-heat recovery.In the above examples the failure to absorb all ofthe available waste heat had serious consequences onthe system supplying the waste heat. A somewhat differentwaste-heat dumping problem occurs when theeffect of excessive waste-heat availability has an adverseaffect on the heat sink. An example would be the useof the cooling air stream from an air-cooled screw-typecompressor for space heating in the winter months. Duringthe summer months all of the compressor cooling airwould have to be dumped to the outdoors in order toprevent overheating of the work space.8.1.10 Open Waste-Heat ExchangersAn open heat exchanger is one where two fluidstreams are mixed to form a third exit stream whoseenergy level (and temperature) is intermediate betweenthe two entering streams. This arrangement has the advantageof extreme simplicity and low fabrication costswith no complex internal parts. The disadvantages arethat (1) all flow streams must be at the same pressure,and (2) the contamination of the exit fluids by either ofthe entrance flows is possible. Several effective applicationsof open waste-heat exchangers are listed below:1. The exhaust steam from a turbine-drivenfeedwater pump ina boiler plant is used to preheatthe feedwater in a deaeratingfeedwater heater.Figure 8.4 System with cold weather condenser pressure control.2. The makeup air for an occupiedspace is tempered by mixing itwith the hot exhaust productsfrom the stack of a gas-firedfurnace in a plenum beforedischarge into the space. Thisrecovery method may be prohibitedby codes because of thedanger of toxic carbon monoxide;a monitor should be usedto test the plenum gases.

200 ENERGY MANAGEMENT HANDBOOK3. The continuous blowdown stream from a boilerplant is used to heat the hot wash and rinse waterin a commercial laundry. A steam-heated storageheater serves as the open heater.8.1.11 Serial Use of Process Air and WaterIn some applications, waste streams of process airand water can be directly used for heating without priormixing with other streams. Some practical applicationsinclude:1. Condenser cooling water from batch coolers useddirectly as wash water in a food-processing plant;2. steam condensate from wash water heaters addeddirectly to wash water in the bottling section of abrewery;3. air from the cooling section of a tunnel kiln used asthe heating medium in the drying rooms of a refractory;4. condensate from steam-heated chemical baths returneddirectly to the baths; and5. the exhaust gases from a waste-heat boiler used asthe heating medium in a lumber kiln.In all cases, the possibility of contamination from amixed or a twice-used heat-transport medium must beconsidered.8.1.12 Closed Heat ExchangersAs opposed to the open heat exchanger, the closedheat exchanger separates the stream containing theheating fluid from the stream containing the heatedfluid, but allows the flow of heat across the separatingboundaries. The reasons for separating the streams maybe:1. A pressure difference may exist between the twostreams of fluid. The rigid boundaries of the heatexchanger are designed to withstand the pressuredifferences.Closed heat exchangers fall into the general classificationof industrial heat exchangers, however theyhave many pseudonyms related to their specific form orto their specific application. They can be called recuperators,regenerators, waste-heat boilers, condensers, tubeand-shellheat exchangers, plate-type heat exchangers,feedwater heaters, economizers, and so on. Whatevername is given all perform one basic function: the transferof heat across rigid and impermeable boundaries.Sections 8.3 and 8.4 provide much more technical detailconcerning the theory, application, and commercialavailability of heat exchangers.8.1.13 Runaround SystemsWhenever it is necessary to ensure isolation ofheating and heated systems, or when it becomes advantageousto use an intermediate transfer medium becauseof the long distances between the two systems, a runaroundheat recovery system is used. Figure 8.5 showsthe schematic of a runaround system which recoversheat from the exhaust stream from the heating and ventilatingsystem of a building. The circulating mediumis a water-glycol mixture selected for its low freezingpoint. In winter the exhaust air gives up some energy tothe glycol in a heat exchanger located in the exhaust airduct. The glycol is circulated by way of a small pump toa second heat exchanger located in the inlet air duct. Theoutside air is preheated with recovered waste-heat thatsubstitutes for heat that would otherwise be added inthe main heating coils of the building’s air handler. Duringthe cooling season the heat exchanger in the exhaustduct heats the exhaust air, and the one in the inlet ductprecools the outdoor air prior to its passing through thecooling coils of the air handler. The principal reason forusing a runaround system in this application is the longseparation distance between the inlet air and the exhaustair ducts. Had these been close together, one air-to-air2. One stream could contaminate the other if allowedto mix. The impermeable, separating boundaries ofthe heat exchanger prevents mixing.3. To permit the use of an intermediate fluid bettersuited than either of the principal exchange mediafor transporting waste heat through long distances.While the intermediate fluid is often steam, glycoland water mixtures and other substances can beused to take advantage of their special properties.Figure 8.5 Runaround heat-recovery system.

WASTE-HEAT RECOVERY 201heat exchanger (with appropriate ducting) could havebeen more economical.Figure 8.6 is the schematic diagram of a runaroundsystem used to recover the heat of condensation froma chemical bath steam heater. In this case the bath isa highly corrosive liquid. A leak in the heater coilswould cause the condensate to become contaminatedand thus do damage to the boiler. The intermediate8.2 THE WASTE-HEAT SURVEY8.2.1 How to Conduct the SurveyThe survey should be carried out as an integralpart of the energy audit of the plant. The survey consistsof a systematic study of the sources of waste heatin the plant and of the opportunities for its use. Thesurvey is carried out on three levels. The first step isthe identification of every energy-containingnonproduct that flows from the plant.Included are waste streams containingsensible heat (substances at elevated temperatures);examples are hot inert exhaustproducts from a furnace or cooling waterfrom a compressor. Also to be included arewaste streams containing chemical energy(waste fuels), such as carbon monoxidefrom a heat-treatment furnace or cupola,solvent vapors from a drying oven, or sawdustfrom a planing mill: Figure 8.7 is anexample of a survey form used for listingFigure 8.6 Runaround heat-recovery-system process steam source. waste streams leaving the plant.The second step is to learn more abouttransport fluid isolates the boiler from a potential source the original source of the waste-heat stream. Informationof contamination and corrosion. It should be noted that should be gathered that can lead to a complete heat balanceon the equipment or the system that produces it.the presence of corrosive chemicals in the bath, whichdictated the choice of the runaround system, are also in Since both potential savings and the capital costs tend tocontact with one side of the condensate heat exchanger. be large in waste-heat recovery situations, it is importantThe materials of construction for that heat exchanger that the data be correct. Errors in characterization of theshould be carefully selected to withstand the corrosion energy-containing streams can either make a poor investmentlook good, or conversely cause a good from that chemical.investmentExitDesignation Location Composition Flow Rate Temperature Heat Rate CommentsFigure 8.7 Waste-Heat Source Inventory

202 ENERGY MANAGEMENT HANDBOOKto look bad and be rejected. Because of the high stakesinvolved, the engineering costs of making the surveymay be substantial. Adequate instrumentation must be installedfor accurately metering flow streams. The acquireddata are used for designing waste-heat-recovery systems.The instruments are then used for monitoring system operationafter the installation has been completed. This isto ensure that the equipment is being operated correctlyand maintained in optimum condition so that full benefitswill be realized from the capital investment. Figure8.8 is a survey form used to gather information on eachindividual system or process unit.8.2.2 MeasurementsBecause waste-heat streams have such variability, itis difficult to list every possible measurement that mightbe required for its characterization. Generally speaking,the characterization of the quantity, quality, andtemporal availability of the waste energy requires thatvolumetric flow rate, temperature, and flow intervalsbe measured. Chapter 6 of the NBS Handbook 121 8 isdevoted to this topic. A few further generalizations aresufficient for planning the survey operation:1. Flow continuity requires that the mass flow rate ofany flow stream under steady-state conditions beconstant everywhere in the stream; that is,where Q is the material density, A the cross-sectionalflow area, and V the velocity of flow normal to thatarea. The equation can also be writtenρinQin = ρoutQout (8.9)where Q is the volumetric flow rate. It is safer, moreconvenient, and usually more accurate to measurelow-temperature flows than those at highertemperatures. Thus in many cases the volumetricflow rate of the cold flow and the temperature ofthe inlet and outlet flows are sufficient to infer thecharacteristics of the waste-heat stream.2. Fuel flows in direct-fired equipment are easily measuredwith volumetric rate meters. Combustionair and exhaust gas flows are at least an order ofmagnitude greater than the associated fuel flows.This effectively precludes the use of the volumetricmeter because of the expense. However ASMEorifice meters, using differential pressure cellsare often used for that purpose if the associatedpressure drop can be tolerated. It is even cheaper,although less convenient, to determine the volumetricproportions of the flue-gas constituents. Usingthis data, the air flow quantity and the flue gasflow rate can be calculated from the combustionequation and the law of conservation of mass.3. The total energy flux of the fluid streams can bedetermined from the volumetric flow rate and thetemperature using the equationH = ρQh (8.10)where H = enthalpy flux, Btu/hrρ = density, lb m /ft 3h = specific enthalpy, Btu/lb mQ = volumetric flow rate, ft 3 /hrThe density, the specific enthalpy, andmany other thermal and physical propertiesas a function of temperature andchemical species is given in tables andgraphs found in a number of engineeringhandbooks 9,10 and other specializedvolumes. 11-13Figure 8.84. In order to complete the wasteheatsurvey for the plant it is not necessaryto completely and permanentlyinstrument all systems of interest. Oneor more gas meters can be temporarilyinstalled and then moved to other locations.In fact, portable instruments canbe used for all measurements. Whileequipment monitoring, should becarried out with permanently installed

WASTE-HEAT RECOVERY 203instruments, compromises may be necessary to h' i = specific enthalpy of ith inflow or outflow, Btu/scfmethane. This assumption is often valid for high-Btu fuel gas. For moreh i = specific enthalpy of ith inflow or outflow, Btu/lb m precision and with other fuel gases, the exact composition should bekeep survey costs reasonable. However, permanentlyinstalled instruments should become a partW = net rate of mechanical or electric work beingtransferred to or from the system, Btu/hrof every related capital-improvement program.n = total number of inlet or outlet paths5. For steady-state operations a single temperaturecan be assigned to each outlet flow stream. Butfor a process with preprogrammed temperatureprofiles in time, an average over each cycle mustbe carefully determined. For a temperature-zoneddevice, averages of firing rate must be determinedcarefully over the several burners.penetrating system boundariesEquation 8.11 constitutes the theoretical basis andthe mathematical model of the heat-balance diagramshown in Figure 8.9. It corresponds to the data requirementsof the survey form shown in Figure 8.8. Usingthe data taken from that form, we compute the separateterms of the heat balance for a hypothetical furnace as8.2.3 Estimation without Measurementfollows:It is risky to base economic predictions used fordecisions concerning expensive waste-heat-recovery H f fuel energy rate = firing rate × HHV(8.12)systems on guesswork. However, when measurementsare not possible, it becomes necessary to rely on the bestapproximations available. The approximations must bewhere HHV is the higher heating value of the fuel nBtu/ft 3 or Btu/gal.made taking full advantage of all relevant data at hand.This should include equipment nameplate data; installation,H f .1 = 403.8 × 10 3 ft 3 /hr × 1030 Btu/ft 3operating, and maintenance literature; productionrecords; fuel and utility invoices; and equipment logs.= 415.9 × 10 6 Btu/hrThe energy auditor must attempt to form a consensusamong those most knowledgeable about the system orpiece of equipment. This can be done by personallyinterviewing engineers, managers, equipment operators,Because this is a dual-fuel installation, we can alsoconstruct a second heat-balance diagram for the alternativefuel:and maintenance crews. By providing iterative feedbackto these experts, a consensus can be developed. However,estimations are very risky. This author, after takingH f .2 = 3162.6 gph × 131,500 Btu/gal= 415.9 × 10 6 Btu/hrevery possible precaution, using all available data andfinding a plausible consensus among the experts, hasfound his economic projections to be as much as 100%more favorable than actual measurements proved. ThoseWriting the combustion equation on the basis of100 ft 3 of dry flue gas from the flue-gas analysis* fornatural gas (fuel no. 1):errors would have been avoided by accurate measurements.8.2.4 Constructing the Heat-Balance DiagramαCh 4 + BO 2 + γN 2 = 7.8CO 2 + 6.3O 2+ 0.5CO + 85.4N 2 + 2 αH 2 OThe first law of thermodynamics as applied to aBecause the chemical atomic species are conserved,steady flow-steady state system is conveniently writtenwe can solve for the relative quantities of air and fuel:nnQ = Σ m 1 h 1 + W = Σ Q 1 h 1 +i =11 (8.11) For carbon: α = 7.8 + 0.5 = 8.3whereFor nitrogen λ = 85.4Q = net rate of heat loss or gain, Btu/hr0.5 2 × 8.3For oxygen: β = 7.8 + 6.3 + —— + ——— = 22.65p i = density of ith inflow or outflow, lb/ft 32 2Q i = volumetric flow rate of ith inflow or outflow, scfh *For simplicity the assumption is made that this natural gas is puredetermined and used in the combustion equation.

204 ENERGY MANAGEMENT HANDBOOKFigure 8.9 Heat-balance diagram for reheat furnace(a) Natural gas.(b) No.2 fuel oil.A volume of air 22.65 + 85.4 ft 3 air— = —————— = —————— = 13.0 ———F volume of fuel 8.3 ft 3 gasH comb. air = A F × fuel firing rate × h' airh’ air at 100°F is found to be 1.2 Btu/ft 3 .H comb.air =13.0× 403.8 10 3 × 1.2=6.3× 10 6 Btu/hrThe specific enthalpy of each of the flue-gas componentsat 2200°F is found from Figure 8.10 to beh' CO2 = 69.3 Btu/scfh' O2 = 45.4 Btu/scfh' CO = 44.1 Btu/scfh' N2 = 43.4 Btu/scf= 54.7 Btu/scfh' H2OH stack gas = fuel firing rate Q co2 × h' co2+ Q O2 × h' O2 + Q CO × h' CO Q N × h' N2+ Q 2O × h' H2O= 403.8 × 10 3 7.8 × 69.3 +6.38.3 8.3 ×45.4+ 0.5 × 44.1 +85.4× 43.4 +2 × 8.3× 54.7 Btu/hr8.3 8.3 8.3= 265.8 × 10 6 Btu/hrBecause each fuel has its own chemical composition,the stack-gas composition will be different foreach fuel, as will the enthalpy flux: thus the calculationshould be repeated for each fuel.Flow path 1 represents the flow of product throughthe furnace.H prod. in = m prod C prod T prod. in= 50 tons × 2000 lb/ton × 0.115 Btu/lb • (100–60)°F= 0.5 × 10 6 Btu/hr

WASTE-HEAT RECOVERY 205Figure 8.10 Heat content vs. temperature.H prod. out = m prod C prod T prod. out= 50 tons × 2000 lb/ton× 0.180 Btu/lb • °F (2000 – 60) °F= 34.9 × 10 6 Btu/hrFlow path 2 is the cooling-water flow for the conveyor.H CW.in = gpm × 60 min/hr × 8.33 lb/gal× 1 Btu/lb • °F × T CW.in= 600 gpm × 60 min/hr × 8.33 lb/gal× 1 Btu/lb • °F (100 – 60) °F = 12.0 × 10 6 Btu/hrH CW. out = gpm × 60 min/hr × 8.33 lb/gal× 1 Btu/lb • °F × T CW.out= 600 gpm × 60 min/hr × 8.33 lb/gal× 1 Btu/lb • °F (191 – 60) °F = 39.3 × 10 6 Btu/hrThe heat losses are then estimated from equation8.11 asQ = H stack gas – H f .1 – H comb. air + H prod. out– H prod. in – H CW. out – H CW.in= (265.8 – 415.9 – 6.3 + 34.9 – 0.5 + 39.3– 12.0)10 6 = – 94.7 × 10 6 Btu/hrwhere Q includes not only the furnace surface losses butany unaccounted for enthalpy flux and all the inaccuraciesof measurement and calculation.Figure 8.9 shows the completed heat-balancediagrams for the reheat furnace. From that diagram weidentify three waste-heat streams, as listed in Table 8.5.8.2.5 Constructing Daily Waste-Heat Sourceand Load DiagramsThe normal daily operating schedule for the furnaceanalyzed in Section 8.2.4 is plotted in Figure 8.11. Itis shown that the present schedule shows furnace operationover two shifts daily, 6 days/week. It is necessaryto identify one or more potential loads for each of thetwo sources.

206 ENERGY MANAGEMENT HANDBOOKIf high-temperature exhaust stack streams fromdirect-fired furnaces are recuperated to heat the combustionair stream, then the load and source diagrams areidentical. This kind of perfect fit enhances the economicsof waste-heat recovery. In order to use the cooling-waterstream it will be necessary to find a potential load. Becauseof the temperature and the quantity of enthalpyflux available, the best fit may be the domestic hot-waterload. The hot water is used for wash facilities for the laborforce at break times and at the end of each shift. Thedaily load diagram for the domestic hot-water system isshown in Figure 8.12. Time coincidence for the load andsource is for two 10-minute periods prior to lunch andfor two 30-minute wash-up periods at the end of theshifts. Because the time coincidence between source andload is for only 1-1/2 hours in each 16 hours, waste-heatrecovery will require heat storage.8.2.6 Conceptual Design of theWaste-Heat-Recovery SystemPrior to equipment design and before a detailedeconomic analyses is performed, it is necessary to developone or more conceptual designs which can serve asa model for the future engineering work. This approachis illustrated by the analyses done in Sections 8.2.4 and8.2.5 for the two waste-heat streams. An excellent referencetext which is useful for the design of waste-heatrecovery systems is Hodge’s Analysis and Design of EnergySystems. 14Stack-Gas StreamClearly, recuperation is the most promising candidatefor heat recovery from high-temperature exhaustgas streams. In the application pictured in Figure 8.13the hot exhaust gases will be cooled by the incomingcombustion air. Because of the temperature of the gasesleaving the furnace, the heat exchanger to be selectedis a radiation recuperator. This is a concentric tubeheat exchanger which replaces the present stack. Theincoming combustion air is needed to cool the base ofthe recuperator and thus parallel flow occurs. Figure8.13 includes a sketch of the temperature profiles forthe two streams. It is seen that in the parallel flow exchanger,heat recovery ceases when the two streams approacha common exit temperature. For a well-insulatedrecuperator the conservation of energy is expressed bythe equationQ stack gas (h' stack gas, in – h' stack gas, out )= Q comb. air (h' comb. air, out – h' comb.air, in )(8.14)Both the right- and left-hand terms represent theheat-recovery rate as well as the decrease in fuel energyrequired. If the burners or associated equipment havemaximum temperature limitations, those temperaturesFigure 8.11 Furnace operating schedule.Figure 8.12 Daily domestic water load.Table 8.5 Waste-Heat Streams from Reheat Furnace Fired with Natural GasExhaust Stack: Cooling Product:Composition Combustion Products Water Steel CastingsTemperature (°F) 2200 191 2000Flow rate 57,000,000 cfh 36,000 gph 50 tons/hrEnthalpy rate (Btu/hr) 265,800,000 39,300,000 34,900,000Percent fuel 64 9 8energy rate

WASTE-HEAT RECOVERY 207different. The annual heat recovery for each fuel wasassumed to be proportional to the total consumption ofeach fuel, so that the heat recovered was found asheat recovered=1.39× 10 9 ft 3 gasyr× 24.6 – 1.2 Btuft 3× 12.95 ft3 airft 3 gas+4.43× 10 6 galyr× 2040.8 ft3 air× 24.6 – 1.2Btugal oil ft 3=4.2× 10 11 +2.1× 10 11 =6.3× 10 11 Btu/yrFigure 8.13 Temperature distribution in recuperator.also become the high limits for the combustion air,which in turn fixes the maximum allowable enthalpyfor the combustion air. Thus the maximum rate of heatrecovery is also fixed. Otherwise, the final temperaturesof the two stream are based on an optimization of theeconomic opportunity. This is so because increased heatrecovery implies increased recuperator area and thusincreased cost. It may also imply higher combustionair temperatures with the resultant increase in fan operatingcosts and additional investment costs in hightemperatureburners, combustion air ducts, and largerfans. In this case we assume that a 100°F temperaturedifference will occur between the preheated combustionair and the stack gases leaving the recuperator. Equation8.14 cannot be used directly because the volume ratesof fuel and air required are reduced with recuperation.However, if the air/fuel ratio is maintained constant,then Q stack/gas /Q comb.air remains almost constant; thenequation 8.14 can be writtenQ stack gash' comb. air, out – h' comb. air, in———— = const. = ———————————— (8.15)Q comb. airh' stack gas, in – h' stack gas, outThis equation can be solved with the help of data fromFigure 8.9 and the temperature relationshipT stack gas, out = T comb. air, out + 100°F (8.16)Separate solutions are required for the primary andalternative fuels. The solutions are found by iteratingequations 8.15 and 8.16 through a range of temperatures.The value of preheat temperature found for natural gasis 1260°F. For fuel oil the temperature is only slightlyThis is an energy savings of6.3 × 10 11—————————————————— = 0.31 or 31%1.39 × 10 9 × 1030 + 4.43 × 10 6 × 131,500The predicted cost savings is$1,308,900 for natural gas950,200 for No. 2 fuel oil——————————————$2,259,100 totalThe complete retrofit installation is estimated tocost less than $2,000,000, and the payback period is lessthan one year. Two points must be emphasized. Theentire retrofit installation must be well engineered as asystem. This includes the recuperator itself as well asmodifications and/or replacement of burners and fans,and the system controls. Only then can the projectedsystem life span be attained and the capital paybackactually realized. The cost of lost product must also befactored into the economic analysis if the installationis planned at a time that will cause a plant shutdown.Economics may dictate a delay for the retrofit until thenext scheduled or forced maintenance shutdown.8.3 WASTE-HEAT EXCHANGERS8.3.1 Transient Storage DevicesThe earliest waste-heat-recovery devices were “ regenerators.”These consisted of extensive brick work,called “checkerwork,” located in the exhaust flues andinlet air flues of high-temperature furnaces in the steelindustry. Regenerators are still used to a limited extentin open hearth furnaces and other high-temperature

208 ENERGY MANAGEMENT HANDBOOKfurnaces burning low-grade fuels. It is impossible toachieve steel melt temperature unless regenerators areused to boost the inlet air temperature. In the processvast amounts of waste heat are recovered which wouldotherwise be supplied by expensive high-Btu fuels. Pairsof regenerators are used alternately to store waste heatfrom the furnace exhaust gases and then give back thatheat to the inlet combustion air. The transfer of exhaustgasand combustion-air streams from one regenerator tothe other is accomplished by using a four-way flappervalve. The design of and estimates of the performanceof recuperators follows the principles presented in Section8.1.7. One disadvantage of this mode of operationis that heat-exchanger effectiveness is maximum only atthe beginning of each heating and cooling cycle and fallsto almost zero at the end of the cycle. A second disadvantageis that the tremendous mass of the checkerworkand the volume required for its installation raises capitalcosts above that for the continuous-type air preheaters.An alternative to the checkerwork regenerator isthe heat wheel. This device consists of a permeable flatdisk which is placed with its axis parallel to a pair ofsplit ducts and is slowly rotated on an axis parallel tothe ducts. The wheel is slowly rotated as it interceptsthe gas streams flowing concurrently through the splitducts. Figure 8.14 illustrates those operational features.As the exhaust-gas stream in the exhaust duct passesthrough one-half of the disk it gives up some of itsheat which is temporarily stored in the disc material. Asthe disc is turned, the cold incoming air passes throughthe heated surfaces of the disk and absorbs the energy.The materials used for the disks include metal alloys,ceramics and fiber, depending upon the temperature ofthe exhaust gases. Heat-exchanger efficiency for the heatwheel has been measured as high as 90% based uponthe exhaust stream energy. Further details concerningthe heat wheel and its applications are given in Section8.4.3.8.3.2 Steady-State Heat ExchangersSection 8.4 treats heat exchangers in some detail.However, several important criteria for selection arelisted below.1. Flow Arrangements. These are characterized as:Parallel flowCounterflowCrossflowMixed flowThe flow arrangement helps to determine theoverall effectiveness, the cost, and the highestachievable temperature in the heated stream. TheSection of glassceramic wheelFanInsulationStable thermal wallFigure 8.14 Heat wheel.Incinerated airGrid burnerPreheated vapor-laden airCeramic wheelVapor-laden airlatter effect most often dictates the choice of flowarrangement. Figure 8.15 indicates the temperatureprofiles for the heating and heated streams, respectively.If the waste-heat stream is to be cooledbelow the cold stream exit, a counterflow heatexchanger must be used.2. Character of the Exchange Fluids. It is necessaryto specify the heated and cooled fluids asto:Chemical compositionPhysical phase (i.e., gaseous, liquid, solid, ormultiphase)Change of phase, if any, such as evaporatingor condensingThese specifications may affect the optimum flowarrangement and/or the materials of construction.8.3.3 Heat-Exchanger EffectivenessThe effectiveness of a heat exchanger is defined asa ratio of the actual heat transferred to the maximumpossible heat transfer considering the temperatures oftwo streams entering the heat exchanger. For a givenflow arrangement, the effectiveness of a heat exchangeris directly proportional to the surface area that separatesthe heated and cooled fluids. The effectiveness of typicalheat exchangers is given in Figure 8.16 in terms of theparameter AU/C min where A is the effective heat-transferarea, U the effective overall heat conductance, and C min

WASTE-HEAT RECOVERY 209the mass flow rate times the specific heat of the fluidwith minimum mc . The conductance is the heat rate perunit area per unit temperature difference. Note that asAU/C min increases, a linear relation exists with the effectivenessuntil the value of AU/C min approaches 1.0. Atthis point the curve begins to knee over and the increasein effectiveness with AU is drastically reduced. Thus onesees a relatively early onset of the law of diminishingreturns for heat-exchanger design. It is implied that onepays heavily for exchangers with high effectiveness.8.3.4 Filtering or FoulingOne of the important heat-exchanger parametersrelated to surface conditions is termed the fouling factor.The fouling of the surfaces can occur because of filmdeposits, such as oil films; because of surface scalingdue to the precipitation of solid compounds from solution;because of corrosion of the surfaces; or becauseof the deposit of solids or liquids from two-phase flowstreams. The fouling factor increases with increased foulingand causes a drop in heat exchanger effectiveness. Ifheavy fouling is anticipated, it may call for the filteringof contaminated streams, special materials of construction,or a mechanical design that permits easy access tosurfaces for frequent cleaning.8.3.5 Materials and ConstructionThese topics have been reviewed in previous sections.In summary:1. High temperatures may require the use of specialmaterials.2. The chemical and physical properties of exchangefluids may require the use of special materials.3. Contaminated fluids may require special materialsand/or special construction.4. The additions of tube fins on the outside, groovedsurfaces or swaged fins on the inside, and treatedor coated surfaces inside or outside may be requiredto achieve compactness or unusually higheffectiveness.8.3.6 Corrosion ControlThe standard material of construction for heatexchangers is mild steel. Heat exchangers made of steelare the cheapest to buy because the material is the leastexpensive of all construction materials and because it isso easy to fabricate. However, when the heat transfermedia are corrosive liquids and/or gases, more exoticmaterials may have to be used. Corrosion tables 15 givethe information necessary to estimate the life of theheat exchanger and life-cycle-costing studies allowFigure 8.15 Cross-flow heat exchanger.

210 ENERGY MANAGEMENT HANDBOOKvalid comparisons of the costs of owning the steel heatexchanger versus one constructed of exotic materials.The problem is whether it will be cheaper to replacethe steel heat exchanger at more frequent intervals orto buy a unit made of more expensive materials, butrequiring less frequent replacement. Mechanical designswhich permit easy tube replacement lower thecost of rebuilding and favor the use of mild steel heatexchangers.Corrosion-resisting coatings, such as the TFE plastics,are used to withstand extremely aggressive liquidsand gases. However, the high cost of coating and thedanger of damaging the coatings during assembly andduring subsequent operation limit their use. One disadvantageof using coatings is that they almost invariablydecrease the overall conductance of the tube walls andthus necessitate an increase in size of the heat exchanger.The decision to use coatings depends first upon theavailability of alternate materials to withstand the corrosionas well as the comparative life-cycle costs, assumingthat alternative materials can be found.Among the most corrosive and widely used materialsflowing in heat exchangers are the chlorides suchas hydrochloric acid and saltwater. Steel and most steelalloys have extremely short lives in such service. Oneclass of steel alloys that have shown remarkable resistanceto chlorides and other corrosive chemicals is calledduplex steels 16 and consists of half-and-half ferrite andaustenitic microstructures. Because of their high tensilestrength, thinner tube walls can be used and this offsetssome of the higher cost of the material.Figure 8.16 Typical heat-exchanger effectiveness.

WASTE-HEAT RECOVERY 2118.3.7 MaintainabilityProvisions for gaining access to the internalsmay be worth the additional cost so thatsurfaces may be easily cleaned, or tubes replacedwhen corroded. A shell and tube heatexchanger with flanged and bolted end capswhich are easily removed for maintenance isshown in Figure 8.17. Economizers are availablewith removable panels and multipleone-piece finned, serpentine tube elements,which are connected to the headers withstandard compression fittings. The tubes canbe removed and replaced on site, in a matterof minutes, using only a crescent wrench.8.4 COMMERCIAL OPTIONS INWASTE-HEAT-RECOVERY EQUIPMENT8.4.1 IntroductionIt is necessary to completely specify all of the operatingparameters as well as the heat exchange capacityfor the proper design of a heat exchanger, or for theselection of an off-the-shelf item. These specificationswill determine the construction parameters and thus thecost of the heat exchanger. The final design will be acompromise among pressure drop (which fixes pump orfan capital and operating costs), maintainability (whichstrongly affects maintenance costs), heat exchanger effectiveness,and life-cycle cost. Additional features, suchas the on-site use of exotic materials or special designsfor enhanced maintainability, may add to the initialcost. That design will balance the costs of operation andmaintenance with the fixed costs in order to minimizethe life-cycle costs. Advice on selection and design ofheat exchangers is available from manufacturers andfrom T.E.M.A.* Industrial Heat Exchangers (17) is an excellentguide to heat exchanger selection and includes adirectory of heat exchanger manufacturers.The essential parameters that should be knownand specified in order to make an optimum choice ofwaste-heat recovery devices are:Temperature of waste-heat fluidFlow rate of waste-heat fluidChemical composition of waste-heat fluidMinimum allowable temperature of waste-heatfluidAmount and type of contaminants in the wasteheatfluid*Tubular Equipment Manufacturers Association, New York, NYFigure 8.17. Shell and tube heat exchanger.Allowable pressure drop for the waste-heat fluidTemperature of heated fluidChemical composition of heated fluidMaximum allowable temperature of heated fluidAllowable pressure drop in the heated fluidControl temperature, if control requiredIn the remainder of this section, some commontypes of commercially available waste-heat recoverydevices are discussed in detail.8.4.2 Gas-to-Gas Heat Exchangers: RecuperatorsRecuperators are used in recovering waste heat tobe used for heating gases in the medium- to high-temperaturerange. Some typical applications are soakingovens, annealing ovens, melting furnaces, reheat furnaces,afterburners, incinerators, and radiant-heat burners.The simplest configuration for a heat exchanger isthe metallic radiation recuperator, which consists of twoconcentric lengths of metal tubing, as shown in Figure8.18. This is most often used to extract waste heat fromthe exhaust gases of a high-temperature furnace forheating the combustion air for the same furnace. The assemblyis often designed to replace the exhaust stack.The inner tube carries the hot exhaust gases whilethe external annulus carries the combustion air from theatmosphere to the air inlets of the furnace burners. Thehot gases are cooled by the incoming combustion air,which then carries additional energy into the combustionchamber. This is energy that does not have to besupplied by the fuel; consequently, less fuel is burnedfor a given furnace loading. The saving in fuel alsomeans a decrease in combustion air, and therefore stacklosses are decreased not only by lowering the stack exit

212 ENERGY MANAGEMENT HANDBOOKFigure 8.18 Metallic radiation recuperator.temperatures, but also by discharging smaller quantitiesof exhaust gas. This particular recuperator gets itsname from the fact that a substantial portion of the heattransfer from the hot exhaust gases to the surface of theinner tube takes place by radiative heat transfer. Thecold air in the annulus, however, is almost transparent toinfrared radiation so that only convection heat transfertakes place to the incoming combustion air. As shownin the diagram, the two gas flows are usually parallel,although the configuration would be simpler and theheat transfer more efficient if counterflow were used.The reason for the use of parallel flow is that the coldair often serves the function of cooling the hottest partof the exhaust duct and consequently extends its servicelife.The inner tube is often fabricated from high-temperaturematerials such as high-nickel stainless steels.The large temperature differential at the inlet causesdifferential expansion, since the outer shell is usually ofa different and less expensive material. The mechanicaldesign must take this effect into account. More elaboratedesigns of radiation recuperators incorporate two sections;the bottom operating in parallel flow, and the uppersection using the more efficient counterflow arrangement.Because of the large axial expansions experiencedand the difficult stress conditions that can occur at thebottom of the recuperator, the unit is often supported atthe top by a freestanding support frame and the bottomis joined to the furnace by way of an expansion joint.A second common form for recuperators is calledthe tube-type or convective recuperator. As seen in theschematic diagram of a combined radiation and convectivetype recuperator in Figure 8.19, the hot gases are carriedthrough a number of small-diameter parallel tubes, whilethe combustion air enters a shell surrounding the tubesand is heated as it passes over the outside of the tubesone or more times in directions normal to the tubes. If thetubes are baffled as shown so as to allow the air to passover them twice, the heat exchanger is termed a two-passconvective recuperator; if two baffles are used, a three-passrecuperator; and so on. Although baffling increases the costof manufacture and also the pressure drop in the air path, italso increases the effectiveness of heat exchange. Tube-typerecuperators are generally more compact and have a highereffectiveness than do radiation recuperators, because of thelarger effective heat-transfer area made possible throughthe use of multiple tubes and multiple passes of the air. Formaximum effectiveness of heat transfer, combinations ofthe two types of recuperators are used, with the convectiontype always following the high-temperature radiationrecuperator.The principal limitation on the heat recovery possiblewith metal recuperators is the reduced life of the liner atinlet temperatures exceeding 2000°F. This limitation forcesthe use of parallel flow to protect the bottom of the liner.The temperature problem is compounded when furnacecombustion air flow is reduced as the furnace loading isreduced. Thus the cooling of the inner shell is reduced andthe resulting temperature rise causes rapid surface deterioration.To counteract this effect, it is necessary to providean ambient air bypass to reduce the temperature of theexhaust gases. The destruction of a radiation recuperatorby overheating is a costly accident. Costs for rebuildingone are about 90% of the cost of a new unit.To overcome the temperature limitations of metalrecuperators, ceramic-tube recuperators have been developedwhose materials permit operation to temperatures of2800°F and on the preheated air side to 2200°F, althoughpractical designs yield air temperatures of 1800°F. Earlyceramic recuperators were built of tile and joined withfurnace cement. Thermal cycling caused cracking of thejoints and early deterioration of the units. Leakage ratesas high as 60% were common after short service periods.Later developments featured silicon carbide tubes joined byflexible seals in the air headers. This kind of design, illustratedin Figure 8.20, maintains the seals at a relatively lowtemperature and the life of seals has been much improved,as evidenced by leakage rates of only a few percent aftertwo years of service.An alternative design for the convective recuperatoris one in which the cold combustion air is heated in a

WASTE-HEAT RECOVERY 213bank of parallel tubes extending into the flue-gasstream normal to the axis of flow. This arrangementis shown in Figure 8.21. The advantages ofthis configuration are compactness and the easeof replacing individual units. This can be doneduring full-load operation and minimizes thecost, inconvenience, and possible furnace damagedue to a forced shutdown from recuperatorfailure.Recuperators are relatively inexpensiveand they do reduce fuel consumption. However,their use may require extensive capitalimprovements. Higher combustion air temperaturesmay require:• burner replacement• larger-diameter air lines with flexible expansionfittings• cold-air piping for cooling high-temperatureburners• modified combustion controls• stack dampers• cold air bleeds• recuperator protection systems• larger combustion air fans to overcomethe additional pressure drops in thesystem.8.4.3 Heat WheelsA rotary regenerator, also called an airFigure 8.19. Combined radiation and convective recuperator.preheater or a heat wheel, is used for lowtomoderately high-temperature waste-heatrecovery. Typical applications are for spaceheating, curing, drying ovens and heat-treatfurnaces. Originally developed as an air preheaterfor utility steam boilers, it was lateradapted, in small sizes, as a regenerator forautomotive turbine applications. It has beenused for temperatures ranging from 68°F to2500°F.Figure 8.22 illustrates the operation ofa heat wheel in an air conditioning application.It consists of a porous disk, fabricatedof material having a substantial specific heat.The disk is driven to rotates between two sideby-sideducts. One is a cold-gas duct and theFigure 8.20. Silicon-carbide-tube ceramic recuperator.other is a hot-gas duct. Although the diagramshows a counterflow configuration, parallel flow can the area of the cold duct, the heat is transferred fromalso be used. The axis of the disk is located parallel to the disk to the cold air. The overall efficiency of heatand on the plane of the partition between the ducts. transfer (including latent heat) can be as high as 90%.As the disk slowly rotates, sensible heat (and in some Heat wheels have been built as large as 70 ft incases, moisture-containing latent heat) is transferred to diameter with air capacities to 40,000 cfm. Multiplethe disk by the hot exhaust gas. As the disk moves into units can be used in parallel. This modular approach

214 ENERGY MANAGEMENT HANDBOOKFigure 8.21 Parallel-tube recuperator.may be used to overcome a mismatch between capacityrequirements and the limited number of sizes availablein commercial units.The limitations on the high-temperature range forthe heat wheel are primarily due to mechanical difficultiesintroduced by uneven thermal expansion of therotating wheel. Uneven expansion can cause excessivedeformations of the wheel that result in the loss of adequategas seals between the ducts and the wheel. Thedeformation can also result in damage due to the wheelrubbing against its retaining enclosure.Heat wheels are available in at least four types:1) A metal frame packed with a core of knitted meshstainless steel, brass, or aluminum wire, 2) A so-calledlaminar wheel fabricated from corrugated materialswhich form many small diameter parallel-flow passages,3) A laminar wheel constructed from a high-temperatureceramic honeycomb, and 4) A laminar wheel constructedof a fibrous material coated with a hygroscopic so thatlatent heat can be recovered.Most gases contain some water vapor since it isa natural component of air and it is also a product ofhydrocarbon combustion. Water vapor, as a componentof a gas mixture, carries with it its latent heat ofevaporation. This latent heat may be a substantial partof the energy contained within the exit-gas streams fromair-conditioned spaces or from industrial processes. Torecover some of the latent heat in the gas stream, using aheat wheel, the sheet must be coated with a hygroscopicmaterial such as lithium chloride (LiCl) which readilyabsorbs water vapor to form a hydrate, which in thecase of lithium chloride is the hydrate LiCl•H 2 O. Thehydrate consists of one mole of lithium chloride chemicallycombined with one mole of water vapor. Thus theweight ratio of water to lithium-chloride is 3:7. In a hygroscopicheat wheel, the hot gas stream gives up someFigure 8.22 Rotary regenerator (heat wheel).part of its water vapor to the lithium-chloride coating;the gases to be heated are dry and absorb some of thewater held in the hydrate. The latent heat in that watervapor adds directly to the total quantity of recoveredheat. The efficiency of recovery of the water vapor inthe exit stream may be as high as 50%.Because the pores or passages of heat wheels carrysmall amounts of gas from the exhaust duct to the intakeduct, cross-contamination of the intake gas can occur. Ifthe contamination is undesirable, the carryover of exhaustgas can be partially eliminated by the addition ofa purge section located between the intake and exhaustducts, as shown in Figure 8.23. The purge section allowsthe passages in the wheel to be cleared of the exhaustgases by introducing clean air which discharges thecontaminant to the atmosphere. Note that additional gasseals are required to separate the purge ducts from theintake and exhaust ducts and consequently add to thecost of the heat wheel.Common practice is to use six air changes of cleanair for purging. This results in a reduction of cross-contaminationto a value as little as 0.04% for the gas and0.2% for particulates in laminar wheels, and less than1.0% total contaminants in packed wheels.If the heated gas temperatures are to be heldconstant, regardless of heating loads and exhaust gastemperatures, the heat wheel must be driven at variablespeed. This requires a variable-speed drive with a speedcontrollerwith an air temperature sensor as the controlelement. When operating with outside air in periods ofsub-zero temperatures and high humidity, heat wheelsmay frost up requiring the protection of an air-preheatsystem. When handling gases containing water-soluble,greasy, or large concentrations of particulates, air filtersmay be required in the exhaust system upstream fromthe heat wheel. These features, however, add to the

WASTE-HEAT RECOVERY 215Figure 8.23Heat wheel with purge section.Figure 8.24 Passive gas to gasregenerator.complexity and the cost of owning and operating thesystem.Contaminant buildup on ceramic heat wheels canoften be removed by raising the temperature of theexhaust stream to exceed the ignition temperature ofthe contaminant. However, heat wheels are inherentlyself-cleaning, because materials entering the wheel fromthe hot-gas stream tend to be swept out by the reverseflow of the cold-gas stream.8.4.4 Passive Air PreheatersPassive gas-to-gas regenerators are available forapplications where cross-contamination cannot be tolerated.One such type of regenerator, the plate-type, isshown in Figure 8.24. A second type, the heat pipe arrayis shown in Figure 8.25. Passive air preheaters areused in the low- and medium-temperature applications.Those include drying, curing, and baking ovens; air preheatersin steam boilers; air dryers; waste heat recoveryfrom exhaust steam; secondary recovery from refractorykilns and reverbatory furnaces; and waste heat recoveryfrom conditioned air.The plate-type regenerator is constructed of alter-nate channels which separate adjacent flows of heatedand heating gases by a thin wall of conducting metal.Although their use eliminates cross-contamination, theyare bulkier, heavier, and more expensive than a heatwheel of similar heat-recovery and flow capacities. Furthermore,it is difficult to achieve temperature controlof the heated gas, while fouling may be a more seriousproblem.The heat pipe is a heat-transfer element that isassembled into arrays which are used as compact andefficient passive gas-to-gas heat exchangers. Figure 8.25shows how the bundle of finned heat pipes extendthrough the wall separating the inlet and exhaust ductsin a pattern that resembles the conventional finned tubeheat exchangers. Each of the separate pipes, however,is a separate sealed element. Each consists of an annularwick on the inside of the full length of the tube,in which an appropriate heat-transfer fluid is absorbed.Figure 8.26 shows how the heat transferred from thehot exhaust gases evaporates the fluid in the wick.This causes the vapor to expand into the center core ofthe heat pipe. The latent heat of evaporation is carriedwith the vapor to the cold end of the tube. There it is

216 ENERGY MANAGEMENT HANDBOOKFigure 8.25 Heat pipe.Figure 8.26 Heat pipe operation.pressure drop is controlled by the spacing of the tubesand the number of rows of tubes. Econ omizers are availableboth prepackaged i n modular sizes and designedand fabricated to custom specifications from standardcomponents. Materials for the tubes and fins can beselected to withstand corrosive liquids and/or exhaustgases.Temperature control of the boiler feedwater is necessaryto prevent boiling in the economizer during lowremovedby transferral to the cold gas as the vapor isrecondensed. The condensate is then carried back inthe wick to the hot end of the tube. This takes place bycapillary action and by gravitational forces if the axis ofthe tube is tilted from the horizontal. At the hot end ofthe tube the fluid is then recycled.The heat pipe is compact and efficient for tworeasons. The finned-tube bundle is inherently a goodconfiguration for convective heat transfer between thegases and the outside of the tubes in both ducts. Theevaporative-condensing cycle within the heat tubes is ahighly efficient method of transferring heat internally.This design is also free of cross-contamination. However,the temperature range over which waste heat can be recoveredis severely limited by the thermal and physicalproperties of the fluids used within the heat pi pes. Table8.6 lists some of the transfer fluids and the temperatureranges in which they are applicable.8.4.5 Gas or Liquid-to-Liquid Regenerators:The Boiler EconomizerThe economizer is ordinar ily constructed as a bundleof finned tubes, installed in the boiler’s breeching.Boiler feedwater flows through the tubes to be heated bythe hot exhaust gases. Such an arrangement is shown inFigure 8.27. The tubes are usually connected in a seriesarrangement, but can also be arranged in series-parallelto control the liquid-side pressure drop. The air-sideTable 8.6 Temperature Ranges for Heat-Transfer FluidsUsed in Heat PipesTemperatureFluid Range (°F) Compatible MetalsNitrogen – 300 to – 110 Stainless steelAmmonia – 95 to + 140 Nickel, aluminum,stainless steelMethanol – 50 to + 240 Nickel, copper,stainless steelWater 40 to 425 Nickel, copperMercury 375 to 1000 Stainless steelSodium 950 to 1600 Nickel, stainless steelLithium 1600 to 2700 Alloy of niobium andzirconiumSilver 2700 to 3600 Alloy of tantalumand tungsten

WASTE-HEAT RECOVERY 217Figure 8.28 Fuel savings from a gas-fired boiler usingeconomizer.Figure 8.27 Boiler economizer.steam demand or in case of a feedwater pump failure.This is usually obtained by controlling the amount ofexhaust gases flowing through the economizer using adamper, which diverts a portion of the gas flow througha bypass duct.The extent of heat recovery in the economizermay be limited by the lowest allowable exhaust gastemperature in the exhaust stack. The exhaust gasescontain water vapor both from the combustion air andfrom the combustion of the hydrogen that is containedin the fuel. If the exhaust gases are cooled below the dewpoint of the water vapor, condensation will occur andcause damage to the structural materials. If the fuel alsocontains sulfur, the sulfur-dioxide will be absorbed bythe condensed water to form sulfuric acid. This is verycorrosive and will attack the breeching downstream ofthe economizer and the stack lines. The dew point ofthe exhaust gases from a natural-gas-fired boiler variesfrom approximately 138°F for a stoichiometric fuel/airmixture, to 113°F for 100% excess air. Because heat-transmissionlosses through the stack cause axial temperaturegradients from 0.2 to 2°F/ft, and because the stack linermay exist at a temperature 50 to 75°F lower than thegas bulk temperature, it is considered prudent to limitminimum stack temperatures to 300°F, or no lower than250°F when burning natural gas. When using the fuelscontaining sulfur, even greater caution is taken. Thismeans that the effectiveness of an economizer is limitedunless the exhaust gases from the boiler are relativelyhot. Figure 8.28 is a graph of the percent fuel savedplotted against percent excess air for a number of stackgas temperatures using natural gas as a boiler fuel. Theplots are based on a 300°F hot-gas temperature leavingthe economizer.8.4.6 Shell-and-Tube or Concentric-TubeHeat ExchangersShell-and-tube and concentric-tube heat exchangersare used to recover heat in the low and mediumrange from process liquids, coolants, and condensatesof all kinds for heating liquids.When the medium containing waste heat is eithera liquid or a vapor that heats a liquid at a differentpressure, a totally exclosed heat exchanger must beused. The two fluid streams must be separated so as tocontain their respective pressures. In the shell-and-tubeheat exchanger, the shell is a cylinder that contains thetube bundle. Internal baffles may be used to direct thefluid in the shell over the tubes in multiple passes. Becausethe shell is inherently weaker than the tubes, thehigher-pressure fluid is usually circulated in the tubeswhile the lower-pressure fluid circulates in the shell.However, when the heating fluid is a condensing vapor,it is almost invariably contained within the shell. If thereverse were attempted, the condensation of the vaporwithin the small-diameter parallel tubes would causeflow instabilities. Shell and tube heat exchangers areproduced in a wide range of standard sizes with manycombinations of materials for the tubes and the shells.The overall conductance of these heat exchangers rangeto a maximum of several hundred Btu/hr ft 2 °F.A concentric-tube exchanger is used when the fluidpressures are so high that a shell design is uneconomical,or when ease of dissembly is paramount. The hotter fluidis almost invariably contained in the inner tube to mini-

218 ENERGY MANAGEMENT HANDBOOKmize surface heat losses. The concentric-tube exchangermay consist of a single straight length, a spiral coil, or abundle of concentric tubes with hairpin bends.Shell-and-tube and concentric-tube heat exchangersare used to recover heat in the low and mediumrange from process liquids, coolants, and condensatesof all kinds for heating liquids.8.4.7 Waste-Heat BoilersWaste-heat boilers are water tube boilers in whichhot exhaust gases are used to generate steam. The exhaustgases may be from a gas turbine, an incinerator,a diesel engine, or any other source of medium- tohigh-temperature waste heat. Figure 8.29 shows a conventional,two-pass waste-heat boiler. When the heatsource is in the medium-temperature range, the boilertends to become bulky. The use of finned tubes extendsthe heat transfer areas and allows a more compact size.If the quantity of waste heat is insufficient for generatinga needed quantity of steam, it is possible to addauxiliary burners to the boiler or an afterburner tothe ducting upstream of the boiler. The conventionalwaste-heat boiler cannot generate super-heated steamso that an external superheater is required if superheatis needed.A more recently designed waste-heat boiler utilizesa finned-tube bundle for the evaporator, an externaldrum, and forced recirculation of the feedwater. Thedesign, which is modular, makes for a compact unitwith high boiler efficiency. Additional tube bundles canbe added for superheating the steam and for preheatingthe feedwater. The degree of superheat which can beachieved is limited by the waste-heat temperature. Thesalient features of the boiler are shown on the schematicdiagram in Figure 8.30.Waste-heat boilers are commercially available incapacities from less than 1000 up to 1 million cfm ofexhaust gas intake.are no different from those used for the analysis of anyother industrial capital project. These techniques arethoroughly discussed in Chapter 4 of this volume. 18The economic potential for this class of systems is oftenlimited by factors that are crucial yet overlooked. Althoughthe capital cost of these systems is proportionalto the peak rate of heat recovery, the capital recoverydepends principally on the annual fuel savings. Thesesavings depend on a number of factors, such as the timedistribution of waste-heat source availability, the timedistribution of heat-load availability, the availability ofwaste-heat-recovery equipment that can perform at thespecified thermal conditions, and the current and futureutility rates and prices of fuel. The inability to accuratelypredict these factors can make the normal investmentdecision-making process ineffectual.There is another important distinction to be madeabout waste-heat recovery investment. When capitalprojects involve production-related equipment, the rate8.4.8 Input-Output Matrix forWaste-Heat-Recovery DevicesTable 8.8 presents the significant attributes of themost common types of industrial heat exchangers. Thismatrix is useful in making selections from competingtypes of heat exchangers for waste-heat recovery.8.5 ECONOMICS OF WASTE-HEAT RECOVERY8.5.1 GeneralEconomic analysis techniques used for analyzinginvestment potential for waste-heat-recovery systemsFigure 8.29 Two-pass waste-heat boiler.

WASTE-HEAT RECOVERY 219NewD.A. TankSteam DrumFigure 8.30 A Recirculation waste-heatboiler.Table 8.8 Operation and Application Characteristics of Industrial Heat ExchangersCommercialHeat-TransferEquipmentSpecificationsfor WasteRecoveryUnitLow temperature:subzero –250°FIntermediate temp:250-1200°FHigh temperature:1200-2000°FRecovers moistureLarge temperaturedifferentials permittedPackaged unitsavailableCan be retrofitNo crosscontaminationCompact sizeGas-to-gasheat exchangeGas-to-liquidheat exchangerLiquid-to-liquidheat exchangerCorrosive gasespermitted withspecial constructionRadiation recuperator × × a × × × ×Convection recuperator × × × × × × × ×Metallic heat wheel × × b × × c × × ×Hygroscopic heat wheel × × × × c × ×Ceramic heat wheel × × × × × × × ×Passive regenerator × × × × × × × ×Finned-tube heat exchanger × × × × × × × × dTube shell-and-tube exchanger × × × × × × × × ×Waste-heat boilers × × × × × × × dHeat pipes × × × e × × × × × ×a Off-the-shelf items available in small capacities only.b Controversial subject. Some authorities claim moisture recovery. Do not advise depending on it.c With a purge section added, cross-contamination can be limited to less than 1% by mass.d Can be constructed of corrosion-resistant materials, but consider possible extensive damage to equipment caused by leaks or tube ruptures.e Allowable temperatures and temperature differential limited by the phase-equilibrium properties of the internal fluid.of capital recovery can be adjusted by manipulatingproduct selling prices. When dealing with waste-heatrecoverysystems, this is generally not an option. Therate of capital recovery is heavily influenced by utilityrates and fuel prices which are beyond the influence ofthe investor. This points out the importance of gatheringsolid data, and using the best available predictions inpreparing analyses for the investment decision process.8.5.2 Effect of Utility Rates and Fuel PricesGood investment potential exists in an economicclimate where energy costs are high relative to equipmentcosts. However, the costs of goods are sharplyinfluenced by energy costs. This would seem to indicatethat the investment potential for waste-heat-recoveryprojects is not going to improve any faster than energycostescalations occur. In one sense this is true of future

220 ENERGY MANAGEMENT HANDBOOKprojects. But because capital projects involve one-timeexpenditures, which are usually financed by fixed-rateloans, the worth of a present investment will benefitfrom the rising costs of energy.8.5.3 Effect of Load and Use FactorsThe load factor is defined as the ratio of averageannual load to rated capacity and the use factor as thefractional part of a year that the equipment is in use. Itis clear that the capital recovery rate is directly proportionalto these factors.8.5.4 Effects of Reduced System LifeWaste-heat recovery equipment is susceptible todamage from natural and human-made environmentalconditions. Damage can result from overheating,freezing, corrosion, collision, erosion, and explosion.Furthermore, capital recovery can never be completedif the equipment fails to achieve its expected life. Onemust either factor the risks of equipment damage intothe economic analysis, or insist that sufficient provisionfor equipment safety be engineered into the systems.References1. E. Cook, “Energy Flow in Industrial Societies,” Scientific America,Vol. 225 (Sept. 1971), p. 135.2. M.H. Ross and R.H. Williams, “The Potential for Fuel Conservation.”Technology Review, Vol. 79. No. 4 (1977), p. 43.3. R.W. Sant, ‘’The Least Cost Energy Strategy: Minimizing ConsumerCosts through Competition,” Report of Energy ProductivityCenter, CMU, 1979, pp. 35-36.4. Annual Energy Review, 1991, U.S. Energy Information Administration,Washington, D.C.5. W.M. Rohsenow and J.P. Hartnett, Eds. Handbook of Heat Transfer,McGraw-Hill, New York, 1978, pp. 18-73.6. B.C. Langley, Heat Pump Technology, 2nd Ed., Prentice-Hall, EnglewoodCliffs, N.J., 1989.7. K. Kreider and M. McNeil, Eds., Waste Heat Management Guidebook,NBS Handbook 121, U.S. Government Printing Office,Washington, D.C., 1976, pp. 121-125.8. K. Kreider and M. McNeil, Eds., Waste Heat Management Guidebook,NBS Handbook 121, U.S. Government Printing Office,Washington, D.C., 1976, pp. 155-184.9. Baumeister and Marks, Eds., Standard Handbook for MechanicalEngineers, 7th ed. McGraw-Hill, New York, 1967.10. J.K. Salisbury, Kent’s Mechanical Engineers’ Handbook, Power, 12thed., Wiley, New York, 1967.11. Thermodynamic and Transport Properties of Steam, ASME, New York,1967.12. Selected Values of Chemical Thermodynamic Properties, NBS TechnicalNote 270-3, NBS, Washington, D.C., 1968.13. Gas Engineers Handbook, American Gas Association, New York,19.14. B.K. Hodge, Analysis and Design of Energy Systems, Prentice-Hall,Englewood Cliffs, NJ, 1985.15. Philip A. Schweitzer, Ed., Corrosion Resistance Tables, Vols, I and II,3rd Edition, Marcel Dekker, New York City, 1991.16. Weymueller, C.R., Duplex Steels Fight Corrosion, Welding Designand Fabrication, August 1990, pp 30-33, Penton, Cleveland, OH.17. G. Walker, Industrial Heat Exchangers, 2nd Ed., Hemisphere, NewYork, NY, 1990.18. K. Kreider and M. McNeil, Eds., Waste Heat Management Guidebook,NBS Handbook 121, U.S. Government Printing Office,Washington, D.C., 1976, pp. 25-51.

CHAPTER 9BUILDING ENVELOPEKEITH E. ELDER, P.E.Iverson Elder Inc.Bothell, Washington9.1 INTRODUCTIONBuilding “Envelope” generally refers to those buildingcomponents that enclose conditioned spaces and throughwhich thermal energy is transferred to or from the outdoorenvironment. The thermal energy transfer rate isgenerally referred to as “ heat loss” when we are trying tomaintain an indoor temperature that is greater than theoutdoor temperature. The thermal energy transfer rate isreferred to as “ heat gain” when we are trying to maintainan indoor temperature that is lower than the outdoortemperature. While many principles to be discussed willapply to both phenomena, the emphasis of this chapterwill be upon heat loss.Ultimately the success of any facility-wide energymanagement program requires an accurate assessmentof the performance of the building envelope. This istrue even when no envelope-related improvements areanticipated. Without a good understanding of how theenvelope performs, a complete understanding of theinteractive relationships of lighting and mechanical systemscannot be obtained.In addition to a good understanding of basic principles,seasoned engineers and analysts have becomeaware of additional issues that have a significant impactupon their ability to accurately assess the performance ofthe building envelope.1. The actual conditions under which products andcomponents are installed, compared to how they aredepicted on architectural drawings.2. The impact on performance of highly conductiveelements within the building envelope; and3. The extent to which the energy consumption of abuilding is influenced by the outdoor weather conditions,a characteristic referred to as thermal mass.It is the goal of this chapter to help the reader developa good qualitative and analytical understandingof the thermal performance of major building envelopecomponents. This understanding will be invaluable inbetter understanding the overall performance of the facilityas well as developing appropriate energy managementprojects to improve performance.9.1.1 Characteristics of Building Energy ConsumptionFigure 9.1 below shows superimposed plots of averagemonthly temperature and fuel consumption for anatural gas-heated facility in the Northwest region of theUnited States.Experienced energy analysts recognize the distinc-Figure 9.1221

222 ENERGY MANAGEMENT HANDBOOKtive shape of the monthly fuel consumption profile andcan often learn quite a bit about the facility just frominspection of this data. For example, the monthly energyconsumption is inversely proportional to the averagemonthly temperature. The lower the average monthlytemperature, the more natural gas appears to be consumed.Figure 9.1 also indicates that there is a period duringthe summer months when it appears no heating shouldbe required, yet the facility continues to consume someenergy. For natural gas, this is most likely that which isconsumed for the heating of domestic hot water, but itcould also be due to other sources, such as a gas range ina kitchen. This lower threshold of monthly energy consumptionis often referred to as the “ base,” and is characterizedby the fact that its magnitude is independent ofoutdoor weather trends. The monthly fuel consumptionwhich exceeds the “base” is often referred to as the “variable”consumption, and is characterized by the fact thatits magnitude is dependent upon the severity of outdoorenvironmental conditions. The distinction between thebase and the variable fuel consumption is depicted belowin Figure 9.2.9.1.2 Quantifying Building Envelope PerformanceThe rate of heat transfer through the building envelopewill be found to be related to the following importantvariables:1. Indoor and outdoor temperature;2. Conductivity of the individual envelope components;and3. The square footage of each of the envelope components.For a particular building component exposed to aset of indoor and outdoor temperature conditions, thesevariables are often expressed in equation form by the following:q = UA(T i – T o ) (9.1)Where:q = the component heat loss, Btu/hrU = the overall heat transfer coefficient, Btu/(hr-ft 2 -°F)A = the area of the component, ft 2T i = the indoor temperature, °FT o = the outdoor temperature, °FFigure 9.2Often the base consumption can be distinguishedfrom the variable by inspection. The lowest monthly consumptionoften is a good indicator of the base consumption.However there are times when a more accurate assessmentis necessary. Section 9.9 describes one techniquefor improving the analyst’s discrimination between baseand variable consumption.The distinction between base and variable consumptionis an important one, in that it is only the variablecomponent of annual fuel consumption which can besaved by building envelope improvements. The more accuratethe assessment of the fuel consumption, the moreaccurate energy savings projections will be.9.1.3 Temperatures for Instantaneous CalculationsThe indoor and outdoor temperatures for equation(9.1) are those conditions for which the heat loss needs tobe known. Traditionally interest has been in the designheat loss for a building or component, and is determinedby using so-called “design-day” temperature assumptions.Outdoor temperature selection should be madefrom the ASHRAE Handbook of Fundamentals, for thegeographic location of interest. The values published inthis chapter are those that are statistically known not tobe exceeded more than a prescribed number of hours(such as 2-1/2 %) of the respective heating or cooling season.For heating conditions in the winter, indoor temperaturesmaintained between 68 and 72°F result in comfortto the greatest number of people. Indoor temperaturesmaintained between 74 and 76°F result in the greatestcomfort to the most people during the summer (cooling)period.

BUILDING ENVELOPE 223We will see in later sections that other forms of temperaturedata collected on an annual basis are useful forevaluating envelope performance on an annual basis.The next section will take up the topic of how theU-factors for various envelope components are determined.9.2 PRINCIPLES OF ENVELOPE ANALYSISThe successful evaluation of building envelopeperformance first requires that the analyst be well-versedin the use of a host of analytical tools which adequatelyaddress the unique way heat is transferred through eachcomponent. While the heat loss principles are similar, thecalculation will vary somewhat from component to component.9.2.1 Heat Loss Through OpaqueEnvelope ComponentsWe have seen from Equation (9.1) that the heat lossthrough a component, such as a wall, is proportionalto the area of the component, the indoor-outdoor temperaturedifference, and U, the proportionality constantwhich describes the temperature-dependent heat transferthrough that component. U, currently described as the“ U-factor” by ASHRAE 2 , is the reciprocal of the totalthermal resistance of the component of interest. If thethermal resistance of the component is known, U can becalculated by dividing the total thermal resistance into“1” as shown:U = 1/R t (9.2)The thermal resistance, R t , is the sum of the individual resistancesof the various layers of material which comprisethe envelope component. R t is calculated by adding themup as follows:R t = R 1 + R 2 + R n +... (9.3)R 1 , R 2 and R n represent the thermal resistanceof each of the elements in the path of the “heat flow.”The thermal resistance of common construction materialscan be obtained from the ASHRAE Handbookof Fundamentals. Other physical phenomenon, suchas convection and radiation, are typically included aswell. For instance, free and forced convection are treatedas another form of resistance to heat transfer, and the“resistance” values are tabulated in the ASHRAE FundamentalsManual for various surface orientations andwind velocities. For example, the outdoor resistance dueto forced convection (winter) is usually taken as 0.17hr/(Btu–ft 2 –°F) and the indoor resistance due to freeconvection of a vertical surface is usually taken to be0.68 hr/(Btu–ft 2 –°F).To calculate the overall U-factor, one typicallydraws a cross-sectional sketch of the building componentof interest, assigns resistance values to the various materiallayers, sums the resistances and uses the reciprocalof that sum to represent U. The total calculation of a wallU-factor is demonstrated in the example below.ExampleCalculate the heat loss for 10,000 square feet of wallwith 4-inch face brick, R-11 insulation and 5/8-inch sheetrock when the outdoor temperature is 20°F and the indoortemperature is 70°F.Figure 9.3In the ASHRAE Handbook we find that a conservativeresistance for brick is 0.10 per inch. Four inchesof brick would therefore have a resistance of 0.40. Sheetrock (called gypsum board by ASHRAE) has a resistanceof 0.90 per inch, which would be 0.56 for 5/8-inch sheetrock. Batt insulation with a rating of R-11 will have a resistanceof 11.0, if expanded to its full rated depth.The first step is to add all the resistances.R iOutdoor Air Film (15 m.p.h.) 0.174-inch Face Brick 0.40R-11 Batt Insulation 11.005/8-inch Gypsum Board 0.56Indoor Air Film (still air) 0.68hr–ft 2 FR t = 12.81 —————BtuThe U-factor is the reciprocal of the total resistance1 1 BtuU o = — = —— = 0.078 ————R t 12.81 hr–ft 2 °F

224 ENERGY MANAGEMENT HANDBOOKThe heat loss is calculated by multiplying the U-factor bythe area and the indoor-outdoor temperature difference.q = 0.078 × 10,000 × (70 – 20)= 39,000 Btu/hrThe accuracy of the previous calculation is dependenton at least two important assumptions. The calculationassumes:1. The insulation is not compressed; and2. The layer(s) of insulation has not been compromisedby penetrations of more highly conductive buildingmaterials.9.2.2 Compression of InsulationThe example above assumed that the insulation isinstalled according to the manufacturer’s instructions.Insulation is always assigned its R-value rating accordingto a specific standard thickness. If the insulation iscompressed into a smaller space than it was rated under,the performance will be less than that published by themanufacturer. For example, R-19 batt insulation installedin a 3-1/2-inch wall might have an effective rating as lowas R-13. Table 9.1 7 is a summary of the performance thatcan be expected from various levels of fiberglass batt insulationtypes installed in different envelope cavities.9.2.3 Insulation PenetrationsOne of the assumptions necessary to justify the useof the one-dimensional heat transfer technique used inequation 9-1 is that the component must be thermallyhomogeneous. Heat is transferred from the warm sideof the component to the colder side and through eachindividual layer in a series path, much like current flowthrough simple electrical circuit with the resistances inseries. No lateral or sideways heat transfer is assumed totake place within the layers. For this to be true, the materialsin each layer must be continuous and not penetratedby more highly conductive elements.Unfortunately, there are very few walls in the realworld where heat transfer can truly be said to be one-dimensional.Most common construction has wood or metalstuds penetrating the insulation, and the presence ofthese other materials must be taken into consideration.Traditionally studs are accounted for by performingseparate U-factor calculations through both wall sections,the stud and the cavity. These two separate U-factors arethen combined in parallel by “weighting” them by theirrespective wall areas. The following example (Figure 9.4)shows how this would typically be done for a wall whosestuds, plates and headers constituted 23% of the totalgross wall area.R- R-Cavity FrameOutdoor Air Film (15 m.p.h.) 0.17 0.174-Inch Face Brick 0.40 0.40R-11 Batt Insulation 11.00 —3-1/2-Inch Wood Framing — 3.595/8-Inch Gypsum Board 0.56 0.56Indoor Air Film (still air) 0.6 0.68hr–ft 2 –°FR t = 12.81 5.40 —————Btuhr–ft 2 –°FU i = 0.078 0.185 —————BtuCombining the two U-factors by weighted fractions:hr–ft 2 –°FU o = 0.77 × 0.078 + 023 × 0.185 = 0.103 —————BtuIn the absence of advanced framing constructiontechniques, wood studs installed 16 inches on-centerTable 9.1 R-Value of fiberglass batts compressed within various depth cavities.

BUILDING ENVELOPE 225Figure 9.4comprise 20-25% of a typical wall (including windowframing, sill, etc.). Wood studs installed 24 inches on centercomprise approximately 15-20% of the gross wall.The use of this method is appropriate for situationswhere the materials in the wall section are sufficientlysimilar that little or no lateral, or “sideways” heat transfertakes place. Of course a certain amount of lateral heattransfer does take place in every wall, resulting in someerror in the above calculation procedure. The amount oferror depends on how thermally dissimilar the variouselements of the wall are. The application of this procedureto walls whose penetrating members’ conductivitiesdeviate from the insulation conductivities by less than anorder of magnitude (factor of 10) should provide resultsthat are sufficiently accurate for most analysis of constructionmaterials.Because the unit R-value for wood is approximately1.0 per inch and fiberglass batt insulation is R 3.1 per inch,this approach is justified for wood-framed building components.of construction.The introduction of the metal stud framing systeminto a wall has the potential to nearly double its heatloss! Just how is this possible, given that a typical metalstud is only about 1/20 of an inch thick? The magnitudeof this effect is counter-intuitive to many practicing designersbecause they have been trained to think of heattransfer through building elements as a one-dimensionalphenomenon, (as in the previous examples). But when ahighly conductive element such as a metal stud is presentin an insulated cavity, two and three-dimensional considerationsbecome extremely important, as illustratedbelow.Figure 9.5 shows the temperature distribution centeredabout a metal stud in a section of insulated wall.The lines, called “ isotherms,” represent regions with thesame temperature. Each line denotes a region that is onedegree Fahrenheit different from the adjacent line. Theselines are of interest because they help us to visualize thecharacteristics of the heat flow through the wall section.9.3 METAL ELEMENTSIN ENVELOPE COMPONENTSMost commercial building construction is notwood-framed. Economics as well as the need for fireratedassemblies has increased the popularity of metalframingsystems over the years. The conductivity ofmetal framing is significantly more than an order of magnitudegreater than the insulation it penetrates. In someinstances it is several thousand times greater. However,until recent years, the impact of this type of constructionon envelope thermal performance has been ignored bymuch of the design industry. Yet infrared photography inthe field and hot-box tests in the laboratory have demonstratedthe severe performance penalty paid for this typeFigure 9.5 Metal stud wall section temperature distribution.

226 ENERGY MANAGEMENT HANDBOOKThe direction of heat flow is perpendicular to the lines ofconstant temperature, and the heat flow is more intensethrough regions where the isotherms are closer together.It is possible to visualize both the direction and intensityof heat flow through the wall section by observing theisotherms alone.If the heat flow was indeed one-dimensional and occurringthrough a thermally homogeneous material, thelines would be horizontal and would show an even andlinear temperature-drop progression from the warm sideof the wall to the cold side. Notice that the isotherms are farfrom horizontal in certain regions around the metal stud.Figure 9.5 shows that the heat flow is not parallel, nor doesit move directly through the wall section. Also note that thearea with the greatest amount of heat flow is not necessarilyrestricted to the metal part of the assembly. The metalstud has had a negative influence on the insulation in theadjacent region as well. This is the reason for the significantincrease in heat loss reported above for metal studs.Clearly a different approach is required to determinemore accurate U-factors for walls with highly conductiveelements.9.3.1 Using Parallel-Path Correction FactorsFortunately, most commercial construction consistsof a limited number of metal stud assembly combinations,so the results of laboratory “hot box” tests of typical wallsections can be utilized to estimate the U-factors of wallswith metal studs. ASHRAE Standard 90.1 recommendsthe following equation be used to calculate the equivalentresistance of the insulation layer installed between metalstuds:Where:Where:R t = R i + R e (9.4)R t = the total resistance of the envelope assemblyR i = the resistance of all series elements exceptthe layer of the insulation & studsR e = the equivalent resistance of the layercontaining the insulation and studs= R insulation × F cF c = the correction factor from Table 9.2The use of the above multipliers in wall U-factorcalculation is demonstrated in Figure 9.6.Table 9.2 Parallel path correction factors.—————————————————————————Size of Framing Insulation CorrectionMembers R-Value Factor, F c—————————————————————————2 × 4 16 in. O.C. R-11 0.502 × 4 24 in. O.C. R-11 0.602 × 6 16 in. O.C. R-19 0.402 × 6 24 in. O.C. R-19 0.45—————————————————————————Figure 9.6R iOutdoor Air Film (15 m.p.h.) 0.174-inch Face Brick 0.40R-11/Metal Stud 16" O.C.= 11.0 × 0.50 = 5.505/8-inch Gypsum Board 0.56Indoor Air Film (sill air) 0.68R t =7.31 hr - ft2 –°FBtuU o =R 1 = 1t 7.31 = 0.137 Btuhr - ft 2 –°FNotice that only one path is calculated through theassembly, rather than two. The insulation layer is simplycorrected by the ASHRAE multiplier and the calculationis complete. Also notice that it is only the metal stud/insulationlayer that is corrected, not the entire assembly.This does not mean that only this layer is affected by themetal studs, but rather that this is the approach ASHRAEStandard 90.1 intends for the factors to give results consistentwith tested performance.The above example shows that the presence of the

BUILDING ENVELOPE 227metal stud has increased the wall heat loss by 75%! Theimpact is even more severe for R-19 walls. The importanceof accounting for the impact of metal studs in envelopeU-factor calculations cannot be overstated.9.3.3 The ASHRAE “Zone Method”The ASHRAE Zone Method should be used for U-factor calculation when highly conductive elements arepresent in the wall that do not fit the geometry or spacingcriteria in the above parallel path correction factor table.The Zone Method, described in the ASHRAE Handbookof Fundamentals, is an empirically derived procedurewhich has been shown to give reasonably accurate answersfor simple wall geometries. It is a structured way tocalculate the heat transmission through a wall using bothseries and parallel paths.The Zone Method takes its name from the fact thatthe conductive element within the wall influences theheat transmission of a particular region or “zone.” Thezone of influence is typically denoted as “Zone A,” whichexperiences a significant amount of lateral conduction.The remaining wall section that remains unaffected iscalled “Zone B.” The width of Zone A, which is denoted“W,” can be calculated using the following equation:W=m + 2d (9.5)Where “m” is the width of the metal element at itswidest point and “d” is the shortest distance from themetal surface to the outside surface of the entire component.If the metal surface is the outside surface of thecomponent, then “d” is given a minimum value of 0.5 (inEnglish units) to account for the air film.For example a 3-1/2-inch R-11 insulation is installedbetween 1.25-inch wide metal studs 16 inches on-center.The assembly is sheathed on both sides with 1/2-inchgypsum board. The variable “m” is the widest part of thestud, which is 1.25 inches. The variable “d” is the distancefrom the metal element to the outer surface of the wall,which in this case, is the thickness of the gypsum wallboard, 0.5 inches. The width of Zone A is:WA = 1.25 + (2 × 0.5) =The width of Zone B is:2.25 inchesWB = 16 – 2.25 = 13.75 inchesFigure 9.7 shows the boundaries of the zones describedabove.Actual step-by-step procedures for performing thecalculations for Zone A and Zone B can be found in theFigure 9.7 Zones A and B.ASHRAE Handbook of Fundamentals as well as otherpublications. 59.3.4 Improving the Performance ofEnvelope Components with Metal ElementsThe impact of metal elements can be mitigated by:1. Installing the insulation outside of the layer containingmetal studs.2. Installing interior and exterior finished materials onhorizontal “hat” sections, rather than directly on thestuds themselves.3. Using expanded channel (thermally improved)metal studs.4. Using non-conductive thermal breaks at least 0.40to 0.50 inches thick between the metal element andthe inside and outside sheathing. The value of thethermal break conductivity should be at least a factorof 10 less than the metal element in the envelopecomponent.9.3.5 Metal Elements in Metal Building WallsMany metal building walls are constructed from acorrugated sheet steel exterior skin, a layer of insulation,and sometimes a sheet steel V-rib inner liner. The innerliner can be fastened directly to the steel frame of thebuilding. The exterior cladding is attached to the innerliner and the structural steel of the building through coldformed sheet steel elements called Z-girts. Fiber glass battinsulation is sandwiched between the inner liner and theexterior cladding. Figures 9.8 and 9.9 show details of atypical sheet steel wall.The steel framing and metal siding materials provideadditional opportunities for thermal short circuitsto occur. The highly conductive path created by the metalin the girts, purlins and frames connected directly to themetal siding can result in even greater thermal short-cir-

228 ENERGY MANAGEMENT HANDBOOKheat loss will be dictated by the spacing of the Z-girts, thegauge of metal used and the contact between the metalgirts and the metal wall panels.9.3.6 Metal Wall PerformanceWhile the performance of metal building walls willvary significantly, depending on construction features,Tables 9.3-9.5 will provide a starting point in predictingthe performance of this type of construction. 69.3.7 Strategies for Reducing Heat Loss inMetal Building WallsFigure 9.8 Vertical section view.Figure 9.9 Horizontal section view.cuiting than that discussed in Section 9.3.4 for metal studsin insulated walls. Because of this unobstructed high conductivitypath, the temperature of the metal element isnearly the same through the entire assembly, rather thanvarying as heat flows through the assembly.In laboratory tests, the introduction of 16 gauge Z-Girts spaced 8 feet apart reduced the R-value of a wallwith no girts from 23.4 to 16.5 hr–ft 2 –°F/Btu. 4 Substitutionof Z-Girts made of 12-gauge steel reduced the R-valueeven further, to 15.2 hr–ft 2 –°F/Btu. The extent of this9.3.7.1 Maintain Maximum Spacing Between Z-GirtsTests on a 10-inch wall 8 with a theoretical R-value of36 hr-ft 2 -°F/Btu demonstrated a measured R-value of 10.2hr-ft 2 -°F/Btu with the girts spaced 2 feet apart. Increasingthe spacing to 8 feet between the girts had a correspondingeffect of increasing the R-value to 16.5 hr-ft 2 -°F/Btu.While typical girt spacing in the wall of a metal buildingis 6 feet on-center, this study does emphasize the desirabilityof maximizing the girt spacing wherever possible.9.3.7.2 Use “Thermal” GirtsThe thermal performance of metal building wallscan be improved by the substitution of thermally improvedZ-girts. The use of expanded “thermal” Z-Girtshas the potential to increase the effective R-value by 15-20% for a wall with Girts spaced 8 feet apart. 4 A significantamount of the metal has been removed from thesethermal girts, which increases the length of the heat flowpath and reduces the girt’s heat flow area. Figure 9.10indicates the basic structure of such a girt.Figure 9.10 Expanded “thermal” girt.Table 9.3 Wall insulation installed between 16 gauge Z-girts.NominalUnbridged Thickness R-Value R-Value R-Value R-Value R-ValueR-Value (Inches) 2' O.C. 4' O.C. 5' O.C. 6' O.C. 8' O.C.6 2 4.9 5.7 5.8 5.9 6.010 3 6.4 7.9 8.3 8.5 8.713 4 7.4 9.4 9.9 10.2 10.619 5-1/2 9.0 12.0 12.9 13.4 14.223 7 10.2 13.8 14.9 15.6 16.536 10 14.1 19.8 21.5 22.8 24.4

BUILDING ENVELOPE 229Table 9.4 Wall insulation installed between 12 gauge Z-girts.NominalUnbridged Thickness R-Value R-Value R-Value R-Value R-ValueR-Value (Inches) 2' O.C. 4' O.C. 5' O.C. 6' O.C. 8' O.C.6 2 4.8 5.6 5.8 5.9 6.010 3 6. 1 7.7 8. 1 8.3 8.613 4 6.9 9.0 9.5 9.9 10.319 5-1/2 8.1 11.1 12.0 12.6 13.423 7 8.7 12.3 13.4 14. 1 15.2Table 9.5 combines test results with calculations toestimate the benefits that may be realized utilizing “thermal”girts.9.3.7.3 Increase Z-Girt-to-MetalSkin Contact ResistanceWhile the Z-Girt itself does not offer much resistanceto heat flow, testing has found that up to half of theresistance of the girt-to-metal skin assembly is attributableto “poor” contact between the girt and the metalskin. Quantitative estimates of the benefit of using lessconductive materials to “break” the interface between thetwo metals are not readily available, but laboratory testshave shown some improvement in thermal performance,just by using half the number of self-tapping screws tofasten the metal skin to the girts.9.3.7.4 Install Thermal Break Between Metal ElementsThe impact of this strategy will vary with the flexibilityoffered by the geometry of the individual wallcomponent, the on-center spacing of the wall girts,thermal break material and contact resistance betweenthe girt and metal wall before consideration of the thermalbreak. Figure 9.11 illustrates the impact of thermalbreaks of varying types and thicknesses installed betweena corrugated metal wall and Z-Girts mounted 6feet on-center.To be effective, reduction by a factor of 10 or morein thermal conductivity between the thermal break andthe metal may be required for a significant improvementin performance. The nominal thickness of the break alsoplays an important role. Tests of metal panels indicatethat, regardless of the insulator, inserts less than 0.40inches will not normally be sufficient to prevent substantialloss of their insulation value.9.3.8 Calculating U-factors for Metal Building WallsASHRAE Standard 90.1 recommends The Johannesson-VinbergMethod 7 for determining the U-factorfor sheet metal construction, internally insulated with ametal structure bonded on one or both sides with a metalskin. This method is useful for the prediction of U-factorsfor metal girt/purlin constructions which can be definedby the geometry described in Figure 9.12.The dimensions of the elements shown, as well astheir conductivities are evaluated in a series of formulations,which are in turn combined in two parallel pathsto determine an overall component resistance. The advantageof this method is that metal gauge as well asthermal break thickness and conductivity can be takeninto account in the calculation. Caution is advised whenusing this method to evaluate components with metalto-metalcontact. The method assumes complete thermalcontact between all components, which is not always thecase in metal building construction. To account for thiseffect, values for contact resistance can be introduced inthe calculation in place of or in addition to the thermalbreak layer.Table 9.5 Wall insulation installed between 16 gauge “thermal” Z-girts.NominalUnbridged Thickness R-Value R-Value R-Value R-Value R-ValueR-Value (Inches) 2' O.C. 4' O.C. 5' O.C. 6' O.C. 8' O.C.6 2 5.6 6.2 6.3 6.4 6.410 3 8.3 9.4 9.6 9.8 10.013 4 10.2 1 1.7 12. 1 12.3 12.619 5-1/2 14.0 16.4 16.9 17.3 17.823 7 16.4 19.5 20.2 20.7 21.3

230 ENERGY MANAGEMENT HANDBOOKFigure 9.11 5 Impact of thermal break on metal wall R-value corrugated metal with Z-girts 6’0" on-center.Figure 9.12 Metal skinned structure w/internal metal element.9.4 ROOFSIn many cases the thermal performance of roofstructures is similar to that of walls, and calculations canbe performed in a similar way to that described in theprevious examples. As was the case with walls, metalpenetrations through the insulation will exact a penalty.With roofs these penetrations will usually take one ofseveral forms. The first form is the installation of battinsulation between z-purlins, similar to that discussed inSection 9.3. The performance of these assemblies will besimilar to that of similarly constructed walls.9.4.1 Insulation Between Structural TrussesAnother common thermal short circuit is that whichoccurs as a result of installing the insulation betweenstructural trusses with metal components (Figure 9.13).Table 9.6 is a summary of the derated R-values thatmight be expected as a result of this type of installationcompared to the unbridged R-values published by insulationmanufacturers.9.4.2 Insulation Installed “ Over-the-Purlin”One of the most economical methods of insulatingthe roof of a metal building, shown in Figure 9.14, is tostretch glass fiber blanket insulation over the purlinsor trusses, prior to mounting the metal roof panels ontop.The roof purlin is a steel Z-shaped member witha flange width of approximately 2-1/2 inches. Thestandard purlin spacing for girts supporting roofsin the metal building industry is 5 feet. Using such amethod, the insulation is compressed at the purlin/panel interface, resulting in a thermal short circuit. Theinsulation thickness averages 2-4 inches, but can rangeup to 6 inches. While thicker blanket insulation can bespecified to mitigate this, the thickness is limited bystructural considerations. With the reduced insulationthickness at the purlin, there is a diminishing returnon the investment in additional insulation. Table 9.7is a summary of the installed R-values that might beexpected using this installation technique.

BUILDING ENVELOPE 231Figure 9.14Figure 9.13Table 9.6 Roof insulation installedbetween metal trusses. 5—————————————————————————Unbridged R-Value R-Value R-ValueR-Value 4 Ft O.C. 10 6 Ft O.C. 11 8 Ft O.C. 115 4.8 4.9 4.910 9.2 9.6 9.715 13.2 14.0 14.320 17.0 18.4 18.925 20.3 22.4 23.230 23.7 26.5 27.6—————————————————————————9.4.2.1 Calculating Roof Performance for“Over-the-Purlin” InstallationsTraditional methods of calculating assembly R-valueswill not result in accurate estimates of thermal performancefor “over-the-purlin” installations. The insulationis not installed uniformly, nor at the thickness requiredto perform at the rated R-value, and it is penetrated withmultiple metal fasteners. The Thermal Insulation ManufacturersAssociation (TIMA) has developed an empiricalformula based on testing of insulation meeting the TIMA202 specification. Insulation not designed to be laminated(such as filler insulation) will show performance up to15% less than indicated by the formula below. Assumingthe insulation meets the TIMA specification, the U-factorof the completed assembly can be estimated as:U = 0.012 +0.2550.31 × R f +t× 1– N L+N×0.198 + 0.065×nL (9.6)Where:L = Length of Building Section, feetN = Number of Purlins or Girts in the L Dimensionn = Fastener Population per Linear Foot of PurlinRf =Sum of Inside andhr – ft 2 °FOutside Air Film R-Values, —————Btut = Pre-installed insulation thickness (for TIMA202 type insulation), inchesExample Calculation Using the TIMA FormulaAssume a metal building roof structure with the followingcharacteristics:Length of Building Section= 100 feetNumber of Purlins= 21 purlinsFastener Population per foot of Purlin = 1 per foothr-ft 2 °FSum of Air Film R-Values = 0.61 + 0.17 = 0.78 ————BtuInsulation thickness= 5 inchesTable 9.7 Roof insulation installed compressed over-the-purlin 5 .NominalUnbridged Thickness R-Value R-Value R-Value R-Value R-ValueR-Value (Inches) 2' O.C. 3' O.C. 4' O.C. 5' O.C. 6' O.C.10 3 5.4 6.5 7.2 7.7 8. 113 4 5.7 7.1 8.0 8.7 9.316 5-1/2 5.9 7.4 8.6 9.5 10.219 6 6.0 7.7 9.0 10.0 10.9

232 ENERGY MANAGEMENT HANDBOOKExample Calculation Using the TIMA FormulaAssume a metal building roof structure with the followingcharacteristics:Length of Building SectionNumber of PurlinsFastener Population per Foot of Purlin= 100 feet= 21 purlins= 1 per foothr-ft 2 °FSum of Air Film R-Values = 0.61 + 0.17 = 0.78 ————BtuInsulation thicknessU = 0.012 +×= 5 inches0.255× 1– 21(0.31 × 0.78 + 6) 105 +210.198 + 0.065 × 1105= 0.0973Btuhr – ft 2 °F9.4.3 Strategies for Reducing Heat Loss inMetal Building RoofsMany of the strategies suggested for metal walls areapplicable to metal roofs. In addition to strategies thatminimize thermal bridging due to metal elements, otherfeatures of metal roof construction can have a significantimpact on the thermal performance.Tests have shown significant variation in the thermalperformance of insulation depending upon the facingused, even though the permeability of the facing itself hasno significant thermal effect. The cause of this differenceis in the flexibility of the facing itself. The improvementin permeability rating produces a higher U-factor ratingdue to the draping characteristics. During installation,the insulation is pulled tightly from eave to eave of thebuilding. The flexibility of vinyl facing allows it to stretchand drape more fully at the purlins than reinforced facing.Figures 9.15 and 9.16 show a generalized illustrationof the difference in drape of the vapor retarders and theeffect on effective insulation thickness.The thermal bridging and insulation compressionissues discussed above make the following recommendationsworth considering.9.4.3.1 Use the “ Roll-Runner” Method ofInstalling “Over-the-Purlin” InsulationThe “Roll-Runner” installation technique helps tomitigate the effect described above, achieving less insulationcompression by improving the draping characteristicsof the insulation. This is done using bands or strapsto support the suspended batt insulation, as illustrated inFigure 9.17.Figure 9.17 Roll Runner MethodFigure 9.15 Vinyl facing.Figure 9.16 Vinyl facing with glass fiber scrim.Figure 9.18 Insulated purlin with “roll-runner” installation.

BUILDING ENVELOPE 2339.4.3.2 Use Thermal Spacers Between Purlin andStanding Seam Roof DeckFigure 9.18 depicts a thermal spacer installed betweenthe top of the purlin and the metal roof deck. The impact ofthis strategy will vary with the geometry of the individualcomponent as well as the location of the spacer.Table 9.8 summarizes the comparative performanceof the two installation techniques discussed above. Inall cases, purlins were installed 5 feet on-center. Whileimprovements of 8 to 19 percent in effective R-value areachieved using the Roll-Runner method, performance of60 to 80 percent can be realized when the purlins were alsoinsulated from direct contact with the metal structure.9.4.3.3 Install Additional UncompressedInsulation Between Purlins or Bar JoistsThis insulation system illustrated in Figure 9.19,is sometimes referred to as “FullDepth/Sealed Cavity,” referring tothe additional layer of insulationthat is installed between purlins orbar joists, and which is not compressedas is the over-the-purlininsulation above it. In this configuration,the main function of theover-the-purlin insulation above isto act as a thermal break.over a crawl space. The problem is that knowledge of thetemperature of the crawl space is necessary to performthe calculation. But the temperature of the crawl space isdependent on the number of exposed crawl space surfacesand their U-factors, as well as the impact of crawl spaceventing, if any. For design-day heat loss calculations, itis usually most expedient to assume a crawl space temperatureequal to the outdoor design temperature. Thiswill very nearly be the case for poorly insulated or ventedcrawl spaces.When actual, rather than worst-case heat loss isneeded, it is necessary to perform a heat balance on thecrawl space. The process is described in the ASHRAEHandbook of Fundamentals. Below is a brief summary ofthe approach.Table 9.8 “Roll runner” methodMfgr’s Nominal Over the Roll Roll-RunnerRated Thickness Purlin Runner with InsulatedR-Value (Inches) Method Method Purlin10 3 7.1 7.7 12.513 4 8.3 9.1 14.319 6 11.1 12.5 20.09.4.3.4 Add Rigid Insulation Outsideof PurlinsThe greatest benefit will bederived where insulation can beadded which is neither compressednor penetrated by conductive elements,as shown in Figure 9.20.Table 9.9 comparatively summarizesthe performance of theabove installation for varying levelsof insulation installed over purlinswith varying on-center spacing.Figure 9.19 Insulation suspension system.9.5 FLOORSFloors above grade and exposed to outdoor air canbe calculated much the same way as illustrated previouslyfor walls, except that the percentages assumed for floorjoists will vary somewhat from that assumed for typicalwall constructions.9.5.1 Floors Over Crawl SpacesThe situation is a little different for a floor directlyFigure 9.20 Insulation suspension system.

234 ENERGY MANAGEMENT HANDBOOKTable 9.9 Insulation “over-the-purlin” w/rigid insulation outside of purlin 14 .UninstalledUnbridged Thickness R-Value R-Value R-Value R-Value R-ValueR-Value (Inches) 2' O.C. 3' O.C. 4' O.C. 5' O.C. 6' O.C.13 2 12.2 12.9 13.3 12.5 13.817 3 12.6 13.7 14.4 14.3 15.320 4 12.9 14.3 15.2 15.5 16.523 5-1/2 13.1 14.6 15.8 16.7 17.426 6 13.2 14.9 16.2 18.3 18.1Heat loss, q floor from a Floor to a Crawl Space:q floor = q perimeter + q ground + q air exchangeU f A f (t i –t c ) = U p A p (t c –t o ) + U g A g (t c –t g )+ 0.67ρc p V c (t c –t o ) (9.7)Where:t i = indoor temperature, °Ft o = outdoor temperature, °Ft g = ground temperature, °Ft c = crawl space temperature, °FA f = floor area, ft 2A p = exposed perimeter area, ft 2A g = ground area (Ag = Af), ft 2U f = floor heat transfer coefficient,Btu/(hr–ft 2 –°F)U g = ground coefficient, Btu/hr–ft 2 °FU p = perimeter heat transfer coefficient,Btu/(hr-ft 2 -°F)V c = volume of the crawl space, ft 3ρc p = volumetric air heat capacity(0.018 Btu/(ft 3 –°F)0.67 = assumed air exchange rate (volume/hour)The above equation must be solved for t c the crawlspace temperature. Then the heat loss from the spaceabove to the crawl space, using the floor U-factor can becalculated.9.5.2 Floors On GradeA common construction technique for commercialbuildings is to situate the building on a concrete slabright on grade. The actual physics of the situation canbe quite complex, but methods have been developedto simplify the problem. In the case of slab-on-gradeconstruction, it has been found that the heat loss is proportionalto the perimeter length of the slab, rather thanthe floor area. Rather than using a U-factor, which isnormally associated with a wall or roof area, we use an“F-factor,” which is associated with the number of linearfeet of slab perimeter. The heat loss is given by the equationbelow:q slab = F × Perimeter × (T inside – T outside ) (9.8)F-factors are published by ASHRAE, and are alsoavailable in many state energy codes. As an example, theF-factor for an uninsulated slab is 0.73 Btu/(hr-ft-°F). Aslab with 24 inches of R-10 insulation installed inside thefoundation wall would be 0.54 Btu/(hr-ft-°F).9.5.3 Floors Below GradeVery little performance information exists on theperformance of basement floors. What does exist is morerelevant to residential construction than commercial. Fortunately,basement floor loss is usually an extremely smallcomponent of the overall envelope performance. For everyfoot the floor is located below grade, the magnitudeof the heat loss is diminished dramatically. Floor heat lossis also affected somewhat by the shortest dimension ofthe basement. Heat loss is not directly proportional to theoutside ambient temperature, as other above grade envelopecomponents. Rather, basement floor loss has beencorrelated to the temperature of the soil four inches belowgrade. This temperature is found to vary sinusoidallyover the heating season, rather on a daily cycle, as the airtemperature does.A method for determining basement floor and wallheat loss is described, with accompanying calculations, inthe ASHRAE Handbook of Fundamentals.9.6 FENESTRATIONThe terms “ Fenestration,” “window,” and “glazing”are often used interchangeably. To describe the importantaspects of performance in this area requires that terms bedefined carefully. “Fenestration” refers to the design andposition of windows, doors and other structural openingsin a building. When we speak of windows, we are actually

BUILDING ENVELOPE 235describing a system of several components. Glazing isthe transparent component of glass or plastic windows,doors, clerestories, or skylights. The sash is a frame inwhich the glass panes of a window are set. The Frame isthe complete structural enclosure of the glazing and sashsystem. Window is the term we give to an entire assemblycomprised of the sash, glazing, and frame.Because a window is a thermally nonhomogeneoussystem of components with varying conductive properties,the thermal performance cannot be accuratelyapproximated by the one-dimensional techniques usedto evaluate common opaque building envelope components.The thermal performance of a window system willvary significantly, depending on the following characteristics:• The number of panes• The dimension of the space between panes• The type of gas between the panes• The emissivity of the glass• The frame in which the glass is installed• The type of spacers that separate the panes of glass9.6.1 Multiple Glass PanesBecause of the low resistance provided by the glazingitself, the major contribution to thermal resistance insingle pane glazing is from the indoor and outdoor airfilms. Assuming 0.17 outdoor and 0.68 indoor air filmresistances, a single paned glazing unit might be expectedto have an overall resistance of better than 0.85 (hr-ft 2 -°F)/Btu, or a U-factor of 1.18 Btu/(hr-ft 2 -°F). The additionof a second pane of glass creates an additional space in theassembly, increasing the glazing R-value to 1.85, whichresults in a U-factor of approximately 0.54 Btu/(hr-ft 2 -°F).Similarly, the addition of a third pane of glass might increasethe overall R-value to 2.85 (hr-ft 2 °F)/Btu, yieldinga U-factor of 0.35 Btu/(hr-ft 2 -°F). This one-dimensionalestimate does not hold true for the entire window unit,but only in the region described by ASHRAE as the “center-of-glass”(see Figure 9.21). Highly conductive framingor spacers will create thermal bridging in much the samefashion as metal studs in the insulated wall evaluated inSection 9.31.Figure 9.21 Window components.9.6.2 Gas Space Between PanesMost multiple-paned windows are filled with dryair. The thermal performance can be improved by thesubstitution of gases with lower thermal conductivities.Other gases and gas mixtures used besides dry air areArgon, Krypton, Carbon Dioxide, and Sulfur Hexafluoride.The use of Argon instead of dry air can improve the“center-of-glass” U-factor by 6-9%, depending on the distancebetween panes, and CO 2 filled units achieve similarperformance to Argon gas. For spaces up to 0.5 inches,the mixture of Argon and SF 6 gas can produce the sameperformance as Argon, and Krypton can provide superiorperformance to that of Argon.9.6.3 EmissivityEmissivity describes the ability of a surface togive off thermal radiation. The lower the emissivity ofa warm surface, the less heat loss that it will experiencedue to radiation. Glass performance can be substantiallyimproved by the application of special low emissivitycoatings. The resulting product has come to be known as“Low-E” glass.Two techniques for applying the Low-E film aresputter and pyrolytic coating. The lowest emittancesare achieved with a sputtering process by magneticallydepositing silver to the glass inside a vacuum chamber.Sputter coated surfaces must be protected within an insulatedglass unit and are often called “soft coat.” Pyrolyticcoating is a newer method which applies tin oxide to theglass while it is still somewhat molten. The pyrolyticprocess results in higher emittances than sputter coating,but surfaces are more durable and can be used for singleglazed windows. While normal glass has an emissivity ofapproximately 0.84, pyrolytic coatings can achieve emissivitiesof approximately 0.40 and sputter coating canachieve emissivities of 0.10 and lower. The emittance ofvarious Low-E glasses will vary considerably betweenmanufacturers.9.6.4 Window FramesThe type of frame used for the window unit willalso have a significant impact on the performance. Ingeneral, wood or vinyl frames are thermally superiorto metal. Metal frame performance can be improvedsignificantly by the incorporation of a “ thermal-break.”This usually consists of the thermal isolation of the cold

236 ENERGY MANAGEMENT HANDBOOKside of the frame from the warm side by means of somelow-conducting material. Estimation of the performanceof a window due to the framing elements is complicatedby the variety of configurations and combinations ofmaterials used for sash and frames. Many manufacturerscombine materials for structural or aestheticpurposes. Figure 9.22 below illustrates the impact thatvarious framing schemes can have on glass performanceschemes.The center-of-glass curve illustrates the performanceof the glazing system without any framing. Noticethat single pane glass is almost unaffected by the framingscheme utilized. This is due to the similar order-of-magnitudethermal conductivity between glass and metal.As additional panes are added, emissivities are loweredand low conductivity gases are introduced, the impact ofthe framing becomes more pronounced, as shown by theincreasing performance “spread” toward the right handside of the chart. Notice that a plain double pane windowwith a wood or vinyl frame actually has similar performanceto that of Low-E glass (hard coating) with Argongas fill and a metal frame. Also note how flat the curveis for metal-framed windows with no thermal break. Itshould be clear that first priority should be given to framingsystems before consideration is given to Low-E coatingsor low conductivity gases.9.6.5 SpacersDouble and triple pane window units usuallyhave continuous members around the glass perimeterto separate the glazing lites and provide an edge seal.These spacers are often made of metal, which increasesthe heat transfer between the panes and degrades theperformance of the glazing near the perimeter. ASHRAEreports this conductive region to be limited to a 2.5-inchband around the perimeter of the glazing unit, and appropriatelydescribes it as the “edge of glass.” Obviously,low conductivity spacers, such as plastic, fiberglass oreven glass, are to be preferred over high conductivityspacers, such as metal. The impact of highly conductivespacers is only felt to the extent that other high performancestrategies are incorporated into the window. Forexample, ASHRAE reports no significant performancedifference between spacers types when incorporated intowindow systems with thermally unimproved frames.When thermally improved or thermally broken framesare incorporated, the spacer material becomes a factor inoverall window performance.The best performing windows will incorporatethermally broken frames, non-metallic spacers betweenthe panes, low-emissivity coatings, or low conductivitygases, such as argon or krypton, between the panes.9.6.6 Advanced Window Technologies9.6.6.1 Suspended Interstitial FilmsHigh performance glazing is available that utilizesone or more Low-E coated mylar or polyester films betweenthe inner and outer glazing of the window unit.High performance is achieved by the addition of a secondair space, another Low-E surface, but without the weightof an additional pane of glass.Figure 9.22 Window system performance comparison.

BUILDING ENVELOPE 2379.6.6.2 Advanced SpacersSignificant improvements in spacer design havebeen made in the past 10-15 years. The “Swiggle Strip”spacer sandwiches a thin piece of corrugated metal betweentwo thicker layers of butyl rubber combined with adesiccant. The conductivity is reported to be one-quarterto one-half of the conductivity of a normal aluminumspacer. Other spacers with conductivities only 6 to 11percent as great as aluminum have been designed usingsilicon foam spacers with foil backing or by separatingtwo conventional aluminum spacers with a separate stripof polyurethane foam.9.6.6.3 Future Window Enhancement TechnologiesElectrochromic glass systems, referred to as “ smartwindows” sandwich indium tin oxide, amorphous tungstentrioxide, magnesium fluoride and gold betweenlayers of heat-resistant Pyrex. Imposing a small voltageto the window changes its light and heat transmissioncharacteristics, allowing radiation to be reflected in thesummer and admitted in the winter.Windows with R-values of R-16 are theoreticallypossible if the air or other insulating gas between windowpanes is replaced with a vacuum. Historically, a permanentvacuum has not been achievable due to the difficultyof forming an airtight seal around the edges of theunit. Work is presently under way on special techniquesto create a leakproof seal.9.6.7 Rating the Performance of Window ProductsWindow energy performance information hasnot been made available in any consistent form, untilrecently. Some manufacturers publish R-values thatrival many insulation materials. Usually the center-ofglassU-factor is published for glazing, independent ofthe impact of the framing system that will eventuallybe used. While the center-of-glass rating alone may beimpressive, it does not describe the performance of theentire window. Other manufacturers provide performancedata for their framing systems, but they do notreflect how the entire product performs. Even whenmanufacturers provide ratings for the whole productdifferent methods are used to determine these ratings,both analytical and in laboratory hot box tests. This hasbeen a source of confusion in the industry, requiringsome comprehensive standard for the reporting of windowthermal performance.The National Fenestration Rating Council (NFRC),sanctioned by the federal government under the EnergyPolicy Act of 1992 was established to develop a nationalperformance rating system for fenestration products.The NFRC has established a program where the factorsthat affect window performance are included in publishedperformance ratings. NFRC maintains a directoryof certified products. Currently many local energy codesrequire that NFRC ratings be used to demonstrate compliancewith their window performance standards.9.6.8 DoorsIn general, door U-factors can be determined ina similar manner to the walls, roofs and exposed floordemonstrated above. Softwoods have R-values around1.0 to 1.3 per inch, while hardwoods have R-values thatrange from 0.80 to 0.95 per inch. A 1-3/8-inch panel doorhas a U-factor of approximately 0.57 Btu/(Hr-ft 2 -°F),while a 1-3/4 solid core flush door has a U-factor of approximately0.40.As has been shown to be the case with some wallsand windows, metal doors are a different story. Thesame issues affecting windows, such as framing andthermal break apply to metal doors as well. The U-factorof a metal door can vary from 0.20 to 0.60 Btu/(Hr-ft 2 -°F) depending on the extent to which the metal-to-metalcontact can be “broken.” In the absence of tested door U-factor data, Table 6 page 24.13 in Chapter 24 of the 1997ASHRAE Handbook of Fundamentals can be used toestimate the U-factor of typical doors used in residentialand commercial construction.9.7 INFILTRATIONInfiltration is the uncontrolled inward air leakagethrough cracks and interstices in a building element andaround windows and doors of a building, caused bythe effects of wind pressure and the differences in theoutdoor/indoor air density or pressure differentials. Theheat loss due to infiltration is described by the followingequation:q infiltration = 0.019 × Q × (T inside – T outside ) (9.9)Where Q is the infiltration air flow in cubic feet per hour.The determination of Q is an extremely impreciseundertaking, in that the actual infiltration for similarbuildings can vary significantly, even though observableparameters appear to be the same.9.7.1 Estimating Infiltration for Residential BuildingsASHRAE suggests that the infiltration rate for aresidence can be estimated as:Q = L [(A(T i – T o ) + (Bv 2 )] 1/2 (9.10)

238 ENERGY MANAGEMENT HANDBOOKWhere:Q = The infiltration rate, ft 3 /hrA = Stack Coefficient, CFM 2 /[in 4 -°F]T i = Average indoor Temperature, °FT o = Average outdoor Temperature, °FB = Wind Coefficient, CFM 2 /[in 4 -(mph) 2 ]v = Average wind speed, mphL = Crack leakage area, in 2and the particular location on the building, which in turnis effected by temperature, distance from the building“neutral plane,” the geometry of the building exteriorenvelope elements, and the interior partitions, and allof their relationships to each other. If infiltration due toboth wind velocity (Q w ) and stack effect (Q s ) can be determined,they are combined as follows to determine theoverall infiltration rate.The method is difficult to apply because:Q ws = Q w 2 +Q s2(9.11)1. L, the total crack area in the building is difficult todetermine accurately;2. The determination of the stack and wind coefficientis subjective;3. The average wind speed is extremely variable fromone micro-climate to another; and4. Real buildings, built to the same standards, do notexperience similar infiltration rates.ASHRAE reports on the analysis of several hundredpublic housing units where infiltration varied from0.5 air changes per hour to 3.5 air changes per hour. Ifthe real buildings experience this much variation, wecannot expect a high degree of accuracy from calculations,unless a significant amount of data are available.However, the studies reported provided some usefulguidelines.In general, older residential buildings withoutweather-stripping experienced a median infiltrationrate of 0.9 air changes per hour. Newer buildings,presumably built to more modern, tighter constructionstandards, demonstrated median infiltration rates of 0.5air changes per hour. The structures were unoccupiedduring tests. It has been estimated that occupants addan estimated 0.10 to 0.15 air changes per hour to theabove results.9.7.2 Estimating Infiltration forComplex Commercial BuildingsInfiltration in large commercial buildings is considerablymore complex than small commercial buildingsor residential buildings. It is affected by both windspeed and “stack effect.” Local wind speed is influencedby distance from the reporting meteorological station,elevation, and the shape of the surrounding terrain. Thepressure resulting from the wind is influenced by the localwind velocity, the angle of the wind, the aspect ratioof the building, which face of the building is impingedWhere:Q ws = The combined infiltration rate due towind and stack effect, ft 3 /hrQ w = The infiltration rate due to wind, ft 3 /hrQ s = The infiltration rate due to stack effect,ft 3 /hrWhile recent research has increased our understandingof the basic physical mechanisms of infiltration,it is all but impossible to accurately calculate anythingbut a “worst-case” design value for a particular building.Techniques such as the air-change method, and generalanecdotal findings from the literature are often the mostpractical approaches to the evaluation of infiltration for aparticular building.9.7.2.1 Anecdotal Infiltration FindingsGeneral studies have shown that office buildingshave air exchange rates ranging from 0.10 to 0.6 airchanges per hour with no outdoor intake. To the extentoutdoor air is introduced to the building and it is pressurizedrelative to the local outdoor pressure, the above rateswill be reduced.Infiltration through modern curtain wall constructionis a different situation than that through windows“punched” into walls. Studies of office buildings in theUnited States and Canada have suggested the followingapproximate leakage rates per unit wall area for conditionsof 0.30 inches of pressure (water gauge):Tight ConstructionAverage ConstructionLeaky Construction0.10 CFM/ft 2 of curtainwall area0.30 CFM/ft 2 of curtainwall area0.60 CFM/ft 2 of curtainwall area9.7.2.2 The Air Change MethodWhile it is difficult to quantitatively predict actualinfiltration, it is possible to conservatively predictinfiltration that will give us a sense of “worst-case.” Thisrequires experience and judgment on the part of the engineeror analyst. ASHRAE Fundamentals Manual publishedguidelines for estimating infiltration on the basis of

BUILDING ENVELOPE 239the number of “air changes.” That is, based on the volumeof the space, how many complete changes of air are likelyto occur within the space of an hour?Table 9.10 is a summary of the recommended airchanges originally published by ASHRAE in 1972.These guidelines have been found to be soundover the years and are still widely used by many practicingprofessionals. The values represent a good startingpoint for estimating infiltration. They can be modifiedas required for local conditions, such as wind velocity orexcessive building stack effect.The above values are based on doors and operablewindows that are not weather-stripped. ASHRAE originallyrecommended that the above factors be reducedby 1/3 for weather-stripped windows and doors. Thiswould be a good guideline applicable to modern buildings,which normally have well-sealed windows anddoors. Fully conditioned commercial spaces are slightlypressurized and often do not have operable sash. Thiswill also tend to reduce infiltration.9.7.3 An Infiltration Estimate ExampleAssume a 20' × 30' room with a 10' ceiling and oneexposed perimeter wall with windows. What would theworst-case infiltration rate be?The total air volume is 20 × 30 × 10 = 6,000 ft 3 . Thetable above recommends using 1.0 air changes per hour.However, assuming modern construction techniques, wefollow ASHRAE’s guideline of using 2/3 of the table values.2/3 × 6,000 ft 3 /hr = 4,000 ft 3 /hrThe heat loss due to infiltration for this space, if theindoor temperature was 70°F and the outdoor temperaturewas 25°F would be:q infiltration = 0.019 × 4,000 × (70 – 25)= 3,420 Btu/hr9.8 SUMMARIZING ENVELOPE PERFORMANCEWITH THE BUILDING LOAD COEFFICIENTIn Section 9.1.2 it was shown that building heatloss is proportional to the indoor-outdoor temperaturedifference. Our review of different building components,including infiltration has demonstrated that eachindividual component can be assigned a proportionalityconstant that describes that particular component’sbehavior with respect to the temperature difference imposedacross it. For a wall or window, the proportionalityconstant is the U-factor times the component area, or“UA.” For a slab-on-grade floor, the proportionality constantis the F-factor times the slab perimeter, or “FP.” Forinfiltration, the proportionality constant is 0.019 time theair flow rate in cubic feet per hour. While not attributedto the building envelope, the effect of ventilation mustbe accounted for in the building’s overall temperaturedependentbehavior. The proportionality constant forventilation is the same for infiltration, except that it iscommonly expressed as 1.10 times the ventilation airflow in cubic feet per minute.The instantaneous temperature-dependent performanceof the total building envelope is simply the sumof all the individual component terms. This is sometimesreferred to as the Building Load Coefficient (BLC). The BLCcan be expressed as:BLC = ΣUA + ΣFP + 0.018 Q INF + 1.1 Q VENT (9.12)Where:ΣUA = the sum of all individual component “UA”productsΣFP = the sum of all individual component “FP”productsQ INF = the building infiltration volume flow rate, incubic feet per hourTable 9.10 Recommended air changes due to infiltration.Type of RoomNumber of AirChanges TakingPlace per HourRooms with no windows or exterior doors 1/2Rooms with windows or exterior doors on one side 1Rooms with windows or exterior doors on two sides 1-1/2Rooms with windows or exterior doors on three sides 2Entrance Halls 2

240 ENERGY MANAGEMENT HANDBOOKQ VENT = the building ventilation volume flow rate, incubic feet per minuteFrom the above and Equation (9.1), it follows thatthe total instantaneous building heat loss can be expressedas:q = BLC(T indoor – T outdoor ) (9.13)9.9 THERMAL “WEIGHT”Thermal weight is a qualitative description of theextent to which the building energy consumption occursin “lock-step” with local weather conditions. Thermal“weight is characterized by the mass and specific heat(heat capacity) of the various components which make upthe structure, as well as the unique combination of internalloads and solar exposures which have the potential to offsetthe temperature-dependent heat loss or heat gain of thefacility.Thermally “Light” Buildings are those whose heatingand cooling requirements are proportional to theweather. Thermally “Heavy” Buildings are those Buildingswhose heating and cooling requirements are notproportional to the weather. The “heavier” the building,the less temperature-dependent the building’s energyconsumption appears to be, and the less accuracy that canbe expected from simple temperature-dependent energyconsumption calculation schemes.The concept of thermal weight is an importantone when it comes to determining the energy-savingpotential of envelope improvements. The “heavier” thebuilding, the less savings per square foot of improved envelopecan be expected for the same “UA” improvement.The “lighter” the building, the greater the savings thatcan be expected. A means for characterizing the thermal“weight” of buildings, as well as analyzing the energysavings potential of “light” and “heavy” buildings will betaken up in Section 9.10.9.10 ENVELOPE ANALYSISFOR EXISTING BUILDINGS9.10.1 Degree DaysIn theory, if one wanted to predict the heat lost by abuilding over an extended period of time, Equation 9.12could be solved for each individual hour, taking into accountthe relevant changes of the variables. This is possiblebecause the change in the value BLC with respect to temperatureis not significant and the indoor temperature (T i )is normally controlled to a constant value (such as 70°F inwinter). That being the case, the total energy transfer couldbe predicted by knowing the summation of the individualdeviations of outdoor temperature (T o ) from the indoorcondition (T i ) over an extended period of time.The summation described has come to be known as“Degree-Days” and annual tabulations of Degree-Daysfor various climates are published by NOAA, ASHRAE,and various other public and military organizations.Historically an indoor reference point of 65°F has beenused to account for the fact that even the most poorlyconstructed building is capable of maintaining comfortconditions without heating when the temperature is atleast 65°F.Because of the impracticality of obtaining hour-byhourtemperature data for a wide variety of locations,daily temperature averages are often used to represent24-hour blocks of time. The daily averages are calculatedby taking the average of daily maximum and minimumtemperature recordings, which in turn are converted oDegree Days. Quantitatively this is calculated as:(T reference – T average )nDegree – Days = ———————— (9.14)24Where “n” represents the number of hours in theperiod for which the Degree Days are being reported,and T reference is a reference temperature at which noheating is assumed to occur. Typically, a reference temperatureof 65°F is used. Units of degree hours can beobtained by multiplying the results of any Degree-Daytabulation by 24.Figure 9.23 shows the monthly natural gas consumptionof a metropolitan newspaper building overlaidon a plot of local heating Degree Days.Because of the clear relationship between heatingDegree Days and heating fuel consumption for manybuildings, such as the above, the following formula hasbeen used since the 1930's to predict the future heatingfuel consumption.q × DD × 24E = ——————— (9.15)∆t × H × EffWhere:E = Fuel consumption, in appropriate units,such as Therms natural gasq = The design-day heat loss, Btu/hrDD = The annual heating Degree-Days (usuallyreferenced to 65°F)24 = Converts Degree-Days to Degree-Hours∆t = The design day indoor-outdoor temperaturedifference, °FH = Conversion factor for the type of fuel usedEff = The annualized efficiency of the heatingcombustion process

BUILDING ENVELOPE 241Figure 9.23.Note that if he BLC is substituted for q/∆t, Equation 9.15becomes:E =BLC × DD × 24H × Eff (9.16)Ignoring the conversion terms, we can see that thisequation has the potential to describe the relationship betweenmonthly weather trends and the building heatingfuel consumption shown in Figure 9.23.As building construction techniques have improvedover the years, and more and more heat-producingequipment has found its way into commercial and evenresidential buildings, a variety of correction factors havebeen introduced to accommodate these influences. TheDegree-Day method, in its simpler form, is not presentlyconsidered a precise method of estimating future buildingenergy consumption. However it is useful for indicatingthe severity of the heating season for a region and itcan prove to be a very powerful tool in the analysis ofexisting buildings (i.e. ones whose energy consumptionis already known).It is possible to determine an effective BuildingLoad Coefficient for a building by analyzing the monthlyfuel consumption and corresponding monthly DegreeDays using the technique of linear regression.follow the weather in an indirect fashion.If we could draw a line that represented the closest“fit” to the data, it might look like the Figure 9.25.As a general rule, the simpler the building and itsheating (or cooling) system, the better the correlation.Buildings that are not mechanically cooled will show aneven better correlation. To put it in terms of our earlierdiscussion regarding thermal “weight,” the “lighter” thebuilding, the better the correlation between buildingenergy consumption and Degree Days. To the extent thisline fits the data, it gives us two important pieces of informationdiscussed in Section 9.1.1.• The Monthly Base Consumption• The Relationship Between the Weather and EnergyConsumption Beyond the Monthly Base, which wehave been calling the Building Load Coefficient.Thermsooooooooooo9.10.2 Analyzing Utility BillingsIf we were to plot the monthly natural gas consumptionand corresponding Degree-Days for a building,the result might look something like Figure 9.24. It canbe seen that the monthly energy consumption appears to00Heating Degree DaysFigure 9.24.

242 ENERGY MANAGEMENT HANDBOOKThermsoooooooooooFigure 9.26.00 Heating Degree DaysFigure 9.25.9.10.3 Using Linear Regression for Envelope AnalysisThe basic idea behind linear regression is that anyphysical relationship where the value of a result is linearlydependent on the value of an independent quantitycan be described by the relationship:y = mx + b (9-17)Figure 9.27.where “y” is the dependent variable, x is the independentvariable, m is the slope of the line that describes the relationship,and b is the y-intercept, which is the value of ywhen x = 0.In the case of monthly building energy consumption,the monthly Degree Days, can be taken as the independentvariable, and the monthly fuel consumption asthe dependent variable. The slope of the line that relatesthe monthly consumption to monthly Degree Days isthe Building Load Coefficient (BLC). The monthly basefuel consumption is the intercept on the fuel axis (or themonthly fuel consumption when there are no heatingDegree Days). The fuel consumption for any month canbe determined by multiplying the monthly Degree Daysby the Building Load Coefficient and adding it to themonthly base fuel consumption.Fuel Consumption = BLC × DD + base energy (9.18)The value of the above is that once the equation isdetermined, projected fuel savings can be made by recalculatingthe monthly fuel consumption using a “conservation-modified”BLC.The determination of the Building Load Coefficientis useful in the analysis of building envelope for a numberof reasons.• Often the information necessary for the analysis ofbuilding envelope components is not available.• If envelope information is readily available and theBuilding Load Coefficient has been determined byanalysis of all the individual components, the regressionderived BLC gives us a “real-world” checkon the calculated BLC.• Linear regression also gives feedback in terms ofhow “good” the relationship is between fuel consumptionand local climate, giving a good indicationas to the thermal “weight” of the building,discussed in Section 9.9.The following is a brief overview of the regressionprocedure as it can be applied to corresponding pairs ofmonthly Degree-Day and fuel consumption data. TheBuilding Load Coefficient which describes the actualperformance of the building can be determined with thefollowing relationship:BLC = n∑D iE i – n∑D i ∑E in∑D 2 2i – ∑D i (9.19)Where:n = the number of degree-day/fuel consumptionpairsD i = the degree days accumulated for an individualmonthE i = the energy or fuel consumed for an individualmonth. The monthly base fuel consumptioncan be calculated as:

BUILDING ENVELOPE 243Base = ∑E in– BLC ∑D in (9.20)The units of the BLC will be terms of the fuel unitsper degree day. The BLC will be most valuable for additionalanalysis if it is converted to units of Btu/(hr-°F) orWatts/°C.The following example demonstrates how the abovemight be used to evaluate the potential for envelope improvementin a building.ExampleA proposal has been made to replace 5,000 ft 2 ofwindows in an electrically heated building. The buildingenvelope has been analyzed on a component-by-componentbasis and found to have an overall analytical BLC= 15,400 Btu/(Hr-°F). Of this total, 6,000 Btu/(Hr-°F) isattributable to the single pane windows, which are assumedto have a U-factor of 1.2 Btu/(hr-ft 2 -°F). A linearregression is performed on the electric utility data andavailable monthly Degree-Day data (65°F reference). TheBLC is found to be 83.96 kWh/degree-day, which can beconverted to more convenient units by the following:kWh Btu 1 DayBLC = 83.96 ———— × 3413 ——— × ————°F – Days kWh 24 hours= 11,940 Btu/(hr–°F)What is the reason for the discrepancy between theBLC calculated component-by-component and the BLCderived from linear regression of the building’s performance?The explanation comes back to the concept of thermalweight. Remember, the more thermally “heavy” thebuilding, the more independent of outdoor conditions itis. The difference between the value above and the calculatedBLC is due in part to internal heat gains in the buildingthat offset some of the heating that would normally berequired.This has significant consequences for envelope retrofitprojects. If the potential savings of a window conservationretrofit is calculated on the basis of the direct “UA”improvement, the savings will be overstated in a buildingsuch as the above. A more conservative (and realistic) estimateof savings can be made by de-rating the theoreticalUA improvement by the ratio of the regression UA to thecalculated UA.9.10.4 Evaluating theUsefulness of the Regression ResultsThe Correlation Coefficient, R, is a useful term thatdescribes how well the derived linear equation accountsfor the variation in the monthly fuel consumption of thebuilding. The correlation coefficient is calculated as follows:R = BLC∑D 2 i – ∑D 2in2∑E 2 i – ∑E in (9.21)The square of the Correlation Coefficient, R, 2 providesan estimate of the number of independent values(fuel consumption) whose variation is explained by theregression relationship. For example, an R 2 value of 0.75tells us that 75% of the monthly fuel consumption datapoints evaluated can be accounted for by the linear equationgiven by the regression analysis. As a general rule ofthumb, an R 2 value of 0.80 or above describes a thermally“light” building, and a building whose fuel consumptionindicates an R 2 value significantly less than 0.80 can beconsidered a thermally “heavy” building.9.10.5 Improving the Accuracy of theBuilding Load Coefficient EstimateAs discussed elsewhere in Section 9.9, one of thereasons a building might be classified as thermally“heavy” would be the presence of significant internalheat gains, such as lighting, equipment and people. To theextent these gains occur in the perimeter of the building,they will tend to offset the heat loss predicted using thetools described previously in this chapter. The greater theinternal heat, relative to the overall building temperaturedependence (BLC), the lower the outside temperature hasto be before heating is required.The so-called “building balance temperature” is thetheoretical outdoor temperature where the total buildingheat loss is equal to the internal gain. We discussedearlier that the basis of most published Degree-Day datais a reference (or balance) temperature of 65°F. If data ofthis sort are used for analysis, the lower the actual buildingbalance temperature than 65°F, the more error will beintroduced into the analysis. This accounts for the low R 2values encountered in thermally heavy buildings.While most degree data are published with 65°F asthe reference point, it is possible to find compiled sourceswith 55°F and even 45°F reference temperatures. Theuse of this monthly data can improve the accuracy of theanalysis significantly, if the appropriate data are utilized.ExampleA regression is performed for a library buildingwith electric resistance heating. The results of the analysisindicate an R 2 of 0.457. In other words, only 46% of

244 ENERGY MANAGEMENT HANDBOOKthe variation in monthly electricity usage is explained bythe regression. The figure below is a plot of the monthlyelectric consumption versus Degree Days calculated for a65°F reference.Figure 9.29.Figure 9.28.The plot following shows the result of the regressionrun using monthly Degree Day referenced to 55°F. Noticein the figure that there are less data points showing. Thisis explained by the observation that all the days with averagetemperatures greater than 55°F are not included in thedata set. The resulting R 2 has increased to 0.656.The next figure shows the result of the regressionrun using monthly Degree Day referenced to 45°F. Theresulting R 2 has increased to 0.884. Notice how few datapoints are left. Normally in statistical analysis great emphasisis placed on the importance of having an adequatedata sample for the results to be meaningful. While wewould like to have as many points as possible in our analysis,the concern is not so great with this type of analysis.Statistical analysis usually concerns itself with whetherthere is a relationship to be found. The type of analysiswe are advocating here assumes that the relationship doesexist, and we are merely attempting to discover the mostaccurate form of the relationship. Another way of sayingthis is that we are using a trial-and-error technique to discoverthe building balance point.The final figure shows the result of the regressionrun with monthly degree days referenced to 40°F. Theresulting R 2 has decreased to 0.535.We have learned from the above that the balancepoint of the building analyzed is somewhere between40° and 50°F, with 45°F probably being pretty close.Additional iterations can be made if more precision isdesired, and if the referenced Degree Day data are available.If the published Degree Day data desired are notavailable, it is a relatively simple matter to construct itFigure 9.30.Figure 9.31.from the average daily temperatures published for thelocal climate. If this is done with a spreadsheet, a tablecan be constructed in such a way that the degree days,BLC, monthly base and R 2 values are all linked to an assumedbalance temperature. This balance temperature ismodified iteratively until R 2 is maximized. Of course any

BUILDING ENVELOPE 245energy saving calculations should consistently use boththe derived BLC and degree days that accompany thecorrelation with the highest R 2 .9.11 ENVELOPE ANALYSIS FOR NEW BUILDINGSEnvelope analysis of new buildings offers a differentkind of challenge to the analyst. While new constructionoffers much greater opportunity for economicalimprovements to the envelope, the analysis is muchmore open-ended than that for existing buildings. Inother words, we do not know the monthly energy consumption(the answer), and are left totally at th e mercyof tools designed to assist us in predicting the future energyconsumption of an as-yet unconstructed building,the necessary details of which constitute thousands ofunknowns.We say the above to emphasize the extreme difficultyof the task, and the inadvisability of harboring any illusionsthat this can be done reliably, short of using hourby-hourcomputer simulation techniques. Even with thepowerful programs currently available, accuracy’s nobetter than 10 to 20 percent should be expected.Nevertheless, we are often called upon to quantifythe benefits of using one envelope strategy in place ofanother. We are able to quantify the difference in annualenergy consumption between two options much morereliably than the absolute consumption. A number oftechniques have been developed to assist in this process.One of the more popular and useful tools is the TemperatureBin Method.9.11.1 The Temperature Bin MethodThe Temperature Bin Method requires that instantaneousenergy calculations be performed at manydifferent outdoor temperature conditions, with theresults multiplied by the number of hours expected ateach temperature condition. The “bins” referred to representthe number of hours associated with groups oftemperatures, and are compiled in 5°F increments. Thehour tabulations are available in annual, monthly andsometimes 8-hour shift totals. All hourly occurrences ina bin are assumed to take place at the bin “center-temperature.”For example, it is assumed that 42°F quantitativelyrepresents the 40-44°F bin.The basic methodology of this method requirescalculating the unique heat loss at each bin by multiplyingthe BLC by the difference between the indoortemperature assumed and the center-temperature foreach respective bin. This result is in turn multiplied bythe number of hours in the bin. The products of all thebins are summed to arrive at the total predicted annualheating energy for the building.The advantage of this “multiple-measure” methodover a “single-measure” method, such as the degreeday method, is the ability to accommodate other temperature-dependentphenomena in the analysis. Forexample, the power requirement and capacity of an airto-airheat pump are extremely temperature dependent.This dependency is easily accommodated by the binmethod.However, just as was the case with the Degree-Daymethod, energy savings predicted by the bin methodmay vary significantly from actual, depending on thebuilding balance temperature (building thermal weight).While the balance temperature cannot be predicted for anew building with the techniques previously discussed,it can be estimated with the following equation.T balance = T indoor – q internalBLC (9.22)Where:T balance = The predicted balance temperature, °FT indoor = The assumed indoor conditioned spacetemperature, °Fq internal =BLCThe assumed internal heat gain in buildingtemperature control zones adjacentto the envelope, Btu/hr= The Building Load Coefficient, BLC,Btu/(hr-°F)More accurate energy predictions will result byomitting calculations for bins whose center temperatureexceeds the assumed balance temperature. For example,no calculation should be performed for the 50-54°F bin,if the predicted balance temperature is 50°F. The centertemperature of 47°F for the 45-49°F bin indicates that a3°F temperature difference is appropriate for that bin(50- 47°F).A complete description of the bin method can befound in the ASHRAE Handbook of Fundamentals.9.12 ENVELOPE STANDARDSFOR NEW & EXISTING CONSTRUCTIONThe ASHRAE/IESNA Standard 90.1, Energy Standard forBuildings Except Low-Rise Residential Buildings is the defactostandard for commercial building efficiency and isthe basis of most local energy codes.

246 ENERGY MANAGEMENT HANDBOOK9.12.1 ASHRAE 90.1 Compliance RequirementsCompliance with the Standard may be demonstratedutilizing the Prescriptive Approach or by performingcalculations utilizing the Building Envelope Tradeoff Option.9.12.1.1 The Prescriptive ApproachComponent criteria necessary for local compliancewith the Prescriptive Approach are tabulated in tables for 26separate climate zones, with the appropriate Climate Zoneselected based on the local heating and cooling degreedays.One of the features of the new standard is the inclusionof many precalculated building components in theAppendix. A variety of common (and some uncommon)envelope assemblies can be referenced and utilized fordemonstrating compliance with the Prescriptive Approach.In most cases, this will mean that no calculations will berequired by the designer to demonstrate compliance.9.12.1.2 The Building Envelope Tradeoff OptionThe Building Envelope Tradeoff Option provides morecompliance flexibility than might be found in the PrescriptiveApproach. This option requires the designer todemonstrate that the proposed building envelope resultsin an Envelope Performance Factor that is lower than thebudgeted one for the project. Because of the impact of themechanical and lighting systems on heating and coolingenergy consumption, the Building Envelope Tradeoff Optionrequires information for the mechanical and lightingsystems, as well as for the envelope components. Due tothe complexity of this calculation, the ENVSTD EnvelopeTradeoff Software, first introduced in 1989, will be availableto automate the tradeoff calculations as well as consolidatethe required reference data.9.13 SUMMARYWhile the above discussion of envelope componentshas emphasized the information needed to performrudimentary heat loss calculations, you’ll find that themore you understand these basics, the more you begin tounderstand what makes an efficient building envelope.This same understanding will also guide you in decidinghow to prioritize envelope improvement projects in existingbuildings.9.14 ADDITIONAL READINGAs you can see from this brief introduction, thebest source of comprehensive information on buildingenvelope issues is the ASHRAE Handbook of Fundamentals.You are encouraged to continue your study of buildingenvelope by reading the following chapters in the 2001ASHRAE Handbook of Fundamentals.Chapter Topic23,24 Thermal Insulation and Vapor Retarders25 Thermal and Water Vapor TransmissionData26 Ventilation and Infiltration27 Climatic Design Information28 Residential Cooling and Heating LoadCalculations29 Nonresidential Cooling and Heating LoadCalculations30 Fenestration9.14.1 References1. American Society of Heating, Refrigerating, and Air ConditioningEngineers, Inc., Heat Transmission Coefficients forWalls, Roofs, Cei lings, and Floors, 1993.2. ASHRAE Handbook of Fundamentals, American Societyof Heating, Refrigerating and Air Conditioning Engineers,Inc., Atlanta, GA, 1993.3. ASHRAE Standard 901.-1989, American Society of Heating,Refrigerating, and Air Conditioning Engineers, Inc.,Atlanta, GA, 1999.4. Brown, W.C., Heat-Transmission Tests on Sheet Steel Walls,ASHRAE Transactions, 1986, Vol. 92, part 2B.5. Elder, Keith E., Metal Buildings: A Thermal PerformanceCompendium, Bonneville Power Administration, ElectricIdeas Clearinghouse, August 1994.6. Elder, Keith E., Metal Elements in the Building Envelope.A Practitioner’s Guide, Bonneville Power Administration,Electric Ideas Clearinghouse, October 1993.7. Johannesson, Gudni, Thermal Bridges in Sheet Metal Construction,Division of Building Technology, Lund Instituteof Technology, Lund, Sweden, Report TVHB-3007, 198l .8. Loss, W., Metal Buildings Study: Performance of Materials andField Validation, Brookhaven National Laboratory, December1987.9. Washington State Energy Code, Washington Associationof Building Officials April 1, 1994

CHAPTER 10HVAC SYSTEMSERIC NEIL ANGEVINE, P.E.Associate ProfessorSchool of ArchitectureOklahoma State UniversityStillwater, OklahomaJENNIFER S. FAIR, P.E.Vice PresidentPSA Consulting EngineersOklahoma City, Oklahoma10.1 INTRODUCTIONThe mechanical heating or cooling load in a buildingis dependent upon the various heat gains and lossesexperienced by the building including solar and internalheat gains and heat gains or losses due to transmissionthrough the building envelope and infiltration (or ventilation)of outside air. The primary purpose of the heating,ventilating, and air-conditioning ( HVAC) system ina building is to regulate the dry-bulb air temperature,humidity and air quality by adding or removing heatenergy. Due to the nature of the energy forces whichplay upon the building and the various types of mechanicalsystems which can be used in non-residentialbuildings, there is very little relationship between theheating or cooling load and the energy consumed bythe HVAC system.This chapter outlines the reasons why energyis consumed and wasted in HVAC systems for nonresidentialbuildings. These reasons fall into a varietyof categories, including energy conversion technologies,system type selection, the use or misuse of outside air,and control strategies. Following a review of the appropriateconcerns to be addressed in analyzing an existingHVAC system, the chapter discusses the aspects ofhuman thermal comfort. Succeeding sections deal withHVAC system types, energy conservation opportunitiesand domestic hot water systems.10.2 SURVEYING EXISTING CONDITIONSAs presented in Chapter 3, the first stage of anyeffective energy management program is an energy auditof the facility in question. In surveying the HVACsystem(s) in a facility, the first step is to find out whatyou have to work with: what equipment and controlsystems exist. It is usually beneficial to divide the HVACsystems into two categories: equipment and systemswhich provide heating and cooling, and equipmentand systems which provide ventilation. It is essentialto fully document the type and status of all equipmentfrom major components including boilers, chillers, coolingtowers and air-handling units to the various controlsystems: thermostats, valves and gauges, whether automatedor manual; in order to later determine what elementscan be replaced or improved to realize a savingin energy consumed by the system.The second step is to determine how the systemis operating. This requires that someone measure theoperating parameters to determine whether the systemactually operates as it was specified to operate. Determinethe system efficiency under realistic conditions.This may be significantly different from the theoretical,or full-load efficiency. Determine how the system is operated.What are the hours of operation? Are changes insystem controls manual or automatic? Find out how thesystem is actually operated, which may differ from howthe system was designed to be operated. It is best to talkto operators and/or users of the system who know a lotmore about how the system operates than the engineersor managers.If the system is no longer operating at designconditions, it is extremely useful to determine whatfactors are responsible for the change. Potential causesof operational changes are modifications in the buildingor system and lapses in maintenance. Have there beenstructural or architectural changes to the building withoutcorresponding changes to the HVAC system? Havethere been changes in building operations? Is the systemstill properly balanced? Has routine maintenance been247

248 ENERGY MANAGEMENT HANDBOOKperformed? Has scheduled preventative maintenancebeen performed?Finally, it is useful to determine whether the systemcan or should be restored to its initial design conditions.If practical, it may be beneficial to carry out the neededmaintenance before proceeding to analyze the system forfurther improvements. However, some older systems areso obviously inefficient that bringing them back to originaldesign parameters is not worth the time or expense.Before continuing with an analysis of the system, itis also useful to determine future plans for the buildingand the HVAC system which can seriously effect theenergy efficiency of system operation. Are there plansto remodel the building or parts of the building? Howextensive are proposed changes? Are changes in buildingoperations planned?Document everything. Only when you have a fullrecord of what the system consists of, how it is operatingand how it is operated, and what changes have beenmade and will be made in the future, can you properlyevaluate the benefit of energy conservation techniqueswhich may be applicable to a particular building system.10.3 HUMAN THERMAL COMFORTThe ultimate objective of any heating, cooling andventilating system is typically to maximize human thermalcomfort. Due to the prevalence of simple thermostatcontrol systems for residential and small-scale commercialHVAC systems, it is often believed that humanthermal comfort is a function solely, or at least primarily,of air temperature. But this is not the case.Human thermal comfort is actually maximized byestablishing a heat balance between the occupant andhis or her environment. Since the body can exchangeheat energy with its environment by conduction, convectionand radiation, it is necessary to look at the factorswhich affect these heat transfer processes along withthe body’s ability to cool itself by the evaporation ofperspiration.All living creatures generate heat by burning food,a process known as metabolism. Only 20 percent of foodenergy is converted into useful work; the remainder mustbe dissipated as heat. This helps explain why we remaincomfortable in an environment substantially cooler thanour internal temperature of nearly 100°F (37°C).In addition to air temperature, humidity, air motionand the surface temperature of surroundings allhave a significant influence on the rate at which thehuman body can dissipate heat. At temperatures belowabout 80°F (27°C) most of the body’s heat loss is by convectionand radiation. Convection is affected mostly byair temperature, but it is also strongly influenced by airvelocity. Radiation is primarily a function of the relativesurface temperature of the body and its surroundings.Heat transfer by conduction is negligible, since we makeminimal physical contact with our surroundings whichis not insulated by clothing.At temperatures above 80°F (27°C) the primaryheat loss mechanism is evaporation. The rate of evaporationis dependent on the temperature and humidity ofthe air, as well as the velocity of air which passes overthe body carrying away evaporated moisture.In addition to these environmental factors, therate of heat loss by all means is affected by the amountof clothing, which acts as thermal insulation. Similarly,the amount of heat which must be dissipated is stronglyinfluenced by activity level. Thus, the degree of thermalcomfort achieved is a function of air temperature,humidity, air velocity, the temperature of surroundingsurfaces, the level of activity, and the amount of clothingworn.In general, when environmental conditions arecool the most important determinant of human thermalcomfort is the radiant temperature of the surroundings.In fact, a five degree increase in the mean-radiant temperatureof the surroundings can offset a seven degreereduction in air temperature.When conditions are warm, air velocity and humidityare most important. It is not by accident that thenatural response to being too warm is to increase airmotion. Similarly, a reduction in humidity will offset anincrease in air temperature., although it is usually necessaryto limit relative humidity to no more than 70% insummer and no less than 20% in winter.There is, of course, a human response to air temperature,but it is severely influenced by these otherfactors. The most noticeable comfort response to airtemperature is the reaction to drift, the change of temperatureover time. A temperature drift of more thanone degree Fahrenheit per hour (0.5°C/hr) will resultin discomfort under otherwise comfortable conditions.Temperature stratification can also cause discomfort,and temperature variation within the occupied space ofa building should not be allowed to vary by more than5 degrees F (3°C).Modern control systems for HVAC systems can respondto more than just the air temperature. One optionwhich has been around for a long time is the humidistat,which senses indoor humidity levels and controls humidification.However, state-of-the-art control systems canmeasure operative temperature, which is the air tempera-

HVAC SYSTEMS 249ture equivalent to that affected by radiation and convectionconditions of an actual environment.* Another usefulconstruct is that of effective temperature, which is a computedtemperature that includes the effects of humidityand radiation. †The location and type of air distribution devicesplay a role equal in importance to that of effectivecontrols in achieving thermal comfort. The discomfortcaused by stratification can be reduced or eliminated byproper distribution of air within the space.In general terms, thermal comfort can be achievedat air temperatures between about 68°F and 80°F, andrelative humidities between 20% and 70%, under varyingair velocities and radiant surface temperatures.Figure 10.1 shows the generalized “comfort zone” ofdry bulb temperatures and humidities plotted on thepsychrometric chart. However it should not be forgottenthat human thermal comfort is a complex functionof temperature, humidity, air motion, thermal radiationfrom local surroundings, activity level and amount ofclothing.Figure 10.1 Comfort zone.*Operative temperature is technically defined as the uniform temperatureof an imaginary enclosure with which an individual experiencesthe same heat by radiation and convection as in the actual environment.†Effective temperature is an empirical index which attempts to combinethe effect of dry bulb temperature, humidity and air motion intoa single figure related to the sensation of thermal comfort at 50%relative humidity in still air.10.4 HVAC SYSTEM TYPESThe energy efficiency of systems used to heat andcool buildings varies widely but is generally a functionof the details of the system organization. On themost simplistic level the amount of energy consumedis a function of the source of heating or cooling energy,the amount of energy consumed in distribution, andwhether the working fluid is simultaneously heated andcooled. System efficiency is also highly dependent uponthe directness of control, which can sometimes overcomesystem inefficiency.HVAC system types can be typically classifiedaccording to their energy efficiency as highly efficient,moderately efficient or generally inefficient. This terminologyindicates only the comparative energy consumptionof typical systems when compared to eachother. Using these terms, those system types classifiedas generally inefficient will result in high energy billsfor the building in which they are installed, while anequivalent building with a system classified as highlyefficient will usually have lower energy bills. However,it is important to recognize that there is a wide rangeof efficiencies within each category, and that a specificenergy-efficient example of a typically inefficient systemmight have lower energy bills than the least efficientexample of a moderately, or even highly efficient typeof system.Figure 10.2 shows the relative efficiency of themore commonly used types of HVAC systems discussedbelow. The range of actual energy consumption for eachsystem type is a function of other design variables includinghow the system is configured and installed ina particular building as well as how it is controlled andoperated.To maximize the efficiency of any type of HVACsystem, it is important to select efficient equipment,minimize the energy consumed in distribution andavoid simultaneous heating and cooling of the workingfluid. It is equally important that the control systemdirectly control the variable parameters of the system.Most HVAC systems include zones, which areareas within the building which may have differentclimatic and/or internal thermal loads and for whichheat can be supplied or extracted independent of otherzones.The four-volume ASHRAE Handbook, publishedsequentially in a four-year cycle by the American Societyof Heating, Refrigerating and Air-ConditioningEngineers, Inc., provides the most comprehensive andauthoritative reference on HVAC systems for buildings.The reader is specifically referred to the ASHRAE

250 ENERGY MANAGEMENT HANDBOOKLeast EfficientMultizoneMost EfficientSingle ZoneVariable Volume(CAV) Dual Duct (VAV)(CAV) Reheat (VAV)InductionFan CoilsUnit VentilatorsIndividual UnitsHeat PumpsConvectorsFigure 10.2 Relative energy efficiency of air-conditioningsystems.Handbook: HVAC Systems and Equipment for additionalinformation on any of the systems described below. 3Additional information regarding the most appropriatetypes of HVAC system for specific applications can befound in the ASHRAE Handbook: HVAC Applications. 210.4.1 All-Air SystemsThe most common types of systems for heating andcooling buildings are those which moderate the air temperatureof the occupied space by providing a supply ofheated or cooled air from a central source via a networkof air ducts. These systems, referred to as all-air systems,increase or decrease the space temperature by alteringeither the volume or temperature of the air supplied.Recalling that the most important determinant ofthermal comfort in a warm environment is air velocity,most buildings which require cooling employ all-airsystems. Consequently, all-air systems are the systemof choice when cooling is required. All-air systems alsoprovide the best control of outside fresh air, air quality,and humidity control. An added benefit of forced airsystems is that they can often use outside air for coolinginterior spaces while providing heating for perimeterspaces. (See §10.5.5; Economizers) The advantages ofall-air systems are offset somewhat by the energy consumedin distribution.All-air systems tend to be selected when comfortcooling is important and for thermally heavy buildingswhich have significant internal cooling loads which coincidewith heating loads imposed by heat loss throughthe building envelope.The components of an all-air HVAC system includean air-handling unit (AHU) which includes a fan, coilswhich heat and/or cool the air passing through it, filtersto clean the air, and often elements to humidify the air.Dehumidification, when required, is accomplished bycooling the air below the dew-point temperature. Theconditioned air from the AHU is supplied to the occupiedspaces by a network of supply-air ducts and air isreturned from conditioned spaces by a parallel networkof return-air ducts. (Sometimes the open plenum abovea suspended ceiling is used as part of the return-airpath.) The AHU and its duct system also includes aduct which supplies fresh outside air to the AHU andone which can exhaust some or all of the return air tothe outside. Figure 10.3 depicts the general arrangementof components in an all-air HVAC system.Single Duct SystemsThe majority of all-air HVAC systems employ asingle network of supply air ducts which provide acontinuous supply of either warmed or cooled air to theoccupied areas of the building.Single Zone—The single duct, single-zone systemis the simplest of the all-air HVAC systems. It is one ofthe most energy-efficient systems as well as one of theleast expensive to install. It uses a minimum of distributionenergy,* since equipment is typically located withinor immediately adjacent to the area which it conditions.The system is directly controlled by a thermostat whichturns the AHU on and off as required by the spacetemperature. The system shown in Figure 10.3 is a singlezone system.Single zone systems can provide either heating orcooling, but provide supply air at the same volume andtemperature to the entire zone which they serve. Thislimits their applicability to large open areas with fewwindows and uniform heating and cooling loads. Typicalapplications are department stores, factory spaces,arenas and exhibit halls, and auditoriums.Variable Air Volume—The variable air volume(VAV) HVAC system shown in Figure 10.4 functionsmuch like the single zone system, with the exceptionthat the temperature of individual zones is controlledby a thermostat which regulates the volume of air thatis discharged into the space. This arrangement allows a*Distribution energy includes all of the energy used to move heatwithin the system by fans and pumps. Distribution energy is typicallyelectrical energy.

HVAC SYSTEMS 251Figure 10.3 Elements of an all-air air-conditioning system.high degree of local temperature control at a moderatecost. Both installation cost and operating costs are onlyslightly greater than the single-zone system.The distribution energy consumed is increasedslightly over that of a single-zone system due to thefriction losses in VAV control devices, as well as the factthat the fan in the AHU must be regulated to balance theoverall air volume requirements of the system. Fan regulationby inlet vanes or outlet dampers forces the fan tooperate at less than its optimum efficiency much of thetime (see Figure 10.5) Consequently a variable speedfan drive is necessary to regulate output volume of thefan. For the system to function properly, it is necessarythat air be supplied at a constant temperature, usuallyabout 55°F (13°C). This requires indirect control of thesupply air temperature with an accompanying decreasein control efficiency.Single-duct VAV systems can often provide limitedheating by varying the amount of constant temperatureair to the space. By reducing the cooling airflow,

252 ENERGY MANAGEMENT HANDBOOKFigure 10.4 Variable air volume system schematic diagram.the space utilizes the lights, people and miscellaneousequipment to maintain the required space temperature.However, if the space requires more heat than can besupplied by internal heat gains, a separate or supplementalheating system must be employed.Single-duct VAV systems are the most versatile andhave become the most widely used of all systems forheating and cooling large buildings. They are appropriatefor almost any application except those requiring ahigh degree of control over humidity or air exchange.Reheat systems—Both the single-zone and singleduct VAV systems can be modified into systems whichprovide simultaneous heating and cooling of multiplezones with the addition of reheat coils for each zone(Figure 10.6). These systems are identical in design to theforegoing systems up to the point where air enters thelocal ductwork for each zone. In a reheat system supplyair passes through a reheat coil which usually containshot water from a boiler. In a less efficient option, anelectrical resistance coil can also be used for reheat. (Seecomments regarding the efficiency of electric resistanceheating in §10.5.6.)A local thermostat in each zone controls the temperatureof the reheat coil, providing excellent control ofthe zone space temperature. Constant air volume (CAV)reheat systems are typically used in situations whichFigure 10.5 Fan power vs. discharge volume characteristics.

HVAC SYSTEMS 253Figure 10.6 Reheat system schematic diagram.require precise control of room temperature and/or humidity,often with constant airflow requirements, suchas laboratories and medical facilities.Both the CAV and VAV reheat systems are inherentlyinefficient, representing the highest level of energyconsumption of the all-air systems. This is due to thefact that energy is consumed to cool the supply air andthen additional energy is consumed to reheat it. In VAVreheat systems, the reheat coil is not activated unlessthe VAV controls are unable to meet local requirementsfor temperature control, and they are therefore somewhatmore energy efficient than CAV reheat systems.Both CAV and VAV reheat systems can also beused with specialized controls to condition spaces withextremely rigid requirements for humidity control, suchas museums, printing plants, textile mills and industrialprocess settings.Multizone—Although commonly misused to indicateany system with thermostatically controlled air-conditioningzones, the multizone system is actually a specifictype of HVAC system which is a variation of thesingle-duct CAV reheat system. In a multizone system,each zone is served by a dedicated supply duct whichconnects it directly to a central air handling unit (Figure10.7).In the most common type of multizone system,the AHU produces warm air at a temperature of about100°F (38°C) as well as cool air at about 55°F (13°C)which are blended with dampers to adjust the supplyair temperature to that called for by zone thermostats.In a variation of this system, a third neutral deck usesoutside air as an economizer to replace warm air in thesummer or cool air in the winter. In another variation,the AHU produces only cool air which is tempered byreheat coils located in the fan room. In this case, the hotdeck may be used as a preheat coil.Multizone systems are among the least energyefficient, sharing the inherent inefficiency of reheat systemssince energy is consumed to simultaneously heatand cool air which is mixed to optimize the supply airtemperature. Since a constant volume of air is suppliedto each zone, blended conditioned air must be suppliedeven when no heating or cooling is required.In addition, multizone systems require a great dealof space for ducts in the proximity of the AHU whichrestricts the number of zones. They also consume a greatdeal of energy in distribution, due to the large quantityof constant volume air required to meet space loads.These drawbacks have made multizone systems nearlyobsolete except in relatively small buildings with only afew zones and short duct runs.

254 ENERGY MANAGEMENT HANDBOOKFigure 10.7 Multizone system schematic diagram.Dual Duct SystemsDual duct systems are similar to the multizoneconcept in that both cool supply air and warm supplyair are produced by a central AHU. But instead of blendingthe air in the fan room, separate hot-air ducts andcold-air ducts run parallel throughout the distributionnetwork and air is mixed at terminal mixing boxes ineach zone (Figure 10.8). The mixing boxes may includean outlet for delivering air directly to the space, or aduct may connect a branch network with air mixed toa common requirement.Dual duct systems require the greatest amount ofspace for distribution ductwork. In order to offset thespatial limitations imposed by this problem, dual ductsystems often employ high velocity/high pressure supplyducts, which reduce the size (and cost) of ductwork,as well as the required floor-to-floor height. Howeverthis option increases the fan energy required for distribution.Their use is usually limited to buildings withvery strict requirements for temperature and or humiditycontrol.Constant Volume Dual Duct—For a long time,the only variation of the dual duct system was a CAVsystem, which functioned very much like the multizonesystem. This system exhibits the greatest energy consumptionof any all-air system. In addition to the energyrequired to mix conditioned air even when no heatingor cooling is required, it requires a great amount ofdistribution energy even when normal pressure and lowair velocities are used. For these reasons it has becomenearly obsolete, being replaced with dual duct VAV orother systems.Dual Duct VAV—Although the dual duct VAV systemlooks very much like its CAV counterpart, it is farmore efficient. Instead of providing a constant volume ofsupply air at all times, the primary method of respondingto thermostatic requirements is through adjustingthe volume of either cool or warm supply air.The properly designed dual duct VAV systemfunctions essentially as two single duct VAV systemsoperating side by side; one for heating and one forcooling. Except when humidity control is required it isusually possible to provide comfort at all temperatureswithout actually mixing the two air streams. Even whenhumidity adjustment is required a good control systemcan minimize the amount of air mixing required.The dual-duct VAV system still requires more distributionenergy and space than most other systems. Thelevel of indirect control which is necessary to produceheated and cooled air also increases energy consumption.Consequently its use should be restricted to applicationswhich benefit from its ability to provide ex-

HVAC SYSTEMS 255Figure 10.8 Dual duct system schematic diagram.ceptional temperature and humidity control and whichdo not require a constant supply of ventilation air.10.4.2 All-Water SystemsAir is not a convenient medium for transportingheat. A cubic foot of air weighs only about 0.074 pounds(0.34 kg) at standard conditions (70°F; 1 atm.). With aspecific heat of about 0.24 Btu/lb°F (0.14 joule/C), onecubic foot can carry less than 0.02 Btu per degree Fahrenheittemperature difference. By comparison, a cubicfoot of water weighs 62.4 pounds and can carry 62.4Btu/ft 3 .Water can be used for transporting heat energyin both heating and cooling systems. It can be heatedin a boiler to a temperature of 140 to 250°F (60-120°C)or cooled by a chiller to 40 to 50°F (4-10°C), and pipedthroughout a building to terminal devices which take inor extract heat energy typically through finned coils.Steam can also be used to transport heat energy.Steam provides most of its energy by releasing thelatent heat of vaporization (about 1000 Btu/lb or 2.3joules/kg). Thus one mass unit of steam provides asmuch heating as fifty units of water which undergo a20°F (11°C) temperature change. However, when watervaporizes, it expands in volume more than 1600 times.Consequently liquid water actually carries more energyper cubic foot than steam and therefore requires the leastspace for piping.All-water distribution systems provide flexiblezoning for comfort heating and cooling and have arelatively low installed cost when compared to all-airsystems. The minimal space required for distributionpiping makes them an excellent choice for retrofitinstallation in existing buildings or in buildings withsignificant spatial constraints. The disadvantage to thesesystems is that since no ventilation air is supplied, allwaterdistribution systems provide little or no controlover air quality or humidity and cannot avail themselvesof some of the energy conservation approaches of all-airsystems.Water distribution piping systems are described interms of the number of pipes which are attached to eachterminal device:One-pipe systems use the least piping by connectingall of the terminal units in a series loop. Since the waterpasses through each terminal in the system, its abilityto heat or cool becomes progressively less at great distancesfrom the boiler or chiller. Thermal control is poorand system efficiency is low.Two-pipe systems provide a supply pipe and a returnpipe to each terminal unit, connected in parallelso that each unit (zone) can draw from the supply as

256 ENERGY MANAGEMENT HANDBOOKneeded. Efficiency and thermal control are both high,but the system cannot provide heating in one zone whilecooling another.Four-pipe systems provide a supply and return pipefor both hot water and chilled water, allowing simultaneousheating and cooling along with relatively highefficiency and excellent thermal control. They are, ofcourse, the most expensive to install, but are still inexpensivecompared to all-air systems.Three-pipe systems employ separate supply pipes forheating and cooling but provide only a single, commonreturn pipe. Mixing the returned hot water, at perhaps140°F (60°C), with the chilled water return, at 55°F(13°C), is highly inefficient and wastes energy requiredto reheat or recool this water. Such systems should beavoided.Radiant HeatingRadiant energy is undoubtedly the oldest methodof centrally heating buildings, dating to the era ofthe Roman Empire. Recalling that the most importantdeterminant of thermal comfort when environmentalconditions are too cool is the radiant temperature ofthe physical surroundings, radiant heating systems areamong the most economical, so long as the means ofproducing heat is efficient.The efficiency of radiant heating is a function primarilyof the temperature, area and emissivity of theheat source and the distance between the radiant sourceand the observer. It is therefore essential that radiantheat sources be located so that they are not obstructedby other objects. Emissivity is an object’s ability to absorband emit thermal radiation, and is primarily relatedto color. Dark objects absorb and emit radiation betterthan light colored objects.There are three categories of radiant heating devices,classified according to the temperature of the sourceof heat. All may employ electric resistance heatingelements, but are more energy-efficient if they employcombustion as a heat source.Low temperature radiant floors employ the entirefloor area as a radiating surface by embedding hot watercoils in the floor. The water temperature is typically lessthan 120°F (50°C). By distributing the heat energy uniformlythough the floor, surface temperature is normallybelow 100°F (40°C).By increasing the temperature of the radiant surfaceits area can be reduced. In medium temperature radiantpanels, hot water circulates through metal panels,heating them to a temperature of about 140°F (60°C).Consequently the panels must be located out-of-reach,usually on the ceiling or on upper walls.High temperature infrared heaters are typicallygas-fired or oil-fired and are discussed below underpackaged systems.Because they are not dependent upon maintaininga static room air temperature, radiant heating systemsprovide excellent thermal comfort and efficiency inspaces subject to large influxes of outside air, such asfactories and warehouses. However they are slow torespond to sudden changes in thermal requirements andmalfunctions may be difficult or awkward to correct.Another drawback to radiant systems is that they promotethe stratification of room air, concentrating warmair near the ceiling.Natural ConvectionThe simplest all-water system is a system of hydronic(hot-water) convectors. In this system hot waterfrom a boiler or steam-operated hot water converteris circulated through a finned tube, usually mountedhorizontally behind a simple metal cover which providesan air inlet opening below the tube and an outletabove. Room air is drawn through the convector bynatural convection where it is warmed in passing overthe finned tube.A variation on the horizontal finned-tube hydronicconvector is the cabinet convector, which occupies lessperimeter space. A cabinet convector would have severalfinned tubes in order to transfer additional heat to the airpassing through it. When this is still insufficient a smallelectric fan can be added, converting the convector to aunit heater. Although an electric resistance element can beused in place of the finned tube, the inefficiency of electricresistance heating should eliminate this option.Hydronic convectors are among the least expensiveheating systems to operate as well as to install. Theiruse is limited, however, to heating only and they do notprovide ventilation, air filtration, nor humidity control.Hydronic convectors and unit heaters may be usedalone in buildings where cooling and mechanical ventilationis not required or to provide heating of perimeterspaces in combination with an all-air cooling system.They are the most suitable type of system for providingheat to control condensation on large expanses of glasson exterior wall systems.Fan-coilsA fan-coil terminal is essentially a small air-handlingunit which serves a single space without a ducteddistribution system; the main difference other than sizeis that fan coils generally do not have outside air andexhaust provisions. One or more independent terminalsare typically located in each room connected to a supply

HVAC SYSTEMS 257of hot and/or chilled water. At each terminal, a fan inthe unit draws room air (sometimes mixed with outsideair) through a filter and blows it across a coil of hot wateror chilled water and back into the room. Condensatewhich forms on the cooling, coil must be collected in adrip pan and removed by a drain (Figure 10.9).Although most fan-coil units are located beneathwindows on exterior walls, they may also be mountedhorizontally at the ceiling, particularly for installationswhere cooling is the primary concern.Technically, a fan-coil unit with an outside air inletis called a unit ventilator. Unit ventilators provide thecapability of using cool outside air during cold weatherto provide free cooling when internal loads exceed theheat lost through the building envelope. See the discussionof economizers, §10.5.5.Fan-coil units and unit ventilators are directlycontrolled by local thermostats, often located withinthe unit, making this system one of the most energyefficient. Drawbacks to their use is a lack of humiditycontrol and the fact that all maintenance must occurwithin the occupied space.Fan-coil units are typically used in buildings whichhave many zones located primarily along exterior walls,such as schools, hotels, apartments and office buildings.They are also an excellent choice for retrofitting air-conditioninginto buildings with low floor-to-floor heights.Although a four-pipe fan-coil system can be used fora thermally heavy building with high internal loads, itFigure 10.9 Fan-coil unit.suffers the drawback that the cooling of interior zonesin warm weather must be carried out through active airconditioning,since there is no supply of fresh (cool) outsideair to provide free cooling separately. They are alsoutilized to control the space temperature in laboratorieswhere constant temperature make-up air is supplied toall spaces.Closed-Loop Heat PumpsIndividual heat pumps (§10.4.4) have a number ofdrawbacks in nonresidential buildings. However, closedloopheat pumps, more accurately called water-to-airheat pumps, offer an efficient option for heating andcooling large buildings. Each room or zone contains awater-source heat pump which can provide heating orcooling, along with air filtration and the dehumidificationassociated with forced-air air-conditioning.The water source for all of the heat pumps in thebuilding circulates in a closed piping loop, connectedto a cooling tower for summer cooling and a boilerfor winter heating. Control valves allow the water tobypass either or both of these elements when they arenot needed (Figure 10.10). The primary energy benefitof closed-loop heat pumps is that heat removed fromoverheated interior spaces is used to provide heat forunderheated perimeter spaces during cold weather.Since the closed-loop heat pump system is an allwater,piped system, distribution energy is low, andsince direct, local control is used in each zone, controlenergy is also minimized, making this system one of themost efficient. Although the typical lack of a fresh-airsupply eliminates the potential for an economizer cycle,the heat recovery potential discussed above more thanmakes up for this drawback.Heat pump systems are expensive to install andmaintenance costs are also high. Careful economicanalysis is necessary to be sure that the energy savingswill be great enough to offset the added installation andmaintenance costs. Closed-loop heat pumps are most applicableto buildings such as hotels which exhibit a widevariety of thermal requirements along with simultaneousheating requirements in perimeter zones and largeinternal loads or chronically overheated areas such askitchens and assembly spaces.10.4.3 Air & Water systems—InductionOnce commonly used in large buildings, inductionsystems employ terminal units installed at the exteriorperimeter of the building, usually under windows. Asmall amount of fresh outside ventilation air is filtered,heated or cooled, and humidified or dehumidified by acentral AHU and distributed throughout the building at

258 ENERGY MANAGEMENT HANDBOOKFigure 10.10 Closed-loop heat pump system schematic diagram.high-velocity by small ducts.In each terminal unit, this primary air is dischargedin such a way that it draws in a much larger volumeof secondary air from the room, which is filtered andpassed through a coil for additional heating or cooling(Figure 10.11). The use of primary air as the motive forceeliminates the need for a fan in the induction unit. Thecooling coil is often deliberately kept at a temperaturegreater than the dew point temperature of the roomair which passes through it, eliminating the need for acondensate drain. Although the standard air-water inductionsystem is a cooling-only system, room terminalscan employ reheat coils to heat perimeter zones.Despite the high pressures and velocities requiredfor the primary air distribution, distribution energy isminimized by the relatively small volume of primary air.But the energy saved in primary air distribution is morethan offset by the energy consumed in the indirect controland distribution of cooling water, making air-waterinduction systems among the least energy efficient.Air-water induction units tend to be noisy and thesystem provides negligible control of humidity. The applicabilityof these systems is limited to buildings withwidely varying cooling or heating loads where humiditycontrol is not necessary, such as office buildings. Concernsabout indoor air quality limits their use as well.10.4.4 Packaged systemsAll of the systems described above may be classifiedas central air-conditioning systems in that theycontain certain central elements, typically including aboiler, chiller and cooling tower. Many large buildingsFigure 10.11 Induction unit.

HVAC SYSTEMS 259provide heating and cooling with distributed systemsof unitary or packaged systems, where each packageis a stand-alone system which provides all of theheating and cooling requirements for the area of thebuilding which it serves. Individual units derive theirenergy from raw energy sources typically limited toelectricity and natural gas.Since large pieces of equipment usually havehigher efficiencies than smaller equipment, it might bethought that packaged systems are inherently inefficientwhen compared to central air-conditioning systems. Yetpackaged systems actually use much less energy. Thereare several reasons for this.First, there is much less energy used in distribution.Fans are much smaller and pumps are essentiallynon-existent. In addition, control of the smaller packagedunits is local and direct. Typically, the unit iseither on or off, which can be a disadvantage when thespace use requires that ventilation air not be turned off.However there are some advantages associated with thiscontrol flexibility. It allows individual thermal controland accurate metering of use. In addition, if equipmentfailure occurs it does not affect the entire building.A third reason for the energy efficiency of packagedsystems involves the schedule of operation. While largeequipment is more efficient overall, it only operates atthis peak efficiency when it is running at full load. Smallpackaged units, due to their on/off operation, run at fullload or not at all. In a central air-conditioning system,the central equipment must run whenever any zone requiresheating or cooling, often far from its peak load,optimum efficiency conditions.A secondary advantage to the use of packagedsystems is the advantage of diversity. The design of alarge central air-conditioning system sometimes requiresthat a compromise be made between the ideal type ofsystem for one part of a building and a different typeof system for another. When packaged systems are employed,parts of a building with significantly differentheating and cooling requirements can be served by differenttypes of equipment. This will always provide improvedthermal comfort, and often results in improvedefficiency as well.Packaged Terminal Air-ConditionersThe most common type of packaged equipmentis the packaged terminal air-conditioner, often called aPTAC or incremental unit, due to the fact that increasesin equipment can be made incrementally. Examples ofPTACs are through-the-wall air-conditioners and singlezonerooftop equipment. Their use is limited to about500 square feet per unit.Individual air-to-air (air-source) heat pumps canalso be installed as a packaged system. A heat pump isessentially a vapor-compression air-conditioner whichcan be reversed to extract heat from the outdoor environmentand discharge it into the occupied space. A significantdrawback to air-source heat pumps is that vaporcompression refrigeration becomes inefficient when theevaporator is forced to extract heat from a source whosetemperature is 30°F (0°C) or below.In large systems, heat pumps can utilize a sourceof circulating water from which to extract heat duringcold weather, so that the evaporator temperature neverapproaches 30°F (0°C). The circulating water would beheated in the coldest weather, and could be cooled bya cooling tower to receive rejected heat during warmweather. These closed-loop heat pumps are discussedunder all-water systems above.Unit HeatersPackaged heating-only units typically utilize electricityor natural gas as their primary source of energy.As discussed in a later section, electricity is the mostexpensive source of heat energy and should be avoided.Natural gas (or liquefied propane) provides a moreeconomical source of heat when used in packaged unitheaters.Fan-forced unit heaters can disperse heat over amuch larger area than packaged air-conditioners. Theycan distribute heat either vertically or horizontally andrespond rapidly to changes in heating requirements.High temperature infrared radiant heaters utilizea gas flame to produce a high-temperature (over 500°F,260°C) source of radiant energy. Although they do notrespond rapidly to changes in heating requirements,they are essentially immune to massive intrusion ofcold outside air. Because they warm room surfaces andphysical objects in the space, thermal comfort returnswithin minutes of an influx of cold air.HVAC systems may be central or distributed;all-air, all-water, or air-water (induction). Each systemtype has advantages and disadvantages, not the leastimportant of which is its energy efficiency. An economicanalysis should be conducted in selecting an HVAC systemtype and in evaluating changes in HVAC systemsin response to energy concerns.10.5 ENERGY CONSERVATION OPPORTUNITIESThe ultimate objective of any energy managementprogram is the identification of energy conservationopportunities (ECOs) which can be implemented to pro-

260 ENERGY MANAGEMENT HANDBOOKduce a cost saving. However, it is important to recognizethat the fundamental purpose of an HVAC system isto provide human thermal comfort, or the equivalentenvironmental conditions for some specific process. It istherefore necessary to examine each ECO in the contextof its effect on indoor air quality, humidity and thermalcomfort standards, air velocity and ventilation requirements,and requirements for air pressurization.10.5.1 Thermal Comfort, Air Quality and AirflowIt is not wise to undertake modifications of anHVAC system to improve energy efficiency withoutconsidering the effects on thermal comfort, air quality,and airflow requirements.Thermal ComfortOne of the most serious errors which can be madein modifying HVAC systems is to equate a change indry bulb air temperature with energy conservation. Itis worthwhile to recall that dry bulb air temperature isnot the most significant determinant of thermal comfortduring either the heating season or cooling season.Since thermal comfort in the cooling season is mostdirectly influenced by air motion cooling energy requirementscan be reduced by increasing airflow and/orair motion in occupied spaces without decreasing drybulb air temperature. During the heating season thermalcomfort is most strongly influenced by radiant heating. Consequently,changes from forced-air heating to radiantheating can improve thermal comfort while decreasingheat energy requirements.The total energy, or enthalpy, associated with achange in environmental conditions includes both sensibleheat and latent heat. Sensible heat is the heat energyrequired to increase dry bulb temperature. The heatenergy associated with a change in moisture content ofair is known as latent heat. (See Figure 10.12)Changes in HVAC system design or operationwhich reduce sensible heating or cooling requirementsmay increase latent heating or cooling energy whichoffsets any energy conservation advantage. This is particularlytrue of economizer cycles (§10.5.5). The use ofcool, but humid outside air in an economizer can actuallyincrease energy consumption if dehumidification requirementsare increased. For this reason, the most reliabletype of economizer control is enthalpy control which preventsthe economizer from operating when latent coolingrequirements exceed the savings in sensible cooling.Air velocity and airflowThe evaluation of energy conservation opportunitiesoften neglects previously established requirementsFigure 10.12 Sensible heat and latent heat.for air velocity and airflow. As discussed above, thevolume of air supplied and its velocity have a profoundinfluence on human thermal comfort. ECOs which reduceairflow can inadvertently decrease thermal comfort. Evenmore important, airflow cannot be reduced below thevolume of outdoor air required by codes for ventilation.The design of an all-air or air-water HVAC systemis much more complex that just providing a supply airduct and thermostat for each space. The completed systemmust be balanced to assure adequate airflow to eachspace, not only to offset the thermal loads, but also toprovide the appropriate pressurization of the space.It is a common practice for supply air to exceedreturn air in selected spaces to create positive pressurization,which minimizes infiltration and preventsthe intrusion of odors and other contaminants fromadjacent spaces. Similarly, negative pressurization canbe achieved by designing exhaust or return airflow toexceed supply air requirements in order to maintain asterile field or to force contaminants to be exhausted.Any alterations in air supply or return requirements upsetthe relationship between supply and return airflowsrequiring that the system be rebalanced.Indoor Air QualityAnother factor which is often neglected in the applicationof ECOs is the effect of system changes on indoorair quality. Indoor air quality requirements are most commonlymet with ventilation and filtration provided byan all-air or air-water HVAC system. When outdoor air

HVAC SYSTEMS 261quantities are altered significantly, the effect on indoor airquality is unknown and must be determined.In a polluted environment, increasing outdoor airvolume, for example with an economizer cycle, either increasesfiltration requirements or results in a deteriorationof indoor air quality. On the other hand, reducing airflowto conditioned spaces due, for example, to a conversion toa variable air volume system, reduces the filtration of recycledindoor air which can likewise produce a reductionin air quality. The concerns regarding indoor air qualityare discussed in more detail in Chapter 17.10.5.2 MaintenanceThe first place to begin in the investigation of anyexisting HVAC system is to review the condition andmaintenance of the system. Inadequate or impropermaintenance is a major cause of system inefficiency. Thesystem may have been designed to be energy-efficientand some ECOs may have been previously implemented,but due to a lack of maintenance the systemno longer functions in an energy-efficient mode.A more comprehensive review of maintenanceprocedures is included in Chapter 14, particularly withregard to boiler maintenance. What follows here a briefsummary of some of the things to watch for in inspectingthe system for inadequate maintenance.The control system (see §10.5.6 and Chapter 22)should be performing its function. Thermostats shouldbe properly calibrated. Time clocks, setback devices,and automatic controls should be intact and operating.Bearings and belts should be in good shape and beltsshould be tight. Bearings and other moving parts shouldbe properly lubricated with the appropriate lubricant.Lubricant selection can actually have a significant effecton system performance and efficiency.Dirt build-up should be periodically removedespecially from heat transfer surfaces, particularly thecondenser coils of air-cooled condensers. Filters shouldbe cleaned or replaced regularly rather than waitingfor them to appear so dirty as to need cleaning. Dirtyfilters increase fan horsepower while reducing airflow.If routine and preventative maintenance has not beenperformed on the system, complete the required maintenancebefore proceeding to determine other applicableECOs.10.5.3 Demand ManagementDemand management (see also Chapter 11) involvesshutting off or deferring the operation of equipmentto minimize the peak electrical load which occursat a given time. This strategy can be extended to shuttingoff or reducing HVAC equipment requirementswhen they are not needed. The most common demandmanagement strategies for HVAC systems are scheduledoperation and night set-back.Operating ScheduleOne of the greatest causes of energy waste is unnecessaryoperation. The most energy-efficient equipmentwill consume excess energy if it runs when it isnot needed. Unnecessary operation tends to be causedby large systems which condition entire buildings orsections of buildings and a lack of control to turn offequipment when it is not needed.When buildings are conditioned by large centralsystems, off-hours use of one area may require theoperation of the entire system to condition a singlespace. Sometimes the installation of a local packagedsystem serving a specialized area can prevent the unnecessaryoperation of a larger system. For example, aradio station located in a large office building installeda packaged air-conditioner to provide cooling for itsstudios which operated 24 hours a day while the officesremained on the building system which operated onlyfrom 7 a.m. until 5 p.m. Despite the fact that the packagedsystem was less efficient than the central buildingsystem, the energy saved was substantial.Exhaust fans, dust collectors, and other smallequipment which serve specific rooms are often leftrunning continuously due solely to the lack of localcontrol to turn them off. Local switching alone isseldom adequate to save energy since it is difficult toget occupants to turn equipment off and sometimesto turn it on. Passive switching or other automaticdevices can eliminate the need to run equipment continuously.For example an exhaust fan for a seldom-occupiedstore room can be controlled to turn on with an occupancysensor or when the lights are turned on and toremain on for a set period of time after the lights areturned off or occupants leave the room. Spring-woundinterval timers are useful for controlling small deviceswhich are turned on manually and then left running.Exhaust fans can be interlocked with the central air-conditioningsystem to shut down when the central systemis turned off.Turning off equipment, particularly central air-conditioningsystems, in buildings which are unoccupiedat night is another obvious energy saver. Simple timeclocks can be used where more sophisticated computercontrolled systems are not used, Even when it is necessaryto keep the system operating to provide minimalcooling during the night or on weekends, exhaust fansand fresh-air ventilation can be shut down when the

262 ENERGY MANAGEMENT HANDBOOKbuilding is essentially unoccupied.Some equipment loads are deferrable in that theycan be turned off for a short interval and then turnedback on with no appreciable impact on operations.Water heating and cooling are two examples, but evenspace heating can be deferred for periods of up to thirtyminutes with no perceptible effect on thermal comfort.During extreme weather outside air can be reduced forshort periods in order to improve heating or cooling efficiency.See the discussion of warm-up and cool-downcycles which follows.Computerized controls can be used to preventlarge pieces of equipment from operating simultaneouslywhich increases peak demand and the expensewhich accompanies increased demand. This is particularlyattractive in large systems of distributed packagedequipment.With adequate storage both heat and “cool” canbe stored in order to take advantage of off-peak utilityrates and to reduce instantaneous heating or coolingrequirements during peak periods. See the discussionof thermal storage in section 10.6.4 and in chapter 19.Night Set-backThe principle of night set-back is to reduce theamount of conditioning provided at night by allowingthe interior temperature to drift naturally to a marginaltemperature during the night and then to recondition itto normal conditions in the morning. With the possibleexception of electric heat pumps, night set-back willalways save energy. The energy saved can be computedfrom the equation:E = [(A per × U) + q vent ] ∆T × hr (10.1)whereE =A per = surface area of perimeter envelope,ft 2U = effective U-value of thermal envelope,Btu/hr ft 2 °Fq vent = ventilation load, Btu/hr°F∆T = night setback temperature difference,degrees Fhr = heating season unoccupied hoursIn a temperate climate this can amount to a savingof 40 percent of the envelope heating energy for abuilding with little thermal mass which is occupied onlyten hours per day. However the amount of energy savedwill not always be large and local utility rate structuresmay be such that the actual cost of the energy saved isless than the cost of the energy to recondition the buildingin the morning.For example, many electric utility companies offertime-of-day rates in which energy used during thenight is charged at a rate much lower than that chargedfor electricity consumed during the day. If an air-conditionedbuilding allowed to heat up during the nightmust be recooled at the day rate in the morning, nightset-back could end up costing more than cooling thebuilding all night.Similarly, the more energy-efficient the building isthe smaller the energy saving will be from night set-backand the less attractive this option will be from a financialstandpoint. Night set-back is marginal at best for cooling,and will be only slightly better for heating unless thebuilding has little thermal mass and significant infiltrationheat loss.Where night set-back is a viable option, a secondaryquestion is whether to turn the system off or simplyturn the thermostat setpoint down (or up). This is nota question which can be answered simply. The answeris dependent on the length of time the system will besetback and how much temperature variation will occurif the system is turned off. The question of whether it iseven advisable to turn the system off is also influencedby humidification and pressurization requirements, airquality concerns, etc. See §10.5.1.Warm-up and Cool-downWhen a building is operated with night set-back,it is not necessary to condition the building until thelast person leaves in the evening. But the system mustbe restarted in the morning so that optimum conditionsare reached before the beginning of the business day. Itis not difficult to determine an appropriate time to letthe system begin to drift in the afternoon, but morningwarm-up (or cool-down) can require substantial timeand energy.One of the best approaches to re-conditioning a buildingafter night set-back is with the use of an optimum-timestart device. This microprocessor-based thermostat comparesthe outside and inside temperatures along with thedesired setpoint during the operating cycle. It determineshow long it will take to re-condition the building to thesetpoint based on previous data and turns the system on atthe appropriate time to reach the setpoint temperature justin time. Since time clocks must be set to turn the system onwith adequate time to re-condition the building on eventhe coldest morning, the optimum-time start device cansave more than half of the warm-up time on days when theoutdoor ambient temperature is moderate.A second method of saving energy during warmupor cool down is with a warm-up/cool-down cycle.

HVAC SYSTEMS 263During the warm-up/cool-down cycle the system recirculatesbuilding return air until a temperature withinone or two degrees of the setpoint is reached, saving theenergy which would be required to heat or cool outsideventilation air. Since outside ventilation air is providedprimarily to meet human fresh-air needs there is noharm in reducing or eliminating it during warm-up orcool-down. However modem building codes prohibiteliminating all ventilation air unless the building is trulyunoccupied.Warm-up/cool-down cycles provide a secondaryadvantage in times of extreme weather. Should the heatingand cooling equipment be unable to maintain thesetpoint temperature due to a lack of capacity to conditionthe ventilation air because of its extreme temperature, thewarm-up/cool-down cycle could close the outside-airdamper and reduce the volume of outdoor air until interiorconditions were restored to near the setpoint. Since thesystem would run continuously in either case, this proceduretechnically doesn’t save energy but it does maintainthermal comfort without expending additional energy.10.5.4 Heat RecoveryHeat recovery systems are used during the heatingseason to extract waste heat and humidity from exhaustair which is used to preheat cold fresh air from outside. Inwarm weather, they may be used to extract heat and humidityfrom warm fresh air and dispel it into the exhaustair stream. Their use is particularly appropriate to buildingswith high outside air requirements. Heat recoverysystems are discussed in more detail in Chapter 8.One special type of heat recovery, dubbed bootstrapheating, recovers waste heat from warm condenserwater to provide nearly free heat to meet perimeterheating needs. This is most applicable to buildings inwhich interior zone cooling exceeds the requirementsfor perimeter heating. Care must be taken in utilizingbootstrap heating to avoid reducing the condenser watertemperature below the chiller equipment’s minimumentering water temperature.10.5.5 EconomizersOne advantage of all-air HVAC systems is thatthey can utilize outside air to condition interior spaceswhen it is at an appropriate temperature. The use ofoutside air to actively cool interior spaces is referred to as aneconomizer, or economizer cycle.Whenever the outside air is cooler than the coolingsetpoint temperature only distribution energy is requiredto provide cooling with outside air. When the outside airtemperature is above the setpoint temperature but lessthan the return air temperature, it requires less coolingenergy to utilize 100 percent outdoor air for supply airthan to condition recycled indoor air. An economizercycle is simply a control sequence that adjusts outsideair and exhaust dampers to utilize 100 percent outside airwhen its temperature makes it advantageous to do so.The energy consumed in heating outside air at sealevel may be computed from the equationq = 1.08 Q•∆T (10.2a)whereq = hourly ventilation heating load, Btu/hrQ = volumetric flow rate of air, cfm∆T = temperature increase, °FThe economizer cycle is most appropriate forthermally massive buildings which have high internalloads and require cooling in interior zones year round.It is ineffective in thermally light buildings and buildingswhose heating and cooling loads are dominated bythermal transmission through the envelope. Economizercycles will provide the greatest benefit in climates havingmore than 2000 heating degree days per year, sincewarmer climates will have few days cold enough topermit the use of outside air for cooling.In theory 100 percent outside air would be utilizedfor cooling whenever the outside air temperature isbelow the return air temperature, and this is done indry climates. But if the outside air has a higher relativehumidity than return air and contains significant latentheat, the economizer would not be activated at this temperature.The simplest type of economizer utilizes a dry-bulbtemperature control which activates the economizer at apredetermined outside dry-bulb temperature, usuallythe supply air temperature, or about 55°F (13°C). Above55°F (13°C), minimum outdoor air is supplied for ventilation.Below 55°F (13°C), the quantity of outdoor air isgradually reduced from 100 percent and blended withreturn air to make 55°F (13°C) supply air.Eq. 10.2a predicts that the energy required to coolroom (return) air 20°F (13°C) from 75°F (24°C) with asupply air temperature of 55°F (13°C) is 1.08 × Q × 20= 21.6 Q. The energy saved [E, Btu/hr] from the useof a simple economizer which blends outside air withreturn to provide a supply air temperature, at sea level,is therefore:E = 21.6 Q(10.2b)Because it is possible to utilize outdoor air attemperatures above 55°F (13°C) to save cooling energy,a modified dry-bulb temperature control can be used. This

264 ENERGY MANAGEMENT HANDBOOKis identical to the simpler dry-bulb temperature controlexcept that when the outside temperature is between55°F (13°C) and a preselected higher temperature basedon the typical humidity, 100 percent outdoor air is used,but cooled to 55°F for supply air.The third and most efficient type of control for aneconomizer cycle is enthalpy control, which can instantaneouslydetermine and compare the amount of energy requiredto cool 100 percent outdoor air with that requiredto cool the normal blend of return air. It then selects thesource which requires the least energy for cooling.Air-handling units which lack adequate provisionfor 100 percent outside air can utilize an alternative“wet-side” economizer. This energy-saving techniqueis a potential retrofit for existing older buildings. Severalvariations on the wet-side economizer exist, all ofwhich allow the chiller to be shut down. The simplestwaterside economizer is a coil in the air-handling unitthrough which cooling tower water can be circulatedto provide free cooling. An alternative arrangement forlarge chilled water systems interconnects the chilledwater circuit directly with the cooling tower, to allowthe chiller to be bypassed. To reduce contamination ofthe chilled water, the cooling tower water should onlybe coupled to the chilled water circuit through a heatexchanger (Figure 10.13).The energy saved [E, kWh] due to the use of a wetsideeconomizer can be determined from a knowledgeof the hours [hr] of chiller operation during when theoutside temperature is below 40°F (4°C), the capacity ofchiller in tons and the kW/ton rating of the chiller:E = hr × tons × kW/ton (10.3)A word of caution is advised. A wet-side economizeruses condenser water at 40-45°F (4-7°C). But thechiller typically operates with condenser water muchhigher than this. In some cases, the lowest condenserwater temperature at which the chiller will operate is65°F (18°C). This means that when the chiller is turnedoff there would be no useful coolant until the coolingtower cooled to 45°F (7°C). This “cool down” period istypically no more than 30 minutes. For additional discussionof the wet-side economizer, see §10.6.6.10.5.6 Control StrategiesControl systems play a large part in the energyconservation potential of a HVAC system. To be effectiveat controlling energy use along with thermal comfortthey must be used appropriately, work properly and beset correctly. Overheated or overcooled spaces not onlywaste energy, they are uncomfortable. An easy way tospot such areas is to walk around the outside of thebuilding and look for open windows which usuallyindicate overheated (or overcooled) spaces.The efficiency of control systems is mostly relatedto the directness of control. Simple controls which turnthe system off and on when needed provide the mostdirect control. Systems in which air and/or water temperaturesare controlled by parameters other than theactual need for heating and cooling may be said to beindirectly controlled, and are inherently less efficient.Direct control systems most commonly employprimarily thermostats which turn equipment on or offor adjust the volume of air supplied to a space on thebasis of room temperature. Direct thermostat control ismost effective in thermally light buildings whose energyrequirements are directly proportional to the exteriorweather conditions. Where thermostats are an appropriatechoice, individual zones must be used for spaceswith different heating or cooling requirements.Many older HVAC systems, particularly heatingonlysystems, were created as single-zone systems forwhich the HVAC system is either on throughout thebuilding or entirely off. Self-contained thermostaticvalves can be used to modify single-zone convectionheating systems and single-zone forced air systems canbe readily modified into zoned VAV or reheat air-conditioningsystems without replacing the costly centralequipment.The selection of set points, even for directly-controlledsystems, is important. Thermostats should be setto maximize thermal comfort rather than relying on previouslyestablished design values. The need for humidificationvaries with outdoor temperature and humidistatsshould not be set at constant levels and forgotten. Theneed for humidification is also a function of occupancyand should be adjusted in assembly spaces when theyare unused or underutilized. The use of outside air forventilation when this air must be conditioned likewisecan be reduced with proper controls to adjust outside-airdampers. Modem microprocessor-based energy managementcontrol systems (EMCS) allow for vastly improveddirect control of HVAC systems.A useful concept made possible by modem electronicsis the use of enthalpy control. Enthalpy control relatesthe energy required to cool air at a given dry-bulbtemperature to the energy required to remove humidityfrom moist outdoor air. Enthalpy controls are most commonlyused to adjust economizer cycles, determiningwhen it is more efficient to recycle warm indoor air thanto condition humid outside air for “free” cooling.Control systems which are not well-understoodor simple to use are subject to misuse by occupants or

HVAC SYSTEMS 265Figure 10.13 Wet-side economizer schematic diagram.building service personnel. For example the controls ona unit ventilator include a room temperature thermostatwhich controls the valve on the heating or cooling coil,a damper control which adjusts the proportion of freshair mixed with recirculated room air, and a low-limitthermostat which prevents the temperature of outsideair from dropping below a preset temperature (usually55 to 60°F; 13 to 16°C). A common error of occupants orbuilding custodians in response to a sense that the airsupplied by the unit ventilator is too cold is to increasethe setpoint on the low-limit thermostat, which preventsfree cooling from outside air or, on systems without acooling coil, prevents cooling altogether. Controls whichare subject to misadjustment by building occupantsshould be placed so that they cannot be tamperedwith.The energy consumption of thermally heavybuildings is less related to either the inside or outsideair temperature. Both the heating and cooling loads inthermally heavy buildings are heavily dependent onthe heat generated from internal loads and the thermalenergy stored in the building mass which may be dis-

266 ENERGY MANAGEMENT HANDBOOKFansLightingPumpsSpacecoolingFigure 10.14 Energy cost distribution for a typicalnon-residential building using an all-air reheat HVACsystem.Other(magnitudeuncertain)Space heatingKitchen& processCooling of reheatDomestichot waterReheatsipated at a later time.In an indirect control system the amount of energyconsumed is not a function of human thermal comfortneeds, but of other factors such as outdoor temperature,humidity, or enthalpy. Indirect control systemsdetermine the set points for cool air temperature, watertemperatures, etc. As a result indirect control systemstend to adjust themselves for peak conditions ratherthan actual conditions. This leads to overheating orovercooling of spaces with less than peak loads.One of the most serious threats to the efficiencyof any system is the need to heat and cool air or watersimultaneously in order to achieve the thermal balancerequired for adequate conditioning of spaces. Figure10.14 indicates that 20 percent of the energy consumedin a commercial building might be used to reheat cooledair, offsetting another 6 percent that was used to cool theair which was later reheated. For the example buildingthe energy used to cool reheated air approaches thatactually used for space cooling.Following the 1973 oil embargo federal guidelinesencouraged everyone to reduce thermostat settings to68°F (20°C) in winter and to increase thermostat settingsin air-conditioned buildings to 78°F (26°C) in summer.[In 1979, the winter guideline was reduced farther to65°F (18°C).] The effect of raising the air-conditioningthermostat on a reheat, dual-duct, or multizone systemis actually to increase energy consumption by increasingthe energy required to reheat air which has beenmechanically cooled (typically to 55°F; 13°C).To minimize energy consumption on these typesof systems it makes more sense to raise the dischargetemperature for the cold-deck to that required to cool perimeterareas to 78°F (26°C) under peak conditions. If thesystem was designed to cool to 75°F (24°C) on a peak dayusing 55°F (13°C) air, the cold deck discharge could beincreased to 58°F (14.5°C) to maintain space temperaturesat no more than 78°F (26°C), saving about $5 per cfmper year. Under less-than-peak conditions these systemswould operate more efficiently if room temperatures wereallowed to fall below 78°F (26°C) than to utilize reheatedair to maintain this temperature.More extensive discussion of energy managementcontrol systems may be found in Chapters 12 and 22.10.5.7 HVAC EquipmentThe elements which provide heating and coolingto a building can be categorized by their intended function.HVAC equipment is typically classified as heatingequipment, including boilers, furnaces and unit heaters;cooling equipment, including chillers, cooling towersand air-conditioning equipment; and air distributionelements, primarily air-handling units (AHUs) and fans.A more lengthy discussion of boilers may be foundin Chapter 6, followed by a discussion of steam andcondensate systems in Chapter 7. Cooling equipment isdiscussed in section 10.6, below. What follows here relatesmostly to air-handling equipment and distributionsystems.Figure 10.14 depicts the typical energy cost distributionfor a large commercial building which employsan all-air reheat-type HVAC system. Excludingthe energy costs associated with lighting, kitchen andmiscellaneous loads which are typically 25-30 percentof the total, the remaining energy can be divided intotwo major categories: the energy associated with heatingand cooling and the energy consumed in distribution.The total energy consumed for HVAC systemsis therefore dependent on the efficiency of individualcomponents, the efficiency of distribution and the abilityof the control system to accurately regulate the energyconsuming components of the system so that energy isnot wasted.The size (and heating, cooling, or air-moving capacity)of HVAC equipment is determined by the mechanicaldesigner based upon a calculation of the peakinternal and envelope loads. Since the peak conditionsare arbitrary (albeit well-considered and statisticallyvalid) and it is likely that peak loads will not occursimultaneously throughout a large building or complex

HVAC SYSTEMS 267requiring all equipment to operate at its rated capacity,it is common to specify equipment which has a totalcapacity slightly less than the peak requirement. Thisdiversity factor varies with the function of the space.For example, a hospital or classroom building will usea higher diversity multiplier than an office building.In sizing heating equipment however, it is not uncommonto provide a total heating capacity from severalunits which exceeds the design heating load by as muchas fifty percent. In this way it is assured that the heatingload can be met at any time, even in the event that oneunit fails to operate or is under repair.The selection of several boilers, chillers, or airhandlingunits whose capacities combine to providethe required heating and cooling capability instead ofsingle large units allows one or more components ofthe system to be cycled off when loads are less than themaximum.This technique also allows off-hours use of specificspaces without conditioning an entire building.Equipment EfficiencyEfficiency, by definition, is the ratio of the energyoutput of a piece of equipment to its energy input, inlike units to produce a dimensionless ratio. Since noequipment known can produce energy, efficiency willalways be a value less than 1.0 (100%).Heating equipment which utilizes electric resistanceappears at first glance to come closest to theideal of 100 percent efficiency. In fact, every kilowatt ofelectrical power consumed in a building is ultimatelyconverted to 3413 Btu per hour of heat energy. Since thisis a valid unit conversion it can be said that electric resistanceheating is 100 percent efficient. What is missingfrom the analysis however, is the inefficiency of producingelectricity, which is most commonly generated usingheat energy as a primary energy source.Electricity generation from heat is typically about30 percent efficient, meaning that only 30 percent of theheat energy is converted into electricity, the rest beingdissipated as heat into the environment. Energy consumedas part of the generation process and energy lostin distribution use up about ten percent of this, leavingonly 27 percent of the original energy available for useby the consumer. By comparison, state-of-the-art heatingequipment which utilizes natural gas as a fuel is morethan eighty percent efficient. Distribution losses in naturalgas pipelines account for another 5 percent, makingnatural gas approximately three times as efficient as aheat energy source than electricity.The relative efficiency of cooling equipment isusually expressed as a coefficient of performance (COP),which is defined as the ratio of the heat energy extractedto the mechanical energy input in like units. Since theheat energy extracted by modem air conditioning farexceeds the mechanical energy input a COP of up to 6is possible.Air-conditioning equipment is also commonlyrated by its energy efficiency ratio (EER) or seasonal energyefficiency ratio (SEER). EER is defined as the ratioof heat energy extracted (in Btu/hr) to the mechanicalenergy input in watts. Although it should have dimensionsof Btu/hr/watt, it is expressed as a dimensionlessratio and is therefore related to COP by the equationEER = 3.41 • COP (10.4)Although neither COP nor EER is the efficiencyof a chiller or air-conditioner, both are measures whichallow the comparison of similar units. The term air-conditioningefficiency is commonly understood to indicatethe extent to which a given air-conditioner performs toits maximum capacity. As discussed below, most equipmentdoes not operate at its peak efficiency all of thetime. For this reason, the seasonal energy efficiency ratio(SEER), which takes varying efficiency at partial loadinto account, is a more accurate measure of air-conditioningefficiency than COP or EER.In general, equipment efficiency is a function ofsize. Large equipment has a higher efficiency than smallequipment of similar design. But the rated efficiency ofthis equipment does not tell the whole story. Equipmentefficiency varies with the load imposed. All equipmentoperates at its optimum efficiency when operated at ornear its design full-load condition. Both overloadingand under-loading of equipment reduces equipment efficiency.This fact has its greatest impact on system efficiencywhen large systems are designed to air-condition anentire building or a large segment of a major complex.Since air-conditioning loads vary and since the designheating and cooling loads occur only rarely under themost severe weather or occupancy conditions, most ofthe time the system must operate under-loaded. Whenselected parts of a building are utilized for off-hoursoperation this requires that the entire building be conditionedor that the system operate far from its optimumconditions and thus at far less than its optimum efficiency.Since most heating and cooling equipment operatesat less than its full rated load during most of theyear, its part-load efficiency is of great concern. Becauseof this, most state-of-the-art equipment operates muchcloser to its full-load efficiency than does older equip-

268 ENERGY MANAGEMENT HANDBOOKment. A knowledge of the actual operating efficiency ofexisting equipment is important in recognizing economicopportunities to reduce energy consumption throughequipment replacement.Distribution EnergyDistribution energy is most commonly electricalenergy consumed to operate fans and pumps, with fanenergy typically being far greater than pump energy exceptin all-water distribution systems. The performanceof similar fans is related by three fan laws which relatefan power, airflow, pressure and efficiency to fan size,speed and air density. The reader is referred to theASHRAE Handbook: HVAC Systems and Equipment foradditional information on fans and the application ofthe fan laws. 3Fan energy is a function of the quantity of airflowmoved by the fan, the distance over which it is moved,and the velocity of the moving air (which influencesthe pressure required of the fan). Most HVAC systems,whether central or distributed packaged systems, allair,all-water, or a combination are typically oversizedfor the thermal loads that actually occur. Thus the fanis constantly required to move more air than necessary,creating inherent system inefficiency.One application of the third fan law describesthe relationship between fan horsepower (energy consumed)and the airflow produced by the fan:W 1 = W 2 × (Q 1 /Q 2 ) 3 (10.5)whereW = fan power required, hpQ = volumetric flow rate, cfmBecause fan horsepower is proportional to the cubeof airflow, reducing airflow to 75 percent of existingwill result in a reduction in the fan horsepower by thecube of 75 percent, or about 42 percent: [(0.75) 3 = 0.422]Even small increases in airflow result in disproportionalincreases in fan energy. A ten percent increase in airflowrequires 33 percent more horsepower [1.103 = 1.33],which suggests that airflow supplied solely for ventilationpurposes should be kept to a minimum.All-air systems which must move air over greatdistances likewise require disproportionate increases inenergy as the second fan law defines the relationshipbetween fan horsepower [W] and pressure [p], whichmay be considered roughly proportional to the lengthof ducts connected to the fan:W 1 = W 2 × (P 1 /P 1 ) 3/2 (10.6)The use of supply air at temperatures of less than55°F (13°C) for primary cooling air permits the use ofsmaller ducts and fans, reducing space requirementsat the same time. This technique requires a complexanalysis to determine the economic benefit and is seldomadvantageous unless there is an economic benefitassociated with space savings.System ModificationsIn examining HVAC systems for energy conservationopportunities, the less efficient a system is, thegreater is the potential for significant conservation tobe achieved. There are therefore several “off-the-shelf”opportunities for improving the energy efficiency ofselected systems.All-air Systems—Virtually every type of all-airsystem can benefit from the addition of an economizercycle, particularly one with enthalpy controls. Systemswith substantial outside air requirements can also benefitfrom heat recovery systems which exchange heatbetween exhaust air and incoming fresh air. This is apractical retrofit only when the inlet and exhaust ductsare in close proximity to one another.Single zone systems, which cannot provide sufficientcontrol for varying environmental conditionswithin the area served can be converted to variable airvolume (VAV) systems by adding a VAV terminal andthermostat for each new zone. In addition to improvingthermal comfort this will normally produce a substantialsaving in energy costs.VAV systems which utilize fans with inlet vanesto regulate the amount of air supplied can benefit froma change to variable speed or variable frequency fandrives. Fan efficiency drops off rapidly when inlet vanesare used to reduce airflow.In terminal reheat systems, all air is cooled tothe lowest temperature required to overcome the peakcooling load. Modern “discriminating” control systemswhich compare the temperature requirements in eachzone and cool the main airstream only to the temperaturerequired by the zone with the greatest requirementswill reduce the energy consumed by these systems.Reheat systems can also be converted to VAV systemswhich moderate supply air volume instead of supply airtemperature, although this is a more expensive alterationthan changing controls.Similarly, dual-duct and multizone systems can benefitfrom “smart” controls which reduce cooling requirementsby increasing supply air temperatures. Hot-decktemperature settings can be controlled so that the temperatureof warm supply air is just high enough to meet

HVAC SYSTEMS 269design heating requirements with 100 percent hot-decksupply air and adjusted down for all other conditionsuntil the hot-deck temperature is at room temperaturewhen outside temperatures exceed 75°F (24°C). Dual ductterminal units can be modified for VAV operation.An economizer option for multizone systems isthe addition of a third “bypass” deck to the multizoneair-handling unit. This is not appropriate as a retrofitalthough an economizer can be utilized to provide colddeckair as a retrofit.All-water systems—Wet-side economizers are themost attractive common energy conservation measureappropriate to chilled water systems. Hot-water systemsbenefit most from the installation of self-contained thermostatvalves, to create heating zones in spaces formerlyoperated as single-zone heating systems.Air-water Induction—Induction systems are seldominstalled anymore but many still exist in olderbuildings. The energy-efficiency of induction systemscan be improved by the substitution of fan-poweredVAV terminals to replace the induction terminals.10.6 COOLING EQUIPMENTThe most common process for producing coolingis vapor-compression refrigeration, which essentiallymoves heat from a controlled environment to a warmer,uncontrolled environment through the evaporation ofa refrigerant which is driven through the refrigerationcycle by a compressor.Vapor compression refrigeration machines aretypically classified according to the method of operationof the compressor. Small air-to-air units mostcommonly employ a reciprocating or scroll compressor,combined with an air-cooled condenser to forma condensing unit. This is used in conjunction with adirect-expansion (DX) evaporator coil placed withinthe air-handling unit.Cooling systems for large non-residential buildingstypically employ chilled water as the medium whichtransfers heat from occupied spaces to the outdoorsthrough the use of chillers and cooling towers.10.6.1 ChillersThe most common type of water chiller for largebuildings is the centrifugal chiller which employs acentrifugal compressor to compress the refrigerant,which extracts heat from a closed loop of water which ispumped through coils in air-handling or terminal unitswithin the building. Heat is rejected from the condenserinto a second water loop and ultimately rejected to theenvironment by a cooling tower.The operating fluid used in these chillers may beeither a CFC or HCFC type refrigerant. Many existingcentrifugal chillers use CFC-11 refrigerants, the manufactureand use of which is being eliminated under theterms of the Montreal Protocol. New refrigerants HCFC-123 and HCFC-134a are being used to replace the CFCrefrigerants but refrigerant modifications to existingequipment will reduce the overall capacity of this equipmentby 15 to 25 percent.Centrifugal chillers can be driven by open orhermetic electric motors or by internal combustionTable 10.1 Summary of HVAC System Modifications for Energy ConservationSystem typeAll-air systems (general):Single zone systemsVariable air volume (VAV) systemsReheat systemsConstant volume dual-duct systemsMultizone systemsAll-water systems:hydronic heating systemschilled water systemsAir-water induction systemsEnergy Conservation Opportunitieseconomizerheat recoveryconversion to VAVreplace fan inlet vane control with variable frequency drive fanuse of discriminating control systemsconversion to VAVuse of discriminating control systemsconversion to dual duct VAVuse of discriminating control systemsaddition of by-pass deck*addition of thermostatic valveswet-side economizerreplacement with fan-powered VAV terminals*Requires replacement of air-handling unit

270 ENERGY MANAGEMENT HANDBOOKengines or even by steam or gas turbines. Natural gasengine-driven equipment sized from 50 to 800 tons ofrefrigeration are available and in some cases are used toreplace older CFC-refrigerant centrifugal chillers. Theseengine-driven chillers are viable when natural gas costsare sufficiently low. Part-load performance modulatesboth engine speed and compressor speed to match theload profile, mainta ining close to the peak efficiencydown to 50 percent of rated load. They can also use heatrecovery options to take advantage of the engine jacketand exhaust heat.Turbine-driven compressors are typically used onvery large equipment with capacities of 1200 tons ormore. The turbine may be used as part of a cogenerationprocess but this is not required. (For a detailed discussionof cogeneration, see Chapter 7.) If excess steam isavailable, in industry or a large hospital, a steam turbinecan be used to drive the chiller. However the higher loadon the cooling tower due to the turbine condenser mustbe considered in the economic analysis.Small water chillers, up to about 200 tons of capacity,may utilize reciprocating or screw compressors andare typically air-cooled instead of using cooling towers.An air-cooled chiller uses a single or multiple compressorsto operate a DX liquid cooler. Air-cooled chillers arewidely used in commercial and large-scale residentialbuildings.Other types of refrigeration systems include liquidoverfeed systems, flooded coil systems and multi-stagesystems. These systems are generally used in large industrialor low-temperature applications.10.6.2 Absorption ChillersAn alternative to vapor-compression refrigerationis absorption refrigeration which uses heat energy todrive a refrigerant cycle, extracting heat from a controlledenvironment and rejecting it to the environment(Figure 10.15). Thirty years ago absorption refrigerationwas known for its low coefficient of performance andhigh maintenance requirements. Absorption chillersused more energy than centrifugal chillers and wereeconomical only if driven by a source of waste heat.Today, due primarily to the restriction on the useof CFC and HCFC refrigerants, the absorption chilleris making a comeback. Although new and improved, itstill uses heat energy to drive the refrigerant cycle andtypically uses aqueous lithium bromide to absorb therefrigerant and water vapor in order to provide a highercoefficient of performance.The new absorption chillers can use steam as aheat source or be direct-fired. They can provide simultaneousheating and cooling which eliminates the needfor a boiler. They do not use CFC or HCFC refrigerants,which may make them even more attractive in yearsto come. Improved safety and controls and better COP(even at part load) have propelled absorption refrigerationback into the market.In some cases, the most effective use of refrigerationequipment in a large central-plant scenario isto have some of each type, comprising a hybrid plant.From a mixture of centrifugal and absorption equipmentthe operator can determine what equipment willprovide the lowest operating cost under different con-Figure 10.15 Simplified absorption cycle schematic diagram.

HVAC SYSTEMS 271ditions. For example a hospital that utilizes steam yearround, but at reduced rates during summer, might usethe excess steam to run an absorption chiller or steamdriventurbine centrifugal chiller to reduce its summertimeelectrical demand charges.10.6.3 Chiller PerformanceMost chillers are designed for peak load and thenoperate at loads less than the peak most of the time.Many chiller manufacturers provide data that identifiesa chiller’s part-load performance as an aid to evaluatingenergy costs. Ideally a chiller operates at a desiredtemperature difference (typically 45-55 degrees F; 25-30degrees C) at a given flow rate to meet a given load.As the load requirement increases or decreases, thechiller will load or unload to meet the need. A resetschedule that allows the chilled water temperature tobe adjusted to meet thermal building loads based onenthalpy provides an ideal method of reducing energyconsumption.Chillers should not be operated at less than 50 percentof rated load if at all possible. This eliminates bothsurging and the need for hot-gas bypass as well as thepotential that the chiller would operate at low efficiency.If there is a regular need to operate a large chiller at lessthan one-half of the rated load it is economical to installa small chiller to accommodate this load.10.6.4 Thermal StorageThermal storage can be another effective way ofcontrolling electrical demand by using stored chilledwater or ice to offset peak loads during the peak demandtime. A good knowledge of the utility consumptionand/or load profile is essential in determining theapplicability of thermal storage. See Chapter 19 for adiscussion of thermal storage systems.10.6.5 Cooling TowersCooling towers use atmospheric air to cool thewater from a condenser or coil through evaporation. Ingeneral there are three types of cooling tower, named forthe relationship between the fan-powered airflow and theflow of water in the tower: counterflow induced draft,crossflow induced draft and counterflow forced draft.The use of variable-speed, two-speed or three-speedfans is one way to optimize the control of the coolingtower in order to reduce power consumption and provideadequate water cooling capacity. As the required coolingcapacity increases or decreases the fans can be sequencedto maintain the approach temperature difference. Formost air-conditioning systems this usually varies between5 and 12 degrees F (3 to 7 degrees C).When operated in the winter, the quantity of airmust be carefully controlled to the point where thewater spray is not allowed to freeze. In cold climates itmay be necessary to provide a heating element withinthe tower to prevent freeze-ups. Although electric resistanceheaters can be used for this purpose it is far moreefficient to utilize hot water or steam as a heat source ifavailable.10.6.6 Wet-side EconomizerThe use of “free-cooling” using the cooling towerwater to cool supply air or chilled water is referred toas a wet-side economizer. The most common and effectiveway of interconnecting the cooling tower water tothe chilled water loop is through the use of a plate-andframeheat exchanger which offers a high heat transferrate and low pressure drop. This method isolates thecooling tower water from the chilled water circuit maintainingthe integrity of the closed chilled water loop.Another method is to use a separate circuit and pumpthat allows cooling tower w ater to be circulated througha coil located within an air-handling unit.The introduction of cooling tower water, intothe chilled water system, through a so-called strainercycle, can create maintenance nightmares and shouldbe avoided. The water treatment program required forchilled water is intensive due to the required cleannessof the water in the chilled water loop.10.6.7 Water treatmentA good water treatment program is essential tothe maintenance of an efficient chilled water system.Filtering the cooling tower water should be evaluated.In some cases, depending on water quality, this can savethe user a great deal of money in chemicals. Pretreatingnew system s prior to initial start-up will also providelonger equipment life and insure proper system performance.Chiller performance is based on given design parametersand listed in literature provided by the chillermanufacturer. The performance will vary with buildingload, chilled water temperature, condenser water temperatureand fouling factor. The fouling factor is the resistancecaused by dirt, scale, silt, rust and other depositson the surface of the tubes in the chiller and significantlyaffects the overall heat transfer of the chiller.10.7 DOMESTIC HOT WATERThe creation of domestic hot water (DHW) representsabout 4 percent of the annual energy consumption

272 ENERGY MANAGEMENT HANDBOOKin typical non-residential buildings. In buildings wheresleeping or food preparation occur, including hotels,restaurants, and hospitals, DHW may account for asmuch as thirty percent of total energy consumption.Some older lavatory faucets provide a flow of 4 to 6gal/min (0.25 to 0.38 l/s). Since hand washing is a functionmore of time than water use, substantial savings canbe achieved by reducing water flow. Reduced-flow faucetswhich produce an adequate spray pattern can reducewater consumption to less than 1 gal/min (0.06 l/s). Flowreducing aerator replacements are also available.Reducing DHW temperature has also been shownto save energy in non-residential buildings. Since mostbuilding users accept water at the available temperature,regardless of what it is, water temperature can bereduced from the prevailing standard of 140°F (60°C)to a 105°F (40°C) utilization temperature saving up toone-half of the energy used to heat the water.Many large non-commercial buildings employ recirculatingDHW distribution systems in order to reduceor eliminate the time required and water wasted influshing cold water from hot water piping. Recirculatingdistribution is economically attractive only where DHWuse is high and/or the cost of water greatly exceeds thecost of water heating. In most cases the energy requiredto keep water in recirculating DHW systems hot exceedsthe energy used to heat the water actually used.To overcome this waste of energy there is a trendto convert recirculating DHW systems to localized pointof-usehot water heating, particularly in buildings whereplumbing facilities are widely separated. In either caseinsulation of DHW piping is essential in reducing thewaste of energy in distribution. One-inch of insulationon DHW pipes will result in a 50% reduction in thedistribution heat loss.One often-overlooked energy conservation opportunityassociated with DHW is the use of solar-heatedhot water. Unlike space-heating, the need for DHW isrelatively constant throughout the year and peaks duringhours of sunshine in non-residential buildings. Yearrounduse amortizes the cost of initial equipment fasterthan other active-solar options.Many of the techniques appropriate for reducingenergy waste in DHW systems are also appropriate forenergy consumption in heated service water systems forindustrial buildings or laboratories.10.8 ESTIMATING HVACENERGY CONSUMPTIONThe methods for estimating building heating andcooling loads and the consumption of energy by HVACsystems are described in Chapter 9.References1. ASHRAE Handbook: Fundamentals, American Society of Heating,Refrigerating and Air-Conditioning Engineers, Inc., Atlanta,1993.2. ASHRAE Handbook: HVAC Applications, American Society of Heating,Refrigerating and Air-Conditioning Engineers, Inc., Atlanta,1995.3. ASHRAE Handbook: HVAC Systems and Equipment, American Societyof Heating, Refrigerating and Air-Conditioning Engineers,Inc., Atlanta, 1992.4. ASHRAE Handbook: Refrigeration, American Society of Heating,Refrigerating and Air-Conditioning Engineers, Inc., Atlanta,1 9 9 4 .

CHAPTER 11E LECTRIC ENERGY MANAGEMENTK.K. LOBODOVSKYBSEE & BSMECertified Energy AuditorState of California11.1 INTRODUCTIONEfficient use of electric energy enables commercial,industrial and institutional facilities to minimize operatingcosts, and increase profits to stay competitive.The majority of electrical energy in the UnitedStates is used to run electric motor driven systems.Generally, systems consist of several components, theelectrical power supply, the electric motor, the motorcontrol, and a mechanical transmission system.There are several ways to improve the systems'efficiency. The cost effective way is to check each componentof the system for an opportunity to reduce electricallosses. A qualified individual should oversee theelectrical system since poor power distribution within afacility is a common cause of energy losses.Technology Update Ch. 18 1 lists 20 items to helpfacility management staff identify opportunities to improvedrive system efficiency.1. Maintain Voltage Levels.2. Minimize Phase Imbalance.3. Maintain Power Factor.4. Maintain Good Power Quality.5. Select Efficient Transformers.6. Identify and Fix Distribution System Losses.7. Minimize Distribution System Resistance.8. Use Adjustable Speed Drives (ASDs) or 2-SpeedMotors Where Appropriate.9. Consider Load Shedding.10. Choose Replacement Before a Motor Fails.11. Choose Energy-Efficient Motors.12. Match Motor Operating Speeds.13. Size Motors for Efficiency.14. Choose 200 Volt Motors for 208 Volt Electrical Systems.15. Minimize Rewind Losses.16. Optimize Transmission Efficiency.17. Perform Periodic Checks.18. Control Temperatures.19. Lubricate Correctly.20. Maintain Motor Records.Some of these steps require the one-time involvementof an electrical engineer or technician. Somesteps can be implemented when motors fail or majorcapital changes are made in the facility. Others involvedevelopment of a motor monitoring and maintenanceprogram.11.2 POWER SUPPLYMuch of this information consists of standardsdefined by the National Electrical Manufacturers Association(NEMA).The power supply is one of the major factors affectingselection, installation, operation, and maintenanceof an electrical motor driven system. Usual service conditions,defined in NEMA Standard Publication MG1,Motors and Generators, 2 include:• Motors designed for rated voltage, frequency, andnumber of phases.• The supply voltage must be known to select theproper motor.• Motor nameplate voltage will normally be lessthen nominal power system voltage.NominalMotor UtilizationPower System (Nameplate) VoltageVoltage (Volts)———————Volts——————————208 200240 230480 460600 5752400 23004160 40006900 660013800 13200• Operation within tolerance of ±10 percent of therated voltage.273

274 ENERGY MANAGEMENT HANDBOOK• Operation from a sine wave of voltage source (notto exceed 10 percent deviation factor).• Operation within a tolerance of ±5 percent of ratedfrequency.• Operation within a voltage unbalance of 1 percentor less.Operation at other than usual service conditions may resultin the consumption of additional energy.11.3 EFFECTS OF UNBALANCED VOLTAGES ONTHE PERFORMANCE OF POLYPHASESQUIRREL-CAGE INDUCTION MOTORS(MG 1-20.56)When the line voltages applied to a polyphaseinduction motor are not equal, unbalanced currents inthe stator windings result. A small percentage of voltageunbalance results in a much larger percentage currentunbalance. Consequently, the temperature rise ofthe motor operating at a particular load and percentagevoltage unbalance will be greater than for the motoroperating under the same conditions with balancedvoltages.Voltages should be evenly balanced as closely asthey can be read on a voltmeter. If the voltages areunbalanced, the rated horsepower of polyphase squirrel-cageinduction motors should be multiplied by thefactor shown in Figure 11.1 to reduce the possibilityof damage to the motor. Operation of the motor withmore than a 5-percent voltage unbalance is not recommended.When the derating curve of Figure 11.1 is appliedfor operation on balanced voltages, the selection andsetting of the overload device should take into accountthe combination of the derating factor appliedto the motor and the increase in current resulting fromthe unbalanced voltages. This is a complex probleminvolving the variation in motor current as a functionof load and voltage unbalance in the addition tothe characteristics of the overload device relative toI MAXIMUM or I AVERAGE . In the absence of specific informationit is recommended that overload devices beselected and/or adjusted at the minimum value thatdoes not result in tripping for the derating factor andvoltage unbalance that applies. When the unbalancedvoltages are unanticipated, it is recommended that theoverload devices be selected so as to be responsive toI MAXIMUM in preference to overload devices responsiveto I AVERAGE .Figure 11.1 Polyphase squirrel-cage induction motorsderating factor due to unbalanced voltage.11.4 EFFECT ON PERFORMANCE—GENERAL (MG 1 20.56.1)The effect of unbalanced voltages on polyphaseinduction motors is equivalent to the introduction of a“ negative-sequence voltage” having a rotation oppositeto that occurring the balanced voltages. This negative-sequencevoltage produces an air gap flux rotating againstthe rotation of the rotor, tending to produce high currents.A small negative-sequence voltage may producecurrent in the windings considerably in excess of thosepresent under balanced voltage conditions.11.4.1 Unbalanced Defined (MG 1 20.56.2)The voltage unbalance in percent may be definedas follows:PercentVoltage = 100 ×UnbalanceMaximum voltage deviationfrom average voltageaverage voltageExample—With voltages of 220, 215 and 210, the averageis 215, the maximum deviation from the average is 5PERCENT VOLTAGE UNBALANCE= 100 * 5/215 = 2.3 PERCENT11.4.2 Torque (MG 1 20.56.3)The locked-rotor torque and breakdown torque aredecreased when the voltage is unbalanced. If the voltageunbalance is extremely severe, the torque might not beadequate for the application.11.4.3 Full-Load Speed (MG 1 20.56.4)The full-load speed is reduced slightly when themotor operates at unbalanced voltages.11.4.4 Currents (MG 1 20.56.5)The locked-rotor current will be unbalanced butthe locked rotor kVA will increase only slightly.

ELECTRIC ENERGY MANAGEMENT 275The currents at normal operating speed withunbalanced voltages will be greatly unbalanced in theorder of 6 to 10 times the voltage unbalance.11.5 MOTORThe origin of the electric motor can be traced backto 1831 when Michael Faraday demonstrated the fundamentalprinciples of electromagnetism. The purposeof an electric motor is to convert electrical energy intomechanical energy.Electric motors are efficient at converting electricenergy into mechanical energy. If the efficiency of anelectric motor is 80%, it means that 80% of electricalenergy delivered to the motor is directly converted tomechanical energy. The portion used by the motor isthe difference between the electrical energy input andmechanical energy output.A major manufacturer estimates that US annualsales exceed 2 million motors. Table 11.1 lists sales volumeby motor horsepower. Only 15% of these salesinvolve high-efficiency motors. 3Table 11.1 Polyphase induction motors annual salesvolume.hp————Units—————1-5 1,300,0007.5-20 500,00025-50 140,00060-100 40,000125-200 32,000———— 250-500 ———— 11,000Total 2,023,000Motor terms are used quite frequently, usually onthe assumption that every one knows what they meanor imply. Such is far too often not the case. The followingsection is a list of motor terms.11.6 GLOSSARY OF FREQUENTLYOCCURRING MOTOR TERMS 4AmpsFull Load AmpsThe amount of current the motor can be expectedto draw under full load (torque) conditions is called FullLoad Amps. It is also known as nameplate amps.Locked Rotor AmpsAlso known as starting inrush, this is the amount ofcurrent the motor can be expected to draw under startingconditions when full voltage is applied.Service Factor AmpsThis is the amount of current the motor will drawwhen it is subjected to a percentage of overload equalto the service factor on the nameplate of the motor. Forexample, many motors will have a service factor of 1.15,meaning that the motor can handle a 15% overload. Theservice factor amperage is the amount of current that themotor will draw under the service factor load condition.Code LetterThe code letter is an indication of the amount ofinrush current or locked rotor current that is required bya motor when it is started. Motor code letters usually appliedto ratings of motors normally started on full voltage(chart below).Code letter Locked rotor* Horsepower HorsepowerkVA per Single-phase Three-phasehorsepower———————————————————————————————————F 5.0 to 5.6 15 upG 5.6 to 6.3 5 7.5 to 10H 6.3 to 7.1 3 5J 7.1 to 8.0 1.5 to 2 3K 8.0 to 9.0 0.75 to 1.00 1.5 to 2L 9.0 to 10.0 0.5 1———————————————————————————————————*Locked rotor kVA is equal to the product of the line voltage times motor current divided by 1,000when the motor is not allowed to rotate; this corresponds to the first power surge required to startthe motor. Locked-rotor kVA per horsepower range includes the lower figure up to but not includingthe higher figure.

276 ENERGY MANAGEMENT HANDBOOKDesignThe design letter is an indication of the shape of thetorque speed curve. Figure 11.2 shows the typical shapeof the most commonly used design letters. They are A,B, C, and D. Design B is the standard industrial duty motorwhich has reasonable starting torque with moderatestarting current and good overall performance for mostindustrial applications. Design C is used for hard to startloads and is specifically designed to have high startingtorque. Design D is the so-called high slip motor whichtends to have very high starting torque but has high slipRPM at full load torque. In some respects, this motorcan be said to have a ‘spongy’ characteristic when loadsare changing. Design D motors particularly suited forlow speed punch press, hoist and elevator applications.Generally, the efficiency of Design D motors at full loadis rather poor and thus they are normally used on thoseapplications where the torque characteristics are of primaryimportance. Design A motors are not commonlyspecified but specialized motors used on injection moldingapplications have characteristics similar to DesignB. The most important characteristic of Design A is thehigh pull out torque.Figure 11.2EfficiencyEfficiency is the percentage of the input power thatis actually converted to work output from the motorshaft. Efficiency is now being stamped on the nameplateof most domestically produced electric motors. See thesection 11.14.746 × HP OutputEfficiency = EFF = ————————Watts InputFrame SizeMotors, like suits of clothes, shoes and hats, comein various sizes to match the requirements of the applications.In general, the frame size gets larger withincreasing horsepower or with decreasing speeds. Inorder to promote standardization in the motor industry,NEMA (National Electrical Manufacturers Association)prescribes standard frame sizes for certain horsepower,speed, and enclosure combinations. Frame size pinsdown the mounting and shaft dimension of standard motors.For example, a motor with a frame size of 56, will alwayshave a shaft height above the base of 3- 1/2 inches.FrequencyThis is the frequency for which the motor is designed.The most commonly occurring frequency in thiscountry is 60 cycles but, internationally, other frequenciessuch as 25, 40, and 50 cycles can be found.Full Load SpeedAn indication of the approximate speed that the motorwill run when it is putting out full rated output torqueor horsepower is called full load speed.High Inertia LoadThese are loads that have a relatively high fly wheeleffect. Large fans, blowers, punch presses, centrifuges,industrial washing machines, and other similar loads canbe classified as high inertia loads.Insulation ClassThe insulation class is a measure of the resistanceof the insulating components of a motor to degradationfrom heat. Four major classifications of insulation areused in motors. They are, in order of increasing thermalcapabilities, A, B, F, and H.Class of Insulation System Temperature, Degrees C—————————— ——————————A 75B 905F 115H 130—————————————————————————

ELECTRIC ENERGY MANAGEMENT 277Load TypesConstant HorsepowerThe term constant horsepower is used in certaintypes of loads where the torque requirement is reduced asthe speed is increased and vice-versa. The constant horsepowerload is usually associated with metal removal applicationssuch as drill presses, lathes, milling machines,and similar types of applications.Constant TorqueConstant torque is a term used to define a load characteristicwhere the amount of torque required to drivethe machine is constant regardless of the speed at whichit is driven. For example, the torque requirement of mostconveyors is constant.PhasePhase is the indication of the type of power supplyfor which the motor is designed. Two major categoriesexist: single phase and three phase. There are some veryspotty areas where two phase power is available but thisis very insignificant.PolesThis is the number of magnetic poles within themotor when power is applied. Poles are always an evennumber such as 2, 4, 6. In an AC motor, the number ofpoles work in conjunction with the frequency to determinethe synchronous speed of the motor. At 50 and 60cycles, common arrangements are:Synchronous speed—————————————————————————Poles 60 Cycles 50 Cycles—————————————————————————2 3600 30004 1800 15006 1200 10008 900 75010 720 600—————————————————————————Power FactorPercent power factor is a measure of a particularmotor’s requirements for magnetizing amperage. Formore information se section 11.7.Power FactorWatts Input(3 phase) = pf = —————————Volts × Amps × 1.73Variable TorqueVariable torque is found in loads having characteristicsrequiring low torque at low speeds and increasingvalues of torque required as the speed is increased. Typicallyexamples of variable torque loads are centrifugalfans and centrifugal pumps.Service FactorThe service factor is a multiplier that indicates theamount of overload a motor can be expected to handle.For example, a motor with a 1.0 service factor cannot be

278 ENERGY MANAGEMENT HANDBOOKexpected to handle more than its nameplate horsepoweron a continuous basis. Similarly, a motor with a 1.15 servicefactor can be expected to safely handle intermittentloads amounting to 15% beyond its nameplate horsepower.SlipSlip is used in two forms. One is the slip RPM whichis the difference between the synchronous speed and thefull load speed. When this slip RPM is expressed as a percentageof the synchronous speed, then it is called percentslip or just ‘slip.’ Most standard motors run with a fullloadslip of 2% to 5%.Synchronous SpeedThis is the speed at which the magnetic field withinthe motor is rotating. It is also approximately the speedthat the motor will run under no load condition. For example,a 4 pole motor running in 60 cycles would have amagnetic field speed of 1800 RPM. The no load speed ofthat motor shaft would be very close to 1800, probably1798 or 1799 RPM. The full load speed of the same motormight be 1745 RPM. The difference between the synchronousspeed of the full load speed is called the slip RPM ofthe motor.TemperatureAmbient Temperature.Ambient temperature is the maximum safe roomtemperature surrounding the motor if it is going to beoperated continuously at full load. In most cases, thestandardized ambient temperature rating is 40°C (104°F).This is a very warm room. Certain types of applicationssuch as on board ships and in boiler rooms, may requiremotors with a higher ambient temperature capabilitysuch as 50°C or 60°C.Temperature Rise.Temperature rise is the amount of temperaturechange that can be expected within the winding of themotor from non-operating (cool condition) to its temperatureat full load continuous operating condition.Temperature rise is normally expressed in degrees centigrade.Time RatingMost motors are rated in continuous duty whichmeans that they can operate at full load torque continuouslywithout overheating. Motors used on certain typesof applications such as waste disposal, valve actuators,hoists, and other types of intermittent loads, will frequentlybe rated in short term duty such as 5 minutes, 15minutes, 30 minutes or 1 hour. Just like a human being,a motor can be asked to handle very strenuous work aslong as it is not required on a continuous basis.TorqueTorque is the twisting force exerted by the shaft or amotor. Torque is measured in inch pounds, foot pounds,and on small motors, in terms of inch ounces.Full Load TorqueFull load torque is the rated continuous torque thatthe motor can support without overheating within itstime rating.Peak TorqueMany types of loads such as reciprocating compressorshave cycling torque where the amount of torque requiredvaries depending on the position of the machine.The actual maximum torque requirement at any point iscalled the peak torque requirement. Peak torque are involvedin things such as punch presses and other types ofloads where an oscillating torque requirement occurs.Pull Out TorqueAlso known as breakdown torque, this is the maximumamount of torque that is available from the motorshaft when the motor is operating at full voltage and isrunning at full speed. The load is then increased until themaximum point is reached. Refer to Figure 11.3.Pull Up TorqueThe lowest point on the torque speed curve for amotor accelerating a load up to full speed is called pull uptorque. Some motors are designed to not have a value ofpull up torque because the lowest point may occur at thelocked rotor point. In this case, pull up torque is the sameas locked rotor torque.Figure 11.3 Typical speed—torque curve.

ELECTRIC ENERGY MANAGEMENT 279Starting TorqueThe amount of torque the motor produces when itis energized at full voltage and with the shaft locked inplace is called starting torque. This value is also frequentlyexpressed as ‘Locked Rotor Torque.’ It is the amount oftorque available when power is applied to break the loadaway and start accelerating it up to speed.VoltageThis would be the voltage rating for which the motoris designed. Section 11.2.11.7 POWER FACTORWHAT IS POWER FACTOR (pf)?It is the mathematical ratio of ACTIVE POWER () toAPPARENT POWER (VA)Active powerpf = ——————————Apparent powerpf angle in degrees = cos –1 θ= W = Cos θACTIVE POWER = W = “real power” = supplied by thepower system to actually turn the motor.REACTIVE POWER = VAR = (W)tan θ = is used strictlyto develop a magnetic field within the motor.or (VA) 2 = (W) 2 + (VAR) 2NOTE: Power factor may be “leading” or “lagging”depending on the direction of VAR flow.CAPACITORS can be used to improve the powerfactor of a circuit with a large inductive load. Currentthrough capacitor LEADS the applied voltage by 90 electricaldegrees (VAC), and has the effect of “opposing”the inductive “LAGGING” current on a “one-for-one”(VAR) basis.WHY RAISE POWER FACTOR (pf)?Low (or “unsatisfactory”) power factor is causedby the use of inductive (magnetic) devices and can indicatepossible low system electrical operating efficiency.These devices are:• non-power factor corrected fluorescent and highintensity discharge lighting fixture ballasts (40%-80% pf)• arc welders (50%-70% pf)• solenoids (20%-50% pf)• induction heating equipment (60%-90% pf)• lifting magnets (20%-50% pf)• small “dry-pack” transformers (30%-95% pf)• and most significantly, induction motors (55%-90%pf)Induction motors are generally the principal causeof low power factor because there are so many in use,and they are usually not fully loaded. The correction ofthe condition of LOW power factor is a problem of vitaleconomic importance in the generation, distribution andutilization of a-c power.MAJOR BENEFITS OFPOWER FACTOR IMPROVEMENT ARE:• increased plant capacity,• reduced power factor “penalty” charges for theelectric utility,• improvement of voltage supply,• less power losses in feeders, transformers and distributionequipment.WHERE TO CORRECT POWER FACTOR?Capacitor correction is relatively inexpensive bothin material and installation costs. Capacitors can beinstalled at any point in the electrical system, and willimprove the power factor between the point of applicationand the power source. However, the power factorbetween the utilization equipment and the capacitor willremain unchanged. Capacitors are usually added at eachpiece of offending equipment, ahead of groups of smallmotors (ahead of motor control centers or distributionpanels) or at main services. Refer to the National ElectricalCode for installation requirements.

280 ENERGY MANAGEMENT HANDBOOKThe advantages and disadvantages of each type ofcapacitor installation are listed below:Capacitor on each piece of equipment (1,2)ADVANTAGES• increases load capabilities of distribution system.• can be switched with equipment; no additionalswitching is required.• better voltage regulation because capacitor use followsload.• capacitor sizing is simplified• capacitors are coupled with equipment and movewith equipment if rearrangements are instituted.DISADVANTAGES• small capacitors cost more per KVAC than largerunits (economic break point for individual correctionis generally at 10 HP).Capacitor with equipment group (3)ADVANTAGES• increased load capabilities of the service,• reduced material costs relative to individual correction• reduced installation costs relative to individualcorrectionDISADVANTAGES• switching means may be required to controlamount of capacitance used.Capacitor at main service (4,5, & 6)ADVANTAGES• low material installation costs.DISADVANTAGES• switching will usually be required to control theamount of capacitance used.• does not improve the load capabilities of the distributionsystem.OTHER CONSIDERATIONSWhere the loads contributing to power factor arerelatively constant, and system load capabilities are nota factor, correcting at the main service could provide acost advantage. When the low power factor is derivedfrom a few selected pieces of equipment, individualequipment correction would be cost effective. Most capacitorsused for power factor correction have built-infusing; if not, fusing must be provided.The growing use of ASDs (nonlinear loads) hasincreased the complexity of system power factor and itscorrections. The application of pf correction capacitorswithout a thorough analysis of the system can aggravaterather than correct the problem, particularly if the fifthand seventh harmonics are present.POWER QUALITY REQUIREMENTS 6The electronic circuits used in ASDs may be susceptibleto power quality related problems if care is nottaken during application, specification and installation.The most common problems include transient overvoltages,voltage sags and harmonic distortion. Thesepower quality problems are usually manifested in theform of nuisance tripping.TRANSIENT OVERVOLTAGES—Capacitors aredevices used in the utility power system to providepower factor correction and voltage stability duringperiods of heavy loading. Customers may also use capacitorsfor power factor correction within their facility.When capacitors are energized, a large transient overvoltagemay develop causing the ASD to trip.VOLTAGE SAGS—ASDs are very sensitive to temporaryreductions in nominal voltage. Typically, voltagesags are caused by faults on either the customer’s or the

ELECTRIC ENERGY MANAGEMENT 281ics which, if severe, can cause motor, transformer andconductor overheating, capacitor failures, misoperationof relays and controls and reduce system efficiencies.Compliance with IEEE-519 “Recommended Practicesand Requirements for Harmonic Control in ElectricalPower Systems” is strongly recommended.11.8 HANDY ELECTRICAL FORMULAS & RULES OF THUMBConversion formulasRules of thumb.At 3600 RPM, a motor develops 1.5 lb.-ft. per HP.At 1800 RPM, a motor develops 3 lb.-ft. per HP.At 1200 RPM, a motor develops 4.5 lb.-ft. per HP.At 550 & 575 Volts, a 3 phase motor draws 1 amp per HP.At 440 & 460 Volts, a 3 phase motor draws 1.25 amp per HP.At 220 & 230 Volts, a 3 phase motor draws 2.5 amp per HP.11.9 ELECTRIC MOTOR OPERATING LOADSMost electric motors are designed to operate at50 to 100 percent of their rated load. One reason is themotors optimum efficiency is generally 75 percent of therated load, and the other reason is motors are generallysized for the starting requirements.Several surveys of installed motors reveal thatlarge portion of motors in use are improperly loaded.Underloaded motors, those loaded below 50 percent ofrated load, operate inefficiently and exhibit low powerfactor. Low power factor increases losses in electricaldistribution and utilization equipment, such as wiring,motors, and transformers, and reduces the load-handlingcapability and voltage regulation of the building’sutility's electrical system.HARMONIC DISTORTION—ASDs introduce harmonicsinto the power system due to nonlinear characteristicsof power electronics operation. Harmonics arecomponents of current and voltage that are multiples ofthe normal 60Hz ac sine wave. ASDs produce harmonelectricalsystem. Typical part-load efficiency and powerfactor characteristics are shown in Figure 11.4.POWER SURVEYPower surveys are conducted to compile meaningfulrecords of energy usage at the service entrance,feeders and individuals loads. These records can beanalyzed to prioritize those areas yielding the greatestenergy savings. Power surveys also provide informationfor load scheduling to reduce peak demand and showoperational characteristics of loads that may suggestcomponent or system replacement to reduce energyconsumption. Only through the measurement of ACpower parameters can true cost benefit analysis be performed.7

282 ENERGY MANAGEMENT HANDBOOKFigure 11.4 Typical part-load efficiency and power factor characteristics11.10 DETERMINING ELECTRICMOTOR OPERATING LOADSDetermining if electric motors are properly loadedenables a manager to make informed decisions aboutwhen to replace them and which replacement to choose.There are several ways to determine motor loads. Thebest and the simplest way is by direct electrical measurementusing a Power Meter. Slip Measurement orAmperage Readings methods can be used to estimatethe actual load.11.11 POWER METERTo understand the electrical power usage of afacility, load or device, measurements must be takenover a time span to have a profile of the unit’s operation.Digital power multimeters, measure Amps, Volts,kWatts, kVars, kVA, Power Factor, Phase Angle andFiring Angle. The GENERAL TEST FORM Figure 11.5provides a format for documentation with correspondingconnection diagrams for various power circuitconfigurations.Such measurements should only beperformed by trained personnelSelection of Equipment for PowerMeasurement or SurveysWhen choosing equipment to conduct a powersurvey, many presentation formats are available includingindicating instruments, strip chart recorders anddigital devices with numeric printout. For most surveyapplications, changing loads makes it mandatory fordata to be compiled over a period of time. This periodmay be an hour, day, week or month. Since it is notpractical to write down varying readings from an indicatingdevice for a long period of time, a chart recorderor digital device with numeric printout is preferred. Ifloads vary frequently, an analog trend recording will beeasier to analyze than trying to interpret several numericreports. Digital power survey monitors are typically lessexpensive than analog recordings systems. Completemicroprocessor based power survey systems capable ofmeasuring watts, VARs, kVA, power factor, watt hours,VAR hours and demand including current transformersare available for under $3000. With prices for memoryand computers going down, digital devices interfacedto disk or cassette storage will provide a cost effectivemethod for system analysis. 7LoadsWhen analyzing polyphase motors, it is importantto make measurements with equipment suited for theapplication. Watt measurements or VAR measurementsshould be taken with a two element device. Power factorshould be determined from the readings of both measurements.When variable speed drives are encountered,it is always preferable to take measurements on the lineside of the controller. When measurements are requiredon the load side of the controller, the instrument specificationsshould be reviewed and if there is a question onthe application the manufacturer should be contacted. 711.12 SLIP MEASUREMENTConditions1. Applied voltage must be within 5% of nameplaterating.2. Should not be used on rewound motors.

ELECTRIC ENERGY MANAGEMENT 283TO MEASURE 1 Phase, 2 Wire 1 Phase, 3 Wire 3 Phase, 3 Wire 3 Phase, 4 Wire 3 Phase, 4 Wire TAPL - 1 to N L - 1 to N A to B Phase A Phase to N B Phase to NVOLTAGE L - 2 to N C to B Phase B Phase to N A Phase to N(V) C Phase to N C Phase to NL - 1 L - 1 A Phase A Phase B PhaseCURRENT L - 2 C Phase B Phase A Phase(A) C Phase C PhaseL - 1 to N L - 1 to N A Phase to NPOWER L - 2 to N B Phase to N(kW)C Phase to NTotal kW Total kW Total kW Total kWL - 1 to N L - 1 to N A Phase to NVOLT-AMPERES L - 2 to N B Phase to NREACTIVEC Phase to N(kVAR) Total kVAR Total kVAR Total kVAR Total kVARL - 1 to N L - 1 to N A Phase to NVOLT-AMPERES L - 2 to N B Phase to N(kVA)C Phase to NTotal kVA Total kVA Total kVA Total kVAL - 1 to N L - 1 to N A Phase to NPOWER FACTOR L - 2 to N B Phase to NPF %C Phase to NCombined PF % Combined PF % Combined PF % Total PF %VOLTAGE LEADCURRENTTRANSFORMERwith white leadtowards loadRE = REDBE = BLUEBK = BLACKWE = WHITESOURCE SOURCE SOURCE SOURCE SOURCEL1 N L1 N L2 L1 L2 L3 L1 L2 L3 N L1 TAP L2 L3RD WE RD WE BERD WE BE RD BE BK WE BE WE RDBKRDRD BK BERD BK BE RD BE BK BE RD BKLOAD LOAD LOAD LOAD LOADFigure 11.5 General test form (for use with power meter).

284 ENERGY MANAGEMENT HANDBOOK3. Motors should be operating under steady loadconditions.4. Should be performed by trained personnel.Note: Values used in this analysis are subject to roundingerrors. For example, full load speed often roundedto the nearest 5 RPM.Procedure1. Read and record the motors nameplate Full LoadSpeed. (RPM)2. Determine Synchronous speed No Load Speed(RPM) (900, 1200, 1800, 3600)3. Measure and record Operating Load Speed withtachometer. (RPM)4. Insert the recorded values in the following formulaand solve.NLS – OLS(% Motor load) = ——————— × 100NLS – FLSWhere:NLS = No load or synchronous speedOLS = Operating load speedFLS = Full load speedExample:Consider a 100 HP, 1800 RM MotorFLS = 1775 RPM, OLS = 1786 RPM1800– 1786(% Motor load) = ——————— × 100 = 561800 – 1775Approximate load on motor = 100 HP × 0.56 = 56 HP11.13 AMPERAGE READINGS 4Conditions1. Applied voltage must be within 5% of nameplaterating.2. You must be able to disconnect the motor fromthe load. (By removing V-belts or disconnecting acoupling).3. Motor must be 7-1/2 HP or larger, 3450, 1725 or1140 RPM.4. The indicated line amperage must be below the fullload nameplate rating.Procedure1. Measure and record line amperage with load connectedand running.2. Disconnect motor from load. Measure and recordthe line amperage when the motor is running withoutload.3. Read and record the motors nameplate amperagefor the voltage being used.4. Insert the recorded values in the following formulaand solve.(2 × LLA) – NLA(% Rated HP) = ———————— × 100(2 × NPA) – NLAWhere:LLA = Loaded Line AmpsNLA = No Load Line Amps (Motordisconnected from load)NPA = Nameplate Amperage (Foroperating voltage)Please note: This procedure will generally yield reasonablyaccurate results when motor load is in the 40 to100% range and deteriorating results at loads below40%.Example:• A 20 HP motor driving a pump is operating on 460volts and has a loaded line amperage of 16.5.• When the coupling is disconnected and the motoroperated at no load the amperage is 9.3.• The motor nameplate amperage for 460 volts is24.0.Therefore we have:Loaded Line Amps LLA = 16.5No Load Amps NLA = 9.3Nameplate Amps NPA = 24.0(2 × 16.5) – 9.3 23.7(%Rated HP) = —————— × 100 = —— × 100 = 61.2%(2 × 24.0) – 9.3 38.7Approximate load on motor = 20 HP × 0.612 = 12.24or slightly over 12 HP11.14 ELECTRIC MOTOR EFFICIENCYThe efficiency of a motor is the ratio of the mechanicalpower output to the electrical power input. Itmay be expressed as:

ELECTRIC ENERGY MANAGEMENT 285Output Input – Losses OutputEfficiency = ——— = —————— = ———————Input Input Output + LossesDesign changes, better materials, and manufacturingimprovements reduce motor losses, makingpremium or energy-efficient motors more efficient thanstandard motors. Reduced losses mean that an energyefficientmotor produces a given amount of work withless energy input than a standard motor. 3In 1989, the National Electrical Manufacturers Association(NEMA) developed a standard definition forenergy-efficient motors. 2How should we interpret efficiency labels?Efficiencies and Different Standards—————————————————————————Standard 7.5 HP motor 20 HP motor—————————————————————————International (IEC 34-2) 82.3% 89 4%British (BS-269) 82.3% 89.4%Japanese (IEC-37) 85.0% 90.4%U.S. (IEEE -112 Method B)* 80.3% 86.9%—————————————————————————The critical part of the efficiency comparison calculationsis that the efficiencies used must be comparable.The Arthur D Little report contained the followinginteresting statement: “Reliable and consistent data onmotor efficiency is not available to motor appliers. Datapublished by manufacturers appears to range from veryconservative to cavalier.”Recognizing that less than a 10 percent spread inlosses is statistically insignificant NEMA has set up efficiencybands. Any motor tested by IEEE - 112, MethodB, will carry the nominal efficiency of the highest bandfor which the average full load efficiency for the modelis equal to or above that nominal.The NEMA nominal efficiency is defined as theaverage efficiency of a large population of motors of thesame design. The spread between nominal efficiency inthe table based on increments of 10 percent losses. Thespread between the nominal efficiency and the associatedminimum is based on an increment of 20 percentlosses.11.14.1 The Following is ReprintedFrom NEMA MG 1-1987Efficiency (MG 1-12.54)Determination of Motor Efficiency andLosses (MG 1-12.54.1)Efficiency and losses shall be determined in accordancewith IEEE Std 112 Standard Test Proceduresfor Polyphase Induction Motors and Generators*. Theefficiency shall be determined at rated output, voltage,and frequency.Unless otherwise specified, horizontal polyphasesquirrel-cage medium motors rated 1 to 125 horsepowershall be tested by dynamometer (Method B) as describedin par. 5.2.2.4 of IEEE Std 112. Motor efficiency shall becalculated using MG 1-12.57 in lieu of Form E of IEEEStd 112. Vertical motors in this horsepower range shallalso be tested by Method B if bearing construction permits;otherwise they shall be tested by segregated losses(Method E) as described in par. 5.2.3.1 of IEEE Std 112,including direct measurement of stray-load loss.The following losses shall be included in determiningthe efficiency:1. Stator I 2 R.2. Rotor I 2 R.3. Core Loss.4. Stray load loss.5. Friction & windage loss. †6. Brush contact loss of wound-rotor machinesPower required for auxiliary items, such as externalpumps or fans, that are necessary for the operationof the motor shall be stated separately.In determining I 2 R losses, the resistance of eachwinding shall be corrected in a temperature equal to anNominalEfficiencyMinimumEfficiency93.6 ————— 20% Greater Losses → 92.4|10% Greater Losses↓93.0*See Referenced Standards, MG 1-1.01†In the case of motors which are furnished with thrust bearings, onlythat portion of the thrust bearing loss produced by the motor itselfshall be included in the efficiency calculation. Alternatively, a calculatedvalue of efficiency, including bearing loss due to external thrustload, shall be permitted to be specified.In the case of motors which are furnished with less than a full set ofbearing, friction and windage losses which are representative of theactual installation shall be determined by (1) calculations or (2) experiencewith shop tested bearings and shall be included in the efficiencycalculations.

286 ENERGY MANAGEMENT HANDBOOKambient temperature of 25°C plus the observed ratedload temperature rise measured by resistance. Whenthe rated load temperature rise has not been measured,the resistance of the winding shall be corrected to thefollowing temperature:————————————————————————————Class of Insulation System Temperature, Degrees C————————————————————————————A 75B 95F 115H 130—————————————————————————This reference temperature shall be used for determiningI 2 R losses at all loads. If the rated temperaturerise is specified as that of a lower class of insulationsystem, the temperature for resistance correction shallbe that of the lower insulation class.NEMA Standard 5-12-1975, revised 6-21-1979; 11-12-1981;11-20-1986; 1-11-1989.Efficiency Of Polyphase Squirrel-cageMedium Motors with Continuous Ratings(MG 1-12.54.2)The full-load efficiency of Design A and B singlespeedpolyphase squirrel-cage medium motors in therange of 1 through 125 horsepower for frames assigned inaccordance with NEMA Standards Publication No. MG 13,Frame Assignments for Alternating Current Integral-horsepowerInduction Motors, (see MG1-1.101) and equivalentDesign C ratings shall be identified on the nameplate bya nominal efficiency selected from the Nominal Efficiencycolumn in Table 11.2 (NEMA Table 12.6A) which shall benot greater than the average efficiency of a large populationof motors of the same design.The efficiency shall be identified on the nameplateby the caption “NEMA Nominal Efficiency” or “NEMANom. Eff.”The full load efficiency, when operating at ratedvoltage and frequency, shall be not less than the minimumvalue indicated in Column B of Table 11.2 (NEMATable 12.6A) associated with the nominal value in ColumnA.Suggested Standard for Future Design 3-16-1977,NEMA Standard 1-17-1980, revised 3-8-1983; 3-14-1991.The full-load efficiency, when operating at ratedvoltage and frequency, shall be not less than the minimumvalue indicated in Column C of Table 11.2 (NEMATable 12.6A) associated with the nominal value in ColumnA.Suggested Standard for Future Design 3-14-1991.Variations in materials, manufacturing processes,and tests result in motor-to-motor efficiency for a largepopulation of motors of a single design is not a uniqueefficiency but rather a band of efficiency. Therefore,Table 11.2 (NEMA Table 12.6A) has been established toindicate a logical series of nominal motor efficienciesand the minimum associated with each nominal. Thenominal efficiency represents a value which should beused to compute the energy consumption of a motor orgroup of motors.Authorized Engineering Information 3-6-1977, revised1-17-1980;1-11-1989.Efficiency Levels of Energy Efficient PolyphaseSquirrel-Cage Induction Motors (MG 1-12.55)The nominal full-load efficiency determined inaccordance with MG 1-12.54.1, identified on the nameplatein accordance with MG 1.12.54.2, and havingcorresponding minimum efficiency in accordance withColumn B of Table 11.2 (NEMA Table 12.6A) shall equalor exceed the values listed in Table 11.3 (NEMA Table12.6B) for the motor to be classified as “energy efficient.”NEMA Std 1-11-1989;3-14-1991.Efficiency Levels of Energy Efficient PolyphaseSquirrel-Cage Induction Motors (MG 1-12.55A)(SUGGESTED STANDARD FOR FUTURE DESIGN)The nominal full-load efficiency determined in accordancewith MG 1-12.54.1, identified on the nameplatein accordance with MG 1-12.54.2 and having minimumefficiency in accordance with Column C of Table 11.2(NEMA Table 12.6A) shall equal or exceed the valueslisted in Table 11.4 (NEMA Table 12.6C) for the motorto be classified as “energy efficient.”Suggested Standard for future design 9-5-1991.11.15 COMPARING MOTORSIt is essential that motor comparison be done on thesame basis as to type, size, load, cost of energy, operatinghours and most importantly the efficiency values such asnominal vs. nominal or guaranteed vs. guaranteed.The following equations are used to compare thetwo motors.For loads not sensitive to motor speed—Note: Replacing a standard motor with an energyefficientmotor in a centrifugal pump or fan applicationcan result in increased energy consumption if energy-efficientmotor operates at a higher RPM.

ELECTRIC ENERGY MANAGEMENT 287Table 11.2 (NEMA Table 12.6A)—————————————————————————Column A Column B* Column C † MinimumNominal Efficiency Minimum Efficiency Efficiencybased on 20% Based on 10% LossLoss Difference Difference—————————————————————————99.0 98.8 98.998.9 98.7 98.898.8 98.6 98.798.7 98.5 98.698.6 98.4 98.598.5 88.2 98.498.4 98.0 98.298.2 97.8 98.098.0 97.6 97.897.8 97.4 97.697.6 97.1 97.497.4 96.8 97.197.1 96.5 96.896.8 96.2 96.596.5 95.8 96.296.2 95.4 95.895.8 95.0 95.495.4 94.5 95.095.0 94.1 94.594.5 93.6 94.194.1 93.0 93.693.6 92.4 93.093.0 91.7 92.492.4 91.0 91.791.7 90.2 91.091.0 89.5 90.290.2 88.5 89.589.5 87.5 88.588.5 86.5 87.587.5 85.5 86.586.5 84.0 85.585.5 82.5 84.084.0 81.5 82.582.5 80.0 81.581.5 78.5 80.080.0 77.0 78.578.5 75.5 77.077.0 74.0 75.575.5 72.0 74.074.0 70.0 72.072.0 68.0 70.070.0 66.0 68.068.0 64.0 66.066.0 62.0 64.064.0 59.5 62.062.0 57.5 59.559.5 55.0 57.557.5 52.5 55.055.0 50.5 52.552.5 48.0 50.550.5 46.0 48.0—————————————————————————*Column B approved as NEMA Standard 3/14/1991† Column C approved as Suggested Standard for Future Designs 3/14/1991

288 ENERGY MANAGEMENT HANDBOOKTable 11.3 (NEMA Table 12.6B) Full-load efficiencies of energy efficient motors.

ELECTRIC ENERGY MANAGEMENT 289Table 11.4 (NEMA Table 12.6C) (Suggested standard for future design)Full-load efficiencies of energy efficient motors.

290 ENERGY MANAGEMENT HANDBOOKSame horsepower—different efficiency.kW saved =hp× 0.746 × 100E STD– 100E EESame horsepower and % load—different efficiencykW saved =hp× 0.746 × L × 100E STD– 100E EEAnnual $ savings due to difference in efficiencyS=hp× 0.746 × L × C × N × 100E STD– 100E EEExampleS = $ Savings (annual) 100hp = Horsepower 100L = % Load 100C = Energy cost ($/kWh) 0.08N = Operating house (annual) 4000E STD = % Efficiency of standard motor 91.7E EE = % Efficiency of energy eff. motor 95.0RPM STD = Speed of standard motor 1775RPM EE = Speed of energy eff. motor 1790For loads sensitive to motor speedAbove equations should be multiplied by speedratio correction factor.SRCF = Speed Ratio Correction Factor3RPM EERPM STDExample:S = 100 × 0.746 × 1 × 0.080 × 4000× (100/91.7-100/95.0) = $904S = 100 × 0.746 × 1 × 0.080 × 4000× (100/91.7-100/95.0) × (1790/1775) 3 = $262$ 642 reduction in expected savings.Relatively minor, 15 RPM, increase in a motor’s rotationalspeed results in a 2.6 percent increase in the loadplaced upon the motor by the rotating equipment.11.16 SENSITIVITY OF LOAD TO MOTOR RPMWhen employing electric motors, for air movingequipment, it is important to remember that the performanceof fans and blowers is governed by certainrules of physics. These rules are known as “The AffinityLaws” or “ The Fan Laws.” There are several parts to it,and are all related to each other in a known manner andwhen one changes, all others change. For centrifugalloads, even a minor change in the motor’s speed translatesinto significant change in energy consumption andis especially troublesome when the additional air flowis not needed or useful. Awareness of the sensitivity ofload and energy requirements to motor speed can helpeffectively identify motors with specific performance requirements.In most cases we can capture the full energyconservation benefits associated with an energy efficientmotor retrofits.Terminology of Load to Motor RPMCFM Fan capacity ( Cubic Feet per Minute)Volume of air moved by the fan per unit oftime.P Pressure. Pressure produced by the fan that canexist whether the air is in motion or confined ina closed duct.HP Horsepower. The power required to drive an airmoving device.RPM Revolutions Per Minute. The speed at which theshaft of air moving equipment is rotating.Affinity Laws or Fan LawsLaw #1 CFM 2CFM 1= RPM 2RPM 1Quantity (CFM) varies as fan speed (RPM)Law #2 P 2P 1= RPM 2 2RPM 12Pressure (P) varies as the square of fan speed (RPM)Law #3 HP 2HP 1= RPM 2 3RPM 13Horsepower (HP) varies as the cube of fan speed (RPM)ExampleFan system 32,000 CFMMotor 20 HP 1750 RPM (existing)Motor 20 HP 1790 RPM (new EE)

ELECTRIC ENERGY MANAGEMENT 291kW = 20 × 0.746 = 14.92 kWNew CFM with new motor = 1790/1750 × 32,000 = 32,731or 2.3% increaseNew HP = (1790/1750)3 × 20 × = 21.4 HP or 7% increase.New kW = 21.4 × 0.746 = 15.96 kW 7% increase in kW andwork performed by motor.Replacing a standard motor with an energy efficientmotor in centrifugal pump or a fan applicationcan result in increased energy consumption if the energyefficient motor operates at a higher RPM. Table 11.5shows how a 10 RPM increase can negate any savingsassociated with a high efficiency motor retrofit.11.17 THEORETICAL POWER CONSUMPTIONFigure 11.6 illustrates the energy saving potentialof the application of Adjustable Speed Drive to an applicationthat traditionally uses throttling control, suchas Discharge Damper, Variable Inlet Vane or Eddy CurrentDrive.From the standpoint of maximum energy conservation,the most optimal method to reduce fan CFM is toreduce the fan’s speed (RPM). This can be accomplishedby changing either, the sheaves of the motor, the sheavesof the fan or by varying fan motor speed.Applications Involving ExtendedPeriods of Light Load Operation 2A number of methods have been proposed toreduce the voltage applied to the motor in response tothe applied load, the purpose of this being to reducethe magnetizing losses during the periods when the fulltorque capability of the motor is not required. Typical ofthese devices is the power factor controller. The powerfactor controller is a device that adjusts the voltage appliedto the motor to approximate a preset power factor.These power factor controllers may, for example,be beneficial for use with small motors operating forextended periods of light loads where the magnetizationlosses are a relatively high percentage of the total loss.Care must be exercised in the application of these controllers.Savings are achieved only when the controlledmotor is operated for extended periods at no load orlight load.Particular care must be taken when consideringtheir use with other than small motors. A typical 10horsepower motor will have idle losses in the order of4 or 5 percent of the rated output. In this size range themagnetization losses that can be saved may not be equalto the losses added by the controller plus the additionalmotor losses caused by the distorted voltage wave forminduced by the controller.Applications Involving Throttling orBy-pass Control. 2Many pump and fan applications involve the controlof flow or pressure by means of throttling or bypassdevices. Throttling and bypass valves are in effect seriesand parallel power regulators that perform their functionby dissipating the difference between source energysupplied and the desired sink energy.These losses can be dramatically reduced by controllingthe flow rate or pressure by controlling thespeed of the pump or fan with a variable speed drive.Figure 11.6 illustrates the energy saving potentialof the application of variable speed drive to an applicationthat traditionally uses throttling control.A simplistic example will serve to illustrate the savingsto be achieved by the use of this powerful energyconservation tool.Assumptions:Line Power Required by Fan at full CFMwithout flow control device 100 HPFull CFM required1000 hours per year75% CFM required 3000 hours per year50% CFM required 2000 hours per yearCost of energy$.06 per kWhrFigure 11.6% Power consumption with various flow control methodsper Figure 11.6.

292 ENERGY MANAGEMENT HANDBOOKTable 11.5 Hourly operating costs.

ELECTRIC ENERGY MANAGEMENT 293100 HP MOTOR; 91.7%EFFICIENT;FL SPEED 1775 RPMOPERATES 1000 HOURS PER YEAR;ENERGY COST $0.08 PER kWh• COST TO OPERATE THIS MOTOR = $ 6,508100 HP MOTOR; 91.7%EFFICIENT;FL SPEED 1775 RPMOPERATES 1000 HOURS PER YEAR;ENERGY COST $0.08 PER kWhWITH EFFICIENCY IMPROVEMENTOF 5%from 91.7% to 96%COST TO OPERATE THIS MOTOR $ 6,198WITH INVESTMENT OF $ ??????$310 SAVINGS100 HP MOTOR; 91.7%EFFICIENT;FL SPEED 1775 RPMOPERATES 1000 HOURS PER YEAR;ENERGY COST $0.08 PER kWhWITH REDUCTION IN OPERATINGHOURSOF 5%from 1000 to 950COST TO OPERATE THIS MOTOR= $6,183$325 SAVINGSWITH INVESTMENT OF $ ??????100 HP MOTOR; 91.7%EFFICIENT;FL SPEED 1775 RPMOPERATES 1000 HOURS PER YEAR;ENERGY COST $0.08 PER kWhWITH LOAD SPEED REDUCTIONOF 5%from 1775 to 1686COST TO OPERATE THIS MOTOR $ 5,580$928 SAVINGSWITH INVESTMENT OF $ ??????

294 ENERGY MANAGEMENT HANDBOOKAnnual cost of Energyhrs x hp x 746 x % energy consumption x $/kW$ = ——————————————————————1000 x 100Annual Cost of Energy Summary:Discharge Damper $24,600Variable Inlet Vane $19,600Eddy Current Drive $15,900Adjustable Speed Drive $13,90011.18 MOTOR EFFICIENCY MANAGEMENTMany think that when one is saying Motor Effi -ciency the logical word to follow is improvement. Whereare we going? How far can we push manufacturers inour quest for the perfect motor?During the 102nd Congress, the Markey Bill, H.R.2451, was introduced. The bill mandated componentefficiency standards for such products as lighting, distributiontransformers and electric A.C. motors.This plan was met with opposition by NEMA andother interested groups. They called for a system approachthat would recognize the complex nature of theproduct involved under the plan. The bill passed by theEnergy & Power Subcommittee on the theory that theelimination of the least efficient component from themarket would ensure that consumers would purchaseand use the most efficient products possible.Although motors tend to be quite efficient inthemselves, several factors can contribute to cost-effectivereplacement or retrofit alternatives to obtainefficiency gain in motors. We are well aware that theelectric motor’s primary function is to convert electricalenergy into mechanical work. It is also importantto remember that good energy management requires aconsideration of the total system of which the motor is apart.Experience indicates that despite heightenedawareness and concern with energy efficiency, the electricmotor is either completely neglected or decisionsare made on the basis of incomplete information. At thispoint I would like to quote me.“Motors Don’t Waste Energy, People Do.”What this really means is that we must start managingefficiency and not just improving the motor. Thisis what will improve your corporate bottom line.11.19 MOTORS ARE LIKE PEOPLEMotors can be managed the same way and withthe same skills as people. There are amazing similarities.I have spent years managing both and find there is verylittle difference between the two.The expectations are the same for one as for theother. The employee’s performance is evaluated toidentify improvement opportunities linking them to organizationalgoals and business objectives. The managermeasures the performance of an employee as an individualand as a member of a team. Why then would itnot work the same with your motors? Motors employedjust as people are and they work as an individual or asa team. They will perform their best if cared for, maintained,evaluated and rewarded.An on going analysis of motor performance preventsmajor breakdown. Performance evaluation of amotor should be done as routinely as it is done on anemployee. Both the motor and an employee are equallyimportant. Applied motor maintenance will keep thebuilding or plant running smoothly with minimal stresson the system or downtime due to failure.MAXIMIZE YOUR EFFECTIVENESSWITH MOTOR EVALUATION SYSTEM11.20 MOTOR PERFORMANCEMANAGEMENT PROCESS (MPMP)The Motor Performance Management Process(MPMP) is designed to be the Motor Manager’s primarytool to evaluate, measure and most importantly manageelectric motors. MPMP focuses on building a strongerrelationship between Motor Manager and the electricmotor employed to perform a task. Specifically, it is alogical, systematic and structured approach to reduceenergy waste. Energy waste reduction is fundamental inbecoming more efficient in an increasingly competitivemarket. The implementation of MPMP is more than agood business practice it is an intelligent managementresource.→NEGLECTING YOUR MOTORSCAN BE A COSTLY MISTAKE←Motor Managers must understand motor efficiency,how it is achieved and how to conduct an economicevaluation. Timely implementation of MPMP would bean effective way to evaluate existing motor performancein a system and identify improvement opportunitieslinking them to organizational goals. The following

ELECTRIC ENERGY MANAGEMENT 295Motor Manager model defines the task of managingand enabling motors to operate in the ways required toachieve business objectives.Motor Desired Business————— ⇒ Motor ⇒ ————— ⇒ —————Manager Operation ObjectivesIt is vital to evaluate the differences between motorsoffered by various manufacturers and only choosethose that clearly meet your operating criteria. An investmentof 20 to 25% for an Energy Efficient motor,over the cost of a standard motor, can be recovered in arelatively short period of time. Furthermore, with somemotors the cost of wasted energy exceeds the originalmotor price even in the first year of operation.11.21 HOW TO START MPMPBegin by conscientiously gathering information oneach motor used in excess of 2000 hours per year.Complete a MOTOR RECORD FORM (Figure 11.7) foreach motor. (A detailed profile of your existing Motor.)• This form must be established for each motor toserve as a performance record. It will help youto understand WHEN, WHERE, and HOW yourmotors are used, and identify opportunities toimprove drive system efficiency.• Each item in this form must be addressed, payingparticular attention to the following items:Motor Location, Application, Energy Cost $/kWh,Operating Speed, Operating Load and Nameplateinformation.• Motors with special electrical designs or mechanicalfeatures should be studied carefully. Some applicationsrequire special attention, such as fans,compressors and pumps.• Motors with a history of repeated repair should beof special interest.• Examine your completed MOTOR RECORDFORM and select the best candidates for possibleretrofit or future replacement.• Check with your local utility regarding the availabilityof financial incentives or motor rebates.Finding the right motor for the application, andcalculating its energy and cost savings can be done withthe MotorMaster 5 software and WHAT IF motor comparisonform Figure 11.8. This form has the capability tocompare several motors and analyze potential savings.System efficiency is theproduct of individual efficiencies.For example, out of$5.00 total motor losses(Scenario 3) only 69 centsis from the motor itself.The belt and fan accountfor $3.89 + .36 = $4.25 ofthe total losses.(The kW amounts shownin the table in the graphicare the sums of the previouscomponents.)

296 ENERGY MANAGEMENT HANDBOOKFigure 11.7 Motor record form.

ELECTRIC ENERGY MANAGEMENT 297Figure 11.8 WHAT IF motor comparison form.

298 ENERGY MANAGEMENT HANDBOOKHOW TO GET AROUND IN THE ‘WHAT IF’ FORM.COLUMNLINE EXPLANATIONEXISTING 1-9 Information can be taken from the Motor Record Form if previously generated. If not available, generatedata.EXISTING 1 ENERGY COST $/kWh Self ExplanatoryEXISTING 2 OPERATING HOURS PER YEAR is very important to be as accurate as possibleEXISTING 3 MOTOR HORSEPOWER Self ExplanatoryEXISTING 4 NO LOAD SPEED RPM (synchronous speed) is usually within 5% of Full Load Speed i.e. 900, 1200, 1800,or 3600 rpm.EXISTING 5 FULL LOAD SPEED is found on the nameplate. This information is important for loads sensitive to motorspeed.EXISTING 6 EFFICIENCY @ 100% LOAD NEMA % efficiency at full load. If motor is relatively new this will be foundon the nameplate, if older, it will be necessary to contact the manufacturer or the MotorMaster data base.(WSEO)EXISTING 7-8 EFFICIENCY @ 75% AND 50% LOAD This information can be obtained from the manufacturer or theMotorMaster data base. (WSEO)EXISTING 9 INVESTMENT/SALVAGE This should include total cost associated with purchase of a motor, such ascost of motor installation, balancing alignment and disposition of existing motor.PROPOSAL 5-9 Information is acquired from the motor manufacturers catalogs or the MotorMaster data base1-5 which contains information on more than 10,000 motors from various manufacturers.EXISTING10-21 The data is calculated using the formulas in the column entitledFORMULA or is automatically calculated if using the What If spreadsheet.PROPOSAL 10-39 The data is calculated using the formulas in the column entitled1-5 FORMULA or is automatically calculated if using the What If spreadsheet.FORMULA10-39 The formulas used for calculating. These formulas may also be used tocreate a spreadsheet similar to ‘What If.’11.22 NAMEPLATE GLOSSARYHP—The number of, or fractional part of a horsepower,the motor will produce at rated speed.RPM—An indication of the approximate speed that themotor will run when it is putting out full rated outputtorque or horsepower is called full load speed.VOLTS—Voltage at which the motor may be operated.Generally, this will be 115 Volts, 230 Volts, 115/230V, or 220/440 V.AMPS—The amount of current the motor can be expectedto draw under full load (torque) conditionsis called full load amps. It is also known asnameplate amps.HZ—Frequency at which the motor is to be operated.This will almost always be 60 Hertz.SERVICE FACTOR—The service factor is a multiplierthat indicates the amount of overload a motor canbe expected to handle. For example, a motor with a1.15 service factor can be expected to safely handleintermittent loads amounting to 15% beyond itsnameplate horsepower.DESIGN—The design letter is an indication of the shapeof the torque speed curve. They are A. B. C andD. Design B is the standard industrial duty motorwhich has reasonable starting torque with moderatestarting current and good overall performancefor most industrial applications. Design C is usedfor hard to start loads and is specifically designedto have high starting torque. Design D is the socalled high slip motor which tends to have veryhigh starting torque but has high slip rpm at fullload torque. Design A motors are not commonlyspecified but specialized motors used on injectionmolding applications.FRAME—Motors, like suits of clothes, shoes and hats,come in various sizes to match the requirementsof the application.

ELECTRIC ENERGY MANAGEMENT 299INSULATION CLASS—The insulation class is a measureof the resistance of the insulating componentsof a motor to degradation from heat. Four majorclassifications of insulation are used in motors.They are, in order of increasing thermal capabilities,A, B, C, and F.TYPE—Letter code that each manufacturer uses to indicatesomething about the construction and thepower the motor runs on. Codes will indicate splitphase, capacitor start, shaded pole, etc.CODE—This is a NEMA code letter designating thelocked rotor kVA per horsepower.PHASE—The indication of the type of power supply forwhich the motor is designed. Two major categoriesexist; single phase and three phase.AMB. DEG. C.— Ambient temperature is the maximumsafe room temperature surrounding the motor if itis going to be operated continuously at full load. Inmost cases, the standardized ambient temperaturerating is 40 degrees C (104 degrees F).TEMPERATURE RISE— Temperature rise is the amountof temperature change that can be expected withinthe winding of the motor from non-operating (coolcondition) to its temperature at full load continuousoperating condition. Temperature rise is normallyexpressed in degrees centigrade.DUTY—Most motors are rated for continuous dutywhich means that they can operate at full loadtorque continuously without overheating. Motorsused on certain types of applications such as wastedisposal, valve actuators, hoists, and other typesof intermittent loads, will frequently be rated forshort term duty such as 5 minutes, 15 minutes, 30minutes, or 1 hour.EFF. INDEX OR NEMA %— Efficiency is the percentageof the input power that is actually convertedto work output from the motor shaft. Efficiencyis now being stamped on the nameplate of mostdomestically produced electric motors.SummaryOver 60 percent of electricity in the United Statesis consumed by electric motor drive systems. Generally,a motor drive system consists of several components; apower supply, controls, the electric motor and the mechanicaltransmission system. The function of an electricmotor is to convert electric energy into mechanical work.During the conversion the only power consumed bythe electric motor is the energy losses within the motor.Since the motor losses are in the range of 5-30% of theinput power, it is important to consider the total systemof which the motor is a part.This chapter deals mostly with electric motor drivesystems and provides practical methods for managingmotors.One of the “MotorManager” methods is the MotorPerformance Management Process (MPMP) whicheffectively evaluates the performance of existing motorsand identifies opportunities to link them to organizationalgoals. It is a primary tool to evaluate, measureand manage motors and a logical, systematic, structuredapproach in reducing energy waste, fundamental to efficiencyin a competitive market. With minor changes,this process can be used to evaluate other electrical andmechanical equipment.Considerable attention must be paid to the efficienciesof all electric equipment being purchased today.This is true not only for motors but for transformers ofall types and other electrical devices.To guard against the waste of electrical energy,manufacturers of dry type transformers are designingthem with lower than normal conductor and total losses.The reduction in these losses also lowers the temperaturerise of the transformer resulting in improved lifeexpectancies as well as a reduction in the air conditioningrequirements.References1. Technology Update Produced by the Washington State EnergyOffice for the Bonneville Power Administration.2. NEMA Standards publication No. MG 10. Energy ManagementGuide for Selection and Use of Polyphase Motors. National ElectricalManufacturers Association, Washington, D.C. 19893. McCoy, G., A. Litman, and J. Douglass. Energy Efficient ElectricMotor Selection Handbook. Bonneville Power Admin. 19914. Ed Cowern Baldor Electric Motors.5. MotorMaster software developed by the Washington StateEnergy Office (WSEO) puts information on more than 10,000three-phase motors at your fingertips. The MotorMaster lets youreview features and compare efficiency, first cost, and operatingcost. The U.S. Department of Energy (DOE) and the BonnevillePower Administration (BPA) funded the development of thesoftware. You may obtain this valuable software or more detailedinformation by contacting WSEO (360) 956-20006. Financing for Energy Efficiency Measures, Virginia Power, POST-AND, REV(0)08-17-93.7. William E. Lanning and Steven E. Strauss Power Survey Techniquesfor electrical loads Esterline Angus Instrument Corp.8. Konstantin Lobodovsky, "Efficient Application of AdjustableSpeed Drives," Motor Manager, Penn Valley, CA, 1999 (530) 432-2237

300 ENERGY MANAGEMENT HANDBOOKAppendixELECTRONIC ADJUSTABLE SPEED DRIVES:ISSUES AND APPLICATIONS*CLINT D. CHRISTENSONOklahoma Industrial Assessment CenterSchool of Industrial Engineering and Management322 Engineering NorthOklahoma State UniversityStillwater, OK 74078INTRODUCTIONElectric motors are used extensively to drive fans,pumps, conveyors, printing presses, and many otherprocesses. A majority of these motors are standard, 3-phase, AC induction motors that operate at a singlespeed. If the process (fan, pump, etc.) is required to operateat a speed different than the design of the motor,pulleys are applied to adjust the speed of the equipment.If the process requires more than one speed during itsoperation, various methods have been applied to allowspeed variation of a single speed motor. These methodsinclude variable pitch pulley drives, motor-generatorsets, inlet or outlet dampers, inlet guide vanes, and VariableFrequency Drives. The following section will brieflydiscuss each of these speed control technologies.Changing the pulleys throughout the day to followdemand is not feasible, but the use of a “Reeves” typevariable pitch pulleys drive was a common application.These drives utilized a wide belt between two pairsof opposing conical pulleys. As the conical pulleys ofthe driven shaft were brought together (moved apart)the pulley diameter would increase (decrease) and decrease(increase) the belt speed and the process speed.This type of system is still extensively in use in thefood and chemical industries where mixing speeds candramatically effect product quality. These systems are amechanical speed adjustment system which has inherentfunction losses and require routine maintenance.Motor-Generator sets were used in the past to convertincoming electricity to a form required in the processincluding changes from AC to DC. The DC outputcould then be used to synchronize numerous dc motors*Facilities today generally have significant harmonics. Especiallywhen capacitors are used in systems with harmonics andvariable frequency drives, professional advice is needed. Alsomotor capability may be a problem. See (8) page 71 for moreinformationat the required speed. This was the common type ofspeed control in printing presses and other “web” typesystems. The use of an Eddy-current clutch would varythe output of the generator to the specific needs of thesystem. The windage and other losses associated withmotors are at least doubled with the generator and theefficiency of the system drops drastically at low loadsituations.A majority of the commercial and industrial fanand pump speed control techniques employed do notinvolve speed control at all. These systems utilize inletdampers, outlet dampers (valves) with or withoutbypass, or inlet guide vanes to vary the flow to theprocess. Inlet dampers, used in fan applications, reducethe amount of air supplied to the process by reducingthe inlet pressure. Outlet dampers (valves) maintainsystem pressure (head) seen by the fan (pump) whilereducing the actual volume of air (liquid) flowing. Inletguide vanes are used in fans similar to inlet dampersbut the guide vanes are situated such that as air flow isreduced, the circular motion of the fan is imparted uponthe incoming air. Each of these control methods operatesto reduce the amount of flow with some reduction inenergy required.Variable Frequency Drives (VFD) change the speedof the motor by changing the voltage and frequency ofthe electricity supplied to the motor based upon systemrequirements. This is accomplished by converting theAC to DC and then by various switching mechanismsinvert the DC to a synthetic AC output with controlledvoltage and frequency [Phipps, 1994]. If this processis accomplished properly, the speed of the motor canbe controlled over a wide variation in shaft speed (0rpm through twice name plate) with the proper torquecharacteristics for the application. The remainder of thispaper will discuss the various issues and applications ofVFD’s. Figure 1 shows the percent power curves versuspercent load for simplified centrifugal air handling fanapplication.VARIABLE FREQUENCY DRIVE TYPESIn order to maintain proper power factor andreduce excessive heating of the motor, the name platevolts per hertz ratio must be maintained. This is themain function of the variable frequency drive (VFD).The four main components that make up AC variablefrequency drives (VFD’s) are the Converter, Inverter,the DC circuit which links the two, and a control unit,shown in Figure 2. The converter contains a rectifierand other circuitry which converts the fixed frequency

ELECTRIC ENERGY MANAGEMENT 301Figure 1. Typical power consumption of various fancontrol systems. (Source: Moses et. al., 1989)AC to DC. The inverter converts the DC to an adjustablefrequency, adjustable voltage AC (both must beadjustable to maintain a constant volts to hertz ratio).The DC circuit filters the DC and conducts the DC tothe inverter. The control unit controls the output voltageand frequency based upon feedback from the process(e.g. pressure sensor). The three main types of inverterdesigns are voltage source inverters, current source inverters,and pulse width modulation inverters. Each willbe briefly discussed in the next section.Inverter TypesThe voltage source inverters (VSI) use a siliconcontrolled rectifier (SCR) to rebuild a pseudosine waveform for delivery to the motor. This is accomplishedwith a six-step voltage inverter with a voltage sourceconverter and a variable voltage DC bus. As with anySCR system, troublesome harmonics are reflected to thepower source. Also the six-step wave form sends currentin pulses which can cause the motor to cog at low frequencies,which can damage keyways, couplings, pumpimpellers, etc. [Phipps, 1994].The current source inverters (CSI) also use an SCRinput from the power source but control the current to themotor rather than the voltage. This is accomplished witha six-step current inverter with a voltage source converterand a variable voltage DC bus. The CSI systems havethe same problems with cogging and harmonics as theVSI systems. Many manufacturers only offer VSI or CSIVFD’s for larger horsepower sizes (over 300 HP).The pulse width modulation (PWM) VFD’s havebecome the state of the art concept in the past severalyears, starting with the smaller hp sizes, and available upto 1500 hp from at least one manufacturer [Phipps, 1994].PWM uses a simple diode bridge rectifier for power inputto a constant voltage DC bus. The PWM inverter developsthe output voltage by producing pulses of varying widthswhich combine to synthesize the desired wave form. Thediode bridge significantly reduces the harmonics fromthe power source. The PWM system produces a currentwave form that more closely matches the power linewave form, which reduces adverse heating. The PWMdrives also have the advantage of virtually constant powerfactor at all speeds. Depending on size, it is possibleto have power factors over 95% [Phipps, 1994]. Anotheradvantage of the PWM VFD’s is that sufficient currentfrequency (~200+ Hz) is available to operate multiplemotors on a single drive, which would be advantageousfor the printing press example discussed earlier. The nextFigure 2. Typical VFD system. (Source: Lobodovsky, 1996)

302 ENERGY MANAGEMENT HANDBOOKsection will discuss the types of loads that require adjustablespeeds that may be controlled by variable frequencydrives.VARIABLE SPEED LOADSThe three common types of adjust able speedloads are variable torque, constant torque, and constanthorsepower loads. A variable torque load requires muchlower torque at low speeds than at high speeds. Withthis type of load, horsepower varies approximately asthe cube of speed and the torque varies approximatelyas the square of the speed. This type of load is used inapplications such as centrifugal fans, pumps, and blowers.A constant torque load requires the same amountof torque at low speed as at high speed. The torqueremains constant throughout the speed range, and thehorsepower increases or decreases in direct proportionto the speed. A constant torque load is used in applicationssuch as conveyors, positive displacement pumps,some extruders, and for shock loads, overloads, or highinertia loads. A constant horsepower load requires hightorque at low speeds, and low torque at high speeds,and therefore constant horsepower at any speed. Constanthorsepower loads are encountered in most metalcutting operations, and some extruders [Lobodovsky,1996]. The savings available from non-centrifugal (constanttorque or constant horsepower) loads are basedprimarily on the VFD’s high efficiency (when comparedto standard mechanical systems), increased power factor,and reduced maintenance costs. The next sectionwill discuss several applications of VFD’s and the issuesinvolved with the application.Figures 3 and 4 show the energy savings potential fora variable speed fan and pump, respectively. The VFDpump is compared with a valve control system whichwould be adjusted to maintain a constant pressure in thesystem. The VFD fan is compared with both the dampercontrol and the inlet guidevane controls. These curvesdo not account for system characteristics (i.e., head orstatic pressure), which would need to be included inan actual design. These figures show that the amountof savings achievable from the VFD is based upon theFigure 3. Energy savings with VFD fan. (Source: Lobodovsky,1996)VARIABLE FREQUENCY DRIVE APPLICATIONSVariable frequency drive systems offer many benefitsthat result in energy savings through efficient andeffective use of electric power. The energy savings areachieved by eliminating throttling, performance, andfriction losses associated with other mechanical or electromechanicaladjustable speed technologies. Efficiency,quality, and reliability can also be drastically improvedwith the use of VFD technology. The application of aVFD system is very load dependent and a thoroughunderstanding of the load characteristics is necessary fora successful application. The type of load (i.e. Constanttorque, variable torque, constant horsepower) shouldbe known as well as the amount of time that the systemoperates (or could operate) at less than full speed.Figure 4. Savings with VFD pump. (Source: Lobodovsky,1996)

ELECTRIC ENERGY MANAGEMENT 303percent volume flow for both cases. The consumptionsavings would be determined by the percent of time ata particular load multiplied by the amount of time atthat particular load.Application IdentificationThere are many instances where a VFD can besuccessfully applied. The situation where the existingequipment already utilizes one of the older speed controltechnologies is the easiest to identify. In the case ofthe printing press that utilizes an Motor-generator setwith a Eddy-current clutch, a single VFD and new ACmotors could replace the MG-set and the DC motors. Amixer utilizing variable pitch pulleys (“Reeves” drive)could be replaced with a VFD which would reduceslippage losses and could dramatically improve productquality (through better control) and reliability (solidstate versus mechanical).Equipment (fans or pumps) utilizing dampers orvalves to reduce flow is another instance where a VFDmay provide a better means of control and energy savings.A variable volume air handler that utilizes damperor inlet guide vane controls could be replaced with a VFDdrive controller. This would reduce the amount of energyrequired to supply the required amount of air to the system.Chiller manufacturers have utilized VFD controllersto replace the standard butterfly inlet valves on centrifugalcompressors. This can significantly reduce the partload power requirements of a chiller which occur for amajority of the operating cycle in most applications.Application AnalysisProject evaluation methods depend in large part onthe size of the project. Smaller and less complex projectsmay only require reviewing specifications, installationsketches and vendors’ quotes. Larger, more complexprojects require more detailed engineering drawings anda drive system specialist will need to review the plan [Lobodovsky,1996]. Once a possible application for a VFDis identified, the load profile (percent load versus time)should be determined. This curve(s) could be developedwith the use of demand metering equipment or processknowledge (less desirable). This curve will be used to determinethe available savings to justify the project as wellas assure that the motor is properly sized. The properload profile (constant torque, variable torque, etc.) can becompared to that loads corresponding VFD load profilein order to develop savings potential. The motor must beevaluated to assure that the VFD is matched to the motorand load, determine the motor temperature requirements,that the minimum motor speeds are met, amongothers. Many VFD manufacturers will require that theexisting motor be replaced with a new model to assurethat the unit is properly sized and not affected by earliermotor misuse or rewinding. At this stage the expertise ofa VFD analyst or sales representative should be broughtinto the project for further design issues and costs. Phippsincludes comprehensive chapters on applying drives tovarious applications where several check lists are included.The next section includes two case studies of applicationsof VFD’s in industry.CASE STUDIESThe following case studies are included as anexample of possible VFD applications and the analysisprocedures undertaken in the preliminary systemsanalysis.Boiler Combustion FanThis application involves the use of a VFD to varythe speed of a centrifugal combustion air intake fan (50nameplate horsepower) on a scotch marine type highpressure steam boiler. The existing system utilizes anactuator to simultaneously vary the amount of gas andair that enters the burner. The air is controlled with theuse of inlet dampers (not guide vanes). As the amountof “fire” is reduced, the damper opening is reduced andvisa versa. The centrifugal fan and continuous variationin fire rate make this a feasible VFD application.The load profile of the boiler and corresponding motordemand (measured with demand metering equipment)is listed in Table 1. This table also includes the correspondingVFD demand requirements (approximatedfrom Figure 4), kW savings, and kWh savings.The annual savings for this example totaled 88,000kWh, which would equate to an annual savings of$4,400 (based upon a cost of energy of $0.05/kWh). Lobodovskyprovides an average estimated installed costof VFD’s in this size range at around $350 per horsepower,or an installed cost of $17,500 (50 hp * $350/hp). Thiswould yield a simple payback of around 4 years. Thisexample does not take into account demand savingswhich may result if the demand reduction correspondswith the plant peak demand. The control of the fan VFDwould be able to utilize the same output signal that theexisting actuator does.Industrial Chiller PlantA different calculation procedure will be used inthe following example. A malting plant in Wisconsinuses seven 550 ton chillers to provide cold water forprocess cooling. Three of the chillers work all of the time

304 ENERGY MANAGEMENT HANDBOOKTable 1. Boiler combustion fan load profile and VFD SavingsOperating Time Percent Full Existing Load w/ Load with VFD Kilowatt Kilowattatload (hours) Load Power 2 Damper (kW) 1 Control (kW) 2 Savings Hour Savings1000 100 37 37 0 02000 105 40 30 10 20,0003000 95 35 20 15 45,0001000 90 33 10 23 23,0001 Measured with a Demand Meter2 Approximated using Figure 488,000 kWhand the other four are operated according to the plant’svarying demand for cooling. The chillers are currentlycontrolled by variable inlet vanes (VIV).The typical load diversity and the power inputrequired for one of the chillers under varying load wereobtained from the manufacturer. In addition, the datain Table 2 give the power input of a proposed variablefrequency drive (also referred to as adjustable speeddrive (ASD)) for one chiller that operates about 6,000hours per year.A weighted average of the load and duty-cyclefractions gives a load diversity factor (ldf) of 64.2 percent.This means that, on the average, the chiller operatesat 64 percent of its full load capacity. A duty-cyclefraction weighted average of the savings attainable bya VFD can be estimated, which is 26.6 percent savingsper year. The energy savings due to avoided cost ofTable 2. Centrifugal chiller load and power input(Source: updated from Moses et Al., 1989)—————————————————————————Load VIV ASD Savings Duty CycleFraction % kW % kW % kW Fraction(1) (2) (3) (4) (5)—————————————————————————0.3 41 17 24 0.080.4 53 22 31 0.100.5 66 33 33 0.130.6 75 40 35 0.180.7 82 54 28 0.220.8 87 63 24 0.150.9 92 78 14 0.091.0 100 100 0 0.05—————————————————————————Column 2. From typical compressor performance with inlet and vaneguide control (York Division of Borg-Warner Corp).Column 3. From Carrier Corporation’s Handbook of Air ConditioningDesign: Comparative Performance of CentrifugalCompressor Capacity Control Methods.Column 5. Actual performance data.—————————————————————————electricity usage are computed as follows: Savings =(0.266)(550 ton)(0.7 kW/ton)(0.642)(6,000 hr/yr) = ldf=394,483 kWh/year The dollar savings at $0.04 per kWhare (394,483 kWh/year)($0.04/kWh) = $15,779/year. Fora 500 horsepower variable frequency drive, the installedcost is estimated at $75,000, based upon Lobodovskyaverage installed cost of $150 per ton for units of thissize. Therefore, the payback period is:($75,000)/($15,779/year) = 4.75 years.VFD ATTRIBUTESThese are just a few of the examples to show savingscalculations. The advantages of VFD’s include otheraspects beyond energy savings, including improvedproductivity, reduced maintenance costs, and improvedproduct quality, among others. The application of VFDsis relatively straight forward but a thorough analysis ofthe existing system and design of the future system isnecessary to assure successful application. The load profileof the existing system is necessary to both determinesavings as well as assure the system is properly sized.The specification of the actual VFD should only be doneby VFD suppliers and/or experts. The list of referencesis a good source of more detailed discussions of each ofthe points discussed in this paper.Attached are some additional forms and informationon variable speed drives. These are reproducedfrom a forthcoming book by Mr. KonstantinLobodovsky.ReferencesLobodovsky, Konstantin K., Motor Efficiency Management & ApplyingAdjustable Speed Drives, April 1996.Moses, Scott A., Wayne C. Turner, Jorge Wong, and Mark Duffer,“Profit Improvement With Variable Frequency Drives, EnergyEngineering. Vol. 86, No. 3, 1989.Phipps, Clarence A. Variable Speed Drive Fundamentals Lilburn, GA, TheFairmont Press, 1 994.

ELECTRIC ENERGY MANAGEMENT 305Typical ASD Equipment CostsTypical ASD Installation CostsTypical ASD Equipment and Installation Costs


ELECTRIC ENERGY MANAGEMENT 307

308 ENERGY MANAGEMENT HANDBOOKALTERNATIVE METHOD OF VARYINGTHE SPEED OF THE DRIVEN LOADPAUL DEWEY BOGGS, IIIK.K. LOBODOVSKYINTRODUCTIONVariable frequency drives (vfd’s), or inverters,have become the universally accepted method of variablespeed control of AC motors and their respectiveloads. This paper addresses an alternative approach toreliable variable speed control of the load and providesa brief look at the developmental history of the magnetic-coupledvariable speed drive, the latest advancesin the technology, the benefits, and the application ofthis technology with variable torque loads such as centrifugalfansBASIC PRINCIPLE OF OPERATIONThe magnetic coupled concept differs vastly fromvariable frequency drives in that there is no electricalpower interruption to the motor. With the motor runningcontinuously, precise speed control is accomplishedby varying the magnetic coupling between the motor’soutput shaft and the load. Most configurations are comprisedof two primary elements:Electromagnet (multi-pole rotor) mounts onto the shaftof the motor.Steel drum (armature/pulley)Drives Basic PartsOne element (input portion) is affixed to the motorsshaft so as to run continuously. The other element(output portion) will have a connection to the drivenload. These two elements are separated by an air gap,and have no other mechanical connection other thensupportive bearings. By applying current to the coilof the electromagnet rotor, a polarized magnetic fieldis produced, creating eddy currents on the surface ofthe drum, magnetically coupling both components andcausing the output portion to turn in the same directionas the motor. The speed of the output is dependent onthe strength of the magnetic field which is proportionatelycontrolled by the amount or current applied to theelectromagnet.EVOLUTION OFTHE MAGNETIC COUPLED DRIVEFoot Mounted StyleIn the 1940’s and 50’s, magnetic coupled devicesknown as eddy current clutches were effectively usedwith AC motors and were quickly becoming a popularmethod of varying the speed of many industrial loads.Although bulky end inefficient, these workhorses werequite reliable and were used in applications such aspunch presses, conveyers, winders, end other machinetool situations. These were oversized foot mountedunits that initially were designed as a separately housedclutch assembly with an input shaft end en output shaftto be coupled in line between the motor and the load.Also offered was motor and clutch combination (onepiece) packaged units. In those days, the primary focuswas on functionality, performance and maintainability,as energy efficiency was not as important a factor as itis today.Shaft-Mounted StylesIn the 1960’s some of the first commercially availablemotor shaft-mounted magnetically coupled driveswere offered to the industry. This new design wasintended for fractional and small integral horsepowerapplications, and was novel in that the drive was totallysupported by the motor shaft. The product didhowever have some drawbacks, namely oversized sliprings end problems with brush alignment. It was theuser's responsibility to align the brush holder with theslip rings. The intent was to mount the brush holderto the existing motor bolts. Unfortunately, motors suppliedby different manufacturers varied significantly andalignment became difficult. Additionally, the slip ringswere provided on the drive at the motor shaft entry side,

ELECTRIC ENERGY MANAGEMENT 309fabricated on a circuit board type material with copperrings racing the motor. Since the diameter of the outerring was larger than the inner ring, the outer brushwould wear out faster. In addition to the uneven endrapid brush wear, the integrity or the circuit board andcopper rings were effected by heat, causing separationof the rings from the base material. This design wasabandoned soon after initial production.In the 1980’s the problem with brush alignmenthad been somewhat resolved by a new shaft-mounteddesign that incorporated a bracket supported by anadditional bearing on the drive which maintained reasonablealignment between the brush holder and theslip rings. This design enjoyed some success in themachine tool industry where reduced run-time hourswas common this basic design was still flawed, as theslip rings were still located in the motor shaft entryside, causing them to be oversized and progressivelylarger to accommodate the higher horsepower motorshafts. This created a major headache in the HVAC airhandler marketplace, as it was soon discovered that 24-hour duty meant brush changes in some cases as oftenas once every two to three months on the larger drives.Although an improvement over previous efforts, thelocation or the bearings being cantilevered to the pulleygrooves, caused premature bearing failure in manyinstances. The additional bearing used to accommodatethe brush holder had an unacceptability high failure rateas well. This high maintenance drive has become virtuallyobsolete in the air handler industry and has beenroutinely replaced by the more efficient and reliable newbrushless designs.Another design consideration places the pulleygrooves out or the outboard side of the drive, a distanceaway from the motor face. This, by far is the poorest orall approaches because it directly jeopardizes the motorbearings’ life expectancy. Since the pulley grooves arenot located over the nema shaft extension, applying fullrated belt tension exceeds the overhung load rating ofthe motor in many cases. This outboard pulley designhas many documented failures in the field, again compoundedby drive hearing failure due to cantileveringeffect, and motor bearing damage as well.Preferred DesignThe most reliable and field proven design distinguishesitself in many ways from the previous concepts.The rotor/coil assembly rotates constantly with themotor shaft. A one-piece drum/pulley portion is theoutput-driving member. The pulley grooves are locatedin-board, closer to the motor face than any other design.The drives’ bearings are located directly under the pulleygrooves so that maximum belt tension can be appliedcontinuously on all models without harming thedrive bearings end yet remains well under the overhungload capacities of the motor. Because the drum is copperlined, the brushless drive runs cooler, and is moreefficient than other magnetic-coupled drives. The drivecoil requires only one third to one fourth the wattageof other models. It has the fewest parts, weighs lessend has the best operating performance or all availabledesigns the unblemished track record approaching fiveyears allows for the longest drive warranty that is availablein the industry, matching the full five year motorwarranty.ENERGY SAVINGSQ. How Does The Magnetic-Coupled Drive Save EnergyIf The Motor Runs Continuously?Because of the nature of the descending torque loaditself, the magnetic-coupled drive takes advantageof the energy savings with variable speed.The magnetic-coupled drive will take advantage ofthe affinity laws on variable torque loads such ascentrifugal fans and pumps and does so withoutaltering the voltage or interrupting power to themotor. Even with the motor running continuously,the kW required by the motor changes according tothe actual load; i.e., loading and unloading of themotor by varying the speed of the load via magneticcoupling between the motor end the load. What youalways pay for is the amount of kW used and evenwith the motor running continuously, the differencein the amount of kW that required of any 3 phasemotor from no load to full load is very significant.As example: 7-1/2 hp motor/3-phase/60hz (typicalblower application)Measured @ full load, max fan speed (typical) = 5.60kWMeasured @ min load, min fan speed = approximately0.40 to 0.50 kWGreater than 10 to 1 kW difference throughout entirespeed range, thanks to the affinity laws. Even greateradvantages from full load to no load are realized asthe motor horsepower size increases. As a result, thepower curves are very similar to vfd’s on variable

310 ENERGY MANAGEMENT HANDBOOKtorque loads and the kW savings are significant aswell. As a general rule, magnetic-coupled drives aremore efficient at the top end of the curve, and thevfd’s are more efficient further down the curve.Q. Does The Magnetic-Coupled Drive Cause AdditionalMotor Heating, Even At Very Low Speeds?The magnetic coupling is electrically isolated fromthe motor and in effect operates as an infinitelyvariable; frictionless clutch, allowing tie motors tooperate as originally designed at full speed continuously,end with pure uninterrupted AC power.Regardless of the drives’ operating speed, the motornever sees any additional heating contributed by thedrive. In fact, the drive itself via slip, and not themotor dissipate any additional heat. Since these newdrives are efficiently sized to handle the full ratedhorsepower of this types of loads, the minimumamount of drive heating is effectively dissipated bythe drives’ own integral fan. Since the motor runscontinuously end the drive is simply controllablycoupling and uncoupling the load, the effect of themotor loading is no difference in operation than ifyou had incorporated an infinite number of pulleysizes to provide variable speed to the load.Q. Power Quality Issues, Harmonics?Since the magnetic-coupled drive does not interruptthe power source to the motor, there are no currentharmonics produced nor is there any resultant voltagedistortion. There is never any need for filters,reactors, or full rated 3 phase isolation transformers.Q. How Far Can The Controller Be Located From TheMotor? Are There Any Limitations?The distance from the controller to the magneticdrive and motor has worked successfully at distancesup to 2000 feet. The only requirement is thatthe two wires that provide power to the drive coilbe sized large enough to allow for any voltage drop(say 14 gauge, typically). No filters or any other devicesare required. There is no concern about evercausing any damage to the motor or drive.Q. What About Nuisance Dropouts?By virtue or the inherently simple design, the driveis always active as long as the motor starter is energized.Transient over voltages, voltage sags, andharmonic distortion from other sources generally donot effect magnetic-coupled drives unless the durationis long enough to drop out the motor startercircuit.Q. Lightning?Since eddy-current drives are isolated from thepower source, they provide the highest level of immunityto the effects of lightning.Q. Retrofit Existing Motors?The shaft-mounted magnetic-coupled drive allowsfor true variable speed retrofit of any existing motor.There is never any need for inverter duty motors.All original motor wiring circuitry can alwaysremain undisturbed.Q. Bypass?A simple mechanical full speed lock-up feature isstandard on all drives. Full speed electrical bypasscan be accomplished with one diode.Q. Is There Any Motor Insulation Failure Or ElectricalPitting Of Motor Bearings?Since die magnetic-Coupled drive does not electricallyconnect to the motor wires, there is never anypossibility or causing any electrically induced harmto the motor windings or the motor bearings.Q. What About Power Factor?Some utility companies in certain locations maychange penalty if the facility’s total measured powerfactor is below an acceptable pre-determined level.In hundreds and hundreds of installations in manyvaried facilities such as colleges, schools, hospitals,government buildings, and shopping malls, powerfactor has never been an issue with the new shaftmountedvariable speed drives. However, in theevent low power factor requires attention, low costpower factor correction capacitors can be easily installedat each motor location as required.Q. Maintainability, In-House?This simple technology does not require highlyskilled personnel to maintain. The same compact

ELECTRIC ENERGY MANAGEMENT 311low cost controller is used on all drive sizes 1through 150 horsepower. Unlike high maintenancebrushes and slip rings, the rotary brushless plug-incoupling cartridge can be swapped out in a matterof seconds.Q. For Applications Other Than Variable Torque Loads,How Does The Magnetic-coupled Drive Perform?In constant torque applications, the drive must beproperly sized to allow for additional drive heatdissipation at the lower speed ranges, however thereare no added demands on the motor other than theconventional no load to full load conditions, thesame as in fixed speed operation considerations.These drives perform very well in many other applicators,such as conveyors, feeders, machine tools,punch-presses, etc., As variable voltage, constantcurrent, or closed loop methods.Q. What’s New On The Horizon?Continued development on the next generation ofself-powered variable speed drives is a priority.These new magnetic coupled drives have their ownsource or power, an integral generator, automaticallyactive when the motor is running. No externalpower source is required. Installation cost is reducedfurther because there is no need for the added costof an electrician. Simply mount the drive to themotor shaft, connect the belts, end then provide astandard 4/20-ma signal to the drive. Loop poweredat any horsepower. Imagine that.CONCLUSIONThe new generation of magnetic-Coupled drivetechnologies, combined with the ease of installationand the industry’s first mega-motor/drive warrantyprovides an attractively reliable and efficient alternativefor variable fan speed control. This rugged producthas been successfully used as upgrade replacements formost of the previously mentioned designs and technologiesin hundreds of belt driven installations. In criticalsituations, where maximum reliability and maximumrun time is paramount; this approach may well be thepreferred alternative.Www.Payback.ComE-Mail: info@Coyoteinc.ComTypical Power ConsumptionFAN SHEAVES SELECTIONVery Important:The fan sheaves should always be selected so thatwhen the fan is at its maximum rpm, the payback driveshould also be operating as closely as possible to its maximumattainable output speed, typically 1600-1700 rpm.*See technical data sheet for the specific paybackmodel’s sheaves sizes and output speed ranges @ givenhp load.By correctly sizing the fan sheaves, the systemwill be more efficient, the drive will run cooler, andthe drive’s bearing life will be optimized. (Selecting toosmall a driven fan pulley will waste energy and createunnecessary heat dissipation in the drive.)For Retrofit Applications:There are two easy methods for determining thenew fan pulley size to match with the selected modelpayback drive’s sheaves. In both cases, be sure to measurethe center to center distance between the existingmotor shaft and fan shaft. Observe the belt take-up adjustmenton the motor base and allow for a mid-range

312 ENERGY MANAGEMENT HANDBOOKBLOCK DIAGRAM—Variable FrequencyDrive (VFD)Typical required configurationfor reductionof harmonic distortionlevels induced by VariableFrequency Drivesin variable speed applications.Block Diagram—Variable Frequency DriveBLOCK DIAGRAM—Magnetic Coupled DriveTypical configuration for magnetic coupledvariable speed drive application. (True ZeroHarmonic Distortion)Block Diagram—Magnetic Coupled Drivetake-up measurement to start with for calculating thenew belt sizes.Always take an original rpm reading of the driven(fan) pulley and amp reading of the existing motor atfull load and continue to monitor the motor amps afterthe retrofit. In all correctly sized applications, operationthroughout the entire speed range should not exceed thefull load amps of the motor.Method 1: Pulley Ratio MethodStep 1: determine the ratio of the existing motor pulleyand fan pulley by dividing the driven (fan) pulleydiameter by the motor pulley diameter.(Driven (Fan) Pulley Diameter)/(Motor Pulley Diameter)= (Working Ratio)Step 2: multiply the listed sheaves diameter of the appropriatemodel payback drive by the derived workingratio to determine the ideal new driven (fan) sheavesdiameter.(Payback Listed Sheaves Dia.) × (Working Ratio) = (CalculatedFan Sheave Diameter.)Step 3: select the nearest size sheaves from a sheavesselection book. Using the new selected sheaves diameterin conjunction with the new drive listed sheaves diameterand the center to center measurement, size the beltsfrom a belt selection guide.Note: taking into account the difference between themotor rpm of 1750 and the drive’s maximum attainableoutput of 1600-1700 rpm, the actual new driven rpm

ELECTRIC ENERGY MANAGEMENT 313may be slightly less than the original by a factor of thisdifference rpm, however mechanical lockup, if requiredin emergency situations should return the driven pulleyclose to the original rpm.Method 2: Rpm Ratio MethodStep 1: measure the existing driven (fan) pulley rpmwith an accurate tachometer instrument.Step 2: divide the payback drive rated rpm (1600-1700)*by the measured (fan) rpm.*See applicable model data sheet for max rated output@ given horsepower.(Payback Max Rated Rpm)/(Existing Driven (Fan) PulleyRpm) = (Rpm Ratio)Step 3: multiply the listed sheaves diameter of the appropriatepayback drive by the calculated rpm ratio todetermine the ideal new driven (fan) sheaves diameter.in conjunction with the payback drive listed sheavesdiameter and the center to center measurement, size thebelts from a belt selection guide.Note: the new driven pulley maximum rpm will bevery closely matched to the original driven pulley rpm,however if mechanical lockup is required for emergencysituations, care should be taken that the additional increasein rpm does not cause the system to exceed itsmaximum capacity or the motor to exceed it’s ratedmotor amps.For new installations: use method 2 for determiningthe correct sheaves size when the correct model paybackdrive and desired maximum driven (fan) rpm isalready known.If you need assistance in correctly sizing your tan pulleyor have any other questions about your application,please call us at: 817-485-3336 or you may e-mail us at:info@coyoteinc.com.(Payback Listed Sheaves Dia.) ¥ (Rpm Ratio) = CalculatedDriven Fan Sheave Diameter.Step 4: select the nearest size sheaves from a sheavesselection book. Using the new selected sheaves diameterPumpDrives basic parts


CHAPTER 12ENERGY MANAGEMENT CONTROL SYSTEMSDALE A. GUSTAVSON, C.E.M.D.A. Gustavson CompanyOrange, California(714) 639-6100TOM LUNNEBERG, P.E.Manager, Energy EngineeringCTG Energetics, Inc.Irvine, CA 92618(949) 790-0010WILLIAM E. CRATTYALFRED R. WILLIAMS, P.E.Energy Conservation Mgmt. Co.Bethel, Conn.Competitive economic pressures on owners to reducebuilding operating expenses are challenging the traditionaldesign and control of heating, ventilating, airconditioning (HVAC) and lighting functions. Facilityowners and operators have strong financial incentivesto match more closely control, zoning and HVAC equipmentsizing to the use of building spaces and outsideenvironmental conditions. This must be done withoutsacrificing comfort and safety. Energy management systemsplay a key roll in meeting this challenge.12.1 ENERGY MANAGEMENT SYSTEMSEnergy management is the control of energy consumingdevices for the purpose of minimizing energydemand and consumption. Manually toggling on andoff devices based upon need is a rudimentary form ofenergy management. The advent of mechanical devicessuch as time clocks for automatic toggling and bimetallicstrip thermostats to control the output of heating andcooling devices along with pneumatic and electricaltransmission systems provided means for developingearly energy management systems in the form of automatictemperature controls. The advent of solid stateelectronic control devices and the increasing powerof the microprocessor based personal computer haveled to dramatic advances in energy management andwhat today is termed the energy management controlsystem (EMCS). The primary difference between earlyautomatic temperature control systems and EMCS isthe application of a broad base of variables throughprogrammable logic controllers to optimize the use ofenergy. While EMCS is used to control building environmentalconditions and industrial processes, this chapterwill deal only with EMCS applications for control ofbuilding HVAC and lighting functions.12.1.1 Direct Digital ControlHVAC building control system manufacturers havegreatly enhanced EMCS by incorporating direct digitalcontrol (DDC). DDC is defined as a digital computerthat measures particular variables, processes this datavia control algorithms and controls a terminal deviceto maintain a given setpoint or the on/off status of anoutput device. The term “digital” refers only to the factthat input/output information is processed digitally andnot that input or output devices are digital. Inputs andoutputs relative to a DDC EMCS can be either digitalor analog. Typically, most inputs are analog signalsconverted to digital signals by the computer while thegreater portion of outputs are likely to be digital (zeroor full voltage).DDC systems use software to program microprocessors,therefore providing tremendous flexibility forcontrolling and modifying sophisticated control applications.Changing control sequences by modifyingsoftware allows the user to improve performance ofcontrol systems throughout a building.DDC EMCSs can be programmed for customizedcontrol of HVAC and lighting systems and performfacility wide energy management routines such as electricalpeak demand limiting, ambient condition lightingcontrol, start/stop time optimization, sitewide chilledwater and hot water reset, time-of-day scheduling andoutdoor air free cooling control. An EMCS using DDCcan integrate automatic temperature control functionswith energy management functions to ensure thatHVAC systems operate in accord with one another forgreater energy savings.The most significant benefits DDC systems offerare: a) the ability to customize the scheduling ofequipment and their component devices to react toever changing conditions of use and weather, and b)315

316 ENERGY MANAGEMENT HANDBOOKadditional control modes (integral and derivative) thatresult in quicker, more accurate control when comparedto pneumatic systems. Pneumatic systems inherently offeronly proportional (how far is the input from setpoint)control in which the terminal control device linearly variesthe output as an input variable changes relative tosetpoint. Linear control results in offset or “hunting” bya terminal control device (valve) as it throttles to controlto setpoint. By adding the integral (how long has theinput been away from setpoint) mode to a proportionalcontroller, offset is minimized as the control point is automaticallyreset while the controller continually shiftsthe throttling range. By adding the derivative (how fastis the input approaching/moving away from setpoint)mode the controller can achieve setpoint much quickerand more accurately whenever load varies. Generally,proportional-integral (PI) control is sufficient for buildingHVAC applications.The primary objective of an EMCS using DDC is tooptimize the control and sequencing of mechanical systems.A DDC EMCS also allows centralized and remotemonitoring, supervision and programming maintenanceof the HVAC and lighting functions. Additionally, suchsystems can lead to improved indoor environmentalcomfort and air quality.Importantly, EMCS can also perform valuable nonenergyrelated tasks (NERTs). These often overlookedDDC applications, or “extra-standard” functions, can bebusiness specific or facility specific such as follows:A design/build EMCS specialist for shoppingmalls adds people counting, vehicle counting, andprecipitation logs to normal EMCS monitoring tohelp mall management lease space and track thesuccess of mall promotions.During a facility appraisal of a year-old drug storedistribution center, a West Coast EMCS consultant/contractor discovered an opportunity to controlthe facility’s conveyor system. The application notonly resulted in energy savings, but also streamlinedshipping.An imaginative plant engineer commissioned thewriting of custom front-end software for a brandname EMCS to perform noise level control in a factory.This ‘stretch’ of the EMCS now helps protectthe hearing of employees.DDC EMCSs are being used to control environmentsin mushroom farms and banana distributioncenters, to control pivot-irrigation systems, and evento control snow-making equipment at ski resorts. Thelimiting of DDC EMCS application strictly to HVAC andlighting control diminishes the financial benefit of thistechnology to the system owner.Sometimes a business-specific or facility-specificapplication will have a greater impact on EMCS justificationthan does the energy saving component. Forexample, a 1990 law change in California requires banksto maintain specified levels of lighting near their automaticteller machines (ATMs). The law was passed inresponse to muggings and killings at poorly lit ATMs.Some banks have responded by randomly installingadditional lighting fixtures. This shortsighted approachstill begs the question, what happens if the lights goout?One possible DDC solution would be to measureand log an ATM’s light level. When lumens drop belowa specified level, that facility’s EMCS could simultaneouslyactivate an emergency backup lighting system,notify an alarm monitoring company, and automaticallydial out to a central EMCS monitoring station to recordand verify that the notification function worked. Analarm light or tone also could be activated at the siteto alert the facility personnel to the problem the nextmorning. All of this by the same microprocessor thatcontrols the HVAC and lighting? Easily! Would it becost effective? That question might best be asked of thebank’s legal department.For more discussion on NERTs and other benefitsof EMCSs, see Section 12.2.12.1.2 HardwareAn EMCS can range from a very simple standaloneunitary microprocessor based controller withfirmware routines (control software logic that the usercannot modify except for setpoints) that provides controlof a terminal unit such as a heat pump to a verysophisticated large building DDC EMCS that interfaceswith fire and security systems. Although this chapterwill generally cite examples from applications inmedium to large facilities, there have been numerousdocumented successes with stand-alone controllers insmall buildings. Fast food and dinner houses, auto dealerships,retail stores, bowling centers, super markets,branch banks, small commercial offices, etc., have allbenefited from this technology. Typically, EMCS installationsat small buildings are ‘design/build’. That is, anEMCS manufacturer, distributor, installing contractor,end user, or some combination thereof select the EMCShardware and decides how it is to be applied.Much study remains to be done to determinethe extent by which firmware routines themselvescontribute to the economics in small building EMCSprojects. Empirical data suggests that while EMCSs in

ENERGY MANAGEMENT CONTROL SYSTEMS 317small facilities do improve the accuracy and responseof mechanical system controls, the energy managementroutines are responsible only partially for the savingswhich are achieved. In applications where unitary controllershave been treated simply as ‘devices’, by-in-largesavings have proven to be less dramatic than where theyhave been treated as ‘systems’ and used as ‘tools forsaving energy’. Better applied, installed, documented,supported, and maintained EMCS projects have yieldedbetter results than those in which ‘black boxes’ werehung on walls in broom closets and left alone to worktheir ‘magic’.What is it that would lead one restaurant chainto bypass or disconnect hundreds of small EMCSs andreturn to time clock and manual control while at anotherchain, using nearly identical systems and strategies, a40% return on investment, better comfort, and morerapid HVAC service response is realized? The oppositeresults suggest that the “ human dimension” is a factorwhich must be more clearly understood and weighed.This is true for small design/build projects as well asfor very large, sophisticated, ‘engineered’ projects. The“people” factor must not be ignored. Different buildingmanagers and occupants have different needs and occupancyhabits. Also, no matter how well a system isdesigned, if the building operators and occupants arenot properly taught how to use and maintain it, theEMCS will never live up to its full potential. For morediscussion on the importance of training and the “humandimension,” see Section 12.3.8.The selection of EMCS type and sophistication forany given application should balance management andcontrol desires with economics. An all encompassingDDC EMCS will provide the best overall control andmanagement capability, but it is also the most expensive.On the other hand, stand-alone controllers are the leastexpensive for individual control applications; they alsorestrict control strategy and management capability. Interms of energy savings, an all encompassing systemmay not provide significantly greater savings than usingstand-alone controllers for all HVAC functions at a givenbuilding, but it surely provides for better and easiermanagement of the mechanical systems. However, withappropriate telecommunications software a system ofunitary stand-alone controllers connected to modemscould prove to be a very effective EMCS for a chain ofwidely dispersed small facilities.There are two basic economic opportunities forapplication of an EMCS. The first is a retrofit of existingbuildings with functioning automatic temperaturecontrol systems. The second is in new construction. Itcould be very difficult to justify removing a functioningautomatic temperature control system in an existingbuilding to install an all encompassing DDC EMCSsolely for energy conservation purposes. For example,an air handling unit is typically managed by a controllerthat monitors signals for outside air temperature,mixed air temperature, discharge air temperature, andreturn air temperature, compares the signals to setpointconditions and in turn transmits a signal to a terminalcontrol device to position the outside air dampers, exhaustdampers, return air dampers, and heating valveor chilled water valve to maintain a predeterminedsetpoint. If the unit has a variable speed drive, humidifieror dehumidifier there are even more conditions tomonitor and control. To remove all of the existing controlsand install an all encompassing DDC EMCS couldbe very costly. An economical alternative is to eliminatethe traditional time clock function from the automatictemperature control system and interface a DDC EMCSin a “supervisory” mode over the existing controller byconnecting key electric/pneumatic relays (EPs) to DDCoutputs and installing strategically placed input sensors.A representative sampling of conditions such as a spacetemperature or discharge air temperature in conjunctionwith global inputs and well designed software in mostcases would provide adequate management to maintaincomfort and save a significant amount of energy.The first cost of a DDC EMCS installation has decreaseddramatically in the past several years, and DDCcurrently costs about the same or less than a pneumaticcontrol system in most new construction applications. Inaddition to this first cost advantage, some advantagesof DDC over pneumatic controls are: 1) more precisecontrol, 2) unlimited customization of control schemesfor energy management and comfort, 3) centralizes andintegrates control and monitoring of HVAC and lighting,4) easier to maintain, and 5) easier to expand and‘grow’ with building size and use.Current day DDC EMCS hardware configurationvaries from manufacturer to manufacturer but has acommon hierarchical configuration of microprocessorbased digital controllers as well as a front-end personalcomputer. This configuration can be categorized by threelevels: 1) terminal equipment level controllers, 2) systemlevel controllers, and 3) the operator interface level.The three levels are networked together via a communicationstrunk which allows information to be sharedbetween other terminal equipment controllers and alllevel controllers.The most important part of a DDC EMCS are thesystem controllers because they monitor and controlmost mechanical equipment. A major feature of systemlevel controllers is their ability to handle multiple control

318 ENERGY MANAGEMENT HANDBOOKloops and functions such as proportional- integral-derivative(PID) control, energy management routines,and alarms while the terminal equipment controllers aresingle control loop controllers with specific firmware.The operator interface level is generally a personalcomputer that serves as the primary means to monitorthe network for specific data and alarms, customize thecontrol software for downloading to specific system controllers,maintain time-of-day scheduling, and generatemanagement reports. However, it is not uncommon fora single personal computer to serve both as the systemcontroller and operator interface.The specific method of communications withinan EMCS is significant because of the amount of databeing processed simultaneously. While methods of datatransfer between DDC controllers and the operator interfacelevel vary from manufacturer to manufacturer,communications protocols can be simplified into twodistinct categories: 1) poll/response, and 2) peer-to-peeror token-ring-passing network.A peer-to-peer network does not have a communicationmaster or center point as does the poll/responsesystem because every trunk device, be it a terminal levelcontroller attached to a systems level controller or asystem level controller networked to other system levelcontrollers, at some point, has a time slot allowing it tooperate as the master in its peer grouping. However,the terminal level controller in a peer-to-peer networkusually is a poll/response device. A system using peerto-peercommunication can offer distinct advantagesover poll/response communication when redundancy ofcritical global data is accommodated. These advantagesare: 1) communication is not dependent on one device,2) direct communication between controllers does notrequire communication through the operator interfacelevel, and 3) global information can be communicatedto all controllers quickly and easily.The speed of system communication in buildingcontrol usually is not an issue with today’s DDCtechnology. However, the response time for reaction tocontrol parameters can be an issue with the applicationof a centralized poll/response system where controland monitoring point densities are very high or alarmresponse time is critical. Generally, peer-to-peer communicationdistributive systems that network system levelcontrollers and terminal equipment level controllersprovide the quickest response time. However, the everincreasing speed of the personal computer is allowingcentralized poll/response systems effectively to expandtheir control and monitoring horizon.Although quantum leaps in desktop computerprocessing horsepower lend some advantages to poll/response systems that use a personal computer for centralizedcontrol, panelized DDC EMCSs are still by farthe most popular approach for providing digital controlof building systems. It used to be the case that using ahost PC for building system automation allowed morecomplicated sequences of operation to be implemented,and greater amounts of trend data to be stored. Today,however, reduction in physical size as well as cost forcomputer memory chips has led to much greater capabilitiesfor stand-alone control units, and as a result thereis not much of a performance gap between centralizedpoll/response and panelized DDC systems.Be it new construction or a retrofit, when selectingan EMCS the actual needs and requirements of theparticular building or campus of buildings must beconsidered. While facility layout and construction typewill influence EMCS configuration, software is the mostcritical element in any EMCS application.12.1.3 SoftwareThe effectiveness of the software control logic iswhat provides the building operator with the benefitsof an energy management system. While color graphicsand other monitoring enhancements are nice featuresto have, the control logic in the system and terminalcontrollers is what improves building systems efficiencyand produces the energy savings.Networking DDC devices has provided buildingoperators with the ability to customize the traditionalenergy management strategies such as time-of-day/holiday scheduling, demand shedding, duty cycling,optimum start/stop and temperature control as wellas the ability to implement energy management techniquessuch as occupied/unoccupied scheduling ofdiscrete building areas, resulting in reduced airflowvolumes and unoccupied period setback strategies thatgreatly reduce operating cost. PID control provides formore accurate, precise and efficient control of buildingHVAC systems. However, software that providesfor adaptive or self tuning control not only enhancessavings but also addresses environmental quality.Adaptive control software monitors the performanceof a particular control loop and automatically adjustsPID parameters to improve performance. This featureimproves control loop response to more complicatedand dynamic processes.If the energy management industry has an Achillesheel it is the specifier who specifies an EMCS byreference to a particular manufacturer’s general hardwarespecifications rather than detailing the specificsof software sequencing logic and system configuration.Specification by manufacturer reference is not a

ENERGY MANAGEMENT CONTROL SYSTEMS 319complement to a hardware manufacturer rather it is anindication of EMCS ignorance on the part of the specifier.HVAC mechanical equipment operation sequencesare becoming increasingly complex. This is a result ofstricter criteria established by national codes (AmericanSociety of Heating, Refrigerating and Air ConditioningEngineers’ Minimum Outdoor Air Requirements Standard62 for example) and by building operators whorequire highly accurate control systems, energy usemonitoring and accounting by zone of use. To meet thechallenges posed by these stricter codes and buildingoperator demands the EMCS specifier must be thoroughlyknowledgeable about all aspects of EMCSs.As suggested in the earlier discussion of the humandimension phenomenon, it is essential that theEMCS specifier look beyond technology or, as the oldsaying goes, look beyond the trees to the forest. That is,a specifier must be knowledgeable about not only thebuilding construction and mechanical and lighting systemsto be controlled but also the businesses and peoplewhich occupy the building. And, of equal importance isthe need to assess the level of training and ongoing supportthat the EMCS owner/operator will need to realizethe anticipated benefit.12.1.4 Control StrategiesA DDC EMCS can serve six basic functions: 1)manage the demand or need for energy at any giventime, 2) manage the length of time that devices consumeenergy, 3) set alarms when devices fail or malfunction, 4)facilitate monitoring of HVAC system performance andthe functioning of other building systems, 5) assist thebuilding operator to administer equipment maintenance,and 6) provide the building/business owner/operatorwith non-energy related tasks (NERTs), i.e., extra-standardfunctions to make the EMCS more effective oradvantageous. There are financial benefits for each ofthese six functions, yet traditionally, only the first twoare quantified in an economic analysis. The industry stillhas work to do to develop new, accurate, reliable modelsfor predicting energy savings by an EMCS. But, to limitthe inquiry to just energy savings falls far short of whatis necessary for EMCS technology to realize anywhereclose to its truly staggering potential.It is the industry’s responsibility to develop modelsfor predicting the value of all EMCS functions. It is toooften left to sales and marketing people to explore and,even more importantly, assert the full range of benefitsof EMCS technology. The energy and engineeringcommunities tend to stop where science ends and speculationbegins. Speculation is a healthy exercise. Here aresome questions which might be worth asking:What is the annual dollar value of an efficientoffice building 95% leased as compared to a lessefficient building 75% leased?What is the annual dollar value of being able toprovide basic comfort troubleshooting by telephoneusing a personal computer and remotecommunication software?What is the annual dollar value of increased productivityin a building, which suffers little or nodiscomfort?What is the value of lowering absenteeism byusing EMCS to maintain indoor air quality monitoringand preventive action?What if one of the above benefits is the differencebetween justifying an EMCS project and not goingahead? The key to making the most of NERTs, orextra-standard EMCS functions is recognizing that: 1)quantified benefits have more impact than unquantified‘hopes’ when buying decisions are made, 2) it is theindustry’s duty to continually observe, track and createnew EMCS ‘by-products’, and 3) only imaginationlimits the range of extra-standard functions which canbe incorporated into the EMCS design. The extent andproficiency of performing these functions depends onthe software programming.The energy management industry has evolvedstandard time-proven software routines that can be usedas a starting point to develop effective programming.These routines are commonly referred to as controlstrategies. While these routines can be applied in virtuallyall EMCS installations they must be customized foreach building. A description of some of these routinesfollows.Daily Scheduling: This routine provides for individual,multiple start and stop schedules for each pieceof equipment, for each day of the week.Holiday Programming: This routine provides formultiple holiday schedules which can be configuredup to a year in advance. Typically, each holiday can beprogrammed for complete shutdown where all zones ofcontrol are maintained at setback levels, or for specialdays requiring the partial shutdown of the facility. Inaddition, each holiday generally can be designated as asingle date or a range of dates.Yearly Scheduling: Typically, any number of controlpoints can be assigned to special yearly scheduling routines.The system operator usually can enter schedulesfor yearly scheduled control points for any date during

320 ENERGY MANAGEMENT HANDBOOKthe year. Depending on the particular EMCS softwareoperating system, yearly scheduled dates may be erasedonce the dates have passed and the schedules have beenimplemented; or, the scheduled dates may repeat in thefollowing year as do scheduled holiday dates.Demand Limiting or Load Shedding: Demand limitingcan be based on a single electric meter or multiplemeters. Generally, loads are assigned to the appropriatemeters in a specified order in which equipment is to beshed. Usually, a control point that is allowed to be shedis given a status as a “round-robin” demand point (firstoff-first on), a “priority” demand point (first off-last on),or a “temperature” demand shed point (load closestto setpoint is shed first). Other parameters per controlpoint include Rated kW, Minimum Shed Time, MaximumShed Time and Minimum Time Between Shed.Minimum On-Minimum Off Times: Normally asystem turns equipment on and off based on temperatures,schedules, duty cycles, demand limits and otherenvironmental parameters. However, mechanical equipmentoften is specified to run for a minimum amountof time once started and/or remain off for a minimumamount of time once shut down. Therefore, this routinegives the operator the ability to enter these minimumtimes for each piece of equipment controlled.Duty Cycling: This routine provides the ability fora control point to be designated for either temperaturecompensated (cycle a piece of equipment on and off tomaintain a setpoint within a dead band) or straight timedependent (cycle a piece of equipment on and off fordistinct time intervals) duty cycling. Control parametersfor temperature compensated duty cycling include TotalCycle Lengths, Long and Short Off Cycles and Hi andLow Temperatures. There are separate sets of parametersfor both heating and cooling. Time dependent dutycycling uses Total Cycle Length and Off Cycle Lengths.Cycles for each load can be programmed in specificminute increments and based on a selectable offset fromthe top of the hour.Optimum Start/Stop: For both heating and coolingseasons DDC systems can provide customized optimumstart/stop routines which take into account outside temperatureand inside zone temperatures when preparingthe building climate for the occupant or shutting the facilitydown at the end of the day. During unoccupied hours(typically at night) the software tracks the rate of heat lossor gain and then utilizes this data to decide when equipmentwill be enabled in order to regain desired climaticconditions by the scheduled time of occupation. The samelogic is used in reverse for optimum stop.Night Setback: This routine allows building lowtemperature limits for nighttime, weekend and holidayhours, as well as parameters and limits for normal occupiedoperation to be user selectable. Night setback isusually programmed to evaluate outside air temperaturein the algorithm.Hot Water Reset: This routine varies the temperatureof the hot water in a loop such that the watertemperature is reduced as the heating requirement forthe building decreases. The reset temperature for nightsetback usually is less than the day reset temperature.Typically, reset is accomplished by controlling a threewayvalve in the hot water loop; however, dependingon boiler type, reset can be accomplished by “floating”the aquastat setpoint. For pneumatic control systems,the primary variable in the reset algorithm is outside airtemperature. For newer systems that have DDC controlson air handling units, it is possible to reset the heatinghot water temperature based on the “worst case”hot water control valve position. Under this strategy,the heating hot water temperature is reset downwardsuntil the air handling unit with the greatest need forheating has its hot water control valve fully open. ForVAV systems that feature DDC control at the zone level,it is also possible to reset the temperature of hot waterfeeding zonal reheat coils based on the “worst case”zone. Under this strategy, each DDC zone controller reportsthe position of its hot water reheat control valve,and the temperature it reset downwards until the zonewith the greatest reheat requirements has its valve fullyopen. This strategy can provide a significant reductionin wasteful reheating of conditioned air.Boiler Optimization: In a facility that has multipleboilers this routine schedules the boilers to maximizeplant efficiency by staging the units to give preferenceto the most efficient boiler, by controlling the burnerfiring mode when desirable, and by minimizing partialloading.Chilled Water Reset: This routine varies the temperatureof the chilled water in a loop such that the watertemperature is increased as the cooling requirement forthe building decreases. This is typically done by controllinga three-way valve in the chilled water loop. Chilledwater reset also can be accomplished by interfacing theEMCS with the chiller controls to reduce the maximumavailable cooling capacity such as during demand limiting.For pneumatic control systems, the primary variablein the reset algorithm is outside air temperature. For newersystems that have DDC controls on air handling units,it is possible to reset the chilled water temperature basedon the “worst case” chilled water control valve position.Under this strategy, the chilled water temperature is resetupwards until the air handling unit with the greatest needfor cooling has its chilled water control valve fully open.

ENERGY MANAGEMENT CONTROL SYSTEMS 321In this way, the chilled water temperature will be reset toas high as possible—which improves chiller capacity andefficiency—without compromising comfort. It must benoted that this strategy must be carefully implementedwith variable flow chilled water systems, as it is possiblefor improvements in chiller efficiency to be eclipsed byincreased pumping energy.Chiller Optimization: In a facility that has multiplechillers this routine schedules the chillers to maximizeplant efficiency by staging the units to give preference tothe most efficient chiller. Unlike the boiler optimizationroutine, however, it can be more efficient to operate achiller plant with units sharing the load rather than fullyloading a unit.Chiller Demand Limiting: This routine limits chillerdemand by interfacing the EMCS with the chiller controlsto reduce the maximum available cooling capacity inseveral fixed steps. The primary variable in the demandlimiting algorithm is kW demand.Free Cooling: Free cooling is the use of outside air toaugment air conditioning or to ventilate a building whenthe enthalpy (total heat content) of the outside air is lessthan the enthalpy of the internal air and there is a desireto cool the building environment. In arid climates, aneconomizer cycle can work well by measuring dry-bulbtemperature only. In climates where humidity is of concern,enthalpy-based controls are preferable as they canprovide greater comfort and increased energy savings.An economizer can be categorized as integrated (meaningit can operate in conjunction with mechanical cooling)or non-integrated (meaning that the outside air damper isfixed at its minimum position when mechanical cooling isrequired).Recirculation: This routine provides for rapidwarm-up during heating and rapid cool-down duringcooling by keeping outside air dampers fully closed andreturn air dampers fully opened during system start-up.Hot Deck/Cold Deck Temperature Reset: This routineselects the zones or areas with the greater heatingand cooling requirements and establishes the minimumhot and cold deck temperature differential which willmeet the requirements in order to maximize system efficiency.(NOTE: Multi-zone system)Motor Speed Control: This routine will vary thespeed of fan and pump motors to reduce air and watervelocity as loads decrease. Speed control can be accomplishedeither by controlling a two speed motor with adigital output or a variable speed drive with an analogoutput.Manual Override: This routine provides separatemanual override schedules to allow for direct control ofequipment for specific periods of time.Duty Logs: DDC systems can track and displayvarious types of information such as last time on,last time off, daily equipment runtimes, temperatures(or other analog inputs) kWh and even remote panelhardware performance. DDC systems can also accumulateand display monthly scheduled and unscheduledruntimes for each piece of equipment controlled ormonitored.Temperature and hardware performance informationcan be monitored and displayed. Change of statecan be logged for all loads and can be reviewed on adaily basis for all loads. In addition, a system can produceline graphics or historical trends of input data andhardware systems information.Energy consumption logs can be kept separatelyfor every pulse meter point. For electricity these energylogs can record data (by the day) such as daily kilowatthour consumption, kilowatt demand, time the peak occurred,selected demand limit, time that any load wasshed, minimum, maximum and average outside temperaturesand degree day information.Alarm Monitoring and Reporting: DDC systemsregister and display alarms for conditions such as 1)manual override of machinery at remote locations, 2)equipment failures, 3) high temperatures, 4) low temperatures,invalid temperatures (sensor is being tamperedwith), and 5) communication problems.12.2 JUSTIFICATION OF EMCSsTraditionally, a commercial building is constructedby one party (a developer) owned by a second party(investors) and operated by a third party (a facilitiesmanagement firm). Typically, the building will haveone electric meter and a central heating plant. Therefore,total energy cost must be included in the rentalstructure. Before the ‘first energy crisis’ and energy wascheap, lessees typically paid a flat rental charge based onsquare footage rented and nobody paid much attentionto operating hours of the HVAC and lighting systems.When energy prices started their rapid upwardspiral building owners and operators implemented measuresto pass the cost increase directly on to the lessee.However, now because of the volume of space availableand other competitive pressures building owners andoperators are being forced to find means of reducingthe energy cost component of the base rent in order toreduce operating costs and entice customers with lowerbase rentals in the lease market.Depending on the type of HVAC and lighting controlsystems installed in a building, energy consumption

322 ENERGY MANAGEMENT HANDBOOKand operating costs can be dramatically different. Therefore,it should go without saying that since operatingcosts are a key ingredient when establishing competitiverentals, the type of control system installed can have asignificant impact on the owner’s income. And, to followthis line of reasoning to its conclusion, since rentalincome is a key ingredient in the appraisal of a commercialfacility, the type of control systems installed canhave a significant impact on the value placed on thebuilding should the owner want to sell.A well-designed and properly operated EMCS is aninvestment which offers the building owner/manager amultitude of benefits, which can oftentimes be bundledand quantified. These benefits include, yet go beyond,impressive payback and return-on-investment figures.In addition to the immediate financial benefits, there arelong-range benefits to managing energy and demand aswell. Such benefits can be wide-ranging, positively affectingsociety in general. Consider the following EMCSfunctions, along with the related benefits:• Manage energy consumption and demand—Loweroperating expenses, higher profits and increasedcompetitiveness, keep energy prices affordable inthe long term. Less smog, reduced acid rain andless global warming.• Optimize operating efficiencies of energy consumingequipment—Same benefits as above plusextended equipment life which further increasesprofits and competitiveness.• Improve comfort—Increases occupant productivityand concentration. Helps occupants feel morealert, rested, make fewer mistakes and performbetter. Leads to a more vital U.S. economy withbetter educated and less stressed populace.• Improve indoor air quality—Saves lives, increasesproductivity, prevents lawsuits, lowers insurancecosts and reduces absenteeism. Contributes to bettereconomical and educational systems.• Activate alarms when equipment malfunctions—Providesfor less costly disruptions ofproductivity due to faster response, increased profitability,extended equipment life and increasedcompetitiveness.• Assist on- or off-site operator administer service/maintenance—Results in better records, historiesfor more efficient, accurate work, less downtime,more productivity and lower cost for service.Keeps polluting vehicles off the street. Leads tohigher profits and increased competitiveness.• Monitor/log building equipment performanceand energy use—Resulting, accumulated dataleads to improved performance of all EMCS functions,thereby increasing magnitude of all benefitsitemized above. Also confirms cost avoidance benefitof project.• Perform NERTs (non-energy related tasks)—Forevery NERT there is a corresponding NEB ( nonenergybenefit). In fact, the value of NERTs canexceed the value of all the other benefits combined.(Refer back to Section 12.1.1 for specific examplesof NERT applications.) Some NEBs can be easilyquantified and others cannot, but they should bearticulated in your investment justification.While some of these benefits may seem far-reaching,different building owners and operators have differentneeds and requirements. Energy management professionalsshould always be on the alert for specific ways anEMCS can benefit a facility, as each case is unique.12.2.1 An EMCS OpportunityFor purposes of discussion, consider a ten story,multiple tenant, commercial office building configuredand occupied as follows as an example of an opportunityfor an EMCS application (Figure 12.1).a. One boiler room in the basement with 2 boilersand 2 redundant sets of circulating pumps (4 inall) to supply hot water to perimeter radiation andheating coils.b. One domestic hot water system for the entirebuilding.c. Four mechanical rooms—a north and a southmechanical room on the 5th floor each with 2 airhandling units (AHUs) supplying the first 5 floorsand a north and a south mechanical room on the10th floor each with 2 AHUs supplying the top 5floors. Each AHU has a preheat coil on the inletside of the supply fan and a chilled water coiland a reheat coil on the discharge side to maintaindesired hot deck and cold deck temperatures fora double duct distribution system. The north mechanicalroom on the 10th floor has a chiller and

ENERGY MANAGEMENT CONTROL SYSTEMS 323Figure 12.1 Ten-story office building.a redundant set of circulators to service the northside AHUs while the south mechanical room onthe 10th floor has a chiller and a redundant set ofcirculators to service the south side AHUs.d. One roof top mounted cooling tower serving bothchillers.e. The building is zoned by AHU (8 zones) and eachroom has a mixing box that is controlled by itsown thermostat to allow heating and cooling asrequired.f. Toilet exhaust is provided by 2 roof mounted exhaustfans that are switched in one of the 10th floormechanical rooms.g. Corridor and lobby lighting is switched from acentral location while space lighting is operated bya wall switch in each space.h. The building is normally occupied between 8:00am and 5:00 p.m. Monday thru Friday and sporadicallyoccupied a various times after normal hoursuntil midnight on work days and occasionally onweekends.

324 ENERGY MANAGEMENT HANDBOOKAn opportunity for use of DDC EMCS in eitherretrofits or new construction can be found when consideringbuilding use. One of the keys to effectiveutilization of an EMCS is zoning. Through flexibleprogram scheduling to control the energy consumingsystems by zone the building operator can realize substantialenergy savings.Most likely an existing building configured andoccupied as the above example building will havemanual lighting control and a mechanical time clockbased pneumatic system controlling HVAC functions asfollows:a. Each mechanical room has a time clock activatedcontrol panel for occupied/unoccupied switchover of day/night air pressures for space thermostatsand start/stop of the mechanical equipment.b. Each AHU has a mixed air temperature controller,a cold deck temperature controller and a hotdeck temperature controller to maintain supply airtemperatures at preset levels.c. Outside air dampers are calibrated to maintaindesired minimum air intake.d. Each zone has a centrally located night thermostatfor its respective AHU.e. In the boiler room there is a summer/winterchange over of the heating and cooling functions.f. Corridor and lobby lighting is controlled by a mechanicaltime clock in the electrical control room.To accommodate after normal hours occupancythe temperature control system provides for occupiedconditions from 7:00 am to midnight on weekdays. Onweekends the system provides for occupied conditionsfrom 8:00 am to 5:00 pm. Toilet exhaust fans run continuously.The corridor and lobby lighting controller isset to maintain lights on from 6:00 am to 1:00 am onweekdays and from 7:00 am to 6:00 pm on weekends.Cleaning crews and tenants are responsible for controlof lighting in the tenant spaces and toilets.In multiple tenant, high rise commercial officebuildings with HVAC and lighting systems configuredand operated in a manner similar to the example facilitythe DDC EMCS provides an opportunity to save a significantamount of energy and recover costs associatedwith after normal hours tenant use.12.2.2 The EMCS RetrofitWithout the benefit of a DDC EMCS the operatorof the example building must incur after-normal hoursoperating costs and allocate them over the entire leasingcost. This unfairly spreads the costs generated by a fewtenants over the lease costs of all tenants. The overalleffect are higher leasing charges for all.The most cost effective way to retrofit a DDCEMCS in a building controlled by a reasonably maintainedpneumatic control system is to interface it withthe existing pneumatics in a supervisory mode by eliminatingnight thermostats, replacing time clock contactswith DDC outputs, direct hard wiring for start/stopcontrol of the equipment not under the influence ofthe pneumatic system (i.e. toilet exhaust fans, domesticservice hot water circulators, and corridor, lobby andexterior lighting); direct hard wiring boilers, chillersand hot water and chilled water circulators for start/stop control, and direct hard wiring to enable/disablethe equipment under control of the pneumatic system(i.e. the AHUs and dampers). Enabled equipment isallowed to operate under the control of the pneumaticsystem. Disabled equipment is shut down (Figure12.2).With a supervisory DDC system the building operatorcan effect a dramatic reduction in energy usageby incorporating a simple timed on/off control strategyfor all HVAC and lighting zones that segregates theoperation of the equipment into discrete tenant zones.The control strategy would shut down the tenant zoneswhen outside of normal lease times. If a tenant wishedto utilize a space after hours, he/she could initial anoverride (using a simple automated telephone interfacefrom his/her office) to signal the system to preservehis/her working environment. Additionally, on weekendsand holidays the entire building can be kept inthe unoccupied mode unless a tenant requires extendedoccupancy of a particular zone. That tenant could initiatean override from any outside telephone to ensureoccupied conditions for his/her space upon arrival atthe building. Whenever an override is initiated, anoperator interface level personal computer could trackthe energy consumption and/or runtimes of that zone’sHVAC equipment for the purpose of billing the tenantan after hours usage charge.Experience at a multiple tenant, high rise commercialoffice building on Long Island, New York, hasshown that this “tenant strategy” not only results in adramatic reduction in energy usage but also allows thebuilding operator to reduce the base rental and recoverafter normal hours operating cost directly from thosetenants responsible for the use.By installing a DDC system as a supervisor of andexisting pneumatic control system the building operator

ENERGY MANAGEMENT CONTROL SYSTEMS 325AI-1AI-2AI-3AI-4AI-5AI-6AI-7AI-8AI-9BI-1BI-2BI-3BI-4Boiler #1 water supply temperatureBoiler #2 water supply temperatureCommon boiler water supply temperatureNorth loop supply temperatureSouth loop supply temperatureNorth loop return temperatureSouth loop return temperatureOutside air temperature sensorAmbient light sensorCirc Pump #1 flowCirc Pump #2 flowCirc Pump #3 flowCirc Pump #4 flowSLCBO-1BO-2BO-3BO-4BO-5BO-6AO-1AO-2BO-7BO-8BO-9BO-10Boiler #1 S/SBoiler #2 S/SCirc Pump #1 S/SCirc Pump #2 S/SCirc Pump #3 S/SCirc Pump #4 S/SReset control northReset control southSummer winter changeoverExterior lightingDomestic hot waterAlarmTELC#1TELC#2TELC#3TELC#4Typical forTELC 2, 3, 4AI 1 Hot deck sensor AHU-1AI 2 Cold deck sensor AHU-1AI 3 Mixed air sensor AHU-1AI 4 Return air sensor AHU-1AI 5 Chiller outlet #1AI 6 SpareBI 7 Chilled water Pump #1 flowBI 8 Chilled water Pump #2 flowBI 9 Cooling water Pump #1 flowBI 10 Cooling water Pump #2 flowBO 1 AHU #1 S/SBO 2 AHU #2 S/SBO 3 Chiler S/SBO 4 Day/night switchover AHU-1 zoneBO 5 Lighting controlBO 6 Toilet exhaustBO 7 AHU damper control (both units)BO 8 Day/night switchover AHU-2 zoneAI-11AI-12AI-13AI-14AI-15AI-16AI-17AI-18AI-19AI-20AI-21AI-22Rm sensor AHU-1Rm sensor AHU-1Rm sensor AHU-2Rm sensor AHU-2Hot deck sensor AHU-2Cold deck sensor AHU-2Mixed air sensor AHU-3Return air sensor AHU-2Rm sensor AHU-1Rm sensor AHU-1Rm sensor AHU-2Rm sensor AHU-2SLC -TELC -AI -BI -AO -BO -S/S -System level controllerTerminal equipment level controllerAnalog inputBinary inputAnalog outputBinary outputStop/startFigure 12.2 Office building supervisory EMCS configuration,

326 ENERGY MANAGEMENT HANDBOOKcan implement traditional control strategies (see section12.1.4) to improve building operating efficiency andachieve additional energy cost reductions. In additionto optimizing heating and cooling functions, the toiletexhaust fans, domestic service hot water and corridorand lobby lighting can be interlocked with zone occupied/unoccupiedstatus to shut down these systemswhen not needed. Pay backs for this type of EMCSretrofit are typically very attractive.12.2.3 New Construction EMCSIf the exampled building were under constructionas a new building, the developer would be remiss notto consider a DDC EMCS as part of the original design.In addition to all of the benefits described above, the allencompassing system would provide added benefit inthe form of PI control of heating and cooling valves anddampers. The design engineer also would be remiss notto consider improvements in the mechanical design thatalong with DDC control could significantly improve theenergy efficiency of the building.By combining more aggressive zoning with DDCEMCS in new construction the building designer has theopportunity to better match control of the HVAC withbuilding usage and optimize equipment sizing to betterfit loads thus improving equipment operating efficiency.However, the bulk of the EMCS opportunity lies in thethousands of existing buildings which are controlledwith pneumatic and electrical/mechanical automatictemperature control systems. Surprisingly enough thereare also buildings still being controlled by hand!12.3 SYSTEMS INTEGRATIONThe performance of an EMCS is directly proportionalto the quality of the designer’s systemsintegration effort. Systems integration can be definedas those activities required to accomplish the precisemonitoring and control desired when incorporating anEMCS into a building’s HVAC and lighting systems.This definition implies that systems integration includesnot only selecting the appropriate EMCS and detailingmonitoring points and control interfaces, but also,customizing the software algorithms to meet the user’sneeds and commissioning the system.12.3.1 Facility AppraisalAn appraisal of the job site is the first step ofEMCS systems integration. The appraisal is generallydone by interviewing the building users to determinebuilding usage patterns (be careful to differentiate betweennormally scheduled usage and after hours usage),to identify any comfort complaints (i.e. are there ‘hot’or ‘cold’ spots in the building, is ventilation adequate,etc.), to assess the attitude that the building maintenancestaff has towards an EMCS, and to determine exactlywhat the building manager expects the system to accomplish.During the facility appraisal, the following peopleshould be interviewed:Tenants (by company, department, job function)Prospective tenantsDelivery servicesSecurity personnelService vendors (including janitorial, HVAC, lighting,electrical, life safety)Building engineers/facility managersCorporate energy manger, if anyBuilding owner/CEO/CFOProperty manager and staffLeasing agents.Perhaps the most overlooked above are leasingagents. These professionals must understand the benefitsof energy efficiency, or they will not be able to sellit to perspective tenants.12.3.2 Equipment ListNext, the designer should develop a complete andaccurate equipment list. In a retrofit situation the listshould include all of the equipment and devices to becontrolled, the present method of control for these items,the operating condition of the items and their controls,and the area each item services. Do not forget to includeexterior and interior lighting on the list.12.3.3 Input/Output Point DefinitionThe third step is to define the input and outputpoints required to achieve the monitoring and control desired,in the most cost-effective or efficient manner. Thisrequires an in-depth appraisal of such items as the equipmentlist, layout of building spaces, method of buildingconstruction, HVAC physical plant makeup and layout,methods of heating and cooling, types of secondaryHVAC distribution systems and controls, and the conditionof the mechanical equipment and existing controls.The definition of a control point should include alisting of all inputs that will influence the functioningof that point. The installation specifications should detailthis information and indicate the nature of the influenceas follows:

ENERGY MANAGEMENT CONTROL SYSTEMS 327Output Point Analog Input Point NotesAHU #1 Outside Air TemperatureRoom #1027 Temperature- - - - - - - - - - - - - - - - - -Room #935 TemperatureRoom #837 Temperature Room of greatestRoom #715 Temperature DemandRoom #625 Temperature- - - - - - - - - - - - - - - - - - -12.3.4 Systems ConfigurationOnce the inputs and outputs have been defined,the designer needs to define any constraints that mightbe placed on the EMCS hardware. Building size, buildingdesign, building location, building use, design of thephysical plant, type and layout of the secondary distributionsystems, design of the electrical system, and typeand condition of existing controls in retrofits as well asthe nature of the personnel operating the building aresome of the items to be considered when configuring anEMCS.Cost of installation can vary significantly dependingon the type of system architecture (centralizedversus distributive), the method of transmittinginformation between system controllers and the input/output devices (direct hardwire versus multiplexingversus powerline carrier), the types of system controllersand input/output devices selected (two wayversus one way communicators) and the amount ofprogramming required (customized algorithms versusfirmware).Knowing who is going to maintain the softwareand perform the necessary administrative functions(i.e. inputting schedule changes, monitoring alarms,and producing reports) not only affects how manybut also where the operator interface level personalcomputer is to be located, and the type of monitoringand alarming that will be needed. This will also determinetraining requirements and the extent of supportservices needed.12.3.5 Specifics of Software LogicThe most critical phase of systems integrationis the software specification. The designer must takecare to be as explicit as he/she can when detailingthe nature of the algorithms to be installed. This canbe done by structuring the software specifications ina manner that clearly defines the algorithms and theirinterrelationships. The designer who simply includes amanufacturer’s general hardware specifications ratherthan detailing the specifics of software sequencing logicmost assuredly dooms the system to failure.APPENDIX A “EMCS Software Specifications” isa sample of what can be described as a minimally acceptablesoftware specification. Ideally, in addition toa descriptive specification, the designer would leavelittle doubt as to exactly what he/she has in mind byproviding a listing, in easy to understand English format,of all system application programs associated witheach piece of equipment. Following is an example ofwhat a listing might look like for control of boilers.HOT WATER BOILER CONTROLWhere there are multiple boilers in a set the software shall change the lead-lag status of boilers 1 and 2 in the set on abiweekly basis (Note the call to the lead-lag algorithm). This specific algorithm calls the general CENTHEAT algorithmwhich assumes that the heating circulation pump is stage one of the central heating plant.For the following control points CONTRACTOR shall apply this HOT WATER BOILER CONTROL algorithm:1. Boiler 12. Boiler 2VARIABLE DEFINITION:User defined variables (operating parameters):HEATSET:SETBACKTEMP:MAXOATHEAT:OCCLOWOAT:OCCHIGHOAT:The heating target temperature (72°F).The minimum set back temperature allowed (55°F).The maximum outside air temperature for heating to be utilized (65°F).The low outside air temperature parameter to be used for hot water reset during occupied time of day (0°F).The high outside air temperature parameter to be used for hot water reset during occupied time of day (60°F).

328 ENERGY MANAGEMENT HANDBOOKOCCLOWSWT:OCCHIGHSWT:UNOCCLOWOAT:UNOCCHIGHOAT:UNOCCLOWSWT:UNOCCHIGHSWT:The low hot water supply temperature parameter to be used for hot water reset during occupied time of day (120°F).The high hot water supply temperature parameter to be used for hot water reset during occupied time of day (190°F).The low outside air temperature parameter to be used for hot water reset during unoccupied time of day (0°F).The high outside air temperature parameter to be used for hot water reset during unoccupied time of day (40°F).The low hot water supply temperature parameter to be used for hot water reset during unoccupied time of day (120°F).The high hot water supply temperature parameter to be used for hot water reset during unoccupied time of day (160°F).Measured variables:ACTUALTEMP:OATEMP:The temperature being controlled by the I/0 point in question. In this case of a central boiler, the ACTUALTEMP shall representthe coolest of all of the zone temperatures. Refer to the Analog Temperature Assignment List to relate the temperature inputsto the control outputs.Outside air temperature.Calculated variables:MANOVRFLAG:MINFLAG:TOD:MODETEST:TURNON:Holds the cue for a manual override of normal program logic, 0 = manually override off, 1 manually override on, 2 = no manualoverride in effect. This is calculated by the MANOVR sub program.Holds the cue for whether a point is currently being controlled by its minimum on or minimum off time, 0 = it is not currentlyaffected by minimum on/off, 1 = it is currently affected by minimum on/off times.Holds a value of 1 if the point is within its time of day schedule, 0 if the point is not within its time of day schedule.Holds the cue for the System Mode, 0 = heating season, 1 = non-heating season. Its contents are determined by the MODETESTalgorithm that is called from this algorithm.Holds a value of 1 if the point should be turned on, 0 if the point should be turned off.PROGRAM CODE BLOCK‘ identify point characteristics500 COOLFLAG = 0 ‘ not a cooling point.HEATFLAG = 1 ‘ heating point.TODFLAG = 1 ‘ time of day point.OPTSTFLAG = 1 ‘ optimum start point.TCFLAG = 1 ‘ temperature control point.OALFLAG = 1 ‘ low outside air temperature point.SETBACKFLAG = 1 ‘ night setback point.FREEZEFLAG = 1 ‘ freeze protection point.1000 TURNON = 0 ‘ assume that it can be turned off.CALL LEAD-LAG ‘ see if it is the lead or lag boiler.CALL MANOVR ‘ manual override.IF MANOVRFLAG = 1 THEN TURNON = 1 : GOTO 3000CALL MINTEST ‘ minimum on/off test.IF MINFLAG = 1 THEN GOTO 3000 ‘ TURNON is set in MINTEST ‘ when MINFLAG is = 1CALL FREZPROT ‘ freeze protection override.IF TURNON = 1 THEN GOTO 3000CALL MODETEST ‘ check to see if in heating or non-heating.IF SYSMODE = 1 THEN TURNON = 0 : GOTO 3000‘ outside air heat lockout.IF OATEMP > MAXOATHEAT AND OALFLAG = 1 THEN TURNON = 0 : GOTO 3000CALL SNOWDAY — IF SNOWDAYFLAG = 1 THEN TOD = 0 : GOTO 2010CALL OPENHOUSE — IF OPHOUSEFLAG = 1 THEN TOD = 1 : GOTO 2000CALL ACTIVITY — IF ACTIVITYFLAG = 1 THEN TOD = 1 : GOTO 2000CALL TOD‘ see if its scheduled now.‘ load appropriate hot water parameters, call outside air‘ reset program to calculate target water temps then call‘ the appropriate algorithm to set TURNON variable.2000 IF TOD = 1 THENLOWOAT = OCCLOWOAT : HIGHSWT = OCCHIGHSWT

ENERGY MANAGEMENT CONTROL SYSTEMS 329HIGHOAT = OCCHIGHOAT : LOWSWT = OCCLOWSWTCALL OARESET‘ if there is no call for heat, leave boiler off.IF ACTUALTEMP > HEATSET THEN TURNON = 0 : GOTO 3000CALL CENTHEATEND IF2010 IF TOD = O THENLOWOAT = UNOCCLOWOAT : HIGHSWT = UNOCCHIGHSWTHIGHOAT = UNOCCHIGHOAT : LOWSWT = UNOCCLOWSWTCALL OARESET‘ if there is no call for heat, leave boiler offIF ACTUALTEMP > SETBACKTEMP THEN TURNON = 0 : GOTO 3000CALL CENTHEATEND IF3000 END BLOCK12.3.6 Control Point Tie-InsTo ensure that the EMCS is interfaced to a device tobe controlled exactly as desired, the integrator should includesketches detailing the specifics of the interface at thecontrol circuit of that device as exampled in Figure 12.3.12.3.7 CommissioningThe final step of systems integration is commissioning.Commissioning should include both the hardwareand software. All of the hardware should be physicallyinspected for correct installation and the wiring testedfor continuity. Inputs should be checked for accuracyagainst a standard. The operation of all outputs shouldbe demonstrated from the operator interface personalcomputer or system controller if no personal computeris installed. This inspection should not be taken lightly.The system will not function as designed if any hardwarecomponent is incorrectly installed or does notfunction as desired.The software should be checked before the systemis put on the line. This requires a careful review of allalgorithms, setpoints and schedules for compliance withthe detailed specifications.APPENDIX B “EMCS Installation Requirements”provides the basic parameters to formalize specificationsto bid the installation of a DDC EMCS. Commissioningprocedures are detailed in sections B15, B16 and B17.HarmonicsModern electronic ballasts used in fluorescent lightingsystems, as well as other digital switching devices,can generate harmonics in the electrical distributionsystem. There have been reported cases of such harmonicscausing malfunctions in older control systems thatcommunicate using power line carrier control (PLC)technology (such systems use the electrical distributionsystem as a means for inter-device communication bysuperimposing their digital “messages” over the 60 Hzpower sine wave). Modern PLC systems that use spreadspectrum technology are much more tolerant of harmonics,and encounter very few problems.Communications TechnologyThough most digital control systems rely on hardwiredconnections between devices, there are alternativetechnologies that can be used to simply installation (particularlyin retrofit applications), provide a communicationlink with remote locations, and increase data transmissionrates. These include:• Power Line Carrier (PLC). PLC systems use the existingelectrical wiring system as a means of achievinginter-device communication. This is accomplishedwhen a device generates a high frequencycommunication signal that is superimposed overthe standard sixty hertz power sine wave. Otherdevices that are connected to the same electrical systemcan analyze the spectral content of the incomingpower, and extract the superimposed “message”from the sixty hertz carrier wave. PLC systems canprovide reduced installation cost because of thereducing wiring requirements, but may be less reliableand slower than hard-wired systems.• Wireless Technology. For locations that are locateda significant distance away from a building, there isthe ability to send and receive control messages usingwireless technology. Some systems are based on

Fig. 12.3.5aHVAC SCHEMATIC KEYM DAMPER MOTORT1 MIXED AIR CONTROLLERT2 OUTSIDE AIR TEMPERATURE LIMIT CONTROLLERT3 HOT DECK TEMPERATURE CONTROLLERT4 COLD DECK TEMPERATURE CONTROLLERT5 INDIVIDUAL ROOM THERMOSTATT6 NIGHT THERMOSTAT [REMOVE]FS FREEZESTATEP-l MAIN AIR CONTROLPE-l PNEUMATIC DAY/NIGHT SWITCHOVER [REMOVE]SR SWITCHING RELAY FOR MINIMUM AIRPRV MINIMUM AIR PRESSURE REDUCING VALVES1 EMCS HOT DECK TEMPERATURE SENSORS2 EMCS COLD DECK TEMPERATURE SENSORS3 EMCS MIXED AIR TEMPERATURE SENSORS4 EMCS RETURN AIR TEMPERATURE SENSOR330 ENERGY MANAGEMENT HANDBOOK

ENERGY MANAGEMENT CONTROL SYSTEMS 331simple radio signals, which are relatively low cost,but slow. Other systems are based on cellular telephonetechnology, which can provide greater rangeand faster data transmission speed, but which maybe more expensive to own and operate.• Fiber Optic Communication. Due to the desire tohave ultra fast data transmission capabilities forcomputer network applications, many buildings andcampuses are now installing fiber optic networks.Such networks—which have sufficient bandwidthto provide telephone, computer network, and EMCScommunication simultaneously—provide the fastestdata transmission rates, but are also expensive to install.Many military bases and college campuses haverecently installed fiber optic “backbones” to allowinter-building computer network communication.Such scenarios present an excellent opportunity toreplace archaic dial-up modem EMCS systems withmodern systems that can provide more elaboratecontrol sequences and better accuracy.Industry Standard Communication ProtocolsIn an effort to achieve true interoperability for multiplepieces of equipment from different manufacturers,there has been significant effort focused on developingstandard communication protocols that all manufacturerswill adhere to. The advantage of so-called “open” protocolsis that, for example, a chiller control panel can sharedata with a variable speed drive controlling a chilled waterpump in order to reduce energy use, with near “plugand play” simplicity. The current market is built aroundproprietary control systems, making communication betweendevices from different manufacturers difficult toaccomplish in some cases. Open protocols are intendedto shift the industry away from proprietary systems thatlimit control system sophistication and user-friendliness.Though two standardized protocols are now widelyknown—BacNET (developed by ASHRAE) and Lonworks(based on the Neuron chip, developed by EchelonCorporation)— and a number of others are being developedand implemented as of this writing, their implementationis still not yet at the scale it needs to be providetrue interoperability for a wide range of system types.For the time being, inter-device communicationis typically accomplished using communication “gateways.”A gateway serves as a translator between thecommunication languages of two different pieces ofequipment, allowing them to share data and operate synergistically.A gateway can either be a stand-alone piece ofhardware, or may be a built-in feature of some DDC panels.Systems integration consultants are often involvedon sophisticated control system projects to ensure that alldevices are able to communicate with one another.In the future, it is likely that use of industry standard("open") communication protocols will achievecritical mass, and as a result nearly all manufacturers willadhere to them. At this point in time, however, their useis still limited.12.3.8 Tackling the Human Dimensionin EMCS SpecificationsTraining should also be considered a part of thecommissioning. In fact, training and long-term supportare often the most overlooked aspects of an EMCSproject. In what form and for whom training should beprovided differs for every project. This is because the“people” involved are different every time.In the end, the performance of an EMCS dependson the operating personnel. Those who are to work withthe system must understand it. The most common causefor the failure of an EMCS to achieve intended results isignorance. If a system is overridden frequently, or if controlsare disconnected, or if parameters and schedulesare not properly maintained the EMCS becomes useless.APPENDIX B includes training and documentation instructionsin sections B10, B11, B12, B13, and B14.Training is not something that can be adequatelydescribed in a few short paragraphs in a specification. Inreality, there needs to be Comprehensive, Explicit, andRelevant Training and Support (CERTs). CERTs can beused to prevent misunderstandings and dissatisfactionon the part of the building owner, operators and tenants.CERTs also prevent system disconnections and “vaporwatts”(energy savings that never materialize).In 12.B11.2, the boiler plate reads, “Contractor shallorient the training specifically to the system installedrather than a general training course.” A case can bemade for expanding this to include business specific andfacility specific instructions. In new construction, thesewould be determined by peering as far into the future aspossible to determine what might be required. The bestapproach here is to ask copious questions about how thefacility will be used, staffed, serviced, and marketed (inthe case of an office building).In retrofit applications, the depth and quality of thequestions asked during the facility appraisal (refer backto Section 12.3.8) heavily influence the detail spelledout in the specification. For example, for the EMCSRETROFIT discussed in 12.2.2, a list of facility specificinstructions might also include the following:Program user-friendly, menu-prompted, afterhoursaccess setup routines.

332 ENERGY MANAGEMENT HANDBOOKProgram user-friendly, menu-prompted, monthlyafter-hours access bill printing routines. Set up ten(10) test access codes which demonstrate optionstenants will have for naming after hours zones.Print out bill examples using hourly rates whichwill be billed in this building.Provide one 2-hour training session in after-hoursaccess code programming and bill printing forbuilding engineer.Provide two 2-hour training sessions includingsystem overview, control strategies, and temperatureand status data interpretation, for propertymanager.Provide two 2-hour training sessions in after-hoursaccess code logic, benefits, options, and programmingfor property manager.Provide one 2-hour training session in after-hoursbill printing for property manager and propertymanagement secretary.Provide color copies of access code maps fortenants to easily determine what areas will be illuminatedand conditioned when accessed afterhours.Conduct a meeting with all tenants’ designatedrepresentatives to explain how to request HVACand/or lighting by telephone and to determine theneed for additional access code sub-zones.Conduct follow-up meeting with all tenants’ designatedrepresentatives to review first month’sbilling and deliver access code maps documentingsub-zones requested in first meeting.Conduct a meeting with janitorial service managementpersonnel to explain the use of access codes(lighting only).Provide two 2-hour training sessions for outsideleasing agents in EMCS benefits (i.e. temperaturecontrol, efficiency, improved maintenance and after-hoursfairness), why a building with an EMCSis different and better than a building without anEMCS.Attend up to two lease negotiation meetingsduring first 12 months after start-up to explainafter-hours and other EMCS benefits to prospectivetenants.Act as guest speaker at a tenant meeting to explainEMCS benefits beyond the after-hours fairness alreadyexplained in private meetings.Conduct one 2-hour meeting to explain EMCSoperation, provide basic documentation, and establishcommunication and service-call protocol withall service vendors (i.e. security, HVAC, lighting,and electrical).The above are only a few of what might be a longlist of services which could be included in an EMCSspecification for a particular building owner to ensurethat projected benefits are realized.Whether one is a consulting engineer, a contractordoing a design/build project, or a facility director writinga guide specification, every effort should be made toanticipate future needs. EMCS projects priced, negotiated,or bid without the benefit of ‘complete’ specificationsare doomed to less than stellar performance at best andtotal failure at worse. The human dimension may be themost challenging aspect of all of the factors to weigh andincorporate in an EMCS design.12.3.9 Equipment and ContractorQualification and SelectionMany EMCS projects have gone astray becausedecisions as to system manufacturer and installer (orbid list) were overly influenced by the features of EMCSequipment. There is a big difference between ‘capabilities’and ‘effective application of appropriate control andfacility specific strategies’.Too much emphasis on a ‘brand’ of EMCS overlooksthe vital role of the contractor in the success orfailure of a project. Even when protected by the mostdetailed of specifications, it pays to look beyond thelow bid and conduct a thorough investigation of theinstallers being considered. Several of the major EMCSmanufacturers also have contracting divisions. Talentand experience often vary immensely from branch officeto branch office. Just because a company has beenin business for 50 to 100 years does not automaticallymean it is the best choice for a particular project. Afterall, none of the ‘people’ who will be involved in theproject have worked for that company for 50 to 100years.Thorough interviews, inspections of previous jobsites, conversations with references, and visits to potentialcontractors’ offices to meet ‘the delivery team’ are

ENERGY MANAGEMENT CONTROL SYSTEMS 333all good ways to increase the chances of quality installation.The delivery team will have a greater impact onthe outcome of a project than the EMCS equipment.A thorough investigation of the entire team should bemade and only the qualified allowed to compete. Thesame argument could easily be made for the selectionof a consulting engineer.EMCS technology does not save energy, not on itsown at any rate. People save energy by using EMCStechnology as a tool. By seeking out the crafts-peopleand the people most focused on saving energy, thechances of a project paying off are significantly greater.Appendix AEMCS SOFTWARE SPECIFICATIONSThis Appendix provides a sample of the descriptiveportion of an EMCS software specification for anew construction junior high school in Connecticut.The school is constructed of red brick, has a pitched,well insulated roof and comprises 140,000 sq. ft. Entryways are double doored. Windows comprise a moderatepercentage of outside wall area and are of the doublepane construction. The building is fully air conditioned.The HVAC systems consists of a central boiler/chillerplant and single duct distribution without static airpressure control. Equipment is as follows: 1) a bank of10 modular oil fired hot water boilers controlled by themanufacturer’s microprocessor controller that stagesthe units on as needed to maintain a temperature resetschedule as determined by the outside air temperature,2) 2 two-stage reciprocating chillers connected in parallel,3) a single cell cooling tower with a two speed fan,4) one condenser water circulator, 5) a pair of spaceheating circulators to circulate hot water, 6) a pair ofspace cooling circulators to circulate chilled water, 7) 17air handling units—10 of which contain a preheat coil,a supply air fan, an outside air damper and a returnair damper supply air to terminal units—7 of the unitswhich contain a cooling coil, a heating coil, a supplyair fan, an outside air damper and a return air dampersupply air directly to the space, 8) one makeup air unitwith an outside air damper, 9) 100 space terminal unitseach containing a reheat coil, a chilled water coil anda damper, 10) an oil fired domestic service hot waterheater and associated circulator to supply heat energy toa heat exchanger in the domestic service hot water storagetank, 11) one circulator to circulate domestic servicehot water through the building, 12) 21 attic exhaust fans,13) 14 cabinet unit heaters, and 14) 27 toilet exhaust fansall to be controlled by a distributive DDC EMCS withpeer-to-peer communication. There are several exhaustfans for ventilating mechanical rooms, fume hoods, sciencerooms, the auditorium stage and storage rooms.These units are locally controlled on an as needed basis.Occupancy sensors perform space lighting control.The reader should note that sequencing logicrequirements are organized into three sections: 1) administrative,2) system, and 3) operating.12.A1.0 GENERAL12.A1.1 The listings in this section are meant to providea logical guide for the sequence of operations to be usedfor each point of control. There are many considerationsother than those listed in this section which must beincorporated into a proper energy management systemcontrol program.12.A1.2 CONTRACTOR must provide programs in thelanguage that is appropriate for the hardware that isused. The programs written in the primary languageof the controller must conform in operation and functionalityto the sequences listed in these specifications.CONTRACTOR IS LIABLE FOR PROOF OF COMPLI-ANCE WITH THESE SPECIFICATIONS. BASIS FORPROOF OF COMPLIANCE WILL BE COMPARINGCONTRACTOR’S WRITTEN PROGRAM CODE WITHTHE SEQUENCES PROVIDED IN THESE SPECIFICA-TIONS AND FIELD TESTING OF ACTUAL SYSTEMOPERATION. EACH CONTROL POINT IN THE SYS-TEM MUST BE EXERCISED IN THE PRESENCE OFOWNER’S REPRESENTATIVE TO PROVE PROPEROPERATION.12.A2.0 ADMINISTRATIVE SEQUENCE LOGIC12.A2.1 BASIC SCHEDULE: Install school occupancyschedules for the building as follows:a. School Year: Provide capability for the user to easilyand conveniently schedule the start and stop datesfor the normal school term (School Year = 1). For thecurrent School Year install August 28 as the start dateand June 18 as the end date.b. School Day: Provide capability for the user to easilyand conveniently schedule a day of the week duringthe School Year for normal school activity (School Day= 1). Provide calendar scheduling for the current schoolyear.

334 ENERGY MANAGEMENT HANDBOOKc. School Hours: Provide capability for the user toeasily and conveniently schedule occupancy hours duringthe School Day for the normal daily school session(School Hours = 1). For the current year install 7:00 amas the start time and 3:00 pm as the end time.d. Regular Office Hours: Provide capability for theuser to easily and conveniently schedule regular occupancyhours during the School Day for the centraladministrative office (Regular Office Hours = 1). For thecurrent School Year install 7:00 am as the start time and5:00 pm as the end time.e. Vacation Office Hours: Provide capability for theuser to easily and conveniently schedule administrativeoffice occupancy hours for days other than a School Day(Vacation Office Hours = 1). For the current year install8:00 am as the start time and 3:00 pm as the end time.12.A2.2 ACTIVITY SCHEDULING: Install capability forthe user to easily and conveniently schedule for up totwelve months in advance two (2) occupancy periods perday in addition to the Basic Schedule for an individualarea activity (XXX Activity = 1). Program the activitiescurrently scheduled on the current school calendar.a. Whenever an activity is scheduled, all equipmentnecessary to maintain normal occupancy conditions forthat area at the date and time of that activity shall beenabled for occupancy.b. The user must be able to easily and convenientlyinitiate, set time parameters and cancel each of the belowlisted scheduled activities as a separate function.1. Clrm East Activity 9. Admin Activity2. Clrm S.E. Activity 10. Main Gym Activity3. Clrm South Activity 11. Aux Gym Activity4. Clrm S.W. Activity 12. Music Area Activity5. Clrm West Activity 13. Cafeteria Activity6. Clrm North Activity 14. Industrial Arts Activity7. Media Activity 15. Open House8. Auditorium Activityc. The following are equipment assignments forthe above scheduled activity areas.1. Points affected by CLRM EAST ACTIVITY:Boilers—Chillers—AHU-12. Points affected by CLRM S.E. ACTIVITY:Boilers—Chillers—AHU-23. Points affected by CLRM SOUTH ACTIVITY:Boilers—Chillers—AHU-34. Points affected by CLRM S.W. ACTIVITY:Boilers—Chillers—AHU-45. Points affected by CLRM WEST ACTIVITY:Boilers—Chillers—AHU-56. Points affected by CLRM NORTH ACTIVITY:Boilers—Chillers—AHU-67. Points affected by MEDIA ACTIVITY:Boilers—Chillers—AHU-78. Points affected by AUDITORIUM ACTIVITY:Boilers—Chillers—AHU-8—AHU-15 DomesticHot Water9. Points affected by ADMIN ACTIVITY:Boilers—Chillers—AHU-910. Points affected by MAIN GYM ACTIVITY:Boilers—Chillers—AHU-10Domestic Hot Water11. Points affected by AUX GYM ACTIVITY:Boilers—Chillers—AHU-11Domestic Hot Water12. Points affected by MUSIC ACTIVITY:Boilers—Chillers—AHU-1213. Points affected by CAFETERIA ACTIVITY:Boilers—Chillers—AHU-13—FCU-2Domestic Hot Water14. Points affected by INDUSTRIAL ARTS ACTIVITY:Boilers—Chillers—AHU-14—MAU-115. Points affected by OPEN HOUSE OVERRIDE:All control points12.A2.3 INDIVIDUAL OVERRIDES: Install timed overridesfor each output point as follows:a. At any time the user must be able to easily andconveniently Enable/on or Disable/off any output pointfor a user selected time period.b. At the end of the override period that pointoverridden shall return to its normal programmed controland the override period shall be reset to zero.

ENERGY MANAGEMENT CONTROL SYSTEMS 335c. An Individual Override shall have priority overthe Snow Day Override.12.A2.4 SNOW DAY OVERRIDE: Install an easy andconvenient method to override off any scheduled occupancyand keep the entire building in the unoccupiedmode (Snow Day =1).a. A Snow Day override shall automatically terminateat 23:59 hours each day.12.A2.5 GLOBAL SOFTWARE POINTS: The user variablesoftware points identified by bold print in thisspecification shall be referenced in the algorithms suchthat the user need only to enter a new value for a namedsoftware point to effect a value change in all algorithmswhich include that software variable.12.A3.0 SYSTEM SEQUENCE LOGIC12.A3.1 SYSTEM MODE: Install capability for the userto easily and conveniently set and vary annual beginningand ending dates for the heating season to preventsimultaneous heating and cooling. Heating equipmentoperation will be determined by the time of year (i.e.space heating boilers and circulators will only operateduring the heating season And be locked out during therest of the year regardless of outside air temperature).Install October 15th to May 15th as initial parametersfor the heating season. Heating season System Mode =0, non-heating season System Mode = 1.12.A3.2 SUMMER/WINTER: Install user adjustable variablesfor heating and cooling operations during buildingoccupancy as follows:a. Whenever System Mode = 0 and the outside airtemperature is less than or equal to the user variableWinter Occupancy Setpoint (i.e. 65°F), summer/wintershall go to winter. If the outside air temperature isgreater than the Winter Occupancy Setpoint and thehighest occupied space temperature + greater than someuser variable setpoint (i.e. 74°F) and Chiller Plant Operationis Enabled/on then summer/winter shall go tosummer to allow the chillers to function. If the outsideair temperature is greater than the Winter OccupancySetpoint and Chiller Plant Operation is Disable/off,summer/winter shall remain in winter to allow maximumventilation with outside air.b. Whenever System Mode = 1 and the outside airtemperature is greater than or equal to the user variableSummer Occupancy Setpoint (i.e. 60°F), summer/wintermode shall go to summer to allow the chillers tofunction. If the outside air temperature is less than theSummer Occupancy Setpoint and the lowest occupiedspace temperature is greater than some user variablesetpoint (i.e. 72°F) and is disabled/off, summer/wintermode shall go to winter to allow maximum ventilationwith outside air. If the outside air temperature is lessthan the Summer Occupancy Setpoint and Chiller PlantOperation is Enabled/on, summer/winter mode shallremain in summer mode.12.A3.3 OPTIMUM START: Install an adaptable optimumstart of all control points for both heating andcooling functions as follows:a. During optimum start for both heating andcooling, outside air dampers shall go fully closed; and,classroom and return air dampers shall go fully open;and, all exhaust fans shall be Disabled/off.b. Optimum start shall be limited to a maximumnumber of hours and be user adjustable.12.A3.4 CHILLER PLANT OPERATION: Install capabilityfor the user to easily and conveniently set and varyannual beginning and ending dates for operation of thechiller plant. Install May 1 to November 10 as initialparameters for Chiller Plant Operation.12.A3.5 DESIGNATED SPACE TEMPERATURE SET-POINTS: Install user variable space temperaturesetpoints as follows:a. The occupied setpoint for all Classrooms, allOffices, the Media Center, the Music Area, the Cafeteriaand Kitchen, The Industrial Arts Area, and theAuditorium shall be 72°F for both heating and coolingfunctions.b. The occupied setpoint for the Locker Roomsshall be 74°F for both heating and cooling functions.c. The occupied setpoint for the Main Gymnasiumand the Auxiliary Gymnasium shall be 68°F for heatingand 74°F for cooling.d. The unoccupied heating Setback Temperatureshall be 55°F.e. The unoccupied cooling Setup Temperatureshall be 80°F.

336 ENERGY MANAGEMENT HANDBOOK12.A3.6 FREE COOLING: Install free cooling into allalgorithms that control air handling unit fans, attic exhaustfans, outside air dampers, return air dampers andspace terminal unit dampers as follows:a. Free cooling shall be Enabled/on to ventilatethe building during unoccupied periods whenever theoutside air temperature is greater than the user variableFree Cool Minimum (i.e. 45°F) and the enthalpy of theoutside air is less than the enthalpy of the inside airand the average building temperature is greater thanthe user variable Free Cool Setpoint (i.e. 72°F).b. Whenever free cooling is Enabled/on all outsideair dampers, return air dampers and space terminalunit dampers shall go to the fully open position; and,all air handling units and attic exhaust fans shall beEnabled/on; and, the chiller plant and boiler plant shallbe Disabled/off locked out.12.A3.7 MINIMUM ON/OFF: Install a user adjustableMinimum On/Off time period (i.e. 5 minutes) in thecontrol program for each piece of mechanical equipment.12.A4.0 OPERATING SEQUENCE LOGIC12.A4.1 AHU START/STOP: Install Enable/Disable ofAir Handling Units as follows:a. Whenever School Hours = 1 and Snow Day = 0all Air Handling Units shall be Enabled/on to start thesupply air fans.b. Whenever Regular Office Hours = 1 or VacationOffice Hours = 1 AHU-9 shall be Enabled/on.c. Whenever an individual area activity is scheduledon (i.e. a certain XXX Activity = 1) and Snow Day= 0 the air handling unit serving that area shall be Enabled/on.d. Whenever School Hours = 0 and the activityflag for an individual area (i.e. XXX Activity = 0) the airhandling unit serving that area shall be Disabled/off.e. Whenever Regular Office Hours = 0 or VacationOffice Hours = 0 and Admin Activity = 0 AHU-9 shallbe Disabled/off.12.A4.2 AUDITORIUM AHU: Install control of the AirHandling Unit for the Auditorium as follows:a. Whenever School Day = 1 and Snow Day = 0and the Auditorium is unoccupied as determined bythe Auditorium occupancy sensors AHU-8 shall be dutycycled as a standby load to maintain the DesignatedSpace Temperature for the Auditorium within the uservariable Space Dead Band (i.e. 6°F).b. If, during the standby mode, the Auditoriumshould become occupied as determined by the Auditoriumoccupancy sensors for a user variable time period(i.e. 5 minuets) AHU-8 shall be Enabled/on to maintainthe Designated Space Temperature for the Auditorium.c. Whenever Auditorium Activity = 1 and SnowDay = 0 AHU-8 shall be Enabled/on to maintain theDesignated Space Temperature for the Auditorium.12.A4.3 MAIN GYMNASIUM AHU: Install control ofthe Air Handling Unit for the Main Gymnasium as follows:a. Whenever School Day = 1 and Snow Day = 0AHU-10 shall be duty cycled to maintain the DesignatedSpace Temperature for the Main Gymnasium within theSpace Dead Band.b. Whenever Main Gym Activity = 1 and SnowDay = 0 AHU-10 shall be Enabled/on to maintain theDesignated Space Temperature for the Main Gymnasium.12.A4.4 MAKEUP AIR UNIT (MAU -1): Install Enable/Disable control of MAU-1 such that MAU-1 will be allowedto operate during School Hours and an IndustrialArts Activity. Actual control of on/off while MAU-1 isEnable/on will be by local toggle switch.12.A4.5 AHU OUTSIDE AIR & RETURN AIR DAMP-ERS: Install control of the Air Handling Unit dampersas follows:a. Whenever a unit is in normal operation (the supplyair fan is functioning and free cool or optimum startis not in effect) and the outside air damper for that unitis not under a lockout the dampers shall be calibratedto maintain the user variable Minimum Fresh Air IntakeVolume (i.e. 20%).b. Whenever a unit is not operating the outsideair damper and return air damper for that unit shall gofully closed.

ENERGY MANAGEMENT CONTROL SYSTEMS 337c. Whenever in System Mode = 0 and the outsideair temperature is equal to or greater than the uservariable Winter Lockout Setpoint (i.e. 20°F) outside airdampers and return air dampers on operating Air HandlingUnits shall be Enabled/on and modulate underthe control of a user adjustable PID loop to maintainthe user variable Mixed Air Temperature (i.e. 55°F) forsupply air to the preheat or heating coil.d. Whenever in System Mode = 1 and the outsideair temperature is equal to or less than the user variableSummer Lockout Setpoint (i.e. 85°F) outside air dampersshall be Enabled/on and go to minimum position; and,return air dampers shall go to the fully open position.e. Whenever the outside air temperature is lessthan the Winter lockout Setpoint or greater than theSummer Lockout Setpoint outside air dampers shall beDisabled/off and go fully closed; and, return air dampersshall go to the fully open position.12.A4.6 AHU PREHEAT COIL CONTROL VALVES:Install control of preheat coil control valves for the AirHandling Units as follows:a. The AHU preheat coil control valve shall gofully open whenever air flow to the coil stops.b. Whenever System Mode = 0 and there is airflow to an AHU preheat coil that coil control valveshall modulate under the control of a user adjustablePID loop to maintain the user variable Discharge AirTemperature (i.e. 60°F).12.A4.7 AHU SPACE HEATING COIL CONTROLVALVES: Install control of heating coil control valvesfor the Air Handling Units as follows:a. The AHU heating coil control valve shall gofully open whenever air flow to the coil stops.b. Whenever System Mode = 0 and there is air flowto the heating coil that coil control valve shall modulateunder the control of a user adjustable PID loop to maintainthe designated Space Temperature Setpoint for thatspace.12.A4.8 AHU COOLING COIL CONTROL VALVES:Install control of cooling coil control valves for the AirHandling Units as follows:a. Whenever System Mode = 0 the cooling coilcontrol valve shall go fully open.b. Whenever System Mode = 1 and there is air flowto the cooling coil that coil control valve shall modulateunder the control of a user adjustable PID loop to maintainthe designated Space Temperature Setpoint for thatspace.c. Whenever System Mode = 1 and air flow to acooling coil stops that AHU cooling control valve shallgo fully closed.12.A4.9 SPACE TERMINAL UNIT DAMPERS & CON-TROL VALVES:a. During an occupancy mode whenever a space isoccupied as determined by that space’s motion sensor,the damper for the terminal unit serving that space shallgo fully open and remain open for the duration of theoccupancy; and, the hot water heating valve or chilledwater cooling valve, as the season may be, shall modulateunder the control of a user adjustable PID loop tomaintain the designated Space Temperature Setpoint forthat space.b. During an occupancy mode whenever a space isunoccupied as determined by that space’s motion sensorthat space’s terminal unit damper and heating or chilledwater cooling valve, as the season may be, shall go fullyclosed.c. If, during an occupancy mode, a space is unoccupiedas determined by that space’s motion sensor andthat space’s temperature should exit the user variableSpace Dead Band the damper for that space’s terminalunit shall go fully open; and, the hot water heatingvalve or chilled water cooling valve, as the season maybe, shall modulate under the control of the PID loopto return that space’s temperature to within the SpaceDead Band.d. When a space is in the unoccupied mode andSystem Mode = 1 and that space’s temperature is lessthan the Setup Temperature that space’s terminal unitdamper and chilled water cooling valve shall go fullyclosed.e. If, during the unoccupied mode and SystemMode = 1, the temperature in a space should exceedthe Setup Temperature, then the terminal unit damperfor that space shall go fully open; and, the chilled watercooling valve for that space shall modulate under thecontrol of the PID loop to cool that space to a temperaturethat is less than the Setup Temperature by an

338 ENERGY MANAGEMENT HANDBOOKamount equal to the user variable Setback Differential(i.e. 2°F).f. If, during the unoccupied mode, the temperaturein a space should fall below the Setback Temperature,then the terminal unit damper for that space shall gofully open; and, the hot water heating valve for thatspace shall modulate under the control of the PID loopto heat that space until the temperature in that spaceexceeds the Setback Temperature by an amount equalto the Setback Differential.12.A4.10 SPACE HEATING CIRCULATORS: Install controlof space heating circulators as follows:a. Whenever the building is occupied during SystemMode = 0 and the outside air temperature is lessthan or equal to the Winter Occupancy Setpoint the leadspace heating circulator shall be Enabled/on.b. Whenever the building is occupied during SystemMode = 0 and the outside air temperature is greaterthan the Winter Occupancy Setpoint the lead space heatingcirculator shall be Disabled/off.c. Whenever the building is unoccupied duringSystem Mode = 0 and the outside air temperature isgreater than or equal to the user variable Freeze Point(i.e. 35°F) the space heating circulators shall be Disabled/off.d. Whenever the building is unoccupied duringSystem Mode = 0 and the outside air temperature isgreater than or equal to the Freeze Point and the lowestspace temperature falls below the Setback Temperaturethe lead space heating circulator shall be Enabled/onuntil all space temperatures exceed the Setback Temperatureby an amount equal to the Setback Differential.e. Whenever the outside air temperature is lessthan the Freeze Point the lead space heating circulatorshall be Enabled/on.f. The circulators will alternate lead/lag every 14days and the then current lagging circulator will Enable/onin the event that the lead circulator should fail.Circulator failure is to be determined by a flow switchat each circulator discharge.12.A4.11 BOILERS: Install control of boiler operation asfollows:a. Unless boiler override is on, boiler operationshall be allowed only if System Mode = 0.b. Whenever a space heating circulator is Enabled/on the boilers shall be Enabled/on to maintain the appropriatesupply water temperature as determined bythe boiler reset controller.c. Whenever space heating circulators are Disabled/offthe boilers shall be Disabled/off.12.A4.12 CHILLED WATER CIRCULATORS: Install controlof chilled water circulators as follows:a. Whenever the building is occupied and summer/winteris in summer and Chiller Plant Operationis Enabled/on the lead circulator shall be Enabled/on.b. Whenever the building is unoccupied and thehighest space temperature in a cooling area is less thanor equal to the Setup Temperature the chilled watercirculators shall be Disabled/off.c. Whenever the building is unoccupied and thehighest space temperature in a cooling area is greaterthan the Setup Temperature the lead chilled water circulatorshall be Enabled/on until all space temperaturesare less than the Setup Temperature by an amount equalto the Setback Differential.d. The circulators will alternate lead/lag every 14days and the then current lagging circulator will Enable/onin the event that the lead circulator should fail.Circulator failure is to be determined by a flow switchat each circulator discharge.12.A4.13 CHILLERS: Install control of chillers as follows:a. Whenever a chilled water circulator is Enabled/on the lead chiller shall be Enabled/on.b. A chiller shall operate to maintain chilled waterdischarge (supply to loop) temperature as determinedby that chiller’s controller.c. During School Hours only, whenever the leadchiller is Enabled/on and the outside air temperature isgreater than the user variable Chiller Maximum OutputPoint (i.e. 85°F) the lag chiller shall be Enabled/on.

ENERGY MANAGEMENT CONTROL SYSTEMS 339d. Whenever chilled water cir culators are Disabled/offthe chillers shall be Disabled/off.e. The chillers will alternate lead/lag every 14days and the then current lagging chiller will Enable/onin the event that the lead chiller should fail.12.A4.14 ATTIC EXHAUST FANS: Install control foroperation of attic exhaust fans as follows:a. Whenever the attic air humidity is greater thanthe user variable Attic Maximum Humidity Point (i.e.60%) attic exhaust fans shall be Enabled/on.12.A4.15 TOILET EXHAUST FANS: Install control foroperation of toilet exhaust fans as follows:a. Whenever the building is occupied all toiletexhaust fans shall be Enabled/on.b. Whenever and activity is scheduled on the toiletexhaust fans serving the building area associated withthat activity shall be Enabled/on.c. Whenever the building is unoccupied toilet exhaustfans shall be Disabled/off.12.A4.16 CABINET UNIT HEATERS: Install control ofthe cabinet unit heaters as follows:a. Whenever the building is occupied during SystemMode = 0 and the outside air temperature is lessthan or equal to the Winter Occupancy Setpoint the unitcabinet heaters shall be Enabled/on; and, the hot waterheating valve shall modulate under the control of a useradjustable PID loop to maintain space temperature at auser variable setpoint (i.e. 65°F).b. Whenever the building is occupied during SystemMode = 0 and the outside air temperature is greaterthan the Winter Occupancy Setpoint the unit cabinetheaters shall be Disabled/off.c. Whenever the building is unoccupied and theoutside air temperature is equal to or greater than theFreeze Point the unit cabinet heaters shall be Disabled/off.d. Whenever the building is unoccupied and theoutside air temperature is less than the Freeze Point theunit cabinet heaters shall be Enabled/on; and, the hotwater heating valve shall modulate under the control ofthe PID loop to maintain space temperature at a uservariable setpoint (i.e. 55°F).12.A4.17 DOMESTIC SERVICE HOT WATER: Installcontrol of the domestic service hot water system asfollows:a. During School Hours and scheduled activitiesfor the Cafeteria, Main Gymnasium and Auxiliary Gymnasiumthe domestic service hot water heater and loopcirculators and burner shall be Enabled/on. The localaquastat will cycle the heater circulator and burner onand off as required to maintain the desired loop temperature.b. During unoccupied periods the domestic servicehot water system shall be Disabled/off.12.A4.18 COOLING TOWER: The cooling tower shallbe interlocked with the chillers such that whenever achiller is Enabled/on the condenser water circulator andcooling tower fan are Enable/on.a. Cooling tower fan speed shall be determined bythe cooling tower controller.12.A4.19 AHU FREEZE PROTECTION: Install freezeprotection for the Air Handling Units and the Make-upAir Unit as follows:a. Whenever in System Mode = 0 and the air temperatureas measured at the face of the discharge sideof a preheat or heating coil is less than the user variableFreeze Stat Point (i.e. 45°F) the supply fan for that unitshall be Disabled/off and the outside air damper shallbe Disabled/off and go fully closed.b. If a unit has been Disabled/off because of freezeprotection, should the air temperature as measured atthe face of the discharge side of a preheat or heatingcoil increase to a value greater than the Freeze Stat Pointplus some user variable differential (i.e. 50°F) that unitshall be returned to its normal programmed control.Appendix BEMCS INSTALLATION REQUIREMENTSIn addition the Software Specifications and appropriatedrawings and sketches necessary to integrate a DDCEMCS into a building, the following can be used as aguide in developing specifications to select a supplierand installer of the desired system.

340 ENERGY MANAGEMENT HANDBOOK12.B1.0 GENERAL12.B1.1 The system within each specific building shallbe configured as a distributive (or central as the casemay be) control and monitoring system. All computingdevices, as designed in accordance with FCC Rules andRegulations Part 15, shall be verified to comply withthe requirements for Class A computing devices. Thesystem shall perform all data consolidation, control, timing,alarm, feedback status verification and calculationfor the system.12.B1.2 The system controller (or controllers as the casemay be) shall function as the overall system coordinator,perform automatic energy management functions,control peripheral devices and perform calculationsassociated with operator interactions, alarm reporting,and event logging.12.B1.3 The system shall have at least 48 hours of batterybackup to maintain all volatile memory and thereal time clock.12.B1.4 The system shall have a battery backed uninterruptiblereal time clock with an accuracy within 10seconds per day.12.B1.5 The system shall operate on 120 VAC, 60 Hzpower. Operation shall continue in accordance with theDetailed Specifications from line voltages as low as 95VAC. Line voltages below the operating range of thesystem shall be considered outages.12.B1.6 The system shall be capable of interfacing withno additional hardware, other than the appropriate cabling,to industry standard terminals/PCs. The interfaceshall support as a minimum: RS232 EIA port, ASCIIcharacter code with baud rate selectable between 300-9600 baud.12.B1.7 The system shall have the capability of connectingto a local terminal/PC or printer to perform allprogramming, modification, monitoring and reportingrequirements of the Detailed Specifications.12.B1.8 The system shall support a Hayes Smartmodemwith full RS232 control signals to communicate withremote terminals and other systems via dial-up voicegrade phone lines at a minimum of 1200 baud.12.B1.9 The system shall accept control programs inaccordance with the Software Specifications, performautomated energy management functions, control peripheraldevices and perform all necessary mathematicalcalculations.12.B1.10 The system central processing unit shall be amicrocomputer of modular design. The word size shallbe 8 bits or larger, with a memory cycle time less than1 micro second. All chips shall be second sourced.12.B1.11 Any control program of a system may affectcontrol of another program if desired. Each programshall have full access to all input/output facilities ofthe system.12.B1.12 The system shall bi-directionally transfer datato input/output units and local control units at a minimumof 720 baud.12.B1.13 The system shall provide signal conditioningto insure integrity.12.B1.14 The system shall be user friendly and shall allowOWNER to designate appropriate nomenclature forall inputs, outputs programs, routines and menus.12.B2.0 I NSTALLATION PARTICULARS12.B2.1 CONTRACTOR shall install all equipment,devices, cabling, wiring, etc., in compliance with manufacturer’srecommendations.12.B2.2 The system shall have key lock protected accessto internal circuitry and any output override switches.12.B2.3 The system controller(s) shall be properly locatedto provide easy access, and to ensure the ambienttemperature range is +32 degrees F to +95 degrees F,10% to 95% relative humidity non-condensing. Locationand accessibility for removal of covers for repairshall be sufficient to facilitate required access by servicepersonnel. Location shall be approved by OWNER’SREPRESENTATIVE before installation.12.B2.4 The load on any output shall not exceed 50% ofthat output’s contact rating.12.B2.5 Only one (1) wire pair per input/output pointshall terminate within any input/output unit enclosure.All relays, wire junctions, terminal strips, transformers,modems and other devices not an integral part ofthe input/output unit shall be located external to theinput/output unit in a separate suitable enclosure(s)properly mounted.

ENERGY MANAGEMENT CONTROL SYSTEMS 34112.B2.6 Each processing unit and controller shall havea separate dedicated electric service of the appropriateamperage.12.B2.7 All electrical relays, electric/pneumatic relays(EP’s) and other control devices shall be mounted inan appropriate enclosure, securely fastened and clearlylabeled as to purpose and item controlled. All enclosuresshall be of adequate size to allow easy access. Aminimum of 70 cubic inches shall be allowed per deviceand the minimum size enclosure shall be 6" × 6" × 4".All relays shall be base mounted and face out towardstheir enclosure opening. Relays not applicable to basemounting shall have screw type wire connections.12.B2.8 Where stranded wire is used all connectionsshall be made with solderless connections.12.B2.9 All wiring and cabling shall be labeled witha logical numbering system. The labels shall remainconsistent from the input/output unit to terminationwith labels being clearly visible at all terminations andpenetrations.12.B2.10 For all new work, temperature sensors on pipingshall be installed in wells.12.B2.11 Where strap-on sensors are used in existingpiping, sensors shall be bedded in heat conductingcompound with a shield and securely fastened to pipeunder insulation with strapping as recommended bymanufacturer. If not present, insulation as approvedby OWNER’S REPRESENTATIVE shall be installed toextend 12" on both sides of sensor. Sensors shall not beinstalled on PVC pipe.12.B2.12 Sensors in ductwork shall protrude into airstream.12.B2.13 Where a system output directly or indirectlycontrols equipment previously controlled by a timeclock, that time clock shall be disconnected or otherwiseremoved from the control circuit of the equipment.12.B2.14 Where a system output directly or indirectlycontrols equipment which has in its control circuit aHand On-Off-Auto controller, the Hand On functionshall be taken out of the control circuit.12.B2.15 All control and sensor wiring in boiler rooms,equipment rooms, electrical rooms and unfinished areasshall be enclosed in metal conduit.12.B2.16 All control and sensor wiring in finished areasshall be enclosed in metal wire mold.12.B2.17 All control and sensor wiring in non-accessiblespaces such as ceilings and tunnels shall be tied in agood workmanship manner acceptable to OWNER’SREPRESENTATIVE (maximum 3' spacing) to a suitablesupport with approved ties. Said wiring shall not lie onceiling tiles. All wiring in plenum spaces shall be listedfor use in same.12.B2.18 Control of boilers, fans, pumps, chillers, compressorsor any other equipment which utilizes electricor fossil fuel motors as a power source shall be totallyby electrical means.a. Control points identified as “Electric” controlpoints shall be wired such that the digital signal sentby a control unit is received directly by the device beingcontrolled. Unless otherwise specified, any use ofpneumatics to interface an “Electric” control point withthe device to be controlled is not acceptable.12.B2.19 Use of pneumatics for control of heating/coolingvalves, hot/chilled water reset valves, dampers orsimilar modulating devices is acceptable.a. In the event that CONTRACTOR should utilizeany component of existing pneumatic controls to interface“Pneumatic” control points with the device(s) tobe controlled, CONTRACTOR shall be responsible forcontinuity of the pneumatic signal such that the signalfrom the control unit is transmitted undiminished andas intended to each device to be controlled.12.B2.20 All conduit, boxes, and enclosures shall bemetal.12.B2.21 All wiring carrying 120 volts or more shall be#12 AWG minimum.12.B2.22 All enclosures shall be labeled on the outsidestating purpose and relays or devices enclosed. Allrelays and other devices shall be labeled inside the enclosuresto clearly designate their use. Relays are to belabeled at the base or adjacent to the base so that therelay can be replaced without losing the label.12.B2.23 Labels shall be of engraved phenolic typeriveted to the enclosure. Minimum letter size shall be1/4". Labels shall be approved by OWNER’S REPRE-SENTATIVE.

342 ENERGY MANAGEMENT HANDBOOK12.B2.24 All control and sensor wiring shall be a minimumof #18 gauge wire. All wire in plenums shall beplenum rated.12.B3.0 G LOBAL DATA EXCHANGE12.B3.1 Each control unit and input/output unit in thesystem shall be able to exchange information betweenother control units, input/output units and the systemcontroller(s) each network scan.12.B3.2 Any network structure shall be transparent suchthat each control unit in the system may store and referenceany global variable available in the network for usein the controller’s calculations or programs.12.B4.0 INPUT/OUTPUT UNITS12.B4.1 Input/output units shall be used to connect thedata environment to the system and contain all necessaryinput/output functions to read field sensors andoperate controlled equipment based on instructionsfrom the central processing unit.12.B4.2 Input/output units shall be fully supervised todetect failures. The input/output units shall report thestatus of all points in its data environment at the rate ofat least once every 60 seconds for temperature control.Slightly longer times may be acceptable for non-criticalloads on an exception basis.12.B4.3 Upon failure of the input/output unit (includingtransmission failure), the input/output unit shall automaticallyforce outputs to a predetermined state.12.B4.4 For system voltages below 95 VAC the input/output unit shall totally shut down and de-energize itsoutputs.12.B5.0 INPUTS12.B5.1 The system shall be capable of receiving the followinginput signals:a. Analog Inputs (AI): The AI function shall monitoreach analog input, perform analog to digital (A/D)conversion, and hold the digital value in a buffer forinterrogation.b. Digital Inputs (DI): The DI function shall acceptdry contact closures.c. Temperature Inputs: Temperature inputs originatingfrom a thermister, shall be monitored andbuffered as an AI, except that automatic conversionto degrees F shall occur without any additional signalconditioning. Cable for temperature input wiring fromsensors that incorporate RFI circuitry in the sensor donot require shielded cable. Cable for temperature inputwiring that does not incorporate RFI circuitry in thesensor shall be shielded.d. Pulse Accumulators: The pulse accumulatorfunction shall have the same characteristics as the DI,except that, in addition, a buffer shall be required tototalize pulses between interrogations. The pulse accumulatorshall accept rates up to 10 pulse/second.12.B6.0 OUTPUTS12.B6.1 The system shall be capable of outputting thefollowing system signals:a. Digital Outputs (DO): The DO output functionshall provide dry contact closures for momentary andmaintained programmable operation of field devices.Any manual overrides shall be reported to the systemat each update.b. Analog Output (AO): The AO output functionmay be a true analog output providing a 0-l0VDC and/or 0-20ma DC signal; or, accept digital data, performdigital to analog (D/A) conversion and output a pulsewithin the range of 0.1 to 3276.6 seconds in order toperform pulse width modulation of modulating equipment.12.B7.0 DEVICES12.B7.1 Temperature Sensors: Temperature sensors shallhave a range suitable to their purpose. For example,typical degree ranges should be as follows: outsideair -40°F to +120°F, indoor spaces +40°F to +100°F, hotwater +80°F to +260°F, chilled water +20°F to +90°F,cold duct air +35°F to +80°F, and hot duct air +60°F to+200°F. RTDs, thermocouples and sensors based on theNational Semiconductor LM 34CZ chip are permittedfor sensing temperature. Thermisters of nominal resistanceof at least 10K ohms may be used for temperaturesensing also. Thermisters shall be encapsulated in epoxyor series 300 stainless steel.12.B7.2 Thermowells: Thermowells shall be series 300stainless steel for use in steam or fuel oil lines, and mo-

ENERGY MANAGEMENT CONTROL SYSTEMS 343nel for use in water lines. Thermowells shall be wroughtiron for measuring flue gases.12.B7.3 Pressure Sensors: Pressure sensors shall withstandup to 150% of rated pressure. Accuracy shall beplus or minus 3% of full scale. Pressure sensors shall beeither capsule, diaphragm, bellows or bourdon.12.B7.4 Pressure Switches: Pressure switches shall havea repetitive accuracy plus or minus 3% of range andwithstand up to 150% of rated pressure. Sensors shall bediaphragm or bourdon tube. Switch operation shall beadjustable over the operating pressure range. The switchshall have an application-rated for C, snapacting wipingcontact of platinum alloy, silver alloy or gold plating.12.B7.5 Watt-hour transducers: Watt-hour transducersshall have an accuracy of plus or minus 1 percent forkW and kWh and shall be internally selectable withoutrequiring the changing of current or potential transformers.12.B7.6 Potential Transformers: Potential transformersshall be in accordance with ANSI C57.13.12.B7.7 Current Transformers: Current transformersshall be in accordance with ANSI C57.13.12.B7.8 Watt-hour Meters with Diamond Register: Metersshall be in accordance with ANSI C12 and havepulse initiators for remote monitoring of watt-hour consumptionand instantaneous demand. Pulse initiatorsshall utilize light emitting diodes and photo-detectors.Pulse initiators shall consist of form C contacts with acurrent rating of 2 amps, 10 VA maximum and a liferating of one billion operations.12.B7.9 Flow Meters: Flow meter outputs shall be compatiblewith the system. Accuracy shall be plus or minus3% of full scale.12.B7.10 Control Relays: Control relay contacts shall berated for the application, with form C contacts, enclosedin a dustproof enclosure. Relays shall have silver cadmiumcontacts with a minimum life span rating of onemillion operations. When required, provide relays onH-0-A board with LEDs and relay status feedback.12.B7.11 Reed Relays: Reed relays shall have been encapsulatedin a glass type container housed in a plastic orepoxy case. Contacts shall be rated for the application.Operating and release times shall be five milliseconds orless. Reed relays shall have a minimum life span ratingof 10 million operations.12.B7.12 Solid State Relays (SSRs): Input/output isolationshall be greater than 10E9 ohms with a breakdownvoltage of 1500V root mean square or greater at 60Hz. The contact life shall be 10 x 10E6 operations orgreater. The ambient temperature range of SSRs shallbe minus 20 to plus 140 degrees F. Input impedanceshall not be less than 500 ohms. Relays shall be ratedfor the application. Operating and release time shall be100 milliseconds or less. Transient suppression shall beprovided as an integral part of the relay.12.B7.13 Surge Protection: Surge protection shall meetthe following minimum criteria.a. Solid state silicon avalanche junction diode deviceb. Response time less than or equal to 5ns (nanoseconds)c. Voltage protection level 300 volts maximumd. Service frequency 60 Hertze. Service voltage 110/120 volts ACf. Life expectancy > 10 yearsg. Ambient operating temperatures 0 to 40 degreesCelsiush. Suppression power 12,000 watts (1 × 1000 microsecondpulse)i. Instant automatic resetj. UL Listed12.B8.0 OPERATING SYSTEM12.B8.1 The operating system shall be real-time operatingsystem requiring no operator interaction to initiateand commence operations. The programming shallinclude:a. Operations and management of all devices;b. Error detection and recovery from arithmeticand logical functions;c. Editing software to allow the user to developand alter application programs; and,d. System self-testing.12.B9.0 SOFTWARE12.B9.1 The software shall be designed to run twentyfour hours a day, three hundred sixty five days a year.It shall continually compile and act upon acquired inputand display pertinent information on system operation.

344 ENERGY MANAGEMENT HANDBOOK12.B9.2 The software shall be user friendly, user interactiveand provide complete environmental and machinestatus information at the touch of a key. It shall allowthe user to change operation parameters through easilyunderstood display prompts.12.B9.3 The software shall be flexible and adaptable.All displays must be clear and concise, using the actualnames of the equipment and areas in question ratherthan employing a cryptic computer code to identifylocations and machinery.12.B9.4 All operator input shall require a minimum ofkeystrokes in response to prompted messages clearlyidentifying the required options for entries.12.B9.5 The software modules and data base shall bestored in nonvolatile or battery hacked memory toprevent loss of the operating parameters in the eventof a power failure. When the system starts, it shall firstreload all the parameters and decide what should beon or off. The system shall then scan the facility, seewhat equipment is actually running and update alltemperatures. If, in fact, the status is no longer what itshould be, the system shall stagger start all appropriateequipment.12.B9.6 The software shall provide telecommunicationcapability for remote support at a minimum of 1200baud. After logging onto the system with a remoteterminal, the remote user must have access to all userdefined variables (operational parameters) as describedhereinafter.12.B9.7 The software shall provide the capability toupload and download all data, programs, I/O pointdatabases and user configurable parameters from theremote terminal.12.B9.8 The software shall be capable of autodialing aminimum of three different phone numbers, in pulseor dial mode.12.B9.9 The software shall update the display of datawith the most current data available within 5 secondsof operator request.12.B9.10 The software shall allow the user to assignunique identifiers of his choice to each connected point.Identifiers shall have at least eight alpha/numeric characters.All reference to these points in programs, reportsand command messages shall be by these identifiers.12.B9.11 The software shall provide a simple methodfor the qualified systems integrator to designate controland monitoring characteristics of each control/monitorpoint. There shall be a menu selection to print the I/Opoint listings with current configuration data.12.B9.12 The software shall be custom configured forthe project and shall conform to the sequences thatare explicitly specified in the Software Specifications.All configuration must be completed by CONTRAC-TOR prior to acceptance. GENERIC SOFTWARE TOBE TOTALLY CONFIGURED BY THE END USER ISUNACCEPTABLE. The operator must be able to easilychange operating setpoints and parameters fromlegible, comprehensive menus, but input/output pointdefinition, control algorithms and machinery interlocksmust be custom tailored by CONTRACTOR to conformto the control sequences outlined. These algorithms andinterlocks are designed to coordinate the operation of allcontrolled machinery for optimum energy conservation.Access to input/output point definition and algorithmselection/alternation shall be available to a qualifiedsystems integrator through password protected routinesand configuration file manipulation, however, thesefunctions shall be invisible to the system operator.12.B9.13 Software shall be modular in nature and easilyexpandable for both more control points and/or morecontrol features.12.B9.14 All input gathered by remote sensors shall beglobally accessible to the system controller(s). Softwaredesignation rather than hardware configuration shalldefine the interrelationships of all input/output points.12.B9.15 The software shall provide at least 3 passwordaccess codes that can be user configurable for leveled accessto menu functions. All password protection shall beenabled or disabled through simple menu selections.12.B9.16 The execution of control functions shall notbe interrupted due to normal user communicationsincluding: interrogation, program entry, printout of theprogram for storage, etc.12.B9.17 In the dynamic display mode the softwareshall have the ability to activate real time logic tracers.These tracers shall display the logical sequences beingexecuted by the program as they occur.12.B10.0 ENERGY MANAGEMENT FUNCTIONS12.B10.1 Scheduling: The software shall provide for

ENERGY MANAGEMENT CONTROL SYSTEMS 345up to 5 (user configurable) multiple start and stopschedule(s) for each piece of machinery, for each dayof the week. In addition, it shall provide for twenty-six(26) special holiday schedules which can be configureda year in advance. The operator shall have the ability todesignate each holiday as either a single date or a rangeof dates. In addition, each of these 26 holidays shall beconfigurable as either a complete shutdown day whereall zones of control are maintained at setback levels, ora special holiday allowing for the partial shut down ofthe facility. Each piece of equipment shall have a uniqueschedule for each of these holidays and the schedulesmust be identical in format and execution to the normalday-of-week schedules.12.B10.2 Demand Shedding (if applicable): The operatorshall be allowed to select the maximum electricaldemand limit for each monitored electrical service in thefacility that the system will strive to protect. The operatormust have the ability to set the order of priority inwhich equipment is to be shed. The operator must alsohave the ability to identify any given control point as a“round-robin” shed point (first off-first on), a “priority”demand point (first off-last on) or a “temperature” demandpoint (load closest to setpoint shed first). Specialuser selectable demand parameters for each load shallbe shed type, shed priority, minimum shed time, maximumshed time, minimum time between shed. Thereshall be a programmable demand shedding schedulethat will determine the time of day during which demandshedding will be active.12.B10.3 Minimum on-minimum off times: The operatorshall have the ability to enter minimum on and minimumoff times for each piece of machinery controlled.This feature shall override all other control parametersto prevent equipment short cycling.12.B10.4 Straight Time Duty Cycling: The operator shallbe able to establish any given control point as a straighttime duty cycled point and to set the cycling parameters.For each cycled load the operator shall be able to designateany two minute segment of the hour as either acycled off or cycled on period. The program shall alwaysbegin the programmed cycle periods at the designatedoffset from the top of the hour. Cycling parameters shallbe able to be copied from point to point, or from onepoint to a group of points. The parameter copy featureshall automatically stagger the cycle period offset fromthe top of the hour. There shall be a programmable dutycycling schedule that will determine the time of dayduring which duty cycling will be active.12.B10.5 Temperature Compensated Duty Cycling: Theoperator shall be able to establish any given controlpoint as a temperature compensated duty cycled pointand to set the cycling parameters for both heating andcooling ramps (if applicable) with a temperature DeadBand. For each cycled load the operator shall be ableto designate high and low temperatures, long off andshort off times for both heating and cooling ramps,and a total cycle length. All time parameters shall beprogrammable in minutes and the software shall verifydata entry to prohibit conflicting time parameters. Thereshall be a programmable duty cycling schedule that willdetermine the time of day during which duty cyclingwill be active.12.B10.6 Optimum Start/Stop: For both heating andcooling seasons, the system shall provide customizedoptimum start/stop routines which take into accountoutside temperature and inside zone temperatureswhen preparing the building climate for occupation orshutting the facility down at the end of the day. Thesoftware shall track the rate of heat loss or gain in eachzone and utilize this data when activating optimumstart/stop routines.12.B10.7 Temperature Control: Building target temperaturesfor night time, weekend and holiday hours,as well as parameters and limits on normal occupiedoperation shall be user selectable. The system will striveto maintain these setpoint temperatures in considerationof other energy management functions such asdemand shedding, duty cycling, optimum start/stop,etc. Temperature setpoint parameters for on/off pointsshall be separate temperatures for heating and cooling(if applicable) with Dead Band. Temperature setpointsfor analog control points shall include minimum outputposition (in %), heating full, heating start, cooling startand cooling full.12.B10.8 Air Conditioning Optimization: Air conditioningshall be balanced and controlled considering thefollowing factors:a. Date (user selectable start and stop dates);b. Time of day;c. Temperature (outside and inside zone temperatures);d. Electrical load (staggered start ups & cycling);e. Short cycle protection (user selectable minimumon and minimum off times); and,f. Economizer control of equipment configured tosupport this type of operation.

346 ENERGY MANAGEMENT HANDBOOK12.B10.9 Freeze Protection: Any control point can be assignedto a special freeze protection menu for use in coldclimate areas. Points so assigned shall be programmedto run based on an outside temperature setpoint. Thisfreeze protection algorithm will supersede all other temperaturecontrol parameters.12.B10.10 Manual Override: Separate software manualoverrides shall be available for each controlled load thatallow the operator to turn on any load for a specificperiod of time.12.B10.11 Daylight Savings Time: Start and stop datesfor Daylight Savings Time shall be programmable andautomatically implemented by the system.12.B10.12 Alarms: The software shall register and recordalarms for the following conditions:a. Failure of controlled equipment to respondto computer commands (where positive feedback isused);b. High analog value (user selectable limits); and,c. Low analog value (user selectable limits).All alarms must be visible and history shall be maintainedfor a user selectable number of events with aminimum storage of 10 events. In addition, the systemmust have the capability to automatically dial out of thebuilding to alert user determined authorities in the eventof specific alarms. The operator shall have the ability toenable and disable automatic alarm printing. Alarm logsshall contain the date, time, location and type of alarmregistered. Also, the software must record the same datafor the time when the alarm cleared.12.B10.13 Monitoring: The software shall track anddisplay various types of information on monitored orcontrolled points including but not limited to:a. Last time onb. Last time offc. Temperatured. Pressuree. AlarmsThe software shall have the capability to track analogvalues on the basis of a user defined time period andstore the information for a user configurable number oftime periods with a minimum storage of 24 periods.12.B11.0 TRAINING12.B11.1 CONTRACTOR shall provide the services ofcompetent instructors who will give full instruction todesignated personnel in the operation, maintenance andprogramming of the system.12.B11.2 CONTRACTOR shall orient the training specificallyto the system installed rather than a generaltraining course. Instructors shall be thoroughly familiarwith the subject matter they are to teach.12.B11.3 CONTRACTOR shall provide a trainingmanual for each student which describes in detail thedata included on each training program. CONTRAC-TOR shall provide equipment and material required forclassroom training.12.B11.4 Training on the functional operation of thesystem shall include:a. Operation of equipment;b. Programming;c. Diagnostics;d. Failure recovery procedure;e. Alarm formats (where applicable);f. Maintenance and calibration; and,g. Trouble shooting, diagnostics and repair instructions.12.B12.0 SYSTEMS MANUAL12.B12.1 CONTRACTOR shall provide three (3) systemmanuals describing programming and testing, startingwith a system overview and proceeding to a detaileddescription of each software feature. The manual shallinstruct the user on programming or reprogrammingany portion of the system. This shall include all controlprograms, algorithms, mathematical equations,variables, setpoints, time periods, messages, and otherinformation necessary to load, alter, test and execute thesystem. The manual shall include:12.B12.2 Complete description of programming language,including commands, editing and writing controlprograms, algorithms, printouts and logs, mathematicalcalculations and passwords.12.B12.3 Instructions on modifying any control algorithmor parameter, verifying errors, status, changing passwordsand initiating or disabling control programs.

ENERGY MANAGEMENT CONTROL SYSTEMS 34712.B12.4 Complete point identification, including terminalnumber, symbol, engineering units and controlprogram reference number.12.B12.5 Field information including location, device,device type and function, electrical parameters and installationdrawing number.12.B12.6 Location identification of system control hardware.12.B12.7 For each system function, a listing of digitaland/or analog hardware required to interface the systemto the equipment.12.B12.8 Listing of all system application programs associatedwith each piece of equipment. This listing shallinclude all control algorithms and mathematical equations.The listing shall be in easy to understand Englishformat. All application programs must be submitted. Noproprietary control algorithms will be accepted.12.B13.0 MAINTENANCE MANUAL12.B13.1 CONTRACTOR shall provide three (3) maintenancemanuals containing descriptions of maintenanceon all system components, including sensors andcontrolled devices. The manual shall cover inspection,periodic preventive maintenance, fault diagnosis andrepair or replacement of defective components.12.B14.0 CONTROL DRAWINGS12.B14.1 CONTRACTOR shall submit the followingShop Drawings for the new energy management systemfor approval prior to start of the work.12.B14.2 Wiring diagram indicating general system layoutfor each system.12.B14.3 Control drawings including outline of systemsbeing controlled and location of devices. Include floorplans showing locations of EMS panels, wiring runs,junction boxes, EP’s, PE’s, etc.12.B15.0 CONTRACTOR’S VERIFICATIONOF SYSTEM INSTALLATION12.B15.1 CONTRACTOR shall provide to OWNER writtencertification in the form prescribed by OWNER’SREPRESENTATIVE that installation of the system issubstantially complete as to all respects of these specificationsprior to final checkout pursuant to section12.B17.0 hereof. This certification shall be submitted toOWNER’S REPRESENTATIVE at least ten (10) calendardays prior to the then current Contract completion date.As part of the certification, CONTRACTOR shall performas listed below.12.B15.2 CONTRACTOR shall verify correct temperaturereadings and scaling of all temperature inputs bycomparing actual space temperature to remote temperatureindication as displayed by the system videomonitor. Both actual and displayed temperatures shallbe recorded in table form and certified as to authenticityby CONTRACTOR. Actual space temperature shall bemeasured by a device traceable to the National Bureauof Standards. Said device will be noted in the certification.12.B15.3 CONTRACTOR shall verify operation of alloutputs by operating each output over its full rangeof operation and observing for correct response at thecontrolled equipment.12.B15.4 CONTRACTOR shall provide to OWNER a listof all existing nonoperational equipment that is interfacedwith the system.12.B15.5 CONTRACTOR shall perform an inspection ofthe completed work for compliance with all provisionsof these specifications and the Software Specifications.12.B16.0 CONTRACTOR’S VERIFICATIONOF SYSTEM PROGRAMMING12.B16.1 CONTRACTOR shall provide to OWNER writtencertification that all system programming is completeas to all respects of these specifications and the SoftwareSpecifications prior to final checkout pursuant to section12.B17.0 hereof. This certification shall be submitted toOWNER’S REPRESENTATIVE at least twenty (20) calendardays prior to the then current Contract completiondate. As part of the certification, CONTRACTOR shallperform as listed below.12.B16.2 CONTRACTOR shall provide to OWNER aprinted copy of all software programming, point designationsand schedules installed into the system.12.B16.3 CONTRACTOR shall demonstrate to OWN-ER’S REPRESENTATIVE all software programming asdetailed in the Software Specifications. The demonstrationshall include but not be limited to verification ofthe following:

348 ENERGY MANAGEMENT HANDBOOKa. Individual equipment control programs;b. Activity schedule/programs;c. Holiday schedule;d. Specific area activity overrides;e. Freeze protection;f. Global override;g. Data logging (trending, last time on, last timeoff); and,h. Software interlocks.12.B17.0 FINAL SYSTEMS CHECKOUT12.B17.1 CONTRACTOR shall arrange with OWNER’SREPRESENTATIVE a final systems checkout to be performedat least five (5) calendar days prior to the thencurrent Contract completion date.12.B17.2 CONTRACTOR shall provide all materials andequipment necessary for final systems checkout.12.B17.3 CONTRACTOR shall provide all personnelnecessary for final systems checkout. OWNER’S REPRE-SENTATIVE will act only as an observer for verificationand not actively participate in this checkout procedure.12.B17.4 CONTRACTOR shall demonstrate operation ofthe following system capabilities and components in thepresence of OWNER’S REPRESENTATIVE:a. Remote terminal access to the central systemcontroller and all of its peers via telephone modem;b. Remote uploading and downloading of allprograms and data points on the energy managementsystem (to include the central system controller and anypeers);c. Local terminal access to the central system controllerand any peers;d. Operation of each output point as per subparagraph12.B17.8 below; and,e. Listing of the status of all input points withverification of dynamic response to changing conditionsas per subparagraph 12.B17.7 below.12.B17.5 CONTRACTOR shall demonstrate properoperation of each input point to OWNER’S REPRE-SENTATIVE. Input points will be verified for accuracy,location, scaling and operation. Operation will be verifiedby including a dynamic change in the ambientcondition sensed by input sensor and tracking thesensors response and subsequent return to the staticcondition.12.B17.6 CONTRACTOR shall demonstrate properoperation of each output point to OWNER’S REPRE-SENTATIVE. Each output point will be verified in four(4) parts as follows:a. Building automation system to the interfacedevice (i.e. relay, EP, etc.); b. Orientation of interfacedevice (normally open, normally closed, etc.); c. Interfacedevice to the specified equipment; and, d. Actualcontrol/operation of specified equipment.12.B17.7 OWNER’S REPRESENTATIVE will conduct aninspection of the physical installation to include, but notbe limited to, the following:a. Mounting of the system controller and peers;b. Labeling of wires and components;c. Neat and orderly wiring and terminations;d. Mounting/location of modem;e. Mounting/location of auxiliary enclosures;f. Access to relays and equipment mounted inauxiliary enclosures;g. Mounting/location of temperature and pressuresensors; and,h. Completion of associated electrical work includingdisconnecting all time clocks associated with HVACequipment and disconnecting “Hand On” position on allassociated Hand On-Off-Auto switches.12.B18.0 AS BUILT DRAWINGS12.B18.1 CONTRACTOR shall provide to OWNER “asbuilt” drawings in conjunction with CONTRACTOR’Ssystems installation certification.12.B18.2 As built drawings shall include floor layoutdrawings including but not limited to:a. Location of all control devices;b. Location of all controlled equipment;c. Location of all system controllers and terminalcontrollers;d. Location of all wire runs and number of conductors;e. All spare wires;f. Location of all circuit and feeds including paneldesignations and circuit breaker numbers;g. Room numbers and or space designations;h. Location of all sensors; and,i. Wire labeling at terminal points.12.B18.3 As built drawings shall include all wiring and

ENERGY MANAGEMENT CONTROL SYSTEMS 349pneumatic schematics showing the systems interfacingfor each piece of controlled equipment.Appendix CTHE DDC DICTIONARYALGORITHM—An assortment of rules or steps presentedfor solving a problem.ANALOG INPUT—A variable input that is sensed, liketemperature, humidity or pressure and is sent to a computerfor processing.ANALOG OUTPUT—A variable signal such as voltage,current pneumatic air pressure which would in turnoperate a modulating motor which will drive valves,dampers, etc.ANALOG TO DIGITAL A/D CONVERTER—An integratedcircuit that takes data such as a temperaturesensor and converts it into digital logic that the microprocessorcan recognize. All analog inputs of a DDCsystem are fed into the microprocessor through an A/Dconverter.BAUD RATE—The speed, in bits per second, at whichinformation travels over a communications channel.BIT—The smallest representation of digital data there is.It has two states; 1 (on) or 0 (off).BITS OF RESOLUTION—A representation of how finelyinformation can be represented. Eight bits of resolutionmeans information may be divided into 256 differentstates, ten bits allows division into 1024 states andtwelve bits allows division into 4096 states.BYTE—Eight bits of digital information. Eight bits canrepresent up to 256 different states of information.CONTROL LOOP—The strategy that is used to make apiece of equipment operate properly. The loop receivesthe appropriate inputs and sets the desired condition.DAISY CHAIN—A wiring scheme where units mustfollow one another in a specific order.DEAD BAND—An area around a setpoint where thereis no change of state.DEAD TIME—A delay deliberately placed between tworelated actions in order to avoid an overlap in operationthat could cause equipment problems.DERIVATIVE CONTROL—A system which changes theoutput of a controller based on how fast a variable ismoving from or to a setpoint.DIGITAL INPUT—An input where there are only twopossibilities, such as on/off, open/closed.DIGITAL COMMUNICATION BUS—A set of wired,usually a twisted pair for DDC controllers. Informationis sent over this bus using a digital value to representthe value of the information.DIGITAL OUTPUT—An output that has two states, suchas on/off, open/closed.DIGITAL TO ANALOG D/A CONVERTER—An integratedcircuit that takes data from the microprocessorand converts it to analog data that is represented by avoltage or current. All analog outputs of a DDC systemgo through a D/A converter.DISTRIBUTED CONTROL—A system where all intelligenceis not in a single controller. If something fails,other controllers will take over to control a unit.EPROM—Erasable Programmable Read Only Memory.An integrated circuit that is known as firmware, wheresoftware instructions have been “burned” into it.EEPROM—Electrically Erasable Programmable ReadOnly Memory. An integrated circuit like the EPROMabove except that the information may be altered electronicallywith the chip installed in a circuit.FCU—Fan Coil Unit. A type of terminal unit.FEEDBACK—The signal or signals which are sent backfrom a controlled process telling the current status.FIRMWARE—Software that has been programmed intoan EPROM.FLOATING POINT—Numerical data that is displayedor manipulated with automatic positioning of the decimalpoint.HP—Heat Pump. A type of terminal unit.

350 ENERGY MANAGEMENT HANDBOOKINPUT—Data which is supplied to a computer or controlsystem for processing.INTEGRAL CONTROL—A system which changes theoutput of a controller based on how long a variable hasbeen offset from a setpoint. This type of control reducesthe setpoint/variable offset.INTELLIGENT BUILDING—A building that is controlledby a DDC system.LAG—The delay in response of a system to a controlledchange.LAN—Local Area Network. A communication linethrough which a computer can transmit and receiveinformation from other computers.MICROPROCESSOR—The brains of all DDC systemsand, for that matter, personal computers. An integratedcircuit which has logic and math functions built into it.If fed the proper instructions, it will perform the definedfunctions.MODEM—A device which allows the computer to communicateover phone lines. It allows the DDC systemto be viewed, operated and programmed through atelephone system.OFFSET—The difference between a variable and itssetpoint.OPERATOR’S TERMINAL—The computer at which theoperator can control a motor or to a relay to start/stopa unit.OUTPUT—Processed information that is sent by a controllerto an actuator to control a motor or to a relay tostart/stop a unit.PI CONTROL—A combination of proportional andintegral control. Adequate for almost all HVAC controlapplications. Control loops using PI control to look atboth how far an input is from setpoint and how long ithas been away from setpoint.PID CONTROL—Proportional plus Integral plus DerivativeControl. A system which directs control loopsto look at how far away an input is from setpoint, howlong it has been at setpoint and how fast it is approaching/movingaway from setpoint.PROPORTIONAL CONTROL—A system which linearlyvaries the output as an input variable changes relativeto setpoint. The farther the input is from setpoint, thelarger the change in the output.PULSE WIDTH MODULATION—A means of proportionallymodulating an actuator using digital outputs.One output opens the motor, another closes it. Usedextensively in VAV boxes for driving the damper motoron the box.RAM—Random Access Memory. A computer chip thatinformation can be written to and read from. Used tostore information needed in control loop calculations.SETPOINT—A value that has been assigned to a controlledvariable. An example would be a cooling setpointof 76 degrees F.SOFTWARE—The list of instructions written by an engineeror programmer that makes a controller or computeroperate as it does.TERMINAL UNIT—Part of the mechanical system thatserves an individual zone, such as VAV box, hydronicheat pump, fan coil, etc.TRANSDUCER—A unit that converts one type of signalto another. An electronic signal to a pneumatic signal.TWISTED PAIR—Two wires in one cable that aretwisted the entire length of the cable. DDC systems oftenuse a twisted pair for communication link betweencontrollers.TWO POSITION—Something that has only two states.A two position valve has two state s, either open orclosed.USER INTERFACE—The operator’s terminal is usuallyreferred to as the user interface. If allows the operatorto interrogate the system and perform desired functions.Most interfaces today are personal computers orminicomputers.VAV—Variable Air Volume. A terminal unit that variesthe amount of air delivered to a space, depending uponthe demand for cooling.WORD—Sixtee n bits of digital information. Sixteen bitscan represent up to 65,536 states of information.

ENERGY MANAGEMENT CONTROL SYSTEMS 351EMCS MANUFACTURERS DIRECTORY(The following list has not been updated as of thisprinting, but will be addressed in the next issue.)A.E.T. Systems, Inc., 77 Accord Executive Park Drive,Norwell, MA 02061617-871-4801Allerton Technologies, 2475 140th Ave. N.E., Bldg. A,Bellevue, WA 98005206-644-9500American Auto-Matrix, One Technology Drive, Export,PA 15632412-733-2000Andover Controls Corp., 300 Brickstone Square, Andover,MA 01810508-470-0555Automated Logic Corp., 1283 Kennestone Circle, Marietta,GA 30066404-423-7474Barber-Coleman Co., 1354 Clifford Ave., Rockford, IL61132815-877-0241Broadmoor Electric Co., 1947 Republic Ave., San Leandro,CA 94577415-483-2000Carrier Corp., One Carrier Place, Farmington, CT06034203-674-3000Chrontrol Corp., 9707 Candida Street, San Diego, CA92826619-566-5656CSI Control Systems International, Inc., 1625 W. CrosbyRoad, Carrollton, TX 75006214-323-1111D.K. Enterprises, Inc., 7361 Ethel Avenue, Suite 1, NorthHollywood, CA 91605818-764-0819Danfoss-EMC, Inc., 5650 Enterprise Pkwy., Ft. Myers,FL 33905813-693-5522Delta Controls, Inc., 13520 78th Avenue, Surrey, B.C.Canada V5W836604-590-8184Dencor, Inc., 1450 W. Evans Avenue, Denver, CO 80223303-922-1888EDA Controls Corp., 6645 Singletree Drive, Columbus,OH 43229614-431-0694Electronic Systems USA, Inc., 1014 East Broadway, Louisville,KY 40204502-589-1000Elemco Building Controls, 1324 Motor Parkway,Hauppauge, NY 11788516-582-8266Energy Control Systems, Inc., 2940 Cole Street, Norcross,GA 30071404-448-0651Functional Devices, Inc., 310 S. Union Street, Russiaville,IN 46979317-883-5538Grasslin Controls Corp., 45 Spear Road, Ramsey, NJ07446201-825-9696Honeywell, Inc., Comm. Bldg. Group, Honeywell Plaza,Minneapolis, MN 55408612-870-5200Honeywell, Inc., Res. & Bldg. Controls, 1985 DouglasDrive North, Minneapolis, MN 55422612-870-5200Integrated Energy Controls (Enercon Data), 7464 78th St.West, Minneapolis, MN 55435612-829-1900ISI Wireless, 3000-D South Highland Drive, Las Vegas,NV702-733-6500ISTA Energy Systems Corp., 407 Hope Ave., P.O. Box618, Roselle, NJ 07203908-241-8880Johnson Controls, Inc., 507 E. Michigan Street, P.O. Box423, Milwaukee, WI 53201414-274-4128Kreuter, Mfg., 5000 Peachtree Ind. Blvd., #125, Norcross,GA 30071404-662-5720Landis & Gyr Powers, 1000 Deerfield Parkway, BuffaloGrove, IL 60089708-215-1000Logic Power, 13500 Wright Circle, Tampa, FL 33626813-854-1588Microcontrol Systems, Inc., 6579 N. Sidney Pl., Milwaukee,WI 53209414-262-3143North American Technologies, Inc., 1300 N. FloridaMango Road, Ste 32, W. Palm Beach, FL 33409407-687-3051Novar Controls Corp., 24 Brown Street, Barberton, OH44203216-745-0074Paragon Electric Company, 606 Parkway Blvd., TwoRivers, WI 54241414-793-1161Phonetics, Inc., 901 Tryens Road, Aston, PA 19014215-558-2700

352 ENERGY MANAGEMENT HANDBOOKProcess Systems, Inc., 100-A Forsythe Hall Dr, Box240451, Charlotte, NC 28224704-588-4660Robertshaw Controls, 1800 Glenside Drive, Richmond,VA 23261804-289-4200Scientific Atlanta, 4300 Northeast Expressway, Atlanta,GA 30340404-449-2902Snyder General, Inc., 13600 Industrial Park Blvd., Minneapolis,MN 55440612-553-5330Solidyne Corp., 1202 Carnegie Street, Rolling Meadows,IL 60008619-427-9640Staefa Control Systems Inc., 8515 Miralani Drive, SanDiego, CA 92126619-530-1000Teletrol Systems, Inc., 324 Commercial Street, Manchester,NH 03101603-645-6061The Watt Stopper, Inc., 296 Brokaw Road, Santa Clara,CA 95050408-988-5331Trane Company, The, 20 Yorkton Court, St. Paul, MN55117612-490-3900Triangle Microsystems, Inc., 456 Bacon Street, Dayton,OH 45402513-223-0373Unity Systems, Inc., 2606 Spring Street, Redwood City,CA 94063415-369-3233Xencom Systems, 12015 Shiloh Road, Suite 155, Dallas,TX 75228214-991-1643Zeta Engineering Corp., 797 Industrial Court, BloomfieldHills, MI 48013313-332-2828References1. Cratty, William E., “Energy Conservation Design and ServicesWith Ventana,” APEC Journal, Spring 1990.2. Crosby, Philip B., Quality Is Free, Penguin Books, 1979.3. Elemco Building Controls, Inc., Elemco Energy Executive II ControlAlgorithms, 1989.4. Gustavson and Gustavson, The Business of Energy Management/DoingIt Right And Making Money At It, Gannam/Kubat Publishing,Inc. 1990.5. Honeywell Inc., Engineering Manual of Automatic Control For CommercialBuildings, 1988.6. Kirkpatrick, Bill, “Computers In EMS/Controls,” ContractingBusiness, September 1990.7. Kirkpatrick, Bill, “Overlay Controls,” Contracting Business, November1990.8. Mehta and Thumann, Handbook of Energy Engineering, FairmontPress, Inc., 1989.9. Patel, Jayant R., “Case Study: Energy Conservation/A ContinualProcess At Defense Contractor’s Manufacturing Facility,” EnergyEngineering, Vol. 88, No. 4, 1991.10. Rosenfeld, Shlomo I., “Worker Productivity: Hidden HVACCost,” Heating/Piping/Air Conditioning, September 1991.11. Rosenfeld, Shlomo I., “Worker Productivity: Hidden HVACCost,” Heating/Piping/Air Conditioning, September 1990.12. Rosenfeld, Shlomo I., “Worker Productivity: Hidden HVACCost,” Heating/Piping/Air Conditioning, September 1989.13. Silveria, Kent B., “The New EMCS Generation Has A Lot To Offer,”Consulting-Specifying Engineering, January 1992.14. Stern and Aronson., Energy Use/The Human Dimension, W.H. Freemanand Company, 1984.15. Thumann, Albert. Plant Engineer and Managers Guide To EnergyConservation, Fairmont Press, Inc. 1989.16. PECI, "Energy Management Systems… A Practical Guide," 1977.17. Star, Richard, "Pneumatic Controls In A Digital Age," Energy andEnvironmental Magazine, Fall 1999.18. Ponds, J. Michael, "NERTS… Non-Energy Relationed Tasks,"Energy and Environmental Magazine, Summer, 1995.19. Gustavson, Dale A., "FMCS: The Human Dimension," ContractingBusiness, September 1998.The authors wish to thank the following organizationsfor making this directory available:Electric Power Research Institute, 3412 HillviewAve., Palo Alto, CA, 94303Plexus Research, Inc., 289 Great Road, Acton,MA 01720

CHAPTER 13LIGHTINGERIC A. WOODROOF, PH.D., CEMJohnson Controls, Inc.JOHN FETTERS, CEM, CLEPEffective Lighting Solutions, Inc.13.1 INTRODUCTIONIn today’s cost-competitive, market-driven economy,everyone is seeking technologies or methods toreduce energy expenses and environmental impact. Becausenearly all buildings have lights, lighting retrofitsare very common and generally offer an attractive returnon investment. Electricity used to operate lighting systemsrepresents a significant portion of total electricityconsumed in the United States. Lighting systems consumeapproximately 20% of the electricity generated inthe United States. 1An attractive feature of lighting retrofits is theytypically provide savings for both kW and kWh charges.Thus, the potential for dollar savings is increased. Manylighting retrofits can also improve the visual environmentand worker productivity. Conversely, if a lighting retrofitreduces lighting quality, worker productivity may dropand the energy savings could be overshadowed by reducedprofits. This was the case with the lighting retrofitsof the 1970s, when employees were left “in the dark” dueto massive de-lamping initiatives. However, due to substantialadvances in technologies, today’s lighting retrofitscan reduce energy expenses while improving lightingquality and worker productivity.This chapter will provide the energy manager witha good understanding of lighting fundamentals, so thathe/she can oversee successful lighting upgrades. TheExample Section, (near the end of this chapter) containsanalyses of a few common lighting retrofits. The SchematicsSection contains illustrations of many lamps, ballastsand lighting systems.13.2.1 Lighting QuantityLighting quantity is the amount of light providedto a room. Unlike light quality, light quantity is easy tomeasure and describe.13.2.1.1 UnitsLighting quantity is primarily expressed in threetypes of units: watts, lumens and foot-candles (fc). Figure13.1 shows the relationship between each unit. The wattis the unit for measuring electrical power. It defines therate of energy consumption by an electrical device whenit is in operation. The amount of watts consumed representsthe electrical input to the lighting system.The output of a lamp is measured in lumens. Forexample, one standard four-foot fluorescent lamp wouldprovide 2,900 lumens in a standard office system. Theamount of lumens can also be used to describe the outputof an entire fixture (comprising several lamps). Thus, thenumber of lumens describes how much light is being producedby the lighting system.The number of foot-candles shows how much lightis actually reaching the workplane (or task). Foot-candlesare the end result of watts being converted to lumens, thelumens escaping the fixture and traveling through the airto reach the workplane. In an office, the workplane is thedesk level. You can measure the amount of foot-candleswith a light meter when it is placed on the work surfacewhere tasks are performed. Foot-candle measurementsare important because they express the “result” and notthe “effort” of a lighting system. The Illuminating EngineeringSociety (IES) recommends light levels for specifictasks using foot-candles, not lumens or watts.13.2 LIGHTING FUNDAMENTALSThis section will introduce the important conceptsabout lighting, and the two objectives of the lighting designer:(1) to provide the right quantity of light, and (2)provide the right quality of light.Figure 13.1 Units of measurement.353

354 ENERGY MANAGEMENT HANDBOOKEfficacySimilar to efficiency, efficacy describes an output/input ratio, the higher the output (while input is keptconstant), the greater the efficacy. Efficacy is the amountof lumens per watt from a particular energy source. Acommon misconception in lighting terminology is thatlamps with greater wattage provide more light. However,light sources with high efficacy can provide more lightwith the same amount of power (watts), when comparedto light sources with low efficacy.Figure 13.2 shows lampefficacies for various lamptypes, based on initial lumenvalues. These systems will bediscussed in greater detail inSection 13.2.3.13.2.1.2 IES RecommendedLight LevelsThe Illuminating EngineeringSociety (IES) is thelargest organized group oflighting professionals in theUnited States. Since 1915, IEShas prescribed the appropriatelight levels for many kindsof visual tasks. Although IESis highly respected, the appropriateamount of light fora given space can be subjective.For many years, lightingprofessionals applied thephilosophy that “more lightis better,” and light levelsrecommended by IES generallyincreased until the 1970s.However, recently it has beenshown that occupant comfortdecreases when a space hastoo much light. Numerousexperiments have confirmedthat the prescribed light levelswere excessive and workerproductivity was decreasingdue to poor visual comfort.Due to these findings, IESrevised their Handbook andreduced the recommendedlight levels for many tasks.The lighting designermust avoid over-illuminatinga space. Unfortunately, this objective can be difficult becauseover-illuminated spaces have become the “norm”in many buildings. Although not optimal, the traditionof excessive illumination can be continued simply due tohabit, or an organization’s reluctance to change. To correctthis trend, the first step for a lighting retrofit shouldbe to examine the existing system to determine if it isover-illuminated. Appropriate light levels for all types ofvisual tasks can be found in the most recent IES Hand-Figure 13.2 Lamp Efficacies for Various Lamp Types. (Source: Effective LightingSolutions, Inc.)————————————————————————————————————Lamp Lamp Watts EfficacyFamily Type (nominal) (LPW)————————————————————————————————————Incandescent Incandescent 3 – 1,500 4 – 23————————————————————————————————————Halogen 42 – 1,500 14 – 22————————————————————————————————————CFL Screw-base 9 – 85 40 – 65————————————————————————————————————Twin-tube Pre-heat (2-pin) 5 – 13 50 – 69————————————————————————————————————Quad Pre-heat (2-pin) 9 – 26 64 – 69————————————————————————————————————Rapid-start (4-pin) 5 – 57 46 – 87————————————————————————————————————Linear Fluorescent T2 6 – 13 52 – 66————————————————————————————————————T5 14 – 35 96 – 104————————————————————————————————————T5HO 24 – 80 83 – 94————————————————————————————————————T8 4’ Standard 32 53 – 88————————————————————————————————————T8 4’ Reduced-Wattage 30 94 - 95————————————————————————————————————T8 4’ High-Performance 32 94 – 98————————————————————————————————————T10 4’ 40 83————————————————————————————————————T12 Reduced-Wattage 34 49 – 85————————————————————————————————————T12 4’ Standard 40 53 – 83————————————————————————————————————T12 8’ Slimline Reduced-Wattage 60 64 – 97————————————————————————————————————T12 8’ Slimline 75 58 – 87————————————————————————————————————T12 8’ HO Reduced-Wattage 95 81 – 92————————————————————————————————————T12 8’ HO (110-w) 110 55 – 85————————————————————————————————————T12 8’ VHO Reduced-Wattage 195 67————————————————————————————————————T12 8’ VHO Fluorescent 215 65 – 70————————————————————————————————————Mercury Vapor Standard 50 – 1,000 32 – 58————————————————————————————————————Self-ballasted 160 – 750 14 - 19————————————————————————————————————Metal Halide Standard Probe-start 50 – 1,500 69 – 115————————————————————————————————————Pulse-start 50 – 1,000 69 – 110————————————————————————————————————Ceramic Arc-tube 20 – 400 83 – 95————————————————————————————————————High-PressureSodium Standard 35 – 1,000 64 – 150————————————————————————————————————Improved CRI 70 – 400 63 – 94————————————————————————————————————Low-PressureSodium 18 – 180 100 – 175————————————————————————————————————Induction QL 55 – 165 64 – 73————————————————————————————————————ICETRON 70 – 150 70 – 79————————————————————————————————————

LIGHTING 355book. Table 13.1 is a summary of some of the IES recommendations.It is important to remember that IES light levels correspondto particular visual tasks. In an office, there aremany tasks: walking around the office, viewing computerscreens, reading and writing on paper. Each task requiresa different light level. In the past, lighting designerswould identify the task that required the most light anddesign the lighting system to provide that level of illuminationfor the entire space. However, as previouslystated, these design methods often lead to environmentsTable 13.1 Recommended light levels for visual tasks.—————————————————————————Guideline IlluminanceRangeBuilding/Space Type(footcandles)—————————————————————————Commercial interiorsArt galleries 30-100Banks 50-150Hotels (rooms and lobbies) 10-50Offices 30-100-Average reading and writing 50-75-Hallways 10-20-Rooms with computers 20-50Restaurants (dining areas) 20-50Stores (general) 20-50Merchandise 100-200Institutional interiorsAuditoriums/assembly places 15-30Hospitals (general areas) 10-15Labs/treatment areas 50-100Libraries 30-100Schools 30-150Industrial interiorsOrdinary tasks 50Stockroom storage 30Loading and unloading 20Difficult tasks 100Highly difficult tasks 200Very difficult tasks 300-500Most difficult tasks 500-1000ExteriorBuilding security 1-5Floodlighting(low/high brightness or surroundings) 5-30Parking 1-5—————————————————————————with excessive brightness, glare and poor worker productivity.If the IES tables are applied for each task (“TaskLighting”), a superior lighting system is constructed. Forexample, in an office with computers there should be upto 30 fc for ambient lighting. Small task lights on deskscould provide the additional foot-candles needed toachieve a total illuminance of 50 to 75 fc for reading andwriting. Task lighting techniques are discussed in greaterdetail in Section 13.3.13.2.2 Lighting QualityLighting quality can have a dramatic influenceon the attitude and performance of occupants. In fact,different “moods” can be created by a lighting system.Consider the behavior of people when they eat in differentrestaurants. If the restaurant is a fast-food restaurant,the space is usually illuminated by bright white lights,with a significant amount of glare from shiny tables.Occupants rarely spend much time there partly becausethe space creates an uncomfortable mood and the atmosphereis “fast” (eat and leave). In contrast, consider anelegant restaurant with a candle-lit tables and a “warm”atmosphere. Occupants tend to relax and take more timeto eat. Although occupant behavior is also linked to interiordesign and other factors, lighting quality representsa significant influence. Occupants perceive and react toa space’s light color. It is important that the lighting designerbe able to recognize and create the subtle aspectsof an environment that define the theme of the space. Forexample, drug and grocery stores use white lights to createa “cool” and “clean” environment. Imagine if thesespaces were illuminated by the same color lights as inan elegant restaurant. How would the perception of thestore change?Occupants can be influenced to work more effectivelyif they are in an environment that promotesa “work-like” atmosphere. The goal of the lighting designeris to provide the appropriate quality of light for aparticular task to create the right “mood” for the space.Employee comfort and performance are worth morethan energy savings. Although the cost of energy forlighting ($.50-$1.00/year/ft 2 ) is substantial, it is relativelysmall compared to the cost of labor ($100-$300/year/ft 2 ).Improvements in lighting quality can yield high dividendsfor businesses because gains in worker productivityare common when lighting quality is improved.Conversely, if a lighting retrofit reduces lighting quality,occupant performance may decrease, quickly off-settingany savings in energy costs. Good energy managersshould remember that buildings were not designed tosave energy, they exist to create an environment where

356 ENERGY MANAGEMENT HANDBOOKpeople can work efficiently. Occupants should be able tosee clearly without being distracted by glare, excessiveshadows or other uncomfortable features.Lighting quality can be divided into four main considerations:Uniformity, Glare, Color Rendering Indexand Coordinated Color Temperature.13.2.2.1 UniformityThe uniformity of illuminance describes how evenlylight spreads over an area. Creating uniform illuminationrequires proper fixture spacing. Non-uniform illuminancecreates bright and dark spots, which can cause discomfortfor some occupants.Lighting designers have traditionally specifieduniform illumination. This option is least risky becauseit minimizes the problems associated with non-uniformillumination and provides excellent flexibility for changesin the work environment. Unfortunately, uniform lightingapplied over large areas can waste large amounts ofenergy. For example, in a manufacturing building, 20%of the floor space may require high levels of illumination(100 fc) for a specific visual task. The remaining 80% ofthe building may only require 40 foot candles. Uniformillumination over the entire space would require 100 fcat any point in the building. Clearly, this is a tremendouswaste of energy and money. Although uniform illuminationis not needed throughout the entire facility, uniformillumination should be applied on specific tasks. For example,a person assembling small parts on a table shouldhave uniform illumination across the table top.13.2.2.2 GlareGlare is a sensation caused by relatively bright objectsin an occupant’s field of view. The key word is relative,because glare is most probable when bright objectsare located in front of dark environments. For example,a car’s high beam headlights cause glare to oncomingdrivers at night, yet create little discomfort during theday. Contrast is the relationship between the brightnessof an object and its background. Although most visualtasks generally become easier with increased contrast,too much brightness causes glare and makes the visualtask more difficult. Glare in certain work environments isa serious concern because it usually will cause discomfortand reduce worker productivity.Visual Comfort Probability (VCP)The Visual Comfort Probability is a rating given toa fixture which indicates the percent of people who arecomfortable with the glare. Thus, a fixture with a VCP =80 means that 80% of occupants are comfortable with theamount of glare from that fixture. A minimum VCP of 70is recommended for general interior spaces. Fixtures withVCPs exceeding 80 are recommended in computer areasand high-profile executive office environments.To improve a lighting system that has excessiveglare, a lighting designer should be consulted. Howeverthere are some basic “rules of thumb” which can assist theenergy manager. A high-glare environment is characterizedby either excessive illumination and reflection, or theexistence of very bright areas typically around fixtures.To minimize glare, the energy manager can try to obscurethe bare lamp from the occupant’s field of view, relocatefixtures or replace the fixtures with ones that have a highVCP.Reducing glare is commonly achieved by using indirectlighting, using deep cell parabolic troffers, or speciallenses. Although these measures will reduce glare,fixture efficiency will be decreased because more lightwill be “trapped” in the fixture. Alternatively, glare canbe minimized by reducing ambient light levels and usingtask lighting techniques.Visual Display Terminals (VDTs)Today’s office environment contains a variety ofspecial visual tasks, including the use of computer monitorsor visual display terminals (VDTs). Occupants usingVDTs are extremely vulnerable to glare and discomfort.When reflections of ceiling lights are visible on the VDTscreen, the occupant has difficulty reading the screen.This phenomena is also called “discomfort glare,” andis very common in rooms that are uniformly illuminatedby fixtures with low a VCP. Therefore, lighting for VDTenvironments must be carefully designed, so that occupantsremain comfortable. Because the location VDTscan be frequently changed, lighting upgrades should alsobe designed to be adjustable. Moveable task lights andfixtures with high VCP are very popular for these types ofapplications. Because each VDT environment is unique,each upgrade must be evaluated on a case-by-case basis.13.2.2.3 ColorColor considerations have an incredible influenceon lighting quality. Light sources are specified based ontwo color-related parameters: the Color Rendering Index(CRI) and the Coordinated Color Temperature (CCT).Color Rendering Index (CRI)In simple terms, the CRI provides an evaluation ofhow colors appear under a given light source. The indexrange is from 0 to 100. The higher the number, the easierto distinguish colors. Generally, sources with a CRI > 75provide excellent color rendition. Sources with a CRI < 55provide poor color rendition. To provide a “base-case,”

LIGHTING 357offices illuminated by most T12 Cool White lamps have aCRI = 62.It is extremely important that a light source with ahigh CRI be used with visual tasks that require the occupantto distinguish colors. For example, a room with acolor printing press requires illumination with excellentcolor rendition. In comparison, outdoor security lightingfor a building may not need to have a high CRI, but alarge quantity of light is desired.Coordinated Color Temperature (CCT)The Coordinated Color Temperature (CCT) describesthe color of the light source. For example, on aclear day, the sun appears yellow. On an over-cast day,the partially obscured sun appears to be gray. These colordifferences are indicated by a temperature scale. The CCT(measured in degrees Kelvin) is a close representation ofthe color that an object (black-body) would radiate at acertain temperature. For example, imagine a wire beingheated. First it turns red (CCT = 2000K). As it gets hotter,it turns white (CCT = 5000K) and then blue (CCT =8000K). Although a wire is different from a light source,the principle is similar.CCT is not related to CRI, but it can influence the atmosphereof a room. Laboratories, hospitals and grocerystores generally use “cool” (blue-white) sources, whileexpensive restaurants may seek a “warm” (yellow-red)source to produce a candle-lit appearance. Traditionally,office environments have been illuminated by Cool Whitelamps, which have a CCT = 4100K. However, a more recenttrend has been to specify 3500K tri-phosphor lamps,which are considered neutral. Table 13.2 illustrates somecommon specifications for different visual environments.13.2.3 Lighting System ComponentsAfter determining the quantity and quality of illuminationrequired for a particular task, most lightingdesigners specify the lamp, then the ballast, and finallythe fixture to meet the lighting needs. The Schematics Section(near the end of this chapter) contains illustrations ofmany of the lamps and systems described in this section.13.2.3.1 LampsThe lamp is the first component to consider in thelighting design process. The lamp choice determines thelight quantity, CRI, CCT, relamping time interval andoperational costs of the lighting system. This section willonly cover the most popular types of lamps. Table 13.3summarizes the differences between the primary lampsand lighting systems.Table 13.2 Sample design considerations for a commercial building.——————————————————————————————————————Light CRI Color GlareOffice Areas Levels Temperature——————————————————————————————————————Executive General 100FC ≥80 3000K VCP≥70Task ≥50FC ≥80 3000K VCP≥70——————————————————————————————————————Private General 30-50FC ≥70 3000-3500K VCP≥70Task ≥50FC ≥70 3000-3500K VCP≥70——————————————————————————————————————Open Plan General 30-50FC ≥70 3000-3500K VCP≥90Computers Task ≥50FC ≥70 3000-3500K VCP≥90——————————————————————————————————————Hallways General 10-20FC ≥70 3000-3500K VCP≥70Task 10-20FC ≥70 3000-3500K VCP≥70——————————————————————————————————————Reception/ General 20-50FC ≥80 3000-5000K VCP≥90Lobby Task ≥50FC ≥80 3000-3500K VCP≥90——————————————————————————————————————Conference General 10-70FC ≥80 3000-4100K VCP≥90Task 10-70FC ≥80 3000-4100K VCP≥90——————————————————————————————————————Open Plan General 30-70FC ≥70 3500-4100K VCP≥80General Task 30-70FC ≥70 3500-4100K VCP≥80——————————————————————————————————————Drafting General 70-100FC ≥70 4100-5000K VCP≥90Task 100-150FC ≥780 4100-5000K VCP≥90——————————————————————————————————————

358 ENERGY MANAGEMENT HANDBOOKTable 13.3 Lamp characteristics———————————————————————————————————————————————————IncandescentHigh-PressureIncluding Tungsten Compact Mercury Vapor Sodium Low-PressureHalogen Fluorescent Fluorescent (Self-ballasted) Metal Halide (Improved Color) SodiumWattages (lamp only) 15-1500 15-219 4-40 40-1000 175-1000 70-1000 35-180———————————————————————————————————————————————————Life (hr) 750-12,000 7,500-24,000 10,000-20,000 16,000-15,000 1,500-15,000 24,000 (10,000) 18,000Efficacy 15-25 55-100 50-80 50-60 80-100 75-140 Up to 180(lumens/W) lamp only (20-25) (67-112)Lumen maintenance Fair to excellent Fair to excellent Fair Very good Good Excellent Excellent(good)Color rendition Excellent Good to excellent Good to Poor to excellent Very good Fair PoorexcellentLight direction control Very good to Fair Fair Very good Very good Very good FairexcellentRelight time Immediate Immediate Imm- 3 seconds 3-10 min. 10-20 min. Less than 1 min. ImmediateComparative fixture cost Low: simple Moderate Moderate Higher than Generally High Highfluorescent higher thanmercuryComparative operating High Lower than Lower than Lower than Lower than Lowest of HID Lowcost incandescent incandescent incandescent mercury types———————————————————————————————————————————————————IncandescentThe oldest electric lighting technology is the incandescentlamp. Incandescent lamps are also the leastefficacious (have the lowest lumens per watt) and havethe shortest life. They produce light by passing a currentthrough a tungsten filament, causing it to become hot andglow. As the tungsten emits light, it gradually evaporates,eventually causing the filament to break. When this happens,the lamps is said to be “burned-out.”Although incandescent sources are the least efficacious,they are still sold in great quantities because ofeconomies of scale and market barriers. Consumers stillpurchase incandescent bulbs because they have low initialcosts. However, if life-cycle cost analyses are used, incandescentlamps are usually more expensive than otherlighting systems with higher efficacies.Compact Fluorescent Lamps (CFLs)Overview of CFLs:Compact Fluorescent Lamps (CFLs) are energy efficient,long lasting replacements for some incandescent lamps.CFLs (like all fluorescent lamps) are composed of twoparts, the lamp and the ballast. The short tubular lampscan last longer than 8,000 hours. The ballasts (plasticcomponent at the base of tube) usually last longer than60,000 hours. Some CFLs can be purchased as self-ballastedunits, which “screw in” to an existing incandescentsocket. For simplicity, this chapter refers to a CFL as alamp and ballast system. CFLs are available in manystyles and sizes.In most applications, CFLs are excellent replacementsfor incandescent lamps. CFLs provide similar lightquantity and quality while only requiring about 20-30%of the energy of comparable incandescent lamps. In addition,CFLs last 7 to 10 times longer than their incandescentcounterparts. In many cases, it is cost-effectiveto replace an entire incandescent fixture with a fixturespecially designed for CFLs.The “New Technololgies” Section contains a more thoroughexplanation of CFLs.FluorescentFluorescent lamps are the most common lightsource for commercial interiors in the U.S. They are repeatedlyspecified because they are relatively efficient,have long lamp lives and are available in a wide varietyof styles. For many years, the conventional fluorescentlamp used in offices has been the four-foot F40T12 lamp,which is usually used with a magnetic ballast. However,these lamps are being rapidly replaced by T8 or T5 lampswith electronic ballasts.The labeling system used by manufacturers may appearcomplex, however it is actually quite simple. For example,with an F34T12 lamp, the “F” stands for fluorescent,the “34” means 34 watts, and the “T12” refers to the tubethickness. Since tube thickness (diameter) is measured in1/8 inch increments, a T12 is 12/8 or 1.5 inches in diameter.A T8 lamp is 1 inch in diameter. Some lamp labels includeadditional information, indicating the CRI and CCT. Usu-

LIGHTING 359ally, CRI is indicated with one digit, like “8” meaning CRI= 80. CCT is indicated by the two digits following, “35”meaning 3500K. For example, a F32T8/841 label indicatesa lamp with a CRI = 80 and a CCT = 4100K. Alternatively,the lamp manufacturer might label a lamp with a lettercode referring to a specific lamp color. For example, “CW”to mean Cool White lamps with a CCT = 4100K.Some lamps have “ES,” “EE” or “EW” printed onthe label. These acronyms attached at the end of a lamplabel indicate that the lamp is an energy-saving type.These lamps consume less energy than standard lamps,however they also produce less light.Tri-phosphor lamps have a coating on the insideof the lamp which improves performance. Tri-phosphorlamps usually provide greater color rendition. A bi-phosphorlamp (T12 Cool White) has a CRI= 62. By upgradingto a tri-phosphor lamp with a CRI = 75, occupants willbe able to distinguish colors better. Tri-phosphor lampsare commonly specified with systems using electronicballasts. Lamp flicker and ballast humming are also significantlyreduced with electronically ballasted systems.For these reasons, the visual environment and workerproductivity is likely to be improved.There are many options to consider when choosingfluorescent lamps. Carefully check the manufacturersspecifications and be sure to match the lamp and ballastto the application. Table 13.4 shows some of the specificationsthat vary between different lamp types.The “New Technololgies” Section contains a more thoroughexplanation of the various fluorescent lamp systemsavailable today.High Intensity Discharge (HID)High-Intensity Discharge (HID) lamps are similarto fluorescent lamps because they produce light by dischargingan electric arc through a tube filled with gases.HID lamps generate much more light, heat and pressurewithin the arc tube than fluorescent lamps, hence thetitle “high intensity” discharge. Like incandescent lamps,HIDs are physically small light sources, (point sources)which means that reflectors, refractors and light pipes canbe effectively used to direct the light. Although originallydeveloped for outdoor and industrial applications, HIDsare also used in office, retail and other indoor applications.With a few exceptions, HIDs require time to warmup and should not be turned ON and OFF for short intervals.They are not ideal for certain applications because,as point sources of light, they tend to produce more definedshadows than non-point sources such as fluorescenttubes, which emit diffuse light.Most HIDs have relatively high efficacies and longlamp lives, (5,000 to 24,000+ hours) reducing maintenancere-lamping costs. In addition to reducing maintenance requirements,HIDs have many unique benefits. There arethree popular types of HID sources (listed in order of increasingefficacy): Mercury Vapor, Metal Halide and HighPressure Sodium. A fourth source, Low Pressure Sodium,is not technically a HID, but provides similar quantitiesof illumination and will be referred to as an HID in thischapter. Table 13.3 shows that there are dramatic differencesin efficacy, CRI and CCT between each HID sourcetype.Mercury VaporMercury Vapor systems were the “first generation”HIDs. Today they are relatively inefficient, provide poorCRI and have the most rapid lumen depreciation rateof all HIDs. Because of these characteristics, other morecost-effective HID sources have replaced mercury vaporTable 13.4 Sample fluorescent lamp specifications.————————————————————————————————MANUFACTURERS’ INFORMATIONF40T12CW F40T10 F32T8Bi-phosphor Tri-phosphor Tri-phosphor————————————————————————————————CRI 62 83 83CCT (K) 4,150 4,100 or 5,000 4,100 or 5,000Initial lumens 3,150 3,700 3,050Maintained lumens 2,205 2,960 2,287Lumens per watt 55 74 71Rated life (hrs) 24,000 48,000 † 20,000Service life (hrs) 16,800 33,600 † 14,000————————† This extended life is available from a specific lamp-ballast combination. NormalT10 lamp lives are approximately 24,000 hours. Service life refers to the typicallamp replacement life.————————————————————————————————

360 ENERGY MANAGEMENT HANDBOOKlamps in nearly all applications. Mercury Vapor lampsprovide a white-colored light which turns slightly greenover time. A popular lighting upgrade is to replace MercuryVapor systems with Metal Halide or High PressureSodium systems.Metal HalideMetal Halide lamps are similar to mercury vaporlamps, but contain slightly different metals in the arctube, providing more lumens per watt with improvedcolor rendition and improved lumen maintenance. Withnearly twice the efficacy of Mercury Vapor lamps, MetalHalide lamps provide a white light and are commonlyused in industrial facilities, sports arenas and other spaceswhere good color rendition is required. They are thecurrent best choice for lighting large areas that need goodcolor rendition.High Pressure Sodium (HPS)With a higher efficacy than Metal Halide lamps,HPS systems are an economical choice for most outdoorand some industrial applications where good colorrendition is not required. HPS is common in parkinglots and produces a light golden color that allows somecolor rendition. Although HPS lamps do not provide thebest color rendition, (or attractiveness) as “white light”sources, they are adequate for indoor applications atsome industrial facilities. The key is to apply HPS in anarea where there are no other light source types availablefor comparison. Because occupants usually prefer “whitelight,” HPS installations can result with some occupantcomplaints. However, when HPS is installed at a greatdistance from metal halide lamps or fluorescent systems,the occupant will have no reference “white light” andhe/she will accept the HPS as “normal.” This techniquehas allowed HPS to be installed in countless indoor gymnasiumsand industrial spaces with minimal complaints.Low Pressure SodiumAlthough LPS systems have the highest efficacy ofany commercially available HID, this monochromic lightsource produces the poorest color rendition of all lamptypes. With a low CCT, the lamp appears to be “pumpkinorange,” and all objects illuminated by its light appearblack and white or shades of gray. Applications are limitedto security or street lighting. The lamps are physicallylong (up to 3 feet) and not considered to be point sources.Thus optical control is poor, making LPS less effective forextremely high mounting heights.LPS has become popular because of its extremelyhigh efficacy. With up to 60% greater efficacy than HPS,LPS is economically attractive. Several cities, such as SanDiego, California, have installed LPS systems on streets.Although there are many successful applications, LPSinstallations must be carefully considered. Often lightingquality can be improved by supplementing the LPS systemwith other light sources (with a greater CRI).13.2.3.2 BallastsWith the exception of incandescent systems, nearlyall lighting systems (fluorescent and HID) require a ballast.A ballast controls the voltage and current that issupplied to lamps. Because ballasts are an integral componentof the lighting system, they have a direct impacton light output. The ballast factor is the ratio of a lamp’slight output to a reference ballast. General purpose fluorescentballasts have a ballast factor that is less than one(typically .88 for most electronic ballasts). Special ballastsmay have higher ballast factors to increase light output,or lower ballast factors to reduce light output. As can beexpected, a ballast with a high ballast factor also consumesmore energy than a general purpose ballast.FluorescentSpecifying the proper ballast for fluorescent lightingsystems has become more complicated than it was 25years ago, when magnetic ballasts were practically theonly option. Electronic ballasts for fluorescent lamps havebeen available since the early 1980s, and their introductionhas resulted in a variety of options.This section describes the two types of fluorescentballasts: magnetic and electronic.MagneticMagnetic ballasts are available in three primarytypes.• Standard core and coil• High-efficiency core and coil (Energy-Efficient Ballasts)• Cathode cut-out or HybridStandard core and coil magnetic ballasts are essentiallycore and coil transformers that are relativelyinefficient at operating fluorescent lamps. Although thesetypes of ballasts are no longer sold in the US, they stillexist in many facilities. The “high-efficiency” magneticballast can replace the “standard ballast,” improving thesystem efficiency by approximately 10%.“Cathode cut-out” or “hybrid” ballasts are high-efficiencycore and coil ballasts that incorporate electroniccomponents that cut off power to the lamp cathodes afterthe lamps are operating, resulting in an additional 2-wattsavings per lamp.

LIGHTING 361ElectronicDuring the infancy of electronic ballast technology,reliability and harmonic distortion problems hamperedtheir success. However, most electronic ballasts availabletoday have a failure rate of less than one percent, andmany distort harmonic current less than their magneticcounterparts. Electronic ballasts are superior to magneticballasts because they are typically 30% more energy efficient,they produce less lamp flicker, ballast noise, andwaste heat.In nearly every fluorescent lighting application,electronic ballasts can be used in place of conventionalmagnetic core and coil ballasts. Electronic ballasts improvefluorescent system efficacy by converting thestandard 60 Hz input frequency to a higher frequency,usually 25,000 to 40,000 Hz. Lamps operating on thesefrequencies produce about the same amount of light,while consuming up to 40% less power than a standardmagnetic ballast. Other advantages of electronic ballastsinclude less audible noise, less weight, virtually no lampflicker and dimming capabilities.T12 and T8 ballasts are the most popular types ofelectronic ballasts. T12 electronic ballasts are designed foruse with conventional (T12) fluorescent lighting systems.T8 ballasts offer some distinct advantages over othertypes of electronic ballasts. They are generally more efficient,have less lumen depreciation, and are availablewith more options. T8 ballasts can operate one, two, threeor four lamps. Most T12 ballasts can only operate one,two or three lamps. Therefore, one T8 ballast can replacetwo T12 ballasts in a 4 lamp fixture.Some electronic ballasts are parallel-wired, so thatwhen one lamp burns out, the remaining lamps in thefixture will continue to operate. In a typical magnetic, (series-wiredsystem) when one component fails, all lampsin the fixture shut OFF. Before maintenance personnel canrelamp, they must first diagnose which lamp failed. Thusthe electronically ballasted system will reduce time todiagnose problems, because maintenance personnel canimmediately see which lamp failed.Parallel-wired ballasts also offer the option of reducinglamps per fixture (after the retrofit) if an area isover-illuminated. This option allows the energy managerto experiment with different configurations of lamps indifferent areas. However, each ballast operates best whencontrolling the specified number of lamps.Due to the advantages of electronically ballastedsystems, they are produced by many manufacturers andprices are very competitive. Due to their market penetration,T8 systems (and replacement parts) are more likelyto be available, and at lower costs.HIDAs with fluorescent systems, High Intensity Dischargelamps also require ballasts to operate. Althoughthere are not nearly as many specification options as withfluorescent ballasts, HID ballasts are available in dimmableand bi-level light outputs. Instant restrike systemsare also available.Capacitive Switching HID FixturesCapacitive switching or “bi-level” HID fixtures aredesigned to provide either full or partial light outputbased on inputs from occupancy sensors, manual switchesor scheduling systems. Capacitive-switched dimmingcan be installed as a retrofit to existing fixtures or as adirect fixture replacement. Capacitive switching HID upgradescan be less expensive than installing a panel-levelvariable voltage control to dim the lights, especially incircuits with relatively few fixtures.The most common applications of capacitive switchingare athletic facilities, occupancy-sensed dimming inparking lots and warehouse aisles. General purpose transmitterscan be used with other control devices such as timersand photosensors to control the bi-level fixtures. Upondetecting motion, the occupancy sensor sends a signal tothe bi-level HID ballasts. The system will rapidly bring thelight levels from a standby reduced level to about 80 percentof full output, followed by the normal warm-up timebetween 80 and 100 percent of full light output.Depending of the lamp type and wattage, thestandby lumens are roughly 15-40 percent of full outputand the standby wattage is 30-60 percent of full wattage.When the space is unoccupied and the system is dimmed,you can achieve energy savings of 40-70 percent.13.2.3.3 Fixtures (aka Luminaires)A fixture is a unit consisting of the lamps, ballasts,reflectors, lenses or louvers and housing. The mainfunction is to focus or spread light emanating from thelamp(s). Without fixtures, lighting systems would appearvery bright and cause glare.Fixture EfficiencyFixtures block or reflect some of the light exiting thelamp. The efficiency of a fixture is the percentage of lamplumens produced that actually exit the fixture in the intendeddirection. Efficiency varies greatly among differentfixture and lamp configurations. For example, usingfour T8 lamps in a fixture will be more efficient than usingfour T12 lamps because the T8 lamps are thinner, allowingmore light to “escape” between the lamps and out ofthe fixture. Understanding fixtures is important because alighting retrofit may involve changing some components

362 ENERGY MANAGEMENT HANDBOOKof the fixture to improve the efficiency and deliver morelight to the task.The Coefficient of Utilization (CU) is the percent oflumens produced that actually reach the work plane. TheCU incorporates the fixture efficiency, mounting height,and reflectances of walls and ceilings. Therefore, improvingthe fixture efficiency will improve the CU.ReflectorsInstalling reflectors in most fixtures can improve itsefficiency because light leaving the lamp is more likelyto “reflect” off interior walls and exit the fixture. Becauselamps block some of the light reflecting off the fixture interior,reflectors perform better when there are less lamps(or smaller lamps) in the fixture. Due to this fact, a commonfixture upgrade is to install reflectors and removesome of the lamps in a fixture. Although the fixture efficiencyis improved, the overall light output from each fixtureis likely to be reduced, which will result in reducedlight levels. In addition, reflectors will redistribute light(usually more light is reflected down), which may createbright and dark spots in the room. Altered light levelsand different distributions may be acceptable, howeverthese changes need to be considered.To ensure acceptable performance from reflectors,conduct a trial installation and measure “before” and“after” light levels at various locations in the room. Don’tcompare an existing system, (which is dirty, old and containsold lamps) against a new fixture with half the lampsand a clean reflector. The light levels may appear to beadequate, or even improved. However, as the new systemages and dirt accumulates on the surfaces, the light levelswill drop.A variety of reflector materials are available: highlyreflective white paint, silver film laminate, and anodizedaluminum. Silver film laminate usually has the highestreflectance, but is considered less durable. Be sure toevaluate the economic benefits of your options to get themost “bang for your buck.”In addition to installing reflectors within fixtures,light levels can be increased by improving the reflectivityof the room’s walls, floors and ceilings. For example, bycovering a brown wall with white paint, more light willbe reflected back into the workspace, and the Coefficientof Utilization is increased.Lenses and LouversMost indoor fixtures use either a lens or louver toprevent occupants from directly seeing the lamps. Lightthat is emitted in the shielding angle or “glare zone”(angles above 45 o from the fixture’s vertical axis) cancause glare and visual discomfort, which hinders theoccupant’s ability to view work surfaces and computerscreens. Lenses and louvers are designed to shield theviewer from these uncomfortable, direct beams of light.Lenses and louvers are usually included as part of a fixturewhen purchased, and they can have a tremendousimpact on the VCP of a fixture.Lenses are sheets of hard plastic (either clear ormilky white) that are located on the bottom of a fixture.Clear, prismatic lenses are very efficient because they trapless light within the fixture. Milky-white lenses are called“diffusers” and are the least efficient, trapping a lot ofthe light within the fixture. Although diffusers have beenroutinely specified for many office environments, theyhave one of the lowest VCP ratings.Louvers provide superior glare control and highVCP when compared to most lenses. As Figure 13.3shows, a louver is a grid of plastic “shields” which blockssome of the horizontal light exiting the fixture. The mostcommon application of louvers is to reduce the fixtureglare in sensitive work environments, such as in roomswith computers. Parabolic louvers usually improve theVCP of a fixture, however efficiency is reduced becausemore light is blocked by the louver. Generally, the smallerthe cell, the greater the VCP and less the efficiency. Deepcellparabolic louvers offer a better combination of VCPand efficiency, however deep-cell louvers require deepfixtures, which may not fit into the ceiling plenum space.Table 13.5 shows the efficiency and VCP for variouslenses and louvers. VCP is usually inversely related tofixture efficiency. An exception is with the milky-whitediffusers, which have low VCP and low efficiency.Light Distribution/Mounting HeightFixtures are designed to direct light where it is needed.Various light distributions are possible to best suit anyvisual environment. With “direct lighting,” 90-100% ofthe light is directed downward for maximum use. With“indirect lighting,” 90-100% of the light is directed tothe ceilings and upper walls. A “semi-indirect” systemFigure 13.3 Higher shielding angles for improved glarecontrol.

LIGHTING 363Table 13.5 Luminaire efficiency and VCP.—————————————————————————Shielding Material Luminaire Visual ComfortEfficiency Probability(%) (VCP)—————————————————————————Clear Prismatic Lens 60-70 50-70Low Glare Clear Lens 60-75 75-85Deep-Cell Parabolic Louver 50-70 75-95Translucent Diffuser 40-60 40-50Small-Cell Parabolic Louver 35-45 99—————————————————————————distributes 60-90% down, with the remainder upward.Designing the lighting system should incorporate the differentlight distributions of different fixtures to maximizecomfort and visual quality.Fixture mounting height and light distribution arepresented together since they are interactive. HID systemsare preferred for high mounting heights since the lampsare physically small, and reflectors can direct light downwardwith a high degree of control. Fluorescent lamps arephysically long and diffuse sources, with less ability tocontrol light at high mounting heights. Thus fluorescentsystems are better for low mounting heights and/or areasthat require diffuse light with minimal shadows.Generally, “high-bay” HID fixtures are designed formounting heights greater than 20 feet high. “High-bay”fixtures usually have reflectors and focus most of theirlight downward. “Low-bay” fixtures are designed formounting heights less than 20 feet and use lenses to directmore light horizontally.HID sources are potential sources of direct glaresince they produce large quantities of light from physicallysmall lamps. The probability of excessive direct glaremay be minimized by mounting fixtures at sufficientheights. Table 13.6 shows the minimum mounting heightrecommended for different types of HID systems.13.2.3.4 Exit SignsRecent advances in exit sign systems have createdTable 13.6 Minimum mounting heights for HIDs—————————————————————————Lamp Typefeet above ground—————————————————————————400 W Metal Halide 161000 W Metal Halide 20200 W High Pressure Sodium 15250 W High Pressure Sodium 16400 W High Pressure Sodium 181000 W High Pressure Sodium 26—————————————————————————attractive opportunities to reduce energy and maintenancecosts. Because emergency exit signs should operate 24hours per day, energy savings quickly recover retrofit costs.There are generally two options, buying a new exit sign, orretrofitting the existing exit sign with new light sources.Most retrofit kits available today contain adaptersthat screw into the existing incandescent sockets. Installationis easy, usually requiring only 15 minutes per sign.However, if a sign is severely discolored or damaged,buying a new sign might be required in order to maintainilluminance as required by fire codes.Basically, there are five upgrade technologies: CompactFluorescent Lamps (CFLs), incandescent assemblies,Light Emitting Diodes (LED), Electroluminescent panels,and Self Luminous Tubes.Replacing incandescent sources with compact fluorescentlamps was the “first generation” exit sign upgrade.Most CFL kits must be hard-wired and can not simplyscrew into an existing incandescent socket. Although CFLkits are a great improvement over incandescent exit signs,more technologically advanced upgrades are availablethat offer reduced maintenance costs, greater efficacy andflexibility for installation in low (sub-zero) temperatureenvironments.As Table 13.7 shows, LED upgrades are the mostcost-effective because they consume very little energy,and have an extremely long life, practically eliminatingmaintenance.Table 13.7 Exit sign upgrades.————————————————————————————————————Light Source Watts Life Replacement————————————————————————————————————Incandescent (Long Life) 40 8 months lampsCompact Fluorescent 10 1.7 years lampsIncandescent Assembly 8 3 + years light sourceLight Emitting Diode (LED) 25 light sourceElectroluminescent 1 8+ years panelSelf luminous (Tritium) 0 10-20 years luminous tubes————————————————————————————————————

364 ENERGY MANAGEMENT HANDBOOKAnother low-maintenance upgrade is to install a“rope” of incandescent assemblies. These low-voltage “luminousropes” are an easy retrofit because they can screwinto existing sockets like LED retrofit kits. However, theincandescent assemblies create bright spots which are visiblethrough the transparent exit sign and the non-uniformglow is a noticeable change. In addition, the incandescentassemblies don’t last nearly as long as LEDs.Although electroluminescent panels consume lessthan one watt, light output rapidly depreciates overtime. These self-luminous sources are obviously the mostenergy-efficient, consuming no electricity. However thespent tritium tubes, which illuminate the unit, must bedisposed of as a radioactive waste, which will increaseover-all costs.13.2.3.5 Lighting ControlsLighting controls offer the ability for systems to beturned ON and OFF either manually or automatically.There are several control technology upgrades for lightingsystems, ranging from simple (installing manualswitches in proper locations) to sophisticated (installingoccupancy sensors).SwitchesThe standard manual, single-pole switch was thefirst energy conservation device. It is also the simplestdevice and provides the least options. One negativeaspect about manual switches is that people often forgetto turn them OFF. If switches are far from roomexits or are difficult to find, occupants are more likelyto leave lights ON when exiting a room. 1 Occupantsdo not want to walk through darkness to find exits.However, if switches are located in the right locations,with multiple points of control for a single circuit,occupants find it easier to turn systems OFF. Onceoccupants get in the habit of turning lights OFF uponexit, more complex systems may not be necessary. Thepoint is: switches can be great energy conservationdevices as long as they are convenient to use them.Another opportunity for upgrading controls existswhen lighting systems are designed such that allcircuits in an area are controlled from one switch, yetnot all circuits need to be activated. For example, acollege football stadium’s lighting system is designedto provide enough light for TV applications. However,this intense amount of light is not needed for regularpractice nights or other non-TV events. Because thelights are all controlled from one switch, every timethe facility is used all the lights are turned ON. Bydividing the circuits and installing one more switchto allow the football stadium to use only 70% of itslights during practice nights, significant energy savingsare possible.Generally, if it is not too difficult to re-circuit apoorly designed lighting system, additional switches canbe added to optimize the lighting controls.Time ClocksTime clocks can be used to control lights when theiroperation is based on a fixed operating schedule. Timeclocks are available in electronic or mechanical styles.However, regular check-ups are needed to ensure thatthe time clock is controlling the system properly. Aftera power loss, electronic timers without battery backupscan get off schedule—cycling ON and OFF at the wrongtimes. It requires a great deal of maintenance time to resetisolated time clocks if many are installed.PhotocellsFor most outdoor lighting applications, photocells(which turn lights ON when it gets dark, and off whensufficient daylight is available) offer a low-maintenancealternative to time clocks. Unlike time clocks, photocellsare seasonally self-adjusting and automatically switchON when light levels are low, such as during rainy days.A photocell is inexpensive and can be installed on eachfixture, or can be installed to control numerous fixtureson one circuit. Photocells can also be effectively used indoors,if daylight is available through skylights.Photocells have worked well in almost any climate,however they should be aimed north (in the northernhemisphere) to “view” the reflected light of the north sky.This way they are not biased by the directionality of east/west exposure or degraded by intense southern exposure.Photocells should also be cleaned when fixtures are relamped.Otherwise, dust will accumulate on the photodiodeaperture, causing the controls to always perceive it isa cloudy day, and the lights will stay ON.The least expensive type of photocell uses a cadmiumsulfide cell, but these cells lose sensitivity after beingin service for a few years by being degraded from theirexposure to sunlight. This decreases savings by keepingexterior lighting on longer than required. To avoid thissituation, cadmium sulfide cells can be replaced with electronictypes that do not lose sensitivity over time. Theseelectronic photocells use solid-state, silicon phototransistorsor photodiodes, which last longer as evidenced bytheir longer warranties—up to 6 years—and can easilypay back before that time with energy and labor savings.Photocells combined with DimmableBallasts to allow Daylight HarvestingDaylight harvesting is a control strategy that can be

LIGHTING 365applied where diffuse daylight can be used effectivelyto light interior spaces. There is a widespread misunderstandingthat daylighting can only be done in areas wherethere is a predominance of sunny, clear days, such asCalifornia or Arizona. In fact, many places with over 50%cloudy days can cost-effectively use daylight controls.Daylight harvesting employs strategically locatedphoto-sensors and electronic dimming ballasts. To effectivelyapply this strategy requires more knowledgethan just plugging a sensor into a dimming ballast. Photosensorsand dimming ballasts form a control system thatcontrols the light level according to the daylight level. Thefluorescent lighting is dimmed to maintain a band of lightlevel when there is sufficient daylight present in the space.The output is changed gradually by a fade control so occupantsare not disturbed by rapid changes in light level.Lumen Depreciation Compensation(an additional benefit of a Daylight Harvesting System)Lighting systems are usually over-designed tocompensate for light losses that normally occur duringthe life time of the system. Alternatively, the “lumendepreciation compensation strategy” allows the designlight level to be met without over-designing, therebyproviding a more efficient lighting system. The controlsystem works in a way similar to daylight harvestingcontrols. A photo-sensor detects the actual light level andprovides a low-voltage signal to electronic dimming ballaststo adjust the light level. When lamps are new androom surfaces are clean, less power is required to providethe design light level. As lamps depreciate in their lightoutput and as surfaces become dirty, the input power andlight level is increased gradually to compensate for thesesources of light loss. Some building management systemsaccomplish this control by using a depreciation algorithmto adjust the output of the electronic ballasts instead ofrelying on photo-sensors.Table 13.8 Estimated % savingsfrom occupancy sensors.—————————————————————————ApplicationEnergy Savings—————————————————————————Offices (Private) 25-50%Offices (Open Spaces) 20-25%Rest Rooms 30-75%Corridors 30-40%Storage Areas 45-65%Meeting Rooms 45-65%Conference Rooms 45-65%Warehouses 50-75%—————————————————————————Occupancy SensorsOccupancy sensors save energy by turning offlights in spaces that are unoccupied. When the sensordetects motion, it activates a control device that turnsON a lighting system. If no motion is detected within aspecified period, the lights are turned OFF until motionis sensed again. With most sensors, sensitivity (the abilityto detect motion) and the time delay (difference intime between when sensor detects no motion and lightsgo OFF) are adjustable. Occupancy sensors are producedin two primary types: Ultrasonic (US) and PassiveInfrared (PIR). Dual-Technology (DT) sensors, thathave both ultrasonic and passive infrared detectors, arealso available. Table 13.8 shows the estimated percentenergy savings from occupancy sensor installation forvarious locations.US and PIR sensors are available as wall-switchsensors, or remote sensors such as ceiling mounted oroutdoor commercial grade units. With remote sensors,a low-voltage wire connects each sensor to an electricalrelay and control module, which operates on commonvoltages. With wall-switch sensors, the sensor and controlmodule are packaged as one unit. Multiple sensors and/or lighting circuits can be linked to one control moduleallowing flexibility for optimum design.Wall-switch sensors can replace existing manualswitches in small areas such as offices, conference rooms,and some classrooms. However, in these applications, amanual override switch should be available so that thelights can be turned OFF for slide presentations and othervisual displays. Wall-switch sensors should have an unobstructedcoverage pattern (absolutely necessary for PIRsensors) of the room it controls.Ceiling-mounted units are appropriate in corridors,rest rooms, open office areas with partitions and anyspace where objects obstruct the line of sight from a wallmountedsensor location. Commercial grade outdoor unitscan also be used in indoor warehouses and large aisles.Sensors designed for outdoor use are typically heavy duty,and usually have the adjustable sensitivities and coveragepatterns for maximum flexibility. Table 13.9 indicates theappropriate sensors for various applications.Ultrasonic Sensors (US)Ultrasonic sensors transmit and receive high-frequencysound waves above the range of human hearing.The sound waves bounce around the room and returnto the sensor. Any motion within the room distorts thesound waves. The sensor detects this distortion andsignals the lights to turn ON. When no motion has beendetected over a user-specified time, the sensor sends asignal to turn the lights OFF. Because ultrasonic sensors

366 ENERGY MANAGEMENT HANDBOOKTable 13.9 Occupancy sensor applications.Private Large Partitioned Conference Rest Closets/ Hallways WarehouseOffice Open Office Room Room Copy Corridors AislesSensor Office Plan Room AreasTechnologyPlan———————————————————————————————————————————————————US Wall Switch • • • •US Ceiling Mount • • • • • •IR Wall Switch • • •IR Ceiling Mount • • • • •US Narrow ViewIR High Mount Narrow View • •Corner Mount Wide- • •View Technology Type•need enclosed spaces (for good sound wave echo reflection),they can only be used indoors and perform betterif room surfaces are hard, where sound wave absorptionis minimized. Ultrasonic sensors are most sensitive tomotion toward or away from the sensor. Applications includerooms with objects that obstruct the sensor’s line ofsight coverage of the room, such as restroom stalls, lockerrooms and storage areas.Passive Infrared Sensors (PIR)Passive Infrared sensors detect differences in infraredenergy emanating in the room. When a person moves,the sensor “sees” a heat source move from one zone tothe next. PIR sensors require an unobstructed view, andas distance from the sensor increases, larger motions arenecessary to trigger the sensor. Applications include openplan offices (without partitions), classrooms and otherareas that allow a clear line of sight from the sensor.Dual-Technology Sensors (DT)Dual-Technology (DT) sensors combine both USand PIR sensing technologies. DT sensors can improvesensor reliability and minimize false switching. However,these types of sensors are still only limited to applicationswhere ultrasonic sensors will work.Occupancy Sensor Effect on Lamp LifeOccupancy Sensors can cause rapid ON/OFFswitching which reduces the life of certain fluorescentlamps. Offices without occupancy sensors usually havelights constantly ON for approximately ten hours per day.After occupancy sensors are installed, the lamps may beturned ON and OFF several times per day. Several laboratorytests have shown that some fluorescent lamps loseabout 25% of their life if turned OFF and ON every threehours. Although occupancy sensors may cause lamp lifeto be reduced, the annual burning hours also decreases.Therefore, in most applications, the time period until relampwill not increase. However, due to the laboratoryresults, occupancy sensors should be carefully evaluatedif the lights will be turned ON and OFF rapidly. The longerthe lights are left OFF, the longer lamps will last.The frequency at which occupants enter a roommakes a difference in the actual percent time savings possible.Occupancy sensors save the most energy when appliedin rooms that are not used for long periods of time.If a room is frequently used and occupants re-enter a roombefore the lights have had a chance to turn OFF, no energywill be saved. Therefore, a room that is occupied onceevery three hours will be more appropriate for occupancysensors than a room occupied once every three minutes,even though the percent vacancy time is the same.Occupancy Sensors and HIDsAlthough occupancy sensors were not primarily developedfor HIDs, some special HID ballasts (bi-level) offerthe ability to dim and re-light lamps quickly. Anotherterm for bi-level HID technology is Capacitive SwitchingHID Fixtures, which are discussed in the HID Ballast Section.Lighting Controls via a Facility Management SystemWhen lighting systems are connected to a FacilityManagement System (FMS), greater control options canbe realized. The FMS could control lights (and otherequipment, i.e. HVAC) to turn OFF during non-working

LIGHTING 367hours, except when other sensors indicate that a spaceis occupied. These sensors include standard occupancysensors or a card access system, which could indicatewhich employee is in a particular part of the facility. Ifthe facility is “smart,” it will know where the employeeworks and control the lights and other systems in thatarea. By wiring all systems to the FMS, there is a greaterability to integrate technologies for maximum performanceand savings. For example, an employee can controllights by entering a code into the telephone system or a computernetwork.Specialized controls for individual work environments(offices or cubicles) are also available. These systemsuse an occupancy sensor to regulate lights, otherelectronic systems (and even HVAC systems) in an energyefficient manner. In some systems, remote controls allowthe occupant to regulate individual lighting and HVACsystems. These customized systems have allowed someorganizations to realize individual productivity gains viamore effective and aesthetic work space environments.13.3 PROCESS TO IMPROVELIGHTING EFFICIENCYThe three basic steps to improving the efficiency oflighting systems:1. Identify necessary light quantity and quality to performvisual task.2. Increase light source efficiency if occupancy is frequent.3. Optimize lighting controls if occupancy is infrequent.Step 1, identifying the proper lighting quantity andquality is essential to any illuminated space. However,steps 2 & 3 are options that can be explored individuallyor together. Steps 2 & 3 can both be implemented, but oftenthe two options are economically mutually exclusive.If you can turn OFF a lighting system for the majorityof time, the extra expense to upgrade lighting sourcesis rarely justified. Remember, light source upgrades willonly save energy (relative to the existing system) whenthe lights are ON.13.3.1 Identify necessary light quantitiesand qualities to perform tasks.Identifying the necessary light quantities for a taskis the first step of a lighting retrofit. Often this step isoverlooked because most energy managers try to mimicthe illumination of an existing system, even if it is over-illuminatedand contains many sources of glare. For manyyears, lighting systems were designed with the belief thatno space can be over-illuminated. However, the “morelight is better” myth has been dispelled and light levelsrecommended by the IES declined by 15% in hospitals,17% in schools, 21% in office buildings and 34% in retailbuildings. 2 Even with IES’s adjustments, there are stillmany excessively illuminated spaces in use today. Energymanagers can reap remarkable savings by simply redesigninga lighting system so that the proper illuminationlevels are produced.Although the number of workplane footcandles areimportant, the occupant needs to have a contrast so thathe can perform a task. For example, during the daytimeyour car headlights don’t create enough contrast to benoticeable. However, at night, your headlights provideenough contrast for the task. The same amount of lightis provided by the headlights during both periods, butdaylight “washes out” the contrast of the headlights.The same principle applies to offices, and otherilluminated spaces. For a task to appear relativelybright, objects surrounding that task must be relativelydark. For example, if ambient light is excessive (150fc) the occupant’s eyes will adjust to it and perceive itas the “norm.” However when the occupant wants tofocus on something he/she may require an additionallight to accent the task (at 200 fc). This excessively illuminatedspace results in unnecessary energy consumption.The occupant would see better if ambient lightwas reduced to 30-40 fc and the task light was used toaccent the task at 50 fc. As discussed earlier, excessiveillumination is not only wasteful, but it can reducethe comfort of the visual environment and decreaseworker productivity.After identifying the proper quantity of light, theproper quality must be chosen. The CRI, CCT and VCPmust be specified to suit the space.13.3.2 Increase Source EfficacyIncreasing the source efficacy of a lighting systemmeans replacing or modifying the lamps, ballasts and/orfixtures to become more efficient. In the past, the term“source” has been used to imply only the lamp of a system.However, due to the inter-relationships betweencomponents of modern lighting systems, we also considerballast and fixture retrofits as “source upgrades.”Thus increasing the efficacy simply means getting morelumens per watt out of lighting system. For example, toincrease the source efficacy of a T12 system with a magneticballast, the ballast and lamps could be replaced withT8 lamps and an electronic ballast, which is a more efficacious(efficient) system.Another retrofit that would increase source efficacy

368 ENERGY MANAGEMENT HANDBOOKwould be to improve the fixture efficiency by installingreflectors and more efficient lenses. This retrofit wouldincrease the lumens per watt, because with reflectors andefficient lenses, more lumens can escape the fixture, whilethe power supplied remains constant.Increasing the efficiency of a light source is one ofthe most popular types of lighting retrofits because energysavings can almost be guaranteed if the new systemconsumes less watts than the old system. With reducedlighting load, electrical demand savings are also usuallyobtained. In addition, lighting quality can be improvedby specifying sources with higher CRI and improvedperformance. These benefits allow capital improvementsfor lighting systems that pay for themselves through increasedprofits.Task lightingAs a subset of Increasing Source Efficacy, “Tasklighting” or “Task/Ambient” lighting techniques involveimproving the efficiency of lighting in an entire workplace,by replacing and relocating lighting systems. Tasklighting means retrofitting lighting systems to provideappropriate illumination for each task. Usually, this resultsin a reduction of ambient light levels, while maintainingor increasing the light levels on a particular task.For example, in an office the light level needed on a deskcould be 75 fc. The light needed in aisles is only 20 fc.Traditional uniform lighting design would create a workplacewhere ambient lighting provides 75 fc throughoutthe entire workspace. Task lighting would create an environmentwhere each desk is illuminated to 75 fc, and theaisles only to 20 fc. Figure 13.4 shows a typical applicationof Task/Ambient lighting.Task lighting upgrades are a model of energy efficiency,because they only illuminate what is necessary.Task lighting designs are best suited for office environmentswith VDTs and/or where modular furniture canincorporate task lighting under shelves. Alternatively,Figure 13.4 Task/ambient lighting.moveable desk lamps may be used for task illumination.Savings result when the energy saved from reducing ambientlight levels exceeds the energy used for task lights.In most work spaces, a variety of visual tasks areperformed, and each employee has lighting preferences.Most workers prefer lighting systems designed with tasklighting because it is flexible and allows individual control.For example, older workers may require greater lightlevels than young workers. Identifying task lighting opportunitiesmay require some creativity, but the potentialdollar savings can be enormous.Task lighting techniques are also applicable in industrialfacilities—for example, high intensity task lightscan be installed on fork trucks (to supplement headlights)for use in rarely occupied warehouses. With this system,the entire warehouse’s lighting can be reduced, saving alarge amount of energy.13.3.3 Optimize Lighting ControlsThe third step of lighting energy management is toinvestigate optimizing lighting controls. As shown earlier,improving the efficiency of a lighting system can save apercentage of the energy consumed while the system is operating.However, sophisticated controls can turn systemsOFF when they are not needed, allowing energy savingsto accumulate quickly. The Electric Power Research Institute(EPRI) reports that spaces in an average office buildingmay only be occupied 60-75% of the time, althoughthe lights may be ON for the entire 10 hour day 3 . Lightingcontrols include switches, time clocks, occupancy sensorsand other devices that regulate a lighting system. Thesesystems are discussed in Section 13.2.3, Lighting SystemComponents.13.4 MAINTENANCE13.4.1 Isolated SystemsMost lighting manuals prescribe specialized technologiesto efficiently provide light for particular tasks.An example is dimmable ballasts. For areas that havesufficient daylight, dimmable ballasts can be used withintegrated circuitry to reduce energy consumption duringpeak periods. Still, though there may be some sheddingof lighting load along the perimeter, these energycost savings may not represent a great percentage of thebuilding’s total lighting load. Further, applications ofspecialized technologies (such as dimmable ballasts) maybe dispersed and isolated in several buildings, which canbecome a complex maintenance challenge, even if lamptypes and locations are recorded properly. If maintenancepersonnel need to make additional site visits to get the

LIGHTING 369right equipment to re-lamp or “fine-tune” special systems,the labor costs may exceed the energy cost savings.In facilities with low potential for energy cost savings,facility managers may not want to spend a greatdeal of time monitoring and “fine-tuning” a lightingsystem if other maintenance concerns need attention. Ifa specialized lighting system malfunctions, repair mayrequire special components, that may be expensive andmore difficult to install. If maintenance cannot effectivelyrepair the complex technologies, the systems will failand occupant complaints will increase. Thus the isolated,complex technology that appeared to be a unique solutionto a particular lighting issue is often replaced with asystem that is easy to maintain.In addition to the often eventual replacement oftechnologies that are difficult to maintain, well intendedrepairs to the system may accidentally result in “snapback.”“Snap-back” is when a specialized or isolatedtechnology is accidentally replaced with a common technologywithin the facility. For example, if dimmable ballastsonly represent 10% of the building’s total ballasts,maintenance personnel might not keep them in stock.When replacement is needed, the maintenance personnelmay accidentally install a regular ballast. Thus, the lightingretrofit has “snapped back” to its original condition.The above arguments are not meant to “shootdown” the application of all new technologies. However,new technologies usually bring new problems. The authorsask that the energy manager carefully consider themaintenance impact when evaluating an isolated technology.Once again, all lighting systems depend on regularmaintenance.13.4.2 Maintaining System PerformanceAs with most manufactured products, lighting systemslose performance over time. This degradation canbe the result of Lamp Lumen Depreciation (LLD), FixtureDirt Depreciation (LDD), Room Surface Dirt Depreciation(RSDD), and many other factors. Several of these factorscan be recovered to maintain performance of the lightingsystem. Figure 13.5 shows the LLD for various types oflighting systemsLamp Lumen Depreciation occurs because as thelamp ages, its performance degrades. LLD can be acceleratedif the lamp is operated in harsh environments, orthe system is subjected to conditions for which it was notdesigned. For example, if a fluorescent system is turnedON and OFF every minute, the lamps and ballasts willnot last as long. Light loss due to lamp lumen depreciationcan be recovered by re-lamping the fixture.Fixture Dirt Depreciation and Room Surface DirtDepreciation block light and can reduce light levels.However, these factors can be minimized by cleaningsurfaces and minimizing dust. The magnitude of thesefactors is dependent on each room, thus recommendedcleaning intervals can vary. Generally it is most economicalto clean fixtures when re-lamping.13.4.3 Group Re-lampingMost companies replace lamps when someone noticesa lamp is burned out. In a high rise building, this couldbecome a full-time job, running from floor to floor, officeto office, disrupting work to open a fixture and replace alamp. However, in certain cases, it is less costly to groupre-lamp on a pre-determined date. Group relamping canFigure 13.5 Lamp lumen depreciation (LLD).

370 ENERGY MANAGEMENT HANDBOOKbe cost-effective due to economies of scale. Replacing alllamps at one time can be more efficient than relamping“one at a time.” In addition, bulk purchasing may alsoyield savings. The rule of thumb is: group relamp at 50%to 70% of the lamp’s rated life. However, depending onsite-specific factors and the lumen depreciation of thelighting system, relamping interval may vary.The facility manager must evaluate their own building,and determine the appropriate relamp interval byobserving when lamps start to fail. Due to variations inpower voltages (spikes, surges and low power), lampsmay have different operating characteristics and livesfrom one facility to another. It is important to maintainrecords on lamp and ballast replacements and determinethe most appropriate relamping interval. This also helpskeep track of maintenance costs, labor needs and budgets.Group relamping is the least costly method torelamp due to reduced time and labor costs. For example,Table 13.10 shows the benefits of group relamping. Asmore states adopt legislation requiring special disposalof lighting systems, group relamping in bulk may offerreduced disposal costs due to large volumes of material.13.4.4 Disposal CostsDisposal costs and regulations for lighting systemsvary from state to state. These expenses should be includedin an economic analysis of any retrofit. If properdisposal regulations are not followed, the EPA couldimpose fines and hold the violating company liable forenvironmental damage in the future.13.5 NEW TECHNOLOGIES & PRODUCTS 1The energy efficient lighting market is extremelycompetitive, forcing manufacturers to develop newproducts to survive. The development is so rapid, it ischallenging to “keep up” with all the latest technologies.This chapter describes the proven technologies, howeverit is good idea to evaluate the latest developments beforeimplementing a lighting system.13.5.1 Fluorescent BallastsMiniaturization of electronic ballasts has been madepossible by the use of integrated circuits and surfacemounttechnologies. The new ballasts are smaller, thinnerand lighter.Low Profile HousingThe familiar “brick” shape and weight of ballastswill soon be gone. Reduced parts count and surfacemount technology have reduced the size of ballasts aswell as improved their reliability. These advances havepermitted housings of lower profile and smaller crosssection.Today, some ballasts have a dimensional cross-Table 13.10 Group relamping example: 1,000 3-Lamp T8 Lensed troffers————————————————————————————————————Spot Relamping Group Relamping(on burn-out) (@ 70% rated life)————————————————————————————————————Relamp cycle 20,000 hours 14,000 hoursAvg. relamps/year 525 relamps/yr 750 relamps/yr (group)52 relamps/yr (spot)Avg. material cost/year $1,050/yr $1,604/yrLamp disposal @ 0.50 ea. $236/yr $375/yrAvg. labor cost/year $3,150/yr $1,437/yrTOTAL EXPENSES: $4,463/yr $3,416/yrAssumptions: Labor: $6.00/lamp $1.50/lampMaterial: $2.00/lamp $2.00/lampOperation: 3,500 hr/yr 3,500 hr/yr————————————————————————————————————1 The majority of this section was provided by John Fetters, Effective Lighting Solutions. ©Effective Lighting Solutions, Inc.

LIGHTING 371section of 30 × 30 mm. The advantages of smaller ballastpackages include lighter weight, less material, and easierhandling and installation. In addition, they fit into thenew low-profile fixtures, especially indirect and directindirectfixtures.Universal Input VoltageMany facilities have different lighting system voltagesin different parts of their buildings. Maintenancepersonnel are slowed in their ballast replacement taskwhen they don’t know the voltage for a particular area ofthe building. However, ballasts with the universal voltagefeature will automatically use any line voltage applied(between 120-277-v). In addition to saving valuablemaintenance time when the labor cost of identifying thevoltage for each ballast to be replaced, or the expense ofdistributor restocking of ballasts ordered with the incorrectvoltage is included, any cost difference is very affordable.In addition, fewer replacement ballast models needto be stocked.Optimizing Ballast SelectionInstant-start ballasts have become the most popularmethod of starting F32T8/RS rapid-start lamps becauseof their lower input watts rating compared withrapid-start systems. However, lamp life can be reducedby up to 25% at short burn cycles when lamps are operatedinstant-start, increasing maintenance costs. Inapplications where short ON/OFF cycles are common,lamp life increases by using program-start or rapid-startballasts, instead of instant-start ballasts. Rapid-startoperation of rapid-start lamps will ensure normal ratedlamp life and program-start ballasts can extend lamp lifeby up to 50%.Dimming electronic ballasts for fluorescent lamps.Electronic ballasts with dimming functions operatefluorescent lamps at high frequency, just like fixed-outputelectronic ballasts. Most dimmable ballasts now haveseparate low-voltage control leads, which can be groupedtogether to create control zones, which are independentof the power zones. Many dimming ballast designsprovide over-voltage protection of the control leads incase line voltage is accidentally applied to the low voltageleads. The control method of choice is 0 to 10vDC,although dimming ballasts are also available, which aredesigned to accept the AC line phase control signals fromincandescent wall-box dimmer controls that dim the fluorescentlamps accordingly.Dimming ballasts are divided into two categories,based on dimming ranges:1. Energy management applications: 100% to 5%2. Architectural dimming applications100% to 1% (or less)Note: Today, dimming ballasts for energy-management applicationsare also being used in applications that formerly requiredan architectural dimming ballast, such as conference rooms.Dimmable ballasts are available for dimming mostlinear fluorescent lamps (1-, 2- or 3-lamp versions) includingT5 HO lamps. Many of these products start the lampsat any dimmer setting, and do not have to be ramped upto full-light output before they dim. Most of the modelsavailable measure less than 15% total harmonic distortion(THD) throughout the dimming range.Conference and presentation rooms have traditionallybeen built with two lighting systems. One, anincandescent system, usually uses recessed cans and isdimmed with wall dimmers. The second is usually a fluorescentnon-dimming system for general lighting. Theincandescent system requires a lot of maintenance, dueto the short life of the incandescent lamps. One solutionto this situation is to remove the overhead incandescentsystem and replace the ballasts in the fluorescent systemwith line-voltage dimming ballasts that can operate fromthe existing incandescent wall-box dimmer(s). The mainbenefit of this improvement is lower maintenance cost fora small investment in ballasts and the electrical maintenancestaff can make the change.Electronic ballasts for compact fluorescent lamps (CFLs)Several manufacturers have dimming ballasts forrapid-start (4-pin) compact fluorescent lamps (CFLs).Most of these offerings are for the higher-wattage CFLs(26 to 57-w). The lowest dimming limit is 5% and thedimming range varies with the manufacturer. Thereare designs to accept the AC phase-control signals fromincandescent wall-box dimmer controls. This makesupgrading an older incandescent downlight system toan energy-efficient CFL system easy, with no new wiringrequired.13.5.2 Fluorescent LampsSeveral smaller, yet brighter fluorescent systems (T2, T5and T5HO) have flourished in recent years. Smaller systemshave been effective in task lighting environments, where lesslight from a single source is needed. The reduction of unnecessarylighting reduces energy expenses.T2 lampsThese are sub-miniature, 1/4” (0.25”) diameterlamps that have side tabs instead of end pins. They are

372 ENERGY MANAGEMENT HANDBOOKavailable in standard fluorescent colors of 3000K, 3500K,and 4100K with CRI in the mid-80s. They are rated witha lamp lumen depreciation of 0.95, and only lose 5% oftheir light output in the first 40% of rated life. T2 lampshave lamp efficacy ratings similar to compact fluorescentin the mid-60s.Low profile fixtures used for task and under-counterlighting, showcase and decorative lighting have beenmade possible by these small diameter lamps. Their principalapplication, however, is for backlighting graphicdisplay panels, which are starting to be done with highperformancelight-emitting diodes (LEDs). The use of T2lamps for this application is not expected to increase.T5 lampsThe T5 lamps come in two distinct and differentfamilies—standard (high-efficiency) and high output(HO). These recently developed lamps should not beconfused with older miniature preheat fluorescent lampsof the same diameter nor with the line of long compactfluorescent lamps of the same diameter.Standard (high-efficiency) T5 linear lampsThese 5/8” diameter lamps (Figure 13-6) areequipped with miniature bi-pin bases and are poweredby electronic ballasts. All the lamps in this family operateon the same current (170 ma) and have the same surfacebrightness for all wattages. For cove and cornice applicationsthis is a distinct advantage.Another reason the T5 lamp is suited for these applicationsis that they are designed to peak in their lumenrating at 35°C (95°F) vs. 25°C (77°F) for T12 and T8lamps. This characteristic provides higher light outputin confined applications where there is little or no aircirculation. In indirect fixtures, this thermal characteristicincreases efficiency and gives more usable lumens perwatt.Standard T5 lamps are 12-18% more efficient than T8lamps (96-106 LPW) and 10-15% more efficient than theT5HO. T5s employ rare-earth phosphors with CRI greaterthan 80 and lamp lumen maintenance rated at 95%.There are 4 sizes of standard T5 lamps as shownin Table 13.11, all rated at 20,000 hours (at 3 hours-perstart).Note that the 28-watt lamp (not quite 4’ long) hasan initial lumen rating the same as a 4’ T8 lamp. However,the millimeter lengths and miniature bi-pin basespreclude their use in standard length linear fluorescentsystems and the high bulb-wall brightness limits their useto high ceiling applications because the visible tubes cancreate too much discomfort glare in low mounting heightapplications.Table 13.11 Standard T5 Lamp Sizes(Source: Effective Lighting Solutions, Inc.)—————————————————————————Nominal Length Lumens LumensWatts mm/(in) (initial) (maintained)—————————————————————————14 549/(21.6) 1,350 1,28321 849/(33.4) 2,100 1,99528 1149/(45.2) 2,900 2,75535 1449/(57.0) 3,650 3,460—————————————————————————High output T5 linear lamps (SEE TABLE 13.12)These T5 lamps are physically the same size as standardT5 lamps, but provide higher lumen output. T5HOlamps generate from 1.5 to 2 times the light output of thestandard T5 and nearly twice the light output (188%) ofT8 and T12 systems with the same number of lamps. OnelampT5 HO fixtures can replace both lamps of 2-lamp T8fixtures.They are approximately 10-15% less efficient thanstandard T5 lamps (83 to 94 lumens per watt) and can beup to 8% less efficient than standard T8 systems. The surfacebrightness varies among various wattages and sincelamps operate on different currents (see Table 13.12), eachlamp wattage requires a unique ballast.T5 HO lamps are available in the three standard fluorescentcolor temperatures (cool—4100K, warm—3000Kand neutral—3500K) and have a color rendering indexgreater than 80. Lumen maintenance is rated at 95%.These lamps are now rated at 20,000 hours. Similar tostandard T5 lamps, T5HO lamps also peak in their lumenrating at 35°C (95°F) vs. 25°C (77°F) for T12 and T8 lamps.This provides higher light output in confined applicationswhere there is little or no air circulation. In indirectfixtures, this thermal characteristic results in increasedefficiency with more usable lumens per watt.T5HO lamps are being used in designs of slim profileindirect fixtures that take advantage of the smallerlamp. Only 1 lamp per 4-ft section is required, replacingTable 13.12 T5HO Lamp Sizes(Source: Effective Lighting Solutions, Inc.)—————————————————————————Nominal Length Lamp Lumens LumensWatts mm/(in) Current (ma) (initial) (maintained)—————————————————————————24 549/(21.6) 300 2,000 1,90039 849/(33.4) 340 3,500 3,32554 1149/(45.2) 460 5,000 4,75080 1449/(57.0) 552 7,500 7,125—————————————————————————

LIGHTING 373designs using 2, 4-ft T8 lamps per 4-ft section. The highbulb wall brightness limits their use in direct applicationsin low ceiling height conditions due to discomfort glareFollowing the trend to fluorescent, T5HO lamps are beingused in high ceiling applications, including high-bayindustrial fixtures.T8 lampsStandard T8 lamps (See Figure 13.6)T8 lamps (1" dia) were originally imported from Europein the early 1980s. The lamps now used in the US aredifferent than their European pre-heat cousins and thereare improved models. T8s have been the lamps of choice(along with the high frequency electronic ballasts thatdrive them) for fluorescent upgrades for several years. T8lamps are available in 2’, 3’, 4’, and 5’ lengths, at 17, 25, 32,and 40-w respectively. These lamps require ballasts thatsupply 265 ma. There are also two versions of 8' retrofitlamps at 59, or 86-w. U-tubes are available in the new 15/8” leg spacing and a retrofit U-tube that has 6” leg spacingthat is used to replace 6" leg-spacing T12 U-tubes.Recent advances in T8 lamps have been in improvementsin color rendering and longer life. Extendedperformance T8 lamps have a life rating of 24,000 (at 3hours per start)—20% longer than standard T8 lamps.These extended performance lamps operate on the sameelectronic ballasts designed to operate standard T8 lamps.Lumen maintenance is rated at 0.94 and it levels off afterthat. Lumen output is slightly higher at 3,000 lumens andCRI is improved to 8. Standard 3000K, 3500K and 4100Kcolors are provided.Figure 13.6 Some T8 Lamp Sizes (Source: Sylvania)Reduced-wattage T8 lampsSometimes called “energy-saving” T8 lamps, theselamps are available in 28 and 30-w vs. the standard 32-wmodels. They are designed to replace reduced-wattage,34-w T12 lamps, when upgrading to electronic ballasts.They are recommended for use only on instant-start electronicballasts to provide the higher open-circuit voltagerequired and they need to be operated above 60°F. In addition,they cost more than standard T8s, but they saveabout 6% over standard lamps, are TCLP compliant, andhave high lumen maintenance (94%).High-performance T8 lamps (See Table 13.13)These high-lumen lamps (3,100 L) are part of adedicated lamp/ballast system that can save about 19%over standard T8 systems with the same light output andtwice the lamp life (on program-start ballasts) comparedto standard instant-start T8 systems. They exhibit highlumen maintenance (95%) and high CRI (86).Table 13.13 High-Performance T8 System Watts(Source: Osram-Sylvania)—————————————————————————Number of Lamps 1 2 3 4—————————————————————————Input (system) Watts 25-w 48-w 72-w 94-w—————————————————————————TCLP compliant fluorescent lampsOver 600 million fluorescent lamp tubes are disposedof every year in the US. Prior to June 1999, theUSEPA required that spent fluorescent lamps that did notpass a Toxicity Characteristic Leaching Procedure (TCLP)were to be treated as hazardous waste because they containedmore than 0.2 mg/liter (ppm) of mercury.Standard fluorescent lamps do not pass the TCLPtest and were required (prior to the Universal WasteRule) to be handled as hazardous waste or recycled byusing expensive hazardous waste haulers and massivedocumentation. Fluorescent lamps are now coveredby the Universal Waste Rule (as of today). The mainresult of the inclusion of fluorescent lamps in the universalwaste rule is to encourage recycling of spentlamps.Lamp manufacturers have reduced the mercurycontent of fluorescent tubes over the past decade toless than half of the original content. In response topublic concern for mercury in the environment, themajor lamp companies started to produce what theyoriginally called “low-mercury” lamps. Philips Lightingusing a proprietary dosing and buffering technologythey call ALTO ® , produced the first low-mercuryfluorescent lamps. 4-foot ALTO fluorescent lamps haveless than 10 mg of mercury and therefore will pass theTCLP test. Other lamp companies have followed thistrend and now these lamps are called “TCLP compliant”lamps to indicate that the lamps are designed topass the federal TCLP (Toxic Characteristic Leaching

374 ENERGY MANAGEMENT HANDBOOKProcedure) test.Low mercury lamps have distinctive colored endcaps, usually colored green. Use of TCLP compliantlamps provides users with normal lamp performance,light output, and life and an environmentally friendlyoption to meet their lighting needs. Although they do notneed to be recycled, many end-users are avoiding anyliability for their lamp disposal and recycling their spentTCLP compliant lamps.13.5.3 Compact Fluorescent LampsImprovements to CFL technologies have been occurringevery year since they became commercially available.Products available today provide higher efficaciesas well as instant starting, reduced lamp flicker, quietoperation, smaller size and lighter weight. DimmableCFLs are now available, and it can be expected that theirperformance will increase with time. The 2700K color (incandescentappearance) has been replaced by the 3000Kfor commercial applications. “Pre-heat” models start byblinking before they stay ON. Older lamps blink morethan new lamps during starting. Rapid-start models startinstantly, with no blinking.Traditional Problems with CFLsCFLs suffer from multiple sensitivities that reducethe light output. They are position-sensitive. Gravitydetermines where the excess mercury “pools,” which affectsthe mercury vapor pressure that determines thelumen output. Lumen ratings published in lamp catalogsare performed according to ANSI testing standards thatrequire the lamp to be in the vertical, “base up” position.In the base-down position some CFLs produce 20% fewerlumens. In the horizontal position, they produce about15% less light. Lamp lumen depreciation for CFLs is oftenmore accelerated than for incandescent sources. CFLs arealso not recommended for wet applications.Additional sensitivities include temperature sensitivitythat reduces the light output when CFLs are operatedabove or below their optimum temperature rating.The loss due to temperature is approximately 15-20% andis most noticeable in enclosed fixtures, such as recesseddownlights due to self-heating. However, when the mercuryused in a CFL is in the form of an amalgam—an alloy of mercuryand other metals, the mercury vapor pressure is reducedwithout affecting the lamp temperature. This technique makesthe lamps less temperature sensitive than conventional CFLsand provides more light at the high and low extremes—above100°F and below 32°F. Amalgam lamps are not easily identified,but most “triple-tubes” are amalgam products.Screw-base CFLsScrew-base CFLs have “Edison” bases and are usedto replace incandescent lamps. They have an integralballast built into the base. Early models had magneticballasts built into the base, however most contemporarymodels have electronic bases, allowing significant sizereduction. Some screw-base CFLs are used in commercialapplications, but most are used in residential lighting.The newest shape is the spiral or spring shape, shown inFigure 13.7. Higher wattage options are shown in Table13.14.2-pin preheat CFLs (Figure 13.8)2-pin CFLs require a separate ballast (which is usuallymagnetic) located in the fixture. Each lamp has a starter,located in the base, which provides pre-heat starting. Table13.15 shows twin-tube and quad pre-heat models.Figure 13.7 Spiral or Spring Shaped CFLsTable 13.14 Large Screw-base Compact Fluorescent Lamps(Source: www.maxlite.com)————————————————————————————————————Shape Lumens Watts LPW M.O.D M.O.L. Replaces————————————————————————————————————Spiral 3,500 55 64 3.5" 10" 200-w incandQuad 4,200 65 65 3.5" 11" 250-w incandQuad 5,500 85 65 3.5" 11.8" 300-w incand————————————————————————————————————

LIGHTING 375Figure 13.8 Twin-tube Pre-heat CFLs (Source: Sylvania)Table 13.15 Pre-Heat Compact Fluorescent Lamps(Source: Effective Lighting Solutions, Inc.)—————————————————————————LampDescription Lumens Watts Lumens/Watt—————————————————————————Twin-tube Pre-heat 250 5 50400 7 57600 9 66900 13 69Quad Pre-heat 575 9 64860 13 661200 18 661800 26 69—————————————————————————4-pin rapid-start CFLs4-pin rapid-start lamps are available in 16, 18, 24,26, 28, 32, and 42-watt models. All commercial-grademodels are rated at 10,000 hours. The majority of theselamps are T5 (5/8" tube diameter), but some are T4(1/2" tube diameter). Maximum overall length (MOL)ranges from 3.5" to 5.5.” The three primary color temperaturesare available—3000K (warm), 3500K (neutral),and 4100K (cool). At least one manufacturer alsoprovides the warmer 2700K color. Rapid-start CFLs aredesigned for operation on electronic ballasts and canbe dimmed when operated on a dimming electronicballast, designed for the appropriate lamp wattage.New generation compact fluorescent lamps (CFL)are a significant improvement over the earlier twinand double twin tube types. Instead of using freemercury, these new CFLs use mercury that has beencombined with other metals to form an amalgam. Theamalgam makes the lamps less sensitive to the effectsof temperature and position.This is an important advantage over standardCFLs and is the reason that many applications usingFigure 13.9 Double Arc-Tube HPS (Source: Sylvania)standard CFLs perform poorly. Amalgam CFLs havestable light output from 23°F to 130°F. Also, amalgamlamps are not position sensitive and exhibit less colorshift than conventional CFLs. They do take slightlylonger to warm up, but they are at full brightness inless than 3 minutes. Unfortunately, manufacturers donot always clearly identify their amalgam lamps, butmost “triple” tubes are amalgam.CFLs are available in higher wattage for use inhigh ceiling downlights. A 32-watt triple-tube, amalgamlamp, rated at 2,400 lumens, provides a systemreplacement for 150-watt incandescent downlights. A42-watt triple-tube, amalgam lamp, rated at 3,200 lumens,allows its use as a system replacement for highwattageincandescent downlights and a 57-w rapidstart, triple-tube, amalgam lamp, rated at 4,300 lumens,is equivalent to a 200-w incandescent lamp.At Lightfair International 2003, Philips Lightingunveiled a new multiple burning position, high lumenoutputPL-H lamp. These 4-pin, rapid start lamps areused with high frequency, electronic ballasts. They arecomposed of 6, T5 limbs, joined with bend-and-bridgetechnology. There are 6 models with wattages rangingfrom 60 to 120-w. Versatile and powerful they have alumen output almost double that of other CFLs, up to9,000 lumens (120-w model), they provide maximumdesign freedom in many areas, including high ceilingindoor and outdoor applications. In addition, the whitelight PL-H range promises stable color rendering, longlife and high lumen maintenance.Dimmable CFLsScrew-base dimmable CFLs were introduced in1996. This lamp is intended to replace incandescentlamps used on wallbox dimming systems. The elec-

376 ENERGY MANAGEMENT HANDBOOKtronic ballast base in this 1-piece lamp responds to thephase change voltage waveform from most existingdimmers and dims the CFL down to 10% light output.The dimmable CFL is available in several wattages,the most common being a 23-watt triple-tube amalgamlamp with a lumen rating of 1500 that will replace 90-watt“A” lamps. The major benefit of this lamp is that dimmingis accomplished on existing dimming circuits withno additional control wiring required.13.5.4 High Intensity Discharge (HID) Systems:Metal Halide SystemsMetal Halide lamps have become more populardue to technological advancements and consumer preferencefor “white light.” Technologically, the “pulsestart”metal halide systems are a significant improvementin efficiency and performance. Like most electronicballasts, these operate at high frequency, providea quicker re-strike time (3-5 minutes) versus standardmetal halide systems (6-10 minutes). The pulse-startsystems maintain CRI and lumen output better overtime.Pulse-start metal halideLow-wattage metal halide (< 175-w) and high-pressuresodium lamps have used pulse-start technology formany years, using a high voltage pulse starter to ignitethe lamps. What is new is the availability of high-wattage,pulse-start metal halide lamps (175-w to 1000-w) thatare quickly replacing standard metal halide lamps. Thereis a new family of arc tubes, called “formed body” that replacethe old pinched seal arc tubes and overcome the disadvantagesof the old design. The starter electrode, foundin standard arc tubes, has been eliminated. The new arctube design features uniform geometry and higher fillpressures. Improved temperature control is achievedwith smaller pinch seals that provide less heat loss, reducinglamp-to-lamp color shift. Formed body arc tubelamps provide a lower ambient temperature limit, -40°Finstead of -30°F for standard arc tube lamps. Faster startingand restarting (re-strike) results from the lower massof the new arc tubes. These changes result in higher lampefficacy (up to 110 lumens per watt), improved lumenmaintenance (up to 80%), consistent lamp-to-lamp color(within 100°K) and 50% faster warm up and re-striketimes (three to five minutes vs. eight to 15 minutes).Ceramic metal halide lamps (CMH)Ceramic arc tube metal halide lamps use the sameceramic material used in high-pressure sodium arctubes—polycrystalline alumina (PCA). PCA reduces thesodium loss through the more porous glass arc tube usedin standard metal halide lamps. This reduces color shiftand spectral variation of standard metal halide lampscaused as the sodium is depleted. Metal halide lampswith ceramic arc tubes are designated either CDM (ceramicdischarge, metal halide) or CMH (ceramic metalhalide) and may also refer to their constant color in theirbrand name. They are available from 20 to 400-watt withcolor temperature of either 3000K (warm) or 4000K (cool)and an average rated life from 6,000 to 15,000 hours, dependingon the wattage. Ceramic metal halide lamps arestarted by a pulse starter like PS metal halide lamps andoperate best on electronic ballasts. The main advantage ofthe combination of CMH lamps and electronic ballasts is10-20% higher lumen output (which also results in a correspondinghigher LPW) and the best color stability.The benefits of CMH lamps include good lamp efficacy(83-95 LPW)—in the same range as older, linearfluorescent lamps; high CRI (83-95); limited color shift(from ± 75K to ± 200K CCT); excellent lamp-to-lamp colorconsistency; and good lumen maintenance (0.70-0.80).Applications for these improved color metal halidelamps include high ceilings such as atria, and lobbies ofhospitality spaces, downlights, and lighting merchandise—anywherethat the higher CRI and color consistencycan be justified. Fade-block models with thin-film coatingson the arc-tube shroud are available for merchandiselighting to help reduce the UV fading of materials. Thereare also HPS replacement lamps that can be used as aninterim solution when converting from a high-pressuresodium system to a white light system.High Pressure Sodium systemsTwo new lamp wattages are available to narrowthe gap between the 400-w and the 1,000-w standardHPS lamps. Both sizes will probably not survive themarket and the 600-watt, 90,000 lumen lamp is not aswidely supported by the lighting industry as the 750-w,105,000 lumen lamp as a good "in between" size. In general,however, all high-pressure sodium lamps are losingground to white-light sources, such as metal halide orfluorescent.Several improvements have been introduced in“new-generation” HPS lamps. The major improvement isthe elimination of end-of-life cycling that is characteristicof standard high-pressure sodium lamps. However, thereare two different design approaches by the three majorlamp companies. Two companies have taken the ‘notification’approach, in which the lamp turns a distinctive bluecolor at end of life. A third company simply shuts off thelamp power at end of life.

LIGHTING 377New HPS lamps have welded bases that replacethe old lead soldered bases. Several new models have reducedor zero mercury content, qualifying them as TCLPcompliant lamps.These lamps sacrifice efficacy and life to achieveCRI rating up to 65. Lamp efficacy ranges from 63 to 94LPW and they have an average rated life of 15,000 hours.They are available in 70, 100, 150, 250, and 400-w models,and have lumen ratings from 4,400 to 37,500.Double arc-tube HPS lampsThese HPS lamps are called standby lamps andhave two arc tubes, welded together in parallel. Howeveronly one arc tube operates when the lamp is ignited.Upon the loss of power, the second arc tube, hot frombeing in close proximity to the first arc tube, comes ONat about 50% light output. It then comes up to full lightoutput within the strike time of the lamp (~4 min max).Standby lamps are used for safety and security applicationsand are popular with prison lighting systems aswell as roadway systems, with a tested life of 40,000hours, reducing maintenance time and labor cost.13.5.5 Induction LightingElectrodeless Induction SystemsSince the introduction of the first electric light, asearch has been on for long-life lighting. The reason forthis search is to reduce the cost associated with changinglighting components at or near their end of ratedlife—maintenance cost.The lamps used in induction systems have no electrodesto wear out as other lamps, such as fluorescent andHID lamps do. The lamps can last much longer withoutelectrodes. Long life is the primary advantage of thesesystems. These systems can provide a good paybackwhere maintenance labor cost is high. When comparedwith other light sources, electrodeless induction systemswill operate 5-8 times longer than fluorescent and metalhalide systems and about 4 times longer than HPS systems.In addition, induction lamps come ON relativelyquickly and have short re-strike time compared with HIDlamps.Instead of using electrodes to generate electronsas is done in fluorescent lamps, electrodeless systemsproduce light by means of induction—the use of an electromagneticfield to induce a plasma gas discharge intoa tube or bulb that has a phosphor coating. No electronsare needed, since the gas discharge is induced into thebulb or tube by a high-frequency electronic generatorthat supplies the electromagnetic field. These systemsprovide white light with a minimum color shift and CRIand lumen depreciation values are similar to fluorescentlamps.Each of the two primary electrodeless system lampshas a unique size and shape and require new fixtures thatare designed to optically match each unique shape. Thereis no common electronics package for these productssince they operate on much different frequencies. Theelectronics package must be fairly close to the glass envelopesand the maximum mounting distance is restrictedto the wire length supplied on the electronics package.These are independent systems and are designed so thatboth the glass envelopes and the electronics are changedout together at end of life.Genura LampGE Lighting developed an electrodeless inductionlamp labeled Genura and introduced it in the US in1995. Genura is a compact R30 reflector lamp witha standard medium base that is intended for use as aretrofit lamp (in place of a 100-watt A lamp, a 75-wattR30 lamp or a 65-watt R30 lamp) in recessed downlights(cans). This product is a lamp and not a system, so it iscovered here before the induction systems.QL Induction Lighting System (Figure 13.10)Philips Lighting developed this induction systemand introduced it to the European market in 1991. In 1992the QL was introduced to the US market. The QL systemis comprised of three components. 1) a high-frequencygenerator, 2) a power coupler and 3) the glass bulb. Thehigh-frequency generator is in a separate electronicspackage that provides the 2.65 MHz current to the powercoupler (antenna) through a coax cable. The power couplersits inside the enclosed glass discharge bulb shapedlike a large A lamp. The bulb, which contains an inert gasFigure 13.10 QL Lighting System (Source: Philips Lighting)

378 ENERGY MANAGEMENT HANDBOOKand a small amount of mercury is attached to the powercoupler by a plastic lamp cap that uses a click system.Like fluorescent lamps, the inside walls of the bulb arecoated with a phosphor coating. When the high frequencyelectromagnetic field is applied to the bulb, the gas isionized and the lamp produces photons at UV frequencyand visible light in the same manner as a fluorescent tube.The photons collide with the phosphor coating and causethe lamp to glow. Full brightness is achieved in 10-15 seconds.The system meets FCC requirements as a low EMIdesign.There are three models—55-w, 3,500 lumens, 85-w,6,000 lumens, and 165-w, 12,000 lumens—with lumen efficacyof 64-73 LPW. The 55-watt model has a maximumoverall diameter (MOD) of 85 mm (~3 3/8"), the 85-wattmodel has a MOD of 111 mm (~ 4 3/8"), and the 165-wmodel has a MOD of 131 mm (~5 5/32"). QL bulbs areavailable in two color temperatures—3000K (warm) and4000K (cool).The main advantage of the QL system is its longlife—average rated life is 100,000 hours. Philips rates lifeas 20% failures at 60,000 hours. This long life advantageis especially important where maintenance cost is high.The current emphasis in the U.S. is in outdoor lightingsystems—street, roadway, and tunnel lighting systems.Several fixture manufacturers have incorporated the QLin their designs.ICETRON SystemThe Inductively Coupled Electrodeless system—ICETRON—was developed by Sylvania. This electrodelesssystem consists of three parts: 1) a unique rectangular‘donut’ shaped bulb—filled with an inert gas and asmall amount of mercury, 2) two ring-shaped ferrite corecouplers—one at each of the short sides of the bulb, and3) a separate high-frequency (200-300 KHz) generator. Aplug-in connector attaches leads from the couplers to theelectronic generator. The driver may be mounted up to 66feet away from the lamp.When the high frequency electromagnetic field is appliedto the donut-shaped bulb between the ferrite cores ateach end, the gas inside the bulb is ionized and produceslight by inducing a circulating current in the bulb, whichgenerates photons at UV frequency. These photons collidewith the phosphor coating and cause the lamp to glow.ICETRON lamps strike and re-strike instantly.There are three ICETRON lamps—70-w, 100-w,and 150-w—and two drivers. Table 13.16 shows the combinationsof lamps and drivers and the resulting systemperformance.The mercury in the glass envelope is in the formof an amalgam, providing a universal burn situation.Table 13.16 ICETRON System Performance(Source: Sylvania)—————————————————————————Lamp Driver System System SystemLumens Watts LPW—————————————————————————70-w 100-w 6,500 82 79100-w 100-w 8,000 107 75100-w 150-w 11,000 157 70150-w 150-w 12,000 157 76—————————————————————————Figure 13.11 ICETRON System (Source: Sylvania)Starting temperatures extend down to –40°F, opening upopportunities for low temperature applications such asfreezers and coolers. The ICETRON bulbs are availablein two color temperatures—3500K (neutral) and4100K (cool) and a CRI of 80. Sylvania rates the lumenmaintenance at 70% at 60,000 hours (60% of rated life).This is a departure from the standard method of ratinglumen maintenance for other light sources (at 40% ratedlife). The lumen maintenance curve shows a lumen maintenancevalue of 75 at 40%. At the rated life of 100,000hours, the lumen maintenance is about 65%.The ICETRON system meets FCC (non-consumer)requirements and has a low EMI design. The principaladvantage of this system is long life—100,000 hours. Acomprehensive warranty covers the system for 60 months.Applications where maintenance is difficult and or costlyare prime candidates for these long life systems.13.5.6 Remote Source Lighting and Fiber OpticsRemote source lighting systems have the lightingsource some distance from the point of delivery. Basically,

LIGHTING 379the light source is connected to a light pipe or fiber optics,which carries the light to the point of application. Remotelighting solutions have become more popular becausethey fill the needs of projects that have hazardous orunderwater environments, walk-in freezers, architecturalrestrictions or special aesthetic objectives. Remote sourcelighting systems offer reduced maintenance costs, becauselamps can be accessed easily and safely. For example,light pipes can be effective in gymnasiums or swimming pools.The uniform lighting also can result in a lower glare than singlebright fixtures.Fiber optics can be used to resolve challenges associatedwith maintaining aesthetics. Light sources canbe installed in rooms outside of a viewing area, with thefiber optics routed through walls (or other obscured spaces—likecrown molding) to the application. Like miniatureflashlights, the fiber optics can be pointed directlyat the needed spot. For example, gallery or church lightingcan be achieved without bulky fixtures getting in the way of theoccupant’s view.13.6 SPECIAL CONSIDERATIONS13.6.1 Rules and RegulationsEPACT-1992The National Energy Policy Act of 1992 (EPACT)was designed to dramatically reduce energy consumptionvia more competitive electricity generation and moreefficient buildings, lights and motors. Because lightingis common in nearly all buildings, it is a primary focusof EPACT. The 1992 legislation bans the production oflamps that have low efficacy or CRI. Table 13.17 indicateswhich lamps are banned and a few options for replacingthe banned systems. From left to right, the table showsseveral options, for each banned system, ranging fromthe most efficient substitute to the minimum compliancesubstitute. Generally, the minimum compliance substitutehas the lowest initial cost, but after energy costs havebeen included, the most efficient upgrades have the lowestlife-cycle costs.Often the main expense with a lighting upgradeis the labor cost to install new products; however, theincremental labor cost of installing high-efficiencyequipment is minimal. So, it is usually beneficial toinstall the most efficient technologies because they willhave the lowest operational and life-cycle costs. EPACTonly eliminates the “bottom of the barrel” in terms ofavailable lighting technology. To keep one step ahead offuture lamp bans, it is a good idea to consider upgradeswith greater efficiencies than the minimum acceptablesubstitute.Federal Fluorescent Ballast RuleAn agreement between lighting manufacturers (representedby the National Electrical Manufacturers Association—NEMA)and energy policy advocates (The AmericanCouncil for an Energy Efficient Economy—ACEEE,The Alliance to Save Energy, and the National ResourcesDefense Council—NRDC) was finalized on September2000 and became law as Part 430—Energy ConservationProgram for Consumer Products. The new standards areexpected to reduce greenhouse gas emissions by 19 millionmetric tons of carbon and by 60,000 tons of nitrousoxide over the next 20 years—the equivalent of eliminatingthe emissions of one million cars for 15 years.The rule promotes T8 electronic systems (withoutcreating efficiency standards for T8 ballasts) by raisingthe minimum BEF for T12 ballasts to a level that can onlybe achieved by electronic ballasts. T12 magnetic ballastsare still allowed, but these are a small fraction of a shrinkingfluorescent magnetic market. They are less efficientand carry a cost premium, so in actual practice they willnot be used.No magnetic ballasts may be manufactured for thecovered lamps (2' U-tubes, 4' rapid-start, 8' instant-start,and 8' HO) after June 30, 2005. Magnetic ballasts for T8lamps can continue to be manufactured for applicationssensitive to infrared (IR) or electromagnetic interference(EMI). Luminaires sold on or after April 1, 2006 that usethe covered T12 lamps must incorporate electronic ballasts.An exception is made for magnetic ballasts used forreplacement purposes in existing installations, which canbe manufactured until June 30, 2010, but must be marked“FOR REPLACEMENT USE ONLY.”There is an implied warning to all fluorescent lampusers that if they have not converted to T8 systems byJune 30, 2010, they will have to use T8 ballasts and lampsfor spot replacement in their existing T12 systems, whichcan only result in compatibility problems and a real maintenanceheadache!Electrical ConsiderationsDue to the increasingly complex lighting productsavailable today, concern about effects on power distributionsystems have risen. In certain situations, lightingretrofits can reduce the power quality of an electricalsystem. Poor power quality can waste energy and thecapacity of an electrical system. In addition, it can harmthe electrical distribution system and devices operatingon that system.Electrical concerns peaked when the first generationelectronic ballasts for fluorescent lamps caused powerquality problems. Due to advances in technology, electronicballasts available today can improve power quality

380 ENERGY MANAGEMENT HANDBOOKTable 13.17 EPACT 1992’s effect: lamp bans and options.(Continued)when replacing magnetically ballasted systems in almostevery facility. However, some isolated problems may stilloccur in electronically sensitive environments such asintensive-care units in hospitals. In these types of areas,special electromagnetic shielding devices are available,and are usually required.The energy manager should ensure that a new systemwill improve the power quality of the electrical system.HarmonicsA harmonic is a higher multiple of the primary frequency(usually 60 Hertz) superimposed on the alternatingcurrent waveform. A distorted 60 Hz current wavemay contain harmonics at 120 Hz, 180 Hz and so on. Theharmonic whose frequency is twice that of the fundamentalis called the “second-order” harmonic. The harmonicwhose frequency is three times the fundamental is the“third-order” harmonic.Highly distorted current waveforms contain numerousharmonics. The even harmonics (second-order,fourth order, etc.) tend to cancel each other’s effects, butthe odd harmonics tend to add in a way that rapidly increasesdistortion because the peaks and troughs of theirwaveforms coincide. Lighting products usually indicatea common measurement of distortion percentage: TotalHarmonic Distortion (THD). Table 13.18 shows the %THD for various types of lighting and office equipment.13.6.2 HVAC EffectsNearly all energy consumed by lighting systems isconverted to light, heat and noise, which dissipate into thebuilding. Therefore, if the amount of energy consumed bya lighting system is reduced, the amount of heat energygoing into the building will also be reduced, and less airconditioningwill be needed. Consequently, the amountof winter-time heating may be increased to compensatefor a lighting system that dissipates less heat.Because most offices use air-conditioning for moremonths per year than heating, a more efficient lightingsystem can significantly reduce air-conditioning costs.In addition, air conditioning (usually electric) is muchmore expensive that heating (usually gas). Therefore, the

LIGHTING 381Table 13.17 EPACT 1992’s effect: lamp bans and options (Conclusion).savings on air-conditioning electricity are usually worthmore dollars than the additional gas cost.13.6.3 The Human AspectRegardless of the method selected for achievingenergy savings, it is important to consider the humanaspect of energy conservation. Buildings and lightingsystems should be designed to help occupants work incomfort, safety and enjoyment. Retrofits that improve thelighting quality (and the performance of workers) shouldbe installed, especially when they save money. The recentadvances in electronic ballast technology offer an opportunityfor energy conservation to actually improveworker productivity. High frequency electronic ballastsand tri-phosphor lamps offer improved CRI, less audiblenoise and lamp flicker. These benefits have been shownto improve worker productivity and reduce headaches,fatigue and absenteeism.

382 ENERGY MANAGEMENT HANDBOOKTable 13.18 Power quality characteristics for different electric devices.Implementation TacticsIn addition to utilizing the appropriate lightingproducts, the implementation method of a lightingupgrade can have a serious impact on its success. Toensure favorable reaction and support from employees,they must be involved in the lighting upgrade. Educatingemployees and allowing them to participate in thedecision process of an upgrade will reduce the resistanceof change to a new system. Of critical importance is themaintenance department, because they will have an importantrole in the future upkeep of the system.

LIGHTING 383Once the decision has been made to upgrade thelighting in a particular area, and a trial installation has receivedapproval, a complete retrofit should be completedas soon as possible. Due to economies of scale and minimalemployee distraction, an all-at-once retrofit is usuallyoptimal. In some cases, an over-night or over-the-weekendinstallation might be preferred. This method wouldavoid possible criticisms from side-by-side comparisonsof the old and new systems. For example, a task lightingretrofit may appear darker than a uniformly illuminatedspace adjacent to it. The average worker who believes“more light is better” might protest the retrofit. However,if the upgrade is done over the weekend, the worker maynot easily notice the changes.13.6.4 Lighting Waste and the EnvironmentUpgrading any lighting system will require disposalof lamps and ballasts. Some of this waste may behazardous and/or require special management. Contactyour state to identify the regulations regarding the properdisposal of lighting equipment in your area.MercuryWith the exception of incandescent bulbs, nearly allgaseous discharge lamps (fluorescent and HIDs) containsmall quantities of mercury that end up in the environment,unless recycled. Mercury is also emitted as a byproductof electricity generation from some fossil-fueledpower plants. Although compact fluorescent lamps containthe most mercury per lamp, they save a great deal ofenergy when compared to incandescent sources. Becausethey reduce energy consumption, (and avoid power plantemissions) CFLs introduce to the environment less thanhalf the mercury of incandescents. 5 Mercury sealed inglass lamps is also much less available to ecosystems thanmercury dispersed throughout the atmosphere. Nevertheless,mercury is not good for our environment and theenergy manager should check local disposal codes—youdon’t want to break the law. Mercury in lamps can be recycled,and regulations may soon require it.PCB BallastsBallasts produced prior to 1979 may contain Polychlorinatedbiphenyls (PCBs). Human exposure to thesepossible carcinogens can cause skin, liver, and reproductivedisorders. Older fluorescent and HID ballasts containhigh concentrations of PCBs. These chemical compoundswere widely used as insulators in electrical equipmentsuch as capacitors, switches and voltage regulators until1979. The proper method for disposing used PCB ballastsdepends on the regulations in the state where the ballastsare removed or discarded. Generators of PCB containingballast wastes may be subject to notification and liabilityprovisions under the Comprehensive EnvironmentalResponse, Compensation and Liability Act of 1980 (CER-CLA)—also known as “Superfund.” 6Generally, the PCB ballast is considered to be ahazardous waste only when the ballast is leaking PCBs.An indication of possible PCB leaking is an oily tar-likesubstance emanating from the ballast. If the substancecontains PCBs, the ballast and all materials it contacts areconsidered PCB waste, and are subject to state regulations.Leaking PCB ballasts must be incinerated at an EPAapproved high-temperature incinerator.Energy Savings and Reduced Power Plant EmissionsWhen appliances use less electricity, power plantsdon’t need to produce as much electricity. Because mostpower plants use fossil fuels, a reduction in electricitygeneration results in reduced fossil fuel combustion andairborne emissions. Considering the different types ofpower plants (and the different fuels used) in differentgeographic regions, the Environmental Protection Agencyhas calculated the reduced power plant emissions bysaving one kWh. Table 13.19 shows the reduction of CO 2,SO 2 and NO x for each kWh per year saved, in differentregions of the US. 713.7 DAYLIGHTINGHuman beings developed with daylight as theirprimary light source. For thousands of years humansevolved to the frequency of natural diurnal illumination.Daylight is a flicker-free source, generally with thewidest spectral power distribution and highest comfortlevels. With the twentieth century’s trend towards largerbuildings and dense urban environments, the developmentand wide spread acceptance of fluorescent lightingallowed electric light to become the primary source inoffices.Daylighting interior spaces is making a comebackbecause it can provide good visual comfort, and it cansave energy if electric light loads can be reduced. Newcontrol technologies and improved daylighting methodsallow lighting designers to conserve energy and optimizeemployee productivity.There are three primary daylighting techniquesavailable for interior spaces: Utilizing Skylights, BuildingPerimeter Daylighting and Building Core Daylighting.Skylights are the most primitive and are the most commonin industrial buildings. Perimeter daylighting is definedas using natural daylight (when sufficient) such thatelectric lights can be dimmed or shut off near windows

384 ENERGY MANAGEMENT HANDBOOKTable 13.19

LIGHTING 385at the building perimeter. Traditionally, the amount ofdimming depends on the interior distance from fenestration.However, ongoing research and application of “CoreDaylighting Techniques” can stretch daylight penetrationdistance further into a room. Core daylighting techniquesinclude the use of light shelves, light pipes, active daylightingsystems and fiber optics. These technologies willlikely become popular in the near future.Perimeter and core daylighting technologies aretechnologies that are being further developed to functionin the modern office environment. The modern office hasmany visual tasks that require special considerations toavoid excess illumination or glare. It is important to properlycontrol daylighting in offices, so that excessive glaredoes not reduce employee comfort or the ability to workon VDTs. A poorly designed or poorly managed daylitspace reduces occupant satisfaction and can increase energyuse if occupants require additional electric light tobalance excessive daylight-induced contrast.Windows and daylighting typically cause an increasedsolar heat gain and additional cooling load forHVAC systems. However, development of new glazingsand high performance windows has allowed designersto use daylighting without severe heat gain penalties.With dynamic controls, most daylit spaces can now havelower cooling loads than non-daylit spaces with identicalfenestration. “The reduction in heat-from-lights due todaylighting can represent a 10% down-sizing in perimeterzone cooling and fans. 8 ” However, because there areseveral parameters, daylighting does not always reducecooling loads any time it displaces electric light. As windowsize increases, the maximum necessary daylightmay be exceeded, creating additional cooling loads.Whether interior daylighting techniques can be economicallyutilized depends on several factors. Howeverthe ability to significantly reduce electric lighting loadsduring utility “peak periods” is extremely attractive.13.8 COMMON RETROFITSAlthough there are numerous potential combinationsof lamps, ballasts and lighting systems, a few retrofitsare very common.OfficesIn office applications, popular and profitable retrofitsinvolve installing electronic ballasts and energyefficient lamps, and in some cases, reflectors. Table 13.20shows how a typical system changes with the addition ofreflectors and the removal or substitution of lamps andballasts. Notice that thin lamps allow more light to exitthe fixture, thereby increasing fixture efficiency. Reflectorsimprove efficiency by greater amounts when thereare less lamps (or thinner lamps) to block exiting lightbeams.The expression (Lumens)/(Fixture watt) is an indicatorof the overall efficiency of the lighting system. It issimilar to the efficacy of a lamp.Indoor/Outdoor IndustrialIn nearly all applications with significant annualoperating hours, mercury vapor systems can be replacedby metal halide (or high output fluorescent systems). Thisretrofit will improve CRI, reduce operating and relampingcosts. In applications where CRI is not critical, HPSsystems (which have a higher efficacy than metal halidesystems) can be used.Almost AnywhereIn nearly all applications where incandescent lampsare ON for more than 5 hours per day, switching to CFLswill be cost-effective.13.8.1 Sample RetrofitsThis section provides the equations to calculatesavings from several different types of retrofits. For eachtype of retrofit, the calculations shown are based on averageconditions and costs, which vary from location tolocation. For example, annual air conditioning hours willvary from building to building and from state to state.The energy costs used in the following examples werebased on $10/kW-month and $.05/kWh. In most industrialsettings, demand is also billed. In the following examplesdemand savings would likely occur in all exceptfor examples # 4 and # 6. To accurately estimate the costand savings from these types of retrofits, simply insert localvalues into the equations.EXAMPLE 1:UPGRADE T12 LIGHTING SYSTEM TO T8A hospital had 415 T12 fluorescent fixtures, whichoperate 24 hours/day, year round. The lamps and ballastswere replaced with T8 lamps and electronic ballasts,which saved about 30% of the energy, and providedhigher quality light. Although the T8 lamps cost a littlemore (resulting in additional lamp replacement costs),the energy savings quickly recovered the expense. In addition,because the T8 system produces less heat, air conditioningrequirements during summer months will bereduced. Conversely, heating requirements during wintermonths will be increased.

386 ENERGY MANAGEMENT HANDBOOKTable 13.20 Fluorescent lighting upgrade options.

LIGHTING 387CalculationskW Savings= (# fixtures) [(Present input watts/fixture)—(Proposedinput watts/fixture)]= (415)[(86 watts/T12 fixture)-(60 watts/T8 fixture)]= 10.8 kWkWh Savings= (kW savings)(Annual Operating Hours)= (10.8 kW)(8,760 hours/year)= 94,608 kWh/yearAir Conditioning Savings= (kW savings)(Air Conditioning Hours/year)(1/AirConditioner’s COP)= (10.8 kW)(2000 hours)(1/2.6)= 8,308 kWh/yearAdditional Gas Cost= (kW savings)(Heating Hours/year)(.003413 MCF/kWh)(1/Heating Efficiency)(Gas Cost)= (10.8 kW)(1,500 hours/year)(.003413 MCF/kWh)(1/0.8)($4.00/MCF)= $ 276/yearLamp Replacement Cost= [(# fixtures)(# lamps/fixture)][((annual operationalhours/proposed lamp life)(proposed lamp cost))—((annual hours operation/present lamp life)(presentlamp cost))]= [(415 fixtures)(2 lamps/fixture)][((8,760 hours/20,000hours)($ 3.00/T8 lamp)) – ((8,760 hours/20,000hours)($ 1.50/T12 lamp))]= $ 545/yearTotal Annual Dollar Savings= (kW Savings)(kW charge)+[(kWh savings)+(Air Conditioningsavings)](kWh cost) -(Additional gas cost)– (lamp replacement cost)= (10.8 kW)($ 120/kW year)+[(94,608 kWh)+(8,308kWh)]($ 0.05/kWh) -($ 276/year) – ($ 545/year)= $ 5,621/yearImplementation Cost= (# fixtures) (Retrofit cost per fixture)= (415 fixtures) ($ 45/fixture)= $ 18,675Simple Payback= (Implementation Cost)/(Total Annual Dollar Savings)= ($ 18,675)/($ 5,621/year)= 3.3 yearsEXAMPLE 2: REPLACE INCANDESCENT LIGHTINGWITH COMPACT FLUORESCENT LAMPSA power plant has 111 incandescent fixtures whichoperate 24 hours/day, year round. The incandescentlamps were replaced with compact fluorescent lamps,which saved over 70% of the energy, and last over tentimes as long. Because the lamp life is so much longer,there is a maintenance relamping labor savings. Airconditioning savings or heating costs were not includedbecause these fixtures are located in a high-bay buildingwhich is not heated or air-conditioned.CalculationsWatts Saved Per Fixture= (Present input watts/fixture) – (Proposed input watts/fixture)= (150 watts/fixture) – (30 watts/fixture)= 120 watts saved/fixturekW Savings= (# fixtures)(watts saved/fixture)(1 kW/1000 watts)= (111 fixtures)(120 watts/fixtures)(1/1000)= 13.3 kWkWh Savings= (Demand savings)(annual operating hours)= (13.3 kW)(8,760 hours/year)= 116,683 kWh/yearLamp Replacement Cost= [(Number of Fixtures)(cost per CFL Lamp)(operatinghours/lamp life)] – [(Number of existing incandescentbulbs)(cost per bulb)(operating hours/lamp life)]= [(111 Fixtures)($10/CFL lamp)(8,760 hours/10,000hours)] – [(111 bulbs)($1.93/type “A” lamp)(8,760hours/750 hours)]= $ – 1,530/year § § Negative cost indicates savings.Maintenance Relamping Labor Savings= [(# fixtures)(maintenance relamping cost per fixture)][((annual hours operation/present lamp life))-((annualhours operation/proposed lamp life))]= [(111 fixtures)($1.7/fixture)][((8,760/750))-((8,760/10,000))]= $ 2,039/yearTotal Annual Dollar Savings= (kWh savings)(kWh cost)+ (kW savings)(kW cost)– (lamp replacement cost) + (maintenance relampinglabor savings)= (116,683 kWh)($.05/kWh)+(13.3)($120/kW year) -(-1,530/year) + (2,039/year)= $ 10,999/year

388 ENERGY MANAGEMENT HANDBOOKTotal Implementation Cost= [(# fixtures)(cost/CFL ballast and lamp)] + (retrofitlabor cost)]= (111 fixtures)($45/fixture)= $ 4,995Simple Payback= (Total Implementation Cost)/(Total Annual DollarSavings)= ($ 4,995)/(10,999/year)= 0.5 yearsEXAMPLE 3: INSTALL OCCUPANCY SENSORSIn this example, an office building has many individualoffices that are only used during portions of theday. After mounting wall-switch occupancy sensors, thesensitivity and time delay settings were adjusted to optimizethe system. The following analysis is based on anaverage time savings of 35% per room. Air conditioningcosts and demand charges would likely be reduced, howeverthese savings are not included.CalculationskWh Savings= (# rooms)(# fixtures/room)(input watts/fixture)(1 kW/1000 watts) (Total annual operatinghours)(estimated % time saved/100)= (50 rooms)(4 fixtures/room)(144 watts/fixture)(1/1000)(4,000 hours/year)(.35)= 40,320 kWh/yearTotal Annual Dollar Savings ($/Year)= (kWh savings/year)(kWh cost)= (40,320 kWh/year)($.05/kWh)= $ 2,016/yearImplementation Cost= (# occupancy sensors needed)[(cost of occupancy sensor)+(installation time/room)(labor cost)]= (50)[($ 75)+(1 hour/sensor)($20/hour)]= $ 4,750Simple Payback= (Implementation Cost)/(Total Annual Dollar Savings)= ($ 4,750)/($ 2,016/year)= 2.4 yearsEXAMPLE 4: RETROFIT EXIT SIGNS WITH L.E.D.sAn office building had 117 exit signs, whichused incandescent bulbs. The exit signs were retrofittedwith LED exit kits, which saved 90% of the energy. Eventhough the existing incandescent bulbs were “long-life”models, (which are expensive) material and maintenancesavings were significant. Basically, the hospital shouldnot have to relamp exit signs for 25 years!CalculationsInput Wattage – Incandescent Signs= (Watt/fixture) (number of fixtures)= (40 Watts/fix) (117 fix)= 4.68 kWInput Wattage – LED Signs= (Watt/fixture) (number of fixtures)= (3.6 Watts/fix) (117 fix)= .421 kWkW Savings= (Incandescent Wattage) – (LED Wattage)= (4.68 kW) – (.421 kW)= 4.26 kWkWh Savings= (kW Savings)(operating hours)= (4.26 kW)(8,760 hours)= 37,318 kWh/yrLamp Replacement Cost= [(Number of LED Exit Fixtures)(cost per LEDFixture)(operating hours/Fixture life)] – [(Numberof existing Exit lamps)(cost per Exit lamp)(operatinghours/lamp life)]= [(117 Fixtures)($ 60/lamp kit)(8,760 hours/219,000hours)] – [(234 Exit lamps)($5.00/lamp)(8,760hours/8,760 hours)]= -$ 889/year § § Negative cost indicates savings.Maintenance Relamping Labor Savings= (# signs)(Number of times each fixture is relamped/yr)(time to relamp one fixture)(Labor Cost)= (117 signs)(1 relamp/yr)(.25 hours/sign)($20/hour)= $585/yearAnnual Dollar Savings= [(kWh savings)(electrical consumption cost)] + [(kWsavings)(kW cost)] + [Maintenance Cost Savings]-[lamp replacement cost]= [(37,318 kWh)($.05/kWh)] + [(4.26 kW)($120/kW yr)]+[$585/yr] – [– $889/yr]= $ 3,851/year

LIGHTING 389Implementation Cost= [# Proposed Fixtures][(Cost/fixture + InstallationCost/fixture)]= [117][$60/fixture + $5/fixture]= $ 7,605Simple Payback= (Implementation Cost)/(Annual Dollar Savings)= ($7,605)/($3,851/yr)= 2 years.EXAMPLE 5: REPLACE OUTSIDE MERCURY VAPORLIGHTING SYSTEM WITH HIGH AND LOW PRES-SURE SODIUM LIGHTING SYSTEMA parking lot is illuminated by mercury vaporlamps, which are relatively inefficient. The existingfixtures were replaced with a combination of High PressureSodium (HPS) and Low Pressure Sodium (LPS)lamps. The LPS provides the lowest-cost illumination,while the HPS provides enough color rendering abilityto distinguish the colors of cars. By replacing the fifty400 watt Mercury Vapor lamps with ten 250 watt HPSand forty 135 watt LPS fixtures, the company savedapproximately $ 2,750/year with an installed cost of$12,500 and a payback of 4.6 years.EXAMPLE 6:REPLACE “U” LAMPS WITH STRAIGHT T8 TUBESThe existing fixtures were 2’ by 2’ Lay-In Trofferswith two F40T12CW “U” lamps, with a standard ballastconsuming 96 watts per fixture. The retrofit was toremove the “U” lamps and install three F017T8 lampswith an electronic ballast, which had only 47 watts perfixture.13.9 SUMMARYIn summary, this chapter will help the energymanager make informed decisions about lighting. Thefollowing “recipe” reviews some of the main points thatinfluence the effectiveness of lighting retrofits.A Recipe for Successful Lighting Retrofits1. Identify visual task—Distinguish between tasksthat involve walking and tasks that involve readingsmall print.2. Identify lighting needs for each task-Use IES tablesto determine target light levels.3. Research available products and lighting techniques—Talkto lighting manufacturers about yourobjectives, let them help you select the products.Perhaps they will offer a demonstration or trial installation.Be aware of the relative costs, especiallythe costs associated with specialized technologies.4. Identify lamps to fulfill lighting needs—Pick thelamp that has the proper CRI, CCT, lamp life andlumen output.5. Identify ballasts and fixtures to fulfill lighting needs.Select the proper ballast factor, % THD, voltage,fixture light distribution, lenses or baffles, fixtureefficiency.6. Identify the optimal control technology: Decidewhether to use IR, US or DT Occupancy Sensors.Know when to use time clocks or install switches.7. Consider system variations to optimize:Employee performance—Incorporate the importanceof lighting quality into the retrofit process.Energy savings-Pick the most efficient technologiesthat are cost-effective.Maintenance—Installing common systems forsimple maintenance, group re-lamping, maintenancetraining.Ancillary effects—Consider effects on the HVACsystem, security, safety, etc.8. Publicize results—As with any energy managementprogram, your job depends on demonstrating progress.By making energy cost savings known to theemployees and upper-level management, all peoplecontributing to the program will know that there isa benefit to their efforts.9. Continually look for more opportunities—The lightingindustry is constantly developing new productsthat could improve profitability for your company.Keep in-touch with new technologies and methodsto avoid “missing the boat” on a good opportunity.Table 13.21 offers a more complete listing of energysaving ideas for lighting retrofits.

390 ENERGY MANAGEMENT HANDBOOK13.10 SCHEMATICS

LIGHTING 391


LIGHTING 393Exit SignsLED• 1.8-3.6 Input Watts/Fixture. (Replaces standard 20-25 watt lamps.)• Convert existing incandescent EXIT signs to use energy efficientLED liehg strips.• Each kit contains two LED light strips and a reflective backing toprovide even light distribution and a new red lens for the fixture.• Estimated life is 25 years.• Complies with OSHA and NFPA requirements.• Available in four base styles to fit existing sockets or as a hard wirekit.• LED light strips emit a bright red light and are not recommendedfor use with green signs.• In addition to DGSC standard warranty, manufacturer's 25-yearwarranty applies.• UL approved.CFLQuick connectin g adapter screws into existing incandescent socket. For usewith medium screw base societs.Two lamp EXIT sign retrofit system, backup lamp will take over if the primarylamp fails.UL approved.• Lamp: 9 watt, twin tube compact fluorescent• Lumens: 600• Lamp Avg Life: 10,000 hours• Ballast Losses: 2 watts/ballast• System Input Watts: 11 watts• Minimum Starting Temperature: 0°F

394 ENERGY MANAGEMENT HANDBOOKTable 13.21 Energy saving checklist.———————————————————————————————————————————————————Lighting Needs* Visual tasks: specification Identify specific visual tasks and locations to determine recommended illuminances for tasksand for surrounding areas.* Safety and aesthetics Review lighting requirements for given applications to satisfy safety and aesthetic criteria.* Over-illuminated In existing spaces, identify applications where maintained illumination is greater than recomapplicationmended. Reduce energy by adjusting illuminance to meet recommended levels.* Groupings: similar Group visual tasks having the same illuminance requirements, and avoid widely separatedvisual tasksworkstations.* Task lighting Illuminate work surfaces with luminaires properly located in or on furniture; provide lowerambient levels.* Luminance ratios Use wall-washing and lighting of decorative objects to balance brightness.Space Design and Utilization* Space plan When possible, arrange for occupants working after hours to work in close proximity to oneanother.* Room surfaces Use light colors for walls, floors, ceilings and furniture to increase utilization of light, and reduceconnected lighting power to achieve required illuminances. Avoid glossy finishes on roomand work surfaces to limit reflected glare.* Space utilization Use modular branch circuit wiring to allow for flexibility in moving, relocating or addingbranch circuit wiring luminaires to suit changing space configurations.* Space utilization: Light building for occupied periods only, and when required for security or cleaning purposesoccupancy(see chapter 31, Lighting Controls).Daylighting* Daylight compensation If daylighting can be used to replace some electric lighting near fenestration during substantialperiods of the day, lighting in those areas should be circuited so that it may be controlledmanually or automatically by switching or dimming.* Daylight sensing Daylight sensors and dimming systems can reduce electric lighting energy.* Daylight control Maximize the effectiveness of existing fenestration-shading controls (interior and exterior) orautomatically by switching or dimming.* Space utilization Use daylighting in transition zones, in lounge and recreational areas, and for functions wherethe variation in color, intensity and direction may be desirable. Consider applications wheredaylight can be utilized as ambient lighting, supplemented by local task lights.Lighting Sources: Lamps and Ballasts* Source efficacy Install lamps with the highest efficacies to provide the desired light source color and distributionrequirements.(Continued)

LIGHTING 395* Fluorescent lamps Use T8 fluorescent and high-wattage compact fluorescent systems for improved source efficacyand color quality.* Ballasts Use electronic or energy efficient ballasts with fluorescent lamps.* HID Use high-efficacy metal halide and high-pressure sodium light sources for exterior floodlighting.* Incandescent Where incandescent sources are necessary, use reflector halogen lamps for increased efficacy.* Compact fluorescent Use compact fluorescent lamps, where possible, to replace incandescent sources.* Lamp wattagereduced-wattage lampsUse reduced-wattage lamps where illuminance is too high.* Control compatibility If a control system is used, check compatibility of lamps and ballasts with the control device.* System change Substitute metal halide and high-pressure sodium systems for existing mercury vapor lightingsystems.Luminaires* Maintained efficiency Select luminaires which do not collect dirt rapidly and which can be easily cleaned.* Improved maintenance Improved maintenance procedures may enable a lighting system with reduced wattage to provideadequate illumination throughout systems or component life.* Luminaire efficiency Check luminaire effectiveness for task lighting and for overall efficiency; if ineffective orreplacement or relocation inefficient, consider replacement or relocation.* Heat removal When luminaire temperatures exceed optimal system operating temperatures, consider usingspecial luminaires to improve lamp performance and reduce heat gain to the space.* Maintained efficiency Select a lamp replacement schedule for all light sources, to more accurately predict light lossfactors and possibly decrease the number of luminaires required.Lighting controls* Switching; local control Install switches for local and convenient control of lighting by occupants. This should be incombination with a building-wide system to turn lights off when the building is unoccupied.* Selective switching Install selective switching of luminaires according to groupings of working tasks and differentworking hours.* Low-voltageswitching systemsUse low-voltage switching systems to obtain maximum switching capability.* Master control system Use a programmable low-voltage master switching system for the entire building to turn lightson and off automatically as needed, with overrides at individual areas.* Multipurpose spaces Install multi-circuit switching or preset dimming controls to provide flexibility when spaces areused for multiple purposes and require different ranges of illuminance for various activities.Clearly label the control cover plates.* “Tuning” illuminance Use switching and dimming systems as a means of adjusting illuminance for variable lightingrequirements.(Continued)

396 ENERGY MANAGEMENT HANDBOOK* Scheduling Operate lighting according to a predetermined schedule, based on occupancy.* Occupant/motion sensors Use occupant/motion sensors for unpredictable patterns of occupancy.* Lumen maintenance Fluorescent dimming systems may be utilized to maintain illuminance throughout lamp life,thereby saving energy by compensating for lamp-lumen depreciation and other light loss factors.* Ballast switching Use multilevel ballasts and local inboard-outboard lamp switching where a reduction in illuminancesis sometimes desired.Operation and Maintenance* Education Analyze lighting used during working and building cleaning periods, and institute an educationprogram to have personnel turn off incandescent lamps promptly when the space is notin use, fluorescent lamps if the space will not be used for 10 min. or longer, and HID lamps(mercury, metal halide, high-pressure sodium) if the space will not be used for 30 min. or longer.* Parking Restrict parking after hours to specific lots so lighting can be reduced to minimum securityrequirements in unused parking areas.* Custodial service Schedule routine building cleaning during occupied hours.* Reduced illuminance Reduce illuminance during building cleaning periods.* Cleaning schedules Adjust cleaning schedules to minimize time of operation, by concentrating cleaning activitiesin fewer spaces at the same time and by turning off lights in unoccupied areas.* Program evaluation Evaluate the present lighting maintenance program, and revise it as necessary to provide themost efficient use of the lighting system.* Cleaning and maintenance Clean luminaires and replace lamps on a regular maintenance schedule to ensure proper illuminancelevels are maintained.* Regular system checks Check to see if all components are in good working condition. Transmitting or diffusing mediashould be examined, and badly discolored or deteriorated media replaced to improve efficiency.* Renovation of luminaries Replace outdated or damaged luminaires with modern ones which have good cleaning capabilitiesand which use lamps with higher efficacy and good lumen maintenance characteristics.* Area maintenance Trim trees and bushes that may be obstructing outdoor luminaire distribution and creatingunwanted shadow.———————————————————————————————————————————————————

LIGHTING 39713.11 GLOSSARY 9AMPERE: The standard unit of measurement for electriccurrent that is equal to one coulomb per second. It definesthe quantity of electrons moving past a given point in acircuit during a specific period. Amp is an abbreviation.ANSI: Abbreviation for American National StandardsInstitute.ARC TUBE: A tube enclosed by the outer glass envelopeof a HID lamp and made of clear quartz or ceramic thatcontains the arc stream.ASHRAE: American Society of Heating, Refrigeratingand Air-Conditioning Engineers.AVERAGE RATED LIFE: The number of hours at whichhalf of a large group of product samples have failed.BAFFLE: A single opaque or translucent element used tocontrol light distribution at certain angles.BALLAST: A device used to operate fluorescent andHID lamps. The ballast provides the necessary startingvoltage, while limiting and regulating the lamp currentduring operation.BALLAST CYCLING: Undesirable condition underwhich the ballast turns lamps ON and OFF (cycles) dueto the overheating of the thermal switch inside the ballast.This may be due to incorrect lamps, improper voltagebeing supplied, high ambient temperature around thefixture, or the early stage of ballast failure.BALLAST EFFICIENCY FACTOR: The ballast efficiencyfactor (BEF) is the ballast factor (see below) divided bythe input power of the ballast. The higher the BEF—withinthe same lamp ballast type—the more efficient theballast.BALLAST FACTOR: The ballast factor (BF) for a specificlamp-ballast combination represents the percentage ofthe rated lamp lumens that will be produced by the combination.CANDELA: Unit of luminous intensity, describing theintensity of a light source in a specific direction.CANDELA DISTRIBUTION: A curve, often on polar coordinates,illustrating the variation of luminous intensityof a lamp or fixture in a plane through the light center.CANDLEPOWER: A measure of luminous intensity of alight source in a specific direction, measured in candelas(see above).COEFFICIENT OF UTILIZATION: The ratio of lumensfrom a fixture received on the work plane to the lumensproduced by the lamps alone. (Also called “CU”)COLOR RENDERING INDEX (CRI): A scale of the effectof a light source on the color appearance of an objectcompared to its color appearance under a reference lightsource. Expressed on a scale of 1 to 100, where 100 indicatesno color shift. A low CRI rating suggests that the colorsof objects will appear unnatural under that particularlight source.COLOR TEMPERATURE: The color temperature is aspecification of the color appearance of a light source, relatingthe color to a reference source heated to a particulartemperature, measured by the thermal unit Kelvin. Themeasurement can also be described as the “warmth” or“coolness” of a light source. Generally, sources below3200K are considered “warm;” while those above 4000Kare considered “cool” sources.COMPACT FLUORESCENT: A small fluorescent lampthat is often used as an alternative to incandescent lighting.The lamp life is about 10 times longer than incandescentlamps and is 3-4 times more efficacious. Also calledPL, Twin-Tube, CFL, or BIAX lamps.CONTRAST: The relationship between the luminance ofan object and its background.DIFFUSE: Term describing dispersed light distribution.Refers to the scattering or softening of light.DIFFUSER: A translucent piece of glass or plastic sheetthat shields the light source in a fixture. The light transmittedthroughout the diffuser will be directed and scattered.DIRECT GLARE: Glare produced by a direct view oflight sources. Often the result of insufficiently shieldedlight sources. (SEE GLARE)DOWNLIGHT: A type of ceiling fixture, usually fullyrecessed, where most of the light is directed downward.May feature an open reflector and/or shielding device.EFFICACY: A metric used to compare light output toenergy consumption. Efficacy is measured in lumens per

398 ENERGY MANAGEMENT HANDBOOKwatt. Efficacy is similar to efficiency, but is expressed indissimilar units. For example, if a 100-watt source produces9000 lumens, then the efficacy is 90 lumens perwatt.ELECTRONIC BALLAST: A ballast that uses semi-conductorcomponents to increase the frequency of fluorescentlamp operation-typically, in the 20-40 kHz range.Smaller inductive components provide the lamp currentcontrol. Fluorescent system efficiency is increased due tohigh frequency lamp operation.ENERGY-SAVING BALLAST: A type of magnetic ballastdesigned so that the components operate more efficiently,cooler and longer than a “standard magnetic”ballast. By US law, standard magnetic ballasts can nolonger be manufactured.ENERGY-SAVING LAMP: A lower wattage lamp, generallyproducing fewer lumens.FLUORESCENT LAMP: A light source consisting of atube filled with argon, along with krypton or other inertgas. When electrical current is applied, the resulting arcemits ultraviolet radiation that excites the phosphorsinside the lamp wall, causing them to radiate visiblelight.FOOTCANDLE (FC): The English unit of measurementof the illuminance (or light level) on a surface. One footcandleis equal to one lumen per square foot.FOOTLAMBERT: English unit of luminance. One footlambertis equal to 1/p candelas per square foot.GLARE: The effect of brightness or differences in brightnesswithin t he visual field sufficiently high to cause annoyance,discomfort or loss of visual performance.HARMONIC: For a distorted waveform, a componentof the wave with a frequency that is an integer multipleof the fundamental.HID: Abbreviation for high intensity discharge. Genericterm describing mercury vapor, metal halide, high pressuresodium, and (informally) low pressure sodiumlight sources and fixtures.HIGH-BAY: Pertains to the type of lighting in an industrialapplication where the ceiling is 20 feet or higher.Also describes the application itself.HIGH OUTPUT (HO): A lamp or ballast designed tooperate at higher currents (800 mA) and produce morelight.HIGH PRESSURE SODIUM LAMP: A high intensitydischarge (HID) lamp whose light is produced by radiationfrom sodium vapor (and mercury).HVAC: Heating, ventilating and air conditioning systems.ILLUMINANCE: A photometric term that quantifieslight incident on a surface or plane. Illuminance is commonlycalled light level. It is expressed as lumens persquare foot (footcandles), or lumens per square meter(lux).INDIRECT GLARE: Glare produced from a reflectivesurface.INSTANT START: A fluorescent circuit that ignites thelamp instantly with a very high starting voltage fromthe ballast. Instant start lamps have single-pin bases.LAMP LUMEN DEPRECIATION FACTOR (LLD): Afactor that represents the reduction of lumen outputover time. The factor is commonly used as a multiplierto the initial lumen rating in illuminance calculations,which compensates for the lumen depreciation. TheLLD factor is a dimensionless value between 0 and 1.LAY-IN-TROFFER: A fluorescent fixture; usually a 2' x4' fixture that sets or “lays” into a specific ceiling grid.LED: Abbreviation for light emitting diode. An illuminationtechnology used for exit signs. Consumes lowwattage and has a rated life of greater than 80 years.LENS: Transparent or translucent medium that altersthe directional characteristics of light passing through it.Usually made of glass or acrylic.LIGHT LOSS FACTOR (LLF): Factors that allow fora lighting system’s operation at less than initial conditions.These factors are used to calculate maintainedlight levels. LLFs are divided into two categories, recoverableand non-recoverable. Examples are lamp lumendepreciation and fixture surface depreciation.LOUVER: Grid type of optical assembly used to controllight distribution from a fixture. Can range from smallcellplastic to the large-cell anodized aluminum louvers

LIGHTING 399used in parabolic fluorescent fixtures.LOW-PRESSURE SODIUM: A low-pressure dischargelamp in which light is produced by radiation from sodiumvapor. Considered a monochromatic light source(most colors are rendered as gray).LUMEN: A unit of light flow, or luminous flux. Thelumen rating of a lamp is a measure of the total lightoutput of the lamp.FIXTURE EFFICIENCY: The ratio of total lumen outputof a fixture and the lumen output of the lamps,expressed as a percentage. For example, if two fixturesuse the same lamps, more light will be emitted from thefixture with the higher efficiency.FIXTURE: A complete lighting unit consisting of a lampor lamps, along with the parts designed to distribute thelight, hold the lamps, and connect the lamps to a powersource. Also called a fixture.MEAN LIGHT OUTPUT: Light output in lumens at40% of the rated life.MERCURY VAPOR LAMP: A type of high intensitydischarge (HID) lamp in which most of the lightis produced by radiation from mercury vapor. Emits ablue-green cast of light. Available in clear and phosphorcoatedlamps.METAL HALIDE: A type of high intensity discharge(HID) lamp in which most of the light is produced byradiation of metal halide and mercury vapors in the arctube. Available in clear and phosphor-coated lamps.OCCUPANCY SENSOR: Control device that turnslights OFF after the space becomes unoccupied. May beultrasonic, infrared or other type.PHOTOCELL: A light sensing device used to control fixturesand dimmers in response to detected light levels.POWER FACTOR: Power factor is a measure of howeffectively a device converts input current and voltageinto useful electric power. Power factor is the ratio ofkW/kVA.RAPID START (RS): The most popular fluorescentlamp/ballast combination used today. This ballastquickly and efficiently preheats lamp cathodes to startthe lamp. Uses a “bi-pin” base.REACTIVE POWER: Power that creates no usefulwork; it results when current is not in phase with voltage.Calculated using the equation:reactive power = V x A x sin φwhere φ is the phase displacement angle.RECESSED: The term used to describe the fixture that isflush mounted into a ceiling.RETROFIT: Refers to upgrading a fixture, room, orbuilding by installing new parts or equipment.ROOT-MEAN-SQUARE (rms): The effective averagevalue of a periodic quantity such as an alternatingcurrent or voltage wave, calculated by averaging thesquared values of the amplitude over one period, andtaking the square root of that average.SPACE CRITERION: A maximum distance that interiorfixtures may be spaced that ensures uniform illuminationon the work plane. The fixture height above thework plane multiplied by the spacing criterion equalsthe center-to-center fixture spacing.SPECULAR: Mirrored or polished surface. The angle ofreflection is equal to the angle of incidence. This worddescribes the finish of the material used in some louversand reflectors.T12 LAMP: Industry standard for a fluorescent lampthat is 12 one-eighths (1.5 inches) in diameter.TANDEM WIRING: A wiring option in which a ballastis shared by two or more fixtures. This reduces labor,materials, and energy costs. Also called “master-slave”wiring.VCP: Abbreviation for visual comfort provability. Arating system for evaluating direct discomfort glare.This method is a subjective evaluation of visual comfortexpressed as the percent of occupants of a space whowill be bothered by direct glare. VCP allows for severalfactors: fixture luminances at different angles of view,fixture size, room size, fixture mounting height, illuminance,and room surface reflectivity. VCP tables are oftenprovided as part of photometric reports for specificfixtures.TOTAL HARMONIC DISTORTION (THD): For currentor voltage, the ratio of a wave’s harmonic contentto its fundamental component, expressed as a percentage.Also called “harmonic factor,” it is a measure of the

400 ENERGY MANAGEMENT HANDBOOKextent to which a waveform is distorted by harmoniccontent.VERY HIGH OUTPUT (VHO): A fluorescent lamp thatoperates at a “ very high” current (1500mA), producingmore light output than a “high output” lamp (800 mA)or standard output lamp (430mA).WATT (W): The unit for measuring electrical power. Itdefines the rate of energy consumption by an electricaldevice when it is in operation. The energy cost ofoperating an electrical device is calculated as its wattagetimes the hours of use. In single phase circuits, itis related to volts and amps by the formula: Volts xAmps x PF = Watts. (Note: For AC circuits, PF must beincluded.)WORK PLANE: The level at which work is done andat which illuminance is specified and measured. For officeapplications, this is typically a horizontal plane 30inches above the floor (desk height).References1. Mills, E., Piette, M., 1993, “Advanced Energy-Efficient LightingSystems: Progress and Potential,” Energy—The International Journal,18 (2) pp. 75.2. Woodroof, E., 1995, A LECO Screening Procedure for Surveyors,Oklahoma State University Press—Thesis, p. 54.3. Bartlett, S., 1993, Energy—The International Journal, 18 p.99.4. Flagello, V, 1995, “Evaluating Cost-Efficient P.I.R. OccupancySensors: A Guide to Proper Application,” Strategic Planning forEnergy and the Environment, 14 (3) p. 52.5. Mills, E., Piette, M., 1993, “Advanced Energy-Efficient LightingSystems: Progress and Potential,” Energy—The International Journal,18 (2), p.90.6. United States EPA Green Lights Program, 1995, Lighting UpgradeManual—Lighting Waste Disposal, p.2.7. United States EPA Green Lights Program, 1995, “Green LightsUpdate” p.2, March 1995.8. R. Rundquist, 1991, “Daylighting controls: Orphan of HVAC design,”ASHRAE Journal,, p. 30, November 1991.9. United States EPA Green Lights Program, 1995, Lighting UpgradeManual—Lighting Fundamentals.

CHAPTER 14ENERGY SYSTEMS MAINTENANCEW.J. KENNEDY, P.E.Department of Industrial EngineeringClemson UniversityClemson, SC 29634-0920Energy systems maintenance is the maintenance ofall systems that use or affect the use of energy. Thesesystems are found in every kind of organization thatuses energy, whether a hospital, a church, a store, auniversity, a warehouse, or a factory. Energy systemsmaintenance includes such routine maintenance tasks aslubrication, examination, and cleaning of electrical contactsand calibrating thermostats, and such non-routinetasks as repainting walls to increase the effective lighting,cleaning fins on condensers, and cleaning damperblades and linkages.A good energy systems maintenance program cansave a company substantial amounts of money in wastedsteam and wasted electricity and in the lost productionand additional expense caused by preventable equipmentbreakdowns. Other benefits include general cleanliness,improved employee morale, and increased safety.In a good maintenance program, planning, scheduling,and monitoring are all carried out in a predictable andwell-organized manner. This chapter is designed to assistthe reader to develop such a program. In addition, aspecial section is included on some recent developmentsand on some of the special problems and opportunitiesassociated with material handling systems.The maintenance part of this chapter is dividedinto three sections. The first section gives a four-stepprocedure for developing a maintenance program forenergy management. This plan relies on a detailedknowledge of the systems within any plant and thecomponents of these systems. The systems are describedand the main components of each system are listed. Thesecond section provides a more detailed description ofthe maintenance of these components and constitutesthe main body of this chapter. The final section describessome of the instruments useful in the maintenance associatedwith energy management.40114.1 DEVELOPINGTHE MAINTENANCE PROGRAMThere are four steps in the development of a maintenanceprogram. Step 1 is to determine the present condition ofthe existing facility. This step includes a detailed examinationof each of the major energy-consuming systems. Theoutput from this examination is a list of the motors, lights,transformers, and other components that make up eachsystem, together with a report of the condition of each.Step 2 is the preparation of a list of routine maintenancetasks with an estimate of the number of times that eachtask must be performed. This list should also include,for each task, the craft, the needed material, the timerequired, and the appropriate equipment. These data arethen incorporated in step 3 into a regular schedule forthe accomplishment of the desired maintenance. Step 4is the monitoring necessary to keep the program in forceonce it has been initiated. These four steps are discussedin more detail in the following sections.14.1.1 Determine the Present Condition of the FacilityThe purpose of this step is to create a startingpoint. The questions to be answered are: (1) Whatequipment and systems are in the building? (2) Whatmaterial is available to describe each system and/orits components? (3) What needs to be done to get theenergy-related systems into working condition and tokeep them that way?This step should incorporate (1) vendor data andoperating specifications for as much of the installedequipment as possible, kept in a notebook, a file cabinet,or a computer data base (depending upon the size andcomplexity of the facility); (2) a diagram of each majorsystem, showing the location of all important equipmentand the direction of all fluid flows; (3) a complete listof all the equipment in the building, showing the name,location, and condition of each item; and (4) a comprehensivelist of maintenance tasks required for each pieceof equipment. This information constitutes an equipmentreference source unique to the equipment in yourfacility. It should be kept current, with each addition

402 ENERGY MANAGEMENT HANDBOOKinitialed and dated, and its location should be knownto all maintenance personnel and their supervisors.The preparation of the notebook and the compilationof the other data is made much easier if separateenergy-related systems are defined within the plantso that the systems can be examined in turn. A suggestedclassification of these systems would constitute(1) the building envelope, including all surfaces of thefacility exposed to the outside; (2) the boiler and steamdistribution system; (3) the heating, ventilating, and airconditioningsystem, together with its controls; (4) theelectrical system; (5) the lights, windows, and adjacentreflective walls, ceilings, and floors; (6) the hot-waterdistribution system; (7) the compressed-air distributionsystem; and (8) the manufacturing system, consisting ofmotors and specialized energy-consuming equipmentused in the creation of products. Each of these systemsshould be inspected, the condition of each part noted,and a diagram drawn if appropriate. The following descriptionsand tables have been prepared to assist youin this inspection.Building EnvelopeThis consists of all parts of the facility that canleak air into or out of any building. Its componentsand appropriate initial maintenance measures are givenin Table 14.1. The envelope should be described by ablueprint of the building, showing locations and compositionsof all outside walls, windows, ceilings, andfloors, and the locations of all outside doors. The primarymalfunction of this system is leakage of air, andthis leakage can often be detected by sight—looking forcracks— or by noting the presence of a draft. Infraredscanning from the outside can also be helpful. The benefitsfrom maintaining this system are (1) a reduction inthe amount of air that must be heated or cooled, and (2)increased comfort due to decreased drafts.Boilers and Steam Distribution Systems(These systems are described in greater detail inChapters 5 and 6). A boiler is often the largest consumerof fuel in a factory or building. Any improvements thatmaintenance can make in its operation are thereforeimmediately reflected in decreased energy consumptionand decreased energy cost. If the steam distribution systemhas leaks or is not properly insulated, these faultscause the boiler to generate more steam than is needed;eliminating these problems saves money. But boilers canmalfunction, and steam leaks can cause severe burns ifmaintenance is performed by untrained personnel. Thefirst step in proper boiler maintenance is usually to getthe boiler and the steam distribution system inspectedby a licensed professional. It is possible, however, toexamine your own boiler and determine whether yoursystem has some of the more conspicuous problems. Toestimate the value of repairs to this system, assume thatany boiler that has not been adjusted for two years canhave its efficiency increased by 25% by a suitable adjustment,with a corresponding decrease in fuel consumptionand in projected fuel costs. In steam distribution,a defective steam trap can typically waste 50,000,000Btu/yr, at a cost of between $100 and $1000, dependingupon the source of fuel. The savings from boilerand steam distribution system maintenance measuresare thus worth pursuing. The system components andan appropriate set of initial maintenance measures aregiven in Table 14.2.Heating, Ventilating, and Air-Conditioning Systems(These systems are described in more detail inChapter 10.) The purposes of these systems are to supplyenough air of the right temperature to keep peoplecomfortable and to exhaust harmful or unpleasant aircontaminants. A complete description of this systemshould include a blueprint with the location of all damp-Table 14.1 Problems and solutions: the building envelope.System Component Problem Initial Maintenance ActionDoor Loose fitting Weatherstripping, new threshold, or frame repairDoes not closeIf problem is caused by air pressure, balance intake andexhaust air; correct door fit; check door-closing hardwareWindow Air leakage Weatherstrip, caulk, or add storm windowBrokenReplace broken panesWall Drafts from wall openings Caulk or seal openings on outside of wallCracksPatch or seal if air is entering or leavingCeiling Drafts around exhaust Caulk or repair flashingpipingRoof Holes Patch or cover

ENERGY SYSTEMS MAINTENANCE 403Table 14.2 Problems and solutions: the boiler and steam distribution system.System Component Problem Initial Maintenance ActionBoiler Inoperable gauges Overhaul boiler controls as soon as possibleMost recent boilerHave boiler adjusted for most efficient firingadjustment at least two years agoScale deposits on waterRemove scale; check water-softening systemside of shutdown boilerBoiler stack temperature Clean tubes and adjust fuel burner (Ref. 1)more than 150°F abovesteam or water temperatureFuel valves leakRepairStack shows black smokeCheck combustion controlsor haze when boileris operatingRust in water gauge Check return line for evidence of corrosion (Ref. 2)Safety valves not checkedHave inspection performed immediatelyor taggedSteam trap Leaks Have inspection performed; repair or replaceSteam valve Leaks RepairSteam line Lines uninsulated Have insulation installedWater hammer notedFix steam trapCondensate return Uninsulated Insulate if hot to touchCondensate tank Steam plumes at tank vents Check and repair leaking steam trapsNo insulationInstall insulationCondensate pumps Excessive noise Check and repairLeaksReplace packing: overhaul or replace pump if necessaryers, fans, and ducts, a complete diagram of the controlsystem showing the location of all gauges, thermostats,valves, and other components, and a list of the correctoperating ranges for each dial and gauge. The descriptivematerial should also include any vendor-suppliedmanuals and the engineering diagrams and reportsprepared when each system was installed or modified.Since many systems have been modified since their installation,it is also desirable to have someone preparea diagram of your system as it is now.Determining the present condition of the systemincludes preparing a diagram of the existing systembased on an actual survey of the system; placing labelson each valve, gauge, and piece of equipment; and examiningeach system component to see if it is working.Table 14.3 gives a list of some of the more expensivetroubles that may be encountered.Significant amounts of money may be saved byproper maintenance of this system. These savings comefrom three sources: reducing the energy used by thesystem and its associated cost, decreasing the amountof unanticipated repair that is necessary in the absenceof good maintenance, and reducing downtime causedwhen the system does not work and conditions becomeeither uncomfortable or unsafe. Since energy maintenanceon this system usually includes maintaining thetemperature established by company policy, and sincecompany policies have increasingly favored a hightemperature threshold for cooling and a low thresholdfor heating, energy maintenance is directly responsiblefor realizing the considerable savings made possible bythese policies. It is not uncommon for a good maintenancepolicy to cause a decrease of 50% in the energyconsumption of a building. In addition to the energycost savings, a good energy maintenance program canspot deterioration of equipment and can enable repairto be scheduled at a time that will not cause extensivedisruption of work. Finally, there have been cases whena buildup of poisonous gas was caused when exhaustfans ceased to function. Such problems can often beavoided with a good maintenance program aimed at theventilation system.The controls for a heating, ventilating, and airconditioningsystem can range from a simple thermostat

404 ENERGY MANAGEMENT HANDBOOKTable 14.3 Problems and solutions: the heating, ventilating, and air-conditioning system.Component Problem Initial Maintenance ActionFilter Excessively dirty Replace or cleanDamper Blocked open or linkage disconnected Check damper controlsLeaks badlyClean and overhaulDuctwork Open joints RepairLoose insulation in duct work Replace and attach firmlyWater leakage or rust spotsRepairCrushedReplaceGrillwork Air flow impossible due to dirt Remove and cleanBlocked by equipmentRemove equipmentFan Motor not hooked up to fan Disconnect motor or install fan beltsExcess noiseCheck bearings, belt tension mountings, dirty bladesInsufficient ventilationCheck fan and surrounding duct work and grill workBelt too tight or too looseAdjust motor mountPulleys misalignedCorrect alignmentPump Hot-water pump is cold Inspect valving; check direction of flow(or vice versa)Blower Not moving air in acceptable Check direction of rotation and change wiring if needed;quantitiesclean if dirtyExcessive noiseCheck bearings, couplerRotation wrong directionCheck wiringShaft does not turn freely by hand Check lubricant; repairChiller Leaks RepairCooling tower Scaling on spray nozzles Remove by chipping or by chemical meansLeaksRepairCold water too warmCheck pumps, fans, and fill; clean louversand fillExcessive water driftCheck drift elimination, metering orifices, and basins (forleaks); check for overpumpingCompressor See Table 14.7Thermostat Temperature reading not accurate CalibrateLeaks water or oil from mounting Check pneumatic control linesto a modern computer-controlled network. The amountof troubleshooting that you can do will be directly proportionalto your knowledge of the system. If you knowa lot about it, you may be able to accomplish a greatdeal, even initially. If your knowledge is limited, getadvice from someone who has installed such systems,obtain all the operating manuals that pertain to yoursystem, and get estimates of the cost needed to bringyour control system up to its designed operation. Youcan save larger amounts of money by simply adjustingcontrols—for example, by using a lower temperature onweekends and when no one is in the facility—than youcan by any other comparable expense of effort. It paysto get your control system in operating condition andto keep it that way.Electrical SystemMany industrial electrical distribution systemsare being used in ways not foreseen by the designers.These changes in use can cause some problems in energyconsumption and in safety. If a motor is operating at alower voltage than it was designed for, it is probablyusing more amperage than was intended and is causingunnecessary losses in transmission lines. If the wires aretoo small for the load, line losses can be large, and firehazards increase significantly. Other problems that cancreate unnecessary energy loss are voltage imbalance inthree-phase motors and leaks from voltage sources toground. 1 Most of these problems are safety hazards aswell as expensive in energy costs, and it is imperativethat they be checked.

ENERGY SYSTEMS MAINTENANCE 405It is desirable to have a qualified electrician orelectrical contractor examine your facility to find safetyproblems. At the same time, a load survey should beperformed to determine whether your wiring, transformers,and switch gear are appropriately sized forthe load they are currently carrying. To supplement thisformal examination, Table 14.4 gives a list of troubleindicators you should look for as you are ascertainingthe current condition of your facility. This list should beused in conjunction with the formal survey.Another problem that may be costing money is alow power factor. The power factor is the ratio betweenthe resistive component of ac power and the total (resistiveand inductive) supplied. Because it costs more toprovide electricity with a low power factor, it is commonfor electrical utilities to make an additional charge if thepower factor is less than some value, typically 75%. Ifyour electrical bill includes a charge for a low powerfactor, it may be worthwhile to have a power-factor meterinstalled and to focus management attention on theproblem of increasing the power factor. Equipment thatcontributes to a low power factor includes welding machines,induction motors, power transformers, electricarc furnaces, and fluorescent light ballasts. A low powerfactor can be corrected with the installation of capacitorsor with a separate power supply for machinery causingthe problem. The maintenance of this equipment thenbecomes another item on the list of scheduled energymanagement maintenance. Power-factor management iscovered in Chapter 11.Lights, Windows, and Reflective SurfacesMaintenance of the lighting adjacent to reflectivewalls, ceilings, and floors serves several purposes inenergy management. The energy consumed by lights issignificant, as is the energy used by the heating, ventilating,and air-conditioning system to remove the heat putinto a building by the lights. But the psychological valuemay have a greater impact. If lights are conspicuouslyabsent from corridors and, where not needed, frommanagement offices, people tend to look upon energyconservation as a program that is being taken seriously,and they begin to take it more seriously themselves.Similarly, a plant where energy management has beenencouraged but where no attention has been paid tolighting is often seen as a plant where energy managementis not taken seriously. (For a more complete discussionof lighting, see Chapter 13.)Many factors modify the effectiveness of the lightingsystem. Of particular importance are the conditionof the lights, the cleanliness of the luminaires, and thecleanliness of the walls, ceiling, and floors. When determiningthe condition of this system, a light meter shouldbe carried as standard equipment. This will show therooms where the IES (Illuminating Engineering Society)or other lighting standards have been exceeded or wherenot enough light is present. There may also be someadditional problems, which are described in Table 14.5.Windows can also be a source of light. If they are usedthis way in your facility, they should be cleaned. If theyare not used as a light source, consider boarding themup to avoid having to remove the heat that they allowin from the sun. The lighting systems in a facility areimportant, both in energy cost and as a morale factor.They deserve your attention.Hot-Water Distribution SystemHot water can have several uses within a facility.It can be used for sterilization, for industrial cleaning,as a source of process heat, or for washing hands. Themaintenance principles for all of these uses are the same,but the temperatures that must be maintained may differ.The purpose of energy systems maintenance in thisarea is to keep the temperatures as low as possible, toprevent leaks, to keep insulation in repair, and to keepheat-transfer surfaces clean. To put this into perspective,Table 14.4 Problems and solutions: the electrical system.Component Problem Initial Maintenance ActionTransformer Leaking oil Have electric company check at onceNot ventilatedInstall ventilation or provide for natural ventilationDirt or grease in transformer and Install air filtering system to insure clean contactscontrol roomWater on control room floorInstall drainage or stop leaks into control roomContact Burned spots Indicates shorting; repair immediately (Ref. 4)Frayed wireMay cause shorting; use tape to secure frayed endsSwitch Sound of arcing, lights flicker Replace

406 ENERGY MANAGEMENT HANDBOOKTable 14.5 Problems and solutions: lights, windows, and reflective surfaces.Component ProblemInitial Maintenance ActionLight Illuminate unused space Remove light and store for later use elsewhereFlickers (fluorescents) Replace quicklyToo little lightIncrease lighting to acceptable levelsBallasts buzzAdjust voltage or change ballast typesSmokingReplace ballast; check contacts and electrical wiring; do not use untilcondition is remediedWall Dirty or greasy CleanPainted with dark paint On next painting use brighter paintFloor Hard to keep clean Examine possibility of changing floor surfaceWindow Dirty Clean, if used for light; otherwise, consider boarding over to preventsolar gain and heat lossa leak of 1 cup/min of water that has been heated from55°F to 180°F uses about 30 million Btu/yr. This is approximatelyequivalent to 30,000 ft 3 of natural gas.Part of the description of this system should includehot-water temperatures throughout the facility. Hot-watertemperatures should be kept at a low level unless thereis some good reason for keeping them high. It is alsounlikely that water needs to be kept hot during weekends.A further area for improved maintenance is in theinsulation of hot-water tanks and lines. If a line or tankis hot to the touch, it should probably be insulated, bothfor the energy saved and for safety. Adding 1 in. of insulationto a water main carrying 150°F water can saveas much as $1.60 per foot per year, depending upon thefuel used as a heat source. Heat-transfer surfaces shouldalso be examined—any fins or radiators that are pluggedup with debris or dirt are causing the hot-water heater toconsume more energy. Table 14.6 gives some additionaltroubleshooting suggestions for the initial maintenance.Air Compressors and the Air Distribution SystemCompressed air usually serves one of three functions:as a control medium, for cleaning, or as a sourceof energy for tools or machines. As a control medium,it serves to regulate various parts of heating, ventilating,and air-conditioning systems. In cleaning, compressedair can be used to dry materials or blow away variouskinds of dirt. And it can be a convenient source of energyfor tools or for various kinds of hydraulic equipment.All three uses are affected badly by line leaks andby poor compressor performance. When a pneumaticcontrol system for building temperature develops leaks,the usual result is that only hot air comes into a room.Thus this kind of leak creates two kinds of energy waste:excess running of the compressor and excess heating. Ifthe air is used for cleaning, the affect of a moderate leakTable 14.6 Problems and solutions: hot-water distribution system.Component Problem Initial Maintenance ActionFaucet Leaks Fix; replace with spring-actuated unitsWater too hot for washing Turn down thermostatPiping Hot to touch Lower water temperature or add insulationLeakingReplaceScale buildupConsider installation of water-softening unitWater storage tank Hot to touch Insulate or lower water temperatureLeaksRepair or replaceElectric boiler Scale buildup Install water-softening unit or start regular flushing operationOn during periods when Install time clocks to regulate useno people are in facilityRadiator Finned surface badly fouled Clean with soap, water, and a brushObstructedRemove obstructions

ENERGY SYSTEMS MAINTENANCE 407Table 14.7 Problems and solutions: air compressors and the air distribution system.Component Problem Initial Maintenance ActionCompressor (Ref. 5) Low suction pressures Look for leaks on low-pressure side; overhaul if necessaryGauges do not workRepair or replaceExcess vibrationCheck mountings and initial installation instructionsCold crankcase heater during Check for lubrication problemscompressor operationLoose wiring or frayed wires RepairLeaks on high-pressure side Examine compressor closely; check gaskets,connections, etc., and replace if necessaryAir line Leaks RepairWater in lineLook for leak at compressorOil in lineLook for lubrication leak in compressormay only be that more air is used; this depends uponthe individual application. If air is used as an energysource for motors or tools, the speed of the motor ortool usually depends upon the air pressure, and anydecrease in this pressure will affect the performanceof the motor. Another thing—controls are particularlyaffected by oil, water, and dust particles in control air,so it is of utmost importance that the compressor bechecked regularly to see that the air cleaner is workingas intended. More detail on compressors can be foundin Chapter 12 of Ref. 5. Table 14.7 gives a guide to theinitial maintenance procedures for air compressors andair distribution systems.The Manufacturing Equipment SystemThis section would not be complete without a discussionof some of the more common equipment usedin manufacturing. Some of this equipment has uniquemaintenance requirements, and these must be obtainedfrom the manufacturer. Much equipment, however, iscommon to many companies or facilities, and this includesmotors, ovens, and time clocks.Motors. Motors can consume excess amounts ofenergy if they are improperly mounted, if they are nothooked up to their load, or, in the case of three-phase motors,if the voltages in the opposing legs are different. Inthis last case, the ASHRAE Equipment Handbook states:With three-phase motors, it is essential that the phasevoltages be balanced. If not, a small voltage imbalancecan produce a greater current unbalance and a muchgreater temperature rise which can result in nuisanceoverload trips or motor failures. Motors should not beoperated where the voltage unbalance is greater than 1%without first consulting the manufacturer. 5Bearings. Bearing wear is one of the more significantfailure types in a motor, and this can be avoided bya procedure devised by Harold Tornberg, a plant managerand former industrial engineer for Safeway Stores,Inc. The Tornberg procedure is to connect a stethoscopeto a decibel meter and to take readings at both ends ofa motor. The reading on the driving end of the motorwill usually be 2 to 3 dB above the reading on the inactiveend. If there is no difference, the bearings on theinactive end probably need to be lubricated or replaced.If the difference is 5 to 6 dB, the bearings on the activeend need to be replaced. If the difference is 7 to 9 dB,the bearing is turning inside the housing, and the motorshould be overhauled as soon as possible. If the differencein readings is more than 9 dB, the motor is on theverge of failure and should be replaced immediately.These procedures were developed on motors in use atSafeway Stores and are in use throughout the Safewaysystem. To adapt these procedures to your facility, buyor make an instrument that allows you to determine thenoise level at a particular point. Then survey the motorsin your facility using the criteria discussed above. Takeone of the motors that the test has indicated as needingrepair into your shop, and tear it down. If you find thatthe failure was worse than these standards indicated,lower the decibel limit needed to indicate a problem. Ifyou find that the failure was not as bad as you thought,raise the standards. In any case, this procedure is onethat can be adapted to your needs to indicate bearingfailures before they occur.Ovens. Ovens use much energy, and many standardoperating practices are available to help decreaseassociated waste. 6 In describing the initial maintenanceactions that must be performed, check the seals, controls,

408 ENERGY MANAGEMENT HANDBOOKrefractory, and insulation. The possible problems areshown in Table 14.8. Correcting these will save energyand improve operating efficiency at the same time.Remember also that heat lost from an oven must beremoved by the air-conditioning system.Time Clocks. Time clocks can be used to significantadvantage in the control of equipment that can beturned off at regular intervals. But two problems mayoccur to eliminate any savings that might otherwise begenerated. First, people can wire around time clocks orotherwise obstruct their operation. Or maintenance personnelcan forget to reset them after power outages. Ifeither of these happens, all the potential energy savingspossible with their use will have been lost. These itemsare included in Table 14.8.14.1.2 Prepare a List of Routine MaintenanceActions with Time Estimates,Materials, and Frequency for EachThis is the second step in the four-step procedurefor developing a maintenance program. The first stepwas to determine the present condition of each systemin the facility, using Tables 14.1 to 14.8 to help locatetrouble spots. The products of this step are: (1) a listof all the equipment in the facility by system; (2) a listof the major one-time maintenance problems associatedwith this equipment; and (3) a notebook with these listsand with the diagrams for each of the major systems.The next major step is to augment the notebook by a listof preventive energy maintenance actions for each systemtogether with an estimate of the materials needed,the time required, and the maintenance frequency foreach piece of equipment. Table 14.9 gives representativeinspection intervals to help you develop your ownprocedures. For convenience, the items of equipment arearranged in alphabetical order, and the maintenance actionsare described only briefly. A more detailed descriptionof the required maintenance for some items is givenin Section 14.2 and necessary instruments are discussedin Section 14.5.Incidentally, the lists described above should bemaintained in a form that maintenance personnel caneasily use. If your personnel are familiar with databasesearching, or if you have a computer system that youhave used, understand, and like, then by all means usea computer system. Such systems have information retrievalcapabilities that can be most helpful, they offerfast access to information, and they are easy to update.They also don't take up much room. On the other hand,they can be a barrier to getting work done. If the softwareis unfamiliar, if the maintenance people don't liketo use it, or if keeping it up to date is difficult, thena manual system is better. The objective is to get themaintenance done, and done right, not to demonstrateanother use for a computer.Time Standards. Accurate time standards formaintenance are difficult to obtain except where themaintenance actions are the same whenever they arerepeated. In this case, predetermined time standardsystems such as MTM and MOST are claimed to work,as are some detailed standards such as those developedby the Navy. 7 But most maintenance actionsinclude troubleshooting. Troubleshooting dependsupon the condition of the equipment, the maintenancehistory, and the skill of the maintenance personnel. Ingeneral, hiring journeymen rather than apprentices isan investment that is well worthwhile. To estimatethe amount of maintenance time that will be needed,consult equipment manufacturers first, then otherusers of the same kind of equipment. Modify theseestimates to include the experience of your personnel,the present condition of the equipment, and the availabilityof necessary repair parts. Then record yourtime estimates, compare these estimates with actualexperience, and revise the estimates to conform withyour actual experience.Table 14.8 Motors, ovens, and time clocks.Component Problem Initial Maintenance ActionMotor Noisy Check bearingsToo hotCheck voltage on both legs of three-phase inputVibratesCheck mountingOven Door gaskets worn ReplaceInsulation or refractory brick missing ReplaceThermostats out of calibrationReplace or recalibrateTime clock Does not work Repair or replaceTime incorrectAdjust clock to correct time

ENERGY SYSTEMS MAINTENANCE 409Table 14.9 Table of preventive maintenance actions.FrequencyAir lines1. Check for leaksBlowersAnnually1. Inspect belts for tension and alignment2. Inspect pulley wheels3. Inspect for dirt and greaseBoilers (see Section 14.2.1)1. Check temperature and pressure Daily2. Clean tubes and other heating surfaces As needed3. Check water gauge glass Daily4. Remove scale Annually5. Perform flue-gas analysis Monthly6. Calibrate controls AnnuallyChillers (Ref. 5)Annually1. Clean condenser and oil cooler2. Calibrate controls3. Check electrical connections4. Inspect valves and bearingsCondensate return system1. Check valves, pumps, and lines AnnuallyCooling coils1. Brush and wash with soap Quarterly or when needed2. Clean drip pan drain When neededCompressors1. Check oil levels Monthly2. Check wiring Annually3. Visual check for leaks Monthly4. Log oil temperature and pressure Monthly5. Remove rust with wire brush Annually6. Replace all drive belts AnnuallyCondenser1. Clean fan Annually2. Brush off coil MonthlyControls1. Calibrate thermostats Semiannually (Ref. 4)2. Get professional inspection of control system According to equipment specifications3. Check gauges to see that readings are in correct range Monthly4. Examine control tubing for leaks MonthlyCooling towers (Ref. 3)1. Inspect for clogging and unusual noise Daily2. Check gear reducer oil Level—weekly. For sludge andwater— monthly3. Inspect for leakage Semiannually4. Tighten loose bolts Semiannually5. Clean suction screen WeeklyDampers1. Check closure Every 6 weeks2. Clean with brush Semiannually

410 ENERGY MANAGEMENT HANDBOOKTable 14.9 (Continued)FrequencyDuctworkSemiannually1. Inspect and refasten loose insulation2. Check and repair leaks3. Inspect for crushed or punctured ducts; repairElectrical system1. Inspect equipment for frayed or burned wiring Semiannually2. Perform electrical load analysis Whenever major equipment changes occur, or every2 yearsFans1. Check fan blades and clean if necessary Semiannually2. Check fan belts for proper tension and wear Monthly3. Check pulleys for wear and alignment Semiannually4. Check for drive noise, loose belts, and excessive SemiannuallyvibrationFaucets1. Check for leaks; replace washers if needed AnnuallyFilters, air1. Replace When dirty, or monthly2. Check for gaps around filters When replacing3. Inspect electrical power equipment (for roll filters; Monthlyfrom Ref. 6)Filter, oil1. Clean and oil Whenever compressor oil is changedGauges1. Check calibration Annually2. Check readings As neededGrillwork1. Remove dirt, grease, bugs Monthly2. Remove obstructions from in front of grill Monthly3. Check air direction Monthly4. Recaulk seams As requiredLeaks1. Check for refrigerant leaks AnnuallyLights, inside1. Perform group relamping See Chapter 132. Perform survey of lighting in actual use Semiannually3. Clean luminaires Office area: every 6 months; laboratories: every2 months; maintenance shops: every6 months; heavy manufacturing areas: every3 months; warehouses: every 12 months4. Replace flickering lights ImmediatelyLights, outside See Chapter 13Motors1. Lubricate bearings When needed2. Check alignment Semiannually3. Check mountings SemiannuallyOvensSemiannually1. Check insulation

ENERGY SYSTEMS MAINTENANCE 411Table 14.9 (Continued)Frequency2. Check controls Annually3. Check firebrick and insulation AnnuallyPiping1. Check mountings Annually2. Check for leaks SemiannuallyPumps1. Check lubrication Monthly2. Examine for leaks SemiannuallySteam traps, (Ref. 8), high pressure (250 psig or more)1. Test Daily-weeklySteam traps, medium pressure (30-250 psig)1. Test Weekly-monthlySteam traps, low pressure1. Test Monthly-annuallySteam traps, all (see Section 14.2.3)1. Take apart and check for dirt and corrosion If test shows problemsThermostats1. Check calibration AnnuallyTime clocks1. Reset After every power outage2. Check and clean AnnuallyTransformers (Ref. 6)1. Inspect gauges and record readings Monthly2. Remove debris Monthly3. Sample and test dielectric Annually4. Remove cover and check for water Every 10 yearsWalls and windows1. Check for air infiltration Semiannually2. Clean Regularly, when dirty14.1.3 Prepare a Maintenance ScheduleSection 14.1.2 gave a list of maintenance actionsneeded for many equipment items, together with approximatemaintenance frequencies. To develop yourown maintenance schedule, first list the equipment youhave in your facility. Then, using the table, estimate themaintenance or inspection frequencies for each unit. Asection of such a table is shown in Table 14.10. Thistable should go into the workbook with manufacturers’manuals described earlier. This master table and theworkbook are used to develop two files: an equipmentfile and a tickler file.The equipment file is a list of cards describingthe maintenance for each unit. A unit may have morethan one card, depending upon the complexity of itsmaintenance. These cards contain troubleshooting directions,a description of how to perform specific maintenanceon each piece of equipment, and a record ofindividual equipment maintenance. The cards shouldbe updated every time maintenance is performed orwhenever a new troubleshooting procedure is devised.Figure 14.1 shows such a card. The tickler file has aseparate card for each day and gives the maintenanceto be performed on that day. This file has two parts.The first part is for actions to be done monthly, quarterly,semiannually, or annually and is arranged bymonths. The second part of the file is for actions to bedone weekly and is arranged by day of the week. Inpreparing this tickler file, items that require the sameskills should be grouped together where possible tosave time and to take advantage of any economies ofscale. (An alternative is to do all tasks in the same areaat the same time.) Figure 14.2 shows an example ofone card from the tickler file. Ideally, all of these files

412 ENERGY MANAGEMENT HANDBOOKshould be on a computer, but seethe note in Section 14.1.2.The final step is to use thetickler and equipment files to set upan operating schedule by assigningeach task to a particular person orcrew.14.1.4 Follow-up and MonitoringManagement action is necessaryto see that the prescribedmaintenance policies are beingfollowed. This step, the fourthin the development of an energymaintenance program, has threeobjectives. First, to ensure thatmaintenance is being performed asscheduled. This objective can be accomplishedby periodic inspectionsof the records and of the action recorded.The second objective is forthe management to get an updateon the condition of the facility soas to anticipate and plan for capitalexpenditures relating to maintenance.This planning requiresknowledge of the current conditionof the plant. The third objective isto update all the files.The energy equipment notebookshould be changed every time a newkind of equipment is installed orwhenever a manufacturer announcesimproved maintenance proceduresfor equipment already on hand.The equipment card file should bechanged at the same time, with abbreviatedversions of any new maintenanceinstructions. The equipmentfile should also be changed wheneversomeone discovers an improved wayof maintaining an item on a card.The tickler file should be changedto reflect the actual needs of thefacility. If, for example, the originalstandard on filter changing was 0.5hr per change, and a new method isdeveloped that enables these to be changed in 0.3 hr, thenew method should be noted on the equipment card fileand the new standard on the tickler card files. Similarly,if the frequency of filter changes was originally listedas every month, and experience shows that 3 months isTable 14.10 Example of equipment maintenance and frequencies.Standard,Equipment Location Action FrequencyAir-handling units Bldg 211 Clean filters 0.7 hr, monthlyClean fan blades 1.2 hr, annuallyCheck ducts1.4 hr, annuallyReplace fan belts 0.6 hr, annuallyCheck belt tension 0.1 hr, monthlyFig. 14.1 Equipment record. (Courtesy of Safeway Stores, Inc.)a better interval, the three months should be notedin the tickler file. By following this procedure, the timestandards and the maintenance frequencies can be keptcurrent, and the maintenance plan can be adapted to eachindividual facility.

ENERGY SYSTEMS MAINTENANCE 41314.2 DETAILED MAINTENANCE PROCEDURESSection 14.1 gave an outline for the development ofa maintenance program for use in energy management.To maintain some of the more complex equipment properly,it is necessary to have more detail than could bepresented in Table 14.9. This section is therefore devotedto a more complete discussion of boilers, pumps, andsteam traps.14.2.1 BoilersMoney spent in proper boiler maintenance is oneof the best investments a company can make. The benefitsare substantial. Suppose, for example, that yourgas bill is $1,500,000 per year and that 90% of this, or$1,350,000, is for your central heating plant. Improvingyour boiler efficiency 5% saves$67,500 per year, in addition toany benefits that you may realizethrough decreased downtime,decreased repair costs, and othernontrivial expenses. To gain thisbenefit requires proper boilermaintenance, either by your ownpersonnel or by a boiler servicecompany hired to provide thisservice on a regular basis. Thefollowing section is designed tohelp you decide what must bedone, how often, and why. Twowarnings must be observed. First,safety precautions must be knownand observed at all times. Makesure that all safety interlocks arefunctioning before doing any workon a boiler. Second, nothing in thefollowing section should be construedas superseding manufacturers’instructions or local safetyand environmental regulations.Keeping these warnings in mind,consider the following problems,their effects, and the best meansof correcting them.The basic reference for thefollowing material is Guidelinesfor Industrial Boiler PerformanceImprovement, published by the EnvironmentalProtection Agency. 9mixed with fuel is directly related to the amount ofexcess air introduced to the boiler, as shown in Figure14.3. The problem of excess O 2 is simply this: If there istoo little air available to combine with fuel, incompletecombustion takes place, the boiler smokes, and hazardsof boiler malfunction increase dramatically (in an oil- orcoal-fired boiler) or the carbon monoxide (CO) concentrationbuilds up (in a gas-fired boiler). If too much air isintroduced, a great deal of the energy in the fuel is usedto heat up the excess air, and the efficiency of the boilerdecreases. These inefficiencies are shown in Figures 14.4and 14.5. The two curves in each figure reflect the rangewithin which the curve of a typical boiler would beexpected to fall. The exact curve for your boiler shouldbe established by a careful test at each of the firing ratesyou anticipate.Excess O 2 in Stack GasThe amount of oxygenFig. 14.2 Tickler file card. (Courtesy of Safeway Stores, Inc.)

414 ENERGY MANAGEMENT HANDBOOKFig. 14.3 Relationship between boiler excess air andstack-gas concentration of excess oxygen. (Adapted fromRef. 9.)The combustion controls on the boiler should beused to decrease the amount of excess air to a pointwhere the amount of excess O 2 represents a compromisebetween the optimal minimum and the point at whichactual waste of energy occurs. The effect of reducingexcess oxygen toward the optimum can be dramatic;if a reduction in excess air from 80% to 20% can beaccompanied by a reduction in stack-gas temperaturefrom 500°F to 400°F, the resulting improvement in boilerefficiency can be 6 to 8% or more. On an annual gas billof $1,000,000, this is a reduction of $60,000 to $80,000 ingas alone.But it is not usually desirable to operate at theminimum indicated on Figures 14.4 and 14.5, for severalreasons. The consequences of exceeding the minimumO2 percentage by a small amount are not as damagingas those of being less than the minimum by a correspondingamount. If the minimum is exceeded slightly,fuel is wasted. If, however, there is insufficient excessair and thus insufficient excess oxygen, smoking or CObuildup can occur, the tubes can become fouled, and theboiler can malfunction. Since there is usually some playin the combustion control system, a setting greater thanthe minimum helps to guard against these problems.Typical values are shown in Table 14.11.The amount of O2 in stack gas can be measuredcontinuously, by a recording gas analyzer, or periodically,by an Orsat analyzer. The advantages of a continuous,mounted unit are that readings are always available toboiler operating personnel for daily boiler maintenanceand for troubleshooting. The cost of an installed gasFig. 14.4 Typical smoke-O 2 characteristic curves for coaloroil-fired industrial boilers. Curve 1, gradual smoke/O 2characteristic; curve 2, steep smoke/O 2 characteristic.(Adapted from Ref. 9.)Fig. 14.5 Typical CO-O 2 , characteristic curves for gasfiredindustrial boilers. Curve 1, gradual CO/O 2 characteristic:curve 2, steep CO/O 2 characteristic. (Adaptedfrom Ref. 9.)analyzing unit will vary from application to application,but typical costs are $200 to $1000. The readings shouldbe logged every hour.Sudden changes in the percentage of excess oxygenfrom hour to hour are unlikely but should be analyzedimmediately if they occur. Some of the more common

ENERGY SYSTEMS MAINTENANCE 415Table 14.11 Typical values for minimum excess O 2 .Typical Minimum Values of Excess O 2Fuel Type at High Firing Rates (%)Natural gas 0.5-3.0Oil fuels 2.0-4.0Pulverized coal 3.0-6.0Coal stoker 4.0-8.0Source: Ref. 9.causes for changes in excess O 2 are tube fouling, changesin fuel or in atmospheric conditions, or damage to thecontrol system. The base for comparison should be establishedby the boiler service personnel or the boilercontractor at the time of initial installation of the boiler,and he or she should be consulted immediately if anyunusual conditions are observed.Excessive Stack-Gas TemperatureThis problem can be caused by water-side or firesidefouling. If fire-side fouling (i.e., a buildup of soot,ash, or other particles) is taking place, the rate of heattransfer from the boiler to the water or steam it is heatingis impeded, and the stack gas is correspondinglyhotter. The effect of soot on fuel consumption is noticeable,according to the American Boiler ManufacturersAssociation (ABMA). 10 Efficiency losses due to soot areapproximately as given in Table 14.12. For this reason,the ABMA recommends that tubes be cleaned onceevery shift where practical. Where such cleaning is notpractical, it is recommended that tubes be cleaned wheneverthe stack temperature rises 75°F.Table 14.12 Typical fuel losses due to soot.Soot Layer onIncrease inHeating Surfaces Fuel Consumption(in.) (%)Source: Ref. 10.1/32 2.51/16 4.41/8 8.5If the fouling is on the water side because of scalebuildup, accumulation of mud or slime, or for someother reason, heat transfer will be impeded as describedabove, and increased stack temperature will result. Theeffect of this scaling is seen in Table 14.13. The detrimentaleffect can be severe.As with excess oxygen, it is advisable to monitorTable 14.13 Effect of water-side scaling.Loss of Heat (%)Thicknessof Scale Soft Hard Hard(in.) Carbonate Carbonate Sulfate1/50 3.5 5.2 3.01/32 7.0 8.3 6.01/25 8.0 9.9 9 01/20 10.0 11.2 11.01/16 12.5 12.6 12.61/11 15.0 14.3 14.31/9 n.a. 16.0 16.0Source: Ref. 10.stack temperature on an hourly basis and to record thereadings. By graphing the readings, it is possible to determinehow much variation is caused by shifting load,when given temperature indicates that something unusualhas happened, and when a gradual temperature riseindicates that it is time to clean the water side or the fireside of the boiler. If soot blowers are being used, the stacktemperature should drop immediately after the tubes arecleaned. If the temperature does not drop, the soot blowersmay not be working or the thermometer may havebecome fouled—in any case, something is wrong.Smoking or Excess COExcess CO, in the case of natural gas, or smokingfor coal or oil fuels, gives an indication that somethinghas changed. Changes in the fuel composition or wearin some component of the burner can cause these problems,or there may be a change in the air supply. In anycase, the problem should be corrected immediately. Astack-gas analyzer can be used to monitor the percentageof CO on a continuous basis.Flame AppearanceThe appearance of the flame can give some valuableinformation. If the pattern is unusual, there may havebeen changes in the burner tips or in other parts of theburner, or there may be a malfunction in a related part ofthe boiler. Also, examining the flame pattern can show ifpart of the boiler is getting overheated. At the same timethe flame is being examined, the inside of the boiler canbe examined as far as possible to see that everything isin order—the stoker (if it is a coal-fired boiler), the refractory,the burners, and so on. The flame check is quickand should not replace the other inspections noted above,but it can provide additional information. Such a checkshould be performed every hour.

416 ENERGY MANAGEMENT HANDBOOKRecord KeepingThe Environmental Protection Agency guidelinesrecommend that a log be kept on each boiler with theinformation shown in Table 14.14. 9 In addition to recordingthis information, plotting it on a graph can giveearly indications of unusual trends or cycles in the data;these patterns can then be incorporated into the generalsystem used to indicate when something is about to gowrong.Maintenance Actions and FrequenciesTable 14.15 gives a list of the most common boilermaintenance actions that need to be performed annually.Table 14.16 gives a checklist of other routine maintenanceitems. If your staff is trained in boiler operationand maintenance this table can be used to help definea pattern of boiler maintenance. If you decide to hireboiler maintenance done by an outside firm, the tablecan help you to determine what they should do and todetermine whether you would gain more from doingthis work with your own personnel.Ultimately, your own maintenance personnel willdo much of the routine maintenance. To be sure theyknow what to expect and how to perform these tasks,have a boiler service representative or contractor traineach person in the more routine kinds of boiler operation,such as reading a sight glass, inspecting the flame,and blowing down the boiler. In particular, they shouldlearn enough about the boiler that they know how tooperate it safely. Keep in mind that many maintenancepeople are undertrained as well as underpaid and thatthey are not familiar with procedures for handlinglive steam. Without proper training, they may also letsight glasses go dry, and they may miss such facts asthat the low-water cutoff valve is not working. Thesemaintenance people can be the most expensive onesyou hire—if their lack of training causes you to lose aboiler.Another note—the given maintenance intervals areaverages. If your boiler uses a great deal of makeupwater, more sludge can develop, and this will have to beremoved by blowing down more often than indicated.If your boiler is a closed system with very few leaks,it is possible that water quality will not be a problem.In that case, the blowdowns can be much less frequent.You must adapt the procedures to your own boiler,preferably with the help of a local professional and thevendor.14.2.2 Package BoilersThe foregoing discussion applies to the operator ofa large boiler with complete controls and one or moreTable 14.14 Boiler information to be logged.General data to establish unit outputSteam flow, pressureSuperheated steam temperature (if applicable)Feedwater temperatureFiring system dataFuel type (in multifuel boilers)Fuel flow rateOil or gas supply pressurePressure at burnersFuel temperatureBurner damper settingsWindbox-to-furnace air pressure differentialOther special system data unique to particular installationAir flow indicationAir preheater inlet gas O 2Stack gas O 2Optional: air flow pen, forced-draft fan damper position,forced-draft fan amperesFlue-gas and air temperatureBoiler outlet gasEconomizer or air heater outlet gasAir temperature to air heaterUnburned combustion indicationCO measurementStack appearanceFlame appearanceAir and flue-gas pressuresForced-draft fan dischargeFurnace boiler outletEconomizer differentialAir heater and gas-side differentialUnusual conditionsSteam leaksAbnormal vibration or noiseEquipment malfunctionsExcessive makeup waterBlowdown operationSoot-blower operationSource: Ref. 9.persons directly responsible for the boiler operation.In many cases, however, needs are met by a packageboiler—a self-contained unit that generally requires littlemaintenance. Depending upon the quantity of fresh waterused annually, water treatment may or may not becritically important. Most of the important maintenanceprocedures are covered in the operating manual thatcomes with the boiler, and this manual should be keptavailable and up to date. Some of the more important

ENERGY SYSTEMS MAINTENANCE 417CheckTable 14.15 Boiler checklist for annual maintenance.Examine for:Safety interlocksOperability—must workBoiler trip circuitsOperabilityBurners1. Oil tip openings Erosion or deposits2. Oil temperatures Must meet manufacturer’s specifications3. Atomizing steam pressure Must meet manufacturer’s specifications4. Burner diffusers Burned or broken, properly located in burner throat5. Oil strainers In place, clean6. Throat refractory In good conditionGas injection system1. Orifices Unobstructed2. Filters and moisture traps In place, clean, and operating3. Burner parts Missing or damagedCoal burners1. Burner components Working properly2. Coal Fires within operating specifications3. Grates Excessive wear4. Stokers Location and operation5. Air dampers Unobstructed, working6. Cinder reinjection system Working, unobstructedCombustion Controls1. Fuel valves Move readily, clean2. Control linkages and dampers Excessive “play”3. Fuel supply inlet pressures on atomizing Meet manufacturer’s specifications steam or air systems4. Controls Smooth response to varying loads5. Gauges Functioning and calibratedFurnace1. Fire-side tube surfaces Soot and fouling2. Soot blowers Operating properly3. Baffling Damaged; gas leaks4. Refractory and insulation Cracks, missing insulation5. Inspection ports CleanWater treatment1. Gauges Working properly2. Blowdown valves Working properly3. Water tanks Sludge4. Water acidity Within specificationsSource: Ref. 9.Table 14.16 Boiler checklist for routine maintenance.ActionCheck safety controlsCheck stack-gas analysisBlow down water in gaugesBlow down sludge from condensate tanksHave water chemistry checkedPerform combustion efficiency check; log resultsCheck and record pressures and readings from boiler gaugesFrequencyDailyDaily or more oftenWeeklyWhenever needed; frequency depends uponamount of makeup water usedQuarterlyDaily or weeklyDaily or weekly

418 ENERGY MANAGEMENT HANDBOOKprocedures for keeping a package boiler running aredescribed below.1. Safety and relief valves should be checked occasionallyto see that they work and that they willreseat properly. They should be checked carefullyto avoid excessive steam loss and scaling. If anysuch valve fails to work properly, the fact shouldbe noted and the valve fixed by a boiler servicerepresentative at the first opportunity.2. Air supply should be kept open so that the boilercan have enough combustion air at all times.Restricting combustion air by blocking the boilerair openings or by blocking all air openings intothe boiler room can create a buildup of carbonmonoxide and/or cause the boiler to operate inefficiently.3. Low-water and high-water gauges and controls shouldbe flushed periodically to remove sludge. If sludgebuilds up in a float-operated valve, the valve can failto operate, with expensive consequences to the boiler.The gauges should be flushed periodically andshould not be allowed to get rusty or clogged.4. Combustion controls should be inspected regularlyand adjusted if sooting or burner wear is takingplace. Sooting causes a great drop in boiler efficiency(see Table 14.12), and burner wear can causeirregular firewall wear, inefficient combustion,and the need for increased boiler maintenance. Itis generally worthwhile to have your combustioncontrols inspected at least annually.5. Tubes can be cleaned if scaling occurs by removinghead plates at either end and running a specialbrush through each tube. For detailed proceduresand necessary safety precautions, see your localboiler service representative. Tubes can be replacedif necessary, but the job calls for special skills andshould be done by someone trained for this job.14.2.3 Steam Traps 8A steam trap is a mechanical device used to removeair, carbon dioxide, and condensed steam and toprevent steam from flowing freely into the condensatereturn system. Steam traps are necessary for severalreasons. If air is not removed from steam, the oxygencan dissolve in low-temperature steam condensate andhelp cause corrosion of the valves, pipes, and coils inthe steam distribution system. If carbon dioxide is notremoved, it can combine with steam condensate toform carbonic acid, another major source of corrosionin the steam distribution system. Air and carbon dioxidealso act as insulators to impede heat transfer fromthe steam; their presence creates a partial pressure thatlowers the steam temperature and the heat-transferrate.Perhaps the main function of steam traps, however,is to permit the removal of steam condensate from asystem while simultaneously preventing the free escapeof steam. In this last function, the energy of the steamis kept within the system, and the amount of the livesteam within the facility is controlled.Steam traps occur as parts of nearly every steamdistribution system. They are often not maintained, andthis lack of maintenance can create a hidden cost that issignificant. The cost of a steam trap failure is dependentupon the failure mode, with a strong dependence uponthe original design. The two main failure modes arefailing open and failing shut, and the design considerationof most maintenance interest is proper drainage.Consider these problems in turn.Problems if Trap Fails OpenIf a trap fails open, live steam flows directly fromthe steam system through the trap, a pressure buildup iscaused in the condensate return system and the condensatereturn lines are heated unnecessarily, or the steamis released directly to the air. In the first case, the backpressure in the condensate lines may cause other steamtraps to fail, and the condensate may not be removedfrom the steam distribution system. In the second case,steam discharging to air costs as much as any other kindof steam leak and creates pressure losses. These costscan be estimated using Table 14.17 and the formula1100 Btu/lb energy cost/million BtuC = waste x ————— x —————————boiler eff. 1,000,000whereC = energy cost per month due tosteam losswaste = number of pounds of steamwasted per month, from Table14.171100 Btu/lb = approximate value for totalheat in saturated steam at 100psiboiler eff. = boiler efficiency, usually 60 to80%energy cost/million Btu = cost that can be obtainedfrom your fuel supplierThese figures are close estimates for steam at 100-psi pressure; for other pressures between 50 and 200 psi,the cost is nearly proportional to the pressure, using 100psi and Table 14.17.

ENERGY SYSTEMS MAINTENANCE 419Table 14.17 Cost of Steam Leaks at 100 psi Assuming aBoiler Efficiency of 80% and Input Energy Cost of $2 perMillion BtuSize of Steam Wasted Total Cost Total CostOrifice (in.) per Month (lb) per Month per Year1/2 835,000 $2,480 $29,7607/16 637,000 1,892 22,7043/8 470,000 1,396 16,7525/16 325,000 965 11,5801/4 210,000 624 7,4883/16 117,000 347 4,1641/8 52,500 156 1,872the condensate return system. Because of the design ofthe bucket, some of the condensate within the bucket isalso emptied at this time, but the bucket is never completelyempty, thus maintaining a water seal over theoutlet and preventing steam from blowing through thetrap. The cycle then repeats.The advantages of this kind of trap are its reliabilityand its general ease of servicing. It can developleaks in the bucket, and pivot wear or the accumulationof dirt can make it unworkable.Problems if Trap Fails ShutIf a steam trap fails shut, it acts like a plug in thatpart of the steam system. Condensate builds up andcools, heat transfer stops, and noncondensable gasesdissolve in the water and create corrosion in the pipesand in the equipment served by the trap. If a blockedtrap causes condensation to build up in a line containingactive steam, the steam may push water ahead of it toform a water hammer. To visualize the effects of waterhammer, think of water at 90 miles/hr colliding withthe inside of a pipe. Severe damage can result.Problems of Improper DrainageIf steam condensate is not drained properly,weight accumulates, heat transfer stops, and freezingmay result. Water weighs 62.4 lb/ft 3 , steam at 100 psiabout 0.26 lb/ft 3 . If a 6-in. pipe is filled with condensaterather than steam, it is carrying 12.25 lb per linear footrather than 0.05 lb. The heat transfer problem has beenpresented earlier. Another problem may be the freezingof pipes, particularly if the system depends upon properoperation of the steam trap to prevent freezing.There are four basic types of steam traps, andeach requires a different troubleshooting procedure. Thefollowing sections show how each trap type operates,what can go wrong with it, and what troubleshootingprocedures are recommended.Open Bucket TrapsIn the trap shown in Figure 14.6, the condensateenters through the top. Since the entering hole is at theside of the bucket, condensate flows into the body ofthe trap, raising the bucket and causing the valve toplug the condensate relief hole. When the bucket hasraised as far as the trap frame will allow, condensatefills up the trap and finally overflows into the bucket,decreasing its buoyancy. When enough water has flowedinto the bucket, it drops, opening the condensate returnvalve. Steam pressure then forces the condensate intoFig. 14.6 Open bucket steam trap. (Courtesy of ArmstrongMachine Works.)Open Bucket Traps, Troubleshooting and Maintenance.The operation of an open bucket trap is characterizedby its regular opening and closing. This createsa noise which can be detected with an industrialstethoscope. When a trap is first installed, the frequencyof clicking should be noted on the body of the trap.If this frequency changes drastically without a correspondingchange in steam pressure, the trap shouldbe taken apart (after isolating it from live steam !) andoverhauled. If the trap is not making noise and is cold,determine whether condensate is getting to the trap.If no condensate or steam is reaching the trap, checkstrainers, valves, and lines upstream of the trap to discoverwhere the blockage is occurring. If condensate isreaching the trap, check the pressure of the steam cominginto the trap. If this pressure is too great, the trapwill not work. If the pressure is correct and the trap isnot working, valve off the trap so that no live steam iscoming from either the steam distribution system or thecondensate return and take off the cover or remove thetrap from the line. Look for dirt, worn parts, or worn

420 ENERGY MANAGEMENT HANDBOOKor plugged orifices. It is also possible that some fluctuationsin steam pressure caused the trap to lose thesmall amount of water in the bucket necessary for itsoperation (the priming). This condition can be detectedif the trap is blowing live steam. Close the inlet valvefor a few minutes to let condensate build up, then letthis condensate into the trap and it should work.Another problem that this trap can have is causedby the need to have the priming water in the trap at alltimes. The trap is automatically vulnerable to freezingif the external temperature gets cold enough and if thesupply of steam is shut off. If such freezing occurs, manytraps have provisions for the admission of live steam tothaw the ice.Note: There is a difference between flash steam andlive steam. Flash steam is formed when condensate at ahigh temperature is exposed to air at a lower temperature,and part of it “flashes” to steam. Flash steam hasvery little pressure and gives an irregular, undirectedflow pattern. Live steam however, is steam under pressureand has a well-defined, consistent pattern. The differenceis significant, since the presence of flash steamis expected, whereas live steam gives an indication thatsomething is wrong.Inverted Bucket TrapsThe inverted bucket trap (Figure 14.7) works likethe open bucket trap described above, with two exceptions.First, the trap is opened and closed by steamlifting a bucket rather than by condensate lowering it.Second, the design of the trap is self-cleaning—dirt isscoured from the bottom of the trap automatically as aresult of the trap design. Thus there is less likelihoodthat the trap will become plugged up with dirt, althougha large dirt particle could still keep the bucket lifted upor the valve jammed open. Both of these problems canbe alleviated by installing and maintaining a strainerupstream of the trap. This trap is maintained using theprocedures outlined for the maintenance of the buckettrap described above. This trap can also freeze or loseits prime.Float and Thermostatic TrapsThe float and thermostatic (F&T) steam trap isoften used where steam pressures vary over a widerange and where gravity may be the only force availableto remove condensate. This trap, shown in Figure 14.8,has a float-controlled valve to let condensate out anda thermostatically controlled valve, shut at condensatetemperatures but open at lower temperatures to let outair and other nondissolved gases. It can be used wherecontinuous condensate flow is desirable, unlike thebucket traps, which yield only intermittent flow.Float and Thermostatic Traps, Troubleshootingand Maintenance Procedures. In case of no steam flow,the float may have developed a leak or have collapsed,or the air valve may be plugged. The mechanism maybe worn and in need of replacement. If the trap isblowing live steam, the air hole may be blocked openbecause of wear or by a piece of scale, or the trap maybe filled with dirt. To check any of these, valve off thetrap and examine or remove the trap. Freezing can alsobe a problem.Fig. 14.7 Inverted bucket steam trap. (Courtesy of ArmstrongMachine Works)Fig. 14.8 Float and thermostatic steam trap. (Courtesy ofArmstrong Machine Works.)

ENERGY SYSTEMS MAINTENANCE 421Disk TrapsAnother common type of steam trap is the disktrap, one example of which is shown in Figure 14.9.When condensate or air enters from the left, it pushes upthe disk and flows out through the passage on the right.When steam enters the trap, it passes around the edgeof the disk and creates a downward pressure on the topof the trap. When this pressure is great enough, the diskdrops and the trap closes. When the trap closes, the diskis no longer heated by the steam, and the steam abovethe trap cools, giving lower pressure against the top ofthe disk. When this pressure is lower than the pressurefrom underneath, the trap opens and the process repeatsitself.Disk Traps, Troubleshooting and MaintenanceProcedures. The contact area between the disk andits seat can become corroded or blocked, and this cancause a problem. Other problems may include installationbackward to the proper direction of steam flow. Tomaintain this trap, first check the installation to see if themanufacturer’s directions have been followed. Then, ifinstallation is correct, valve off the trap from live steam,remove the top, and check on the condition of the diskand its contact surface. Unlike the bucket traps, this traphas no prime, so losing its prime is not a problem.Thermostatic TrapsThis trap operates on the general principles illustratedin Figure 14.10. In this trap, the bellows orbimetallic strips are extended when hot and contractedwhen cold. Since condensate is cold (relative to steam),its pressure causes the tap to contract, opening the outletand allowing condensate and noncondensable gasessuch as air and carbon dioxide to escape. When thesehave left the trap, steam comes in and heats up the bellowsor bimetallic strip. This expands, closing the outlet.As steam condenses, the quantity of condensate buildsup, the valve opens again, and the cycle repeats. Thistrap can be used to give a continuous flow of condensateif desired.Thermostatic Traps, Troubleshooting and Maintenance.Depending upon the mounting of the trap,dirt and scale may not be a problem in the outlet. Ifthe bellows leaks and fills with condensate, the trapmay fail instead. If a bimetallic strip is used instead ofbellows, corrosion can be a problem. Like most othertraps, failure can be caused in this trap if scale or dirtgets between the valve and its seat. Also, the dirt screencan get plugged up, causing the trap to fail closed.Steam Distribution System, Maintenance FrequencyThe steam distribution system, including piping,steam-heated equipment, steam traps, and the condensatereturn system, should be checked regularly for leaksand to ensure correct operation. The frequency of inspectionshould vary with the age of the installation—in anew system, dirt and other foreign matter in the pipeswill be deposited in strainers, steam traps, and the bottomof steam-heated equipment. After the installationof a new system, it thus makes sense to check steamtraps, strainers, and drains frequently, perhaps every 3months. As the system gets older, scale can become aproblem if the water is not properly treated. If scaling isa problem, the steam distribution system will have to bemaintained more often than without scaling—semiannuallyrather than annually, for example, on steam trapinspection. The entire steam system should be inspectedcarefully at least once each year for leaks and to keepsteam traps and steam-heated equipment operating correctly.When this inspection and maintenance procedureis followed regularly, the amount of steam needed willbe reduced, and the cost savings will be substantial.14.3 MATERIALS HANDLING MAINTENANCEMuch energy is used in materials handling. Propermaintenance can reduce this amount of energy in theoperation of lift trucks, conveyors, and the ancillary bat-Fig. 14.9 Disk-type steam trap. (Courtesy of ArmstrongMachine Works.)Fig. 14.10 Thermostatic steam trap (bimetal).

422 ENERGY MANAGEMENT HANDBOOKteries, motors, fans, and tires. The proper maintenanceactions are described in great detail in the operating andmaintenance manuals of equipment manufacturers, andmuch of this material is condensed from those publications.For detailed instructions on the maintenance of aparticular equipment item, refer to the manual for thatitem. The material in this chapter can, however, be usedto develop check-off lists and to give a good idea of thetypes and amounts of maintenance required.14.3.1 Electric Lift Trucks 11,12The general maintenance principles that apply toan electric lift truck are (1) keep the battery charged, (2)make sure that all fluid levels are where they should be,(3) check the brakes and keep them working correctly,(4) inspect the electrical equipment and keep it running,and (5) lubricate the truck according to manufacturer’sspecifications as follows.BatteryThe battery should be inspected at least weeklywith specific attention paid to the specific gravity, thecontacts, and the electrolyte level. The specific gravityshould be between 1.250 and 1.275 and should not varymore than 0.020 between cells. Special care must betaken to see that the charger is connected up with correctpolarity and the correct voltage, and smoking shouldbe prohibited near the charging so as not to ignite thehydrogen gas liberated during charging. If the batteryfails to hold a charge, a test with a hydrometer can oftendetect whether a particular cell is at fault; if no cell isclearly bad, have a load test performed by someone whoknows the necessary safety precautions.The battery contacts should be inspected daily forcorrosion. At the same time, the wires leading to the contactsshould also be inspected for cracked insulation andfor other signs of wear. The level of electrolyte shouldbe inspected daily and any deficiencies made up usingdistilled water.Hydraulic ComponentsThe level of hydraulic oil should be inspected dailyand filled if it is below the lower mark on the dipstick.It should be changed at least every 2000 hours. As thedaily hydraulic level inspection is being performed, theoperator should also check all hydraulic lines for leaksand fix them (or have them fixed) immediately.The brakes should also be checked daily to see thatthe shoes are not dragging and that the brake pedal hasthe right amount of play and resistance to foot pressure.Brake shoes should be inspected and adjusted at leastonce a year.TiresIf the tires on a particular lift truck are pneumatic,it is important that they be kept inflated to the pressuresrecommended in the operating manual and thatthe inflation be checked at least weekly Underinflationcan lead to excessive wear. Underinflation also increaseselectrical consumption. If the tires are made of hard rubber,they should be inspected daily so that the operatorknows their condition. If a tire has large chunks out ofit, it may constitute a safety hazard, and the irregularride that this problem generates may be damaging tothe material being transported. If tires show irregularwear, the front end may be out of line—this should bechecked and fixed.General LubricationGeneral lubrication should follow the scheduleand procedures laid out in the operation manual for theparticular lift truck being maintained. This lubricationshould also include inspection, adjustment, and oilingof the hoist chain.14.3.2 Lift Trucks, Non-electricThe main points to maintain on a lift truck poweredby diesel, gasoline, or LPG are the engine, thehydraulic system, the transmission, and the brakingsystem. Batteries are important, but their care is not ascentral to the maintenance of these lift trucks as it is tothe maintenance of electric-powered trucks.EngineSpecial attention should be paid to the oil level,the radiator, and fan belts. The oil level should beinspected daily, and the oil and oil filter should bechanged every 150 hours (unless the environment isvery dirty or muddy, in which case the oil shouldbe changed more often). The oil level on the dipstickshould be between the high and low marks—if it istoo low, more should be added. At the same time thischeck is being made, the operator should look for anyoil leaks. The operator should also try to keep track ofthe frequency of adding oil as warning of unseen leaks.The coolant level in the radiator should be checked beforeeach day’s work, either by looking at the coolantlevel in the recovery bottle or by carefully removing theradiator cap to check the level there. Fan belts shouldbe checked periodically. If they are too tight, bearingwear will be excessive; if they are too loose, insufficientcooling and/or generator power will take place. Moredetails on the maintenance of diesel engines are givenin Section 14.4.

ENERGY SYSTEMS MAINTENANCE 423The Hydraulic SystemIf the hydraulic system fails on a lift truck, peoplecan get hurt. Any hydraulic pumps that are run withoutfluid can burn out, causing a significant and usuallyavoidable expense. The way that such an expense canbe avoided is by careful checking of the hydraulic levelsdaily and by daily inspection of all hydraulic connectionsfor leaks.TransmissionThe fluid level in the transmission should also bechecked daily. This minor inconvenience can save thecost of a new transmission and the loss of the vehicleduring the repair time.BrakesPart of the daily check should be a quick checkof the foot brake and the parking brake. The footbrakepedal should move a short distance—1/4 to l/2 in.,depending upon the truck model—before the brakeengages, and the brake should respond firmly to pressure.The parking brake should be checked to see if itprovides the braking necessary to act as a backup forthe foot brakes. Air brakes should have the drain cockopened until no more water escapes with the air.14.3.3 Conveyor SystemsConveyor belts, in-floor towlines, and similarchain- or belt-driven equipment have three importantareas where proper maintenance can save energy andmoney. These are in controls, drive motors and gear,and the actual moving belt or line.Conveyor ControlsOne way to save energy on controls is to install andmaintain controls that cause the conveyor to move onlywhen there is material to be moved. This measure hassaved significant amounts of energy in coal mines andin pneumatic conveying systems, for example, wherein one case the total energy consumed was reduced by90%. The additional wear and tear on starting motorswas not found to be significant when these motors weremaintained according to the original procedures specifiedat the time of purchase of the equipment. Addinga load-related starting control saves on conveyor wearand on each part of the equipment that is no longeroperating continuously.Drive Motors and GearsDrive motors should be checked regularly usingthe Tornberg procedure for motors described in Section14.1.1. When the motor is being inspected, the rest of thedriving gear should also be inspected, with the detailsbeing dependent upon the power transmission systembeing used by the conveyor.Conveyor Belts or LinesThe belt, towline, or other equipment used tomove the material should be checked regularly. Thekinds of problem to include in the inspection are wornbelts; noisy rollers, indicating possible bearing problems;belts that are too tight or too loose; and any place thatshows conspicuous wear. Such an inspection should beperformed regularly and coupled with regular lubricationof appropriate points. Manufacturing specificationsshould be consulted as a source for specific lubricationdirections for all conveying equipment.14.4 TRUCK OPERATION AND MAINTENANCE 13Most organizations use trucks in some form,whether for coal haulage, parcel delivery, or earth moving.Wherever trucks are used, there are opportunitiesfor saving energy and money. The three general rules forachieving these savings are: (1) match the equipment tothe load, (2) keep the equipment tuned up and in goodrepair, and (3) turn it off if you do not need to have itrunning. The following section gives more details onthese rules as they relate to general operation and tothe operation of diesel and gasoline engines. The informationpresented here has been distilled from manualsproduced by Cummins, Caterpillar, Detroit Diesel, andHyster.14.4.1 Management DecisionsThe person supervising a company with a vehiclefleet can have a large impact upon the costs of runningthat fleet in at least five different areas: (1) matching thevehicle to the job, (2) maintenance of the vehicle yard,(3) instituting a system of fuel accountability, (4) takingadvantage of unavoidable delays to create more efficientoperation, and (5) keeping all vehicles tuned up.Matching the Vehicle to the JobThis takes two forms. First is the policy that assignssmall passenger or pickups to personnel on onepersontrips; delivery vans for local deliveries, where thedemand calls for more capacity than a pickup provides;and semi-trucks only where justified by the amount ofmaterial to be moved. Within each category the samerationale is applied—use the larger vehicles only whentheir use is necessitated by the load. Once the vehicleshave been chosen or assigned based on their intended

424 ENERGY MANAGEMENT HANDBOOKuse, the second rule can come into play—if you do notneed all the power of the vehicle now, see if it can bemodified so that the maximum power used is lowered.This can happen where large earth-moving or transportingequipment has been designed for a 10% grade andwhere actual conditions do not have more than a 7%grade. If the actual requirement is less than the equipmentwas designed for, the governor or the fuel pumpcan be modified to limit the horsepower available to theoperator.Maintaining the Vehicle YardMud resists vehicle travel and makes maintenanceinto a chore. The effect of mud, whether it is on a haulroad or in a vehicle yard, can be seen in Figure 14.11. If asignificant part of your fleet travel is along roads underyour maintenance or in a yard you control, this figurecan help you decide whether to surface the road or yardor how often to grade the road. Particularly where yourvehicles have 30% or more of their travel off the road,the savings can be substantial.Instituting Fuel AccountabilityIf your company does not have a system accountingfor the use of every gallon of fuel, you probablycould benefit from instituting such a program. Theprogram has two benefits: It enables you to determinewhich of your vehicles is using the most fuel, and itdecreases losses to employee vehicles. If you can locatethe vehicles that are least efficient, you can start theprocess to find out whether something is wrong with thevehicle, whether some operating procedures need to bechanged, or whether the particular design is inefficient.In any case, you have a better idea of where your vehiclefuel is going.Taking Advantage of BottlenecksIf your facility has certain inherent bottlenecks,such as a power shovel that has less capacity than thetrucks serving it, it may be possible to have these workto your advantage. In the example of the shovel, fuelcan be saved by slowing down on the return trip fromthe dumping area, using the time that would be wastedwaiting for the shovel in the additional time along thehaul road. Make adjustments to the vehicle engine ifoperators are unwilling to conform to these practices.Keep Vehicles Tuned UpWhen a 1% improvement in vehicle efficiency cancreate a savings of $1 per day for each vehicle, it makessense to keep vehicles tuned up. Keeping a vehicle fleet intune saves three ways: in energy consumption, in frequencyof downtime and the repairs involved, and in the cost ofrepairs that do have to be made. The tune-up procedurefor each vehicle is prescribed by the manufacturer.14.4.2 Daily Maintenance PracticesIn addition to the general maintenance practices describedabove, daily procedures performed by the operatorscan create substantial savings in fuel and operating costs. Inaddition to the mandatory inspection of such safety itemsas brakes, the horn, and lights (if appropriate), a daily inspectionshould be made of tires, batteries, air hoses, fueland lubricant levels, and air filters.TiresProper inflation of tires reduces rolling resistance andthus reduces energy consumption. Figure 14.11 shows atypical curve relating inflation and fuel efficiency. In additionto the fuel savings, proper inflation increases tirelife and decreases the total cost per mile ofrunning each vehicle.BatteriesIf one cell of a battery is weaker than theother cells, the generator uses more powerthan if all cells were in good condition. Thisadditional power cuts into fuel efficiency. Toavoid this problem, the water level in eachcell should be checked daily and filled withdistilled water whenever the level is belowthe prescribed fill mark, and the batteryvoltages should be checked frequently.Fig. 14.11 Gasoline consumption as a function of rolling resistance.Air HosesThese should be checked for leaks as soonas they are under pressure. Air leaks cause

ENERGY SYSTEMS MAINTENANCE 425the air compressor to use more power than necessary andcreate another waste of fuel.Fuel and Lubricant LevelsIt is generally recommended that fuel tanks not befilled to more than 95% of the tank capacity. The extra 5% isto allow for expansion of the fuel as it heats up. If the tanksare filled to the top, this expansion can cause an overflowof fuel, a waste and possible safety hazard. Lubricant levelsshould be checked routinely to detect leaks and to keepproper lubrication in the engine, transmission, and othersystems.Air FiltersAn engine uses many gallons of air for every gallonof fuel burned, so keeping the air clean is critical. If the airfilter gets clogged, the engine must use additional amountsof energy to overcome the resistance and thus extra fuelis consumed; if the air filter gets too dirty, the engine willnot run. Also, particles of the filter or dirt particles may bedrawn into various parts of the engine, causing unnecessarywear.14.4.3 Operating PracticesIn addition to the daily maintenance check describedabove, each operator can do several other things to decreasethe fuel consumption of his or her vehicle. These measuresinclude eliminating unnecessary idling, warming the enginecorrectly, choosing the most economical speeds for a givenload, keeping the hydraulic system above stall speed, andkeeping the cooling temperature within specified limits. Thefollowing paragraphs discuss these actions in more detail.with a reputable dealer to find whether this warm-upperiod is necessary.Choosing the Most Economical SpeedEvery vehicle operator should know the most economicalrpm range to achieve good fuel economy whiledelivering the desired horsepower. The best rpm rangecan be determined from performance curves, such asFigure 14.12, and from the fuel consumption map, suchas Figure 14.13. If these curves describe the engine operatedby a particular operator, that person should knowthat operating at 1700 rpm rather than 2300 rpm savesapproximately 8% in fuel. Similarly, mechanical driveequipment should be operated in the highest gear practical.Since the economical operation of a vehicle fleetmay enable the company to stay in business, saving thisamount of fuel should be important to any operator.Operating the Hydraulic System at the Right SpeedCare should be taken to operate the hydraulicsystem with enough power that the pumps do not stall.Stalling a hydraulic pump creates heat but does not getwork done. Similarly, if operating a converter, the operatorshould downshift to get more power rather thanallow the converter to stall.Maintaining Recommended Coolant TemperaturesThe operator should keep the coolant temperaturein the range recommended by the manufacturers inorder to achieve maximum fuel economy.Eliminating Unnecessary IdlingThe three largest makers of diesels are unanimous incondemning unnecessary idling as a waste of energy andas hard on the engine, although some idling is necessary.If equipment will be idled for more than 10 to 15minutes, the cost for shutdown and restarting the enginewill generally be less than the cost of fuel for idling. Theengine should not be shut down immediately after a hardrun, however, but should be allowed to idle for a few minutesto allow engine temperatures to drop. Otherwise, asurge of hot coolant to the heads can cause damage. Also,turbocharger damage can occur if coolant oil and air floware shut off too abruptly.Warming Up the EngineWhen starting, the engine should be warmed sothat the coolant is in the recommended operating rangebefore putting a load on the vehicle. This is recommendedfor diesel engines. For gasoline engines, checkFig. 14.12 Typical diesel performance curve. (Courtesy ofCummins Engine Company.)

426 ENERGY MANAGEMENT HANDBOOK14.5.1 Electrical MeasurementsElectrical measurements can be used to assessthe condition of individual items of equipment and toanalyze the energy consumption patterns of the entirefacility. Troubleshooting and monitoring of individualunits can be accomplished with portable wattmeters,multimeters, and power-factor meters. Many of theseinstruments are available in recording models alsofor permanent installations for monitoring changes inoverall consumption. Safety precautions must be knownand carefully observed in the installation and use of allequipment, particularly that involving electricity.MultimetersA multimeter is capable of measuring voltage, current,and resistance. These meters come in many ranges.One of the more common types is the clip-on metercapable of measuring 0 to 300 A, 0 to 600 V, and 0 to1000 Q. This meter typically sells for approximately $50and is easily carried in a belt holster. It can be used tocheck many parts of the electrical system and to makeelectrical checks on motors.Fig. 14.13 Fuel consumption map. (Courtesy of CumminsEngine Company.)14.5 MEASURING INSTRUMENTSInstruments serve two functions in maintenance—checking equipment condition and on-line monitoring.It is often impossible to know whether some equipmentis functioning properly without measuring a temperature,pressure, or other parameter associated with theoperation of the equipment. For example, the only indicationthat a steam trap is not working properly maybe the absence of a regular clicking as the trap opensand closes, and this noise may be detectable only witha stethoscope. A similar use of instruments is checkingon the effectiveness of repairs. Another function ofinstrumentation is to observe major system operatingparameters so that equipment can be operated with theminimum effective use of energy. Instrumentation forthis purpose is generally mounted permanently, withgauges readable either at the equipment or at somecentral location.Measuring instruments used in maintenance canbe classified by type or by function—by type accordingto what they measure, by function according to whatsystem they serve. A classification of instruments intothese categories is shown in Table 14.18, followed bydescriptions of each instrument in the table.Power-Factor MeterThis meter is used to determine the ratio betweenthe resistive component and the total (resistive andinductive) electric power supplied. It can be used tocheck the power factor on equipment and to determinewhether a power-factor-improvement program is working.Recording AmmeterThis records the electrical consumption over a prescribedperiod. Its most important function is to enablesomeone to determine the magnitude and timing of peakload demand and the magnitude of the underlying baseload (the equipment that is on all the time). A recordingammeter can also help determine the effectiveness of anongoing energy management program. It can be used toassist operating personnel in shifting peak loads and indetermining the magnitude and timing of any secondarypeaks. Such a meter, installed, costs $200 to $1000 andcan provide a payback period of one to two months.Similar meters are available to record watts, volts, andpower factor.WattmeterA wattmeter is used to determine directly theamount of power used by a piece of equipment or by afacility. It can be used to analyze electrical consumptionby enabling maintenance personnel to first determinethe total power being used in an area, then find what

ENERGY SYSTEMS MAINTENANCE 427Table 14.18 Instruments for use in energy management.System Instrument Portable Mounted Parameter MeasuredBuilding envelope Infrared photography × Heat lossBoilersSee Chapter 5 andSection 14.5Steam traps Thermometer (sensitive) × Temperature differences betweeninlet and outletStethoscope × Opening and closing noiseHeating, ventilating, Flow hood × Air flow ratesand air conditioning Pitot tube × Air flow ratesInclined-tube manometer × Pressure differential between twopointsBourdon tube × × PressurePocket thermometer × Temperature of rooms, pipes, etc.Orifice flowmeters × Flow rates of air or steamPsychrometer × × HumidityElectrical Ammeter, recording × Electricity usage, peaksAmmeter, clip-on, with × Voltage, current, resistanceprobesWattmeter, recording × Power consumptionPower-factor meter × × Electrical power factorLighting Industrial lightmeter × Light levelsHot water Thermometer × × TemperatureAir compressors Pressure gauges × Oil pressure, air pressureStethoscope with gauge × Bearing wear on motorsMultimeters × Voltage balance on three-phasemotorsStrobe light × Motor vibrationInfrared film and camera × Bearing wearequipment is using the power, then turn off machinesand shift loads from peak periods to off-peak periods.14.5.2 Light MeasurementsReducing unnecessary lighting has psychologicalimpacts that can far exceed any direct impact on energyconsumption. To convince people that their areasare overlit or underlit, however, it is necessary to havewell-established lighting standards and a way to compareexisting conditions against those standards. Thestandards have been provided by the Illuminating EngineeringSociety in the IES Handbook and are discussedin Chapter 13 of this book.Industrial Light MeterThis instrument is typically designed to measureincide nt light directly and can be used to measurereflected light and light transmittance as well. An industriallight meter differs from a photographic lightmeter in two ways: (1) it reads directly in footcandlesin scales that are approximately linear rather than thelogarithmic scales used in photographic light meters; (2)it is designed to receive light from wide angles ratherthan focusing on a particular part of the viewing field.Such a meter is small, light, and inexpensive, and sufficientlyuseful that it should be part of the equipmentof every maintenance supervisor.14.5.3 Pressure MeasurementsIf equipment is not operating in its proper pressurerange, it can be damaged or the equipmen t it servescan be damaged. This is true whether the pressurebeing measured comes from oil (compressors or lifttrucks), fuel (boilers), steam (boilers), or other sources.Measuring pressure using portable equipment generallyrequires that the equipment to be measured be equippedwith a fitting specifically designed for pressure-sensingequipment. With such fittings in place, it is possible touse a Bourdon tube or a diaphragm gauge for routineinspections. Where information is to be available at

428 ENERGY MANAGEMENT HANDBOOKany time, or when the readings are to be taken so oftenthat portable pressure-sensing equipment becomes anuisance, it may be desirable to permanently install aBourdon tube or a diaphragm gauge, or to rely on amanometer.Bourdon GaugeThis common pressure gauge consists of a curvedtube closed at one end with the other end connected tothe pressure to be measured. When the pressure insidethe tube is greater than the pressure outside the tube,the tube tends to straighten, and the amount of changein length or curvature can be translated directly into agauge reading. Su ch gauges are available in many pressureranges and accuracies.Diaphragm GaugeIf the pressure inside a bellows or on one side of adiaphragm is greater than the pressure outside the bellowsor on the other side of the diaphragm, the bellowsor the diaphragm will move. The amount of movementis related to the pressure between the inside and outsidethe bellows (or on the different sides of the diaphragm).These pressure gauges are also very common.ManometerIf a glass tube has liquid in it, and if one end isopen to the air and the other end is exposed to a pressureother than air pressure, the end with the higherpressure will have a lower liquid level. This kind ofgauge can be mount ed across a filter bank to indicatewhen a clogged filter is causing pressure to build up,or it can be mounted in a place where a much higherpressure is to be measured. If the tube is inclined, asmaller pressure difference is detected, as in the inclined-tubemanometer. These gauges are easy to read,easy to maintain (the glass must be kept clean, and theinlet and outlet holes must be kept clear of debris), andinexpensive.14.5.4 Stack-Gas AnalysisAs explained in Section 14.2.1, proper maintenanceof boilers depends heavily on knowledge of stack-gascomposition. If there is too much molecular oxygen,the boiler is operating inefficiently; if there is too muchcarbon monoxide or too much smoke, the boiler is operatinginefficiently and creating an operating hazard.Thus it is important to keep track of O 2 , CO, and smoke.Three types of monitoring equipment meet these needs:Orsat kits, permanently mounted meters, and smokedetectors.Orsat ApparatusThe Orsat analysis kit consists of three tubes filledwith potassium hydroxide, potassium pyrogallate, andcuprous chloride, respectively. Flue gas is introducedinto all three tubes, and the amount of carbon dioxide,oxygen, and carbon monoxide removed in the first,second, and third tubes, respectively, indicates the proportion of those gases in the flue gas. Any gas remainingis assumed to be nitrogen. This apparatus is capable ofaccurate readings when representative samples of fluegas have been taken from different points in the fluestream, when the apparatus does not leak, and whenthe operator is well trained in its use. 14Smoke DetectorsSmoke detectors work by comparing the amountof light going through a sample of smoke with somestandard shades. If a Ringlemann scale is used, thesmoke number is between 1 and 4, if the Bacharachscale, between 1 and 9. Smoke detectors can be eitherportable or permanently mounted.14.5.5 Temperature MeasurementTemperature measurements are essential to energymanagement and proper maintenance in at least foursituations: for comfort, to determine where heat is leakingfrom a building, to define abnormally hot areas ina machine, and to use in the analysis of boiler operationsand of industrial operations using process heat.Temperatures in these situations call for instrumentssuch as a pocket thermometer, infrared photography,and permanently mounted devices possibly using thermocouples.Pocket ThermometersEvery person who must set thermostats in abuilding and listen to the complaints of uncomfortablepeople needs one of these. It is possible to get a ruggedthermometer, small enough to fit into a shirtpocket and accurate to within ± 1/2°F between 0 and220°F, for $12 to $16. Such a thermometer can b e usedto calibrate thermostats, and to check complaints. Byplacing the heat-sensing end on a hot pipe, it is possibleto estimate the temperature of the material in the pipeor to determine that it is beyond the temperature rangeof the thermometer. This thermometer can also be usedto check the temperature of water coming from variouspoints of a culinary hot-water system. The thermometeris small and inexpensive, and its value as a tool in energymanagement mak es it indispensable.

ENERGY SYSTEMS MAINTENANCE 429Infrared EquipmentThis equipment works by sensing infrared radiation,a kind of radiation given off by warm materials.Infrared equipment, whether photographic or usinga television-like device, senses differences in temperature.It can therefore detect heat leaks that are notrevealed by a visual inspection. Infrared examinationshave been useful in detecting areas of major heat loss inbuildings or between buildings where steam leaks wereoccurring. Such inspections have also proven valuablein detecting areas of unseen friction in motors andthus finding problem areas before the problems havebecome major.Thermocouples 15To cre ate a thermocouple, two wires or strips ofdifferent material are joined at both ends. Then one endis kept at a constant temperature and the other end isused in the temperature probe. If the ends are at differenttemperatures, a voltage difference will occur between theends. Measuring this voltage provides a way to estimat ethe temperature difference between the two ends. Instrumentsusing this principle are found in many places,particularly where a permanent meter is desired for temperaturein a remote or inaccessible place.14.5.6 Velocity and Flow-Rate MeasurementIn many situations, it is necessary to know the flowrate of some substance, such as air or steam, to determinewhere energy is being used. For example, a factorywith several buildings and a central steam system maynot be metered so t hat the steam consumption of eachbuilding can be determined. As another example, it isgenerally difficult to estimate the amount of air beingmoved by a fan in an air-conditioning system withoutmaking some measurement of air velocity. Three typesof measuring instruments used are flow hoods, pitottubes, and orifice plates.Flow HoodsA flow hood resembles an inverted pyramid withthe top replaced by a small cube. The inverted pyramidis made of cloth treated to minimize air leaks. Thesmall cube is a turbine that generates current which ismeasured by an attached meter. In practice, the openingof the hood is placed over the grill emitting theair. Air is forced to the base of the pyramid, turns theturbine, and generates electricity. The meter is calibratedin ft/min (or m/sec). Since the cross-sectionalarea is known at the point where the velocity is beingmeasured, the number of ft 3 /min (or m 3 /sec) can beimmediately calculated.Pitot TubesThe pitot tube operates on the principle that airflow across the end of an open tube creates a pressuredrop, and a measurement of this pressure drop can beconverted into a measurement of the air velocity at theend of the tube. Pitot tubes have been used in applicationsranging from air flow rates from a duct to determiningthe air speed of an airplane. They can be used toestimate the air flow velocity across a duct and may bemore convenient than flow hoods for some applications.Care must be taken, however, that enough readings a remade to give a representative velocity profile—one readingis not enough.Orifice Plates 15An orifice plate consists of a disk with a hole ofknown diameter mounted in a pipe or duct with a manometerattached to the pipe upstream and downstreamfrom the orifice plate. Since the hole in the orifice plateis always smaller than the inside diameter of the pipe,the pressure downstream of the orifice plate is smallerthan the pressure upstream of the plate. This pressuredifference can be used together with the diameter of theorifice and the inside diameter of the pipe and the de n-sity of the material to give a value for the flow rate.14.5.7 Vibration MeasurementVibration is found in most mechanical devices thatmove. Sometimes this noise is helpful, as in the caseof the noise caused by a bucket steam trap openingand closing. Often, however, an increase in vibrationof a machine is an indication that something is goingwrong. Among the many instruments that can be usedto check vibrations, two are of particular value in themaintenance associated with energy management: thestethoscope and the stroboscope.StethoscopeA stethoscope bri ngs noise from a particular placeto the ears of the observer and by isolating the sourceseems to amplify the noise. Two valuable uses for thisare in the Tornberg procedure for monitoring motorbearings, described in Section 14.1, 1, and in the procedurefor monitoring steam trap operation of steam trapsdescribed in Section 14.2.3.StroboscopeA stroboscope is a flashing light whose flashes occurat regular intervals that can be changed at the willof the operator. When the flashes occur at the same timethat a particular part of rotating machinery passes onepoint, the flashes make the machinery appear to stand

430 ENERGY MANAGEMENT HANDBOOKstill. Thus any lateral vibrations of the equipment appearin slow motion and can be examined in detail. Thisprocedure also shows cracks that open up only when themachinery is in motion. By detecting incipient troublebefore it causes the equipment to be shut down, thestroboscope enables maintenance personnel to order replacementparts and to schedule corrective maintenanceat convenient times rather than at the times dictated byequipment breakdowns.14.6 SAVING ENERGY DOLLARS INMATERIALS HANDLING AND STORAGEThe earlier parts of this chapter have dealt withvarious topics in energy-efficient maintenance ofeq uipment and facilities. Two other topics of interestnot covered elsewhere in this handbook are energymanagement in materials handling and new devicesfor energy cost savings. This section covers the first ofthose topics.Section 14.3 discussed the maintenance of certainkinds of materials handling equipment, specifically lifttrucks and conveyor systems. In addition to propermaintenance, however, there are many operating changesthat together can save a large fraction of the materialshandling energy cost. These cost savings can be realizedif a systematic approach is followed. One approach is toanswer these four questions:1. What is energy being used for now, and is all ofthis use productive?2. How much electricity, gas, and oil is used now, andcan this use or its cost be reduced by modifyingequipment usage?3. Can the working hours or the locations of peoplebe changed to reduce energy requirements withoutadverse impacts?4. How can the energy management program bemonitored? These questions are discussed in Sections14.6.1 to 14.6.4.14.6.1 Analyzing Present Energy UsageEnergy is used in three ways in materials handlingand storage: to move material, to conditionspaces for material, and to condition spaces for peoplewho are moving or storing material. Energy used tomove material includes fuel or electricity for lift trucks,diesel or electricity for conveyor power, or cranes andelectricity for automatic storage and retrieval systems.Examples of energy to condition spaces for material includesrefrigeration for vegetables, controlled temperaturestorage for electric components, and materials andventilation of areas used to store volatile liquids. Theenergy used to condition spaces for people includesheating, cooling, and ventilating a warehouse to makeit comfortable for persons working there, air conditioninga cab of a large crane, and providing a comfortableworking area for the secretaries supporting the materialshandling and storage functions. These three functionsalso interact—if gasoline is used in a warehouselift truck, the amount of ventilation air required forpersonnel is 8000 ft 3 /min per lift truck, a requirementthat is not present for electric lift trucks. (Nonelectrictrucks may, however, have offsetting advantages forindividual applications.) Energy used for conditioningmaterial storage spaces can also interact in other wayswith energy use d to condition space for people—forexample, waste heat from a refrigerator has been usedto heat office space.14.6.2 Walk-Through AuditThe first step in analyzing energy usage in yourmaterials handling and storage system is to find outwhat equipment is being used to move material and,when it is turned on, to check the conditions that arebeing maintained for material. Then determine wherethe people are located in the system and what conditionsare being maintained for their benefit. To do this,three walk-through audits are recommended. The firstaudit is performed during working hours in an attemptto find practices that can be improved. For this audit,you should be equipped with an industrial light meter(about $20 to $30) and a pocket thermometer ($10 to$15). As you walk around the facility, you should writeor record potential improvements for later analysis.These improvements can be found in answer to questionssuch as these:1. Does equipment need to be idling when no one isusing it, and, if so, what is a reasonable maximumfor an idling period?2. Is the temperature being maintained unnecessarilyhigh or low—is it necessary to air-condition thisspace in the summer or to heat it in the winter?3. Is all the lighting necessary? Standard guidelinesfor warehouse lighting in the IES Handbook 16 rangefrom 5 to 50 footcandles, depending upon whetherthe storage is in active use and upon the amount ofvisual effort needed to distinguish one item fromanother.4. Is a great deal of conditioned air lost whenevertrucks unload at the facility?The second walk-through audit is performed when

ENERGY SYSTEMS MAINTENANCE 431office personnel have left for the day. The intent of thisaudit is to discover equipment and lights that are onbut are not needed for the security of the building or itscontents. This audit should focus on equipment, lights,and unnecessary heating and cooling of office spaces.The use of a night setback—setting the thermostat to55°F at night—can save as much as 20% of the bill foroffice heating and cooling.The third audit is the 2 a.m. audit, performedsometime after most people have left the building. Inthis audit, look for motors that are on unnecessarily,for lights that are on but are not needed for securitypurposes, and for temperatures that are higher or lowerthan they need to be.Completion of the three walk-through audits givesa qualitative survey of the equipment being used, lightinglevels and temperatures being maintained, a nd someoperating improvements that can be instituted. This isa good start. More information is necessary, however,before an energy management program can achieve itspotential, and this is provided by the next step.14.6.3 Finding and Analyzing Improvementsthat Cost MoneyWalk-through audits can uncover operating practicesthat can be improved, but a different kind ofanalysis is needed to discover possible capital-intensiveimprovements. This analysis has three parts: examinationof past energy bills, use of a checklist, and economicjustification of the most promising improvements. (Theeconomic analysis is discussed in Chapter 4.)Examining Past BillsOne purpose in examining energy bills is to determinethe total amount that can be saved. If your billfor fuel for vehicles is $25,000 per year, then $25,000 isthe upper annual limit on savings. Another purpose forexamining bills is to find what factors are significantin the billing for your facility. If, for example, you arebeing charged for a low power factor on your electricalbill, power-factor improvement may be worthwhile. Ifyour electrical bill shows a factor labeled “power” or“demand,” you are being charged for the peak poweryou use. To find whether peak electrical demand presentsan opportunity for cost savings, compute the loadfactor by the formulakWh used in billing periodload factor = ——————————————————peak demand x hours in billing period(The billing period in days is given on the bill.) If theload factor is less than 70 to 80%, there are significantperiods of high electrical usage. To determine whenthese occur, have your electrical utility install a recordingammeter or install one yourself. When you findthe peak usage times, examine your equipment to findwhat causes the peaks and see if some of this can berescheduled at off-peak times.A third purpose for examining past bills is to establisha base for comparison for your energy managementprogram. For your continued economic health itis essential that you manage your energy consumption.In order to find the measures that have worked foryour facility, however, you need to be able to show thatthese measures have caused you to fall below previousenergy usage. If measures you institute do not cause theconsumption to change significantly, you know that youneed to do more.Possible Areas for Energy Consumption ImprovementAs soon as your bills have been examined, you arein a position to examine possible areas where moneyinvested can have large returns in energy cost containment.(Much of this material was taken, with permission,from Ref. 17.) These areas, in abbreviated form,are presented in Tables 14.19 to 14.22.14.6.4 MonitoringWhen the energy consumption base has been establishedand the energy management plan has begunto take effect, energy consumption must be monitored.Monitoring serves three functions: to evaluate progress,to show which measures work, and to show whichmeasures do not work.The first step in a monitoring program is to choosesome measure of energy consumption to monitor. Eachkind of energy used in the facility should be monitored,with separate graphs for gasoline, propane, gas, andelectric usage (kWh) and electric demand (kW). Theseshould be plotted for the 12 months previous to theinstallation of the energy management plan, where possible,to establish a base against which to judge improvement.If your facility does not have a separate electricmeter or separate accounting procedures for fuel, it isprobably worthwhile to install them so you can knowexactly what effect your program is having on yourenergy consumption.Once the energy management plan has gone intoeffect, the next step is to graph the consumpt ion, monthby month, to see whether the program has helped and, ifso, how much. If the program has not decreased energyconsumption, you can use the monitoring equipment toanalyze your energy use in more detail—to find whatareas or pieces of equipment are using the energy. Then

432 ENERGY MANAGEMENT HANDBOOKTable 14.19 Materials handling energy savings: doors.ProblemPeople use truck bay doors to enter andleave buildingOpen passages between heated and cooled areasTruck bays open to outside airSolutionInstall personnel doorsBenefits include less loss of heated or cooled airUse strip curtain doors if lift trucks use the passageway, unlessthere is a pressure difference between the areasInstall adjustable dock cushionsSource: Ref. 17.Table 14.20 Materials handling energy savings: industrial trucks.ProblemVentilation is used primarily to get rid oftruck fumesExcessive demand charge for electric powerExcessive fuel consumptionSolution1. Modify trucks to reduce ventilation requirements if possible2. Replace diesel or propane trucks by electric trucks3. Replace with more efficient trucks1. Have electric trucks charged during off-peak hours2. Install drop-in dc generators1. Reduce truck idle time to manufacturer’s specifications2. Use smaller trucks where possible3. Use a communication system between dispatcher and trucks toreduce unloaded travel time4. Use improved ignition and carburetion devices5. Use more energy-efficient truck tiresSource: Ref. 17.Table 14.21 Materials handling energy savings: hoists and cranes.ProblemCranes powered by compressed airIdlingToo many movesMost loads too small for full useof hoist or crane capacityCranes running during off hoursSolutionUse electrical hoists and cranes (5 hp on compressor = 1 hp on hoist)Use switches to turn crane off when not in use for a preset time (5-10 min)Package in larger unit loadsSize the crane or hoist to load moved most oftenTurn off as part of regular closing check listSource: Ref. 17.Table 14.22 Materials handling energy savings: conveyors.ProblemSolutionExcessive idling1. Install controls to keep conveyors running only when loaded2. Wire conveyors to light switchMotors running at less than top load 1. Downsize motors and install slow-start controlsMotors burn out 1. Institute preventive maintenance program for motors (see Section 14.1.1)2. Replace single-phase with three-phase motors

ENERGY SYSTEMS MAINTENANCE 433concentrate on these, and start again.When this monitoring has shown that a particularmeasure is particularly effective, find out why, and copyits best features elsewhere. If a measure did not work,find out why and make sure that the bad features of itare not being duplicated elsewhere.14.7 RECENT DEVELOPMENTSTwo recent developments that have had dramaticimpacts on energy management are the maturing of computer-basedenergy maintenance management systems andthe increased availability of low-cost remotely accessiblesensors.14.7.1 Energy Maintenance Management SystemsAn ideal maintenance management system wouldhave these attributes:The most recent energy costs and production datawould be available, broken down as much as possible intocosts by areas or by product, depending upon the amountof metering that is installed. Variance reports should beavailable on request to use in comparing actual and projectedenergy costs.A complete file of maintenance tasks should be available,with the following information for each:• Identification number of equipment to be maintained• Maintenance tasks to be performed, with instructionsand check lists for each task• Frequency of maintenance• Expected maintenance duration• Skills needed• Equipment needed for fault diagnosis and for repair• Priority and justification if appropriate• Repair parts neededThis file can then be used to generate daily maintenanceschedules and to project needs for repair parts.With sufficient additional cost and production data, it canalso be used to estimate the total cost and benefit of themaintenance function as performed, and it can be used forplanning.It should be possible to incorporate a daily log ofpreventive and repair maintenance actions performed. Thisfile would include, for each unit of equipment maintained,the following:• Identification number of equipment to be maintained• Date and time• Person in charge• Equipment condition• Maintenance tasks performed• Time needed for repair• Repair parts used• Additional relevant notesThese characteristics should be built into a managementinformation system in such a way that reports andgraphs are easily composed and retrieved. Such systemshave existed for some time, and, when properly designed,they can provide a great deal of information quickly.But the problem with such a system lies in making itusable by maintenance technicians, and in keeping it up todate. (Note: the following description comes from Guideto Energy Management, by Capehart, Turner, and Kennedy;Fairmont Press, 2000).“Most maintenance people don’t like spread sheets.The information is important, but it will not be collectedif collecting it is more trouble than it appears to be worth.There is a solution—the hand held computer. One company(and by now probably three or four more) has developeda system with the following characteristics:• It is easy to use by the technician in the field. Thetechnician can record the conditions he/she has foundand the actions taken to remedy the problem, withoutwriting anything down.• A permanent record is kept for each machine or areamaintained, and this record can be accessed in thefield with a minimum of effort.• The equipment with the records is lightweight andpocket-portable.• Data from monitoring equipment can be incorporatedinto the equipment database easily.• All equipment and labor data are available for moresophisticated analysis at a central site, and the analysisresults can be immediately available for use in thefield.The technology for this system includes1. A computer capable of handling a large database andmany requests for service quickly. Such computersare readily available and not expensive.2. A portable hand-held transmitter/receiver capable ofdisplaying information sent from the main computerand of sending information back to it in a form that the

434 ENERGY MANAGEMENT HANDBOOKmain computer could understand and analyze. Suchequipment is also available now and is compatiblewith most large maintenance management systemssuch as those by SAP.3. The software and hardware to tie the computer andthe field units together.Such a system has been demonstrated by Datastream(Greenville, SC) (web address: www.dstm.com), FieldData Specialists, Inc. (www.trapbase.com) and others. Theadvantages of such a system are many. First, the informationis likely to be more accurate and more complete thaninformation from paper-based systems. Second, it is possibleto tailor the analysis of a particular machine problem tothe particular machine, knowing the history of repairs onthat machine. Third, good data can be kept for use in spareparts inventory calculations so that this element of repairdelay is eliminated. Finally, it is easy for the techniciansin the field to use, so its use is more likely than for a morecumbersome system or one based on paper.”14.7.2 New Sensors and Monitoring Equipment• Data loggers: Recently, inexpensive battery-poweredinstruments have become available that record measurements(temperature, relative humidity, light intensity,on/off, open/closed, voltage and events) overtime. These data loggers (the name used by the OnsetComputer Corporation) are small, battery-powereddevices that are equipped with a microprocessor, datastorage and sensor. Most data loggers utilize turnkeysoftware on a personal computer to initiate thelogger and view the collected data. Such equipmentis available from the Onset Computer Corporation,at www.onsetcomp.com. The data loggers can keeptrack of the condition of equipment and, when usedwith appropriate software, can project times whenpreventive maintenance should be performed.• Ultrasonic detectors for steam trap testing. Oneclassical way to check steam traps was the “screwdrivermethod,” where a technician put one endof the screwdriver on the steam trap and the otherend in his ear. He then listened for noises of steamand condensate flow and for unusual trap noises.This method has been made more sophisticated bythe use of ultrasonic detectors—stethoscopes withamplifiers and monitors, with software to translateall noises into the audible range. They usually haveheadphones and frequently have meters to indicatethe frequencies being detected. Ultrasonic detectorscan be used to monitor the operation of steam trapsand to detect and localize steam leaks. The cost ofsteam saved by fixing one steam trap or one leak canfrequently pay for the entire cost of the detectors.Information on ultrasonic systems can be found atwww.enerchecksystems.com (Enercheck Systems,Inc.) or from other vendors.14.7.3 Monitoring Systems for EnergyMaintenance ManagementIn addition to the sensors and monitoring equipmentdescribed above, several companies specialize in systemsdesigned to show an operator the condition of equipmentin the boiler room and related areas. Many of these unitsare available off-the-shelf. These include:• Hot water reset and boiler cutout controls (www.mge.com). These adjust the heating water temperatureto a lower temperature when the ambientoutside air is warm, and they turn off the heatingsystem when it is no longer necessary.• Boiler controls. These have been around for a longtime, with the displays continually advancing in theinformation and timeliness they display. Some of thevendors for these are Foster-Wheeler, Honeywell,Warrick, McDonnell Miller, Tekmar, Yokogawa,Omega Engineering, Primary Flow Signal, Inc., andothers. Parameters controlled and displayed typicallyinclude water levels, steam pressures within aboiler and in the steam distribution system, steamand condensate flows, and temperatures. In addition,it is possible to get monitors for pH, conductivity,and flue gas emissions including NO x , CO,and O 2 . The sensor information from these systemscan be transmitted to displays at the operating stationsthrough hard wired circuits or by the use ofradio frequency transmitters and receivers. Thesedisplays assist significantly in managing the boilerand steam distribution system, and they providecontinuous data for use in boiler maintenance.• Boiler sequencing controls. The question of whichboilers to turn on for the most efficient operationis addressed with the use of boiler sequencingcontrols.• Steam traps. Maintaining steam traps has been acumbersome process, in part because it has not beeneasy to determine when a steam trap is working.Armstrong has tackled this problem with their SteamEye system where traps are tested continuouslyand failed traps are reported by a radio transmitter.

ENERGY SYSTEMS MAINTENANCE 435By having a probe in the steam trap, such a unit cansignal if the trap is operating normally, if steam isblowing through the trap without any condensing,or if the trap is flooded with condensate and inoperative.The results of monitoring are displayed inthree files: monitored traps, non-critical trap faults,and new critical trap faults. Since this monitoringsystem saves much of the time needed to determinethe location and nature of steam trap problems, itappears to be a significant advance over the timewhen each trap needed to be examined by either astethoscope or by an ear held on one end of a longscrewdriver. The Armstrong company web site iswww.armstrong-intl.com.Most of the large boiler companies supply the customerwith custom-tailored display and data communicationsystems. Most companies that design and build boilerspay particular attention to boiler control systems. A recentbrochure put out by The McBurney Corporation, whoseareas of expertise include waste and by-product boilersystems, has displays for the Boiler Fuel Delivery system,the Boiler Steam and Feed water treatment system, andseparate modules for the turbine generator, fuel delivery,boiler burner management, and combustion control.(www.mcburney.com) The units in each of these modules arecontrolled by programmable logic controllers that arelinked to the main control display.References1. National Electrical Contractors Association and NationalElectrical Manufacturers Association, Total Energy Management:a Practical Handbook on Energy Conservation andManagement, National Electrical Contractors Association,1976.2. Operation and Maintenance Manual for Steel Boilers (IncludingWater Treatment), The Hydrionics Institute, 34 RussoPlace, Berkeley Heights, N.J. 07922, 1964.3. “Operation and Maintenance Instructions, Class 600 IndustrialCross-Flow Cooling Towers,” The Marley Co.,5800 Foxridge Drive, Mission, Kans. 66202.4. New HVAC Energy Saving Developments, Air DistributionAssociates, Inc., 955 Lively Boulevard, Wood Dale, Ill.60191.5. ASHRAE Handbook and Product Directory, 1979, Equipment,American Society of Heating, Refrigerating and Air ConditioningEngineers, Inc., New York, 1979.6. Thomas F. Sack, A Complete Guide to Building and PlantMaintenance, Prentice-Hall, Englewood Cliffs, N.J., 1971.7. Unit Price Standards Handbook, Depts. of the Navy, theArmy, and the Air Force, NAVFAC P-716.8. Steam Conservation Guidelines for Condensate Drainage, ArmstrongMachine Works, Three Rivers, Mich. 49093.9. Dale E. Shore and Michael W. McElroy, Guidelines forIndustrial Boiler Performance Improvement, EPA Report EPA-600/8-77-003a, Jan. 1977, U.S. Environmental ProtectionAgency, Industrial Environmental Research Laboratory,Research Triangle Park, N.C.10. W.H. Axtman, Boiler Fuel Management and Energy Conservation,American Boiler Manufacturers Association, Suite700, 1500 Wilson Boulevard, Arlington, VA 22209.11. Owner’s and Operator’s Guide, E-Series Sit Drive, HysterCompany, Danville, IL 61832, 1976.12. Owner’s and Operator’s Guide, Challenger, Hyster Company,Danville, IL 61832, 1975.13. Off-Highway Fuel Savings Guide, Cummins Engine Co., Inc.,Columbus, Ind., 1977.14. Automative Marketing Bulletin 78-25, Cummins Engine Co.,Inc., Columbus, IN.15. ASHRAE Handbook and Product Directory, 1977 Fundamentals,American Society of Heating, Refrigerating and AirConditioning Engineers, Inc., New York, 1978.16. “IES Recommended Levels of Illumination,” IES LightingHandbook, 5th ed., Illuminating Engineering Society ofNorth America, New York, 1977.17. Wayne C. Turner, “The Impact of Energy on Material Handling,”presented at the 1980 Materials Handling InstituteSeminar, St. Louis, MO.18. The MP5 Enterprise System, by Datastream Systems, Inc.,50 Datastream Plaza, Greenville, SC 29605.


CHAPTER 15INDUSTRIAL INSULATIONJAVIER A. MONT, PH.D., CEMJohnson Controls, Inc.Industrial Global AccountsChesterfield, MissouriMICHAEL R. HARRISONManager, Engineering and Technical ServicesJohns-Manville Sales Corp.Denver, ColoradoThermal insulation is a mature technology that haschanged significantly in the last few years. What hasnot changed is the fact that it still plays a key role inthe overall energy management picture. In fact, the useof insulation is mandatory for the efficient operation ofany hot or cold system. It is interesting to consider thatby using insulation, the entire energy requirements of asystem are reduced. Most insulation systems reduce theunwanted heat transfer, either loss or gain, by at least90% as compared to bare surfaces.Since the insulation system is so vital to energy-efficientoperations, the proper selection and application of thatsystem is very important. This chapter describes the variousinsulation materials commonly used in industrialapplications and explores the criteria used in selectingthe proper products. In addition, methods for determiningthe proper insulation thickness are developed,taking into account the economic trade-offs betweeninsulation costs and energy savings.15.1 FUNDAMENTALS OFTHERMAL INSULATION DESIGN THEORYThe basic function of thermal insulation is to retardthe flow of unwanted heat energy either to or from aspecific location. To accomplish this, insulation productsare specifically designed to minimize the three modes ofheat transfer. The efficiency of an insulation is measuredby an overall property called thermal conductivity.15.1.1 Thermal ConductivityThe thermal conductivity, or k value, is a measureof the amount of heat that passes through 1 squarefoot of 1-inch-thick material in 1 hour when there isa temperature difference of 1°F across the insulationthickness. Therefore, the units are Btu-in./hr ft 2 °F. Thisproperty relates only to homogeneous materials andhas nothing to do with the surfaces of the material.Obviously, the lower the k value, the more efficient theinsulation. Since products are often compared by thisproperty, the measurement of thermal conductivity isvery critical. The American Society of Testing Materials(ASTM) has developed sophisticated test methods thatare the standards in the industry. These methods allowfor consistent evaluation and comparison of materialsand are frequently used at manufacturing locations inquality control procedures.ConductionHeat transfer in this mode results from atomic ormolecular motion. Heated molecules are excited and thisenergy is physically transferred to cooler molecules byvibration. It occurs in both fluids (gas and liquid) andsolids, with gas conduction and solid conduction beingthe primary factors in insulation technology.Solid conduction can be controlled in two ways: by utilizinga solid material that is less conductive and by utilizingless of the material. For example, glass conductsheat less readily than steel and a fibrous structure hasmuch less through-conduction than does a solid mass.Gas conduction does not lend itself to simple modification.Reduction can be achieved by either reducing thegas pressure by evacuation or by replacing the air witha heavy-density gas such as Freon®. In both cases, theinsulation must be adequately sealed to prevent reentryof air into the modified system. However, since gas conductionis a major component of the total thermal conductivity,applications requiring very low heat transferoften employ such gas-modified products.ConvectionHeat transfer by convection is a result of hotfluid rising in a system and being replaced by a colder,heavier fluid. This fluid heats, rises, and carries moreheat away from the heat source. Convective heat transferis minimized by the creation of small cells within whichthe temperature gradients are small. Most thermal insulationsare porous structures with enough density to437

438 ENERGY MANAGEMENT HANDBOOKblock radiation and provide structural integrity. As such,convection is virtually eliminated within the insulationexcept for applications where forced convection is beingdriven through the insulation structure.RadiationElectromagnetic radiation is responsible for muchof the heat transferred through an insulation and increasesin its significance as temperatures increase.The radiant energy will flow even in a vacuum and isgoverned by the emittance and temperature of the surfacesinvolved. Radiation can be controlled by utilizingsurfaces with low emittance and by inserting absorbersor reflectors within the body of the insulation. The coredensity of the material is a major factor, with radiationbeing reduced by increased density. The interplaybetween the various heat-transfer mechanisms is veryimportant in insulation design. High density reducesradiation but increases solid conduction and materialcosts. Gas conduction is very significant, but to alter itrequires permanent sealing at additional cost. In addition,the temperature in which the insulation is operatingchanges the relative importance of each mechanism.Figure 15.1 shows the contribution of air conduction,fiber conduction, and radiation in a glass fiber insulationat various densities and mean temperatures.15.1.2 Heat TransferThere are many texts dedicated to the physics ofheat transfer, some of which are listed in the references.In its simplest form, however, the basic law of energyFig. 15.1 Contribution of each mode of heat transfer. (From Ref. 15.)flow can be stated as follows:A steady flow of energy through any medium of transmissionis directly proportional to the force causing the flow and inverselyproportional to the resistance to that force.In dealing with heat energy, the forcing functionis the temperature difference and the resistance comesfrom whatever material is located between the two temperatures.temperature differenceheat flow = ———————————resistance to heat flowThis is the fundamental equation upon which allheat-transfer calculations are based.Temperature DifferenceBy definition, heat transfer will continue to occuruntil all portions of the system are in thermalequilibrium (i.e., no temperature difference exists). Inother words, no amount of insulation is able to provideenough resistance to totally stop the flow of heat as longas a temperature difference exists. For most insulationapplications, the two temperatures involved are theoperating temperature of the piping or equipment andthe surrounding ambient air temperature.Thermal ResistanceHeat flow is reduced by increasing the thermalresistance of the system. The two types of resistancescommonly encountered are massand surface resistances. Most insulationsare homogeneous and as suchhave a thermal conductivity or kvalue. Here the insulation resistance,R I = tk/k, where tk represents thethickness of the insulation. In casesof nonhomogeneous products suchas multifoil metallic insulations, thethermal properties of the productsat their actual finished thicknessesare expressed as conductances ratherthan conductivities based on a 1-in.thickness. In this case the resistanceR I = 1/C, where C represents themeasured conductance.The other component of insulationresistance is the surface resistance,R s = 1/f, where f represents the surfacefilm coefficient. These valuesare dependent on the emittance of

INDUSTRIAL INSULATION 439the surface and the temperature difference between thesurface and the surrounding environment.Thermal resistances are additive and as such arethe most convenient terms to deal with. Following areseveral expressions for the heat-transfer equation, showingthe relationships between the commonly used R,C, and U values. For a single insulation with an outerfilm:∆tQ = ————R I + R s∆t= —————tk/k + 1/f∆t= ——————1/C I + 1/f= U ∆tWhere the insulation isnot homogeneous1 1where U = ———– = ————R total R I + R sU is termed the overall coefficient of heat transmissionof the insulation system.15.2 INSULATION MATERIALSMarketplace needs, in conjunction with activeresearch and development programs by manufacturers,are responsible for a continuing change in insulationmaterials available to industry. Some products havebeen used for decades, whereas others are relativelynew and are still being evaluated. The following sectionsdescribe the primary insulation materials availabletoday, but first, the important physical properties will bediscussed.15.2.1 Important PropertiesEach insulation application has a unique set ofrequirements as it relates to the important insulationproperties. However, certain properties emerge as beingthe most useful for comparing different productsand evaluating their fitness for a particular application.Table 15.1 lists the insulation types and product propertiesthat are discussed in detail below. One area thatwill not be discussed is industrial noise control. Thermalinsulations are often used as acoustical insulations fortheir absorption or attenuation properties. Many textsare available for reference in this area.Table 15.1 Industrial insulation types and properties.

440 ENERGY MANAGEMENT HANDBOOKTemperature-Use RangeSince all products have a point at which they becomethermally unstable, the upper temperature limitof an insulation is usually quite important. In somecases the physical degradation is gradual and measuredby properties such as high-temperature shrinkage orcracking. In such cases, a level is set for the particularproperty and the product is rated to a temperature atwhich that performance level is not exceeded. Occasionally,the performance levels are established by industrystandards, but frequently, the manufacturers establishtheir own acceptance levels based on their own researchand application knowledge.In other cases, thermal instability is very rapidrather than gradual. For example, a product containingan organic binder may have a certain temperature atwhich an exothermic reaction takes place due to a toorapidbinder burnout. Since this type of reaction can becatastrophic, the temperature limit for such a productmay be set well below the level at which the problemwould occur.Low-end temperature limits are usually not specifiedunless the product becomes too brittle or stiff and,as such, unusable at low temperatures. The most seriousproblem with low-temperature applications is usuallyvapor transmission, and this is most often related tothe vapor-barrier jacket or coating rather than to theinsulation. In general, products are eliminated fromlow-temperature service by a combination of thermalefficiency and cost.Thermal ConductivityThis property is very important in evaluating insulationssince it is the basic measure of thermal efficiency,as discussed in Section 15.1.1. However, a few pointsmust be emphasized. Since the k value changes withtemperature, it is important that the insulation meantemperature be used rather than the operating temperature.The mean temperature is the average temperaturewithin the insulation and is calculated by summing thehot and cold surface temperatures and dividing by 2: (t h+ t s )/2. Thermal conductivity data are always publishedper mean temperature, but many users incorrectly makecomparisons at operating temperatures.A second concern relates to products which havek values that change with time. In particular, foamproducts often utilize an agent that fills the cells witha gas heavier than air. Shortly after manufacture, someof this gas migrates out, causing an increase in thermalconductivity. This new value is referred to as an “agedk” and is more realistic for design purposes.Compressive StrengthThis property is important for applications wherethe insulation will see a physical load. It may be a fulltimeload, such as in buried lines or insulation supportsaddles, or it may be incidental loading from foot traffic.In either case, this property gives an indication ofhow much deformation will occur under load. Whencomparing products it is important to identify the percentcompression at which the compressive strengthis reported. Five and 10% are the most common, andproducts should be compared at the same level.Fire Hazard ClassificationInsulation materials are involved with fire in twoways: fire hazard and fire protection. Fire protection refersto the ability of a product to withstand fire exposurelong enough to protect the column, pipe, or vessel it iscovering. This topic is discussed in Section 15.3.1.Fire hazard relates to the product’s contributionto a fire by either flame spread or smoke development.The ASTM E-84 tunnel test is the standard methodfor rating fire hazard and compares the FS/SD (flamespread/smoke developed) to that of red oak, whichhas a 100/100 rating. Typically, a 25/50 FHC is specifiedwhere fire safety is an important concern. Certainconcealed applications allow higher ratings, while themost stringent requirements require a noncombustibleclassification.Cell StructureThe internal cell structure of an insulation is a primaryfactor in determining the amount of moisture theproduct will absorb as well as the ease in which vaporwill pass through the material. Closed-cell structurestend to resist both actions, but the thickness of the cellwalls as well as the base material will also influence thelong-term performance of a closed-cell product. In milddesign conditions such as chilled-water lines in a reasonableambient, closed-cell products can be used withoutan additional vapor barrier. However, in severe conditionsor colder operating temperatures, an additionalvapor barrier is suggested for proper performance.Available FormsAn insulation material may be just right for aspecific application, but if it is not manufactured in aform compatible with the application, it cannot be used.Insulation is available in different types (Ref. 19).Loose-fill insulation and insulating cements. Loosefillinsulation consists of fibers, powders, granules ornodules which are poured or blown into walls or otherirregular spaces. Insulating cements are mixtures of a

INDUSTRIAL INSULATION 441loose material with water or other binder which areblown on a surface and dried in place.Flexible, semirigid and rigid insulation. Flexibleand semirigid insulation, which are available in sheets orrolls, are used to insulate pipes and ducts. Rigid insulationis available in rectangular blocks, boards or sheets andare also used to insulate pipes and other surfaces. Themost common forms of insulation are flexible blankets,rigid boards and blocks, pipe insulation half-sections andfull-round pipe sections.Formed-in-place insulation. This type of insulationcan be poured, frothed or sprayed in place to form rigidor semirigid foam insulation. They are available as liquidcomponents, expandable pellets or fibrous materialsmixed with binders.Removable-reusable insulation covers. Used toinsulate components that require routine maintenance(like valves, flanges, expansion joints, etc.). These coversuse belts, Velcro or stainless steel hooks to reduce theinstallation time.Other PropertiesFor certain applications and thermal calculations,other properties are important. The pH of a material isoccasionally important if a potential for chemical reactionexists. Density is important for calculating loads onsupport structures and occasionally has significance withrespect to the ease of installation of the product. The specificheat is used together with density in calculating theamount of heat stored in the insulation system, primarilyof concern in transient heat-up or cool-down cycles.15.2.2 Material DescriptionCalcium SilicateThese products are formed from a mixture of limeand silica and various reinforcing fibers. In general,they contain no organic binders, so they maintain theirphysical integrity at high temperatures. The calciumsilicate products are known for exceptional strength anddurability in both intermediate- and high-temperatureapplications where physical abuse is a problem. In addition,their thermal performance is superior to otherproducts at the higher operating temperatures.Glass FiberFiberglass insulations are supplied in more forms,sizes, and temperature limits than are other industrialinsulations. All of the products are silica-based and rangein density from 0.6 to 12 lb/ ft 3 . The binder systems employedinclude low-temperature organic binders, hightemperatureorganic/antipunk binders, and needledmats with no binders at all. The resulting products includeflexible blankets, semirigid boards, and preformedone-piece pipe covering for a very wide range of applicationsfrom cryogenic to high temperature. In general, thefiberglass products are not considered load bearing.Most of the organic binders used begin to oxidize(burn out) in the range 400 to 500°F. The loss of bindersomewhat reduces the strength of the product in thatarea, but the fiber matrix composed of long glass fibersstill gives the product good integrity. As a result,many fiberglass products are rated for service above thebinder temperature, and successful experience indicatesthat they are completely suitable for numerous applications.Mineral Fiber/Rock WoolThese products are distinguished from glass fiberin that the fibers are formed from molten rock or slagrather than silica. Most of the products employ organicbinders similar to fiberglass but the very high temperature,high-density blocks use inorganic clay-type binders.The mineral wool fibers are more refractory (heatresistant) than glass fibers, so the products can be usedto higher temperatures. However, the mineral wool fiberlengths are much shorter than glass and the products docontain a high percentage of unfiberized material. As aresult, after binder burnout, the products do not retaintheir physical integrity very well and long-term vibrationor physical abuse will take its toll.Cellular GlassThis product is composed of millions of completelysealed glass cells, resulting in a rigid insulation that istotally inorganic. Since the product is closed cell, it willnot absorb liquids or vapors and thus adds security tocryogenic or buried applications, where moisture is alwaysa problem. Cellular glass is load bearing, but alsosomewhat brittle, making installation more difficult andcausing problems in vibrating or flexing applications.At high temperatures, thermal-shock cracking can be aproblem, so a cemented multilayer construction is used.The thermal conductivity of cellular glass is higher thanfor most other products, but it has unique features thatmake it the best product for certain applications.E xpanded PerliteThese products are made from a naturally occurringmineral, perlite, that has been expanded at ahigh temperature to form a structure of tiny air cellssurrounded by vitrified product. Organic and inorganicbinders together with reinforcing fibers are used to holdthe structure together. As produced, the perlite materialshave low moisture absorption, but after heating and

442 ENERGY MANAGEMENT HANDBOOKoxidizing the organic material, the absorption increasesdramatically. The products are rigid and load bearingbut have lower compressive strengths and higher thermalconductivities than the calcium silicate productsand are also much more brittle.Plastic FoamsThere are three foam types finding some use inindustrial applications, primarily for cold service. Theyare all produced by foaming various plastic resins.Polyurethane/lsocyanurate Foams. These twotypes are rigid and offer the lowest thermal conductivitysince they are expanded with fluorocarbon blowingagents. However, sealing is still required to resist themigration of air and water vapor back into the foamcells, particularly under severe conditions with largedifferentials in vapor pressure. The history of urethanesis plagued with problems of dimensional stability andfire safety. The isocyanurates were developed to improveboth conditions, but they still have not achievedthe 25/50 FHC (fire hazard classification) for a fullrange of thickness. As a result, many industrial userswill not allow their use except in protected or isolatedareas or when covered with another fire-resistant insulation.The advantage of these foam products is theirlow thermal conductivity, which allows less insulationthickness to be used, of particular importance in verycold service.Phenolic Foam. These products have achieved therequired level of fire safety, but do not offer k valuesmuch different from fiberglass. They are rigid enoughto eliminate the need for special pipe saddle supportson small lines. However, the present temperature limitsare so restrictive that the products are primarily limitedto plumbing and refrigeration applications.Polyimide FoamsPolyimide foams are used as thermal and acousticalinsulation. This material is fire resistant (FS/SD) of10/10) and lightweight, so it requires fewer mechanicalfastening devices. Thermal insulation is availablein open-cell structure. Temperature stability limits itsapplication to chilled water lines and systems up to100°F.Elastomeric Cellular PlasticThese products combine foamed resins withelastomers to produce a flexible, closed-cell material.Plumbing and refrigeration piping and vessels are themost common applications, and additional vapor-barrierprotection is not required for most cold service conditions.Smoke generation has been the biggest problemwith the elastomeric products and has restricted theiruse in 25/50 FHC areas. To reduce installation costs,elastomeric pipe insulation is available in 6-ft long, presplittubular sections with a factory-applied adhesivealong the longitudinal joint.RefractoriesInsulating refractories consist primarily of twotypes, fiber and brick.Ceramic Fiber. These alumina-silica products areavailable in two basic forms, needled and organicallybonded. The needled blankets contain no binders andretain their strength and flexibility to very high temperatures.The organically bonded felts utilize various resinswhich provide good cold strength and allow the felts tobe press cured up to 18 lb/cu.ft. density. However, afterthe binder burns out, the strength of the felt is substantiallyreduced. The bulk ceramic fibers are also used invacuum forming operations where specialty parts aremolded to specific shapes.Insulating Firebrick. These products are manufacturedfrom high-purity refractory clays with aluminaalso being added to the higher temperature grades.A finely graded organic filler which burns out duringmanufacture provides the end product with a well-designedpore structure, adding to the product’s insulatingefficiency. Insulating firebricks are lighter and thereforestore less heat than the dense refractories and are superiorin terms of thermal efficiency.Protective Coatings and JacketsAny insulation system must employ the propercovering to protect the insulation and ensure long-termperformance. Weather barriers, vapor barriers, rigidand soft jackets, and a multitude of coatings exist forall types of applications. It is best to consult literatureand representatives of the various coating manufacturersto establish the proper material for a specific application.Jackets with reflective surfaces (like aluminumand stainless steel jackets) have low emissivity (ε). Forthis reason, reflective jackets have lower heat loss thanplain or fabric jackets (high emissivity). In hot applications,this will result in higher surface temperaturesand increase the risk of burning personnel. In coldapplications, surface temperatures will be lower, whichcould cause moisture condensation. Regarding jacketingmaterial, existing environment and abuse conditionsand desired esthetics usually dictate the propermaterial. Section 15.3.3 will discuss jacketing systemstypically used in industrial work.

INDUSTRIAL INSULATION 44315.3 INSULATION SELECTIONThe design of a proper insulation system is a twofoldprocess. First, the most appropriate material must beselected from the many products available. Second, theproper thickness of material to use must be determined.There is a link between these two decisions in that oneproduct with superior thermal performance may requireless thickness than another material, and the thicknessreduction may reduce the cost. In many cases, however,the thermal values are so close that the same thicknessis specified for all the candidate materials. This sectiondeals with the process of material selection. Section 15.4addresses thickness determination.15.3.1 Application RequirementsSection 15.2.1 discussed the insulation propertiesthat are of most significance. However, each applicationwill have specific requirements that are used to weighthe importance of the various properties. There are threeitems that must always be considered to determinewhich insulations are suitable for service. They are operatingtemperature, location or ambient environment,and form required.Operating TemperatureThis parameter refers to the hot or cold service conditionthat the insulation will be exposed to. In the eventof operating design temperatures that may be exceededduring overrun conditions, the potential temperatureextremes should be used to assure the insulation’s performance.Cryogenic (–455 to –150°F). Cryogenic service conditionsare very critical and require a well-designed insulationsystem. This is due to the fact that if the systemallows water vapor to enter, it will not only condenseto a liquid but will subsequently expand and destroythe insulation. Proper vapor barrier design is critical inthis temperature range. Closed-cell products are oftenused since they provide additional vapor resistance inthe event that the exterior barrier is damaged or inadequatelysealed.For the lowest temperatures where the maximumthermal resistance is required, vacuum insulations areoften employed. These insulations are specially designedto reduce all the modes of heat transfer. Multiple foilsheets (reduced radiation) are separated by a thin matfiller of fiberglass (reduced solid conduction) and arethen evacuated (reduced convection and gas conduction).These “ super insulations” are very efficient as long as thevacuum is maintained, but if a vacuum failure occurs, theadded gas conduction drastically reduces the efficiency.Finely divided powders are also used for bulk,cavity-fill insulation around cryogenic equipment. Withthese materials, only a moderate vacuum is required,and in the event that the vacuum fails, the powder stillacts as an insulation. It is, however, very importantto keep moisture away from the powders, as they arehighly absorbent and the ingress of moisture will destroythe system.Some plastic foams are suitable for cryogenicservice, whereas others become too brittle to use. Theymust all have additional vapor sealing since high vaporpressures can cause moisture penetration of the cellwalls. Closed-cell foamed glass (cellular glass) is quitesuitable for this service in all areas except those requiringgreat thermal efficiency. Since it is not evacuated andhas solid structure, the thermal conductivity is relativelyhigh.Because of the critical nature of much cryogenicwork, it is very common to have the insulation systemspecifically designed for the job. The increased use ofliquefied gases (natural and propane) together withcryogenic fluids in manufacturing processes will requirecontinued use and improvement of these systems.Low Temperature (–150 to 212°F). This temperaturerange includes the plumbing, HVAC, and refrigerationsystems used in all industries from residentialto aerospace. There are many products available in thisrange, and the cost of the installed thermal efficiency is alarge factor. Products typically used are glass fiber, plasticfoams, phenolic foam, elastomeric materials, and cellularglass. In below-ambient conditions, a vapor barrieris still required, even though as the service approachesambient temperature, the necessary vapor resistancebecomes less. Above-ambient conditions require littlespecial attention, with the exception of plastic foams,which approach their temperature limits around 200°F.Because of the widespread requirement for theplumbing and HVAC services within residential and commercialbuildings, the insulations are subject to a varietyof fire codes. Many codes require a flame spread ratingless than 25 for exposed material and smoke ratings from50 to 400, depending on location. A composite rating of25/50 FHC is suitable for virtually all applications, witha few applications requiring non-combustibility.Intermediate Temperature (212 to 1000°F). Thegreat majority of steam and hot process applicationsfall within this operating range. Refineries, powerplants, chemical plants, and manufacturing operationsall require insulation for piping and equipment at thesetemperatures. The products generally used are calciumsilicate, glass fiber, mineral wool, and expanded perlite.Most of the fiberglass products reach their temperature

444 ENERGY MANAGEMENT HANDBOOKlimit somewhere in this range, with common breakpointsat 450, 650, 850, and 1000°F for various products.There are two significant elements to insulationselection at these temperatures. First, the thermal conductivityvalues change dramatically over the range ofmean temperatures, especially for light-density productsunder 18lb/cu.ft. This means, for example, that fiberglasspipe covering will be more efficient than calciumsilicate for the lower temperatures, with the calciumsilicate having an advantage at the higher temperatures.A thermal conductivity comparison is of value in makingsure that the insulation mean temperature is usedrather than the operating temperature.The second item relates to products that use organicbinders in their manufacture. All the organics willburn out somewhere within this temperature range, usuallybetween 400 and 500°F. Many products are designedto be used above that temperature, whereas others arenot. This is mentioned here only to call attention to thefact that some structural strength is usually lost withorganic binder.High Temperature (1000 to 1600°F). Superheatedsteam, boiler exhaust ducting, and some process operationsdeal with temperatures at this level. Calcium silicate,mineral wool, and expanded perlite products arecommonly used together with the lower-limit ceramicfibers. Except for a few clay-bonded mineral wool materials,these products reach their temperature limits inthis range. Thermal instability, as shown by excessiveshrinkage and cracking, is usually the limiting factor.Refractory (1600 to 3600°F). Furnaces and kilns insteel mills, heat treating and forging shops, as well as inbrick and tile ceramic operations, operate in this range.Many types of ceramic fiber are used, with aluminasilicafibers being the most common. Insulating firebrick,castables, and bulk-fill materials are all necessary formeeting the wide variety of conditions that exist inrefractory applications. Again, thermal instability is thecontrolling factor in determining the upper temperaturelimits of the many products employed.LocationThe second item to consider in insulation selectionis the location of the system. Location includes many factorsthat are critical to choosing the most cost-effectiveproduct for the life of the application. Material selectionbased on initial price only without regard to locationcan be not only inefficient, but dangerous under certainconditions.Surrounding Environment. For an insulation toremain effective, it must maintain its thickness and thermalconductivity over time. Therefore, the system musteither be protected from or able to withstand the rigorsof the environment. An outdoor system needs to keepwater from entering the insulation, and in most areas,the jacketing must hold up under radiant solar load.Indoor applications are generally less demanding withregard to weather resistance, but there are washdownareas that see a great deal of moisture. Also, chemicalfumes, atmospheres, or spillage may seriously affect certainjacketing materials and should be evaluated priorto specifications. Direct burial applications are normallysevere, owing to soil loading, corrosiveness, and moisture.It is imperative that the barrier material be sealedfrom groundwater and resistant to corrosion. Also, theinsulation must have a compressive strength sufficientto support the combined weight of the pipe, fluid, soilbackfill, and potential wheel loads from ground traffic.Another concern is insulation application on austeneticstainless steel, a material subject to chloride stresscorrosioncracking. There are two specifications mostfrequently used to qualify insulations for use on thesestainless steels: MIL-1-24244 and Nuclear RegulatoryCommission NRC Reg. Guide 1.36. The specificationsrequire, first, a stress-corrosion qualification test on actualsteel samples; then, on each manufacturing lot to becertified, a chemical analysis must be performed to determinethe amount of chlorides, fluorides, sodium, andsilicates present in the product. The specific amounts ofsodium and silicates required to neutralize the chloridesand fluorides are stated in the specifications.There are many applications where vibration conditionsare severe, such as in gas turbine exhaust stacks.In general, rigid insulations such as calcium silicatewithstand this service better than do fibrous materials,especially at elevated temperatures. If the temperatureis high enough to oxidize the organic binder, the fibrousproducts lose much of their compressive strength andresiliency. On horizontal piping, the result can be anoval-shaped pipe insulation which is reduced in thicknesson the top of the pipe and sags below the undersideof the pipe, thus reducing the thermal efficiency of thesystem. On vertical piping and equipment with pinnedoninsulation, the problem of sag is reduced, but thevibration can still tend to degrade the integrity of theinsulation.Location in a fire-prone area can affect the insulationselection in two ways. First, the insulation systemcannot be allowed to carry the fire to another area; thisis fire hazard. Second, the insulation can be selectedand designed to help protect the piping or equipmentfrom the fire. There are many products available forjust fireproofing such areas as structural steel columns,

INDUSTRIAL INSULATION 445but in general they are not very efficient thermalinsulations. When an application requires bothinsulation during operation and protection duringa fire, calcium silicate is probably the bestselection. This is due to the water of hydrationin the product, which must be driven off beforethe system will rise above the steam temperature.Other high-density, high-temperature productsare used as well. With all the products it is importantthat the jacketing system be designedwith stainless steel bands and/ or jacketing sincethe insulation must be maintained on the pipingin order to protect it. Figure 15.2 shows fire testresults for three materials per the ASTM E-119fire curve and indicates the relative level of fireprotection provided by each material.A final concern deals with the transportof volatile fluids through piping systems. Whenleaks occur around flanges or valves, these fluidscan seep into the insulation. Depending on the internalinsulation structure, the surface area may be increasedsignificantly, thus reducing the fluid’s flash point. If thiscritical temperature drops below the operating temperatureof the system, autoignition can occur, thus creatinga fire hazard. In areas where leaks are a problem, eithera leakage drain must be provided to remove the fluid orelse a closed-cell material such as cellular glass shouldbe used, since it will not absorb the fluid.The previous discussion is intended to draw attentionto specific application requirements, not necessarilyto determine the correct insulation to be used. Each situationshould be evaluated for its own requirements, andin areas of special concern (auto-ignition, fire, etc.) themanufacturer’s representative should be called upon toanswer questions specific to the product.Resistance to Physical Abuse. Although this issueis related to location, it is so important that it needs itsown discussion. In commercial construction and manylight industrial facilities, the pipe and equipment insulationis either hidden or isolated from any significantabuse. In such cases, little attention need be given tothis issue. However, in most heavy industrial applications,physical abuse and the problems caused by it arematters of great concern.Perhaps by definition, physical abuse differs fromphysical loading in that loading is planned and designedfor, whereas abuse is not. For example, with cold piping,pipe support saddles are often located external tothe insulation and vapor barrier. This puts the combinedweight of the pipe and fluid onto the lower portion ofthe insulation. This is a designed situation, and a rigidmaterial is inserted between the pipe and the saddleFig. 15.2 Fire resistance test data for pipe insulation. (Used bypermission from Ref. 6.)to carry the load. However, if a worker decides to usethe insulated pipe as a scaffold support, a walkway, ora hoist support, the insulation may not be designed tosupport such a load and damage will occur. A quickwalk-through of any industrial facility will show muchevidence of “unusual” or “unanticipated” abuse. Inpoint of fact, many users have seen so much of this thatthey now design for the abuse, having determined thatit is “usual” for their facility.The effects of abuse are threefold. First, dented andcreased aluminum jacketing is unsightly and lowers theoverall appearance of the plant. Nonmetal jacketsmay become punctured and torn. The second point isthat wherever the jacketing is deformed, the materialunder it is compressed and as a result is a much poorerinsulation since the thickness has been reduced. Finally,on outside lines some deformation will undoubtedlyoccur at the jacketing overlap. This allows for water toenter the system, further degrading the insulation andreducing the thermal efficiency.In an effort to deal with the physical abuse problem,some specifications call for all horizontal piping tobe insulated with rigid material while allowing a fibrousoption on the vertical lines. Others modify this specificationby requiring rigid insulation to a height of 6 to 10 fton vertical lines to protect against lateral abuse. Still, infacilities that have a history of a rough environment, it ismost common to specify the rigid material for all pipingand equipment except that which is totally enclosed orisolated.As previously mentioned, the primary insulationmaterial choice is between rigid and nonrigid materials.Calcium silicate, cellular glass, and expanded perlite

446 ENERGY MANAGEMENT HANDBOOKproducts fit the rigid category, whereas most mineralwool and fiberglass products are nonrigid. Over theyears, the calcium silicate products have become the standardfor rigorous services, combining good thermal efficiencywith exceptional compressive strength and abuseresistance. The maintenance activities associated withrigid insulations are significantly less than the maintenanceand replacement needs of the softer insulations inabuse areas. The costs associated with this are discussedin a later section. However, it is also recognized that often,maintenance activities are lacking, which results ina deteriorated insulation system operating at reduced efficiencyfor a long period of time.Form RequiredThe third general category to consider is the insulationform required for the application. Obviously, pipeinsulation and flat sheets are manufactured for specificpurposes, and the lack of a specific form eliminates thatproduct from consideration. However, there are subtledifferences between form that can make a significantdifference in installation costs and system efficiency.On flat panels, the two significant factors are panelsize and the single-layer thickness available. A fibrous4 × 8 ft sheet is applied much more rapidly than four 2× 4 ft sheets, and it is possible that the number of pinsrequired might be reduced. In regard to thickness, if onematerial can be supplied 4 in. thick as a single layer asopposed to two 2-in.-thick layers, the first option willresult in significant labor savings. The same holds truefor 18-in.- vs. 12-in.-wide rigid block installation.Fibrous pipe insulations have three typical forms:one-piece hinged snap-on, two piece halfsections, andflexible blanket wraparound. For most pipe sizes, theone-piece material is the fastest to install and may notrequire banding if the jacket is attached to the insulationand secured to itself. Two-piece products must be wiredin place and then subsequently jacketed in a separateoperation. Wrap-around blankets are becoming morepopular, especially for large-diameter pipe and smallvessels. They come in standard roll lengths and are cutto length on the job site.Rigid insulations also have different forms, whichvary with the manufacturer. The two-piece half-sectionpipe insulation is standard. However, these sections canbe supplied prejacketed with aluminum, which in effectgives a one-piece hinged section that does not need aseparate application of insulation and jacket. Also, thicknessesup to 6 in. are available, eliminating the need fordouble-layer applications where they are not requiredfor expansion reasons. The greatest diversity comes inthe large-diameter pipe sizes. Quads (quarter sections)are available and are both quicker to install and thermallymore efficient than is scored block bent aroundthe pipe. Similarly, curved radius blocks are availablefor sizes above quads and provide a better fit than doesflat beveled block or scored block.The important point is that the available formsof insulation may well affect the decision as to whichmaterial to select and which manufacturer to purchasethat material from. It is unwise to assume that all manufacturersoffer the same sizes and forms or that the costto install the product is not affected by its form.15.3.2 Cost FactorsSection 15.3.1 dealt with the process of selectingthe materials best suited for a specific application. Insome cases the requirements are so stringent that onlyone specific material is acceptable. For most situations,however, more than one insulation material is suitable,even though they may be rank-ordered by anticipatedperformance. In these cases, several cost factors shouldbe considered to determine which specific material(and/or manufacturer) should be selected to provide thebest system for the lowest cost.Initial CostIn new construction, the owner is usually interestedprimarily in the installed cost of the insulationsystem. As long as the various material options providesimilar thermal performance, they can be compared onan equal basis. The contractor, on the other hand, ismuch more concerned about the insulation form andits effect on installation time. It may be of substantialbenefit to the contractor to utilize a more costly materialthat can be installed more efficiently for reduced laborcosts. In a highly competitive market, these savings needto be passed through to the owner for the contractor tosecure the job. The point is that the lowest-cost materialdoes not necessarily become the lowest-cost installedmaterial, so all acceptable alternatives should be evaluated.Maintenance CostTo keep their performance and appearance atacceptable levels, all insulation systems must be maintained.This means, for example, that outdoor weatherprotection must be replaced when damaged to preventdeterioration of the insulation. If left unattended, theentire system may need to be replaced. In a high-abusearea, a nonrigid insulation may need to be replaced quitefrequently in order to maintain performance. Aestheticsoften play an important part in maintenance activities,depending on the type of operation and its location. In

INDUSTRIAL INSULATION 447such cases there is benefit in utilizing an insulation thatmaintains its form and if possible aids the jacketing orcoating in resisting abuse.The trade-off comes between initial cost and maintenancecost in that a less costly system may well requiregreater maintenance. Unfortunately, the authorities forinitial construction and ongoing maintenance are oftensplit, so the owner may not be aware of the future consequencesof the initial system selection. It is imperativethat both aspects be viewed together.Lost Heat CostIf the various suitable insulations are properlyevaluated, a more thermally efficient product should requireless thickness to meet the design parameters However,if a common thickness is specified for all products,there can be a substantial difference in heat loss or gainbetween the systems In such a case, the more efficientproduct should receive financial credit for transferringless heat, and this should be considered in the overallcost calculationsReferring to the previous discussion on maintenancecosts, there was an underlying assumption thatmaintenance would be performed to the extent that theoriginal thermal efficiency would be maintained. In reality,maintenance is usually not performed until the situationis significantly deteriorated and sometimes not eventhen. The result of this is reduced thermal efficiencyfor much of the life of a maintenance-intensive system.It is very difficult to assign a figure to the amount ofadditional heat transfer due to deterioration. In a wetclimate, for example, a torn jacket will allow moistureinto the system and drastically affect the performance.Conversely, in a dry area, the insulation might maintainits performance for quite some time. Still, when dealingwith maintenance costs, it is a valid concern that systemsin need of maintenance generally are transferringmore heat at greater cost than are systems requiring lessmaintenance.Design LifeThe anticipated project life is the foundation uponwhich all costs are compared. Since there are trade-offsbetween initial cost and ongoing maintenance and heatlosscosts, the design life is important in determiningthe total level of the ongoing costs. To illustrate, considerthe difference between designing a 40-year powerplant and a two-year experimental process. Assumingthat the insulation in the experimental process will bescrapped at the close of the project, it makes no senseto use a more costly insulation that has lower maintenancerequirements, since those future benefits willnever be realized. Similarly, utilizing a less costly butmaintenance-intensive system when the design life is 40years makes little sense, since the additional front-endcosts could be regained in only a few years of reducedmaintenance costs.15.3.3 Typical ApplicationsThis section is designed to give a brief overviewof commonly used materials and application techniques.For a detailed study of application, techniques, and recommendations,see Ref. 7, as well as the guide specificationssupplied by most insulation manufacturers.The Heat PlantBoilers are typically insulated with fiberglass ormineral wool boards, with some usage of calcium silicateblock when extra durability is desired. Powerhouseboilers are normally insulated on-site with the fibrousinsulation being impaled on pins welded to the boiler.Box-rib aluminum is then fastened to the stiffeners orbuckstays as a covering for the insulation. In most commercialand light industrial complexes, package boilersare normally used. These are insulated at the factory,usually with fiberglass or mineral wool.Breechings and other high-temperature duct workare insulated with calcium silicate (especially wheretraffic patterns exist), mineral wool, and high-temperaturefiberglass. On very large breechings, prefabricatedpanels are used, as discussed in the following paragraph.H-bar systems supporting the fibrous materialsare common, with the aluminum lagging fastened tothe outside of the H-bar members. Also, many installationsutilize roadmesh over the duct stiffeners, creatingan air space, and then wire the insulation to the meshsubstrate. Indoors, a finish coat of cement may be usedrather than metal lagging.Precipitators are typically insulated with prefabricatedpanels filled with mineral wool or fiberglassblankets. For large, flat areas, such panels provide veryefficient installations, as the panels are simply securedto the existing structure with self-tapping screws. H-barand Z-bar systems are also used to contain the fibrousboards.Steam piping insulation varies with temperatureand location, as discussed earlier. Calcium silicate wiredin place and then jacketed with corrugated or plainaluminum is very widely used. The jacketing is eitherscrewed at the overlap or banded in place. Fiberglass isused extensively in low-pressure steam work in areas oflimited abuse. Mineral wool and expanded perlite canalso be used for higher-temperature steam, but calciumsilicate is the standard.

448 ENERGY MANAGEMENT HANDBOOKProcess WorkHot process piping and vessels are typically insulatedwith calcium silicate, mineral wool, or high-temperaturefiberglass. Horizontal applications are generally subject tomore abuse than vertical and as such have a higher usageof calcium silicate. Many vessels have the insulationbanded in place and then the metal lagging banded inplace separately. Most vessel heads have a cement finishand may or may not be subsequently covered with metal.Recent product developments have provided a fiberglasswraparound product for large-diameter piping and vessels.This flexible material conforms to the curvature andneed only be pinned at the bottom of a horizontal vessel.Banding is then used to secure both the insulation andthe jacketing. In areas of chemical contamination, stainlesssteel jacketing is frequently used.Cold process vessels and piping also use a variety ofinsulations, depending on the minimum temperatureand the thermal efficiency required. Cellular glass iswidely used in areas where the closed-cell structure isan added safeguard (the material is still applied with avapor-barrier jacket or coating). It is also used whereverthere is a combined need for closed-cell structure andhigh compressive strength. However, the polyurethanematerials are much more efficient thermally, and incryogenic work, maximum thermal resistance is oftenrequired. Multiple vapor barriers are used with the urethanesto prevent the migration of moisture throughoutthe entire system. In all cold work, the workmanship,particularly on the outer vapor barrier, is extremely critical.There are many other specially engineered systemsfor cryogenic work, as discussed in Section 15.3.1.Fluid storage tanks located outdoors are typicallyinsulated with fiberglass insulation. Prefabricated panelsare either installed on studs or banded in place. Also, thejacketing can be banded on separately over the insulation.A row of cellular glass is placed along the base ofthe tank to prevent moisture from wicking up into thefiberglass. Sprayed urethane is also used on tanks thatwill not exceed 200°F but a trade-off exists between costefficiency and the long-term durability of the system.Tank roofs are a problem because of the need fora rigid walking surface as well as a lagging systemthat will shed water. Many tops are left bare for thisreason, whereas others utilize a spray coating of corkfilledmastic, which provides only minimal insulation.Rigid fiberglass systems can be made to work with awell-designed covering system that drains properly. Themost secure system is to use a built-up roofing systemsimilar to those used on flat-top buildings. The installationis generally more costly, but acceptable long-termperformance is much more probable.HVAC SystemDuct work constructed of sheet metal is usuallywrapped with light-density fiberglass with a preappliedfoil and kraft facing. The blanket is overlappedand then stapled, with tape or mastic being applied ifthe duct flow is cold and a vapor barrier is required.Support pins are required to prevent sag on the bottomof horizontal ducts. Fiberglass duct liner is used insidesheet-metal ducts to provide better sound attenuationalong the duct; this provides a thermal benefit as well.For exposed duct work, a heavier-density fiberglassboard may be used as a wrap to provide a more acceptableappearance. In all cases, the joints in the sheetmetal ducts should be sealed with tape or caulking tominimize air leakage and allow the transport of air tothe desired location, rather than losing much of it alongthe run.Rigid fiberglass duct board and round duct are alsoused in many low-pressure applications. These productsform the duct itself as well as providing the thermal,acoustical, and vapor-barrier requirements mostoften needed. The closure system used to join the ductsections also acts to seal the system for minimum airleakage.Chillers and chilled water expansion tanks are usuallyinsulated with closed-cell elastomeric sheet to preventcondensation on the equipment. The joints are sealedand a finish may or may not be applied to the outside,depending on location.Piping for both hot and cold service is normallyinsulated with fiberglass pipe insulation. On cold work,the vapor-barrier jacket is sealed at the overlap witheither an adhesive or a factory-applied self-seal lap. Ifstaples are used, they should be dabbed with mastic tosecure the vapor resistance. Aluminum jacketing is oftenused on outside work with a vapor barrier applied beneathif it is cold service. Domestic hot- and cold-waterplumbing and rain leaders are also commonly insulatedwith fiberglass. Insulation around piping supports takesmany forms, depending on the nature of the hanger orsupport. On cold work, the use of a clevis hanger on theoutside of the insulation requires a high-density insert tosupport the weight of the piping. This system eliminatesthe problem of adequately sealing around penetrationsof the vapor barrier.15.4 INSULATION THICKNESS DETERMINATIONThis section presents formulas and graphical proceduresfor calculating heat loss, surface temperature,temperature drop, and proper insulation thickness. Over

INDUSTRIAL INSULATION 449Q p = Q fthe last years, computer programs that perform thesecalculations are more readily available to customers (see2π 212= heat flux through a pipe,Q F = heat flux through a flat surface, Btu/hr the inner surface film coefficient, the same reasoningft 2 applies.section 15.5.4). But still, it is important to understand thebasics for their use. Although the overall objective is toBtu/hr lin. ftdetermine the right amount of insulation that should be A = area of insulation surface, ft 2used, some of the equations use thickness as an input L = length of piping, lin. ftvariable rather than solving for it. However, all thecalculations are simply manipulations or further refinementsof the equation in Section 15.1.2:Q T– = Q F × A or Q p × L = total heat loss, Btu/hrQ =∆tH = time, hr————R I + R C sp = specific heat of material. Btu/lb• °Fρ = density, lb/ft 3Following is a list of symbols, definitions, andM = mass flow rate of a material, lb/hrunits to be used in the heat-transfer calculations.∆ = difference by subtraction, unit lesst a = ambient temperature, °FRH = relative humidity, %t s = surface temperature of insulation next to ambient,°FDP = dew-point temperature, °Ft h = hot surface temperature, normally operatingtemperature (cold surface temperaturein cold applications), °F15.4.1 Thermal Design ObjectiveThe first step in determining how much insulationto use is to define what the objective is. There are manyreasons for using insulation, and the amount to be usedwill definitely vary based on the objective chosen. Thek = thermal conductivity of insulation alwaysfour broad categories, which include most applications,determined at mean temperature, Btu-in./hr ft 2 are (1) personnel protection, (2) condensation control, (3)°Fprocess control, and (4) economics. Each of these is discussedin detail, with sample problems leading throught m = (t h + t s )/2 = mean temperature of insulation,°Fthe calculation sequence.t h = (t in + t out )/2 = average hot temperaturewhen fluid enters at one temperature andleaves at another, °F15.4.2 Fundamental ConceptsThermal EquilibriumA very important law in heat transfer is that undertk = thickness of insulation, in.steady-state conditions, the heat flow through anyportion of the insulation system is the same as the heatr 1 = actual outer radius of steel pipe or tubing, flow through any other part of the system. Specifically,in.the heat flow through the insulation equals the heat flowfrom the surface to the ambient, so the temperature differencer 2 = (r 1 + tk) = radius to outside of insulationon piping, in.for each section is proportional to the resistancefor each section:Eq tk = r 2 Ln (r 2 /r 1 ) = equivalent thickness of insulationon a pipe, in.heat flow = ————————————temperature differenceresistance to heat flowf = surface air film coefficient, Btu/hr ft 2 °Ft h – t a t h – t s t s – t aQ = ——— = ——— = ———R s = 1/f = surface resistance, hr ft 2 R°F/BtuI + R s R I R sR IBecause all of the heat flows Q are equal, this relationshipis used to check surface temperature or other= tk/k = thermal resistance of insulation, hrft 2 °F/Btuinterface temperatures. For an analysis concerned with

450 ENERGY MANAGEMENT HANDBOOKt h – t a= t h – t s1= t s1 – t s2= t s2 – t aR s1 + R 1 + R s2 R s1 R 1 R s2Q == 3.86 actual outside diameterOr for a system with two insulation materials involved,the interface temperature t if between the materials isinvolved.tQ = h – t kEq tk/k +R stQ = h – t a= t h – t if= t if – t s= t s – t a= t if – t aR I1 + R I2 + R s R I2 R I2 R s R I2 + R sIt should be apparent that the heat flow Q is also equalfor any combination of ∆t and R values, as shown bythe last equivalency above, which utilized two parts ofthe system instead of just one.Finally, it is of critical importance to calculate theR I values using the insulation mean temperature, notthe operating temperature. The mean temperature is thesum of the temperatures on either side of the insulationdivided by 2. Again for the last set of equivalencies:t m for R I1 = t h + t if2t m for R I2 = t if – t s2Pipe vs. Flat Calculations— Equivalent ThicknessBecause the radial heat flows in a path from asmaller-diameter pipe, through the insulation, and thenoff a larger-diameter surface, a phenomenon termed“equivalent thickness” (Eq tk) occurs. Because of the geometryand the dispersion of the heat to a greater area,the pipe really “sees” more insulation than is actuallythere. When the adjustment is made to enter a greaterinsulation thickness into the calculation, the standardflat geometry formulas can be used by substituting Eqtk for tk into the equations.The formula for equivalent thickness isEq tk = r 2 ln r 2r 1where r 1 and r 2 are the inner and outer radii of theinsulation system. For example, an 8-in. IPS with 3-in.insulation would lead to an equivalent thickness as follows(8-in. IPS has 8.625 in. actual outside diameter):r 1 = 8.625 = 4.31 Table 15.22r 2 = r 1 + tk = 4.31 +3=7.31Eq tk = 7.31 ln 7.314.31 = 7.31 In 1.70This Eq tk can hen be used in the flat geometry equationby substituting Eq tk for tk.The example above used an even insulation thicknessof 3 in. Some products are manufactured to sucheven thicknesses, and Table 15.2 lists the Eq tk for suchproducts. However, many products are manufactured to“simplified” thicknesses, which allow a proper fit whennesting double-layer materials. ASTM-C-5859 liststhese standard dimensions, and Table 15.3 shows Eqtk values for the simplified thicknesses. Figure 15.3 alsoshows the conversion for any thickness desired and willbe used later in the reverse fashion.Surface ResistanceThere is always diversity of opinion when it comesto selecting the proper values for the surface resistanceR s . The surface resistance is affected by surface emittance,surface air velocity, and the surrounding environment.Heat-transfer texts have developed procedures forcalculating Rs values, but they are all based on speculatedvalues of emittance and air velocity. In actuality,the emittance of a surface often changes with time,temperature, and surface contamination, such as dust.As a result, it is unnecessary to labor over calculatingspecific R s values, when the conditions are estimates atbest.Table 15.4 lists a series of R s values based on threedifferent surface conditions and the temperature differencebetween the surface and ambient air. Also includedare single-point R s values for three different surface airvelocities. See the note at the bottom of Table 15.4 relatingto the effect of R s on heat-transfer calculations.15.4.3 Personnel ProtectionWorkers need to be protected from high-temperaturepiping and equipment in order to prevent skinburns. Before energy conservation analyses becamecommonplace, many insulation systems were designedsimply to maintain a “safe-touch” temperature on theouter jacket. Now, with energy costs so high, personnelprotection calculations are generally limited to temporaryinstallations or waste-heat systems, where theenergy being transferred will not be further utilized.Normally, safe-touch temperatures are specifiedin the range 130 to 150°F, with 140°F being usedmost often. It is important to remember that the

INDUSTRIAL INSULATION 451surface temperature is directly related tothe surface resistance R s , which in turndepends on the emittance of the surface.As a result, an aluminum jacket will behotter than a dull mastic coating overthe same amount of insulation. This isdemonstrated below.CalculationThe objective is to calculate the amount ofinsulation required to attain a specific surfacetemperature. As noted earlier,Therefore,t h – t sR 1= t s – t aR sR 1 = R st h – t st s – t a= tk kflat =Eq tkkpipeFig. 15.3 Equivalent thickness chart. (From Ref. 16.)Therefore,tk or Eq tk = kR st h – t st s – t aTable 15.2 Equivalent thickness values for even insulation thicknesses.Actual Thickness (in.)Nominal PipeSize (in.) r 1 1 1-1/2 2 2-1/2 3 3-1/2 41/2 0.420 1.730 2.918 4.238 5.662 7.172 8.755 10.4023/4 0.525 1.626 2.734 3.966 5.297 6.712 8.199 9.7471 0.658 1.532 2.563 3.711 4.953 6.275 7.665 9.1171-1/4 0.830 1.447 2.405 3.472 4.626 5.856 7.153 8.5071-1/2 0.950 1.403 2.321 3.342 4.449 5.629 6.872 8.1712 1.188 1.337 2.195 3.148 4.177 5.276 6.436 7.6482-1/2 1.438 1.287 2.099 2.997 3.968 5.001 6.093 7.2343 1.750 1.242 2.012 2.858 3.771 4.742 5.768 6.8403-1/2 2.000 1.217 1.959 2.772 3.649 4.582 5.564 6.5924 2.250 1.194 1.916 2.704 3.549 4.448 5.396 6.3864-1/2 2.500 1.178 1.880 2.645 3.464 4.337 5.253 6.2115 2.781 1.163 1.846 2.590 3.388 4.231 5.118 6.0436 3.313 1.138 1.799 2.510 3.270 4.071 4.911 5.7907 3.813 1.120 1.761 2.453 3.184 3.956 4.759 5.6048 4.313 1.108 1.737 2.407 3.116 3.863 4.644 5.4529 4.813 1.097 1.714 2.369 3.056 3.783 4.541 5.33010 5.375 1.088 1.693 2.333 3.007 3.714 4.450 5.21411 5.875 1.079 1.675 2.305 2.972 3.663 4.383 5.12312 6.375 1.076 1.662 2.286 2.936 3.619 4.321 5.04814 7.000 1.069 1.647 2.265 2.900 3.569 4.258 4.96916 8.000 1.059 1.639 2.231 2.858 3.504 4.178 4.86618 9.000 1.053 1.622 2.206 2.822 3.449 4.110 4.776201 0.000 1.048 1.608 2.188 2.789 3.411 4.051 4.71124 12.000 1.040 1.589 2.163 2.736 3.347 3.971 4.59830 15.000 1.032 1.572 2.122 2.704 3.281 3.874 4.497Source: Ref. 1 6.

452 ENERGY MANAGEMENT HANDBOOKTable 15.3 Equivalent thickness values for simplified insulation thicknesses.Actual Thickness (in.)Nominal PipeSize (in.) r1 1 1-1/2 2 2-1/2 3 3-1/2 41/2 0.420 1.730 3.053 4.406 6.787 8.253 9.972 12.7123/4 0.523 1.435 2.660 3.885 5.996 7.447 8.965 10.6421 0.638 1.715 2.770 4.013 5.358 6.702 8.112 9.5811-1/4 0.830 1.281 2.727 3.333 4.552 5.777 7.070 8.4201-1/2 0.950 1.457 2.382 4.025 5.253 6.476 7.759 9.1792 1.188 1.438 2.367 3.398 4.446 5.561 6.733 8.0272-1/2 1.438 1.383 2.765 3.657 4.737 5.815 7.015 8.1953 1.750 1.286 2.114 2.968 3.889 4.868 5.965 7.0463-1/2 2.000 1.625 2.459 3.258 4.166 5.251 6.266 7.2564 2.230 1.281 2.010 2.806 3.659 4.059 5.577 6 5434-1/2 2.300 1.564 2.351 3.152 4.905 4.962 5.907 7 0805 2.781 1.202 1.893 2.639 3.489 4.339 5.230 6.4616 3.313 1.138 1.799 2.555 3.317 4.122 5.237 6.0157 3.813 1.804 2.495 3.230 4.153 4.969 5.8218 4.313 1.776 2.445 3.391 4.010 4.842 5.7689 4.813 1.752 2.579 3.232 3.971 4.786 5.58310 5.375 1.810 2.457 3.108 3.850 4.591 5.36111 5.875 1.793 2.428 3.140 3.793 4.519 5.27112 6.375 1.777 2.405 3.103 3.745 4.456 5.24114 7.000 1.647 2.265 2.900 3.569 4.258 4.96916 8.000 1.639 2.231 2.858 3.504 4.178 4.86618 9.000 1.622 2.206 2.822 3.449 4.110 4.77620 10.000 1.608 2.188 2.789 3.411 4.051 4.71124 12.000 1.589 2.163 2.736 3.347 3.971 4.59830 15.000 1.572 2.122 2.704 3.281 3.874 4.497Source: Ref. 16.Table 15.4 R s Values a (hr • ft 2 °F/Btu).Still AirPlain, Fabric,t s – t a Dull Metal: Aluminum: Stainless Steel:(°F) ε = 0.95 ε = 0.2 ε = 0.410 0.53 0.90 0.8125 0.52 0.88 0.7950 0.50 0.86 0.7675 0.48 0.84 0 75100 0.46 0.80 0 72With Wind VelocitiesWind Velocity(mph)5 0.35 0.41 0.4010 0.30 0.35 0.3420 0.24 0.28 0.27Source: Courtesy of Johns-Manville, Ref. 16.a For heat-loss calculations, the effect of R s is small compared to R I , so the accuracy of R s is not critical. For surface temperature calculations, Rs isthe controlling factor and is therefore quite critical. The values presented in Table 15.4 are commonly used values for piping and flat surfaces. Moreprecise values based on surface emittance and wind velocity can be found in the references.

INDUSTRIAL INSULATION 453Example. For a 4-in. pipe operating at 700°F in an 85°Fambient temperature with aluminum jacketing over theinsulation, determine the thickness of calcium silicatethat will keep the surface temperature below 140°F.Since this is a pipe, the equivalent thickness must firstbe calculated and then converted to actual thickness.STEP 1. Determine k at t m = (700 + 140)/2 = 420°F. k =0.49 from Table 15.1 or appendix Figure 15.A1 for calciumsilicate.the insulation equal to the heat loss off the surface, followingthe discussion in Section 15.4.2.Figure 15.4 will be used for several different calculations.The following example gives the four-stepprocedure for achieving the desired surface temperaturefor personnel protection. The accompanying diagramoutlines this procedure.STEP 2. Determine R s from Table 15.4 for aluminum.t s – t a = 140 – 85 = 55. So R s = 0.85.STEP 3. Calculate Eq tk:Eq tk = (0.49)(0.85)700 – 140—————140 – 85= 4.24 in.STEP 4. Determine the actual thickness from Table 15.2.The effect of 4.24 in. on a 4-in. pipe can be accomplishedby using 3 in. of insulation.Note: Thickness recommendations are alwaysincreased to the next 1-in. increment. If a surface temperaturecalculation happens to fall precisely on an evenincrement (such as 3 in.), it is advisable to be conservativeand increase to the next increment (such as 3-1/2in.). This reduces the criticality of the Rs number used.In the preceding example, it would not be unreasonableto recommend 3-1/2 in. of insulation, since it was foundto be so close to 3 in.To illustrate the effect of surface type, consider hesame example with a mastic coating.Example. From Table 15.4, R s = 0.50, so700 – 140Eq tk = (0.49)(0.50) —————140 – 85= 2.49 in.This corresponds to an actual thickness requirementon a 4-in. pipe of 2 in. This compares with 3 in.required for an aluminum-jacketed system. It is of interestto note that even though the aluminum system hasa higher surface temperature, the actual heat loss is lessbecause of the higher surface resistance value.Graphical MethodThe calculations illustrated above can also be carriedout using graphs which set the heat loss throughExample. We follow the procedure of the first example,again using aluminum jacketing.STEP 1. Determine t s – t a , 140 – 85 = 55°F.STEP 2. In the diagram, proceed vertically from (a)of ∆t = 55 to the curve for aluminum jacketing (b).STEP 2a. Although not required, read the heat lossQ = 65 Btu/hr ft 2 ) (c).STEP 3. Proceed to the right to (d), the appropriatecurve for t h – t s = 700 – 140 = 560°F. Interpolate betweenlines as necessary.STEP 4. Proceed down to read the required insulationresistance R t = 8.6 at (e). Since R = tk/k or Eq tk/k,tk or Eq tk = R I k700 + 140t m = ——————— = 420°F2k = 0.49 from appendix Figure 15.A1 andtk or Eq tk = (8.6) (0.49) = 4.21 in.which compares well with the 4.24 in. from the earliercalculation.The conversion of Eq tk to actual thickness requiredfor pipe insulation is done in the same manner,using Figure 15.3.A better understanding of the procedure involvedin utilizing this quick graphical method will be obtained

454 ENERGY MANAGEMENT HANDBOOKafter working through the remainder of the calculationsin this section.15.4.4 Condensation ControlOn cold systems, either piping or equipment, insulationmust be employed to prevent moisture in thewarmer surrounding air from condensing on the coldersurfaces. The insulation must be of sufficient thicknessto keep the insulation surface temperature above thedew point of the surrounding air. Essentially, the calculationprocedures are identical to those for personnelprotection except that the dew-point temperature issubstituted for the desired surface temperature. (Note:The surface temperature should be kept 1 or 2° abovethe dew point to prevent condensation at that temperature.)Dew-Point DeterminationThe condensation (saturation) temperature, or dewpoint, is dependent on the ambient dry-bulb and wetbulbtemperatures. With these two values and the use ofa psychrometric chart, the dew point can be determined.However, for most applications, the relative humidity ismore readily attainable, so the dew point is determinedusing dry-bulb temperature and relative humidity ratherthan wet-bulb. Table 15.5 is used to find the proper dewpointtemperature.CalculationThis equation is identical to the previous surfacetemperatureproblem except that the surface temperaturets now takes on the value of the dew point of theambient air. Also, t h now represents the cold operatingtemperature.ttk or Eq tk = kR h –t sst s –t aFig. 15.4 Heat loss and surface temperature graphical method. (From Ref. 16.)

INDUSTRIAL INSULATION 455Example. For a 6-in.-diameter chilled-water line operatingat 35°F in an ambient of 90°F and 85% RH,determine the thickness of fiberglass pipe insulationwith a composite kraft paper jacket required to preventcondensation.STEP 1. Determine the dew point (DP) using eithera psychrometric chart or Table 15.5. DP at 90°F and 85%RH = 85°F. (In Step 5, the thickness is rounded up, whichyields a higher temp.)STEP 2. Determine k at t m = (35 + 85)/2 = 60°F. k at60°F = 0.23, from Table 15.1 or appendix Figure 15.A2.STEP 3. Determine Rs from Table 15.4. ∆t here is(t a , – t s ) rater than (t s – t a ), t a – t s = 90 – 85 = 5°F, R s =0.54.STEP 5. Determine the actual thickness from Figure15.2 for 6-in. pipe, 1.24 in. Eq tk. The actual thicknessis 1.5 in.Graphical MethodThe graphical procedures are as described inSection 15.4.3. As the applications become colder, it isapparent that the required insulation thicknesses willbecome larger, with R I values toward the right side ofFigure 15.4. It is suggested that the graphical procedurenot be used when the resulting R I values must be determinedfrom a very flat portion of the (t h – t s ) curve(anytime the numbers are to the far right of Figure 15.4).It is difficult to read the graph with sufficient accuracy,particularly in light of the simplicity of the mathematicalcalculation.STEP 4. Calculate Eq tk.Eq tk = 0.23 0.54=1.24in.35 – 8585 – 90Thickness Chart for Fiberglass Pipe InsulationTable 15.6 gives the thickness requirements for fiberglasspipe insulation with a white, all-purpose jacketin still air. The calculations are based on the lowestTable 15.5 Dew-point temperature.Dry-BulbPercent Relative HumidityTemp.(°F) 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 1005 – 35 – 30 – 25 – 21 – 17 – 14 – 12 – 10 – 8 – 6 – 5 – 4 – 2 – 1 1 2 3 4 510 – 31 – 25 – 20 – 16 – 13 – 10 – 7 – 5 – 3 – 2 0 2 3 4 5 7 8 9 1015 – 28 – 21 – 16 – 12 – 8 – 5 – 3 – 1 1 3 5 6 8 9 10 12 13 14 1520 – 24 – 16 – 8 – 4 – 2 2 4 6 8 10 11 13 14 15 16 18 19 2025 – 20 – 15 – 8 – 4 0 3 6 8 10 12 15 16 18 19 20 21 23 24 2530 – 15 – 9 – 3 2 5 8 11 13 15 17 20 22 23 24 25 27 28 29 3035 – 12 – 5 1 5 9 12 15 18 20 22 24 26 27 28 30 32 33 34 3540 – 7 0 5 9 14 16 19 22 24 26 28 29 31 33 35 36 38 39 4045 – 4 3 9 13 17 20 23 25 28 30 32 34 36 38 39 41 43 44 4550 – 1 7 13 17 21 24 27 30 32 34 37 39 41 42 44 45 47 49 5055 3 11 16 21 25 28 32 34 37 39 41 43 45 47 49 50 52 53 5560 6 14 20 25 29 32 35 39 42 44 46 48 50 52 54 55 57 59 6065 10 18 24 28 33 38 40 43 46 49 51 53 55 57 59 60 62 63 6570 13 21 28 33 37 41 45 48 50 53 55 57 60 62 64 65 67 68 7075 17 25 32 37 42 46 49 52 55 57 60 62 64 66 69 70 72 74 7580 20 29 35 41 46 50 54 57 60 62 65 67 69 72 74 75 77 78 8085 23 32 40 45 50 54 58 61 64 67 69 72 74 76 78 80 82 83 8590 27 36 44 49 54 58 62 66 69 72 74 77 79 81 83 85 87 89 9095 30 40 48 54 59 63 67 70 73 76 79 82 84 86 88 90 91 93 95100 34 44 52 58 63 68 71 75 78 81 84 86 88 91 92 94 96 98 100105 38 48 56 62 67 72 76 79 82 85 88 90 93 95 97 99 101 103 105110 41 52 60 66 71 77 80 84 87 90 92 95 98 100 102 104 106 108 110115 45 56 64 70 75 80 84 88 91 94 97 100 102 105 107 109 111 113 115120 48 60 68 74 79 85 88 92 96 99 102 105 107 109 112 114 116 118 120125 52 63 72 78 84 89 93 97 100 104 107 109 111 114 117 119 121 123 125

456 ENERGY MANAGEMENT HANDBOOKtemperature in each temperature range. Three temperature/humidityconditions are depicted.15.4.5 Process ControlIncluded under this heading will be all the calculationsother than those for surface temperature andeconomics. It is often necessary to calculate the heat flowthrough a given insulatio n thickness, or conversely, tocalculate the thickness required to achieve a certain heatflow rate. The final situation to be addressed deals withtemperature drop in both stagnant and flowing systems.Heat Flow for a Specified ThicknessCalculation Equations. Again, the basic equationfor a single insulation material ist h – t aQ F = ————R I + R sExample. For an 850°F boiler operating indoors in an80°F ambient temperature insulated with 4 in. of calciumsilicate covered with 0.016 in. aluminum jacketing,determine the heat loss per square foot of boiler surfaceand the surface temperature.STEP 1. Find k for calcium silicate at t m . Assumethat t s = 140°F. Then t m = (850 + 140)/2 = 495°F, kat 495°F = 0.53, from Table 15-1 or appendix Figure15.A1.STEP 2. Determine R s for aluminum from Table15.4. t s – t a = 140 – 0 = 60°F, so R s = 0.85.STEP 3. Calculate R I = 4/0.53 = 7.5.STEP 4. Calculate850 – 80QF = —————— = 92 Btu/hr • ft27.5 + 0.85STEP 5. Calculate the surface temperature ts, asfollows:R s × Q F = t s – t a(R s × Q F ) + ta = t st s = (0.85 × 92) + 80= 158°FSTEP 6. Calculate t m to check assumption and tocheck the k value used.t m =850 – 802= 504°FSince k at 504°F = k at 495°F (assumed) = 0.53, the assumptionis okay. A check on R s can also be made basedon the calculated surface temperature.STEP 7. If the assumption is not okay, recalculateusing a new k value based on the new t m .The Q F used above is for flat surfaces. In determiningheat flow from a pipe, the same equations are usedwith Eq tk substituted for tk in the R I calculation asdiscussed in Section 15.4.2. Often, it is desired to expresspipe heat losses in terms of Btu/hr-lin.-ft rather thanTable 15.6 Fiberglass pipe insulation: minimum thickness to prevent condensation a.80°F and 90% RH 80°F and 70% RH 80°F and 50% RHOperating PipeTemperature Pipe Size Thickness Pipe Size Thickness Pipe Size Thickness(°F) (in.) (in.) (in.) (in.) (in.) (in.)0-34 Up to 1 2 Up to 8 1 Up to 8 1/21-1/4 to 2 2-l/2 9-30 1-1/2 9-30 12-1/2 to 8 39-30 3-1/235-49 Up to 1-1/2 1-1/2 Up to 4 1/2 Up to 30 1/22-8 2 412-309-30 350-70 Up to 3 1-1/2 Up to 30 1/2 Up to 30 1/23-1/2 to 20 221 -30 2-l/2Source: Courtesy of Johns-Manville, Ref. 16.a Based on still air and AP Jacket.

INDUSTRIAL INSULATION 457Btu/hr ft 2 . This is termed Qp, withQ P - Q F2πr 212Graphical Method. Figure 15.4 may again be usedin lieu of calculations. The main difference from theprevious chart usage is that surface temperature is nowan unknown, and must be determined such that thermalequilibrium exists.Example. Determine the heat loss from the side wallsof a vessel operating at 300°F in an 80°F ambient temperature.Two inches of 3-lb/ft 3 fiberglass is used withaluminum lagging.STEP 1. Assume a surface temperature t s = 120°F.STEP 2. Calculatet h = t s 300 + 120t m = ——— = ————— = 210°F2 2Determine k from appendix Figure 15.A3 at 210°F. k =0.27.STEP 3. Calculate R I = tk/k = 2/0.27 = 7.41.STEP 4. Go to position (a) on the chart shownfor RI = 7.41 and read vertically to (b), where t h – t s =180°F.200.5°F. Then find a new k = 0.26, which gives a new RI= 2/126 = 7.69.STEP 8. Return to step 4 with the new RI andproceed. This example shows the insensitivity of heatloss to changes in surface temperature since the new Q= 22 Btu/hr ft 2 .For pipe insulation, the same procedure is followedexcept that RI is calculated using the equivalentthickness. Also, conversion to heat loss per linear footmust be done separately after the square-foot loss isdetermined.Thickness for a Specified Heat LossAgain, a surface temperature t s must first be assumedad then checked for accuracy at the end of thecalculation.From Section 15.4.2,Q = t h – t sR Itk or Eq th = kt h – t sQ= t h – t stk/kwhere k is determined at t m = (t h + t s )/2.Exa mple. How much calcium silicate insulation is requiredon a 650°F duct in an 80°F ambient temperatureif the maximum heat loss is 50 Btu/hr ft 2 ? The insulationwill be finished with a mastic coating.STEP 1. Assume that t s = 105°F. So t m = (650 +105)/2 = 377°F. k from Table 15-1 or appendix Figure15.A1 at 377°F = 0.46.STEP 2. Find tk as follows:tk = kt h – t sQ=0.46650 – 10550STEP 5. Read to the left to (c) for heat loss Q = 24Btu/hr ft 2 .STEP 6. Read down from the proper surface curvefrom (d) to (e), which represents ts – ta, to check thesurface-temperature assumption. For aluminum, ts – ta(chart) is 21°F, compared with the 120 – 80 = 40°F assumption.STEP 7. Calculate a new surface temperature 80 +21 = 101°F; then calculate a new tm, = (300 + 101)/2 ==5.01in.STEP 3. Check surface temperature assumption byt s = (Q × R s ) + t a using R s = 0.52. From Table 15.4 for amastic finish,t s = 50(0.52) + 80= 106°F(Note that this in turn changes the t s – t a from 40 to25, which changes R s from 0.49 to 0.51, which is insignificant.)

458 ENERGY MANAGEMENT HANDBOOKFor a graphical solution to this problem, Figure 15.4is again used. It is simply a matter of reading across thedesired Q level and adjusting the t s and R I values to reachequilibrium. Thickness is then determined by tk = kR I .Temperature Drop in a SystemThe following discussion is quite simplified and isnot intended to replace the service of the process designengineer. The material is presented to illustrate howinsulation t ies into the process design decision.Temperature Drop in Stationary Media overTime. The procedure calls for standard heat-flow calculationsnow tied into the heat content of the fluid. Toillustrate, consider the following example.Example. A water storage tank is calculated to have asurface area of 400 ft 2 and a volume of 790 ft 3 . Howmuch will the temperature drop in a 72-hr period withan ambient temperature of 0°F, assuming that the initialwater temperature is 50°F? The tank is insulated with2-in. fiberglass with a mastic coating.Before proceeding, realize that the maximum heattransfer will occur when the water is at 50°F. As it dropsin temperature, the heat-transfer rate is reduced due to asmaller temperature difference. As a first approximation,it is reasonable to use the maximum heat transfer basedon 50°F. Then if the temperature drop is significant, anaverage water temperature can be used in the seconditeration.STEP 1. Assume a surface temperature, calculatethe mean temperature, find the k factor from Table 15.1or appendix Figure 15A.3, and determine R s from Table15.4. With t s = 10°F, m = 30°F, k = 0.22, and R s = 0.53.= 2080 × 72 = 149,760 Btu. Divide this by the availableheat per 1°F drop:149,760 Btu——————— = 3.04°F drop49,296 Btu/°FThis procedure may also be used for fluid lyingstationary in a pipeline. In this case it is easiest to doall the calculations for 1 linear foot rather than for theentire length of pipe.One conservative aspect of this calculation is thatthe heat capacity of the metal tank or pipe is not includedin the calculation. Since the container will haveto decrease in temperature with the fluid, there is actuallymore heat available than was used above.Temperature Drop in Flowing Media. There aretwo common situations in this category, the first involvingflue gases and the second involving water or otherfluids with a thickening or freezing point. This sectiondiscusses the flue-gas problem and the following section,freeze protection.A problem is encountered with flue gases that havefairly high condensation temperatures. Along the lengthof a duct run, the temperature will drop, so insulation isadded to control the temperature drop. This calculationis actually a heat balance between the mass flow rate ofenergy input and the heat loss energy outlet.For a round duct of radius r 1 and length L, gas entersat t h , and must not drop below t min (the dew point).The flow rate is M lb/hr and the gas has a specific heatof C p Btu/lb °F. Therefore, the maximum allowable heatloss in Btu/hr isQ t = MC p ∆t = MC p (t h – t min )STEP 2. Calculate heat loss with fluid at 50°F.Q F =50 – 0= 5.2 Btu/hr • ft22/0.22 + 0.53Q F = Q F × A = 5.2 400 = 2080 Btu/hrAlso,whereQ T = Q P × L = t h – t aR I – R s× 2Ér 212 × Lt h =t in + t out2STEP 3. Calculate the amount of heat that must belost for the entire volume of water to drop 1°F.Available heat per °F= volume × density × specific heat= 790 ft 3 × 62.4 lb/ft 3 × 1 Btu/lb F= 49,296 Btu/°FSTEP 4. Calculate the temperature drop in 72 hrby determining the total heat flow over the period: Q(A conservative simplification would be to set t h = t insince the higher temperature, t in , will cause a greaterheat loss.)To simplify on large ducts, assume that r 1 = r 2(ignore the insulation-thickness addition to the surfacearea). Therefore,andt h – t aR I – R s× 2Ér 112 × L = MC p t h – t min

INDUSTRIAL INSULATION 459R I + R s =Therefore,t h – t at h – t min× 2Ér 112 × LMC P570 – 60 2π24R I = ——————————— × ——— × 90 – 0.52(66,878) (0.18) (575–565) 12= 4.79 – 0.52= 4.27R I =t h – t aMC P t h – t min× 2Ér 112 × L – R s= tk k ∴ tk = R I × kExample. A 48-in.-diameter duct 90 ft long in a 60°Fambient temperature has gas entering at 575°F and15,000 cfm. The gas density standard conditions is 0.178lb/ft 3 and the gas outlet must not be below 555°F. Cp= 0.18 Btu/lb °F. Determine the thickness of calciumsilicate required to keep the outlet temperature above565°F, giving a 10°F buffer to account for the interiorfilm coefficient. A more sophisticated approach calculatesan interior film resistance R s (interior) instead ofusing a 10°F or larger buffer. The resulting equation forQp would beQ p =t h – t a× 2Ér 2R s inferior + R I + R s12This equation, however, will not be used.STEP 1. Determine t h the average gas temperature,= (575 + 565)/2 = 570°F. (A logarithmic mean could becalculated for more accuracy, but it is usually not necessary.)STEP 2. Determine M lb/hr. The flow rate is 15,000cfm of hot gas (570°F). At standard conditions 1 atm,70°F), the flow rate must be determined by the absolutetemperature ratio:t h + 46070 + 460 = 15,000std. flowStd. flow = 15,00070 + 460570 + 460= 6262 cfm std. gas or scfmM = 6262 cfm × 0.178 lb/ft 3 × 60 min/hr= 66,878 lb/hrSTEP 3. Determine Rs from Table 15.4 assuming t s= 80°F and a dull surface R s = 0.5.STEP 4. Calculate R I .STEP 5. Calculate the thickness. Assume that t s =80°F.570 + 80t m = ———— = 325°F20.45 for calcium silicate fromk at 325°F = Appendix Figure 15.A1.tk = R I × k = 4.27 × 0.45= 1.93 in.STEP 6. The thickness required for this applicationis 2 in. of calcium silicate. Again, a more conservativerecommendation would be 2-1/2 in.Note: The foregoing calculation is quite complex.It is, however, the basis for many process control andfreeze-prevention calculations. The two equations for Q,can be manipulated to solve for the following:Temperature drop, based on a given thickness andflow rate.Minimum flow rate, based on given thickness andtemperature drop.Minimum length, based on thickness, flow rate,and temperature drop.Freeze Protection. Four different calculations canbe performed with regard to water-line freezing (or theunacceptable thickening of any fluid).1. Determine the time required for a stagnant, insulatedwater line to reach 32°F.2. Determine the amount of heat tracing required toprevent freezing.3. Determine the flow rate required to prevent freezingof an insulated line.4. Determine the insulation required to prevent freezingof a line with a given flow rate.Calculations 1 and 2 relate to Section 15.4.5, where wedealt with stationary media. To apply the same principlesto the freeze problems, the following modificationsshould be made.a. In calculation 1, the heat transfer should be basedon the average water temperature between thestarting temperature and freezing:

460 ENERGY MANAGEMENT HANDBOOKt h = t start +322b. Rather than solving for temperature drop, giventhe number of hours, the hours are determinedbased onavailable heat Btuhours to freeze = —————— ————heat loss/hr Btu/hrwhere available heat is WCp ∆t, withW = lb of waterC p = specific heat of water (1 Btu/lb °F)∆t = t start – 32c. In calculation 2, the heat-loss value should becalculated based upon the minimum temperatureat which the system should stay, for example,35°F. The heat tracing should provide enoughheat to the system to offset the naturally occurringlosses of the pipe. Heat-trace calculations arequite complex and many variables are involved.References 8 and 10 should be consulted for thistype of work.Calculations 3 and 4 relate to Section 4.5.3,dealing with flows. In the case of water, theminimum temperature can beset at 32°F and theheat-transfer rate is again on an operating averagetemperaturet h = t start +322The equations given can be manipulated to solve forflow rate or insulation thickness.As an aid in estimating the amount of insulationfor freeze protection, Table 15.7 shows both the hours tofreezing and the minimum flow rate to prevent freezingbased on different insulation thicknesses. These figuresare based on an initial water temperature of 42°F, an ambienttemperature of – 10°F, a surface resistance of 0.54,and a thermal conductivity for fiberglass pipe insulationof k = 0.23.15.4.6 Operating ConditionsLike all other calculations, heat-transfer equationsyield results that are only as accurate as the inputvariables used. The operating conditions chosen for theheat-transfer calculations are critical to the result, andvery misleading conclusions can be drawn if improperconditions are selected.The term “operating conditions” refers to the environmentsurrounding the insulation system. Some of thevariable conditions are operating temperature, ambienttemperature, relative humidity, wind velocity, fluid type,mass flow rate, line length, material volume, and others.Since many of these variables are constantly changing,the selection of a proper value must be made on somelogical basis. Following are three suggested methods fordetermining the appropriate variable values.1. Worst Case. If a severe failure might occur withinsufficient insulation, a worst-case approach is probablywarranted. For example, freeze protection shouldobviously be based on the historical temperature extremesrather than on yearly averages. Similarly, exteriorcondensation control should be based on both ambienttemperature and humidity extremes in addition to thelowest operating temperature. The ASHRAE Handbook ofFundamentals as well as U.S. Weather Bureau data giveproper design conditions for most locales. In processareas, an appropriate example involves flue-gas condensation.Here the minimum flow rate is the most criticaland should be used in the calculation.As a general rule, worst-case conditions will resultin greater insulation thickness than will average conditions.In some cases the difference is very substantial,so it is important to determine initially if a worst-casecalculation is required.2. Worst Season Average. When a heating orcooling process is only operating part of a year, it issensible to consider the average conditions only duringthat period of time. However, in year-round operations,a seasonal average is also justified in many cases. Forexample, personnel protection requires a maximumsurface temperature that is dependent on the ambientair temperature. Taking the average summer dailymaximum temperature is more practical than takingthe absolute maximum ambient that could occur. Thefollowing example illustrates this.Example. Consider an 8-in.-diameter, 600°F waste-heatline operating indoors with an average daily high of80°F (but occasionally it will be 105°F). To maintain thesurface below 135°F, 2 in. of calcium silicate is requiredwith the 80°F ambient, whereas 3-1/2 in. is requiredwith the 105°F ambient. The difference is significant andmust be weighed against the benefit of the additionalinsulation in terms of worker safety.3. Yearly Average. Economic calculations forcontinuously operating equipment should be based

INDUSTRIAL INSULATION 461Table 15.7 Hours to freeze and flow rate required to prevent freezing a .1 in 2 in. 3 inNominalPipe Size Hours to Hours to Hours to(in.) Freeze gpm/100 ft Freeze gpm/100 ft Freeze gpm/100 ft1/2 0.30 0.087 0.42 0.282 0.50 0.0533/4 0.47 0.098 0.66 0.070 0.79 0.0581 0.66 0.113 0.96 0.078 1.16 0.0651-1/2 0.90 0.144 1.35 0.096 1.67 0.0782 1.72 0.169 2.64 0.110 3.31 0.0882-1/2 2.13 0.195 3.33 0.124 4.24 0.0983 2.81 0.228 4.50 0.142 5.80 0.1104 3.95 0.279 6.49 0.170 8.49 0.1305 5.21 0.332 8.69 0.199 11.54 0.1506 6.48 0.386 10.98 0.228 14.71 0.1707 7.66 0.437 13.14 0.255 17.75 0.1898 8.89 0.487 15.37 0.282 20.89 0.207Source: Ref. 16.a Calculations based on fiberglass pipe insulation with k = 0.23, initial water temperature of 42°F, and ambient air temperatureof – 10°F. Flow rate represents the gallons per minute required in a 100-ft pipe and may be prorated for longer or shorteron yearly average operating conditions rather thanon worst-case design conditions. Since the intent is tomaximize the owner’s financial return, an average conditionwill not overstate the savings as the worst case orworst season might. A good approach to process workis to calculate the economic thickness based on yearlyaverages and then check the sufficiency of that thicknessunder the worst-case design conditions. That way, bothcriteria are met.15.4.7 Bare-Surface Heat LossIt is often desirable to determine if any insulationis required and also to compare bare surface losses withthose using insulation. Table 15.8 gives bare-surfacelosses based on the temperature difference between thesurface and ambient air. Actual temperature conditionsbetween those listed can be arrived at by interpolation.To illustrate, consider a bare, 8-in.-diameter pipe operatingat 250°F in an 80°F ambient temperature. ∆t = 250– 80 = 170°F. Q for ∆t of 150°F = 812.5 Btu/ hr-lin.-ft;Q for ∆t of 200°F = 1203 Btu/hr lin. ft. Interpolatingbetween 150 and 200°F givesQ 170 = Q 150 + (2/5)(Q 200 – Q 150 )= 812.5 + 0.4(1203 – 812.5)= 968.7 Btu/hr lin. ft15.5 INSULATION ECONOMICSThermal insulation is a valuable tool in achievingenergy conservation. However, to strive for maximumenergy conservation without regard for economics isnot acceptable. There are many ways to manipulate thecost and savings numbers, and this section explains thevarious approaches and the pros and cons of each.15.5.1 Cost ConsiderationsSimply stated, if the cost of insulation can berecouped by a reduction in total energy costs, the insulationinvestment is justified. Similarly, if the cost ofadditional insulation can be recouped by the additionalenergy-cost reduction, the expenditure is justified. Thereis a significant difference between the “full thickness”justification and the “incremental” justification. This isdiscussed in detail in Section 15.5.3. The following discussionswill generally use the incremental approach toeconomic evaluation.Insulation CostsThe insulation costs should include everything thatit takes to apply the material to the pipe or vessel and toproperly cover it to finished form. Certainly, it is morecostly to install insulation 100 ft in the air than it is fromground level, and metal jackets are more costly than allpurposeindoor jackets. Anticipated maintenance costs

462 ENERGY MANAGEMENT HANDBOOKTable 15.8 Heat loss from bare surfaces a .Temperature Difference (°F)Normal PipeSize (in.) 50 100 150 200 250 300 350 400 450 500 550 600 700 800 900 10001/2 22 47 79 117 162 215 279 355 442 541 650 772 1,047 1,364 1,723 2,1233/4 27 59 99 147 203 269 349 444 552 677 812 965 1,309 1,705 2,153 2,6541 34 75 124 183 254 336 437 555 691 846 1,016 1,207 1,637 2,133 2,694 3,3201-1/4 42 94 157 232 321 425 552 702 873 1,070 1,285 1,527 2,071 2,697 3,406 4,1981-1/2 49 107 179 265 367 487 632 804 1,000 1,225 1,471 1,748 2,371 3,088 3,899 4,8062 61 134 224 332 459 608 790 1,004 1,249 1,530 1,837 2,183 2,961 3,856 4,870 6,0022-l/2 74 162 271 401 556 736 956 1,215 1,512 1,852 2,224 2,643 3,584 4,669 5,896 7,2673 89 197 330 489 677 897 1,164 1,480 1,841 2,256 2,708 3,219 4,365 5,685 7,180 8,8493-1/2 102 225 377 558 773 1,024 1,329 1,690 2,102 2,576 3,092 3,675 4,984 6,491 8,198 10,1004 115 254 424 628 869 1,152 1,496 1,901 2,365 2,898 3,479 4,135 5,607 7,304 9,224 11,3704-1/2 128 282 471 698 965 1,280 1,662 2,113 2,628 3,220 3,866 4,595 6,231 8,116 10,250 12,6305 142 313 524 776 1,074 1,424 1,848 2,350 2,923 3,582 4,300 5,111 6,931 9,027 11,400 14,0506 169 373 624 924 1,279 1,696 2,201 2,799 3,481 4,266 5,121 6,086 8,254 10,750 13,580 16,7307 195 430 719 1,064 1,473 1,952 2,534 3,222 4,007 4,910 5,894 7,006 9,501 12,380 15,630 19,2608 220 486 813 1,203 1,665 2,207 2,865 3,643 4,531 5,552 6,666 7,922 10,740 13,990 17,670 21,7809 246 542 907 1,343 1,859 2,464 3,198 4,066 5,057 6,197 7,440 8,842 11,990 15,620 19,720 24,31010 275 606 1,014 1,502 2,078 2,755 3,576 4,547 5,655 6,930 8,320 9,888 13,410 17,470 22,060 27,18011 300 661 1,106 1,638 2,267 3,005 3,901 4,960 6,169 7,560 9,076 10,790 14,630 19,050 24,060 29,66012 326 718 1,202 1,779 2,463 3,265 4,238 5,338 6,701 8,212 9,859 11,720 15,890 20,700 26,140 32,21014 357 783 1,319 1,952 2,703 3,582 4,650 5,912 7,354 9,011 10,820 12,860 17,440 22,710 28,680 35,35016 408 901 1,508 2,232 3,090 4,096 5,317 6,759 8,407 10,300 12,370 14,700 19,940 25,970 32,790 40,41018 460 1,015 1,698 2,514 3,480 4,612 5,987 7,612 9,467 11,600 13,930 16,550 22,450 29,240 36,930 45,51020 510 1,127 1,885 2,790 3,862 5,120 6,646 8,449 10,510 12,880 15,460 18,380 24,920 32,460 40,990 50,52024 613 1,353 2,263 3,350 4,638 6,148 7,980 10,150 12,620 15,460 18,570 22,060 29,920 38,970 49,220 60,66030 766 1,690 2,827 4,186 5,795 7,681 9,971 12,680 15,770 19,320 23,200 27,570 37,390 48,700 61,500 75,790Flat 98 215 360 533 738 978 1,270 1,614 2,008 2,460 2,954 3,510 4,760 6,200 7,830 9,650Source: Ref. 16.a Losses given in Btu/hr lin. ft of bare pipe at various temperature differences and Btu/hr-ft 2 for flat surfaces. Heat losses were calculated forstill air and ε = 0.95 (plain, fabric or dull metals).should also be included based on the material and applicationinvolved. The variations in labor costs due toboth time and base rate should be evaluated for eachparticular insulation system design and locale. In otherwords, insulation costs tend to be job specific as well asbeing differentiated by product.Lost Heat CostsReducing the amount of unwanted heat loss is thefunction of insulation, and the measurement of this isin Btu. The key to economic analyses rests in the dollarvalue assigned to each Btu that is wasted. At the veryleast, the energy cost must include the raw-fuel cost,modified by the conversion efficiency of the equipment.For example, if natural gas costs $2.50/million Btu and itis being converted to heat at 70% efficiency, the effectivecost of the Btu is 2.50/0.70 = $3.57/million Btu.The cost of the heat plant is always a point of discussion.Many calculations ignore this capital cost on thebasis that a heat plant will be required whether insulationis used or not. On the other hand, the only purposeof the heat plant is to generate usable Btus. So the cost ofeach Btu should reflect the capital plant cost ammortizedover the life of the plant. The recent trend that seems mostreasonable is to assign an incremental cost to increasesin capital expenditures. This cost is stated as dollars per1000 Btu per hour. This gives credit to a well-insulatedsystem that requires less Btu/hr capacity.Other CostsAs the economic calculations become more sophisticated,other costs must be included in the analysis. Themajor additions are the cost of money and the tax effectof the project. Involving the cost of money recognizesthe real fact that many projects are competing for eachinvestment dollar spent.Therefore, the money used to finance an insulationproject must generate a sufficient after-tax return or the

INDUSTRIAL INSULATION 463money will be invested elsewhere to achieve such a return.This topic, together with an explanation of the useof discount factors, is discussed in detail in Chapter 4.The effect of taxes can also be included in theanalysis as it relates to fuel expense and depreciation.Since both of these items are expensed annually, theafter-tax cost is significantly reduced. The final examplein Section 15.5.3 illustrates this.15.5.2. Energy Savings CalculationsThe following procedure shows how to estimatethe energy cost savings resulting from installing thermalinsulation.ProcedureSTEP 1. Calculate present heat losses (Q Tpres ). Youcan use one of the following methods to calculate theheat losses of the present system:• Heat flow equations. These equations are in Section15.4.2.• Graphical method. Consists of Steps 1, 2 and 2a ofthe graphical method presented in Section 15.4.3.• Table values. Table 15.8 presents heat losses valuesfor bare surfaces (dull metals).STEP 2. Determine insulation thickness (tk).Using Section 15.4, you can determine the insulationthickness according to your specific needs. Dependingon the pipe diameter and temperature, the first inch ofinsulation can reduce bare surface heat losses by approximately85-95% (Ref. 20). Then, for a preliminaryeconomic evaluation, you can use tk = 1-in. If the evaluationis not favorable, you will not be able to justify athicker insulation. On the other hand, if the evaluationis favorable, you will need to determine the appropriateinsulation thickness and reevaluate the investment.STEP 3. Calculate heat losses with insulation(Q Tins ). Use the equations from Section 15.4.5.STEP 4. Determine heat loss savings (Q Tsavings ).Subtract the heat losses with insulation from the presentheat losses (Q Tsavings = Q Tpres - Q Tins )STEP 5. Estimate fuel cost savings. Estimate theamount of fuel used to generate each Btu wasted anduse this value to calculate the energy cost savings. Withthis savings, you can evaluate the insulation investmentusing any appropriate financial analysis method (seeSection 15.5.3).Example. For the example presented in section 15.4.3,determine the fuel cost savings resulting from insulatingthe pipe with 3-1/2 in. of calcium silicate.Data• Pipe data: 4-in pipe operating at 700°F in an 85°Fambient temperature.• Jacket type: Aluminum.• Pipe length: 100-ft.• Operating hours: 4,160 hr/yr• Fuel data: Natural gas, burned to heat the fluid inthe pipe at $3/MCF. Efficiency of combustion isapproximately 80%STEP 1. Determine present heat loss. From Table15.8 (4-in. pipe, temperature difference = t s – t a = 700– 85 = 615°F), heat loss = 4,356 Btu/hr-lin.ft. Then,Q Tpres = (heat loss/lin.ft)(length)= (4,356 Btu/hr-lin.ft.)(100 ft)= 435,600 Btu/hrSTEP 2. Determine insulation thickness. In thisexample, the surface temperature has to be below 140°F,which is accomplished with an insulation thickness = tk= 3.5-in.STEP 3. Determine heat losses with insulation. Forthis example, we need to calculate the heat losses fortk = 3.5-in following the procedure outlined in Section15.4.5.1) From the example in Section 15.4.3, t s = 140 °F, k= 0.49 and R s = 0.85.2) From Table 15.2, Eq tk for 3-1/2-in insulation n a4-in. pipe = 5.396 in. Then,R I =Eq tk/k =5.396/0.49= 113) Calculate heat loss Q F :700 – 85Q F = —————— = 52 Btu/hr ft 211 + 0.854) Calculate surface temperature t s :t s = t a + R s × Q F = 85 + (0.85 × 52) = 129°F5) Calculate t m = (700+129)/2 = 415°F. The insulationthermal conductivity at 415°F is 0.49, which is close

464 ENERGY MANAGEMENT HANDBOOKenough to the assumed value (see Appendix 15.1).Then, Q F = 52 Btu/hr ft 2 .6) Determine the outside area of insulated pipe. FromTable 15.2, pipe radius = rl = 2.25-in., then, outsideinsulated area (ft 2 ):= 2π (rl+tk)(length)/(12 in./ft)= 2π (2.25 in+3.5 in.)(100 ft)/(12 in./ft)= 301 ft 27) Calculate heat losses with insulation:Q Tins = (Q F )(outside area)= (52 Btu/hr ft 2 )(301 ft 2 )= 15,652 Btu/hrSTEP 4. Determine heat losses savings Q Tsavings:Q Tsavings = (Q Tpres – Q Tins )(hr/yr)= (435,600–15,652 Btu/hr)(4,160 hr/yr)(1 MMBtu/10 6 Btu)= 1,747 MMBtu/yrSTEP 5. Determine fuel cost savings. Assuming 1MCF = 1 MMBtu:Fuel savings= (Q Tsavings )(conversion factor)/(combustion efficiency)= (1,747 MMBtu/yr)(1 MCF/MMBtu)/(0.8)= 2,184 MCF/yrThen,Fuel cost savings = (fuel savings) (fuel cost)= (2,184 MCF/yr) ($3/MCF)= $6,552/yr15.5.3 Financial Analysis Methods—Sample CalculationsChapter 4 offers a complete discussion of thevarious types of financial analyses commonly used inindustry. A review of that material is suggested here, asthe methods discussed below rely on this basic understanding.To select the proper financial analysis requires anunderstanding of the degree of sophistication requiredby the decision maker. In some cases, a quick estimate ofprofitability is all that is required. At other times, a verydetailed cash flow analysis is in order. The importantpoint is to determine what level of analysis is desiredand then seek to communicate at that level. Followingis an abbreviated discussion of four primary methods ofevaluating an insulation investment: (1) simple payback;(2) discounted payback; (3) minimum annual cost usinga level annual equivalent; and (4) present-value costanalysis using discounted cash flows.Economic CalculationsBasically, a simple payback period is the time requiredto repay the initial capital investment with theoperating savings attributed to that investment. Forexample, consider the possibility of upgrading a presentinsulation thickness standard.ThicknessCurrent UpgradedStandard Thickness Difference—————————————————————————Insulationinvestment ($) 225,000 275,000 50,000Annual fuel cost ($) 40,000 30,000 10,000investment difference 50,000Simple payback = —————————— = ———— = 5.0 yearsannual fuel saving 10,000—————————————————————————This calculation represents the incremental approach,which determines the amount of time to recover theadditional $50,000 of investment.In the following table, the full thickness analysisis similar except that the upgraded thickness numbersare now compared to an uninsulated system with zeroinsulation investment.Uninsulated UpgradedSystem Thickness Difference—————————————————————————Insulationinvestment ($) 0 275,000 275,000Annual fuel cost ($) 340,000 30,000 310,000275,000Simple payback = ————— = 0.89 year310,000—————————————————————————The magnitude of the difference points out the dangerin talking about payback without a proper definitionof terms. If in the second example, management hada payback requirement of 3 years, the full insulation

INDUSTRIAL INSULATION 465investment easily complies, whereas the incrementalinvestment does not. Therefore, it is very important tounderstand the intent and meaning behind the paybackrequirement.Although simple payback is the easiest financialcalculation to make, its use is normally limited to roughestimating and the determination of a level of financialrisk for a certain investment. The main drawback withthis simple analysis is that it does not take into accountthe time value of money, a very important financialconsideration.Time Value of MoneyAgain, see Chapter 4. The significance of the costof money is often ignored or underestimated by thosewho are not involved in their company’s financial mainstream.The following methods of financial analysis areall predicated on the use of discount factors that reflectthe cost of money to the firm. Table 15.9 is an abbreviatedtable of present-value factors for a steady incomestream over a number of years. Complete tables arefound in Chapter 4.Discounted PaybackAlthough similar to simple payback, the utilizationof the discount factor makes the savings in future yearsworth less in present-value terms. For discounted payback,then, the annual savings times the discount factormust now equal the investment to achieve payback inpresent-value dollars. Using the same example:—————————————————————————ThicknessCurrent UpgradedStandard Thickness DifferenceInsulationinvestment ($) 225,000 275,000 50,000Annual fuel cost ($) 40,000 30,000 10,000Now, payback occurs when:investment = discount factor × annual savings50,000 = (discount factor) × 10,000so solving for the discount factor,investment 50,000discount factor = —————— = ——— = 5.0annual savings 10,000—————————————————————————For a 15% cost of money, read down the 15% columnof Table 15.9 to find a discount factor close to 5.The corresponding number of years is then read to theleft, approximately 10 years in this case. For a cost ofmoney of only 5%, the payback is achieved in about 6years. Obviously, a 0% cost of money would be the sameas the simple payback calculation of 5 years.Minimum Annual Cost AnalysisAs previously discussed, an insulation investmentmust involve a lump-sum cost for insulation as well as astream of fuel costs over the many years. One method ofputting these two sets of costs into the same terms is tospread out the insulation investment over the life of theproject. This is done by dividing the initial investmentby the appropriate discount factor in Table 15.9. Thisproduces a “level annual equivalent” of the investmentfor each year which can then be added to the annualfuel cost to arrive at a total annual cost.Utilizing the same example with a 20-year projectlife and 10% cost of money:ThicknessCurrent UpgradedStandard Thickness—————————————————————————Insulation investment ($) 225,000 275,000For 20 years at 10%, the discount factor is 8.514 (Table15.9), soEquivalent annual insulation costs 225,000 275,0008.514 8.514= 26,427 32,300Annual fuel cost ($) 40,000 30,000Total annual cost ($) 66,427 62,300—————————————————————————Therefore, on an annual cost basis, the upgraded thicknessis a worthwhile investment because it reduces theTable 15.9 Present-Value Discount Factors for an Incomeof $1 Per Year for the Next n Years—————————————————————————Cost of Money at:Years 5% 10% 15% 20% 25%—————————————————————————1 .952 .909 .870 .833 .8002 1.859 1.736 1.626 1.528 1.4403 2.723 2.487 2.283 2.106 1.9524 3.546 3.170 2.855 2.589 2.3625 4.329 3.791 3.352 2.991 2.68910 7.722 6.145 5.019 4.192 3.57115 10.380 7.606 5.847 4.675 3.85920 12.460 8.514 6.259 4.870 3.954—————————————————————————

466 ENERGY MANAGEMENT HANDBOOKannual costs by $4127.Now, to illustrate again the importance of usinga proper cost of money, change the 10% to 20% andrecompute the annual cost. The 20% discount factor is4.870.ThicknessCurrent UpgradedStandard ThicknessEquivalent annual insulation cost ($) 225,000 275,0004.870 4.870= 46,201 56,468Add the annual fuel cost ($) 40,000 30,000Total annual cost 86,201 86,468In this case, the higher cost of money causes the upgradedannual cost to be greater than the current cost,so the project is not justified.Present-Value Cost AnalysisThe other method of comparing project costs is tobring all the future costs (i.e., fuel expenditures) back totoday’s dollars by discounting and then adding this tothe initial investment. This provides the total presentvaluecost of the project over its entire life cycle, andprojects can be chosen based on the minimum presentvaluecost. This discounted cash flow (DCF) techniqueis used regularly by many companies because it allowsthe analyst to view a project’s total cost rather than justthe annual cost and assists in prioritizing among manyprojects.Thickness Current UpgradedStandard Thickness—————————————————————————Annual fuel cost ($) 40,000 30,000For 20 years at 10% the discount factor is 8.514 (Table15.9), soPresent value of fuelcost over 20 years 40,000 × 8.514 30,000 × 8.514= 340,560 255,420Insulation investment ($) 225,000 275,000Total present-value cost of $565,460 $530,420insulation project—————————————————————————Again, the lower total project cost with the upgradedthickness option justifies that project.So far, the effect of taxes and depreciation has beenignored so as to concentrate on the fundamentals. However,the tax effects are very significant on the cash flowto the company and should not be ignored. In the casewhere the insulation investment is capitalized utilizinga 20-year straight-line depreciation schedule and a 48%tax rate, the following effects are seen (see table at topof next page).This illustrates the significant impact of both taxesand depreciation. In the preceding analysis, the PVbenefit of upgrading was (565,460 – 530,420) = $35,040.In this case, the cash flow benefit is reduced to (356,116– 351,626) = $4490.The final area of concern relates to future increasesin fuel costs. So far, all the analyses have assumed aconstant stream of fuel costs, implying no increase inthe base cost of fuel. This assumption allows the useof the PV factor in Table 15.9. To accommodate annualfuel-price increases, either an average fuel cost over theproject life is used or each year’s fuel cost is discountedseparately to PV terms. Computerized calculationspermit this, whereas a manual approach would be extremelylaborious.15.5.4 Economic Thickness (ETI) CalculationsSection 15.5.3 developed the financial analysesoften used in evaluating a specific insulation investment.As presented, however, the methods evaluateonly two options rather than a series of thickness options.Economic thickness calculations are designed toevaluate each 1/2-in. increment and sum the insulationand operating costs for each increment. Then the optionwith the lowest total annual cost is selected as theeconomic thickness. Figure 15.5 graphically illustratesthe optimization method. In addition, it shows the effectof additional labor required for double- and triple-layerinsulation applications.Mathematically, the lowest point on the total-costcurve is reached when the incremental insulation costequals the incremental reduction in energy cost. Bydefinition, the economic thickness is:that thickness of insulation at which the cost of the nextincrement is just offset by the energy savings due to thatincrement over the life of the project.Historical developmentA problem with the McMillan approach was the largenumber of charts that were needed to deal with all the operatingand financial variables. In 1949, Union Carbide Corp.in a cooperation with West Virginia University establisheda committee headed by W.C. Turner to establish practicallimits for the many variables and to develop a manual forperforming the calculations. This was done, and in 1961,the manual was published by the National InsulationManufacturers Association (previously called TIMA and

INDUSTRIAL INSULATION 467ThicknessCurrent StandardUpgraded Thickness1. Annual energy cost ($) 40,000 30,0002. After-tax energy cost ($) ((1) × (1.0 – 0.48)) 20,800 15,6003. Insulation depreciation ($ tax benefit)(225,000/20 yr)(0.48) 5,400(275,000/20 yr)(0.48) 6,6004. Net annual cash costs [$; (2) – (3)] 15,400 9,0005. Present-value factor for 20 years at 10% = 8.5146. Present value of annual cash flows [$; (4) × (5)] 131,116 76,6267. Present value of cash flow for insulation purchase ($) 225,000 275,0008. Present-value cost of project [$; (6) + (7)] 356,116 351,626Fig. 15.5 Economic thickness of insulation (ETI) concept.now NAIMA, North American Insulation ManufacturersAssociation). The manual was entitled How to DetermineEconomic Thickness of Insulation and employed a numberof nomographs and charts for manually performing thecalculations.Since that time, the use of computers has greatlychanged the method of ETI calculations. In 1973, TIMAreleased several programs to aid the design engineer inselecting the proper amount of insulation. Then in 1976,the Federal Energy Administration (FEA) published a nomographmanual entitled Economic Thickness of IndustrialInsulation (Conservation Paper #46). In 1980, these manualmethods were computerized into the “Economic Thicknessof Industrial Insulation for Hot and Cold Surfaces.” Throughthe years, NAIMA developed a version for personal computers;the newest program was renamed 3EPLUS andcalculates the ETI thickness of insulation.Perhaps the most significant change occurring is thatmost large owners and consulting engineers are developingand using their own economic analysis programs,

468 ENERGY MANAGEMENT HANDBOOKspecifically tailored to their needs. As both heat-transferand financial calculations become more sophisticated, theseprograms will continue to be upgraded and their usefulnessin the design phase will increase.Nomograph MethodsA nomograph methods is not presented here, but theinterested reader can review the following references:• FEA manual (Ref. 12). This manual provides a fairlycomplete but time-consuming nomograph method.• 1972 ASHRAE Handbook of Fundamentals, Chapter17 (Ref. 13) which provides a simplified, one-pagenomograph. This approach is satisfactory for a quickdetermination, but it lacks the versatility of the morecomplex approach. The nomograph has been eliminatedin the latest edition and reference is made tothe computer analyses and the FEA manual.Computer ProgramsSeveral insulation manufacturers offer to run theanalysis for their customers. Also, computer programs suchas the 3EPLUS are available for customers who want torun the analysis on their own. The 3EPLUS software is anETI program developed by the North American InsulationManufacturers Association and the Steam Challenge Program.The program, available for free download (Ref. 14),calculates heat losses, energy and cost savings, thicknessfor maximum surface temperature and optimum thicknessof insulation.All the insulation owning costs are expressed on anequivalent uniform annual cost basis. This program usesthe ASTM C680 method for calculating the heat loss andsurface temperatures. Each commercially available thicknessis analyzed, and the thickness with the lowest annualcost is the economic thickness (ETI).Figure 15.6 shows the output generated by the NAIMA3EPLUS program. The first several lines are a readout ofthe input data. The different variables used in the programallow to simulate virtually any job condition. The sameprogram can be used for retrofit analyses and bare-surfacecalculations. There are two areas of input data that are notfully explained in the output. The first is the installed insulationcost. The user has the option of entering the installedcost for each particular thickness or using an estimatingprocedure developed by the FEA (now DOE).The second area that needs explanation is the insulationchoice, which relates to the thermal conductivity of thematerial. The example in Figure 15.6 shows the insulationas Glass Fiber Blanket. The program includes the thermalconductivity equations of several generic types of thermalinsulation, which were derived from ASTM materialsspecifications. The user has the option of supplying thermalconductivity data for other materials.The lower portion of the output supplies seven columnsof information. The first and second columns areinput data, while the others are calculated output. Theprogram also calculates the reduction in CO 2 emissions byinsulating to economic thickness. The meaning of columnstwo to seven of the output are explained below.Annual Cost ($/yr). This is the annual operating costincluding both energy cost and the amortized insulationcost. Tax effects are included. This value is the one thatdetermines the economic thickness. As stated under thecolumns, the lowest annual cost occurs with 2.50 in. ofinsulation which is the economic thickness.Payback period (yr). This value represents thediscounted payback period of the specific thickness ascompared to the reference thickness. In this example, thereference thickness variable was input as zero, so the paybackis compared to the uninsulated condition.Present Value of Heat Saved ($/ft). This gives theenergy cost savings in discounted terms as compared tothe uninsulated condition. As discussed earlier, the firstincrement has the most impact on energy savings, but thefurther incremental savings are still justified, as evidencedby the reduction of annual cost to the 2.50-in. thickness.Heat Loss (Btu/ft). This calculation allows the user to checkthe expected heat loss with that required for a specific process.It is possible that under certain conditions a thicknessgreater than the economic thickness may be required toachieve a necessary process requirement.Surface Temperature (°F). This final output allowsthe user to check the resulting surface temperature to assurethat the level is within the safe-touch range. The ETIprogram is very sophisticated. It employs sound methodsof both thermal and financial analysis and provides outputthat is relevant and useful to the design engineer and owner.NAIMA makes this program available to those desiring tohave it on their own computer systems. In addition, severalof the insulation manufacturers offer to run the analysisfor their customers and send them a program output.

INDUSTRIAL INSULATION 469Figure 15.6 NAIMA 3E computer program output.APPENDIX 15.1Typical Thermal Conductivity Curves Used in SampleCalculations*Fig. 15.A1 Calcium silicate.Fig. 15.A2 Fiberglass pipe insulation.Fig. 15.A3 Fiberglass board, 3 lb/ft 3 .*Current manufacturers’ data should always be used for calculations.The savings for the economic thickness is 49.77 $/ln ft/yr and the reductionin Carbon Dioxide emissions is 1608 lbs/lnft/yr.

470 ENERGY MANAGEMENT HANDBOOKReferences1. American Society for Testing and Materials, Annual Book of ASTMStandards: Part 18—Thermal and Cryogenic Insulating Materials;Building Seals and Sealants; Fire Test; Building Constructions;Environmental Acoustics; Part 17—Refractories, Glass and OtherCeramic Materials; Manufactured Carbon and Graphite Products.2. W.H. McADAMS, Heat Transmission, McGraw-Hill, New York,1954.3. E.M. SPARROW and R.D. CESS, Radiation Heat Transfer, McGraw-Hill, New York, 1978.4. L.L. BERANEK, Ed., Noise and Vibration Control, McGraw-Hill,New York, 1971.5. F.A. WHITE, Our Acoustic Environment, Wiley, New York, 1975.6. M. KANAKIA, W. HERRERA, and F. HUTTO, JR., “Fire ResistanceTests for Thermal Insulation,” Journal of Thermal Insulation, Apr.1978, Technomic, Westport, Conn.7. Commercial and Industrial Insulation Standards, Midwest InsulationContractors Association, Inc., Omaha, Neb., 1979.8. J.F. MALLOY, Thermal Insulation, Reinhold, New York, 1969.9. American Society for Testing and Materials, Annual Book of ASTMStandards, Part 18, STD C-585.10. J.F. MALLOY, Thermal Insulation, Reinhold, New York, 1969, pp.72-77, from Thermon Manufacturing Co. technical data.11. L.B. McMlLLAN, “Heat Transfer through Insulation in the Moderateand High Temperature Fields: A Statement of Existing Data,” No.2034, The American Society of Mechanical Engineers, New York,1934.12. Economic Thickness of Industrial Insulation, Conservation PaperNo. 46, Federal Energy Administration, Washington, D.C., 1976.Available from Superintendent of Documents, U.S. GovernmentPrinting Office, Washington, D.C. 20402 (Stock No. 041-018-00115-8).13. ASHRAE Handbook of Fundamentals, American Society of Heating,Refrigerating and Air Conditioning Engineers, Inc., New York,1972, p. 298.14. NAIMA 3 E’s Insulation Thickness Computer Program, NorthAmerican Insulation Manufacturers Association, 44 Canal CenterPlaza, Suite 310, Alexandria, VA 22314.15. P. Greebler, “Thermal Properties and Applications of High TemperatureAircraft Insulation,” American Rocket Society, 1954.Reprinted in Jet Propulsion, Nov.-Dec. 1954.16. Johns-Manville Sales Corporation, Industrial Products Division,Denver, Colo., Technical Data Sheets.17. ASHRAE Handbook of Fundamentals, American Society of Heating,Refrigerating and Air Conditioning Engineers, Inc., Atlanta, GA,1992, p.22.16.18. W.C.Turner and J.F. Malloy, Thermal Insulation Handbook, RobertE. Krieger Publishing Co. And McGraw Hill, 1981.19. Ahuja, A., “Thermal Insulation: A Key to Conservation,” Consulting-SpecifyingEngineer, January 1995, p. 100-108.20. U.S. Department of Energy, “Industrial Insulation for SystemsOperating Above Ambient Temperature,” Office of Industrial Technologies,Bulletin ORNL/M-4678, Washington, D.C., September1995.

CHAPTER 16USE OF ALTERNATIVE ENERGYJERALD D. PARKERProfessor Emeritus, Oklahoma State UniversityStillwater, OklahomaProfessor Emeritus, Oklahoma Christian UniversityOklahoma City, Oklahoma16.1 INTRODUCTIONAny energy source that is classified as an “alternativeenergy source” is that because, at one time it was notselected as the best choice. If the original choice of anenergy source was a proper one the use of an alternativeenergy source would make sense only if some conditionhas changed. This might be:1. Present or impending nonavailability of the presentenergy source2. Change in the relative cost of the present and thealternative energy3. Improved reliability of the alternative energysource4. Environmental or legal considerationsTo some, an alternative energy source is a nondepletingor renewable energy source, and, for manyit is this characteristic that creates much of the appeal.Although the terms “ alternative energy source” and“renewable energy sources” are not intended by thiswriter to be synonymous, it will be noted that some ofthe alternative energy sources discussed in this sectionare renewable.It is also interesting that what we now think of asalternative energy sources, for example solar and wind,were at one time important conventional sources of energy.Conversely, natural gas, coal, and oil were, at sometime in history, alternative energy sources. Changes inthe four conditions listed above, primarily conditions 2and 3, have led us full circle from the use of solar andwind, to the use of natural gas, coal, and oil, and backagain in some situations to a serious consideration ofsolar and wind.In a strict sense, technical feasibility is not a limitationin the use of the alternative energy sources thatwill be discussed. Solar energy can be collected at anyreasonable temperature level, stored, and utilized in avariety of ways. Wind energy conversion systems arenow functioning and have been for many years. Refusederivedfuel has also been used for many years. What isimportant to one who must manage energy systems arethe factors of economics, reliability, and in some cases,the nonmonetary benefits, such as public relations.Government funding for R&D as well as tax incentivesin the alternative energy area dropped sharplyduring the decade of the eighties and early nineties. Thiscaused many companies with alternative energy productsto go out of business, and for others to cut back on productionor to change into another product or technologyline. Solar thermal energy has been hit particularly hardin this respect, but solar powered photovoltaic cells havehad continued growth both in space and in terrestrialapplications. Wind energy systems have continued to beinstalled throughout the world and show promise of continuedgrowth. The burning of refuse has met with someenvironmental concerns and strict regulations. Recyclingof some refuse materials such as paper and plastics hasgiven an alternative to burning. Fuel cells continue toincrease in popularity in a wide variety of applicationsincluding transportation, space vehicles, electric utilitiesand uninterruptible power supplies.Surviving participants in the alternative energybusiness have in some cases continued to grow and to improvetheir products and their competitiveness. As someor all of the four conditions listed above change, we willsee rising or falling interest on the part of the government,industry and private individuals in particular alternativeenergy systems.16.2 SOLAR ENERGY16.2. 1 Availability“ Solar energy is free!” states a brochure intendedto sell persons on the idea of buying their solar products.“There’s no such thing as a free lunch” shouldcome to mind at this point. With a few exceptions, onemust invest capital in a solar energy system in orderto reap the benefits of this alternative energy source.In addition to the cost of the initial capital investment,one is usually faced with additional periodic or random471

472 ENERGY MANAGEMENT HANDBOOKcosts due to operation and maintenance. Provided thatthe solar system does its expected task in a reasonablyreliable manner, and presuming that the conventionalenergy source is available and satisfactory, the importantquestion usually is: Did it save money compared to theconventional system? Obviously, the cost of money, thecost of conventional fuel, and the cost and performanceof the solar system are all important factors. As a firststep in looking at the feasibility of solar energy, we willconsider its availability.Solar energy arrives at the outer edge of the earth’satmosphere at a rate of about 428 Btu/hr ft 2 (1353W/m 2 ). This value is referred to as the solar constant.Part of this radiation is reflected back to space, partis absorbed by the atmosphere and re-emitted, andpart is scattered by atmospheric particles. As a result,only about two-thirds of the sun’s energy reaches thesurface of the earth. At 40° north latitude, for example,the noontime radiation rate on a flat surface normal tothe sun’s rays is about 300 Btu/hr • ft 2 on a clear day.This would be the approximate maximum rate at whichsolar energy could be collected at that latitude. A solarcollector tracking the sun so as to always be normalto the sun’s rays could gather approximately 3.6 × 10 3Btu/ft 2 -day as an absolute upper limit. To gather 1 millionBtu/day, for example, would require about 278 ft 2(26 m 2 ) of movable collectors, collecting all the sunlightthat would strike them on a clear day.Since no collector is perfect and might collect only70% of the energy striking it, and since the percentsunshine might also be about 70%, a more realistic areawould be about 567 ft 2 (53 m 2 ) to provide 1 millionBtu of energy per day. In the simplest terms, would thecost of constructing, operating, and maintaining a solarsystem consisting of 567 ft 2 of tracking solar collectorsjustify a reduction in conventional energy usage of 1million Btu/day? Fixed collectors might be expected todeliver approximately 250,000 Btu/yr for each squarefoot of surface.A most important consideration which was ignoredin the discussion above was that of the system’sability to use the solar energy when it is available. Aspace-heating system, for example, cannot use solarenergy in the summer. In industrial systems, energy demandwill rarely correlate with solar energy availability.In some cases, the energy can be stored until needed,but in most systems, there will be some available solarenergy that will not be collected. Because of this factor,particular types of solar energy systems are most likelyto be economically viable. Laundries, car washes, motels,and restaurants, for example, need large quantitiesof hot water almost every day of the year. A solar water-heatingsystem seems like a natural match for suchcases. On the other hand, a solar system that furnishesheat only during the winter, as for space heating, mayoften be a poor economic investment.The amount of solar energy available to collect ina system depends upon whether the collectors moveto follow or partially follow the sun or whether theyare fixed. In the case of fixed collectors, the tilt fromhorizontal and the orientation of the collectors may besignificant. The remainder of this section considers theenergy available to fixed solar collecting systems.Massive amounts of solar insolation data havebeen collected over the years by various governmentand private agencies. The majority of these data arehourly or daily solar insolation values on a horizontalsurface, and the data vary considerably in reliability.Fixed solar collectors are usually tilted at some anglefrom the horizontal so as to provide a maximum amountof total solar energy collected over the year, or to providea maximum amount during a particular season ofthe year. What one needs in preliminary economic studiesis the rate of solar insolations on tilted surfaces.Figure 16.1 shows the procedure for the conversionof horizontal insolation to insolation on a tiltedsurface. The measured insolation data on a horizontalsurface consist of direct radiation from the sun anddiffuse radiation from the sky. The total radiation mustbe split into these two components (step A) and eachcomponent analyzed separately (steps B and C). In addition,the solar energy reflected from the ground andother surroundings must be added into the total (stepD). Procedures for doing this are given in Refs. 1 to 4.Figure 16.1 Conversion of horizontal insolation to insolation on tilted surface.

USE OF ALTERNATIVE ENERGY 473A very useful table of insolation values for 122cities in the United States and Canada is given in Ref.5. These data were developed from measured weatherdata using the methods of Refs. 2 and 3 and are only asreliable as the original weather data, perhaps ± 10%. Asummary of the data for several cities is given in Table16.1.One of the more exhaustive compilations of U.S.solar radiation data is that compiled by the NationalClimatic Center in Asheville, North Carolina, for theDepartment of Energy. Data from 26 sites were rehabilitatedand then used to estimate data for 222 stations,shown in Figure 16.2. A summary of these data is tabulatedin a textbook by Lunde. 6 It should be rememberedthat measured data from the past do not predict whatwill happen in the future. Insolation in any month canbe quite variable from year to year at a given location.Another approach is commonly used to predictinsolation on a specified surface at a given location.This method is to first calculate the clear-day insolation,using knowledge of the sun’s location in the sky at thegiven time. The clear-day insolation is then correctedby use of factors describing the clearness of the sky ata given location and the average percent of possiblesunshine.The clear-sky insolation on a given surface is readilyfound in references such as the ASHRAE Handbookof Fundamentals. A table of percent possible sunshine forseveral cities is given in Table 16.2.16.2.2 Solar CollectorsA wide variety of devices may be used to collectsolar energy. A general classification of types is givenin Figure 16.3. Tracking-type collectors are usually usedwhere relatively high temperatures (above 250°F) arerequired. These types of collectors are discussed at theend of this section. The more common fixed, flat-platecollector will be discussed first, followed by a discussionof tube-type or mildly concentrating collectors.The flat-plate collector is a device, usually faced tothe south (in the northern parts of the globe) and usuallyat some fixed angle of tilt from the horizontal. Itspurpose is to use the solar radiation that falls upon itto raise the temperature of some fluid to a level abovethe ambient conditions. That heated fluid, in turn, maybe used to provide hot water or space heat, to drive anengine or a refrigerating device, or perhaps to removemoisture from a substance. A typical glazed flat-platesolar collector of the liquid type is shown in Figure16.4.The sun’s radiation has a short wavelength and easilypasses through the glazing (or glazings), with onlyabout 10 to 15% of the energy typically reflected and absorbedin each glazing. The sunlight that passes throughis almost completely absorbed by the absorber surfaceand raises the absorber temperature. Heat loss out theback from the absorber plate is minimized by the use ofinsulation. Heat loss out the front is decreased somewhatby the glazing, since air motion is restricted. The heatedFigure 16.2 Weatherstations for whichrehabilitated measured(asterisks) and deriveddata have been collected.[From SOLMET,Volume 1, and InputData for Solar Systems,Nov. 1978, prepared byNOAA for DOE, Interagencyagreement E(49-26)-1041. Some dataare given in Ref. 6.]

474 ENERGY MANAGEMENT HANDBOOKTable 16.1 Average Daily Radiation on Tilted Surfaces for Selected CitiesAverage Daily Radiation (Btu/day ft 2 ).City Slope Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.Albuquerque, NM hor. 1134 1436 1885 2319 2533 2721 2540 2342 2084 1646 1244 103430 1872 2041 2295 2411 2346 2390 2289 2318 2387 2251 1994 178040 2027 2144 2319 2325 2181 2182 2109 2194 2369 2341 2146 194250 2127 2190 2283 2183 1972 1932 1889 2028 2291 2369 2240 2052vert. 1950 1815 1599 1182 868 754 795 1011 1455 1878 2011 1927Atlanta, GA hor. 839 1045 1388 1782 1970 2040 1981 1848 1517 1288 975 74030 1232 1359 1594 1805 1814 1801 1782 1795 1656 1638 1415 111340 1308 1403 1591 1732 1689 1653 1647 1701 1627 1679 1496 118850 1351 1413 1551 1622 1532 1478 1482 1571 1562 1679 1540 1233vert. 1189 1130 1068 899 725 659 680 811 990 1292 1332 1107Boston, MA hor. 511 729 1078 1340 1738 1837 1826 1565 1255 876 533 43830 830 1021 1313 1414 1677 1701 1722 1593 1449 1184 818 73640 900 1074 1333 1379 1592 1595 1623 1536 1450 1234 878 80350 947 1101 1322 1316 1477 1461 1494 1448 1417 1254 916 850vert. 895 950 996 831 810 759 791 857 993 1044 842 820Chicago, IL hor. 353 541 836 1220 1563 1688 1743 1485 1153 763 442 28030 492 693 970 1273 1502 1561 1639 1503 1311 990 626 38440 519 716 975 1239 1425 1563 1544 1447 1307 1024 662 40350 535 723 959 1180 1322 1341 1421 1363 1274 1034 682 415vert. 479 602 712 746 734 707 754 806 887 846 610 373Ft. Worth, TX hor. 927 1182 1565 1078 2065 2364 2253 2165 1841 1450 1097 89830 1368 1550 1807 1065 1891 2060 2007 2097 2029 1859 1604 138840 1452 1601 1803 1020 1755 1878 1845 1979 1995 1907 1698 148850 1500 1614 1758 957 1586 1663 1648 1820 1914 1908 1749 1549vert. 1315 1286 1196 569 728 679 705 890 1185 1459 1509 1396Lincoln, NB hor. 629 950 1340 1752 2121 2286 2268 2054 1808 1329 865 62930 958 1304 1605 1829 2004 2063 2088 2060 2092 1818 1351 102740 1026 1363 1620 1774 1882 1909 1944 1971 2087 1894 1450 111350 1068 1389 1597 1679 1724 1720 1763 1838 2030 1922 1512 1170vert. 972 1162 1156 989 856 788 828 992 1350 1561 1371 1100Los Angeles, CA hor. 946 1266 1690 1907 2121 2272 2389 2168 1855 1355 1078 90530 1434 1709 1990 1940 1952 1997 2138 2115 2066 1741 1605 143940 1530 1776 1996 1862 1816 1828 1966 2002 2037 1788 1706 155050 1587 1799 1953 1744 1644 1628 1758 1845 1959 1791 1762 1620vert. 1411 1455 1344 958 760 692 744 918 1230 1383 1537 1479New Orleans, LA hor. 788 954 1235 1518 1655 1633 1537 1533 1411 1316 1024 72930 1061 1162 1356 1495 1499 1428 1369 1456 1490 1604 1402 100940 1106 1182 1339 1424 1389 1309 1263 1371 1451 1626 1464 105850 1125 1174 1292 1324 1256 1170 1137 1259 1381 1610 1490 1082vert. 944 899 847 719 599 546 548 647 843 1189 1240 929Portland, OR hor. 578 872 1321 1495 1889 1992 2065 1774 1410 1005 578 50830 1015 1308 1684 1602 1836 1853 1959 1830 1670 1427 941 94140 1114 1393 1727 1569 1746 1739 1848 1771 1680 1502 1020 104250 1184 1442 1727 1502 1622 1594 1702 1673 1651 1539 1073 1116vert. 1149 1279 1326 953 889 824 890 989 1172 1309 1010 1109Source: Ref. 5.

USE OF ALTERNATIVE ENERGY 475Table 16.2 Mean percentage of possible sunshine for selected U.S. cities.Station Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. AnnualAlbuquerque, NM 70 72 72 76 79 84 76 75 81 80 79 70 76Atlanta, GA 48 53 57 65 68 68 62 63 64 67 60 47 60Boston, MA 47 56 57 56 59 62 64 63 61 58 48 48 57Chicago, IL 44 49 53 56 63 69 73 70 65 61 47 41 59Ft. Worth, TX 56 57 65 66 67 75 78 78 74 70 63 58 68Lincoln, NB 57 59 60 60 63 69 76 71 67 66 59 55 64Los Angeles, CA 70 69 70 67 68 69 80 81 80 76 79 72 73New Orleans, LA 49 50 57 63 66 64 58 60 64 70 60 46 59Portland, OR 27 34 41 49 52 55 70 65 55 42 28 23 48Source: Ref. 7.Figure 16.3 Types of solar collectors.absorber plate also radiates energy back toward the sky,but this radiation is longer-wavelength radiation andmost of this radiation not reflected back to the absorberby the glazing is absorbed by the glazing. The heatedglazing, in turn, converts some of the absorbed energyback to the air space between it and the absorber plate.The trapping of sunlight by the glazing and the consequentheating is known as the “ greenhouse effect.”Energy is removed from the collector by the coolantfluid. A steady condition would be reached when theabsorber temperature is such that losses to the coolantand to the surroundings equal the energy gain from thesolar input. When no energy is being removed from thecollectors by the coolant, the collectors are said to be atstagnation. For a well-designed solar collector, that stagnationtemperature may be well above 300°F. This mustbe considered in the design of solar collectors and solarsystems, since loss of coolant pumping power might beexpected to occur sometime during the system lifetime. Atypical coolant flow rate for flat-plate collectors is about0.02 gpm/ft 2 of collector surface (for a 20°F rise).The fraction of the incident sunlight that is collectedby the solar collector for useful purposes is called thecollector efficiency. This efficiency depends upon severalvariables, which might change for a fixed absorber platedesign and fixed amount of back and side insulation.These are:1. Rate of insolation2. Number and type of glazing3. Ambient air temperature4. Average (or entering) coolant fluid temperatureA typical single-glazed flat-plate solar collectorefficiency curve is given in Figure 16.5. The measuredperformance can be approximated by a straight line. The

476 ENERGY MANAGEMENT HANDBOOKleft intercept is related to the product τα, where τ is thetransmittance of the glazing and α is the absorptance ofthe absorber plate. The slope of the line is related to themagnitude of the heat losses from the collector, a flatterline representing a collector with reduced heat-losscharacteristics.A comparison of collector efficiencies for unglazed,single-glazed, and double- glazed flat-plate collectors isshown in Figure 16.6. Because of the lack of glazingreflections, the unglazed collector has the highest efficienciesat the lower collector temperatures. This factor,combined with its lower cost, makes it useful forswimming pool heating. The single-glazed collector alsoperforms well at lower collector temperatures, but likethe unglazed collector, its efficiency drops off at highercollection temperatures because of high front losses.The double-glazed collector, although not performingtoo well at lower temperatures, is superior at the highertemperatures and might be used for space heating and/or cooling applications. The efficiency of an evacuatedtube collector is also shown in Figure 16.6. It can be seenthat it performs very poorly at low temperatures, butbecause of small heat losses, does very well at highertemperatures.A very important characteristic of a solar collectorsurface is its selectivity, the ratio of its absorptance α s forsunlight to its emittance ε for long-wavelength radiation.A collector surface with a high value of α s /ε is called aselective surface. Since these surfaces are usually formedby a coating process, they are sometimes called selectivecoatings. The most common commercial selective coatingis black chrome. The characteristics of a typical blackchrome surface are shown in Figure 16.7, where α λ =ε λ , the monochromatic absorptance and monochromaticFigure 16.4 Typical double-glazedflat-plate collector, liquid type, internallymanifolded. (Courtesy LOF.)

USE OF ALTERNATIVE ENERGY 477Figure 16.5 Efficiency of a typical liquid-type solar collectorpanel.emittance of the surface. Note that at short wavelengths(~ 0.5 μ), typical of sunlight, the absorptance is high.At the longer wavelengths (~2 μ and above), where theabsorber plate will emit most of its energy, the emittanceis high. Selective surfaces will generally perform betterthan ordinary blackened surfaces. The performance of aflat black collector and a selective coating collector arecompared in Figure 16.8. The single-glazed selective collectorperformance is very similar to the double-glazednonselective collector. Economic considerations usuallylead one to pick a single-glazed, selective or a doubleglazed,nonselective collector over a double-glazed, selectivecollector, although this decision depends heavilyupon quoted or bid prices.Air-type collectors are particularly useful wherehot air is the desired end product. Air collectors havedistinct advantages over liquid-type collectors:1. Freezing is not a concern.2. Leaks, although undesirable, are not as detrimentalas in liquid systems.3. Corrosion is less likely to occur.Figure 16.6 Comparison of collector efficiencies forvarious liquid type collectors.Air systems may require large expenditures offan power if the distances involved are large or if thedelivery ducts are too small. Heat-transfer rates to airare typically lower than those to liquids, so care mustbe taken in air collectors and in air heat exchangers toprovide sufficient heat-transfer surface. This very ofteninvolves the use of extended surfaces or fins on the sidesof the surface, where air is to be heated or cooled. Typicalair collector designs are shown in Figure 16.9.Flat-plate collectors usually come in modulesFigure 16.7 Characteristics of a typical selective (blackchrome) collector surface.Figure 16.8 Comparison of the efficiencies of selectiveand nonselective collectors.

478 ENERGY MANAGEMENT HANDBOOKFigure 16.10 Examples of collectors hooked in parallel. (a)Internally manifolded. (b) Externally manifolded.Figure 16.9 Typical air collector designs. (a) Finnedsurface. (b) Corrugated surface. (c) Porous matrix.about 3 ft wide by 7 ft tall, although there is no standardsize. Collectors may have internal manifolds orthey may be manifolded externally to form collectorarrays (Figure 16.10). Internally manifolded collectorsare easily connected together, but only a small numbercan be hooked together in a single array and still havegood flow distribution. Small arrays (5 to 15) are oftenpiped together with similar arrays in various series andparallel arrangements to give the best compromise betweennearly uniform flow rates in each collector, and assmall a pressure drop and total temperature rise as canbe attained. Externally manifolded collectors are easilyconnected in balanced arrays if the external manifoldis properly designed. However, these types of arraysrequire more field connections, however, have more exposedpiping to insulate, and are not as neat looking.The overall performance of a collector array, measuredin terms of the collector array efficiency, may bequite a bit less than the collector efficiency of the individualcollectors. This is due primarily to unequal flowdistribution between collectors, larger temperature risesin series connections than in single collectors, and heatlosses from the connecting piping. A good array designwill minimize these factors together with the pumpingrequirements for the array.Concentrating collectors provide relatively hightemperatures for applications such as air conditioning,power generation, and the furnishing of industrial orprocess heat above 250°F (121°C). They generally cannotuse the diffuse or scattered radiation from the skyand must track so that the sun’s direct rays will beconcentrated on the receiver. The theory is simple. Byconcentrating the sun’s rays on a very small surface,heat losses are reduced at the high temperature desired.An important point to make is that concentratingcollectors do not increase the amount of energy abovethat which falls on the mirrored surfaces; the energy ismerely concentrated to a smaller receiver surface.A typical parabolic trough-type solar collectorarray is shown in Figure 16.11. Here the concentratingsurface or mirror is moved, to keep the sun’s raysconcentrated as much as possible on the receiver, in thiscase a tube through which the coolant flows. In somesystems the tube moves and the mirrored surfaces remainfixed.This type collector can be mounted on an east-westaxis and track the sun by tilting the mirror or receiverin a north-south direction (Figure 16.12a). An alternativeis to mount the collectors on a north-south axis andtrack the sun by rotating in an east-west direction(Figure 16.12b). A third scheme is to use a polar mount,aligning the trough and receiver parallel to the earth’spole, or inclined at some angle to the pole, and trackingeast to west (Figure 16.12c). Each has its advantages anddisadvantages and the selection depends upon the application.A good discussion of concentrating collectorsis given in Ref. 8.Fully tracking collectors may be a parabolic diskwith a “point source” or may use a field of individualnearly flat moving mirrors or heliostats, concentratingtheir energy on a single source, such as might be installed

USE OF ALTERNATIVE ENERGY 479Figure 16.11 Typical parabolic troughtypesolar collector array (Suntec,Inc.).Figure 16.12 Trough-type collectorarrangements for sun tracking. (a)N-S horizontal E-W tracking. (b) E-W horizontal N-S tracking. (c) Polaraxis E-W tracking.give the best combination of surface area and pressuredrop. Air flow must be down for storing and up forremoval if this type system is to perform properly. Horizontalair flow through a storage bed should normallybe avoided. An air flow rate of about 2 cfm/ft 2 of collectoris recommended. The amount of storage requiredin any solar heating system is tied closely to the amountof collector surface area installed, with the optimumamount being determined by a computer calculation.As a rule of thumb, for rough estimates one should useabout 75 lb of rock per square foot of air-type collectors.If the storage is too large, the system will not be ableto attain sufficiently high temperatures, and in addition,heat losses will be high. If the storage is too small, thesystem will overheat at times and may not collect andstore a large enough fraction of the energy available.The most common solar thermal storage system isone that uses water, usually in tanks. As a rule the waterstorage tank should contain about 1.8 gal/ft 2 of collectorsurface. Water has the highest thermal storage capabilona tower (a power tower). Computers usually controlthe heliostat motion. Some trough-type collectors arealso fully tracking, but this is not too common. All partialand fully tracking collectors must have some device tolocate the sun in the sky, either by sensing or by prediction.Tracking motors, and in some cases flexible or movableline connections, are additional features of trackingsystems. Wind loads can be a serious problem for anysolar collector array that is designed to track. Ability towithstand heavy windloads is perhaps the biggest singleadvantage of the flat-plate, fixed collector array.16.2.3 Thermal Storage SystemsBecause energy demand is almost never tied to solarenergy availability, a storage system is usually a partof the solar heating or cooling system. The type of storagemay or may not depend upon the type of collectorsused. With air-type collectors, however, a rock-bed typeof storage is sometimes used (Figure 16.13). The rocksare usually in the size range 3/4 to 2 in. in diameter to

480 ENERGY MANAGEMENT HANDBOOKFigure 16.14 External heat exchanger between collectorsand main storage.Figure 16.13 Rock-bed-type storage system.ity of any common single-phase material per unit massor per unit volume. It is inexpensive, stable, nontoxic,and easily replaced. Its main disadvantage is its highvapor pressure at high temperatures. This means thathigh pressures must be used to prevent boiling at hightemperatures.Water also freezes, and therefore in most climates,the system must either (1) drain all of the collector fluidback into the storage tank, or (2) use antifreeze in thecollectors and separate the collector fluid from the storagefluid by use of a heat exchanger.Drain-down systems must be used cautiously becauseone failure to function properly can cause severedamage to the collectors and piping. It is the more usualpractice in large systems to use a common type of heatexchanger, such as a shell-and-tube exchanger, placedexternal to the storage tank, as shown in Figure 16.14.Another method, more common to small solar systems,is to use coils of tubing around the tank or inside thetank, as shown in Figure 16.15.In any installation using heat exchangers betweenthe collectors and storage, the exchanger must have sufficientsurface for heat transfer to prevent impairmentof system performance. Too small a surface area in theexchanger causes the collector operating temperature tobe higher relative to the storage tank temperature, andthe collector array efficiency decreases. As a rough ruleof thumb, the exchanger should be sized so as to givean effectiveness of at least 0.60, where the effectivenessis the actual temperature decrease of the collector fluidpassing through the exchanger to the maximum possibletemperature change. The maximum possible would bethe difference between the design temperature of the collectorfluid entering the exchanger and the temperatureentering from the storage tank.Stratification normally occurs in water storageFigure 16.15 Internal heat exchanger between collectorand storage medium.systems, with the warmest water at the top of the tank.Usually, this is an advantage, and flow inlets to thetank should be designed so as not to destroy this stratification.The colder water at the bottom of the tank isusually pumped to the external heat exchanger and thewarmer, returning water is placed at the top or near thecenter of the tank. Hot water for use is usually removedfrom the top of the tank.Phase-change materials (PCMs) have been studiedextensively as storage materials for solar systems. Theydepend on the ability of a material to store thermal energyduring a phase change at constant temperature.This is called latent storage, in contrast to the sensiblestorage of rock and water systems. In PCM systemslarge quantities of energy can be stored with little or no

USE OF ALTERNATIVE ENERGY 481change in temperature. The most common PCMs are theeutectic salts. Commercial PCMs are relatively expensiveand, to a certain extent, not completely proven as tolifetime and reliability. They offer distinct advantages,however, particularly in regard to insulation and spacerequirements, and will no doubt continue to be givenattention.Figure 16.16 Installation of a differential temperature controllerin a liquid heating system.16.2.4 Control SystemsSolar systems should operate automatically withlittle attention from operating personnel. A good controlsystem will optimize the performance of the system withreliability and at a reasonable cost. The heart of anysolar thermal collecting system is a device to turn onthe collector fluid circulating pump (and other necessarydevices) when the sun is providing sufficient insolationso that energy can be collected and stored, or used. Withflat-plate collectors it is common to use a differentialtemperature controller (Figure 16.16), a device with twotemperature sensors. One sensor is normally located onthe collector fluid outlet and the other in the storagetank near the outlet to the heat exchanger (or at the levelof the internal heat exchanger). When the sun is out, thefluid in the collector is heated. When a prescribed temperaturedifference (about 20°F) exists between the twosensors, the controller turns on the collector pump andother necessary devices. If the temperature differencedrops below some other prescribed difference (about 3to 5°F), the controller turns off the necessary devices.Thus clouds or sundown will cause the system to shutdown and prevent not only the unnecessary loss of heatto the collectors but also the unnecessary use of electricity.The distinct temperature difference to start and tostop is to prevent excessive cycling.Differential temperature controllers are availablewith adjustable temperature difference settings and canalso be obtained to modulate the flow of the collectorfluid, depending upon the solar energy available.Controllers for high-temperature collectors, suchas evacuated tubes and tracking concentrators, sometimesuse a light meter to sense the level of sunlightand turn on the pumps. Some concentrating collectorsare inverted for protection when light levels go belowa predetermined value.In some systems the storage fluid must be keptabove some minimum value (e.g., to prevent freezing).In such cases a low-temperature controller is needed toturn on auxiliary heaters if necessary. A high-temperaturecontroller may also be needed to bypass the collectorfluid or to turn off the system so that the storage fluidis not overheated.Figure 16.17 shows a control diagram for a solarheatedasphalt storage system (Figure 16.18) in which thefluid must be kept between two specified temperatures.Solar heat is used whenever it is available (collector pumpon). If the storage temperature drops below the specifiedminimum, the pump and an electric heater are turned onto circulate electrically heated fluid to the tank. If the tankfluid gets too warm, the system shuts off. Almost any requiredcontrol pattern can be developed for solar systemsusing the proper arrangement of a differential temperaturecontroller, high- and low-temperature controllers,relays, and electrically operated valves.16.2.5 Sizing and EconomicsAn article on how to identify cost-effectivesolar-thermal applications is given in the ASHRAEJournal 9 . In almost any solar energy system thelargest single expense are the solar collector panelsand support structure. For this reason the systemis usually “sized” in terms of collector panel area.Pumps, piping, heat exchangers, and storage tanksare then selected to match.Very rarely can a solar thermal system provide100% of the energy requirements for a givenapplication. The optimum-size solar system is theone that is the most economical on some chosenbasis. The computations may be based on (1) lowestlife-cycle cost, (2) quickest payout, (3) best rateof return on investment, and, (4) largest annualsavings. All of these computations involve the initialinstalled cost, the operating and maintenancecosts, the life of the equipment, the cost of money,the cost of fuel, and the fuel escalation rate, in ad-

482 ENERGY MANAGEMENT HANDBOOKFigure 16.17 Control system for the solar-heated asphalt storage tank of Figure 16.Figure 16.18 Flow schematic of a solar-heated asphalt storagedition to computations involving the amount of energyfurnished by the solar system.A typical set of calculations might lead to the resultsshown in Figure 16.19, the net annual savings peryear versus the collector area, with the present cost offuel as a parameter. 10 Curve "a" represents a low fuelcost, the net savings is negative, and the system wouldcost rather than save money. Curve "b" represents aslightly higher fuel cost where a system of about 800 ft 2of collectors would break even.Curves "c" and "d," representing even higher fuelcosts, show a net savings, with optimum savings occurringat about 1200 and 2000 ft 2 , respectively.High interest rates tend to reduce the economic viabilityof solar systems. High fuel costs obviously havethe opposite effect, as does a longer life of the equipment.Federal and state tax credits would also have animportant effect on the economics of solar energy as analternative energy source. Technical improvements andlower first costs can obviously have an important effecton the economics, but contributions of these two factorshave been relatively slow in coming.16.2.6 Solar CellsSolar cells use the electronic properties of semiconductormaterial to convert sunlight directly intoelectricity. They are widely used today in space vehiclesand satellites, and in terrestrial applications requiringelectricity at remote locations. Since the conversion isdirect, solar cells are not limited in efficiency by theCarnot principle. A wide variety of text under titlessuch as solar cells, photovoltaics, solar electricity, andsemiconductor technology are available to give detailsof the operating principles, technology and system applicationsof solar cells.Most solar cells are very large area p-n junction

USE OF ALTERNATIVE ENERGY 483of importance to utilizing solar cells are shown in Figure16.20b. The short-circuit current I sc is, ideally, equal tothe light generated current I L . The open-circuit voltageV oc is determined by the properties of the semiconductor.The particular point on the operating curve wherethe power is maximum, the rectangle defined by V mpand I mp will have the greatest area. The fill factor FFis a measure of how “square” the output characteristicsare. It is given by:V mp I mpFF = ————V oc I scFigure 16.19 Collector area optimization curves for a typicalsolar heating system. Ref. 10.diodes. Figure 16.20a. A p-n junction has electronicasymmetry. The n-type regions have large electrondensities but small hole densities. Electrons flow readilythrough the material but holes find it very difficult.P-type material has the opposite characteristic. Excesselectron-hole pairs are generated throughout the p-typematerial when it is illuminated. Electrons flow from thep-type region to the n-type and a flow of holes occursin the opposite direction. If the illuminated p-n junctionis electrically short circuited a current will flow in theshort-circuiting lead. The normal rectifying current-voltagecharacteristic of the diode is shown in Figure 16.20b.When illuminated (insulated) the current generated bythe illumination is superimposed to give a characteristicwhere power can be extracted.The characteristic voltage and current parametersIdeally FF is a function only of the open-circuitvoltage and in cells of reasonable efficiency has a valuein the range of 0.7 to 0.85.Most solar cells are made by doping silicon, thesecond most abundant element in the earth’s crust. Sandis reduced to metallurgical-grade silicon, which is thenfurther purified and converted to single-crystal siliconwafers. The wafers are processed into solar cells whichare then encapsulated into weatherproof modules. Boronis used to produce p-type wafers and phosphorus isthe most common material used for the n-type impurity.Other types of solar cells include CdS, CdTe, and GaAs.Theoretical maximum conversion efficiencies of thesesimple cells are around 21 to 23 percent.Major factors which, when present in real solarcells, prohibit the attainment of theoretical efficienciesinclude reflection losses, incomplete absorption, onlypartial utilization of the energy, incomplete collection ofelectron-hole pairs, a voltage factor, a curve factor andinternal series resistance.Thin-film solar cells have shown promise in reducingthe cost of manufacturing and vertical junctionBlocksElectron FlowElectronsBlocksHole FlowDarkV ocVpnI LII mpV mpIlluminatedHolesI L Conventional Current DirectionI scOperatingpoint for max.power(a)(b)Figure 16.20 Nomenclature of solar cells.

484 ENERGY MANAGEMENT HANDBOOKcells have been shown to have high end-of-life efficiencies.Solar cells are subject to weathering and radiationdamage. Care must be taken in solar arrays to avoidpoor interconnection between cells, and increased seriesresistance due to deterioration of contacts.Solar cells are arranged in a variety of series andparallel arrangements to give the voltage-current characteristicsdesired and to assure reliability in case ofindividual cell failure. Fixed arrays are placed at someoptimal slope and usually faced due south in the northernhemisphere. Large arrays are usually placed on astructure allowing tracking of the sun similar to thoseused for concentrating solar thermal collectors. In somearrays the sunlight is concentrated before it is allowedto impinge on the solar cells. Provision must be madefor thermal energy removal since the solar cell typicallyconverts only a small fraction of the incident sunlightinto electrical power. Increasing temperature of the cellhas a dominant effect on the open circuit voltage, causingthe power output and efficiency to decrease. Forsilicon cells the power output decreases by 0.4 to 0.5 %per degree Kelvin increase.Provisions must usually be made for convertingthe direct current generated by the array into the moreuseful alternating current at suitable frequency and voltage.In many systems where 24 hour/day electricity isneeded some type of storage must be provided.16.3 WIND ENERGYWind energy to generate electricity is most feasibleat sites where wind velocities are consistently high andreasonably steady. Ideally these sites should be remotefrom densely populated areas, since noise generation,safety, and disruption of TV images may be problems.On the other hand the generators must be close enoughto a consumer that the energy produced can be utilizedwithout lengthy transmission. An article in the EPRIjournal (11) gives a good update on wind energy in theelectric utility industry as of 1999. Another very usefulsource of information about wind energy is availablefrom the American Wind Energy Association (12) andfrom its web site. This group publishes the AWEA WindEnergy Weekly and maintains an archive of back issues.According to Awea 3,600 megawatts of new wind energycapacity were installed in 1999 worldwide, bringingtotal installed capacity of 13,400 MW. In the UnitedStates 895 MW of new generating capacity was addedbetween July 1998 and June 1999. In addition more than180 MW of equipment was installed in repowering (replacing)older wind equipment. Some of the growth hasbeen due to supportive government policies at both thestate and federal levels, some due to the technology'ssteadily improving economics, and some due to electricutilities developing "green" policies for customerspreferring nonpolluting sources. Early growth was inthe mountain passes of California. More recently rapidgrowth in wind energy has occurred in Minnesota andIowa as a result of legislative mandates. Other states areexpected to follow.The seacoast of Europe, where strong winds blowconsistently, continues to be a popular siting for windturbines. European manufacturers account for 90 percentof the turbines installed worldwide.Cost of wind-powered electricity has fallen byabout 80% since the early 1980's and is expected tocontinue to fall as the technology develops. The averagecost in 2000 was in the 5 cents/kWh range. To be competitivewith conventional sources this cost will have tobe cut approximately in half.16.3. 1 AvailabilityA panel of experts from NSF and NASA estimatedthat the power potentially available across the continentalUnited States, including offshore sites and theAleutian arc, is equivalent to approximately 10 5 GWof electricity. 13 This was about 100 times the electricalgenerating capacity of the United States. Figure 16.21shows the areas in the United States where the averagewind velocities exceed 18 miles/hr (6 meters/sec) at 150ft (45.7 m) above ground level. As an approximation, thewind velocity varies approximately as the 1/7 power ofdistance from the ground.The power that is contained in a moving air streamper unit area normal to the flow is proportional to thecube of the wind velocity. Thus small changes in windvelocity lead to much larger changes in power available.The equation for calculating the power density of thewind isP 1— = — ρV 3 (16.1)A 2where P = power contained in the windA = area normal to the wind velocityρ = density of air (about 0.07654 lbm/ft 3or 1.23 kg/m 3 )V= velocity of the air streamConsistent units should be selected for use in equation16.1.It is convenient to rewrite equation 16.1 asPA = KV 3 (16.2)

USE OF ALTERNATIVE ENERGY 485Figure 16.21 Areas in the United States where average wind speeds exceed 18 miles/hr (8 miles/sec) at 150 ft (45.7m) elevation above ground level. (From Ref. 13.)If the power density P/A is desired in the unitsW/ft 2 , then the value of K depends upon the units selectedfor the velocity V. Values of K for various unitsof velocity are given in Table 16.3.The fraction of the power in a wind stream that isconverted to mechanical shaft power by a wind deviceis given by the power coefficient Cp.It can be shown that only 16/27 or 0.5926 ofthe power in a wind stream can be extracted by awind machine, since there must be some flow velocitydownstream from the device for the air to move out ofthe way. This upper limit is called the Betz coefficient(or Glauert’s limit). No wind device can extract thistheoretical maximum. More typically, a device mightextract some fraction, such as 70%, of the theoreticallimit. Thus a real device might extract approximately(0.5926)(0.70) = 41% of the power available. Such adevice would have an aerodynamic efficiency of 0.70and a power coefficient of 0.41. The power conversioncapability of such a device could be determined by usingequation 16.2 and Table 16.3. Assume a 20-mile/hrwind. ThenPA actual= 5.08 × 10 –3 20 3 0.41) = 16.7 W/ft 2Notice that for a 30-mile/hr wind the power conversionTable 16.3 Values of K to Give P/A (W/ ft 2 ) in Equation16.2 a .Units of V K—————————————————————————ft/sec 1.61 × 10 –3miles/hr 5.08 × 10 -3km/hr 1.22 × 10 -3m/sec 5.69 × 10 -2knots 7.74 × 10 –3a To convert w/ft 2 to W/m 2 , multiply by 10.76.capability would be 56.2 W/ft 2 , or more than three timesas much.Because the power conversion capability of a winddevice varies as the cube of the wind velocity, one cannotpredict the annual energy production from a winddevice using mean wind velocity. Such a predictionwould tend to underestimate the actual energy available.16.3.2 Wind DevicesWind conversion devices have been proposed andbuilt in a very wide variety of types. The most generaltypes are shown in Figure 16.22. The most common type

486 ENERGY MANAGEMENT HANDBOOKis the horizontal-axis, head-on type, typical of conventionalfarm windmills. The axis of rotation is parallel tothe direction of the wind stream. Where the wind directionis variable, the device must be turned into the wind,either by a tail vane or, in the case of larger systems,by a servo device. The rotational speed of the single-,double-, or three-bladed devices can be controlled byfeathering of the blades or by flap devices or by varyingthe load.In most horizontal-axis wind turbines, the generatoris directly coupled to the turbine shaft, sometimesthrough a gear drive. In the case of the bicycle multibladedtype, the generator may be belt driven off therim, or the generator hub may be driven directly off therim by friction. In the later case there is no rotationalspeed control except that imposed by the load.In the case of a vertical-axis wind turbine (VAWT)such as the Savonius or the Darrieus types, the directionof the wind is not important, which is a tremendousadvantage. The system is more simple and there are nostresses created by yawing or turning into the wind asoccurs on horizontal-axis devices. The VAWT are alsolighter in weight, require only a short tower base, andcan have the generator near the ground. VAWT enthusiastsclaim much lower costs than for comparablehorizontal-axis systems.The side wind loads on a VAWT are accommodatedby guy wires or cables stretched from the groundto the upper bearing fixture.The Darrieus-type VAWT can have one, two, three,or more blades, but two or three are most common. Thecurved blades have an airfoil cross section with very lowstarting torque and a high tip-to-wind speed.The Savonius-type turbine has a very high startingtorque but a relatively low tip-to-wind speed. Itis primarily a drag-type device, whereas the Darrieustype is primarily a lift-type device. The Savonius andthe Darrieus types are sometimes combined in a singleturbine to give good starting torque and yet maintaingood performance at high rotational speeds.Figure 16.23 shows the variation of the powercoefficient Cp as the ratio of blade tip speed to windspeed varies for different types of wind devices. It canbe seen that two-blade types operating at relatively highspeed ratios have the highest value of Cp, in the rangeof 0.45, which is fairly close to the limiting value of theBetz coefficient (0.593). The Darrieus rotor is seen tohave a slightly lower maximum value, but like the twobladetype, performs best at high rotational speeds. TheAmerican (bicycle) multi-blade type is seen to performbest at lower ratios of tip to wind speed, as does theSavonius.For comparison, in a 17-mile/hr (7.6-meters/sec)wind, a 2000-kW horizontal-axis wind turbine wouldhave a diameter of 220 ft (67 m) and a 2000-kW Darrieustype would have a diameter of 256 ft (78 m) and wouldstand about 312 ft (95 m) tall. 1216.3.3 Wind SystemsBecause the typical wind device cannot furnishenergy to exactly match the demand, a storage systemand a backup conventional energy source may be madea part of the total wind energy system (Figure 16.24).The storage system might be a set of batteries and thebackup system might be electricity from a utility. Insome cases the system may be designed to put electricalpower into the utility grid whenever there is a surplusand to draw power from the utility grid wheneverthere is a deficiency of energy. Such a system must besynchronized with the utility system and this requireseither rotational speed control or electronic frequencycontrol such as might be furnished by a field-modulatedgenerator.Economics favors the system that feeds surpluspower into the utility grid over the system with storage,but the former does require reversible metering devicesand a consenting utility. Some states have and othersprobably will pass laws that require public utilities toaccept such power transfers.16.3.4 Wind Characteristics—SitingThe wind characteristics given in Figure 16.21 aresimple average values. The wind is almost always quitevariable in both speed and direction. Gusting is a rapidup-and-down change in wind speed and/or direction.An important characteristic of the wind is the numberof hours that the wind exceeds a particular speed. Thisinformation can be expressed as speed-duration curves,such as those shown in Figure 16.25 for three sites inthe United States. These curves are similar to the loaddurationcurves used by electric utilities.Because the power density of the wind dependson the cube of the wind speed, the distribution of annualaverage energy density of winds of various speedswill be quite different for two sites with different averagewind speeds. A comparison between sites havingaverage velocities of 13 and 24 miles/hr (5.8 and 10.7meters/sec) is given in Figure 16.25. The area under thecurve is indicative of the total energy available per unitarea per year for each case.Sites should be selected where the wind speed is ashigh and steady as possible. Rough terrain and the presenceof trees or building should be avoided. The crestof a well-rounded hill is ideal in most cases, whereas

USE OF ALTERNATIVE ENERGY 487Figure 16.22 Types of wind-conversion devices. (From Ref. 13.)

488 ENERGY MANAGEMENT HANDBOOKFigure 16.23 Typical pressure coefficients of several wind turbine devices.(From Ref. 13.)Figure 16.24 Typical WECS with storage. (From Ref. 13.)a peak with sharp, abrupt sides might be very unsatisfactory,because of flow reversals near the ground.Mountain gaps that might produce a funneling effectcould be most suitable.16.3.5 Performance of Turbines and SystemsThere are three important wind speeds that mightbe selected in designing a wind energy conversion system(WECS). They are (1) cut-in wind speed, (2) ratedwind speed, and (3) cut-off wind speed. The names aredescriptive in each case. The wind turbine is kept fromturning at all by some type of brake as long as the windspeed is below the cut-in value. The wind turbine isshut off-completely at the cut-off wind speed to preventdamage to the turbine. The rated wind speed is thelowest speed at which the system can generate its ratedpower. If frequency control were not important, a windturbine would be permitted to rotate at a variable speedas the wind speed changed. In practice, however, sincefrequency control must be maintained, the wind turbinerotational speed might be controlled by varying the loadon the generator when the wind speed is between the

USE OF ALTERNATIVE ENERGY 489content of the wind at that site is low for speedsbelow the cut-in speed and somewhat above therated speed.Another useful curve is the actual annualpower density output of a WECS (Figure 16.28).The curve shows the hours that the device wouldactually operate and the hours of operation at fullrated power. The curve is for a system with a ratedwind speed of 30 miles/hr (13.4 meters/sec),a cut-in velocity of 15 miles/hr (6.7 meters/sec)and a cut-off velocity of 60 miles/hr (26.8 meters/sec) with constant output above 30 miles/hr.Figure 16.25 Annual average speed-duration curves for threesites. (From Ref. 13.)16.3.6 Loadings and AcousticsBlades on wind turbine devices have a varietyof extraneous loads imposed upon them.Rotor blades may be subject to lead-lag motions,flapping, and pitching. These motions and someof their causes are shown in Figure 16.29. Theseloads can have a serious effect on the systemperformance, reliability, and lifetime.Acoustics can be a serious problem withwind devices, especially in populated areas. TheDOE/ NASA device at Boone, North Carolina,caused some very serious low-frequency (~1 Hz)noises and was taken out of service.The most promising wind systems from aneconomic standpoint appear to be mid-size, propeller-typesystems, located in large numbers atone site and controlled from a central terminal.Most systems will likely be owned by an electricutility or sell their power to a utility.Figure 16.26 Comparison of distribution of annual average energydensity at two sites. (From Ref. 13.) (a) V avg = 13 miles/hr.(b) V avg = 24 miles/hr.cut-in and rated speed. When the wind speed is greaterthan the rated speed but less than cut-out speed, thespin can be controlled by changing the blade pitch onthe turbine. This is shown in Figure 16.27 for the 100-kWDOE/NASA system at Sandusky, Ohio. A system suchas that shown in Figure 16.27 does not result in largelosses of available wind power if the average energy16.4 REFUSE-DERIVED FUEL16.4.1 Process WastesTypical composition of solid waste is shownin Table 16.4. It can be seen that more than 70%by weight is combustible. More important, morethan 90% of the volume of typical solid wastecan be eliminated by combustion. Burning wasteas fuel has the advantage of not only replacingscarce fossil fuels but also greatly reducing theproblem of waste disposal.Solid wastes affect public health, the environment,and also present an opportunity for reuse orrecycling of the material. Managing this in an optimumway is sometimes called integrated solid waste management.A textbook on that subject (15) provides muchmore detail than can be furnished in this brief handbookdiscussion. That reference estimates that between2500 and 7750 pounds of waste is generated per person

490 ENERGY MANAGEMENT HANDBOOKFigure 16.27 Power output of a100-kW WECS at various windspeeds. (From Ref. 13.)Table 16.4 Typical composition of solid waste.Figure 16.28 Actual annual power density output of aWECS. (From Ref. 13.)each year in the United States. Included in this is typically2225 pounds of municipal waste, 750 pounds ofindustrial waste, and between 250 and 3000 pounds ofagricultural waste per person each year.The total mass of solid wastes in the United Statesreached more than 41 billion tons in 1971 14 and is probablymore than double that amount today. It was estimatedthat each person in the United States consumed660 Ib of packaging material in 1976, and uses over 5lb/day of waste products.The heating value of the refuse would be an importantconsideration in any refuse-derived fuel (RDF)application. Typical heating values of solid waste refusecomponents are given in Table 16.5. Other values aregiven in Ref. 16.Food wastes—12% by weightGarbage (10%)Fats (2%)Noncombustibles—24% by weightAshes (10%)Metals (8%): cans, wire, and foilGlass and ceramics (6%): bottles primarilyRubbish—64% by weightPaper (42%): various types, some with fillersLeaves (5%) Grass (4%)Street sweepings (3%)Wood (2.4%): packaging, furniture, logs,twigsBrush (1.5%)Greens (1.5%)Dirt (1%)Oil, paints (0.8%)Plastics (0.7%): polyvinyl chloride, polyethylene,styrene, etc., as found in packaging,housewares, furniture, toys, and nonwovensyntheticsRubber (0.6%): shoes, tires, toys, etc.Rags (0.6%): cellulose, protein, and wovensyntheticsLeather (0.3%): shoes, tires, toys, etc.Unclassified (0.6%)Source: Ref. 14.16.4.2 Refuse PreparationThere are several routes by which waste can beused to generate steam or electricity. The possible pathsfor municipal wastes are shown in Figure 16.30. In thepast the most common method was for the refuse to

USE OF ALTERNATIVE ENERGY 491Figure 16.29 Rotor blade motions andtheir causes. (From Ref. 13.)Figure 16.30 Possible paths to convert municipal refuse to fuel. (From Ref. 15.)be burned unprepared in a waterwall steam generator.This technology is simple and well developed andcosts can be accurately predicted. Another approach isfor the refuse to be placed in a landfill and gas formedfrom decomposition of the organic material recoveredand burned. This is not efficient and requires land foruse in the landfill. Refuse may be given some treatment,such as shredding and separation, and then burned in awaterwall steam generator.The more sophisticated methods involve sometype of treatment after shredding to change the refuseinto a more desirable fuel form. This may involve convertingthe shredded refuse into a gas or liquid or intosolid pellets.

492 ENERGY MANAGEMENT HANDBOOKTable 16.5 Typical Heating Values of Solid Waste RefuseComponents a MJ/kg Btu/lbDomestic refuseGarbage 4.23 1,820Grass 8.88 3,820Leaves It 39 4,900Rags (cotton, linen) 14.97 6,440Brush, branches 16.60 7,140Paper, cardboard, cartons, bags 17.81 7,660Wood, crates, boxes, scrap 18.19 7,825Industrial scrap and plastic refuseBoot, shoe trim, and scrap 19.76 8,500Leather scrap 23.24 10,000Cellophane 27.89 12,000Waxed paper 27.89 12,000Rubber 28.45 12,240Polyvinyl chloride 40.68 17,500Tires 41.84 18,000Oil, waste, fuel-oil residue 41.84 18,000Polyethylene 46.12 19,840AgriculturalBagasse 8.37-15.11 3,600-6,500Bark 10.46-12.09 4,500-5,200Rice hulls 12.15-15.11 5,225-6,500Corncobs 18.60-19.29 8,000-8,300CompositeMunicipal 10.46-15.11 4,500-6,500Industrial 15.34-16.97 6,600-7,300Agricultural 6.97-13.95 3,000-6,000Source: Ref. 14.a Calorific value in MJ/kg (Btu/lb) as fired.The shredding, which can be done wet or dry(as received), converts the refuse into a relatively homogeneousmixture. The shredding is usually done byhammermills or crushers. This shredding operation iscostly both in terms of energy and maintenance. Onereference gives 1977 maintenance costs of 60 cents/tonof waste. 16Problems with fire, explosions, vibrations, andnoise are common in the shredding operation.Density separation increases the fuel’s heatingvalue, minimizes wear on transporting and boiler heattransfersurfaces, and makes the ash more usable. Resourcerecovery can be an important by-product, withseparation of metals and glass for resale.16.4.3 Pyrolysis and Other ProcessesPyrolysis is the thermal decomposition of materialin the absence of oxygen. The product can be a liquidor a gas suitable for use as a fuel. There are many pyrolysisprojects in the research and development stage.An apparently successful pyrolytic heat-recovery systemis described in Ref. 17. The new plant saved $53,000 thefirst year while disposing of 90% of the firm’s waste. Itwas expected that the system would pay out in approximatelythree years. Emissions were said to be belowstandards set by EPA. A second pyrolytic heat recoveryis being installed to take care of expanding needs.Anaerobic digestion processes, similar to thoseused in wastewater treatment facilities, can also be usedto convert the shredded, separated waste into a fuel.About 3 scf of methane can be produced from about 1 lbof refuse. In this process the shredded organic materialis mixed with nutrients in an aqueous slurry, heated toabout 140°F, and circulated through a digester for severaldays. The off-gas has a heating value of about 600Btu/scf but can be upgraded to nearly pure methane.Solid fuel pellets can also be prepared from refusewhich are low in inorganics and moisture and withheating values around 7500 to 8000 Btu/lbm (17,000 to19,000 kJ/kg). Some pellets have been found to be toofibrous to be ground in the low- and medium-speedpulverizers that might normally be found in coal-firedplants.16.4.4 Refuse CombustionThe major problems in firing refuse in steam generatorsseems to be fouling of heat-transfer surfaces andcorrosion. Fouling is caused by slag and fly-ash deposition.It is reduced by proper sizing of the furnace, byproper arrangement of heat-transfer surfaces, and byproper use of boiler cleaning equipment. 17Corrosion in RDF systems is usually due to:1. Reducing environment caused by stratification orimproper distribution of fuel and air.2. Halogen corrosion caused by presence of polyvinylchloride (PVC) in the refuse.3. Low-temperature corrosion caused when some surfacein contact with the combustion gases is belowthe dew-point temperature of the gas.It appears that many existing coal-fired boilers canbe modified to use suitably prepared refuse as a fuel.One type of refuse that is both abundant and readilyusable in certain types of furnaces and steam generatorsis tires. Paper mills, cement plants and electric utilitiesare among those who have discovered this abundantand readily available fuel, with a higher heating valuethan coal and with less production of pollutants such asnitrous oxides and ash. 18Old tires have been a problem in landfills becausetheir shape permits the trapping of water and methane.Tire piles which catch fire are difficult to extinguish and

USE OF ALTERNATIVE ENERGY 493may burn for days, spewing black smoke into the airand oozing oil into the ground. Burning tires as fuelsolves the difficult problems created by the U.S. inventoryof old tires, estimated to be about 850 million,and the 250 million tires added to the piles each year.Added incentives using old tires are offered by moststate governments and tire collectors typically charge atipping fee for taking a used tire. The company or utilityoperating the furnace can assume responsibility fordirect collection or hire that done by a vendor.Most of the experience in using tires as a fuel inthe U.S. electric utility industry has been with cyclonefiredboilers, which make up less than 10 percent ofutility capacity. Little modification is required in thesetype boilers except for the conveyer system bringing thefuel to the combustion chambers. The metal bead wirearound the rim of the tires must be removed prior tocombustion and any other metal left after burning mustbe removed magnetically. The tires are typically burnedin cyclone-fired boilers as chips about 1 inch square andas a small percentage of the total fuel. Slightly largersizes have been successfully burned in stoker-fired unitswhere the fuel sits on a moving grate near the bottomof the boiler.Tires are more difficult to burn successfully in pulverized-coalboilers except as whole tires in wet-bottomfurnaces operating at high temperatures, around 3200°F.In one project whole tires were fed into the boiler at 10second intervals by a conveyor and lock hopper system.There appears to be good promise of burning tires influidized-bed combustion systems, particularly if theyare designed in advance to handle such fuel. In the nearfuture it is likely that all tires being discarded will findtheir way directly into use in some material manufacturing,such as road surfacing, or will be used as a fuel.The split between the various uses will be determinedprimarily by the economics.16.5 FUEL CELLSAll energy conversion processes that utilize theconcept of a heat engine are limited in their thermalefficiency by the Carnot principle. The fuel cell, sinceit is a direct conversion device, has the advantage ofnot being limited by that principle. The fuel cell is anelectrochemical device in which the chemical energy ofa conventional fuel is converted directly and efficientlyinto low voltage direct-current electrical energy. It canbe thought of as a primary battery in which the fuel andoxidizer are stored external to the battery and are fed toit as needed.In the last decade the fuel cell has emerged as oneof the most promising alternative energy technologies.Because of their modular capabilities fuel cells can bebuilt in a wide range of sizes from 200 kW units topower an individual building to 100 megawatt plantsto add baseload capacity to an electric utility system.Small plants can operate with efficiencies similar tothose of large plants. They can produce high gradewaste heat for use in cogeneration or in space heatingapplications, yielding total energy efficiencies approaching85 percent. Their reliability makes them useful asuninterruptible power systems for hospitals, communicationand computer companies, and hotels. They makelittle noise, pose little danger to those around and aregenerally acceptable in close quarters. They can use avariety of fuels and change out between fuels can beaccomplished rapidly. They can provide VAR control,they have a quick ramp rate and they can be remotelycontrolled in unattended operation.It is estimated that by the year 2010 there will beapproximately 130 gigawatts of new electrical generatingcapacity (26). The fuel cells will have opportunity tobe a part of this expanding capacity in all of the areasof repowering, new central power plants , industrialgeneration and in commercial/ residential generation.Many improvements have been brought forth for usein commercial fuel cells. In the demonstration systemsthere do not seem to be any serious technical barriers.A major effort will be directed to making the cells moreeconomically competitive and reliable with suitable lifetimesof operation.Fundamentals of fuel cell operation are describedin texts on direct energy conversion, such as the oneby Angrist (19) and in the Fuel Cell Handbook (20). Aschematic of a fuel cell is given in Figure 16.31. The fuel,which is in gaseous form, diffuses through the anodeand is oxidized. This releases electrons to the externalcircuit. The oxidizer gas diffuses through the cathodewhere it is reduced by the electrons coming from the anodethrough the external circuit. The resulting oxidationproducts are carried away. In contrast to the combustionin a heat engine, where electrons pass directly from thefuel molecules to the oxidizer molecules, the fuel cellkeeps the fuel molecules from mixing with the oxidizermolecules. The transfer of electrons takes place throughthe path containing the load.The theoretical efficiency of a simple fuel cell isthe maximum useful work we can obtain (the changein Gibbs free energy) divided by the heat of reaction.Since the change in Gibbs free energy is given by ∆G =∆H – T∆S, the efficiency of the simple cell is:

494 ENERGY MANAGEMENT HANDBOOKFigure 16.31 Simple schematic of a fuel cell.η = G H=1–T SHIf any heat is being rejected by the cell T∆S is notzero and the cell efficiency will be less than zero. Otherfactors that must be considered in real fuel cells arelosses associated with attendant accessories, undesirablereactions taking place in the cell, some hindrance of thereaction at the anode or cathode, a concentration gradientin the electrolyte, and Joule heating in the electrolyte.The difference between the maximum useful work andthe actual work must appear as rejected heat, in keepingwith the first law of thermodynamics. A voltageefficiency is defined by:η υ = V acVThe Faradaic or current efficiency η F is also definedfor the fuel cell, and is the fraction of the reactionwhich is occurring electrochemically to give current. Thepart of the chemical free energy that actually results inelectrical energy is the product of η υ η F .The ideal fuel would be hydrogen, which if usedwith pure oxygen as the oxidizer, would give pure wateras the product of combustion.In non-space (utility) operation the four mostpromising types of fuel cells may be classified by theirelectrolytes and the corresponding temperatures atwhich they operate. These are:ElectrolyteTemperaturephosphoric acid fuel cell (PAFC) 200°Cproton exchange membrane (PEM) 80°Cmolten carbonate fuel cells (MCFC) 650°Csolid oxide fuel cell (SOFC) 1000°COperating temperature has a very significant effecton materials of construction, fabrication methods, andthe way in which a unit may be applied.The phosphoric acid (PAFC) is the only type of fuelcell that is in commercial production. The other threeare in various stages of research and development anddemonstration. The MCFC and SOFCs are of interestsince they promise higher efficiencies and lower first

USE OF ALTERNATIVE ENERGY 495costs than the PAFC. The PEMs are primarily suited forresidential/business and transportation application.The most common PAFC applications today arecogeneration units placed on site and fired with naturalgas. Promising fuels for the near future applicationsinclude methanol and coal.The proton exchange membrane (PEM) fuel cellsare a recent development, and have a solid polymermembrane electrolyte that can be operated at 80°C. Onemajor developer has been Ballard Power Systems ofCanada. Ballard is utilizing plastic for the gas manifoldsand other components in an attempt to reduce costs. TheChicago Transit Authority have been using this type ofcell in some of their buses.The molten carbonate fuel cells (MCFC) operate at650°C, high enough to allow the cell’s rejected heat tobe used in a thermal steam cycle. Another advantage ofthe temperature is that reforming can take place withinthe cell, using a reforming catalyst, thus improving thecell efficiency. The cell operation does not require theuse of precious metal catalysts and major cell-stackcomponents can be stamped from less expensive metals.Corrosion prevention and selection of materials forconstruction are the major technical challenges.Full-scale demonstrations are now being testedand show promising results both in performance andcost. Earlier demonstration tests pinpointed problemsthat seem to be solvable by improved control strategies,operating procedure and equipment design.The solid oxide fuel cell (SOFC) can be fabricatedin a variety of shapes due to the use of a solid ceramicelectrolyte. Corrosion problems are alleviated and naturalgas can be used directly without external reforming.The high temperatures do lead to problems in thermalexpansion mismatches and sealing between cells.Developments on SOFCs have taken place continuouslyfor a longer period of time than any of the other celltypes mentioned above. Westinghouse has been a primarydeveloper and has shown their ability to operatethese cells for long periods of time. State-of-the-art fuelcell technology is being demonstrated in a 160-kilowattplant. These plants allow more flexibility in the choice offuels and show excellent performance in combined cycleoperation. The SOFCs can easily follow changing loadrequirements by adjusting air and fuel flows.In March of 2000 the U.S. Department of Energy announcedthat FuelCell Energy, Inc. (FCE) had completedone year of commercial design validation and endurancetesting of a 250 kilowatt-class Direct FuelCell (DFC)power plant in Connecticut. Internal conversion of fuelgas to hydrogen occurs directly instead of externally ina separate unit. This promises reduces costs and makesmore efficient use of what would have been wasted excessheat. The demonstration involved a record runningtime of 8600 hours for a carbonate fuel cell fuel cell stack.The stack was the largest of its type ever assembled up tothis time. The next step toward commercialization will befield trials of the packaged submegawatt product.Several organizations have formed to promote fuelcell research, development and commercialization. TheWorld Fuel Cell Council (21) was founded as a nonprofitassociation in 1991 by a number of fuel cell manufacturersand material suppliers. Its objective is to promotethe most rapid commercialization of fuel cell technologyworldwide. Members of the Council include companiesinvolved in the development and use of a variety of fuelcell technologies for both stationary and mobile applications.The U.S. Fuel Cell Council (22) is an industry associationdedicated to fostering the commercializationof fuel cells in the United States. It provides technicaladvice, conducts outreach activities to inform userindustries, works to raise public awareness, conductseducation programs, provides networking opportunities,and establishes links to comparable activities in theU.S. and elsewhere.The Fuel Cell Commercialization Group (FCCG)has a mission to commercialize carbonate fuel cells forpower generation (23). FCCG members are electric andgas utilities and other energy users that have recognizedthe opportunity and the value of early involvement inthe development and commercialization of this verypromising technology. The FCCG is working to designa multi-megawatt carbonate fuel cell power plant thatmeets utilities’ needs for a commercial product, availablein the year 2002. FCCG members provide productdefinition, information exchange, and other marketfeedback critical to the commercialization process, andwill purchase the first fuel cell power plants producedby FCE. The Alliance to Commercialize CarbonateTechnology (24), a membership alliance of utilities andindustry, was created to help bring the next generationof fuel cell technology, known as molten carbonate fuelcells (MCFC), into commercial markets.The mission of the Nation Fuel Cell Research Center(NFCRC) is to promote and support the genesis of afuel cell industry by providing technological leadershipwithin a vigorous program of research, development anddemonstration (25). NFCRC’s goal is to become a focalpoint or “technological incubator” for advancing fuel celltechnology.A large amount of support for fuel cell research anddevelopment is provided by the United States Departmentof Energy (26). The department maintains an informative

496 ENERGY MANAGEMENT HANDBOOKweb site, which can lead interested parties into the widerange of activities in this area. (http://www.fe. doe. gov)References1. S.E. Fuller, “The Relationship between Diffuse, Total and ExtraterrestrialSolar Radiation,” Journal of Solar Energy and Technology,Vol. 18, No. 3 (1976), pp. 259263.2. B.V.H. Liu and R.C. Jordan, “Availability of Solar Energy forFlat-Plate Solar Heat Collectors,” Application of Solar Energy forHeating and Cooling of Buildings, ASHRAE GRP-170, AmericanSociety of Heating, Refrigerating and Air Conditioning Engineers,Inc., New York, 1977.3. S.A. Klein, “Calculations of Monthly Average Insolation on TiltedSurfaces,” Vol. 1, Proceedings of the ISES and SESC Joint Conference,Winnipeg, 1976.4. D.W. Ruth and R.E. Chant, “The Relationship of Diffuse Radiationto Total Radiation in Canada,” Journal of Solar Energy andTechnology, Vol. 18, No. 2 (1976), pp. 153-154.5. “Introduction to Solar Heating and Cooling Designing and Sizing,”DOE/CS-0011 UC-59a, b, c, Aug. 1978.6. Peter J. Lunde, Solar Thermal Engineering, Wiley, New York,1980.7. Climatic Atlas of the United States, U.S. Government Printing Office,Washington, D.C., 1968.8. A.B. Meinel and M.P. Meinel, Applied Solar Energy, an Introduction,Addison Wesley, Reading, Mass., 1976.9. Maria Telkes, “Solar Energy Storage,” ASHRAE Journal, Sept.1974, pp. 38-44.10. ERDA Facilities Solar Design Handbook, ERDA 77-65, Aug. 1977.11. Taylor Moore, “Wind Power: Gaining Momentum”, EPRI journal,Winter 1999, EPRI, Palo Alto, CA.12. American Wind Energy Association, 122 C Street NW, Washington,DC, 20001, e-mailwindmail@awea.org.13. F.R. Eldridge, Wind Machines, NSF-RA-N-75-051, prepared for NSFby the Mitre Corporation, Oct. 1975, U.S. Government PrintingOffice, Washington, D.C. (Stock No. 038000-00272-4).14. R.F. Rolsten et al., “Solid Waste as Refuse Derived Fuel,” NuclearTechnology, Vol. 36 (mid-Dec. 1977), pp. 314-327.15. G. Tchobanoglous, H. Theisen, and S.A. Vigil, Integrated SolidWaste Management, McGraw-Hill, New York, 1993.16. “Power from Wastes-A Special Report,” Power, Feb. 1975.17. A.W. Morgenroth, “Solid Waste as a Replacement Energy Source,”ASHRAE journal, June 1978.18. J.H. Hirschenhofer, D.B. Stauffer, R.P. Engleman, M.G. Klett, FuelCell Handbook, Business /Technology Books, November 1998.19. World Fuel Cell Council, Kroegerstrasse 5, D-60313 Frankfurt/M,Germany .20. U.S. Fuel Cell Council, 1625 K. Street NW Suite 75, Washington,DC, 200021. Fuel Cell Commercialization Group, 1800 M Street, NW, Suite300, Washington, DC 20036-5802.22. The Alliance to Commercialize Carbonate Technology, 1110Capitol Way S., Suite 307, Olympia, Washington 98501.23. National Fuel Cell Research Center, University of California,Irvine, Engineering Laboratory Facility, Irvine, California 92697-3550.24. Fuel Cell Systems, U.S. Department of Energy, Federal EnergyTechnology Center, 3610 Collins Ferry Road, Morgantown, WV26507-0880

CHAPTER 17I NDOOR AIR QUALITYH.E. BARNEY BURROUGHSAtlanta, Ga.17.1 INTRODUCTION AND BACKGROUNDIAQ and Energy Management are in an undeniablylinked relationship. The purpose of this chapter is toprovide the energy manager with a general and overallunderstanding of the IAQ issue. From this understandingwill come the knowledge of the delicate balancebetween IAQ and energy management and, especially,the confidence that they can work together synergisticallyand positively to create buildings that operate bothhealthfully and efficiently.Many authors and experts focus on the mid-70swith its energy crisis as the spawning ground of today’sIAQ problems. Skyrocketing energy costs led to tightbuilding construction which resulted in drastic reductionsin ventilation air and infiltration. This yieldedthe early expression “ Tight Building Syndrome.” Unfortunately,this label was not only misleading, it wasinadequate in providing a full explanation for the issue.It resulted in the focus of blame on ventilation and its inherentenergy cost as being the “cause all”—”fix all”— ofindoor air quality. This is simply not the case, then ornow as we enter the new millennium.There were a number of other trends and changesthat were happening at the same time. These combinedand intermingled to cause today’s situation of manyexisting or potential problem buildings. Let’s examinethese issues so that we have a better understanding ofhow the IAQ situation developed.17.1.1 Impact of “Least Cost” Design and Construction“Value Engineering” has lost its once positive connotation.Now we know it really means devalued and“cheapened.” This “ least cost” mentality led to many design/constructiondecisions and tactics that save on firstcost. However, both the owner and the occupants toooften pay a deferred price in increased life/cycle costs,discomfort, illness and health costs, and low productivity.In the extreme, first cost savings return their dubious“value” multifold in the form of lost lease income,expensive remediation, and even litigation.17.1.2 Technical Advances inOffice Equipment and ProcessesWhat modern office could survive without thehigh speed/volume copier, the fax, the laser printer, andthe desktop computer terminal? In addition to their heatload, these bring ozone, Volatile Organic Compounds(VOCs), and submicron respirable particles from theprint toner into the indoor environment.17.1.3 Changes in Office Layout and FurnishingsHigh density office layout is now possible through“office systems.” In addition to the VOC contaminantoutgassing of their fabric and fiberboard components,work stations can add unpredicted barriers to air flowand ventilation effectiveness. Further, they facilitateincreased occupancy density which can exceed originalshell building design parameters.17.1.4 Shift of Materials used in ConstructionPolymers and plastics are now widely used inconstruction components as glues, binders, and soilingretardants. Typical are formaldehyde (HCHO) andother aldehydes which are ubiquitous components offurniture, fabric, particle board, and plastic surfaces.There is a widening usage of fabrics and fleecy materialsin the space. Polymeric glues and high solvent-loadedadhesives were also widely used to attach them. Theusage of carpeting is now widespread in schools, forexample, where hard surfaces previously prevailed. Porouswall dividers and wall coverings add to this fabricburden and contribute to both outgassing and “sinks.”An example is 4PC (4-phenyl-cyclo-hexane) which wasa component in the carpet backing. Exposure to highconcentration peaks or prolonged exposure over time tothese chemicals can trigger a health response risk knownas MCS (Multiple-Chemical-Sensitivity).17.1.5 Focus on Energy Conservation in HVAC SystemsMany well-meaning energy managers and engineersdesignated energy conservation above comfort497

498 ENERGY MANAGEMENT HANDBOOK(remember the “Building Emergency Thermostat Settings”[BETS] of the Carter era). This “feeding frenzy”of energy saving brought on VAV (Variable Air Volume)distribution that was designed to completelyshutdown supply air to zones with low demand. Thisresulted in lowered or nonexistent ventilation effectiveness.This meant that conditioned dilution air was notbeing delivered to the occupied zone. This allows airwithin the space to become “aged” or stale and stuffyand the system to go out of proper air balance. Unventilatedspace also becomes a “sink” for contaminants.Contaminants remaining in the space deposit onto andinto fleecy and porous surfaces. Then, they re-emergeover time by outgassing due to changes in temperature,humidity, and airflow.17.1.6 Focus on VentilationBecause of the energy concern and its resultingpredominance, ventilation, and the initial label of “ TightBuilding Syndrome” received undue attention. This isnot to say that ventilation is not integrally involved inmany IAQ problems. However, this early focus overshadowedthe roles of source control and filtration ascontrol mechanisms. It also minimized the many otheraspects of good IAQ such as maintenance, moisturecontrol, housekeeping, commissioning, and other facilitymanagement practices.17.1.7 Prevailing National PrioritiesThe early focus on energy and the overall environmentlessened under the Reagan Administration withthe de-emphasis on regulation. However, the Federalregulators focused on Radon and Asbestos as being seriousindoor environmental threats to the general publichealth and welfare. Heavy regulations were createdregarding asbestos in public buildings, such as schoolsand commercial buildings. The early focus of the regulationswas mitigation through removal. This createdtremendous financial burden on both the private andpublic building management segment. Of course, theregulations and their interpretation have since softened.Their impact, however, was to financially burden facilityand plant management teams and to sensitize decisionmakers against environmental concerns.17.1.8 “No Maintenance” MentalityThe traditional “if it ain’t broke, don’t fix it” attitudematched with the “least cost” perspective hascontributed heavily to the current status of buildingenvironmental quality. Inadequate, deferred, or nonexistentmaintenance can destroy the finest of designs.Most IAQ experts agree that buildings that do notwork properly due to poor upkeep are the most likelycandidates for Sick Building Syndrome. EPA researchersclaim over half of the complaint buildings they visit areplagued with poor maintenance.17.1.9 Time Marches OnFew energy managers today remember theOPEC crisis that began the energy panics in 1974.Buildings constructed in the early stages of this periodhave considerable age. Systems, controls, decorations,roofs, elevators, and HVAC systems, have been subjectedto decades of use and abuse. Poor maintenance,poor filtration, low ventilation rates, water leaks andleast cost purchasing decisions have taken their toll. Itwas revealed, for example, in the trial concerning theEPA headquarters building in Washington, D.C. thatthe ductwork had only 1/32" of uniform dirt buildup.Yet, this equated to tons of particulate contaminationthat had to be cleaned out of the building distributionsystem. This had accumulated over years of operatingwith low efficiency blanket media particulate filters. Theaging process alone is a major factor to deal with whenconsidering IAQ problems.17.1.10 Low Filtration is No FiltrationLow efficiency disposable type furnace filters orblanket media were and are widely used in commercialbuildings. They represent the least cost, however, theyalso represent least performance in providing contaminantcontrol and protection of the systems, the building,and the occupants. The resulting poor filtration allowedcontamination accumulation on cooling coils, ductwork,and the facility. This robs efficiency and energy because ofoperating with fouled inefficient systems, like dirty coils.Poor filtration also provides nutrients for microbial growthas well as a source of odor and VOC sinks. Furthermore,the conditioned and occupied space is not protected fromthe pollutants generated in the outdoor environment.17.1.11 “It’s No Problem Boss!”The natural response by Facility Managers toIAQ complaints was “frustration mixed with skepticism”and a “tendency to downplay complaints” 1 .Handling of IAQ problems was, therefore, slow or nonexistent.Thus, the real problems are complicated by theperception of lack of concern, lack of response, lack ofcommunications and just poor human relations. Thisinappropriate response adds human anger to the healtheffects equation and it usually equals litigation.17.1.12 Public Awareness prevailsIAQ issues have been in all the pertinent trade

INDOOR AIR QUALITY 499publications, most network TV news shows and manyhometown newspapers. The general public was conditionedto environmental issues with the focus on radon,asbestos, PCB, insecticides, fertilizers, ozone and lead.Problem buildings like EPA buildings and numeroushigh profile public courthouses make great headlines.Thus, your tenants, your employees, even yourfamily and children know about air quality in theirspace. The general public is therefore suspicious of“places that make them sick” and “chemical smells” and“black stuff that comes out of the ductwork or sticks towalls.” This awareness has raised the expectation of occupantsregarding the quality of the air in their personalspace.17.1.13 Summary of TrendsIt should be obvious that there are a number ofissues and related trends that worked together to bringus the IAQ issue. The interesting point is that many ofthese influences are totally under the control of the facilityand plant management team. Like the philosopher ofthe Okefenokee (Walt Kelly's Pogo) once stated. “We hasfound the enemy—and he is us.”17.2 WHAT IS THE CURRENT SITUATION?Have we learned anything? This discussion isintended to brief the reader on our current knowledgeof the IAQ issue—now called IEQ (Indoor EnvironmentalQuality).17.2.1 Symptoms of IEQWe still lack a full understanding of the healtheffect, linkages and consequences of indoor air contaminantexposure. This is especially true of the morecontroversial ailments, such as MCS (Multiple ChemicalSensitivity). However, we do have a good workingunderstanding of the health related symptoms that arereported in a sick building. Listed in Table 17.1 2 , are thehealth effects that are typical of occupant complaints.What makes them symptoms of “ Sick Building Syndrome”is that they occur only when the individual isexposed to the building environment. When no longerin the exposure environment, the symptoms tend to goaway or lessen dramatically. If the symptoms persisteven after leaving the space and they fit an identifiablemedical diagnostic pattern, then the ailment is likely tobe a “ Building Related Illness” such as Legionnaire’sDisease or Hypersensitivity Pneumonitis. Unfortunately,some of these symptoms are vague and are easilyconfused with other causes, such as the common cold,preexisting allergies or the prevailing “flu bug.” Thisfuels the response of facility managers to treat the occupantcomplaints as “walk-in” problems rather thancaused by the building space.Table 17.1 Typical IEQ Symptoms———————————————Burning eyesHeadache & sinusitisRunny noseTightness in chestCough & hoarsenessLethargy & fatigueGeneral malaiseSore throatItching skinVarious combinations of the above17.2.2 Causes of IEQ ProblemsThe more common and obvious causes of IEQproblems are known and are listed in Table 17.2. However,they seldom represent a single causative elementbecause most problem buildings exhibit multiple andintermingled causes. Further, all of the involved causeshave to be brought under control to assure that thebuilding returns to a nonproblem status. We still lackclear linkages between specific cause and the resultinghealth effect. Yet, we know that the ultimate and overridingcause of IEQ problems is the unusual or elevatedconcentration of components in the air supplied to thespace. The typical air components noted in Table 17.3 areprevalent in most all indoor air and exposure to themin routine concentrations does not bring about healtheffects or discomfort. However, it is the elevated concentrationsof these same components that contaminatethe space and bring on health complaints. Thus, it iscontaminant amplification that is the primary IEQ cause.Therefore, the search for the specific cause and effect ina specific building site should focus on those elementsthat impact the contaminant amplification and retention,such as water and dirt as amplifiers of fungal growth.When we fully understand these factors, we can thendiagnose building related complaints. They also provideinsight regarding what makes one building “sick” andanother “healthy.”17.2.3 No Clear Definition of “Acceptable” AirWe still lack a clearly stated and measurable definitionof “ acceptable indoor air quality.” The primarycurrent parameter is tied to an outdoor air ventilationrate prescribed by the ASHRAE Ventilation Standard62 4 . This rate is, in turn, tied to the human occupancy

500 ENERGY MANAGEMENT HANDBOOKTable 17.2 Typical IEQ Causes——————————————Low ventilation ratesInadequate ventilation efficiencyPoor quality makeup airPoor filtrationInadequate humidity controlWater incursionPoor maintenanceMicrobiological accumulation and growthInternal pollutantsGround gasSpace usage and activityConcurrent stressorsTable 17.3 Typical Air Contaminants—————————————————Respirable particlesViable particlesAllergensSporesOdorsCorrosive gasesVOCs & FormaldehydeHydrocarbonsReactive inorganic compoundsOxidantslevel of the building and is stated in CFM per occupant.This has proven to be an inadequate yardstick becauseit ignores the contaminant load of the building, its furnishings,and the activities within the space. Further, itpromotes the philosophy that air quality and comfort issolely a function of the ventilation rate. This is definitelynot the case as has been discussed previously. Thus,designers, building owners, and facility managers needa better target guideline than furnished by Standard 62for defining and determining acceptable indoor air quality.The Standard specifically defines it as "air in whichthere are no known contaminants at harmful concentrationsas determined by cognizant authorities and withwhich a substantial majority (80% or more) of the peopleexposed do not express dissatisfaction." Thus, the issuebecomes a complaint based and a comfort based issue.The 20% dissatisfaction factor also becomes irrelevantwhen we realize that just one highly vocal or highlyaffected individual can trigger an IAQ incident. Thecurrent ASHRAE Standing Standard Project Committee(SSPC) is revising Standard 62 through addenda thatare issued under the continuos maintenance process.They are wrestling with the rationale for establishingmore meaningful and appropriate ventilation rates. Inthe meantime, it is very important for the IEQ managerto be aware of actual ventilation rates and related occupancypopulations within the facility.17.2.4 Litigation as the DriverWith the failure of repeated legislative initiativesand the sidetrack of proposed OSHA regulations, theremaining driver of IEQ is the risk of litigation. Lawsuitsin both the worker’s compensation and civil courtsabound. Often, they are highly publicized and involvemillions of dollars in liability and mitigation. The specterof litigation is, therefore, a powerful and destructionforce at work in the IEQ arena. A legacy of the proposedOSHA regulation 5 persists, however, in the almost universalresponse of the building management communityto the smoking issue. After only the announcementof a proposed ruling that would ban smoking in theworkplace, the voluntary control of tobacco smokingin public space was widespread. This provides someindication of the power of regulation and the influencethat it could impose on the indoor air quality issue.This powerful influence provides some guidance to thepotential outcome of the ASHRAE Standard 62 revisionprocess as it issues addenda for public review. This maybe particularly true for this standard; addenda are beingdrafted in mandatory code language. Managers ofenergy and IEQ should track standards and regulationspertaining to IEQ very closely and carefully.17.3 SOLUTIONS AND PREVENTIONOF IEQ PROBLEMSThe following discussions provide both practicaland philosophical aspects to the prevention and mitigationof the causes of IEQ complaints.17.3.1 Understand the Issue!Some experts in the field have re-expressed theissue in the broader term “IEQ” (Indoor EnvironmentalQuality) and this is now used in this text. This recognizesthat the perceived quality of the conditioned spacealso includes the broader issues of lighting, sound, ergonomics,and psycho/social issues.Even as a complex and multidisciplined issue, IEQcontrol and prevention reduces to simple principles.These principles are inherent in the definition of airconditioning as “the control of temperature, humidity,distribution, and cleanliness of air supplied to occupiedspace.” When a building HVAC system fails to addressany aspect of this definition, then complaints, discom-

INDOOR AIR QUALITY 501fort, and health effects can result. Cleanliness of the airis probably the most overlooked but perhaps the mostcrucial aspect in commercial buildings. Thus, an overridingcause of complaints and adverse health effectsfrom IEQ is an abnormal or excessive exposure to accumulatedairborne contaminants.The tactics to control excessive contaminant buildupin conditioned space are: dilution, extraction, andsource control. 4Dilution is the reduction of contaminant load byaddition of clean or at least lesser contaminated air, i.e.,ventilation.Extraction is the reduction of contaminant load bysubtraction of contaminants by means of filtration.Source Control is the reduction of contaminant loadby elimination by means of exhaust or selection of lowemitting components and practices.Most of the management, control, or prevention suggestionsreflect attention to all three of these controltactics. Of these, however, the most cost effective issource control and the least cost effective is ventilation.Thus, filtration and source control should receivepriority when analyzing mitigation alternatives from acost effectiveness standpoint.17.3.2 Treat the Issue as Reality!An IEQ complaint may be real or it may be imagined—butit is real to the individual reporting the complaint.Thus, potential IEQ complaints require promptand positive response from the facility managementteam. This sends a proactive message to employeesand tenants that their comfort, health, and feelingsare important to you. This avoids frustration and angerexperienced by occupants and avoids a legitimatecharge of negligence on your part. Complaints shouldbe logged and follow-up documented which aids inproviding information that is helpful in the diagnosticsprocess. Frequency, nature, location, timing and affectedindividuals provide guidance in determining the causalfactors of the incidents. This also helps in the determinationof the severity and urgency of the situation.17.3.3 Put Someone in Charge and Make Them Smart!Appoint an experienced member of the facilitymanagement team as your IEQ Manager. Many timesthis is the Energy Manager because of the close relationshipbetween IEQ and energy management issues.This is consistent with the recommendation in the EPABuilding Owners Manual. This individual should be theclearing house for IEQ complaints, investigation, problemresolutions, and record keeping. This also becomesthe individual who receives training to become the inhouse“guru.” The IEQ manager should become veryfamiliar with causes, prevention, and mitigation tactics.ASHRAE, IFMA, BOMA, EPA, AIHA, and AWMA areassociations or organizations that are sources of informationand training on IEQ matters. In addition, the AEE(Association of Energy Engineers) offers a course onIEQ Fundamentals that leads to a certification (CertifiedIndoor Air Quality Professional—CIAQP).17.3.4 Establish Proactive Monitoring!I now hear building managers say “Sure, we respondto IEQ complaints”! That is not proactive but“reactive response.” Proactive response takes effort, time,and money just like preventative maintenance. Periodicfacility inspections and assessments will provide baselineair quality performance data. It can also reveal imminentproblems that can be proactively mitigated before occupantcomplaints occur. This entails a physical inspectionwalk-through of the entire facility to identify potentialIEQ causes and practices before they manifest into complaints.It also may be advisable to perform baselineair diagnostics on key environmental parameters suchas temperature and relative humidity, Carbon Dioxide,Carbon Monoxide, Total Volatile Organics, AirborneMicrobials, Airborne Particles, and Formaldehyde. Thiswill enable the Facility Manager to establish “norms” forthe facility that can be used for comparative purposes ifcomplaints occur and performance levels alter.17.3.5 Use ASHRAE Standard 62Really use it, which means to go beyond the ventilationtables. Table 17.4 4 numerically designates outdoorair levels for ventilation using the Ventilation RateProcedure of Standard 62-1999. Most designers stop atthis table and never go beyond. Thus, they don’t realizethat the Standard also covers such issues as outdoor airquality, ventilation effectiveness, monitoring, maintenance,filtration, humidity control, and source control.The Indoor Air Quality alternative procedure allows forconsideration of energy management concerns throughoutdoor air reduction and air cleaning. Because of thisbroad coverage, Standard 62 becomes an excellent overallprotocol for addressing many of the issues discussedabove, including energy management.17.3.6 Test, Adjust and Balance Again!Unfortunately, “TAB” work is usually considereda one-time event—done at system start-up. Then thereport is filed away. However, a building is a dynamic

502 ENERGY MANAGEMENT HANDBOOKmodel, changing as it matures through its life cycle ofchanging layouts, tenants, turnover and usage. TABbecomes an effective tool to assure ongoing maximizedsystem performance and ventilation effectiveness.17.3.7 Exploit Filtration!Perhaps the most under-utilized of the preventiontactics is filtration, both gas phase and particulate.High efficiency filtration addresses contaminant controlby extraction, can cost-effectively substitute for ventilationto save energy, and can act as source control. Moreoften than not, urban outdoor air does not even meetthe ambient air quality levels prescribed by Standard62. Though Standard 62 is largely silent on filtration, anew addendum will incorporate a minimum particulatefilter efficiency. This will be set at the particle extractionperformance of not less than the ASHRAE medium efficiencyfilter. This is the minimum efficiency value thatdefines a good pleated filter. Filtration can also cleansethe outdoor dilution air to eliminate it as a contaminantsource and enhance its effectiveness as a space dilutant.Further, self-contained air cleaners can provide localizedair quality improvement using both gaseous and par-Table 17.4 Excerpts from Table II - ASHRAE Standard 62-1999———————————————————————————————————————————————————Estimated Outdoor Air RequirementsMaximum Occupancy cfmApplication P/1000 ft 2 or 100 m 2 person cfm/ft 2 Comments———————————————————————————————————————————————————Offices———————————————————————————————————————————————————Office space 7 20 Some office equipment may require localexhaust.Reception areas 60 15Telecommunication centers 60 20and data entry areasConference rooms 50 20 Supplementary smoke-removal equipmentmay be required.———————————————————————————————————————————————————Public Spaces———————————————————————————————————————————————————Corridors and utilities 0.05Public restrooms, cfm/wc 50 Mechanical exhaust with no recirculationor urinalis recommended. Normally supplied bytransfer air, local mechanical exhaust:with no recirculation recommendedLocker and dressing rooms 0.5Smoking lounge 70 60Elevators 1.00 Normally supplied by transfer air.———————————————————————————————————————————————————Retail Stores, Sales Floors,and Show Room Floors———————————————————————————————————————————————————Basement and street 30 0.30Upper floors 20 0.20———————————————————————————————————————————————————

INDOOR AIR QUALITY 503ticulate filter components. Tables 17.5 and 17.6 3 provideadditional usages and benefits from more extensive usageof filtration.Table 17.5 Usages of Filtration——————————————Protect mechanical equipment such as heat exchangecoilsProtect systems such as air distribution ductwork, outlets,and porous componentsProtect occupied space such as wall and ceiling surfacesProtect occupants from contaminant exposureProtect processes such as pharmaceuticals and electronicchip fabricationProvide clean makeup air to use as high quality dilutionairProtect environment from contaminated exhaustProvide source control to avoid reintrainment into thereturn public airAugment ventilation air through treatment of the returnairTable 17.6 Benefits of Filtration——————————————Increased system efficiency from clean componentsIncrease system lifeLower maintenance costsLower housekeeping costsAvoidance of product failureEnhanced energy managementLower health risk and costsIncreased productivity and reduced absenteeismAvoiding efficiency losses from the insulating effectof dirt build-up on heat transfer surfaces—HVAC coils17.3.8 Use Lifelong Commissioning!Thought by many to be just system start-up, commissioningis emerging as an essential component of thesuccessful quality construction or renovation process.The commissioning process really commences with thefirst vision of a construction process, and then proceedsthroughout the life of the building. (See ASHRAE Guideline#1, Commissioning). 7 There is a very strong rolefor the commissioning agent to assure that air quality isbuilt into and retained within the building constructionprocess.17.3.9 Leverage the Use of Source Control!Controlling the contaminant, whether it be anodor or a toxic substance, is most effectively and efficientlydone at the source. This can mean localizedexhaust of contaminants to atmosphere. It can meanlocalized spot filtration using air cleaners. It can meana rigorous review of all MSDS records and restrictingintroduction of purposeful, but harmful, products andchemicals to the space. This includes pesticides androutine housekeeping chemicals. It can also mean areview of construction and buildout components foroutgassing properties and/or potential. Specific controltactics are scheduled in Table 17.7 9 and are quotedfrom Managing Indoor Air Quality, an excellent referencebook for the energy manager.17.3.10 Look at Controls!As the nerve center of the building, controls playa major role in effective and efficient building management…getting the right air to the right place at theright time. They both monitor and control response tothe varying conditions within and without the buildingenvelope. They can impact air quality in the way thatoutdoor air levels are managed, distribution systemsrespond, and mechanical equipment reacts to both thechanging dynamics of the building and the needs of theoccupants.17.3.11 Diligent Maintenance!One of the most frequently cited deficiencies inair quality, proper operation and maintenance can bean effective deterrent to IEQ problems. Cleanlinessand avoidance of water in the systems are critical tomaintaining high air quality. Poor maintenance canunfortunately negate the finest of designs and constructionefforts. Likewise, housekeeping and janitorialprotocols should be examined. Cleaning chemicals,vacuuming techniques, and other purposeful practicessuch as pest control, should all be scrutinized to makesure that they are lowering and not aggravating contaminantlevels.17.3.12 Perform Life Cycle Costing!As opposed to “least cost” thinking, (see Chapter4) life cycle costing must re-emerge as the real costdecision model. For example, high efficiency filtersare more expensive first cost than cheap furnacefilters. Yet, they are more cost effective when the energyimpact, health impact, and long life cycle issuesare considered. Recognize the real “life” cycle andthe ecological and environmental cost of a productfrom its cradle to its grave. And the “Green Building”folks would like us to think in terms of cradleto cradle—including the cost and ecological impact ofrecycling. For example, in some urban communities,furnace filters have to be demanufactured at great cost

504 ENERGY MANAGEMENT HANDBOOKTable 17.7 Typical Control TacticsContaminant Sources Control Techniques———————————————————————————————————————————————————Asbestos Furnace, pipe, wall ceiling insulation 1. Enclose: shieldfireproofing, acoustical and floor tiles 2. Encapsulate; seal3. Remove4. Label ACM5. Use precautions against breathingwhen disturbed———————————————————————————————————————————————————Bioaerosols Wet insulation, carpet, ceiling tile, wall Use effective filters. Check and cleancoverings, furniture, air conditioners, any areas with standing water. Be suredehumidifiers, cooling towers, drip pans, condensate pans drain and are clean.and cooling coils. People, pets, plants, Treat with algicides. Maintain humidinsects,and soilifiers and dehumidifiers. Check &clean duct linings. Keep surfaces clean.———————————————————————————————————————————————————Carbon Monoxide (CO) Vehicle exhaust, esp. attached garages; Check and repair furnaces, flues, heatunvented kerosene heaters andexchangers, etc. for leaks. Use onlygas appliances; tobacco smoke; malfunc- vented combustion appliances. Be suretioning furnacesexhaust from garage does not enter airintake.———————————————————————————————————————————————————Combustion Products Incomplete combustion process Use vented appliances and heaters.(NO x and CO)Avoid air from loading docks and garagesentering air intake. Check HVACfor leaks regularly, repair promptly. Useplants as air cleaners. Filter particulates.———————————————————————————————————————————————————Environmental Tobacco Passive smoking; sidestream and main- Eliminate smoking; confine smokersSmoke (ETS) stream smoke to designated areas; or isolate smokerswith direct outside exhaust. Increaseventilation. Filter contaminants.———————————————————————————————————————————————————Formaldehyde Building products; i.e., paneling,, Selective purchasing of materials with(HCHO) materials with lower particleboard, lower formaldehyde emissions. Barrierplywood urea-formaldehyde insulation coatings and sealants. New constructionas well as fabrics and furnishingscommissioning.———————————————————————————————————————————————————Radon Soil around basements and slab on grade Ventilate crawl spaces. Ventilate subslab.Seal cracks, holes, around drainpipes. Positive pressure in tight basements.NOTE: Increased ventilation does notnecessarily reduce radon levels.———————————————————————————————————————————————————Volatile Organic Solvents in adhesives cleaning agents, Avoid use of solvents and pesticidesCompounds (VOCs) paints, fabrics, tobacco smoke, indoors. If done, employ time of useandlinoleum, pesticides, gasoline, photo- excessive ventilations. Localizedcopying materials, refrigerants,exhaust near source when feasible.building materialSelective purchasing.Increase ventilation.———————————————————————————————————————————————————

INDOOR AIR QUALITY 505to be disposed in solid waste handling sites. Likewise,air quality concerns and their resulting impact on thehuman asset base must be considered when design/construct decisions are made. Shortcuts made forreasons of budget, time, or convenience, will demandpayment in due course.17.3.13 Prepare for Regulation and Litigation Now!The proactive Facility Team will prepare for thiseventuality by bringing your facilities and your practicesinto compliance with current “state of the art” and todocument your efforts.It is never too soon to prepare for the eventualityof litigation. The best defense against either litigation orregulation is to proactively follow the recommendationsin this chapter. Then document, document, document!Document your design path! Document your decisionpath!! Document your baseline performance path usingproactive response to the IEQ issue!!!References1. Building Operating Management Magazine, August, 19942. Burroughs, H.E. The Role of Risk Assessment as a Mitigation Protocolfor IAQ Causes. World Energy Engineering Congress, Associationof Energy Engineers Conference, 1991.3. Burroughs, H.E. Air Filtration. A New Look At An Existing IAQAsset. Carrier Global Engineering Conference, 1994.4. ASHRAE Standard 62-1999, Ventilation for Acceptable IndoorAir Quality. American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc., 1791 Tullie Circle, NE, Atlanta, GA.303295. Department of Labor. Federal Register (OSHA), Occupational Safetyand Health Administration, Pages 1 5969 -1 6039.6. Building Air Quality: A Guide For Building Owners And FacilityManagers. US EPA, 401 M St., Washington, DC 20460.7. ASHRAE Guideline 1-1997, Guideline for Commissioning ofHVAC Systems. American Society of Heating, Refrigerating, andAir-Conditioning Engineers, Inc. 1791 Tullie Circle, NE, Atlanta,GA. 30329.8. ASHRAE 52.2-1999, Methods of Testing General Ventilation Air-Cleaning Devices for Removal Efficiency by Particle Size.9. Burroughs, H.E. and Hansen, Shirley J., Managing Indoor Air Quality,Second Edition, 1998. The Fairmont Press, Lilburn, GA.


CHAPTER 18ELECTRIC AND GAS UTILITY RATESFOR COMMERCIAL AND INDUSTRIAL CUSTOMERSLYNDA J. WHITERICHARD A. WAKEFIELDJAIRO A. GUTIERREZCSA Energy ConsultantsArlington, VA18.1 INTRODUCTIONPurpose and LimitationsThe main focus of this chapter on rates is to provide informationon how an average commercial or industrial customercan identify potential rate-related ways of reducingits energy costs. The basic costs incurred by electric andgas utilities are described and discussed. How these costsare reflected in the final rates to commercial and industrialcustomers is illustrated. Some examples of gas andelectric rates and how they are applied are included. Inaddition, this chapter identifies some innovative ratesthat were developed by electric and gas utilities as a responseto the increasing pressure for the development ofa more competitive industry.Because of the breadth and complexity of the subjectmatter, the descriptions, discussions and explanationspresented in this chapter can not cover every specificsituation. Energy consumption patterns are often uniqueto a particular commercial and/or industrial activity andtherefore case by case evaluations are strongly suggested.The purpose is to present some general cost backgroundand guidelines to better understand how to identify potentialenergy cost savings measures.18.1.2 General InformationHistorically, electric and gas utility rate structureswere developed by the utilities themselves within a muchless complex regulatory environment, by simply consideringmarket factors (demand) as well as cost factors(supply). Today the increasing pressures to develop morecompetitive markets have forced utilities to reconsidertheir traditional pricing procedures. Other factors affectingtoday’s electric and gas markets include rising fuelprices, environmental concerns, and energy conservation507mandates. These factors and pressures have affected gasand electric utility costs and hence their rates to their finalcustomers.In general, electric and gas rates differ in structureaccording to the type and class of consumption. Differencesin rates may be due to actual differences in thecosts incurred by a utility to serve one specific customervs. another. Utility costs also vary according to the timewhen the service is used. Customers using service at offpeakhours are less expensive to serve than on-peak users.Since electricity cannot be stored, and since a utility mustprovide instantaneous and continuous service, the size of ageneration plant is determined by the aggregate amount ofservice taken by all its customers at any particular time.The main cost elements generally included in ratemakingactivities are: energy costs, customer costs, and demandcosts. Each of these is discussed in the next section.18.2 UTILITY COSTSUtilities perform their activities in a manner similarto that of any other privately-owned company. The utilityobtains a large portion of its capital in the competitivemoney market to build its system. It sells a service to thepublic. It must generate enough revenues to cover its operatingexpenses and some profit to stay in business andto attract capital for future expansions of its system.In general there are two broad types of costs incurredby a utility in providing its service. First, there arethe fixed capital costs associated with the investment inthe facilities needed to produce (or purchase) and deliverthe service. Some of the expenses associated with fixedcapital costs include interest on debts, depreciation, insurance,and taxes. Second, there are the expenses associatedwith the operation and maintenance of those samefacilities. These expenses include such things as salariesand benefits, spare parts, and the purchasing, handling,preparing, and transporting of energy resources. Therates paid by utility customers are designed to generatethe necessary revenues to recover both types of costs.Both capital and operation and maintenance costs are

508 ENERGY MANAGEMENT HANDBOOKallocated between the major cost elements incurred by autility.18.2.1 Cost ComponentsThe major costs to a utility can be separated intothree components. These include customer costs, energy/commodity costs, and demand costs. These cost componentsare briefly described below.Customer CostsCustomer costs are those costs incurred in the connectionbetween customer and utility. They vary with thenumber of customers, not with the amount of use by thecustomer. These costs include the operating and capitalcosts associated with metering (original cost and on-goingmeter-reading costs), billing, and maintenance ofservice connections.Energy/Commodity CostsEnergy and commodity costs consist of costs thatvary with changes in consumption of kilowatt-hours(kWh) of electricity or of cubic feet of gas. These are thecapital and operating costs that change only with theconsumption of energy, such as fuel costs and productionsupplies. They are not affected by the number of customersor overall system demand.Demand CostsElectric utilities must be able to meet the peak demand—theperiod when the greatest number of customersare simultaneously using service. Gas utilities mustbe responsive to daily or hourly peak use of gas. In eithercase, the utility will need to generate or purchase enoughpower to cover its firm customers’ needs at all times. Demand-relatedcosts are dependent upon overall systemrequirements. Demand costs can be allocated in manydifferent ways, but utilities tend to allocate on-peak load.Included in these costs are the capital and operating costsfor production, transmission, and storage (in the case ofgas utilities) that vary with demand requirements.18.2.2 Allocation of CostsOnce all costs are identified, the utility must decidehow to allocate these costs to its various customer classes.How much of each cost component is directly attributableto serving a residential, a lighting, or a manufacturingcustomer? In answer to this question, each utility performsa cost-of-service study to devise a set of allocationfactors that will allow them to equitably divide thesecosts to the various users. After the costs are allocated,the utility devises a rate structure designed to collectsufficient revenue to cover all its costs, plus a fair rate ofreturn (currently, this is running between 10 and 14% ofthe owners’ equity.)18.3 RATE STRUCTURES18.3.1 Basic Rate StructureThe rate tariff structure generally follows the majorcost component structure. The rates themselves usuallyconsist of a customer charge, an energy charge, and ademand charge. Each type of charge may consist of severalindividual charges and may be varied by the time orseason of use.Customer ChargeThis is generally a flat fee per customer ranging fromzero to $25 for a residential customer to several thousanddollars for a large industrial customer. Some utilities basethe customer charge to large industrial customers on thelevel of maximum annual use.Energy ChargeThis is a charge for the use of energy, and is measuredin dollars per kilowatt-hour for electricity, or indollars per therm or cubic foot of gas. The energy chargeoften includes a fuel adjustment factor that allows theutility to change the price allocated for fuel cost recoveryon a monthly, quarterly, or annual basis without resortingto a formal rate hearing. This passes the burden of variablefuel costs (either increases or decreases) directly tothe consumer. Energy charges are direct charges for theactual use of energy.Demand ChargeThe demand charge is usually not applied to residentialor small commercial customers, though it is notalways limited to large users. The customer’s demand isgenerally measured with a demand meter that registersthe maximum demand or maximum average demand inany 15-, 30-, or 60-minute period in the billing month. Forcustomers who do not have a demand meter, an approximationmay be made based on the number of kilowatthours consumed. Gas demand is determined over anhour or a day and is usually the greatest total use in thestated time period.Another type of demand charge that may be includedis a reactive power factor charge; a charge for kilovoltampreactive demand (kVAR). This is a method used tocharge for the power lost due to a mismatch between theline and load impedance. Where the power-factor chargeis significant, corrective action can be taken, for exampleby adding capacitance to electric motors.

ELECTRIC AND GAS UTILITY RATES FOR COMMERCIAL AND INDUSTRIAL CUSTOMERS 509Demand may be “ratcheted” back to a period ofgreater use in order to provide the utility with revenuesto maintain the production capabilities to fulfill thegreater-use requirement. In other words, if a customeruses a maximum demand of 100 kW or 100 MMBtu onemonth, then uses 60 kW or 60 MMBtu for the next sixmonths, he/she may have to pay for 100 kW or MMBtueach month until the ratchet period (generally 12 months)is over.18.3.2 VariationsUtilities use a number of methods to tailor theirrates to the needs of their customers. Some of the differentstructures used to accomplish this include: seasonalpricing; block pricing; riders; discounts; and innovativerates.Seasonal PricingCosts usually vary by season for most utilities.These variations may be reflected in their rates throughdifferent demand and energy charges in the winter andsummer. When electric utilities have a seasonal variationin their charges, usually the summer rates are higher thanthe winter rates, due to high air conditioning use. Gasutilities will generally have winter rates that are higherthan summer rates, reflecting increased space-heatinguse.Block PricingEnergy and demand charges may be structured inone of three ways: 1) a declining block structure; 2) aninverted block structure; or 3) a flat rate structure. Aninverted block pricing structure increases the rate as theconsumption increases. A declining block pricing methoddecreases the rate as the user’s consumption increases.When a rate does not vary with consumption levels it isa “flat” structure. With the declining or inverted blockstructures, the number of kWh, MMBtu, or therms used isbroken into blocks. The unit cost (cents per kWh or centsper MMBtu or therm) is lower or higher for each succeedingblock.A declining block reflects the fact that most utilitiescan generate additional electricity or provide additionalgas for lower and lower costs—up to a point. The capitalcosts of operation are spread over more usage. The invertedblock structure reflects the fact that the incremental costof production exceeds the average cost of energy. Hence,use of more energy will cause a greater cost to the utility.Most utilities offer rates with more than one blockpricing structure. A utility may offer some combinationof inverted, declining and flat block rates, often reflectingseasonal energy cost differentials as well as use differentials.For example, a gas utility may use an inverted blockpricing structure in the winter that reflects the higher energycosts in that period, but use either a flat or decliningblock pricing structure in the summer when energy costsare lower.RidersA “Rider” modifies the structure of a rate based onspecific qualifications of the customer. For example, acustomer may be on a general service rate and subscribeto a rider that reduces summer energy charges where theutility is granted physical control of the customers airconditioning load.DiscountsThe discount most often available is the voltagediscount offered by electric utilities. A voltage discountprovides for a reduction in the charge for energy and/ordemand if the customer receives service at voltages abovethe standard voltage. This may require the customer toinstall, operate and maintain the equipment necessary toreduce the line voltage to the appropriate service voltage.Each customer must evaluate the economics of the discountsagainst the cost of the equipment they will have toprovide.Innovative RatesIncreased emphasis on integrated resource planning,demand-side management and the move to a morecompetitive energy marketplace has focused utility attentionon innovative rates. Those rates designed to changecustomer load use, help customers maintain or increasemarket share, or provide the utility with a more efficientoperating arena are innovative. Most rates offered todayfit into the innovative category.18.4 INNOVATIVE RATE TYPESUtilities have designed a variety of rate types toaccomplish different goals. Some influence the customerto use more or less energy or use energy at times thatare helpful to the utility. Others are designed to retain orattract customers. Still others are designed to encourageefficient use of energy. The following are some of the innovativerate types that customers should know about.Time-of-Use Rates are used for the pricing ofelectricity only. The primary purpose of the time-of-use(TOU) rate is to send the proper pricing signals to the consumerregarding the cost of energy during specific timesof the day. Generally, a utility’s daytime load is higherthan its nighttime load, resulting in higher daytime pro-

510 ENERGY MANAGEMENT HANDBOOKduction costs. Proper TOU price signals will encouragecustomers to defer energy use until costs are lower. TOUrates are usually offered as options to customers, thoughsome utilities have mandatory TOU rates.End-Use Rates, these rates include air-conditioning,all-electric, compressed natural gas, multi-family, spaceheating,thermal energy storage, vehicle fuel and waterheatingrates. These rates are all intended to encouragecustomers to use energy for a specific end-use.Financial Incentive Rates include rates such asresidential assistance, displacement, economic development,and surplus power rates. Assistance rates providediscounts to residential customers who meet specificlow-income levels, are senior citizens or suffer from somephysical disability. Displacement rates are offered by electricutilities to customers who are capable of generatingtheir own electricity. The price offered to these customersfor utility-provided power is intended to induce thecustomer to “displace” its own generated electricity withutility-provided electricity. Economic development ratesare generally offered by utilities to provide economic incentivesfor businesses to remain, locate, or expand intoareas which are economically distressed. This type of rateis an attempt to attract new customers into the area and toget existing customers to expand until the area is revitalized.Surplus power rates are offered to large commercialand industrial customers. They are offered gas or electriccapacity at greatly reduced prices when the utility has anexcess available for sale.Interruptible Rates generally apply to commercialand industrial customers. The utilities often offer severaloptions with respect to the customer’s ability to interrupt.Prices vary based on the amount of capacity that isinterruptible, the length of the interruption, and the notificationtime before interruption. Such interruptions aregenerally, but not always, customer controlled. In addition,the total number of interruptions and the maximumannual hours of interruption may be limited.18.5 CALCULATION OF A MONTHLY BILLFollowing is the basic formula used for calculatingthe monthly bill under a utility rate. The sum of thesecomponents will result in the monthly bill.1) Customer Charge• Customer charge = fixed monthly charge2) Energy Charge• Energy charge = dollars × energy use• Energy/Fuel Cost Adjustment = dollars × energyuse3) Demand Charge• Demand charge = dollars × demand• Reactive Demand Charges (electric only) =dollars × measured kilovolt-ampere reactivedemand4) Tax/Surcharge• Tax/surcharge = either sum of one or more ofitems 1-3 above multiplied by tax percentage,dollars × energy use, or dollars × demandSome examples illustrating the calculation of amonthly bill follow. These examples are actual rate schedulesused by the utilities shown. The information regardingthe consumption of electricity used in the sample calculationsare based on the typical figures shown in Table18.1.18.5.1 Commercial General Service withDemand ComponentThe commercial general service rate often involvesthe use of demand charges. Table 18.2 provides rate datafrom Public Service Electric & Gas Company. A samplebill is calculated using the data from Table 18.1 for a conveniencestore.Energy Usage – 17,588 kWh; Billing Demand – 30 kW;Season – Summer (June)Customer Charge: $3.74Energy Charge: $1,661.0117,588 kWh ×.09444Energy Cost Adjustment: –$302.3717,588 kWh × Demand Charge: $262.34(1 kW × 4.53) + (29 kW × 8.89)Total Monthly Charge: $1,624.72Energy Usage Measured in kWh per kW DemandThe standard measure of electric energy blocks is inkilowatt hours. An alternative measure used by electricutilities for commercial and industrial rates is energy perunit of demand (e.g., kWh per kW). This block measurementis illustrated by Virginia Power’s Schedule GS-2Intermediate rate in Table 18.3. We are calculating the billfor October use. According to Table 18.1, the conveniencestore used 16,003 kWh in October and had a billing demandof 32 kW. The bill for this use is as follows.Customer Charge: $21.17Energy Charge:Block 1 - (150 kWh × 32 kW),or 4,800 kWh @ $0.04840 = $232.32

ELECTRIC AND GAS UTILITY RATES FOR COMMERCIAL AND INDUSTRIAL CUSTOMERS 511Table 18.1 Typical usage patterns.Convenience Store Office Building Shopping CenterkW kWh kW kWh kW kWhJAN 28 12,049 660 194,500 1,935 892,000FEB 30 13,097 668 215,500 1,905 795,000MAR 31 15,001 668 223,500 1,740 719,000APR 32 13,102 600 177,500 1,515 633,000MAY 30 14,698 345 101,000 1,425 571,000JUN 30 17,588 293 95,000 1,455 680,000JUL 32 17,739 375 118,000 1,440 661,000AUG 35 17,437 420 156,500 1,395 667,000SEP 31 18,963 555 130,000 1,455 733,000OCT 32 16,003 540 207,000 1,455 646,000NOV 30 17,490 570 169,500 1,545 675,000DEC 29 12,684 600 172,500 1,740 768,000Table 18.2 Commercial general service with demand component.————————————————————————————————Company:Public Service Electric & Gas CompanyRate Class:CommercialRate Type:General ServiceRate Name:Schedule GLPService:ElectricEffective Date: 01/01/94Qualifications: General purposes where demand is less then 150 kW.————————————————————————————————Winter SummerOct-May Jun-SepCustomer Charge: $3.74 $3.74Minimum Charge: $3.74 $3.74Energy Cost Adjustment: -$0.019037 –$0.017192Tax Rate: $0.00 $0.00Surcharge: $0.00 $0.00# of Energy Blocks: 1 1Block 1 Size: > 0 > 0Block 1 Energy Charge: $0.09444 $0.09444No. of Demand Blocks: 2 2Block 1 Size: 1 1Block 2 Size: > 1 > 1Block 1 Demand Charge: $4.53 $4.53Block 2 Demand Charge: $7.84 $8.89————————————————————————————————BILLING DEMAND: Estimated by dividing kWh usage by 100 or, if metered,greatest average 30-minute demand in the month.————————————————————————————————

512 ENERGY MANAGEMENT HANDBOOKBlock 2 - (150 kWh × 32 kW),or 4,800 kWh @ $0.02712 = $130.18Block 3 - (all remaining kWh),or 6,403 kWh @ $0.01173 = $75.11Energy Cost Adjustment: (16,003 × $0.01418) $226.92Demand Charge: (32 kW × $5.023) $160.74Subtotal: $846.44Tax Rate:25% of first $50 of subtotal $12.5012% on remainder of subtotal $95.57Total Monthly Charge: $954.51Time-of-Use RatesTime-of-use rates are calculated very differentlyfrom general service rates. The customer’s use must berecorded on a time-of-use meter, so that billing can becalculated on the use in each time period. In the followingsample calculation from Long Island Lighting Company(Table 18.4), we assume the customer has responded tothe price signal and has relatively little on-peak usage.Energy Usage - 1,200 kWh; Season - Summer; On-PeakPeriod - 10:00 AM-8:00 PM, Monday-Friday; On-PeakUsage - 12.5%Customer Charge: $9.79Energy Charge: $176.53150 on-peak kWh @ $0.3739 = $56.091050 off-peak kWh @ $0.1147 = $120.44Energy Cost Adjustment: $2.88Tax: $10.02Total Monthly Charge: $199.2218.5.2 Gas General Service RatesGas general service rates are calculated in a similarmanner to the electric general service rates. Gas ratesare priced in dollars per MMBtu, or in dollars per MCF,depending upon the individual utility’s unit of measurement.Table 18.3 Energy usage measured in kWh per kW of demand.——————————————————————————————————————————————Company: Virginia Electric & Power CompanyRate Class: CommercialRate Type: General ServiceRate Name: Schedule GS-2 - IntermediateEffective Date: 1/01/94Qualifications: Non-residential use with at least three billing demands => 30 kW in the current andprevious 11 billing months; but not more than two billing months of 500 kW or more.——————————————————————————————————————————————OCT-MAYJUN-SEPCustomer Charge: $21.17 $21.17Minimum Charge: See Note See NoteEnergy Cost Adjustment: $0.01418 $0.01418Tax Rate: See Note See NoteNo. of Energy Blocks: 3 3Block 1 Size: 150 kWh/kW demand 150 kWh/kW demandBlock 2 Size: 150 kWh/kW demand 150 kWh/kW demandBlock 3 Size: > 300 kWh/kW demand > 300 kWh/kW demandBlock 1 Energy Charge: $0.0484 $0.0484Block 2 Energy Charge: $0.02712 $0.02712Block 3 Energy Charge: $0.01173 $0.01173No. of Demand Blocks: 1 1Block 1 Size: > 0 > 0Block 1 Demand Charge: $5.023 $6.531——————————————————————————————————————————————MINIMUM CHARGE: Greater of: 1) contract amount; or 2) sum of customer charge, energy charge andadjustments, plus $1.604 times the maximum average 30-minute demand measured in the month. TAX: Taxis 25% of the first $50, and 12% of the excess.BILLING DEMAND: Maximum average 30-minute demand measured in the month, but not less than themaximum demand determined in the current and previous 11 months when measured demand has reached500 kW or more.——————————————————————————————————————————————

ELECTRIC AND GAS UTILITY RATES FOR COMMERCIAL AND INDUSTRIAL CUSTOMERS 513Table 18.4 Electric time-of-use rate.————————————————————————————————Company: Long Island Lighting CompanyRate Class: ResidentialRate Type: Time-of-UseRate Name: Schedule SC 1-VMRP, Rate 2Effective Date: 04/11/95Qualifications:Use for all residential purposes where consumption is 39,000 kWh orless for year ending September 30, or under 12,600 kWh for June throughSeptember.————————————————————————————————OCT-MAYJUN-SEPCustomer Charge: $9.79 $9.79Minimum Charge: $16.53 $16.53Energy Cost Adjustment: $0.0024 $0.0024Tax Rate: 5.29389% 5.29389%No. of Energy Blocks: 1 1Block 1 Size: > 0 > 0Block 1 Energy Charge:On-Peak: $0.1519 $0.3739Off-Peak: $0.0978 $0.1147————————————————————————————————TAX: applied to total billPEAK PERIOD:On-Peak Hours: 10 a.m.-8 p.m., MON-FRI.Off-Peak Hours: All remaining hours.————————————————————————————————Commercial General ServiceRate with Demand ComponentGas rates for the commercial or industrial customermay involve a demand charge. The method for calculatingthis type of bill is done in the same manner as anelectric rate with a demand charge. The example in Table18.5 comes from Northern Illinois Gas Company.Energy Usage - 11,500 MMBtu; Billing Demand - 400MMBtu; Season - WinterCustomer Charge: $325.00Energy Charge: $4,899.00Purchased Gas Adjustment(Energy Cost Adjustment): $28,221.00Demand Charge: $2,828.00Tax: $1,849.92Total Charge: $38,122.9218.6 CONDUCTING A LOAD STUDYOnce a customer understands how utility rates areimplemented, he/she can perform a simple load studyto make use of this information. A load study will helpthe energy user to identify his load patterns, amount andtime of occurrence of maximum load, and the load factor.This information can be used to modify use in ways thatcan lower electric or gas bills. It can also help the customerto determine the most appropriate rate to use.The basic steps of a utility load study are shown inTable 18.6.The first step is to collect historical load data. Pastbills are one source for this information. One year of datais necessary to identify seasonal patterns; two or moreyears of data is preferable. Select a study period that isfairly representative of normal consumption conditions.The next step is to organize the data so that usepatterns are evident. One way to analyze the data is toplot the kWh usage, the maximum demand, and the loadfactor. The load factor is the ratio of the average demandto the maximum demand. The average demand is determinedby the usage in kWh divided by the total numberof hours (24 × number of days) in the billing period. Thenumber of days in the billing period may vary dependingon how often the meter is read. (See Table 18.7.)

514 ENERGY MANAGEMENT HANDBOOKTable 18.5 Commercial gas rate with demand component.———————————————————————————————————————Company: Northern Illinois Gas CompanyRate Class: Commercial/Industrial LargeRate Type: General ServiceRate Name: Schedule 6Effective Date: 09/01/89Qualifications: General commercial use. All charges are shown per MMBtu.———————————————————————————————————————WinterSummerCustomer Charge: $325.00 $325.00Minimum Charge: $3,000.00 $3,000.00Purchased Gas Adjustment: $2.454 $2.61Tax Rate: 5.1% 5.1%No. of Energy Blocks: 1 1Block 1 Size: > 0 > 0Block 1 Energy Charge: $0.426 $0.426No. of Demand Blocks: 1 1Block 1 Size: > 0 > 0Block 1 Demand Charge: $7.07 $5.388———————————————————————————————————————TAX: applied to the total bill and is the lower of 5% of revenues or $0.24 per MMBtu,plus 0.10%.BILLING DEMAND: per MMBtu of customer’s Maximum Daily Contract Quantity.The PGA and Demand charges shown here change monthly. Winter charges applied inJanuary 1994; and Summer charges applied in July 1993.———————————————————————————————————————Table 18.6 Basic steps for conducting a loadanalysis.1) Collect historical load data• compile data for at least one year2) Organize data by month for• kWh consumption• maximum kW demand• load factor3) Review data for• seasonal patterns of use• peak demands4) Determine what demand or use can be eliminatedor reduced5) Review load data with utilityNext, review the data. Seasonal variations will beeasily pinpointed. For example, most buildings will showa seasonal trend with two peaks. One will occur in winterand another during summer, reflecting seasonal heatingand cooling periods. There may be other peaks due tosome aspect of some industrial process, such as a cannerywhere crops are processed when they are harvested.In Figure 18.1, kWh, maximum kW, and average kWare plotted from the data in Table 18.1 for the shoppingTable 18.7 Load factor calculation.—————————————————————————average demand (kW)Load Factor = ———————————, wheremaximum demand (kW)kWh usageAverage Demand = ———————(24) × (number of days)Example: December office building load from Table 18.1172,500 kWhAverage Demand = ————————— = 231.9 kW(24) × (31)231.9 kWLoad Factor = ——————— = 0.39600 kW—————————————————————————center and the office building.Note that the shopping center has a dominant winterpeak—it is in the winter that the maximum kWh andkW are used for the year. The load factor ranges between0.54 and 0.70. Overall, the average demand for this customeris about 60% of the peak demand.The office building shows a different pattern. Theload factor ranges between 0.35 and 0.52. This reflects thefact that the office building is really used less than half

ELECTRIC AND GAS UTILITY RATES FOR COMMERCIAL AND INDUSTRIAL CUSTOMERS 515Figure 18.1 Monthly load profile.the time. Generally, working hours span from 8 a.m. to 6p.m. Although some electrical load continues during thenight hours, it is not as intense as during the normal officehours.If the load factor were 1, this would imply uniformlevels of use—in effect, a system that was turned on andleft running continuously. This may be the case withsome manufacturing processes such as steel mills andrefineries.The fourth step is to determine what demand or usecan be eliminated, reduced, or redirected. How can theshopping center reduce its energy costs? By reducing orshifting the peak demand it can shave demand costs. Althoughoverall consumption is not necessarily reduced,the demand charge is reduced. Where demand ratchetsare in place, shaving peak demand may result in savingsover a period of several months, not just the month ofuse. One way to shift peak demand is to install thermalstorage units for space cooling purposes; this will shiftday time load to night time, giving the customer an overallhigher load factor. This may qualify the customer forspecial rates from the utility as well.Where there may not be much that can be doneabout the peak demand, (in a high load factor situation)more emphasis should be placed on methods to reduceusage. Some examples: turn up the thermostat at nightduring the summer, down during the winter; install motiondetectors to turn off unnecessary lights; turn off otherequipment that is not in use.Where the customer is charged for electric serviceon a time-of-use basis, a more sophisticated load studyshould be performed. The data collected should consistof hourly load data over at least one year. This data canbe obtained through the use of recording meters. Onceacquired, the data should be organized to show use patternson a monthly basis with Monday through Friday (orSaturday, depending on the customer’s uses) use plottedseparately from weekend use. Review of this data shouldshow where shaving or shifting energy or demand canlower overall electricity bills.Once the customer has obtained a better understandingof his energy usage patterns, he can discusswith his utility how to best benefit from them. The utilitymost likely will be interested because it will also receivesome benefits. The customer can consider implementingcertain specific measures to better fit in the utility’s loadpattern, and at the same time improve his energy use. Thecustomer’s benefit will generally be associated with less

516 ENERGY MANAGEMENT HANDBOOKenergy-related costs. Table 18.8 contains some examplesof options that can be taken by commercial and industrialcustomers and the effect of those options on the utility.18.7 EFFECTS OF DEREGULATIONON CUSTOMER RATES18.7.1 Gas and Electric Supply DeregulationIn the period since 1980, many changes have eitheroccurred or begun to occur in the structure of the nation’selectric and gas supply industries. These changes have alreadybegun to affect the rate types and structures for U.S.gas and electricity consumers. In the natural gas industry,well-head prices were deregulated as a result of theNatural Gas Policy Act of 1978 and the subsequent NaturalGas Well-Head Decontrol Act of 1989. Subsequently,FERC introduced a number of restructuring rules (OrderNos. 436, 500, and 636) that dramatically change theregulation of the nation’s pipelines and provide accessfor end-users to transport gas purchased at the well-head.In the electric industry, supply deregulation commencedwith passage of the Public Utility Regulatory Policies Actof 1978, which encouraged electric power generation bycertain non-utility producers. The Energy Policy Act of1992 further deregulated production and mandated opentransmission access for wholesale transfers of electricitybetween qualified suppliers and wholesale customers.These legal and regulatory changes will have asignificant and lasting effect on the rate types and ratestructures experienced by end-users. In the past, most gasand electric customers paid a single bundled rate that reflectedall costs for capacity and energy, storage, delivery,and administration. Once customers are given the opportunityto purchase their gas and electric resources directlyfrom producers, it then becomes necessary to unbundlethe costs associated with production from the costs associatedwith transportation and delivery to end users. Thisunbundling process has already resulted in separate ratesfor many services whose costs were previously combinedin the single unit price for either gas or electricity.18.7.2 Effect on Gas RatesMuch of the discussion in Section 18.3 of this chapterpertains to bundled rates for gas. However, as a resultof unbundling, many utilities are now offering customersfour separate services, including balancing, procurement,storage, and transportation of gas. Gas balancing ratesprovide charges for over- or under-use of customerownedgas over a specified period of time. When thecustomer has the utility procure gas for transportationto the customer, gas procurement rates are charged. Gasstorage rates are offered to customers for the storage ofcustomer-owned gas. Gas transportation rates are offeredto commercial, industrial and non-utility generator customersfor the transportation and delivery of customerownedgas. In addition to these rates, there is the actualcost of purchasing the gas to be transported. Gas procurement,balancing, storage and transportation rates haveincreased in usage as the structure of the gas industry hasevolved.Two other types of gas rates also are evolving as aresult of industry deregulation. These include negotiatedgas rates and variable gas rates. The former refers to ratesthat are negotiated between individual customers andthe utility. Such rates are often subject to market conditions.The latter, variable gas service rates, refer to ratesthat vary from month to month. A review of all of the gasservice rates collected by the Gas Research Institute (GRI)in 1994 indicated that 52% of the gas utilities surveyedoffered at least one type of variable pricing. Such ratesare often indexed to an outside factor, such as the priceof gasoline or the price of an alternative fuel, and theyusually vary between established floor and ceiling prices.The most common types of variable rates are those offeredfor transportation services.Table 18.8 Customer options and their effects on utility.———————————————————————————————————————————OPTIONS———————————————————————————————————————————Commercial Industrial Utility Effect———————————————————————————————————————————Accept direct control of Subscribe to interruptible Reduction of load duringwater heaters rates peak periods———————————————————————————————————————————Store hot water to Add nighttime operations Builds load during off-peakincrease space heatingperiods———————————————————————————————————————————

ELECTRIC AND GAS UTILITY RATES FOR COMMERCIAL AND INDUSTRIAL CUSTOMERS 51718.7.3 Effect on Electric RatesIn the past, most U.S. electric customers have paid asingle bundled rate for electricity. Many of these customerspurchased from a utility that produced, transmitted,and delivered the electricity to their premises. In othercases, customers purchased from a distribution utilitythat had itself purchased the electricity at wholesale froma generating and transmitting utility. In both of thesecases, the customer paid for electricity at a single rate thatdid not distinguish between the various services requiredto produce and deliver the power. In the future, as a resultof the deregulation process already underway, thereis a far greater likelihood that initially large customers,and later many smaller customers, will have the abilityto select among a number of different suppliers. In mostof these cases, however, the transmission and delivery ofthe purchased electricity will continue to be a regulatedmonopoly service. Consequently, future electricity consumersare likely to receive separate bills for:• electric capacity and energy;• transmission; and• distribution.In some cases, a separate charge may also be madefor system control and administrative services, dependingon exact industry structure in the given locality. Foreach such charge, a separate rate structure will apply. Atpresent, it appears likely that there will be significant regionaland local differences in the way these rates evolveand are implemented.GLOSSARYThere are a few terms that the user of this documentneeds to be familiar with. Below is a listing of commonterms and their definitions.Billing Demand: The billing demand is the demand thatis billed to the customer. The electric billing demandis generally the maximum demand or maximumaverage measured demand in any 15-, 30-, or 60-minute period in the billing month. The gas billingdemand is determined over an hour or a day andis usually the greatest total use in the stated timeperiod.British Thermal Unit (Btu): Quantity of heat needed tobring one pound of water from 58.5 to 59.5 degreesFahrenheit under standard pressure of 30 inches ofmercury.Btu Value: The heat content of natural gas is in Btu percubic foot. Conversion factors for natural gas are:• Therm = 100,000 Btu;• 1 MMBtu = 1,000,000 Btu = 1 Decatherm.Contract Demand: The demand level specified in a contractualagreement between the customer and theutility. This level of demand is often the minimumdemand on which bills will be determined.Controllable Demand: A portion or all of the customer’sdemand that is subject to curtailment or interruptiondirectly by the utility.Cubic Foot: Common unit of measurement of gas volume;the amount of gas required to fill one cubicfoot.Curtailable Demand: A portion of the customer’s demandthat may be reduced at the utility’s direction.The customer, not the utility, normally implementsthe reduction.Customer Charge: The monthly charge to a customer forthe provision of the connection to the utility and themetering of energy and/or demand usage.Demand Charge: The charge levied by a utility for metereddemand of the customer. The measurement ofdemand may be either in kW or kVA.Dual-Fuel Capability: Some interruptible gas rates requirethe customer to have the ability to use a fuelother than gas to operate their equipment.Energy Blocks: Energy block sizes for gas utilities are eitherin MCFs or in MMBtus. The standard measuresof energy block sizes for electric utilities are kWhs.However, several electric utilities also use an energyblock size based on the customers’ demand level(i.e. kWh per kW). Additionally, some electric utilitiescombine the standard kWh value with the kWhper kW value.Energy Cost Adjustment (ECA): A fuel cost factorcharged for energy usage. This charge usually varieson a periodic basis, such as monthly or quarterly.It reflects the utilities’ need to recover energy relatedcosts in a volatile market. It is often referred to as thefuel cost adjustment, purchased power adjustmentor purchased gas adjustment.Excess or Non-Coincidental Demand: Some utilitiescharge for demands in addition to the on- or offpeakdemands in time-of-use rates. An excess demandis demand used in off-peak time periods thatexceeds usage during on-peak hours. Non-coincidentaldemand is the maximum demand measuredany time in a billing period. This charge is usually inaddition to the on- or off-peak demand charges.Firm Demand: The demand level that the customer canrely on for uninterrupted use.Interruptible Demand: All of the customer’s demandmay be completely interrupted at the utility’s direc-

518 ENERGY MANAGEMENT HANDBOOKtion. Either the customer or the utility may implementthe interruption.MCF: Thousand (1000) cubic feet.MMCF: Million (1,000,000) cubic feet.Minimum Charge: The minimum monthly bill that willbe charged to a customer. This generally is equal tothe customer charge, but may include a minimumdemand charge as well.Off-Peak Demand: Greatest demand measured in theoff-peak time period.On-Peak Demand: Greatest demand measured in the onpeaktime period.Ratchet: A ratchet clause sets a minimum billing demandthat applies during peak and/or non-peak months.It is usually applied as a percentage of the peak demandfor the preceding season or year.Reactive Demand: In electric service, some utilities havea special charge for the demand level in kilovoltamperesreactive (kVAR) that is added to the standarddemand charge. This value is a measure of thecustomer’s power factor.Surcharge: A charge levied by utilities to recover fees orimposts other than taxes.Therm: A unit of heating value equal to 100,000 Btu.Transportation Rates: Rates for the transportation ofcustomer-owned gas. These rates do not includepurchase or procurement of gas.Voltage Discounts: Most electric utilities offer discountedrates to customers who will take service at voltagesother than the general distribution voltages. Thevoltages for which discounts are generally offeredare Secondary, Primary, Sub-transmission andTransmission. The actual voltage of each of theselevels vary from utility to utility.References1. Acton, J.P., Gelbard, E.H., Hosek, J.R., & Mckay, D.J. (1980, February).British Industrial Response to the Peak-Load Pricing of Electricity.The Rand Corporation, R-2508-DOE/DWP.2. David, A.K., & Li, Y.Z. (1991, November). A Comparison ofSystem Response For Different Types of Real-Time Pricing. IEEEInternational Conference on Advances In Power System Control,Operation and Management. Hong Kong, p. 385-390.3. Anonymous (1997, August). Energy User News. Chilton Co., p.32.4. EPRI (1980, October). Industrial Response To Time Of Day Pricing—ATechnical and Economic Assessment Of Specific LoadManagement Strategies. Gordian Associates, EA-1573, ResearchProject 1212-2.5. Hanser, P., Wharton, J., & Fox-Penner, P. (March 1, 1997). RealtimePricing—Restructuring’s Big Bang? Public Utilities Fortnightly,135 (5), p. 22-30.6. Kirsch, L. D., Sullivan, R.L., & Flaim, T.A. (1988, August). DevelopingMarginal Costs For Real-Time Pricing. IEEE Transactions onPower Systems, 3 (3), p. 1133-1138.7. Mykytyn Consulting group, Inc. (1997). Electric Utilities andTariffs. PowerRates [Online]. Available: http://www.mcgi.com/pr/samples/utility_list.html [November 4, 1997].8. O’Sheasy, M. (MIKE.T.OSHEASY@gpc.com). (1998, March 12).RTP. E-mail to Mont, J. (mjavier@okstate.edu).9. Tabors, R.D., Schweppe, F.C., & Caraminis, M.C. (1989, May).Utility Experience with Real-Time Rates. Transactions on PowerSystems, 4(2), p. 463-471.10. Tolley, D.L. (1988, January). Industrial Electricity Tariffs. PowerEngineering Journal. p. 27-34.

CHAPTER 19THERMAL ENERGY STORAGECLINT CHRISTENSONJohnson Controls, Inc.19.1 INTRODUCTIONA majority of the technology developed for energymanagement has dealt with the more efficient consumptionof electricity, rather than timing the demand forit. Variable frequency drives, energy efficient lights,electronic ballasts and energy efficient motors are a fewof these consumption management devices. These techniquesoften only impact a small portion of the facilitiesdemand (when compared to say the mechanical coolingequipment), which is normally a major portion of thefacilities overall annual electric bill. The management ofdemand charges deals very little with conservation ofenergy, but mainly with the ability of a generator to supplypower when needed. It is this timing of consumptionthat is the basis of demand management and the focus ofthermal energy storage (TES).Experts agree that demand management is actuallynot a form of energy conservation but a form of cost management.Throughout the 1980’s and most of the 1990’s,Demand Side Management (DSM) was done by utilitiesin order to manage generating capacity and costs by promotingdemand reduction though incentives (financialrewards) and disincentives (rate structures). Most of theincentive programs have ceased due to surplus generationcapacity and the approach of retail electrical deregulation.A deregulated market place will surely impact thecost of energy for many customers and if the commoditypricing experiments of the recent past are any indicationof the future, demand costs, and thus demand management,will remain as an important cost control strategyfor utilities and the energy users.Utilities often charge more for energy and demandduring certain periods in the form of on-peak rates andratchet clauses. The process of managing the generationcapacity that a particular utility has “on-line” involvesthe utilization of those generating units that producepower most efficiently first since these units would havethe lowest avoided costs (ultimately the actual cost ofenergy). When the loads are approaching the connectedgeneration capacity of the utility, additional generating519Figure 19.1 Typical office building chiller consumptionprofile.units must be brought on line. Each additional unit has anincrementally higher avoided cost since these “peakingunits” units are less efficient and used less often. This hasprompted many organizations to implement some formof demand management.Thermal energy storage (TES) is the concept ofgenerating and storing energy in the form of heat or coldfor use during peak periods. For the profile in Figure19.1, a cooling storage system could be implemented toreduce or eliminate the need to run the chillers duringthe on-peak rate period. By running the chillers duringoff-peak hours and storing this capacity for use during the onpeakhours, a reduction in energy costs can be realized. If thistype of system is implemented during new constructionor when equipment is being replaced, smaller capacitychillers can be installed, since the chiller can spread theproduction of the total load over the entire day, ratherthan being sized for peak loads.

520 ENERGY MANAGEMENT HANDBOOKThermal energy storage has been used for centuries,but only recently have large electrical users taken advantageof the technique for cost management. The processinvolves storing Btu’s (or lack of Btu’s) for use when eithera heat source or a heat sink is required. The use of eaves,root cellars, ground coupled heat pump systems, andadobe type thermal mass could all be considered forms ofthermal storage. Today, the ability to take advantage of asource of inexpensive energy (whether waste heat sourceor time based rate structure) for use during a later time ofmore expensive energy has extended the applications ofTES. For this particular chapter, the focus of discussion willconcentrate on the storage of cooling capacity and the storageof heat will not be considered. The two main drivingforces behind the storage of cooling capacity, rate structureand cooling system management, will be discussed in thefollowing paragraphs.Often the chiller load and efficiency follow the chillerconsumption profile, in that the chiller is running athigh load, i.e. high efficiency, only a small portion of theday. This is due to the HVAC system having to producecooling when it is needed as well as to be able to handleinstantaneous peak loads. With smaller chiller systemsdesigned to handle the base and peak loads during offpeakhours, the chillers can run at higher average loadsand thus higher efficiencies. Appendix A following thischapter lists several manufacturers of thermal energystorage systems.Thermal energy storage also has the ability to balancethe daily loads on a cooling system. Conventionalair conditioning system must employ a chiller largeenough to handle the peak cooling demand as it occurs.This mandates that the cooling system will be required tooperate in a load following mode, varying the output ofthe system in response to changes in the cooling requirements.In systems that operate within a one or two shiftoperation or those that are much more climatically based,can benefit from the smoothing characteristics of TES. Aschool for example that adds a new wing, could utilizethe existing refrigeration system during the evening togenerate cooling capacity to be stored for use during theday. Although additional piping and pumping capacitywould need to be added to the addition, new chiller capacitymay not have to be added. A new construction projectthat would have similar single shift cooling demandprofile could utilize a smaller chiller in combination withstorage to better balance the chiller operation. This couldsignificantly reduce the capital cost of the renovation inaddition to any rate based savings as discussed above.Companies often control the demand of electricityby utilizing some of the techniques listed above and otherconsumption management actions which also reducedemand. More recently the ability to shift the time whenelectricity is needed has provided a means of balancingor shifting the demand for electricity to “off-peak” hours.This technique is often called demand balancing or demandshifting. This demand balancing may best be seenwith the use of an example 24-hour chiller consumptionplot during the peak day, Figure 19.1 and Table 19.1. Thisfacility exhibits a typical single shift building load profile.Note that the load listed in this table for the end of hour1 identifies the average load between midnight and 1:00a.m., and that for end of hour 2 is the average load between1 and 2 a.m., and so on. This example will employa utility rate schedule with a summer on-peak demandperiod from 10 am to 5:59 p.m., an 8-hour period. Movingload from the on-peak rate period to the off-peak periodcan both balance the demand and reduce residual ratch-Table 19.1 Example chiller consumption profile————————————————————————Chiller Consumption ProfileChiller LoadEnd————————————————————————of Hour (Tons) Rate1 100 Reg2 120 Reg3 125 Reg4 130 Reg5 130 Reg6 153 Reg7 165 Reg8 230 Reg9 270 Reg10 290 Reg11 340 On-Peak12 380 On-Peak13 450 On-Peak14 490 On-Peak15 510 On-Peak16 480 On-Peak17 410 On-Peak18 360 On-Peak19 250 Reg20 210 Reg21 160 Reg22 130 Reg23 125 Reg————————————————————————24 115 RegDaily Total 6123 Ton-HrsDaily————————————————————————Avg. 255.13 TonsPeak Total 3420 Ton-HrsPeak————————————————————————Demand 510 Tons

THERMAL ENERGY STORAGE 521eted peak charges. Thermal energy storage is one methodavailable to accomplish just that.19.2 STORAGE SYSTEMSThere are two general types of storage systems, onesthat shut the chiller down during on-peak times and runcompletely off the storage system during that time areknown as “full storage systems.” Those designed to havethe chiller run during the on-peak period supplementingthe storage system are known as “ partial storagesystems.” The full storage systems have a higher firstcost since the chiller is off during peaking times and thecooling load must be satisfied by a larger chiller runningfewer hours and a larger storage system storing the excess.The full storage systems do realize greater savingsthan the partial system since the chillers are completelyturned off during on-peak periods. Full storage systemsare often implemented in retrofit projects since a largechiller system may already be in place.A partial storage system provides attractive savingswith less initial cost and size requirements. New constructionprojects will often implement a partial storagesystem so that the size of both the chiller and the storagesystem can be reduced. Figures 19.2 and 19.3 and Tables19.2 and 19.3 demonstrate the chiller load required to satisfythe cooling needs of the office building presented inFigure 19.1 for the full and partial systems, respectively.Column 2 in these tables represents the building coolingload each hour, and column 3 represents the chiller outputfor each hour. Discussion of the actual calculationsthat are required for sizing these different systems is includedin a subsequent section. For simplicity sake, thesenumbers do not provide for any system losses, which willalso be discussed in a later section.The full storage system has been designed so thatthe total daily chiller load is produced during the offpeakhours. This eliminates the need to run the chillersduring the on-peak hours, saving the increased rates fordemand charges during this period and as well as anyfuture penalties due to ratchet clauses. The partial storagesystem produces 255.13 tons per hour during the entireday, storing excess capacity for use when the buildingdemand exceeds the chiller production. This providesthe ability to control the chiller load, limit the peak chillerdemand to 255.13 kW, * and still take advantage of the offpeakrates for a portion of the on-peak chiller load.3.517*assuming COP = 3.5, then kW/ton = ——— = 1.0 kW/tonCOPTable 19.2 Full storage chiller consumption profile.Chiller Consumption Profile—Full Storage System1 2 3 4—————————————————————————End of Cooling Chiller RateHour (Tons) Load (Tons) Load—————————————————————————21 100 382.69 Reg2 120 382.69 Reg3 125 382.69 Reg4 130 382.69 Reg5 130 382.69 Reg6 153 382.69 Reg7 165 382.69 Reg8 230 382.69 Reg9 270 382.69 Reg10 290 382.69 Reg11 340 0 On-Peak12 380 0 On-Peak13 450 0 On-Peak14 490 0 On-Peak15 510 0 On-Peak16 480 0 On-Peak17 410 0 On-Peak18 360 0 On-Peak19 250 382.69 Reg20 210 382.69 Reg21 160 382.69 Reg22 130 382.69 Reg23 125 382.69 Reg24 115 382.69 Reg—————————————————————————Without storage With storageDaily Total (Ton-Hrs) 6123 6123Daily Avg (Tons): 255.13 1 255.13—————————————————————————Peak Total (Ton-Hrs) 3420 3 0 4Peak Demand (Tons) 510 3 0—————————————————————————41 6123 Ton-Hr—————— = 255.13 Avg Tons24 Hours2 6123 Ton-Hr—————— = 382.69 Avg Tons16 Hours3 This peak load is supplied by the TES, not the chiller4 This is the chiller load and peak during on-peak periodsAn advantage of partial load systems is that theycan provide a means of improving the performance of asystem that can handle the cumulative cooling load, butnot the instantaneous peak demands of the building. Insuch a system, the chiller could be run nearer optimalload continuously throughout the day, with the excesscooling tonnage being stored for use during the peak

522 ENERGY MANAGEMENT HANDBOOKTable 19.3 Partial storage chiller consumption profile.Figure 19.2 Full storage chiller consumption profile.Figure 19.3 Partial storage chiller consumption profile.Chiller Consumption Profile—————————————————————————Partial Storage System—————————————————————————1 2 3 4Hour of Cooling Chiller RateDay Load (Tons) Load (Tons)—————————————————————————11 100 255.13 Reg2 120 255.13 Reg3 125 255.13 Reg4 130 255.13 Reg5 130 255.13 Reg6 153 255.13 Reg7 165 255.13 Reg8 230 255.13 Reg8 270 255.13 Reg10 290 255.13 Reg11 340 255.13 On-Peak12 380 255.13 On-Peak13 450 255.13 On-Peak14 490 255.13 On-Peak15 510 255.13 On-Peak16 480 255.13 On-Peak17 410 255.13 On-Peak18 360 255.13 On-Peak19 250 255.13 Reg20 210 255.13 Reg21 160 255.13 Reg22 130 255.13 Reg23 125 255.13 Reg—————————————————————————24 115 255.13 RegWithout storage With storageDaily Total (Ton-Hrs) 6123 6123Daily—————————————————————————Avg (Tons): 255.13 255.13Peak Total (Ton-Hrs): 3420 2 2041 3Peak Demand (Tons): 510 2 255.13 3—————————————————————————1 6123 Ton-Hr—————— = 255.13 Avg Tons24 Hours2 This peak load is supplied by the TES, supplemented by the chiller3 This is the chiller load and peak during on-peak period.periods. An optional method for utilizing partial storageis a system that already utilizes two chillers. The dailycooling load could be satisfied by running both chillersduring the off-peak hours, storing any excess coolingcapacity, and running only one chiller during the on-peakperiod, to supplement the discharge of the storage system.This also has the important advantage of offering areserve chiller during peak load times. Figure 19.4 shows

THERMAL ENERGY STORAGE 523the chiller consumption profile for this optional partialstorage arrangement and Table 19.4 lists the consumptionvalues. Early and late in the cooling season, the partialload system could approach the full load system characteristics.As the cooling loads and peaks begin to decline,the storage system will be able to handle more of theon-peak requirement, and eventually the on-peak chillercould also be turned off. A system such as this can bedesigned to run the chillers at optimum load, increasingefficiency of the system.19.3 STORAGE MEDIUMSThere are several methods currently in use to storecold in thermal energy storage systems. These are water,ice, and phase change materials. The water systems simplystore chilled water for use during on-peak periods.Ice systems produce ice that can be used to cool the actualchilling water, utilizing the high latent heat of fusion.Phase change materials are those materials that exhibitproperties, melting points for example, that lend themselvesto thermal energy storage. Figure 19.5a representsthe configuration of the cooling system with either a wateror phase change material thermal storage system andFigure 19.5b represents a general configuration of a TESutilizing ice as the storage medium The next few sectionswill discuss these different mediums.Figure 19.4 Optional partial storage chiller profile.19.3.1 Chilled Water StorageChilled water storage is simply a method of storingchilled water generated during off-peak periods ina large tank or series of tanks. These tanks are the mostcommonly used method of thermal storage. One factor tothis popularity is the ease to which these water tanks canbe interfaced with the existing HVAC system. The chillersare not required to produce chilled water any colderthan presently used in the system so the system efficiencyis not sacrificed. The chiller system draws warmer waterfrom one end of the system and this is replaced withchilled water in the other. During the off-peak chargecycle, the temperature of the water in the storage willdecline until the output temperature of the chiller systemis approached or reached This chilled water is then withdrawnduring the on-peak discharge cycle, supplementingor replacing the chiller(s) output.Facilities that have a system size constraint such aslack of space often install a series of small insulated tanksthat are plumbed in series. Other facilities have installeda single, large volume tank either above or below ground.The material and shape of these tanks vary greatly frominstallation to installation. These large tanks are often designedvery similar to municipal water storage tanks. Themain performance factors in the design of these tank systems,either large or multiple, is location and insulation.An Electric Power Research Institute’s (EPRI) CommercialCool Storage Field Performance Monitoring Project(RP-2732-05) Report states that the storage efficiencies oftanks significantly decrease if tank walls were exposedto sunlight and outdoor ambient conditions and/or hadlong hold times prior to discharging 7 . To minimize heatgain, tanks should be out of the direct sun wheneverpossible. The storage efficiency of these tanks is also decreasedsignificantly if the water is stored for extendedperiods.One advantage to using a single large tank ratherthan the series of smaller ones is that the temperature differentialbetween the warm water intake and the chilledwater outlet can be maintained. This is achieved utilizingthe property of thermal stratification where the warmerwater will migrate to the top of the tank and the colderto the bottom. Proper thermal stratification can only bemaintained if the intake and outlet diffusers are locatedat the top and bottom of the tank and the flow rates ofthe water during charge and discharge cycles is keptlow This will reduce a majority of the mixing of the twotemperature waters. Another method used to assure thatthe two temperature flows remain separated is the useof a movable bladder, creating a physical partition. Onetop/bottom diffuser tank studied in the EPRI study used

524 ENERGY MANAGEMENT HANDBOOKTable 19.4 Partial storage chiller consumption profile.1 2 3 4Hour of Day Cooling Load Chiller Load 1,2 Rate—————————————————————————(Tons) (Tons)1 100 306 Reg2 120 306 Reg3 125 306 Reg4 130 306 Reg5 130 306 Reg6 153 306 Reg7 165 306 Reg8 230 306 Reg9 270 306 Reg10 290 306 Reg11 340 153 On-Peak12 380 153 On-Peak13 450 153 On-Peak14 490 153 On-Peak15 510 153 On-Peak16 480 153 On-Peak17 410 153 On-Peak18 360 153 On-Peak19 250 306 Reg20 210 306 Reg21 160 306 Reg22 130 306 Reg23 125 306 Reg—————————————————————————24 115 306 RegWithout storage With storageDaily Total (Ton-Hrs): 6123 6123Daily—————————————————————————Avg (Tons): 255.13 255.13On-Peak (Ton-Hrs): 3420 3 1225 4Peak—————————————————————————Demand (Tons): 510 3 153 41 (6123 Ton-Hr) (2 Chillers Operating)———————————————————— = 306 Tons(16 Hours)(2 Chillers) + (8 Hours)(1 Chiller)Figure 19.5a Water & eutectic storage system configuration.2 (6123 Ton-Hr) (1 Chiller Operating)———————————————————— = 153 Tons(16 Hours)(2 Chillers) + (8 Hours)(1 Chiller)3 This peak load is supplied by the TES, supplemented by the chiller.4 This is the chiller load and peak during the on-peak period.—————————————————————————a thermocouple array, installed to measure the chilledwater temperature at one foot intervals from top to bottomof the tank. This tank had a capacity of 550,000 gallonsand was 20 feet deep but had only a 2.5 foot blendzone over which the temperature differential was almost20 degrees 7 .The advantages of using water as the thermal storagemedium are:Figure 19.5b Ice storage system configuration.

THERMAL ENERGY STORAGE 5251. Retrofitting the storage system with the existingHVAC system is very easy,2. Water systems utilize normal evaporator temperatures,3. With proper design, the water tanks have good thermalstorage efficiencies,4. Full thermal stratification maintains chilled watertemperature differential, maintaining chiller loadingand efficiencies, and5. Water systems have lower auxiliary energy consumptionthan both ice and phase change materials sincethe water has unrestricted flow through the storagesystem.19.3.2 Ice StorageIce storage utilizes water’s high latent heat of fusionto store cooling energy. One pound of ice stores 144Btu’s of cooling energy while chilled water only contains1 Btu per pound –°F 7,8 . This reduces the required storagevolume approximately 75% 7 if ice systems are used ratherthan water. Ice storage systems form ice with the chillersystem during off-peak periods and this ice is used togenerate chilled water during on-peak periods.There are two main methods in use to utilize ice foron-peak cooling. The first is considered a static system inwhich serpentine expansion coils are fitted within a insulatedtank of cooling water. During the charging cycle,the cooling water forms ice around the direct expansioncoil as the cold gases pass through it (see Figure 19.5b).The thickness of the ice varies with the ice building time(charge time) and heat transfer area. During the dischargecycle, the cooling water contained in the tank is used tocool the building and the warmer water returned fromthe building is circulated through the tank, melting theice, and using its latent heat of fusion for cooling.The second major category of thermal energy storagesystems utilizing ice can be considered a dynamicsystem. This system has also been labeled a plate ice makeror ice harvester. During the charging cycle the coolingwater is pumped over evaporator “plates” where ice isactually produced. These thin sheets of ice are fed into thecooling water tank, dropping the temperature. Duringon-peak periods, this chilled water is circulated throughthe building for cooling. This technology is considereddynamic due to the fact that the ice is removed from theevaporator rather than simply remaining on it.Static ice storage systems are currently available infactory-assembled packaged units which provide easeof installation and can provide a lower initial capitalcost. When compared to water storage systems, the sizeand weight reduction associated with ice systems makesthem very attractive to facilities with space constraintsOne main disadvantage to ice systems is the fact that theevaporator must be cold enough to produce ice. Theseevaporator temperatures usually range from 10° to 25°while most chiller evaporator temperatures range from42° to 47° 9 . This required decrease in evaporator temperatureresults in a higher energy demand per ton causingsome penalty in cooling efficiency. The EPRI Project reportedthat chillers operating in chilled water or eutecticsalt (phase change material) used approximately 20% lessenergy than chillers operating in ice systems (0.9 vs. 1.1kW/ton) 7 . The advantages of using ice as the thermalstorage medium are:1. Retrofitting the storage system with the existingHVAC chilled water system is feasible,2. Ice systems require less space than that required bythe water systems,3. Ice systems have higher storage but lower refrigerationefficiencies than those of water, and4. Ice systems are available in packaged units, due tosmaller size requirements19.3.3 Phase Change MaterialsThe benefit of capturing latent heat of fusion whilemaintaining evaporating temperatures of existing chillersystems can be realized with the use of phase changematerials. There are materials that have melting pointshigher than that of water that have been successfullyused in thermal energy storage systems. Several of thesematerials fall into the general category called “ eutecticsalts” and are salt hydrates which are mixtures of inorganicsalts and water. Some eutectic salts have melting(solidifying) points of 47° 7 , providing the opportunityfor a direct retrofit using the existing chiller system sincethis is at or above the existing evaporator temperatures.In a thermal storage system, these salts are placed in plasticcontainers, which are immersed within an insulatedchilled water tank. During the charging cycle, the chilledwater flows through the gaps between the containers,freezing the salts within them. During the on-peakdischarge, the warmer building return water circulatesthrough the tank, melting the salts and utilizing the latentheat of fusion to cool the building. These salt solutionshave latent heat of fusion around 40 Btu/lb 9 .This additional latent heat reduces the storage vol-

526 ENERGY MANAGEMENT HANDBOOKume by 66% of that required for an equivalent capacitywater storage system 9 . Another obvious benefit of usingeutectic salts is that the efficiency of the chillers is notsacrificed, as stated earlier, since the phase change occursaround normal evaporator suction temperatures. Oneproblem with the eutectic salt systems is that the auxiliaryenergy consumption is higher since the chilled watermust be pumped through the array of eutectic blocks.The auxiliary energy consumption of the ice systems ishigher than both the water and eutectic salt systems sincethe chilled water must be pumped through the ice systemcoils, nozzles, and heat exchangers. The EPRI studyfound that the chilled water systems had an average auxiliaryenergy use of 0.43 kWh/Ton-Hr compared to thephase change systems (eutectic and ice) average auxiliaryenergy use of 0.56 kWh/Ton-Hr 5 . The advantages of usingeutectic salts as the thermal storage medium are thatthey:1. can utilize the existing chiller system for generatingstorage due to evaporator temperature similarity,2. require less space than that required by the watersystems, and3. have higher storage and equivalent refrigeration efficienciesto those of water.19.4 SYSTEM CAPACITYThe performance of thermal storage systems dependsupon proper design. If it is sized too small or toolarge, the entire system performance will suffer. The followingsection will explain this sizing procedure for theexample office building presented earlier. The facility hasa maximum load of 510 tons, a total cooling requirementof 6,123 Ton-Hours, and a on-peak cooling requirementof 3,420 Ton-Hours. This information will be analyzed tosize a conventional chiller system, a partial storage system,a full storage system, and the optional partial storagesystem. These results will then be used to determinethe actual capacity needed to satisfy the cooling requirementsutilizing either a chilled water, a eutectic salt, oran ice thermal storage system. Obviously some greatlysimplifying assumptions are made.19.4.1 Chiller System CapacityThe conventional system would need to be able tohandle the peak load independently, as seen in Figure19.1. A chiller or series of chillers would be needed toproduce the peak cooling load of 510 tons. Unfortunately,packaged chiller units usually are available in incrementsthat mandate excess capacity but for simplicity one 600-ton chiller will be used for this comparison. The conventionalchiller system will provide cooling as it is neededand will follow the load presented in Figure 19.1 andTable 19.1.To determine the chiller system requirement of acooling system utilizing partial load storage, further analysisis needed. Table 19.1 showed that the average coolingload of the office building was 255.13 tons per hour. Theideal partial load storage system will run at this load (seeFigure 19.3 and Table 19.3). The chiller system wouldneed to be sized to supply the 255.13 tons per hour, soone 300-ton chiller will be used for comparison purposes.Table 19.5 shows how the chiller system would operate at255.13 tons per hour, providing cooling required for thebuilding directly and charging the storage system withthe excess. Although the storage system supplements thecooling system for 2 hours before the peak period, thecooling load is always satisfied.Comparing the peak demand from the bottomsof columns 2 and 3 of Table 19.5 shows that the partialstorage system reduced this peak load almost 50% (510– 255.13 = 254.87 Tons). Column 4 shows the tonnage thatis supplied to the storage system and column 5 shows theamount of cooling contained in the storage system at theend of each hour of operation. This system was design sothat there would be zero capacity remaining in the thermalstorage tanks after the on-peak period. The valuescontained at the bottom of Table 19.5 are the total storagerequired to assure that there is no capacity remainingand the maximum output required from storage. Thesevalues will be utilized in the next section to determine thestorage capacity required for each of the different storagemediums.The full storage system also requires some calculationsto determine the chiller system size. Since the chillerswill not be used during the on-peak period, the entiredaily cooling requirement must be generated during theoff-peak periods. Table 19.1 listed the total cooling load as6,123 Ton-Hours for the peak day. Dividing this load overthe 16 off-peak hours yields that the chillers must generate383 tons of cooling per hour (6,123 Ton-Hours/16hours). A 450-ton chiller will be utilized in this situationfor comparison purposes. Table 19.6 shows how the chillersystem would operate at 383 tons per hour, providingcooling required for the building directly and chargingthe storage system with the excess.Comparing the peak demand with and withoutstorage in Table 19.6 shows that the full storage systemeliminated all load from the on-peak period. Column 4shows the tonnage that is supplied to the storage system

THERMAL ENERGY STORAGE 527Table 19.5 Partial storage operation profile.Thermal Storage Operation Profile———————————————————————————————————————————Partial Storage System1 2 3 4 5 6End of Cooling Chiller Capacity to Capacity StorageHour Load Load Storage In Storage Cycle———————————————————————————————————————————(Tons) (Tons) (Ton-Hrs) (Ton-Hrs)1 100 255.13 155 696 Charge2 120 255.13 135 831 Charge3 125 255.13 130 961 Charge4 130 255.13 125 1086 Charge5 130 255.13 125 1211 Charge6 153 255.13 102 1314 Charge7 165 255.13 90 1404 Charge8 230 255.13 25 1429 Charge9 270 255.13 -15 1414 Discharge10 290 255.13 -35 1379 Discharge11 340 255.13 -85 1294 Discharge12 380 255.13 -125 1169 Discharge13 450 255.13 -195 974 Discharge14 490 255.13 -235 740 Discharge15 510 255.13 -255 485 Discharge16 480 255.13 -225 260 Discharge17 410 255.13 -155 105 Discharge18 360 255.13 -105 0 Discharge19 250 255.13 5 5 Charge20 210 255.13 45 50 Charge21 160 255.13 95 145 Charge22 130 255.13 125 271 Charge23 125 255.13 130 401 Charge———————————————————————————————————————————24 115 255.13 140 541 ChargeWithout storage With storageDaily Total (Ton-Hrs): 6123 6123Daily Avg (Tons): 255.13 255.13———————————————————————————————————————————Peak Total (Ton-Hrs): 3420 2041 Storage Total = 1429Peak Demand (Tons): 510 255.13 Peak Storage Output = 255———————————————————————————————————————————Column 4 = Column 3 – Column 2Column 5(n) = Column 5(n–1) + Column 4(n)and column 5 shows the amount of cooling containedin the storage system at the end of each hour of operation.This system was designed so that there would be 0capacity remaining in the thermal storage tanks after theon-peak period, as shown at the bottom of Table 19.6. Thevalues in Table 19.6 will be utilized in the next section todetermine the storage capacity required for each of thedifferent storage mediums.The optional partial storage system is a blend of thetwo systems presented earlier. Values given in Table 19.7and Figure 19.4 are one combination of several possibilitiesthat would drop the consumption and peak demandduring the on-peak period. Once again this system hasbeen designed to run both chillers during off-peak hoursand run only one during on-peak hours. Benefits of thisarrangement are that the current chiller system could beused in combination with the storage system and thatthe storage system does not require as much capacity asthe full storage system. Also, a reserve chiller is availableduring peak-load times.Comparing the peak demand with and withoutstorage in Table 19.7 shows that the optional partial stor-

528 ENERGY MANAGEMENT HANDBOOK———————————————————————————————————————————Table 19.6 Full storage operation profile.Thermal Storage Operation Profile———————————————————————————————————————————Full Storage System1 2 3 4 5 6Hour of Cooling Chiller Capacity to Capacity StorageDay Load Load Storage In Storage Cycle———————————————————————————————————————————(Tons) (Tons) (Ton-Hrs) (Ton-Hrs)1 100 383 283 1589 Charge2 120 383 263 1852 Charge3 125 383 258 2109 Charge4 130 383 253 2362 Charge5 130 383 253 2615 Charge6 153 383 230 2844 Charge7 165 383 218 3062 Charge8 230 383 153 3215 Charge9 270 383 113 3327 Charge10 290 383 93 3420 Charge11 340 0 -340 3080 Discharge12 380 0 -380 2700 Discharge13 450 0 -450 2250 Discharge14 490 0 -490 1760 Discharge15 510 0 -510 1250 Discharge16 480 0 -480 770 Discharge17 410 0 -410 360 Discharge18 360 0 -360 0 Discharge19 250 383 133 133 Charge20 210 383 173 305 Charge21 160 383 223 528 Charge22 130 383 253 781 Charge23 125 383 258 1038 Charge———————————————————————————————————————————24 115 383 268 1306 ChargeWithout storage With storageDaily Total (Ton-Hrs): 6123 6123Daily———————————————————————————————————————————Avg (Tons): 255.13 255.13Peak Total (Ton-Hrs): 3420 0 Storage Total = 3420Peak———————————————————————————————————————————Demand (Tons): 510 0 Peak Storage Output = 510Column 4 = Column 3 – Column 2Column 5(n) = Column 5(n–1) + Column 4(n)age system reduces the peak load from 510 tons to 153tons, or approximately 70% during the on-peak period.Column 4 shows the tonnage that is supplied to the storagesystem and column 5 shows the amount of coolingcontained in the storage system at the end of each hour ofoperation. This system was designed so that there wouldbe zero capacity remaining in the thermal storage tanksafter the on-peak period. The values contained at the bottomof Table 19.7 are the total storage capacity requiredand the maximum output required from storage. Thesevalues will be utilized in the next section to determine thestorage capacity required for each of the different storagemediums. Table 19.8 summarizes the performanceparameters for the three configurations discussed above.The next section summarizes the procedure used to determinethe size of the storage systems required to handlethe office building.19.4.2 Storage System CapacityEach of the storage mediums has different size requirementsto satisfy the needs of the cooling load. Thissection will describe the procedure to find the actual

THERMAL ENERGY STORAGE 529Table 19.7 Optional partial storage operation profile.——————————————————————————————————————————————————————————————————————————————————————Thermal Storage Operation Profile—Optional Partial Storage System1 2 3 4 5 6End of Cooling Chiller Capacity to Capacity StorageHour Load Load Storage In Storage Cycle(Tons) (Tons) (Ton-Hrs) (Ton-Hrs)———————————————————————————————————————————1 100 306 206 1053 Charge2 120 306 186 1239 Charge3 125 306 181 1420 Charge4 130 306 176 1597 Charge5 130 306 176 1773 Charge6 153 306 153 1926 Charge7 165 306 141 2067 Charge8 230 306 76 2143 Charge9 270 306 36 2179 Charge10 290 306 16 2195 Charge11 340 153 -187 2008 Discharge12 380 153 -227 1782 Discharge13 450 153 -297 1485 Discharge14 490 153 -337 1148 Discharge15 510 153 -357 791 Discharge16 480 153 -327 464 Discharge17 410 153 -257 207 Discharge18 360 153 -207 0 Discharge19 250 306 56 56 Charge20 210 306 96 152 Charge21 160 306 146 298 Charge22 130 306 176 475 Charge23 125 306 181 656 Charge24 115 306 191 847 Charge———————————————————————————————————————————Without storage With storageDaily Total (Ton-Hrs): 6123 6123Daily Avg (Tons): 255.13 255.12———————————————————————————————————————————Peak Total (Ton-Hrs): 3420 1225 Storage Total = 2195Peak Demand (Tons): 510 153 Peak Storage Output = 357———————————————————————————————————————————Column 4 = Column 3 - Column 2Column 5(n) = Column 5(n-1) + Column 4(n)volume or size of the storage system for the partial loadsystem for each of the different storage mediums. Thedesign of the chiller and thermal storage system mustprovide enough chilled water to the system to satisfy thepeak load, so particular attention should be paid to thepumping and piping. Table 19.9 summarizes the size requirementof each of the three different storage options.To calculate the capacity of the partial load storagesystem, the relationship between capacity (C), mass (M),specific heat of material (Cp), and the coil temperaturedifferential (T 2 –T 1 ) shown in Figure 19.5a will be used:C = M Cp (T 2 –T 1 )where: M = lbmCp = Btu/lbm °R(T 2 –T 1 ) = °RThe partial load system required that 1,429 Ton-Hrs bestored to supplement the output of the chiller during onpeakperiods. This value does not allow for any thermalloss which normally occurs. For this discussion, a conservativevalue of 20% is used, which is an average suggestedin the EPRI report 7 . This will increase the storage

530 ENERGY MANAGEMENT HANDBOOKTable 19.8 System performance comparison.———————————————————————————————————————————SYSTEM—————————————————————————Conventional Partial Full OptionalPERFORMANCE PARAMETERS No Storage Storage Storage Partial———————————————————————————————————————————Overall Peak Demand (Tons) 510 255.13 383 306On-Peak, Peak Demand (Tons) 510 255.13 0 153On-Peak Chiller Consumption 3,420 2,041 0 1,225(Ton-Hrs)Required Storage Capacity 1 — 1,379 3,420 2,195(Ton-Hrs)MAXIMUM STORAGE OUTPUT 1 — 255 510 357(Tons)———————————————————————————————————————————1 Values from Table 19.5, 19.6, and 19.7. Represent the capacity required to be supplied by the TES.requirements to 1,715 Ton-Hrs and chilled water storagesystems in this size range cost approximately $200/Ton-Hr including piping and installation 5 . Assuming thatthere are 12,000 Btu’s per Ton-Hr, this yields:C = (1,715 Ton-Hrs)*(12,000 Btu/Ton-Hr)= 20.58 × 10 6 Btu’s.Assuming (T 2 –T 1 ) = 12° and Cp = 1 Btu/lb m °R, the relationbecomes:CM = —————— =Cp(T 2 – T 1 )20.58 × 10 6 Btus—————————— = 1.72 x 10 6 lbm H 2 O1 Btu/lb m –°R)(12°R)1.72 x 10 6 lb mVolume of Water = Mass/Density = ————————62.5 lb m /Ft 3= 27,520 Ft 3 or1.72 × 10 6 lb m——————— = 206,235 gal.8.34 lb m /galSizing the storage system utilizing ice is completedin a very similar fashion. The EPRI study states that theice storage tanks had average daily heat gains 3.5 timesgreater than the chilled water and eutectic systems dueto the higher coil temperature differential (T 2 –T 1 ) Toallow for these heat gains a conservative value of 50%will be added to the actual storage capacity, which is anaverage suggested in the EPRI report 7 This will increasethe storage requirements to 2,144 Ton-Hrs. Assumingthat there are 12,000 Btu’s per Ton-Hr, this yields: (2,144Ton-Hrs)*(12,000 Btu’s/Ton-Hr) = 25.73 × 10 6 Btu’s. Theice systems utilize the latent heat of fusion so the C 1 nowbecomesC 1 = Latent Heat = 144 Btu/lb m .Because the latent heat of fusion, which occurs at32°F, is so large compared to the sensible heat, the sensibleheat (Cp) is not included in the calculation. The massof water required to be frozen becomes:25.73 × 10 6 BtusM = C/C 1 = ———————— = 1.79 × 10 5 lb m H 2 O(144 Btu/lb m )Mass 1.79 × 10 5 lb mVolume of Ice = ———— = ——————Density 62.5 lb m /Ft 3= 2,864 Ft 3This figure is conservative since the sensible heathas been ignored but calculates the volume of ice neededto be generated. The actual volume of ice needed willvary and the total amount of water contained in thetank around the ice coils will vary greatly. The ability topurchase pre-packaged ice storage systems makes theirsizing quite easy For this situation, two 1,080 Ton-Hr icestorage units will be purchased for approximately $150/

THERMAL ENERGY STORAGE 531Table 19.9 Complete system comparison.———————————————————————————————————————————SYSTEM———————————————————————————Conventional Partial Full OptionalPerformance Parameters No Storage Storage Storage Partial———————————————————————————————————————————CHILLERSIZE (# and Tons) 1 @ 600 1 @ 300 1 @ 450 2 @ 175COST($) 180,000 90,000 135,000 105,000WATER STORAGECapacity (Ton-Hrs) — 1,715 4,104 2,634Volume (cubic feet) — 27,484 65,769 42,212Volume (gallons) — 205,635 492,086 315,827Cost per Ton-Hr ($) — 200 135 165Storage cost ($) — 343,000 554,040 434,610ICE STORAGECapacity (Ton-Hrs) 2,144 5,130 3,293# and size (Ton-Hrs) — 2 @ 1,080 4 @ 1,440 3 @ 1,220Ice volume (cubic feet) — 2,859 6,840 4,391Cost per Ton-Hr (S) — 150 150 150Storage cost ($) 1 — 324,000 864,000 549,000EUTECTIC STORAGECapacity (Ton-Hrs) 1,715 4,104 2,634Eutectic vol (cubic feet) — 8,232 19,699 12,643Cost per Ton-Hr ($) — 250 200 230Storage cost ($) — 428,750 820,000 605,820———————————————————————————————————————————1 (2 units)(1,080 Ton-Hrs/units)($150/Ton-Hr) = $324,000Note: The values in this table vary slightly from those in the text from additional significant digits.Ton-Hr including piping and installation 4 (note that thisprovides 2,160 Ton-Hrs compared to the needed 2,144Ton-Hrs).Sizing the storage system utilizing the phase changematerials or eutectic salts is completed just as the ice storagesystem. The EPRI study states that the eutectic saltstorage tanks had average daily heat gains approximatelythe same as that of the chilled water systems. To allowfor these heat gains a conservative value of 20% is addedto the actual storage capacity 5 . This increases storage requirementsto 1,715 Ton-Hrs. Assuming there are 12,000Btu’s per Ton-Hr, this yields: (1,715 Ton-Hrs)*(12,000Btu’s/Ton-Hr) = 20.58 × 10 6 Btu’s.The eutectic system also utilizes the latent heat offusion like the ice system and the temperature differentialshown in Figure 19.5a is not used in the calculation. TheC 1 now becomes:C 1 = Latent Heat = 40 Btu/lbm20.58 × 10 6 BtusM = C/C 1 = ———————— = 5.15 × 10 5 lbm(40 Btu/lmb)Volume of Mass 5.15 × 10 5 lbmEutectic Salts (assuming = ———– = ———————density = water) Density 62.5 lbm/ft 3= 8,232 ft 3The actual volume of eutectic salts needed wouldneed to be adjusted for density differences in the variouscombinations of the salts. Eutectic systems have not beenstudied in great detail and factory sized units are notyet readily available. The EPRI report 7 studied a systemthat required 1,600 Ton-Hrs of storage which utilizedapproximately 45,000 eutectic “bricks” contained in an

532 ENERGY MANAGEMENT HANDBOOK80,600 gallon tank of water. For this situation, a similareutectic storage unit will be purchased for approximately$250/Ton-Hr including piping and installation. The ratioof Ton-Hrs required for partial storage and the requiredtank size will be utilized for sizing the full and optionalpartial storage systems.Table 19.9 summarizes the sizes and costs of thedifferent storage systems and the actual chiller systemsfor each of the three storage arrangements. The valuespresented in this example are for a specific case and eachapplication should be analyzed thoroughly. The cost perton hour of a water system dropped significantly as thesize of the tanks rises as will the eutectic systems since theengineering and installation costs are spread over morecapacity. Also we ignored the sensible heat of the ice andeutectic systems.19.5 ECONOMIC SUMMARYTable 19.9 covered the approximate costs of each ofthe three system configurations utilizing each of the threedifferent storage mediums. Table 19.8 listed the variouspeak day performance parameters of each of the systemspresented. To this point, the peak day chiller consumptionhas been used to size the system. To analyze the savingspotential of the thermal storage systems, much moreinformation is needed to determine daily cooling andchiller loads and the respective storage system performance.To calculate the savings accurately, a daily chillerconsumption plot is needed for at least the summer peakperiod. These values can then be used to determine thechiller load required to satisfy the cooling demands.Only the summer months may be used since most of thecooling takes place and a majority of the utilities “timeof use” charges (on-peak rates) are in effect during thattime. There are several methods available to estimate orsimulate building cooling load. Some of these methodsare available in a computer simulation format or can alsobe calculated by hand.For the office building presented earlier, an alternativemethod will be used to estimate cooling savings.An estimate of a monthly, average day cooling load willbe used to compare the operating costs of the respectivecooling configurations. For simplicity, it is assumed thatthe peak month is July and that the average cooling day is90% of the cooling load of the peak day. The average coolingday for each of the months that make up the summercooling period are estimated based upon July’s averagecooling load. These factors are presented in Table 19.10 forJune through October 11 . These factors are applied to thehourly chiller load of the average July day to determine theseason chiller/TES operation loads. The monthly averageday, hourly chiller loads for each of the three systems arepresented in Table 19.11. The first column for each monthin Table 19.11 lists the hourly cooling demand. The chillerconsumption required to satisfy this load utilizing each ofthe storage systems is also listed. This table does not accountfor the thermal efficiencies used to size the systemsbut for simplicity, these values will be used to determinethe rate and demand savings that will be achieved afterimplementing the system. The formulas presented for thepeak day thermal storage systems operations have beenused for simplicity. These chiller loads do not representthe optimum chiller load since some of partial systemsapproach full storage systems during the early and latecooling months. The bottom of the table contains the totalsfor the chiller systems. These totaled average day valueswill now be used to calculate the savings. The differencebetween the actual cooling load and the chiller load is theapproximate daily savings for each day of that month.A hypothetical southwest utility rate schedule willbe used to apply economic terms to these savings. Theelectricity consumption rate is $0.04/kWh and the de-Table 19.10 Average summer day cooling load factors.———————————————————————————————————————————MONTH kW FACTOR 1 PEAK TONS 2 kWh FACTOR 1 Ton-Hrs/day 3———————————————————————————————————————————JUNE 0.8 360 0.8 4,322JULY 1 450 1 5,403AUGUST 0.9 405 0.9 4,863SEPT 0.7 315 0.7 3,782OCT 0.5 225 0.5 2,702———————————————————————————————————————————1 kW and kWh factors were estimated to determine utility cost savings.2 The average day peak load is estimated to be 90% of the peak day. The kW factor for each month is multiplied by thepeak months average tonnage. For JUNE: PEAK TONS = (0.8)*(450) = 3603 The average day consumption is estimated to be 90% of the peak day. The kWh factor for each month is multipliedby the peak months average consumption. For JUNE: CONSUMPTION = (0.8)*(5,403) = 4,322

THERMAL ENERGY STORAGE 533Table 19.11 Monthly average day chiller load profiles.mand rate during the summer is $3.50/kW per month forthe peak demand during the off-peak hours and $5.00/kW per month for the peak demand during the on-peakhours. These summer demand rates are in effect fromJune through October. This rate schedule only providessavings from balancing the demand, although utilitiesoften have cheaper off-peak consumption rates. It can beseen that the off-peak demand charge assures that the demandis leveled and not merely shifted. This rate schedulewill be applied to the total values in Table 19.11 andmultiplied by the number of days in each month to determinethe summer savings. These savings are contained inTable 19.12. The monthly average day loads in Table 19.11are assumed to be 90% of the actual monthly peak billingdemand, and are adjusted accordingly in Table 19.12. Thetotal monthly savings for each of the chiller/TES systemsis determined at the bottom of each monthly column.These cost savings are not the only monetary justificationfor implementing TES systems. Utilities often ex-tend rebates and incentives to companies installing thermalenergy storage systems to shorten their respectivepayback period. This helps the utility reduce the need tobuild new generation plants. The southwest utility servingthe office building studied here offers $200 per designday peak kW shifted to off-peak hours up to $200,000.19.6 CONCLUSIONSThermal energy storage will play a large role in thefuture of demand side management programs of bothprivate organizations and utilities. An organization thatwishes to employ a system wide energy managementstrategy will need to be able to track, predict and controltheir load profile in order to minimize utility costs. Thismanagement strategy will only become more critical aselectricity costs become more variable in a deregulatedmarket. Real time pricing and multi-facility contracts will

534 ENERGY MANAGEMENT HANDBOOKfurther enhance the savings potential of demand management,within which thermal energy storage shouldbecome a valuable tool.The success of the thermal storage system and theHVAC system as a whole depend on many factors:• The chiller load profile,• The utility rate schedules and incentive programs,• The condition of the current chiller system,• The space available for the various systems,• The selection of the proper storage medium, and• The proper design of the system and integration ofthis system into the current system.Thermal storage is a very attractive method for anorganization to reduce electric costs and improve systemmanagement. New installation projects can utilize storageto reduce the initial costs of the chiller system as well assavings in operation. Storage systems will become easier tojustify in the future with increased mass production, technicaladvances, and as more companies switch to storage.References1. Cottone, Anthony M., “Featured Performer: Thermal Storage,” inHeating Piping and Air Conditioning, August 1990, pp. 51-55.2. Hopkins, Kenneth J., and James W. Schettler, “Thermal StorageEnhances Heat Recovery,” in Heating Piping and Air Conditioning,,March 1990, pp. 45-50.3. Keeler, Russell M., “Scrap DX for CW with Ice Storage,” in HeatingPiping and Air Conditioning,, August 1990, pp. 59-62.4. Lindemann, Russell, Baltimore Aircoil Company, PersonnelPhone Interview, January 7, 1992.5. Mankivsky, Daniel K., Chicago Bridge and Iron Company, PersonnelPhone Interview, January 7, 1992.6. Pandya, Dilip A., “Retrofit Unitary Cool Storage System,” inHeating Piping and Air Conditioning,, July 1990, pp. 35-37.7. Science Applications International Corporation, Operation Performanceof Commercial Cool Storage Systems Vols. 1 & 2, ElectricPower Research Institute (EPRI) Palo Alto, September 1989.8. Tamblyn, Robert T., “Optimizing Storage Savings,” in HeatingPiping and Air Conditioning,, August 1990 pp. 43-46.Table 19.12 Summer monthly system utility costs and TES savings.Table 19.13 Available demand management incentives.———————————————————————————————————————————SystemConventional Partial Full OptionalPerformance Parameters No Storage Storage Storage Partial———————————————————————————————————————————Actual On-Peak Demand 1 (kW) 510 255 0 153On-Peak Demand Shifted 2 (kW) 255 510 357Utility Subsidy 3 ($) 51,000 102,000 71,400———————————————————————————————————————————1 Yearly design peak demand from Table 19.8.2 Demand shifted from design day on-peak period. For partial: 510 kW - 255 kW = 255 kW.3 Based upon $200/kW shifted from design day on-peak period. For partial: 255 kW * $200/kW = $51,000.

THERMAL ENERGY STORAGE 535Appendix 19-APartial list of manufacturers of thermal storage systems. Source: Energy User News, Vol. 22,No. 12, December 1997.Manufacturer Storage Type Capacity (ton-hours)——————————————————————————————————————————————Applied Thermal Technologies Ice, Ice coil 450——————————————————————————————————————————————Baltimore Aircoil Co. Ice, gylcol solid ice, ice coil 237-761——————————————————————————————————————————————Berg Chilling Systems Ltd. ice coil 100-10,000+——————————————————————————————————————————————Calmac Manufacturing Corp ice, glycol solid ice, eutectic 570——————————————————————————————————————————————CBI Walker Inc. water, hot water 2,000 -120,000+——————————————————————————————————————————————Chester-Jensen Co. Inc. ice coil 12-1,200——————————————————————————————————————————————Chicago Bridge & Iron Co. water 500+——————————————————————————————————————————————Cryogel ice, encapsulated ice 100-40,000——————————————————————————————————————————————Delta-Therm Corp.unlimited——————————————————————————————————————————————Dunham-Bush Inc. ice 120, 180, 240——————————————————————————————————————————————FAFCO Inc. Ice, gylcol solid ice 125, 250, 375, 500——————————————————————————————————————————————Group Thermo Inc. water, hot water to 1,800 GPH——————————————————————————————————————————————Henry Vogt Machine Co. ice unlimited——————————————————————————————————————————————Morris & Associates ice 50 - 200 tons/day——————————————————————————————————————————————Natgun Corp. ice, water 2,000 and up——————————————————————————————————————————————Paul Mueller Co. ice slurry 3-1000+——————————————————————————————————————————————Perma Pipewater——————————————————————————————————————————————Phoenix Thermal Storage ice, hot water 3 to 5——————————————————————————————————————————————Precision Parts Corp. hot water 440-17,800 gallons——————————————————————————————————————————————Reaction Thermal Systems ice 242-4,244——————————————————————————————————————————————Steffes ETS Inc. ceramic brick 1.32-9 kW——————————————————————————————————————————————Steibel Eltron Inc.ceramic brick——————————————————————————————————————————————Store-More ice coil 270-20,000ice, water, encapsulated ice,——————————————————————————————————————————————The Trane Co. glycol solid ice, ice coil 60-145,000——————————————————————————————————————————————Turbo Refrigeration ice 10-340——————————————————————————————————————————————Vogt Tube Ice ice unlimited——————————————————————————————————————————————York international Corp. ice 200 - 60,000——————————————————————————————————————————————9. Thumann, Albert, Optimizing HVAC Systems The Fairmont Press,Inc., 1988.10. Thumann, Albert and D. Paul Mehta, Handbook of Energy Engineering,The Fairmont Press, Inc., 1991.11. Wong, Jorge-Kcomt, Dr. Wayne C. Turner, Hemanta Agarwala,and Alpesh Dharia, A Feasibility Study to Evaluate Different Optionsfor Installation of a New Chiller With/Without Thermal Energy StorageSystem, study conducted for the Oklahoma State Office BuildingsEnergy Cost Reduction Project, Revised 1990.

536 ENERGY MANAGEMENT HANDBOOKAppendix 19-BPartial list of Utility Cash Incentive Programs.Source: Dan Mankivsky, Chicago Bridge & Iron, August 1991.STATECASH INCENTIVE- Electric Utility $/kW Shifted Maximum———————————————————————————————————————————ARIZONA- Arizona Public Service 75-125 no limit- Salt River Project 60-250 no limitCALIFORNIA- American Public Utilities Dept. 60 50,000- L.A. Dept of Water & Power 250 40% cost- Pacific Gas & Electric 300 50%-70%- Pasadena Public Utility 300 no limit- Riverside Public Utility 200 no limit- Sacramento Municipal Util Dist. 200 no limit- San Diego Gas & Electric 50-200 no limit- Southern California Edison 100 300,000DISTRICT OF COLUMBIA- Patomac Electric Power Co. 200-250 no limitFLORIDA- Florida Power & Light Co. 250/ton no limit- Florida Power Corp. 160-180 25%- Tampa Electric Co. 200 no limitINDIANA- Indianapolis Power & Light 200 no limit- Northern Indiana Public Service 200/tonMARYLAND- Baltimore Gas & Electric 200 no limit- Patomac Electric Power Co. 200-250 no limitMINNESOTA- Northern States Power 400/ton no limitNEVADA- Nevada Power 100-150 no limitNEW JERSEY- Atlantic Electric 150 200,000- Jersey Central Power & Light 300 250,000- Orange & Rockland Utilities 250 no limit- Public Service Electric & Gas 125-250 no limit(Continued)

THERMAL ENERGY STORAGE 537STATECASH INCENTIVE- Electric Utility $/kW Shifted Maximum———————————————————————————————————————————NEW YORK- Central Edison Gas & Electric 25/Ton-Hr equip cost- Consolidated Edison Co. 600 no limit- Long Island Lighting Co. 300-500 no limit- New York State Electric & Gas 113 no limit- Orange & Rockland Utilities 250 no limit- Rochester Gas & Electric 200-300 70,000NORTH DAKOTA- Northern States Power 400/ton no limitOHIO- Cincinnati Gas L Electric 150 no limit- Toledo Edison 200-250 —OKLAHOMA- Oklahoma Gas & Electric 125-200 225,000PENNSYLVANIA- Metropolitan Edison 100-250 40,000- Orange & Rockland Utilities 250 no limit- Pennsylvania Electric 250 no limit- Pennsylvania Power & Light 100 no limit- Philadelphia Electric 100-200 25,000SOUTH DAKOTA- Northern States Power 400/ton no limitTEXAS- Austin Electric Department 300 150,000- El Paso Electric Company 200 no limit- Gulf States Utilities 250 —- Houston Lighting & Power 350 —- Texas Utilities (Dallas Power, Texas 125-250 no limitElectric Service, and Texas Power & Light)WISCONSIN- Madison Gas & Electric 60 - 80 no limit- Northern States Power 175 no limit- Wisconsin Electric Power 350 no limit*Note: Some states have additional programs not listed here and some of the listed programshave additional limitations.


CHAPTER 20CODES, STANDARDS & LEGISLATIONALBERT THUMANN, P.E., CEMAssociation of Energy EngineersAtlanta, GACLINT CHRISTENSONJohnson Controls, Inc.Editor's Note: EPACT-2005 was just being made federallaw as this edition was being finalized. Look for detailsof EPACT-2005 legislation in future editions.—————————————————————————This chapter presents an historical perspective on keycodes, standards, and regulations which have impactedenergy policy and are still playing a major role in shapingenergy usage. The Energy Policy Act of 1992 is farreaching and its implementation is impacting electricpower deregulation, building codes and new energyefficient products. Sometimes policy makers do not seethe far reaching impact of their legislation. The EnergyPolicy Act for example has created an environmentfor retail competition. Electric utilities will drasticallychange the way they operate in order to provide powerand lowest cost. This in turn will drastically reduce utilitysponsored incentive and rebate programs which haveinfluenced energy conservation adoption.resource planning; allow efficiency programs to be atleast as profitable as new supply options; and encourageimprovements in supply system efficiency.Equipment Standards• Establishes efficiency standards for: commercial heatingand air-conditioning equipment; electric motors;and lamps.• Gives the private sector an opportunity to establishvoluntary efficiency information/labeling programsfor windows, office equipment and luminaires, or theDept. of Energy will establish such programs.Renewable Energy• Establishes a program for providing federal supporton a competitive basis for renewable energy technologies.Expands program to promote export of theserenewable energy technologies to emerging marketsin developing countries.Alternative Fuels• Gives Dept. of Energy authority to require a privateand municipal alternative fuel fleet program starting in1998. Provides a federal alternative fuel fleet programwith phased-in acquisition schedule; also providesstate fleet program for large fleets in large cities.20.1 THE ENERGY POLICY ACT OF 1992This comprehensive legislation is far reaching andimpacts energy conservation, power generation, andalternative fuel vehicles as well as energy production.The federal as well as private sectors are impacted bythis comprehensive energy act. Highlights are describedbelow:Energy Efficiency ProvisionsBuildings• Requires states to establish minimum commercialbuilding energy codes and to consider minimumresidential codes based on current voluntary codes.Utilities• Requires states to consider new regulatory standardsthat would: require utilities to undertake integrated539Electric Vehicles• Establishes comprehensive program for the researchand development, infrastructure promotion, andvehicle demonstration for electric motor vehicles.Electricity• Removes obstacles to wholesale power competitionin the Public Utilities Holding Company Act by allowingboth utilities and non-utilities to form exemptwholesale generators without triggering the PUHCArestrictions.Global Climate Change• Directs the Energy Information Administration toestablish a baseline inventory of greenhouse gasemissions and establishes a program for the voluntaryreporting of those emissions. Directs the Dept.of Energy to prepare a report analyzing the strategies

540 ENERGY MANAGEMENT HANDBOOKfor mitigating global climate change and to develop aleast-cost energy strategy for reducing the generationof greenhouse gases.Research and Development• Directs the Dept. of Energy to undertake researchand development on a wide range of energy technologies,including: energy efficiency technologies,natural gas end-use products, renewable energyresources, heating and cooling products, and electricvehicles.20.2 STATE CODESThe Energy Policy Act of 1992 called for states toestablish minimum commercial building energy codesand to consider the same for residential codes. Prior tothis regulation, many states had some level of energyefficiency included in building codes (ASHRAE 90-80,CA Title 24, etc.), but most did not address the advancesin equipment, materials or designs that wouldimpact energy usage. A 1991 study by the Alliance toSave Energy found that most states employed codesthat were very outdated, which may have initiated thatportion of EPACT-1992.The development of efficiency standards normallyis undertaken by a consortium of interested partiesin order to assure that the performance level is economicallyattainable. The groups for building efficiencystandards are made up of building designers, equipmentsuppliers, construction professionals, efficiencyexperts, and others. There are several trade groups andresearch institutions that have developed standardsas well as some states that developed their own. Theapproved standards are merely words on paper untila state or local agency adopts these standards into aparticular building code. Once this occurs, officials(state or local) have the authority to inspect and assurethat the applicable codes are enforced during designand construction.The main organization responsible for developingbuilding systems and equipment standards, at leastin the commercial sector is the American Society ofHeating, Refrigeration, and Air-conditioning Engineers( ASHRAE).More than three quarters of the states have adoptedASHRAE Standard 90-80 as a basis for theirenergy efficiency standard for new building design.The ASHRAE Standard 90-80 is essentially “prescriptive”in nature. For example, the energy engineer usingthis standard would compute the average conductivevalue for the building walls and compare it against thevalue in the standard. If the computed value is abovethe recommendation, the amount of glass or buildingconstruction materials would need to be changed tomeet the standard.Most states have initiated “ Model Energy Codes”for efficiency standards in lighting and HVAC. Probablyone of the most comprehensive building efficiencystandards is California Title 24. Title 24 establishedlighting and HVAC efficiency standards for new construction,alterations and additions of commercial andnon-commercial buildings.ASHRAE Standard 90-80 has been updated into twonew standards:ASHRAE 90.1-1999 Energy Efficient Design of NewBuildings Except New Low-Rise ResidentialBuildingsASHRAE 90.2-1993 Energy Efficient Design of NewLow Rise Residential BuildingThe purposes of ASHRAE Standard 90.1-1999 are:(a)(b)(c)set minimum requirement for the energy efficientdesign of new buildings so that they may be constructed,operated, and maintained in a mannerthat minimizes the use of energy without constrainingthe building function nor the comfortor productivity of the occupants.provide criteria for energy efficient design andmethods for determining compliance with thesecriteria.provide sound guidance for energy efficient design.In addition to recognizing advances in the performanceof various components and equipment, theStandard encourages innovative energy conservingdesigns. This has been accomplished by allowingthe building designer to take into consideration thedynamics that exist between the many componentsof a building through use of the System PerformanceMethod or the Building Energy Cost Budget Methodcompliance paths. The standard, which is cosponsoredby the Illuminating Engineering Society of NorthAmerica, includes an extensive section on lighting efficiency,utilizing the Unit Power Allowance Method.The standard also addresses the design of the followingbuilding systems:

CODES, STANDARDS & LEGISLATION 541• Electrical power,• Auxiliary systems including elevators and retailrefrigeration,• Building envelope,• HVAC systems,• HVAC equipment,• Service water heating and equipment, and• Energy Management.ASHRAE has placed 90.1 and 90.2 under continuousmaintenance procedures by a Standing StandardProject Committee, which allows corrections and interpretationsto be adopted through addenda.20.3 MODEL ENERGY CODEIn 1994, the nation’s model code organizations,Council of American Building Officials (CABO), BuildingOfficials and Code Administrators International(BOCA), International Conference of Building Officials(ICBO), and Southern Building Codes CongressInternational (SBCCI), created the International CodeCouncil (ICC). The purpose of the new coalition wasto develop a single set of comprehensive buildingcodes for new residential and commercial buildings,and additions to such buildings. The 2000 InternationalEnergy Conservation Code (IECC) was published inFebruary of 2000 along with ten other codes, collectivelycreating the 2000 Family of International codes.These codes are the successor to the 1998 IECC and the1995 Model Energy Code (MEC) as well as all of theprevious MECs.The IECC establishes minimum design and constructionparameters for energy-efficient buildingsthrough the use of prescriptive and performance basedprovisions. The 2000 IECC has been refined and simplifiedin response to the needs of the numerous users ofthe model energy code. It establishes minimum thermalperformance requirements for building ceilings, walls,floors/foundations, and windows, and sets minimumefficiencies for lighting, mechanical and power systemsin buildings. Currently EPACT-1992 references MEC 95as the recommended building efficiency code. The Departmentof Energy is considering certifying the 2000IECC as the most cost-effective residential energy-efficiencystandard available. Once this determination isannounced, EPACT-1992 requires states to determinethe appropriateness of revising their residential energycodes to meet or exceed the 2000 IECC.The publication of the 2000 IECC offers states andlocal jurisdictions the opportunity to apply for financialand technical assistance offered by DOE’s BuildingStandards and Guidelines Program. If the standardsare codified by these entities, their code enforcementagencies will have opportunities to utilize the supportinfrastructure already established by the nationalmodel code organizations. More information can beobtained at the building Codes Assistance Projects website: www. crest. org/efficiency/.20.4 FEDERAL ENERGYEFFICIENCY REQUIREMENTSThe federal sector is a very large consumer ofenergy in the United States. There are actually over500,000 federal buildings with a combined energy costof $10 billion per year. Managers and operators of theseinstallations (mostly Department of Defense and PostalService) have very little incentive to conserve energy orimprove efficiency. Any work that is accomplished towardthese goals would have normally been kept in thecoffers and consumed by other functions as unencumberedfunds. The OPEC oil embargo brought into focusthe impact of energy costs and the US dependence onforeign sources of energy upon our economy. In 1975,the Energy Policy and Conservation Act directed thefederal government to develop mandatory standardsfor agency procurement policies with respect to energyefficiency; and, develop and implement a 10-year planfor energy conservation in federal buildings, includingmandatory lighting, thermal and insulation standards.This act was formalized with the Energy Conservationand Production Act in 1977, which established a 10%savings goal by 1985 over a 1975 baseline. The NationalEnergy Conservation Policy Act of 1978 further definedthe federal energy initiative with the following stipulations:• Establishes the use of Life-cycle-cost (LCC) methodof project analysis,• Establishes publication of Energy PerformanceTargets,• Requires LCC audits and retrofits of federal buildingsby 1990,• Establishes Federal Photovoltaic Program,• Buildings exceeding 1000 square feet are subjectto energy audits, and• Establishes a Federal Solar Program.

542 ENERGY MANAGEMENT HANDBOOKIn 1988 the Federal Energy Management ImplementationAct (FEMIA 1988) amended the FederalEnergy Initiative by removing the requirements toperform the LCC audits by 1990 and extended thedeadline of 10 percent savings goals to 1995. FEMIA alsoallowed the Secretary of Energy to set the discount rateused in LCC analysis and directed the various federalagencies to establish incentive for energy conservation.The National Defense Authorization Acts for FY 89, 90,and 91 added the following provisions:• Establishes incentive for shared energy savingscontracts in DOD, allowing half of first year savingsto be used for welfare, morale, and recreationactivities at the facility. The other half to be used foradditional conservation measures.• Expands DOD’s shared energy savings incentive toinclude half of first 5 years of savings.• Requires the Secretary of Defense to develop planfor maximizing Cost effective energy savings,develop simplified contracting method for sharedenergy savings, and report annually to congress onprogress.• Expands DOD incentives to participate in utilityrebate programs and to retain two-thirds of fundssaved.The President has power to invoke their ownstandards, in the form of Executive Orders, underwhich, agencies of the federal government must adhere.Presidents Bush and Clinton have both furtherincreased and extended the efficiency improvementsrequired to be undertaken by the federal sector. OnJune 3, 1999, President Clinton signed the order titled,“Greening the Government Through Efficient EnergyManagement.” The order required federal agencies toachieve by 2010:• 35% greater energy efficiency in buildings relativeto 1985 levels, and• 30% cut in greenhouse gas emissions from building-relatedenergy use relative to 1990.The order also directs agencies to maximize the use ofenergy savings performance contracts and utility contracts,in which private companies make energy improvements onfederal facilities at their own expense and receive a portionof the resulting savings. Life cycle cost analysis must beused so agencies see the long term savings from energyinvestments rather than merely the low bidder selectioncriteria. The order requires that everything from light bulbsto boilers be energy efficient as well as the use of renewableenergy technologies and sources such as solar, wind,geothermal and biomass. This order also mandated that theDOE, DOD and GSA provide relevant training or trainingmaterials for those programs that they make available to allfederal agencies relating to energy management strategiescontained in this order. A complete text of E.O. 13123 canbe found on the FEMP Web site www.greeninginterior.doi.gov/13123.html.20.5 INDOOR AIR QUALITY (IAQ) STANDARDS 1Indoor Air Quality (IAQ) is an emerging issue ofconcern to building managers, operators, and designers.Recent research has shown that indoor air is oftenless clean than outdoor air and federal legislation hasbeen proposed to establish programs to deal with thisissue on a national level. This, like the asbestos issue,will have an impact on building design and operations.Americans today spend long hours inside buildings,and building operators, managers and designers mustbe aware of potential IAQ problems and how they canbe avoided.IAQ problems, sometimes termed “ Sick BuildingSyndrome,” have become an acknowledged health andcomfort problem. Buildings are characterized as sickwhen occupants complain of acute symptoms such asheadache, eye, nose and throat irritation, dizziness,nausea, sensitivity to odors and difficulty in concentrating.The complaints may become more clinicallydefined so that an occupant may develop an actualbuilding-related illness that is believed to be related toIAQ problems.The most effective means to deal with an IAQproblem is to remove or minimize the pollutant source,when feasible. If not, dilution and filtration may be effective.Dilution (increased ventilation) is to admit moreoutside air to the building, ASHRAE’s 1981 standardrecommended 5 CFM/person outside air in an officeenvironment. The new ASHRAE ventilation standard,62-1989, now requires 20 CFM/person for offices if theprescriptive approach is used. Incidentally, it was theenergy cost of treating outside air that led to the 1981standard. The superseded 1973 standard recommended15-25 CFM/person.1 Source: Indoor Air Quality: Problems & Cures, M. Black & W. Robertson,Presented at 13th World Energy Engineering Congress.

CODES, STANDARDS & LEGISLATION 543Increased ventilation will have an impact onbuilding energy consumption However, this cost neednot be severe. If an airside economizer cycle is employedand the HVAC system is controlled to respondto IAQ loads as well as thermal loads, 20 CFM/personneed not be adhered to and the economizer hours willhelp attain air quality goals with energy savings at thesame time.In the fall of 1999 ASHRAE Standard 62-1999was issued, “Ventilation for Acceptable Indoor AirQuality.” The new standard contained the entire 1989version, which remains unchanged, along with fournew addenda. The reference in the 89 standard thatthe ventilation levels could accommodate a moderateamount of smoking was removed, due to troubles withsecondhand tobacco smoke. The new standard also removedreference to thermal comfort, which is coveredby other ASHRAE Standards. Attempts were made toclarify the confusion concerning how carbon dioxidecan be used to determine air contamination. A statementwas also added to assure that designers understandthat merely following the prescribed ventilationrates does not ensure acceptable indoor air quality. TheStandard was added to the continuous review process,which will mandate firms keep up with the perpetualchanges, corrections and clarifications. There are manyissues that are still under review as addendums to thestandard. The types of buildings that are covered werelimited to commercial and institutional, and the methodsof calculation of the occupancy levels have beenclarified. ASHRAE offers a subscription service thatupdates all addendum and interpretations. One of themain issues that should be considered during designof HVAC systems is that the outdoor air ventilation isrequired to be delivered cfm, which may be impactedwith new variable volume air handling systems.Energy savings can be realized by the use ofimproved filtration in lieu of the prescriptive 20 CFM/person approach. Improved filtration can occur at theair handler, in the supply and return ductwork, or inthe spaces via self-contained units. Improved filtrationcan include enhancements such as ionization devices toneutralize airborne biological matter and to electricallycharge fine particles, causing them to agglomerate andbe more easily filtered.The Occupational Safety and Health Administration(OSHA) announced a proposed rule on March 25,1994 that would regulate indoor air quality (IAQ) inworkplaces across the nation. The proposed rule addressesall indoor contaminants but a significant stepwould ban all smoking in the workplace or restrictit to specially designed lounges exhausted directlyto the outside. The smoking rule would apply to allworkplaces while the IAQ provisions would impact“non-industrial” indoor facilities.There is growing consensus that the most promisingway to achieve good indoor air quality is throughcontaminant source control. Source control is more costeffective than trying to remove a contaminant once ithas disseminated into the environment. Source controloptions include chemical substitution or product reformulating,product substitution, product encapsulation,banning some substances or implementing material emissionstandards. Source control methods except emissionstandards are incorporated in the proposed rule.20.6 REGULATIONS & STANDARDSIMPACTING CFCsFor years, chlorofluorocarbons (CFCs) were usedin air-conditioning and refrigeration systems. However,because CFCs are implicated in the depletion of theearth’s ozone layer, regulations required the completephaseout of the production of new CFCs by the turn ofthe century. Many companies, like DuPont, developedalternative refrigerants to replace CFCs. The need foralternatives will become even greater as regulatorycutbacks cause continuing CFC shortages.Air-conditioning and refrigeration systems designedto operate with CFCs will need to be retrofitted(where possible) to operate with alternative refrigerantsso that these systems can remain in use for theirintended service life.DuPont and other companies are commercializingtheir series of alternatives—hydrochlorofluorocarbon(HCFC) and hydrofluorocarbon (HFC) compounds. SeeTable 20.1.The Montreal Protocol which is being implementedby the United Nations Environment Program(UNEP) is a worldwide approach to the phaseoutof CFCs. A major revision to the Montreal Protocolwas implemented at the 1992 meeting in Copenhagenwhich accelerated the phaseout schedule.The reader is advised to carefully consider both the“alternate” refrigerants entering the market place andthe alternate technologies available. Alternate refrigerantscome in the form of HCFCs and HFCs. HFCs havethe attractive attribute of having no impact on the ozonelayer (and correspondingly are not named in the CleanAir Act). Alternative technologies include absorptionand ammonia refrigeration (established technologiessince the early 1900’s), as well as desiccant cooling.Taxes on CFCs originally took effect January 1,

544 ENERGY MANAGEMENT HANDBOOKTable 20.1 Candidate Alternatives for CFCs in Existing Cooling Systems———————————————————————————————————————————————————CFC Alternative Potential Retrofit ApplicationsCFC-11 HCFC-123 Water and brine chillers; process coolingCFC-12 HFC-134a or Auto air conditioning; medium temperature commercial foodTernarydisplay and transportation equipment; refrigerators/freezers;Blendsdehumidifiers; ice makers; water fountainsCFC-114 HCFC-124 Water and brine chillersR-502 HFC-125 Low-temperature commercial food equipment———————————————————————————————————————————————————1990. The Energy Policy Act of 1992 revised and furtherincreased the excise tax effective January 1, 1993.Another factor to consider in ASHRAE Guidelines3-1990—Reducing Emission of Fully Halogenated Chlorofluorocarbon(CFC) Refrigerants in Refrigeration andAir-Conditioning Equipment and Applications:The purpose of this guideline is to recommendpractices and procedures that will reduceinadvertent release of fully halogenated chlorofluorocarbon(CFC) refrigerants during manufacture,installation, testing, operation, maintenance, anddisposal of refrigeration and air-conditioningequipment and systems.The guideline is divided into 13 sections. Highlightsare as follows:The Design Section deals with air-conditioningand refrigeration systems and components andidentifies possible sources of loss of refrigerants toatmosphere. Another section outlines refrigerant recoveryreuse and disposal. The Alternative Refrigerantsection discusses replacing R11, R12, R113, R114, R115and azeotropic mixtures R500 and R502 with HCFCssuch as R22.20.7 REGULATORY AND LEGISLATIVEISSUES IMPACTING AIR QUALITY20.7.1 Clean Air Act AmendmentOn November 15, 1990, the new Clean Air Act(CAA) was signed by President Bush. The legislationincludes a section entitled Stratospheric OzoneProtection (Title VI). This section contains extraordinarilycomprehensive regulations for the productionand use of CFCs, halons, carbon tetrachloride, methylchloroform, and HCFC and HFC substitutes. Theseregulations will be phased in over the next 40 years,and they will impact every industry that currently usesCFCs.The seriousness of the ozone depletion is suchthat as new findings are obtained, there is tremendouspolitical and scientific pressure placed on CFC end-usersto phase out use of CFCs. This has resulted in theU.S., under the signature of President Bush in February1992, to have accelerated the phaseout of CFCs.20.7.2 Kyoto ProtocolThe United States ratified the United Nations’Framework Convention on Climate Change, whichis also known as the Climate Change Convention, onDecember, 4, 1992. The treaty is the first binding internationallegal instrument to deal directly with climatechange. The goal is to stabilize green house gases inthe atmosphere that would prevent human impacton global climate change, The nations that signed thetreaty come together to make decisions at meetings callConferences of the Parties. The 38 parties are groupedinto two groups, developed industrialized nations(Annex I countries) and developing countries (Annex11). The Kyoto Protocol, an international agreementreached in Kyoto in 1997 by the third Conference ofthe Parties (COP-3), aims to lower emissions from twogroups of three greenhouse gases: Carbon dioxide,methane, and nitrous oxide and the second group ofhydrofluorocarbon (HFCs), sulfur hexafluoride andperfluorocarbons. Emissions are meant to be reducedand limited to levels found in 1990 or 1995, dependingupon the gases considered. The requirements willimpact future clean air amendments, particularly forpoint sources. These requirements will further impactthe implementation of distributed generation sources,which are discussed in the following section.

CODES, STANDARDS & LEGISLATION 54520.8 REGULATORY AND LEGISLATIVE ISSUESIMPACTING COGENERATION &INDEPENDENT POWER PRODUCTION 2Federal, state and local regulations must be addressedwhen considering any cogeneration project.This section will provide an overview of the federalregulations that have the most significant impact oncogeneration facilities.20.8.1 Federal Power ActThe Federal Power Act asserts the federalgovernment’s policy toward competition and anticompetitiveactivities in the electric power industry.It identifies the Federal Energy Regulatory Commission(FERC) as the agency with primary jurisdictionto prevent undesirable anti-competitive behavior withrespect to electric power generation. Also, it providescogenerators and small power producers with a judicialmeans to overcome obstacles put in place byelectric utilities.20.8.2 Public Utility Regulatory Policies Act (PURPA)This legislation was part of the 1978 NationalEnergy Act and has had perhaps the most significanteffect on the development of cogeneration and otherforms of alternative energy production in the pastdecade. Certain provisions of PURPA also apply tothe exchange of electric power between utilities andcogenerators.PURPA provides a number of benefits to thosecogenerators who can become Qualifying Facilities(QFs) under the act. Specifically, PURPA• Requires utilities to purchase the power made availableby cogenerators at reasonable buy-back rates.These rates are typically based on the utilities’ cost.• Guarantees the cogenerator or small power producerinterconnection with the electric grid and the availabilityof backup service from the utility.• Dictates that supplemental power requirements ofcogenerator must be provided at a reasonable cost.• Exempts cogenerators and small power producersfrom federal and state utility regulations and associatedreporting requirements of these bodies.In order to assure a facility the benefits ofPURPA, a cogenerator must become a Qualifying2 Source: Georgia Cogeneration Handbook, published by the Governor’sOffice of Energy Resources.Facility. To achieve Qualifying Status, a cogeneratormust generate electricity and useful thermal energyfrom a single fuel source. In addition, a cogenerationfacility must be less than 50% owned by an electricutility or an electric utility holding company. Finally,the plant must meet the minimum annual operatingefficiency standard established by FERC when usingoil or natural gas as the principal fuel source. Thestandard is that the useful electric power output plusone half of the useful thermal output of the facilitymust be no less than 42.5% of the total oil or naturalgas energy input. The minimum efficiency standardincreases to 45% if the useful thermal energy is lessthan 15% of the total energy output of the plant.20.8.3 Natural Gas Policy Act (NGPA)The major objective of this legislation was tocreate a deregulated national market for natural gas.It provides for incremental pricing of higher costnatural gas supplies to industrial customers who usegas, and it allows the cost of natural gas to fluctuatewith the cost of fuel oil. Cogenerators classified asQualifying Facilities under PURPA are exempt fromthe incremental pricing schedule established for industrialcustomers.20.8.4 Resource Conservation andRecovery Act of 1976 (RCRA)This act requires that disposal of non-hazardoussolid waste be handled in a sanitary landfill insteadof an open dump. It affects only cogenerators withbiomass and coal-fired plants. This legislation hashad little, if any, impact on oil and natural gas cogenerationprojects.20.8.5 Public Utility Holding Company Act of 1935The Public Utility Holding Company Act of1935 (the 35 Act) authorizes the Securities and ExchangeCommission (SEC) to regulate certain utility“holding companies” and their subsidiaries in a widerange of corporate transactions.The Energy Policy Act of 1992 creates a newclass of wholesale-only electric generators—“ exemptwholesale generators” (EWGs)—which are exemptfrom the Public Utility Holding Company Act (PUH-CA). The Act dramatically enhances competition inU.S. wholesale electric generation markets, includingbroader participation by subsidiaries of electric utilitiesand holding companies. It also opens up foreignmarkets by exempting companies from PUHCA withrespect to retail sales as well as wholesale sales.

546 ENERGY MANAGEMENT HANDBOOK20.8.6 Moving towards a deregulatedelectric power marketplaceEPACT-1992 set into motion a widespreadmovement for utilities to become more competitive.Retail wheeling proposals were set into motion instates such as California, Wisconsin, Michigan, NewMexico, Illinois and New Jersey. There are many issuesinvolved in a deregulated power marketplaceand public service commission rulings and litigationwill certainly play a major role in the power marketplaceof the future. Deregulation has already broughtabout several important developments:• Utilities will need to become more competitive.Downsizing and minimization of costs includingelimination of rebates are the current trend. Thistranslates into lower costs for consumers. Forexample Southern California Edison announcedthat the system average price will be reducedfrom 10.7 cents/kWh to lower than 10 cents bythe year 2000. This would be a 25% reductionafter adjusting for inflation.• Utilities will merge to gain a bigger marketshare. For example, Wisconsin Electric PowerCompany merged with Northern States Power;this merger of two utilities resulted in a savingsof $2 billion over 10 years.• Utilities are forming new companies to broadentheir services. Energy service companies, financialloan programs and purchasing of related companiesare all part of the new utility strategy.• In 1995 one hundred power marketing companieshave submitted applicants to FERC. Powermarketing companies will play a key role inbrokering power between end users and utilitiesin different states and in purchasing of newpower generation facilities.• Utilities will need to restructure to take advantageof deregulation. Generation Companiesmay be split away from other operating divisionssuch as transmission and distribution.Vertical disintegration will be part of the newutility structure.• Utilities will weigh the cost of repowering andupgrading existing plants against purchasingpower from a third party.Chapter 24 discusses many more issues on thetopic of electrical deregulation.20.9 OPPORTUNITIES IN THE SPOT MARKET 3Basics of the Spot MarketA whole new method of contracting has emergedin the natural gas industry through the spot market.The market has developed because the Natural GasPolicy Act of 1978 (NGPA) guaranteed some rightsfor end-users and marketers in the purchasing andtransporting of natural gas. It also put natural gassupplies into a more competitive position with deregulationof several categories.The Federal Energy Regulatory Commission(FERC) provided additional rulings that facilitatedthe growth of the spot market. These rulings includedprovisions for the Special Marketing Programsin 1983 (Order 2346) and Order 436 in 1985, whichencouraged the natural gas pipelines to transport gasfor end-users through blanket certificates.The change in the structure of markets in thenatural gas industry has been immense in terms ofboth volumes and the participants in the market. Byyear-end 1986, almost 40% of the interstate gas supplywas being transported on a carriage basis. Notonly were end-users participating in contract carriage,but local distribution companies (LDCs) wereaccounting for about one half of the spot volumes oninterstate pipelines.The “spot market” or “direct purchase” marketrefers to the purchase of gas supplies directly from theproducer by a marketer, end-user or LDC. (The term“spot gas” is often used synonymously with “best effortsgas,” “interruptible gas,” “direct purchase gas”and “self-help gas.”) This type of arrangement cannotbe called new because the pipelines have always soldsome supplies directly to end-users.The new market differs from the past arrangementsin terms of the frequency in contracting andthe volumes involved in such contracts. Anothercharacteristic of the spot market is that contractsare short-term, usually only 30 days, and on aninterruptible basis. The interruptible nature of spotmarket supplies is an important key to understandingthe operation of the spot market and the costsof dealing in it. On both the production and transportationsides, all activities in transportation orpurchasing supplies are on a “best efforts” basis.This means that when a cold snap comes the directpurchaser may not get delivery on his contractsbecause of producer shutdowns, pipeline capacityand operational problems or a combination of theseproblems. The “best efforts” approach to dealingcan also lead to problems in transporting supplies

CODES, STANDARDS & LEGISLATION 547when demand is high and capacity limited.FERC’s Order No. 436The impetus for interstate pipeline carriage camewith FERC’s Order No. 436, later slightly changedand renumbered No. 500, which provided more flexibilityin pricing and transporting natural gas. Inpassing the 1986 ruling, FERC was attempting to getout of the day-to-day operations of the market andinto more generic rule making. More significantly,FERC was trying to get interstate pipelines out of themerchant business into the transportation business—astep requiring a major restructuring of contracting inthe gas industry.FERC has expressed an intent to create a morecompetitive market so that prices would signal adjustmentsin the markets. The belief is that directsales ties between producers and end-users willfacilitate market adjustments without regulatoryrequirements clouding the market. As more gas isderegulated, FERC reasoned that natural gas priceswill respond to the demand: Lower prices wouldassist in clearing excess supplies; then as marketstightened, prices would rise drawing further investmentinto supply development.FERC Order No. 636Order 636 required significant “Restructuring”in interstate pipeline services, starting in the fall of1993. The original Order 636:• Separates (unbundles) pipeline gas sales from transportation• Provides open access to pipeline storage• Allows for “no notice” transportation service• Requires access to upstream pipeline capacity• Uses bulletin boards to disseminate information• Provides for a “capacity release” program to temporarilysell firm transportation capacity• Pregrants a pipeline the right to abandon gas sales• Bases rates on straight fixed variable (SFV) design• Passes through 100% of transition costs in fixedmonthly charges to firm transport customersFERC Order No. 636AOrder 636A makes several relatively minorchanges in the original order and provides a greatdeal of written defense of the original order’s terms.The key changes are:• Concessions on transport and sales rates for apipeline’s traditional “small sales” customers (likemunicipalities).• The option to “release” (sell) firm capacity for lessthan one month—without posting it on a bulletin boardsystem or bidding.• Greater flexibility in designing special transportationrates (i.e., off-peak service) while still requiringoverall adherence to the straight fixed variable ratedesign.• Recovery of 10% of the transition costs from theinterruptible transportation customers (Part 284).Court action is still likely on the Order. Further,each pipeline will submit its own unique tariffto comply with the Order. As a result, additionalchanges and variations are likely to occur.20.10 THE CLIMATIC CHANGE ACTION PLANThe Climatic Change Action Plan was establishedApril 21, 1993 and includes the following:• Returns U.S. greenhouse gas emissions to 1990levels by the year 2000 with cost effective domesticactions.• Includes measures to reduce all significantgreenhouse gases, carbon dioxide, methane,nitrous oxide, hydrofluorocarbons and othergases.20.11 SUMMARYThe dynamic process of revisions to existingcodes plus the introduction of new legislation willimpact the energy industry and bring a dramaticchange. Energy conservation and creating new powergeneration supply options will be required to meetthe energy demands of the twenty-first century.


CHAPTER 21NATURAL GAS PURCHASINGCAROL FREEDENTHALJOFREEnergy ConsultingHouston, Texas21.1 PREFACEThis is the second full revision for this chapter,Natural Gas Purchasing. Chapter 21 was originally writtenwhen the book was published in 1993. Rewrite forthe first revision was a completely new effort done in1996. As of 2000, the industry has continued to changeand is still in the conversion from a federally regulated,price-controlled business to an economically dynamic,open industry, and this is a completely revised writing.Changes are continuing to shape the industry differently,especially when coupled with the changes coming fromthe potential decontrol of the electric power industry.To make even more changes, the impact of ECommercebusiness-to-business is beginning to play a role in thisindustry. When this revision was started, only one companyoffered the web for gas marketing. As of 2000, fiveadditional companies had launched ECommerce business-to-businessnatural gas trading.The old natural gas business is really a new business.Its structure goes back 150 years but it is more likea new industry. It has the typical growth and turmoil ofa new business. Energy products, especially natural gasand electricity, are new businesses as the country goesinto the new millennium. Newly “reformed” companies,new marketing organizations, new systems affecting gasmarketing, and even, a new industry structure makes itnecessary to start from scratch in writing the revision forthis chapter.Like the new millennium, the natural gas industryand equally as important, the total energy business isgoing through its own transition. Change will continueas companies and businesses try different strategies.ECommerce will play a major role in the industry’stransition. This phase of the transition is amorphous andmakes it difficult to predict the exact course of eventsfor the future. Things that appeared far-out years agoare becoming closer to reality. The newest buzzwords,“distributive electricity” includes the use of fuel cells andsmall dual cycle turbine driven generators by residential549and small commercial users. Both of these are becomingeconomically feasible. The impact on the gas and electricindustries is unknown. This is a time of change for thenew energy business. Marketing and supplying energyproducts like natural gas and electricity will go throughmany changes before optimum conditions are found.A few things are for sure. Natural gas is becomingthe major fuel for stationary power uses in the UnitedStates. Long dominated by oil products for this use, nowgas is becoming the leader. Coal continues as a majorfuel source for electric generation. Consumption of coalfor power generation has reached record levels in recentyears but environmental concerns and the required highcapital for new coal burning generating plants will reducecoal’s market share. The public’s dislike of nuclear powerand the high costs to build plants with the safety desiredmeans no growth for this industry. A new philosophywill have to be developed by society recognizing safetyand environmental benefits of atomic power before newnuclear facilities will be built.The natural gas industry, just like the power industry,which is going through its own decontrol activities,change will be a way of life always. Companies in theenergy field and in associated areas such as communications,financial, systems, etc. will continue to merge,acquire, spin off, and change their structure and goals.As the country goes into the new millennium, these areindustries in transition and will change along with thegrowth industries in cyberspace. A big difference fromthe old, staid and conservative electric and gas utilitiesof the prior century! Change and growth are the way.Regardless of this, one factor continues to dominate.The profit motive is still the driving force of theindustry today. It will not change but will continueinto the future. Economics will govern change andbe the basis for decision making. All the transformations—buyingand selling of companies, new marketingcompanies, new systems for handling the merged assets,etc. will all be subject to one metric; is it profitable? Already,some acquisitions made by large electric and gascompanies to bring together various parts of the energyindustry have come apart because the final economicsdid not pass muster.The purpose of this chapter is to give the fuel buyer,for any operations or industry, the knowledge and infor-

550 ENERGY MANAGEMENT HANDBOOKmation needed to buy natural gas for fuel. The buyer maybe in a large petrochemical plant where natural gas is amajor raw material or may be the commercial user havinghundreds of apartments needing gas for heat and hotwater or plant operator where the gas is used for processsteam. It might be a first time experience or an on-goingjob for the buyer. This chapter will give the backgroundand information to find natural gas supplies for any needat the lowest cost and highest service and security. Thechapter will include information on history of the industry,sources of supply, transportation, distribution, storage,contracts, regulatory, and financial considerationsneeded to buy natural gas.21.2 INTRODUCTIONNatural gas is predominately the compound “methane,”CH 4 . It has the chemical structure of one carbonatom and four hydrogen atoms. It is the simplest of thecarbon based chemicals and has been a fuel for industry,for illumination, and for heating and some cooling ofhomes, offices, schools, and factories. Natural gas is also,a major fuel for generating electricity. In addition to fueluses, gas is a major feedstock for the chemical industryin making such products and their derivatives as ammoniaand methanol. Natural gas is used in refining andchemical plants as a source for hydrogen needed by theseprocessing businesses. Through the reforming process,hydrogen is stripped from the methane leaving carbondioxide, which has its usefulness in chemical manufacturingor use, by itself in cooling, carbonated drinks, orcrude oil recovery.The term “natural gas industry” includes the people,equipment, and systems starting in the fields wherethe wells are located and the natural gas is produced. Itincludes other field tasks as gathering, treating, and processing.Transportation to storage or to interstate or intrastatepipelines for further transportation to the marketarea storage, or to the distribution system for delivery tothe consumer and the burner tip, are part of the system.The burner tip might be in a boiler, hot water heater, combustionengine, or a chemical reactor to name a few of themany uses for natural gas.Natural gas is produced in the field by drilling intothe earth’s crust anywhere from a couple of thousandfeet to five miles in depth. Once the gas is found and thewell completed to bring the gas to the earth’s surface, it istreated if necessary to remove acid impurities and again,if necessary, processed to take out liquid hydrocarbons oflonger carbon chains than methane, which has a singlecarbon atom. After processing, the gas is transported inpipelines to consuming areas where distribution companieshandle the delivery to the specific consumer.In addition to the people and companies directlyinvolved in the production, transportation, storage, andmarketing of natural gas, there are countless other businessesand people involved in assisting the gas industryto complete its tasks. There are systems companies, regulatoryand legal professionals, financial houses, banks,and a host of other businesses assisting the natural gasindustry. Figure 21.1 shows the many parts of the industryas it is known today. The money flowing through themajor sections of the industry are shown in Figure 21.2.The $85 billion industry shown in the diagram only representsthe functions in getting natural gas, the commodity,to market and consumption. Not included in the overallindustrial revenues are the moneys generated by the salesand resale of gas before its consumption, the processingand marketing of natural gas liquids coming from thegas, and the financial markets where gas futures andother financial instruments are sold and traded. These arebig businesses also. Estimates are that the physical gasis traded three to four times before consumption. In thefinancial markets, gas volumes 10 to 12 times the amountof gas consumed on an average day are traded daily.Figure 21.3 should be of most interest to the naturalgas buyer as it depicts the various sales points, stages,and handling the gas goes through in getting from thewellhead to the burner tip—from the wellhead to theconsumer. As one can see in the diagram, there are manyalternate paths the gas can travel before coming to itsend use as a fuel or feedstock for chemical manufacturing.Each one of the stages on the flowsheet representsan added value point in the travel to consumption. Rawgas coming from the wellhead many times has sufficientquality to go directly into a transporting pipeline for deliveryto the consuming area. Sometimes the gas needsfield treating and/or processing to meet pipeline specificationsfor acceptance into the pipeline.The gas industry is the oldest utility except forwater and sanitation. In the middle of the 19th century,many large cities used a synthetic gas made from passingsteam over coal to light downtown areas and providecentral heating systems. Big cities like Baltimore, NewYork, Boston, and many more cities and municipalitiesused gas for illumination. Many utilities from that periodexist today and are still gas and electric suppliers to theareas they serve.In the early days of the gas business, there was nonatural gas, as known today. Instead, these utilities produceda synthetic gas for both the illumination and thecentral heating systems. The synthetic gas, sometimescalled “water gas” because of the method of producing it,

NATURAL GAS PURCHASING 551Figure 21.1 Natural Gas Industry Flowsheet.Figure 21.2 Gas Industry Money Flow for Business Activities.

552 ENERGY MANAGEMENT HANDBOOKFigure 21.3 Wellhead to Consumer Flowsheethad bad attributes—it contained a high content of hydrogenand carbon monoxide, two bad actors for a gas usedin homes, businesses, and factories. People died whenexposed to it because of the carbon monoxide, and buildingsblew-up because of the hydrogen when free gas fromleaks or pipe ruptures was ignited. When natural gascame on the scene in the early 1900s, where it was available,it quickly replaced the old manufactured gas. Aboutthe same time, advances were made in electricity so thatcities and municipalities changed to electricity for lightingand illumination. Natural gas quickly lost its marketfor municipal lighting.Natural gas was originally an unwanted by-productfrom the oil fields. Problem was getting rid of it. Flaringwas used, but this was a waste of good natural resources.Around the beginning of the century, associated gas fromOhio oil fields was shipped to Cleveland in wooden pipesto replace the then used synthetic gas. In the early days ofthe industry, the limitations to greater uses of natural gaswere that gas was produced in only certain parts of thecountry and transportation was available for only veryshort distances. Market penetration was thwarted by theability to ship it. There were no long distance pipelines inthe early days of the industry. Natural gas made a greatreplacement for the synthetic counterpart—methane isessentially safe as far as toxicity and is much safer asfar as explosion. Gas’ growth was dependent on buildinglong distance pipelines. Not until the 1930s did theindustry have the capability of making strong enough,large steel piping needed for the long-distance pipelines.Completion of major interstate pipelines to carry gas fromproducing regions to consumers was the highlight of the1930s to the start of World War II in the early 1940s.Pipeline construction came to a halt and was dormantuntil the war’s end. Construction went full forceafter the war to insure delivering the most economicaland easiest fuel to America’s homes, commercial facilitiesand industrial players. Even today with the start of thenew millennium, some areas of the U.S. still do not havea fully developed natural gas distribution and deliverysystem. Areas in the West where population is sparse,parts of the Northeast where oil prices were too competitiveto delivered gas prices, and other parts of the countrylacking distribution systems for the same reasons are stillwithout natural gas. Many of these use what is called“bottled gas,” a mixture of propane and butane or propaneonly for home heating and other critical uses. Justrecently, new supplies and pipelines were developed tobring natural gas to the Northeast U.S. from Canada. Additionaldistribution systems will bring more gas to morecustomers through the country from the tip of Florida tothe North Central and West Northern states.Ever since natural gas became available for fuel, itwas under some form of government economic control.Through the tariff mechanism for pricing natural gas,the government had the power to make gas prices moreor less attractive to competing fuels. Further, with thegovernment controlling wellhead prices and slow tomake changes in prices as conditions changed, it becamedifficult and economically undesirable to expand naturalgas production. Government price controls hampered thegrowth of the U.S. natural gas business. The gas shortagesof the mid-1970s are an example of government controlstifling expansion and growth. There was no shortage ofgas reserves, only a shortage of incentives for producersto develop and supply the gas. The free market builds itsown controls to foster competition and growth.Congress passed the Natural Gas Policy Act of 1978to change the policy of government economic control. Afew years of transition were needed before significantchanges began in the industry. Real impact started in1985. Even today, the industry is still in transition. Thefederal decontrol changed interstate marketing andmovement of natural gas. Gas at the local levels wherethe state Public Utility Commission or similar local governmenthas control, is still heavily regulated. Decontrolat the federal level is slowly filtering down to local agencies.As of 2002, some states began moving to “opentransportation” rules. A current obstacle to the swifterimplementation of rules at the state and local levels is thetie of gas and electricity as utilities within state regulatorycontrol. With the electric industry going through its own“decontrol,” many wanted to see the much larger electricindustry work out the utility problems. Then gas couldfollow with less negotiating and discussion. The electrictimetable is now years behind its planned evolution andthis has slowed gas local control further.

NATURAL GAS PURCHASING 553With the price of gas changing each year, the totalindustry value changes. The industry in nominal annualterms is roughly a $100 Billion business. Electricityis around $230 Billion. Many electric companies that wereboth gas and electricity utilities even before deregulation,have bought major natural gas pipelines or gas distributors.Large electric companies bought into the natural gasindustry whether they purchased transporters, distributors,or marketing companies. Interestingly, in a relativelyfew years, some of these combinations have come apartbecause of poor profitability.Electric and gas utility companies have gone aftertransportation and marketing companies. Surprisingly,none of the expanding companies have sought to buy, atthe beginning of the gas business, the oil and gas explorationand production companies (E&P companies). Theseare the companies looking for natural gas and then producingit. While all of the transporting companies, whetherlong distance or distribution in nature and, further, whetherelectric and/or natural gas in business, have shied awayfrom the production companies, other E&P companieshave merged or acquired smaller operations to add to thetotal capability of the company. The significant changesduring the 1990's saw major E&P companies acquire evenmajor and independent E&P assets.21.3 NATURAL GAS AS A FUELWhy has natural gas grown in popularity? Whatmakes it a fuel of choice in so many industries as the newmillennium begins? What shortcomings does it have?Figure 21.4 shows the change in basic fuels mix used inthe U.S. in 1985 and 1999. Nuclear, which started in 1960,enjoyed a period of rapid growth. The high costs for all thesafety engineered into the plants had made it an uneconomicalsystem towards the end of the century. There areno nuclear plants scheduled for construction. Even someof those completed and running and some with the initialconstruction still in progress were shut down or convertedinto natural gas fired units. The only change that will beseen in nuclear generation of electricity is plant efficiencieswill be improved for the units continuing to operate.Coal usage in the U.S. has grown in recent yearswith record coal production in the late 1990s. Coal isby far the major fuel used for electric generation, commandinga 56% market share. It has many negativeproperties like the need for railroads for transportation,high pollution from the burner after-products, and poorhandling characteristics including being dirty, losses onstorage, and the difficulties of moving a solid material,including the disposal of the remaining ash. Still, coalhas a number of things going for it which will keep coalin use for many years to come. The ready availabilityand abundance are major merits. The stability of coalprices will always give coal a place in the market. Figure21.5 shows the comparison in prices among coal, naturalgas, and oil products for the period 1985 through 1999.Coal at about a dollar per million British units (MMBtu)is not only much cheaper per unit of energy, but also hasthe advantages of availability and abundance. Coal willslowly lose position because of its disadvantages of pollutionand higher costs to meet changing standards andhigh capital costs for building new generating plants.Petroleum products have lost market share in thelater years because of their costs and the dependenceof the U.S. on foreign suppliers for crude and crudeoil products. Oil products used for electric generationinclude distillate fuel oil, a relatively lightweight oil,which during the refining process can have most of thesulfur removed during that process. Low sulfur fuelsare desirable to keep emissions low for environmentalreasons. The other major oil product used is residualfuel oil, the bottom of the barrel from the refining process.This is a heavy, hard to transport fuel with manyundesirable ingredients that become environmentalproblems after combustion. Many states have put costlytariffs on using residual fuel oil because of its environmentalharm when used.Figure 21.4 U.S. Basic Fuels 1985 & 1999 (Quadrillion Btu)

554 ENERGY MANAGEMENT HANDBOOKNatural gas is the nation’s second largest source offuel and a major source of feedstock for chemical production.Plentiful supplies at economically satisfactoryprices, a well developed delivery system of pipelinesto bring gas from the wellhead to the consumer, and itsenvironmental attractiveness has made natural gas thechoice of fuel for many applications. Going into the newmillennium, natural gas will be a popular fuel. As a fuelfor industry for heating and generating electricity and asa feedstock for chemicals, natural gas is very attractive.For residential and commercial applications, the securityof supply and efficiency in supplying makes it an idealfuel. Even though natural gas is a fossil fuel, it has thelowest ratio of combustion-produced carbon dioxide toenergy released. Carbon dioxide is the biggest culprit inthe concern for global warming.Natural gas consumption data are followed in fourmajor areas by the Federal Energy Information Agency inaddition to its listing the data for natural gas used in thefields for lease and plant fuel and as fuel for natural gaspipelines; residential, commercial, industrial, and electricgeneration. Natural gas demand has always, in moderntimes led the amount of gas produced except for the mid-1970s when the country experienced a severe natural gassupply shortage. In those years, while there were morethan sufficient reserves in the ground to meet demand,the control of gas pricing by the federal governmentstymied the initiative of producers to meet demand. Potentialsupply was available but the lack of profit incentiveprevented meeting demand in those years. Demandincreased because of changes and shortages in crudeoil supplies. Early 1970s were the start of the change incrude pricing and the country was faced with decreasedsupplies from foreign producers. Crude prices doubledalmost overnight, but because natural gas was price controlledand could not meet the rising prices, supplies inthe interstate market suffered.The major market for natural gas is the industrialsector. Residential is next and commercial and electricgenerating take about the same amount. Figure 21.6graphically depicts the share each market took for 1999.The residential market is basically for home heating andhot water fuel. The commercial market is for space heating.Use of natural gas in industrial plants when used forspace heating is included in this category. The industrialcategory covers all other uses of natural gas in industryand includes gas used by industrial locations for powergeneration until earlier 2000. All power generation isnow included in the category of electric generation. Themajor demand factor in all categories is weather. Residentialand commercial consumption are most affectedby weather since these two categories reflect space heating.Electric generation is weather sensitive also since thesummer electric load is responsive to the air conditioningload needed for the hot weather. Even though the industrialload is not as sensitive to weather as are other categories,it does reflect the additional heating load neededfor the process industries when temperatures fall and rawmaterials including process air and/or water are muchcolder.Natural gas has tremendous potential to gain evengreater use in the generation of electricity in several ways.First, it could be the choice fuel to replace aging nuclearplants that will not be re-certified as they age. FurtherFigure 21.5 Fuel Prices for Generating Electricity 1995-1999

NATURAL GAS PURCHASING 555NOTE: INDUSTRIAL INCLUDES INDEPENDENT POWER PRODUCTIONJOFREE HOUSTON, TX 77002 CF 20 JUN 2000Figure 21.6 Natural Gas Markets 1999as coal plants age and need replacement or need to bereplaced because of environmental causes, natural gas isthe ideal fuel. It is easier to get to the plant and handle inthe plant, the environmental needs are much smaller, andthe capital required for the generating plant and facilitiesis much lower. Natural gas is the fuel of choice among thefuels currently available.Even if the electric systems in effect now were tochange to more “distributive” in nature, such as fuel cellsor small, dual cycle gas turbines, natural gas would be theideal fuel. Some planners see fuel cells or turbines beingused by residential units so that each household couldhave its own source of electricity. When houses neededadditional power, they would draw it from the utilitylines. When the fuel cell produces more than needed, theutility would take the excess. Most fuel cell work todayinvolves hydrogen and oxygen as the combined fuels foroperation. Natural gas could be the source of hydrogen.Since many homes already have natural gas piped to thehouse, it would be easy to handle this new fuel to makeelectric power locally. In addition, distributive powergeneration could use small, gas turbines for power supply.Commercial users would be possible users of thesesystems also.21.3.1 SupplyNatural gas is a product coming from the earth. Asdiscussed previously, the major component of naturalgas is the chemical compound methane, CH4. Methaneis the product formed when organic matter like treesand foliage decays without sufficient oxygen available tocompletely transform the carbon in these materials to carbondioxide. Theory is natural gas deep in the ground is aproduct of decaying material from past millions of yearsof the earth’s history. Chemical elements available as thematter decayed gives the methane such contaminants ashydrogen sulfide, carbon dioxide, nitrogen, and manymore compounds and elements. Natural gas comes fromshallow depths as little as a few thousand feet into theearth and as deep as 20 to 25 thousand feet. Natural gaswells are drilled on dry land and on water covered land.Current drilling in the Gulf of Mexico deep waters is inwater depths up to around 3,000 feet.Natural gas quantities are measured in two sets ofunits. The volume of the gas at standard conditions is onemeasure. Basically, at standard conditions of temperatureand pressure, the amount of natural gas in a volume of acubic foot is a standard measure. Since a cubic foot is arelatively small volume when talking of natural gas, theusual term is a thousand cubic feet (Mcf). Still as a volumemeasurement, the next largest unit would be a millioncubic feet (MMcf) which is a thousand, thousand cubicfeet. A billion cubic feet is expressed as Bcf and a trillionis Tcf.Since natural gas is not a pure compound but a mixtureof many products formed from the decaying organicmatter, the energy content of each cubic foot at standardconditions is another method of measuring natural gasquantities. The energy units used in the U.S. are Britishthermal units (Btu), the amount of heat needed to raise apound of water one degree Fahrenheit at standard conditionsof pressure and at 60 degrees Fahrenheit. A typicalcubic foot of gas, if of pure methane, would have about1000 Btu per cubic foot (Btu/cf). Gas coming from wellscan range from very low heat contents (200 to 300 Btu/cf)because of non-combustible contaminants like oxygen,carbon dioxide, nitrogen, water, etc. to energy contents of1500 to 1800 Btu/cf. The additional heat comes from liquidhydrocarbons of higher carbon contents entrained inthe gas. The higher carbon content molecules are knownas “natural gas liquids” (NGLs). Also, other combustiblegases like hydrogen sulfide contained in natural gas canraise the heat content of the gas produced.Data from the Federal Energy Information Agency(EIA) show an “average” cubic foot of gas produced inthe U.S. as dry natural gas in recent years would havehad an average of 1,028 Btu/cf. Gas is treated and/orprocessed to remove the contaminants lowering or raisingthe Btu quantity per cubic foot to meet pipeline specificationsfor handling and shipping and the gas. Pipelinequality natural gas is 950 to 1150 Btu per cubic foot.A frequently used term to describe the energy contentof natural gas when sold at the local distributionlevel, such as residential, commercial or small industrialusers, is the “therm.” A therm is equivalent to 100,000Btu. Ten therms would make a “dekatherm” (Dt) andwould be equivalent to a million Btu (MMBtu). The thermmakes it easier when discussing smaller quantities ofnatural gas.When exploration and production companies searchfor gas in the ground, they refer to the quantities located

556 ENERGY MANAGEMENT HANDBOOKas reserves. This is a measure of the gas the companiesexpect to be able to produce from the fields where the gaswas found. Through various exploration methods—basicgeophysical studies of the ground and surroundingareas to the final steps of development wells are usedfor more accurately pin-pointing reserve volumes. Reservesare the inventory these companies hold and fromwhich gas is produced to fill market needs. In 1997, theU.S. government’s Department of Energy showed U.S.natural gas reserves in the order of magnitude of 170 trillioncubic feet (Tcf) of economically recoverable reserves.Without any replacement, this would be a 5- to 7-year lifeof existing reserves at current consumption rates. U.S.exploration and production companies are continuouslylooking for new reserves to replace the gas taken fromthe ground for current consumption. From 1994 to 1997,producers found reserves equal or more in volume to gasproduced during that year. The reserve volumes are fromareas where gas is already being produced and representa very secure number for the amount of gas thought tobe in the ground and economically feasible to produce.These are called recoverable reserves based on producedand flowing gas.The next level of measuring reserves is gas heldbehind these recoverable producing reserves. A littleless secure and a little more speculative but, still a goodchance of producing as designated. Using this category,just for the U.S., there are enough gas reserves for 25 to35 years depending on the amount consumed each year.There are abundant gas reserves in North America toassure a steady supply for the near term and future. Inaddition to the U.S. reserves, gas in Mexico and Canadaare considered a part of the U.S. supply or the total NorthAmerican supply. Mexico contributes very little to the USsupply at this time because its gas production and transportationsystems are limited. As gas demand and pricesincrease, Mexico could play an important role as an U.S.supplier. As already noted, considerable amounts of gascome from Canada.In addition to these two levels of gas reserves,there are additional categories “possible” or potential ofreserves. These become more speculative but are still animportant potential supply for the future. Some of thesemay become more important sooner than expected. Agood example is the gas supplies coming from coal seamsources. Considerable gas is produced in New Mexicofrom these sources which were not expected to be suchlarge suppliers until much later in time. Additionalpotential supplies but with long lead times for furtherdevelopment is gas from hydrates and gas from sourcesdeeper in the Gulf Coast.Natural gas produced from wells where crude oilis the major product is termed “associated gas.” Roughly40% of the gas produced in the U.S. comes from associatedwells while the rest comes from wells drilled specificallyfor natural gas. Only differences between the gas producedfrom the two types of wells are the associated wellsgas might contain greater amounts of what has been mentionedpreviously as “natural gas liquids” (NGL). Theseliquids are organic compounds with a higher number ofcarbon elements in each of the molecules making up thatcompound and are entrained in the gas as minute liquiddroplets. Methane, which is the predominant compoundin natural gas, has one carbon and four hydrogens in themolecule. The two-carbon molecule is called ethane, thethree-carbon molecule is propane, four-carbon moleculeis butane, and the fifth, is pentane. All molecules withmore than five carbons are collected with the pentanesand the product is called “pentane plus.” It is also knownas “natural gasoline” which must be further refined beforeit can be used as motor fuel. The NGL are removedby physical means either through absorption in an organicsolvent or through cryogenically cooling the gas streamso that the liquids can be separated from the methane andeach other.There are markets for the individual NGL products.The ethane is used by the chemical industry for makingplastics. Propane is also used in the chemical industrybut finds a significant market as fuel. Butanes go to thechemical and fuels market and the pentanes plus arebasically feedstock for the motor fuels production fromrefineries. The overall NGL market is about a $10 to 15Billion a year business depending on the product prices.Prices for NGL vary as the demand varies for each of thespecific products and bear little relationship to the pricepaid for the natural gas. When gas prices are high andNGL prices are low, profitability on the NGL is very poor.At the times, when the profitability is poor, the ethanewill be re-injected back into the natural gas stream andsold with the gas to boost the heat content of the gas.A second difference between associated and gaswell gas is strictly of a regulatory nature. Gas from associatedwells is produced with no quantity regulations sothat the maximum amount of crude oil can be producedfrom the well. Gas from “gas only” wells depending onthe state where produced, may be subject to productionrestrictions because of market, conservation, or otherconditions. Major natural gas producing areas in the U.S.are Texas, Louisiana, Oklahoma, and New Mexico. Thesestates, including the offshore areas along the Gulf Coaststretching from Alabama to the southern tip of Texas, accountfor over 80% of the gas produced in the country.Figure 21.7, North American Gas Producing Areas, showsthe gas producing states in the United States and the im-

NATURAL GAS PURCHASING 557port locations for Canadian gas and for LNG receivingterminals. Other states with significant gas productionare California, Wyoming, Colorado, New York, Pennsylvania,Alabama, Mississippi, and Michigan. A totalof 18 states supply commercial quantities of natural gasaccording to the Federal EIA.A major supplier of gas for the U.S. is Canada.While imports do come from other countries, Canada byfar, is the major supplier to the lower 48 states. Naturalgas coming from Canada is transported by pipeline intothe U.S. The small amounts of gas coming from Mexicoalso travel by pipeline. Imports from other countries intothe U.S. are transported as liquefied natural gas (LNG).Here natural gas at the producing country is cooled andcompressed until it is liquid. The reduction in volumeis roughly 20 times the original volume. The liquefiedgas with its reduced volume is now economically sizedfor shipping. The liquefied gas is transported betweencountries in large vessels, which are essentially very largecryogenically insulated, floating containers. The LNG isreceived at terminals in the U.S. where it is re-vaporizedto gas. During this step, large quantities of refrigerationare available from the expanding liquid to gas. The cooling“energy” is sold and used in commercial applicationsto recoup some of the costs in making the gas into LNG.There are currently four terminals in the U.S. for receivingand handling LNG. These are in Boston, Lake Charles,LA, Baltimore, MD, and off the coast of Georgia at ElbaIsland. The Baltimore and Georgia locations were shutdown years ago when natural gas prices would not justifyLNG sales. Current plans are to reopen both facilitiesshortly.Overall imports into the U.S. have grown considerablysince the mid-1980s when only 843 Bcf wereimported in 1985. Natural gas imports in 1999 increasedfor the 13th consecutive year to 3,548 Bcf, 16.0 percent ofU.S. PRODUCTIONIMPORTSTOTAL18,659 Bcf3,538 Bcf22,197 BcfFigure 21.7 North American Gas Producing Areas in 1999

558 ENERGY MANAGEMENT HANDBOOKtotal U.S. gas supply. Canada supplied 93.9 percent of thetotal imports in 1999. Of the total imports, only 4.5 percentwere received as LNG. Canada did much in the late1990s to expand the pipeline systems bringing gas to theU.S. Additional pipelines are scheduled for completionearly in the new millennium. Most Canadian productionis in the provinces of Alberta and British Columbia. Newproduction did come on from the Eastern Coast late inthe last century and was imported into the U.S. from theMaritime Provinces. Since 1985, Canadian imports havemore than quadrupled and Canada plays a major rolein the expected additional supply needed to meet thedemand for the years to come. Estimates are Canadiangas volumes will increase insuring the supply of gas forU.S. demand in future years. The Alliance Pipeline wascompleted in late 2000 and added 1.3 Bcf/day of supplyto the U.S. Already, Canadian gas makes up a significantportion of the gas going to the U.S. Northeast. Figure 21.7shows the major importing locations for gas coming intothe U.S. from Canada.While natural gas imports into the U.S. as LNG weresmall in comparison to the total gas imported in 1999, theamount coming in 1999 was roughly three times that receivedthe prior year. Equally important, the number ofcountries supplying LNG to the U.S. increased from threeto six. Algeria continued to be the major supplier with 75Bcf in 1999 but recently completed production facilitiesin Trinidad supplied 49 Bcf in the same year. Plans areto make all the terminals in the U.S. operative so that additionalLNG supplies can be expected. Locations of allterminals are shown in Figure 21.721.3.2 TransportationNatural gas in the United States is transported almostexclusively by pipeline. From the time the natural gasleaves the wellhead, whatever route it takes in getting tothe burner tip, it is through a pipe! Short or long distance,regardless, natural gas is transported in pipe. Only exceptionsare the few times compressed natural gas is transportedby truck for short distances. And, in some locationswhere gas is liquefied (LNG) for storage for use duringpeak demand times, the LNG is moved by truck also.Movement of gas through these two means is insignificantin the overall picture of transporting natural gas.When talking of transporting natural gas throughpipelines, there are three main groups of pipelines to beconsidered:Gathering System: These are the pipelines in thefield for collecting the gas from the individual wells andbringing it to either a central point for pick up by thelong-haul pipeline or to a central treating and/or processingfacility.Long-haul transportation: This is the pipeline pickingup the gas at the gathering point, or if a highly productivewell near a pipeline, from the well itself and movingthe gas to a city-gate for delivery to the distributioncompany or to a sales point for a large user where the gasis delivered directly to the consumer. The long-haul pipelinecan be either an interstate pipeline that crosses fromone state into another or an intrastate pipeline wherethe transportation is only within the state where the gaswas produced. The interstate pipelines are economicallycontrolled by the Federal Energy Regulatory Commission(FERC). The operating regulations fall under theDepartment of Transportation (DOT). The EnvironmentalRegulatory Agency has jurisdiction regardless of thetype of pipeline in regard to environmental matters. Theinterstate pipelines are still economically regulated bythe Federal Energy Regulatory Commission (FERC) sincethese are utilities engaged in interstate commerce.Intrastate pipelines are economically regulated bystate agencies. Utilities are granted a license to operate incertain areas and are allowed to make a rate of return ontheir invested capital. This is different from the non-regulatedbusinesses where they compete to make profits fromthe operations. As utilities, the rates for transportation areset through regulatory procedures. The pipeline makes arate case for presentation to the FERC for authorizationto charge the rates shown in the case. The pipeline is allowedto recover all of its costs for transporting the gasand make a return on the invested capital of the pipeline.Natural gas pipelines offer essentially two basic typesof rates for transporting natural gas: firm and interruptible.With firm transportation, the transportation buyer isguaranteed a certain volume capacity daily for the gas itwants transported. The buyer is obligated to pay a portionof the transportation charge regardless whether itsuses the volume or not on a daily basis. This is called a“demand charge” and is a part of the transportation tariff.The second part of the tariff is the commodity charge andis a variable charge depending on how much gas is transportedby the pipeline.Pipelines also offer an “interruptible” tariff wherespace is on a “first come-first served” basis. Interruptibletransportation carries no guarantee to the party buyingthe transportation that space in the pipeline will be availablewhen needed. The tariff here is usually very close tothe commodity rate under the firm transportation.The methodology of the ratemaking procedure usedto recover the pipeline’s costs and rate of return is suchthat when a pipeline sells all of its firm transportation, itwill make its allowed rate of return. A pipeline can legallyexceed its accepted rate of return based on its handlingof the firm and interruptible transportation. Typically,

NATURAL GAS PURCHASING 559the pipeline has about 80% of its volume contracted infirm transportation. When a firm transporter does notuse its full capacity, the pipeline can mitigate the costs tothat pipeline by selling its firm transportation to anothertransporter as interruptible transportation.Many of the transportation contracts for firm transportationare terminating in the 2000 period. With thechanges in the marketing system and the shift in themerchant role, some pipelines may have difficulty in fillingtheir firm transportation sufficiently, This may bringsome reduction in transportation costs which the gasbuyer may be able to exploit. Further, the gas buyer attimes can use what is called “back hauling” to get a lowerrate for gas transportation. An example of this mightbe gas coming from Canada through the North CentralU.S. area such as Chicago. A buyer for this gas might belocated in the Southwest, say in Texas. Rather than shipgas from Chicago to Texas and have to pay the full tariff,a shipper might exchange gas in Texas for the gas tocome from Chicago to Texas. In turn, the gas coming fromCanada would be sold in the Chicago area as “Texas” gas.Here the shipper would pay the much lower fee for the“paper transportation” of the gas volumes. This would bea back haul arrangement.The interstate pipeline community is relativelysmall. Many of the pipelines have merged or were acquiredby other utilities since the regulatory changesin the industry took the merchant function from themand made them strictly transporters. There are 25 majorinterstate pipelines moving gas from the production areasof the country to the consumer. These are owned orcontrolled by only 13 companies. Table 21.1, U.S. InterstateNatural Gas Pipelines, lists the major U.S. interstatepipelines, and the parent company having ownership. Inall likelihood, even more mergers and acquisitions willoccur to bring the number of separate companies evenlower.Intrastate pipeline companies are within the statewhere the gas is produced. Many of these have miles ofpipeline comparable to the interstate systems but, do notcross state lines. Within the state, these pipelines serve thesame mission as the interstate pipelines; bringing the gasfrom the field whether the well or gathering point to thecity gate for distribution by the local distributor or directlyto a large consumer. Some of the larger ones for the gasproducing states are listed in Table 21.2. While the pipelinesthemselves are no longer sellers of natural gas, thebuyer should review the pipelines’ systems to see if thereis a close connection possible so a direct supply might bemade from the pipeline to the consumer. In cases where apipeline is close to a plant or other large user, a marketeror the buyer itself can make arrangements for the shorthaulpipeline to bring gas from the transporting pipelineto the facility.Pipeline transportation might include more thanone pipeline to complete the shipment from well toburner tip. Who pays for the transportation at each stepis open to negotiation between the gas supplier and thebuyer. Usually, the producers are responsible for the gatheringand field costs of getting the gas to the transportationpipeline’s inlet, which may be on the pipeline or at aterminal point, sometimes designated as a “hub.” Manytimes when the transporting pipeline goes through aproducing field, the producer will only be responsible forgathering charges to get the gas from the wellhead to thefield’s central point for discharge into the pipeline’s inlet.The gathering and field charges along with the transportationto the transporting pipeline inlet is what makes thedifference between wellhead gas prices and “into pipe”gas prices.Table 21.1 U.S. Major Interstate Natural Gas Pipelines———————————————————————————————PARENTPIPELINEPIPELINE COMPANY HEADQUARTERS———————————————————————————————Panhandle Pipeline CMS Energy Houston, TXTrunkline PipelineHouston, TXANR Pipeline Coastal Corp. Houston, TXDetroit, MICIG PipelineColorado Springs, COColumbia Gas Tran’n Columbia Energy Co. Charleston, WVColumbia Gulf Trans’nHouston, TXCNG Pipeline Dominican Energy Pittsburgh, PAAlgonquin Gas Trans’n Duke Energy Boston, MATexas Eastern PipelineHouston, TXEl Paso Pipeline El Paso Energy Houston, TXSonat GasHouston, TXTennessee PipelineHouston, TXFlorida Gas (50%) Enron Corporation Houston, TXNorthern Natural GasOmaha, NETranswestern PipelineHouston, TXNGPL Kinder Morgan Houston, TXGateway United Koch Industries Houston, TXWiliston Basin MDU Resources Bismarck, NDNational Fuel Gas National Fuel Gas Buffalo, NYNorthern Border Northern Border Omaha, NEPCT Pacific Gas & Electric San Francisco, CAQuestar Pipeline Questar Energy Salt Lake City, UTMississippi River Reliant Industries St. Louis, MONoram PipelineHouston, TXNorthwest Pipeline Williams Companies Salt Lake City, UTTexas Gas PipelineOwensboro, KYTransco PipelineHouston, TXWilliams Gas PipelineTulsa, OK———————————————————————————————JOFREEHOUSTON, TX CF 20JUN2000

560 ENERGY MANAGEMENT HANDBOOKWho pays for the transportation charges from thetransporting pipeline’s pick-up to the city gate or distributioncompany’s inlet, even if it includes more than onetransporting pipeline, is negotiable between the seller ormarketing company and the buyer. The marketing companyselling the gas might quote a delivered price to thebuyer, especially, if the seller is holding transportationrights with the pipeline handling the transportation. Ifthe buyer has transportation rights, he might take the gasFOB (Free on Board, the point where title transfers andwhere transportation charges to that point are includedin the sales price) at the transportation pipeline’s inlet.These are all part of the marketing and negotiating inmoving gas from the field to the city gate and/or the consumer.What are typical prices for transporting natural gasfrom producing area to consumers in various parts of thecountry where there is no intrastate gas? The buyer canget detailed information from the pipeline tariffs whichcan be gotten from the FERC and other sources like tradeletters and magazines.Pipeline rates or tariffs are set by the regulatoryagencies involved. There is some negotiation possible.Still, the gas in different locations will have a value basedon market conditions regardless of transportation rates.This is called "basis differential." Some typical basis differentialsbetween hubs and major markets are shown inFigure 21.8. These were developed from published pricesgiven in trade publications for a several month period toget representative values.For natural gas to be carried in transportationpipelines, it must meet certain conditions of quality andcomposition. This was previously referred to as "pipelinespecifications." These standards include the heating contentof the gas per unit volume; i.e. British thermal unitsper cubic feet (Btu/cf). Typically, pipeline quality gas willTable 21.2 U.S. Major Natural Gas Intrastate Pipelines—Summer 2000———————————————————————————————————————————————————STATE PIPELINE PARENT HEADQUARTERS———————————————————————————————————————————————————ALABAMA Southeast Alabama Gas Southeast Alabama Gas Andalusia, Al———————————————————————————————————————————————————CALIFORNIA Pacific Gas Trans’n Pacific Gas & Electric Co. San Francisco, CA——————————————————————————————————————————Southern California Gas Sempr Energy Los Angeles, CA———————————————————————————————————————————————————LOUISIANA Chandeleur Pipeline Co. Chandeleur Pipeline Co. Woodlands, TX——————————————————————————————————————————Louisiana Interstate Pln AEP Corp. Alexandria, LA——————————————————————————————————————————Mid Louisiana. Gas Co. Midcoast Energy Resources Houston, TX——————————————————————————————————————————NEW MEXICO Gas Company of New Mexico Public Service Co. of New Mexico Albuquerque, NM———————————————————————————————————————————————————OKLAHOMA Enogex, Inc. Enogex, Inc. Oklahoma City, OK——————————————————————————————————————————Oneoak Gas Tran’n Oneoak Inc. Tulsa, OK———————————————————————————————————————————————————TEXAS Aquilia Gas Pipeline Utilicorp Omaha, NE——————————————————————————————————————————Ferguson-Burleson County Gas Mitchell Energy & Dev’t Corp. Woodlands, TX——————————————————————————————————————————Houston Gas Pipeline Enron Energy Houston, TX——————————————————————————————————————————Lone Star Gas Pipeline Ensearch Dallas, TX——————————————————————————————————————————Midcon Texas Pipeline Midcon Texas Houston, TX——————————————————————————————————————————PG&E Texas Pipeline PG&E Houston, TX——————————————————————————————————————————Westar Transmission Kinder-Morgan Houston, TX——————————————————————————————————————————Winnie Pipeline Co. Mitchell Energy & Dev’t Corp. Woodlands, TX———————————————————————————————————————————————————JOFREEHOUSTON, TX CF 20JUN2000

NATURAL GAS PURCHASING 561be around 1,000 Btu/cf. Gas coming out of the well, canrange from very low values to over 1,500 to 1,600 Btu/cf.The lower values come from gas having contaminantslike carbon dioxide or nitrogen in the stream while thehigher values come from the gas containing entrainedliquid hydrocarbons or hydrogen sulfide. The contaminantsare removed in treating, for the hydrogen sulfideand other acid impurities, and processing facilities for theliquid hydrocarbons such as ethane, propane, etc. Typically,pipeline quality gas will run around 1,000 Btu/cfwith a range of from 950 to 1150 Btu/cf. The exact amountis measured in the stream as the gas is sold on a Btu basis.Typical other specifications for pipeline transmission ofnatural gas are given in Table 21.3.Distribution: Once the natural gas is moved fromthe producing area it can travel from a few miles to thousandsof miles in getting to its destination. The usual terminatingpoint for the gas is at a city gate where the localdistribution company (LDC) delivers it to the individualuser whether it is a commercial, residential, or industrialconsumer. In some cases where the consumer is a largeindustrial or an electric generating plant, the gas might goTable 21.3 Natural Gas Interstate Pipeline Specifications.—————————————————————————Contaminants may not exceed the following levels:—————————————————————————20 grains of elemental sulfur per 100 cubic feet1 grain of hydrogen sulfide per 100 cubic feet7 pounds of water per million cubic feet3 percent of carbon dioxide by volumeOther impurity (i.e. oxygen, nitrogen, dirt, gum,etc.) if their levels exceed amounts that the buyer must incurcosts to make the gas meet pipeline specifications.—————————————————————————Source: Handbook on Gas Contracts, Thomas G. Johnson, IEDPress, Inc. Oklahoma City, OK. 1982, page 63Figure 21.8 Typical Natural Gas Basis Differentials between Hubs and Major Market Points.

562 ENERGY MANAGEMENT HANDBOOKdirectly from the long haul transporter to the consumer.There are hundreds of distribution companies in thecountry. Some are investor owned utilities while manyare municipality owned and operated. Some are co-opsformed for distributing the gas.The trade association representing this group ofgas companies almost exclusively is the American GasAssociation, headquartered outside of Washington, DC.Information and data on the industry as a whole, and ondistribution companies can be obtained from this organization.Its address and web site are listed in Table 21.4.The local distribution company is usually regulatedby the state regulatory agency such as the Public ServiceCommission. It may also be under local regulation bythe city or municipality it serves. This group of naturalgas transporters is yet to be deregulated throughout thecountry. Some states, Georgia the most notable, havepassed new regulations much like the decontrol of thenational pipelines. In these locations, the transporter isstrictly a mover of gas and has no merchant function. Itmay have a subsidiary or affiliated company doing themerchant function or marketing of the gas. The eventualresult of deregulation at this level will be for local distributioncompanies to offer open access to their transportationfacilities. Each state will have to make its decisionas to whether the LDC is freed from the merchant role orretains it if only in part along with offering open transportationfor other merchants to move gas to the finalconsumer.The odorizing of natural gas so that its presencecan be detected easily since natural gas as such is anodorless gas, is usually done by the local distributioncompany before distributing the gas. The odorant is asulfur containing hydrocarbon with an obnoxious odorthat can be detected by human smell even when used invery small, minute quantities in the gas. While it is commonlythought all natural gas must be odorized when it issold to the user, this is not necessarily correct. Gas goingto industrial uses where the sulfur containing materialgiving the odor could be harmful to the process neednot be odorized. There are both federal and state regulationsgoverning the odorization. In buying natural gas,the buyer should insure the contract includes provisionfor adding the odorant and whose responsibility it is forproper addition and monitoring.21.3.3 EconomicsNatural gas prices were originally set by the federalregulatory agency having jurisdiction over natural gas.The original methodology for price setting was much likethe rate of return methodology for pipeline transportationtariffs. This was a direct function of the believed costsof finding, developing and producing natural gas. Asdiscussed previously, the low prices paid at the wellheadprevented the natural gas industry from maintaining thenecessary supply and caused the dire gas shortages of themid-1970s. After natural gas prices were decontrolled,and natural gas became a true commodity, prices are aTable 21.4 Federal Agencies & Natural Gas Trade AssociationsORGANIZATION INFORMATION & SERVICES WEBSITE————————————————————————————————————————————————————————FEDERAL & MAJOR STATE AGENCIES FOR NATURAL GAS & ENERGY REGULATION1 Department of Commerce Information on offshore production of gas and oil www.doc.gov.2 Department of Energy (DOE) Information on energy products; supply, demand,Energy Information Agency consumption, prices, www.eia.doe.gov3 Department of Transportation Regulates the safety of pipelines used in transporting www.dot.govnatural gas.4 Federal Energy Regulatory Regulates natural gas pipeline tariffs and facilities.Commissionwww.ferc.fed.us5 Louisiana Office of Regulatory board for Louisiana natural gas operations.Conservationwww.dnr. state.la.us6 New Mexico Public Regulation Regulatory board for New Mexico.Commissionwww.nmprc.state.nm.us7 Oklahoma Conservation Regulatory board for Oklahoma.Commissionwww. okcc. state. ok. us8 Texas Railroad Commission Regulatory board for Texas. www.rrc.state.tx.us(Continued)

NATURAL GAS PURCHASING 563ORGANIZATION INFORMATION & SERVICES WEBSITE————————————————————————————————————————————————————————NATURAL GAS & RELATED ENERGY TRADE ASSOCIATIONS1 American Gas Association Trade organization on natural gas; major source of(AGA) information on gas local distribution companies. www.aga.org2 American Petroleum Institute Represents the nations oil and gas industries(API)www.api.org3 Association of Energy Engineers Organization supplying information and services forenergy efficiency, energy services, deregulation,facilities, management, etc.www.aeecenter.org4 Canadian Energy Research Responsible for Canadian energyInstitute (CERI) research. www.ceri.ca5 Edison Electric Institute Represents electric industry; major areais investor -owned electric companies.www.eei.org6 Gas Industry Standards Board Industry forum for development of(GISB)natural gas measurement methods andstandards for gas transmission.www.gisb.org7 Gas Processors Association Trade association for natural gas processors, agroup of companies extracting gas liquids fromnatural gas streams and marketing the products.www.gasprocessors.com8 GasMart Annual marketing meeting for natural gas suppliers,transporters, customers, and marketers.www.gasmart.com9 Gas Research Institute Develops technical solutions for naturalgas and related energy markets.www.gri.org10 Interstate Natural Gas Voice of the interstate natural gas system includingAssociation of Americathe pipelines and companies supplyingnatural gas.www.ingaa.org11 National Energy Marketers National non-profit trade association representingAssociation all facets of the energy business. www.energymarketers.com12 Natural Gas Information & Sweb-site dedicated to natural gasEducation Resources education and history. www.naturalgas.org13 Natural Gas Supply Represents independent and integratedAssociation (NGSA) producers and marketers of natural gas. www.ngsa.org14 Southern Gas Association Links people, ideas, and information for transmission,(SFA )distribution, and marketing of natural gas to allcustomers served by member companies.www.southerngas.org—————————————————————————————————————————————————————————JOFREE HOUSTON, TX 77002 CF 21JUN2000

564 ENERGY MANAGEMENT HANDBOOKreflection of the normal economic factors impacting commoditypricing.The price for natural gas at the burner tip is dependenton many things—market conditions, supply/demandbalances, economic conditions, and many more includingthe activity of natural gas financial markets, pricesfor competing fuels, etc. In the early stages of the industry,because natural gas was considered a burdensome byproduct of the crude oil industry, it was sold for very lowprices. When crude oil was around $2/barrel (B) or about30 cents per million British thermal units (MMBtu), naturalgas under federal price control sold for a penny or two perthousand cubic feet or roughly the same per MMBtu. Inactual heating value, a thousand cubic feet of natural gashas close to a million Btu. A barrel of fuel oil is 42 gallonsof oil and about six million Btu depending on the grade offuel oil. On an economic basis of energy content, naturalgas prices for a thousand cubic feet compared to a 42 gallonbarrel of oil, should be close to one-sixth the value ofthe oil, i.e. an $18 barrel of oil would be equivalent to $3/Mcf or $3/MMBtu natural gas. Very seldom has the price,even after decontrol, reached this ratio. Instead, the valueof gas has run about half or about one-twelfth or one-tenththe value of the oil product. In 2000, oil prices (West TexasIntermediate, WTI) were around $35/B. At the same timenatural gas prices were over $5.00/MMBtu. Gas priceswere about one-sixth the value of oil in dollars per barrel,very close to the energy equivalent of competing fuel, residualfuel oil. For the first time gas at the wellhead was atparity with crude oil in $/Btu.The fear in 2000 was gas supplies would not meetdemand. Gas demand is increasing as more and morepower plants are being completed that will use naturalgas as fuel. This will do much to balance the gas demandbetween summer and winter. In summer the gas will goas fuel for electric generation to meet the hot weather requirementsfor power for air-conditioning, and in winterit will go as fuel for heating. Many buyers fear the highsummer demand will prevent storage filling believednecessary for the winter heating season.Pricing is not a logical phenomenon. Data and basicconsiderations can help in predicting prices but the finalprice is very dependent on perception—market perceptionat the time. Too many of the variables are unknownprecisely enough for pricing to be a scientific conclusion.Forecasting prices is art. Perception of the value based onsupply/demand parameters sets the price. The marketitself will do a lot to raise or lower the price. Further, thelarge financial market compared to the physical marketfor natural gas has an immense impact on the prevailingprice. Gas prices can “spike” for many reasons—real orperceived. Hurricanes, hot weather spells, changes in theeconomy, etc. can make prices go up or down quickly andsignificantly. Short-term changes are always a possibility.Seasonality at times has little bearing on the currentprice. Natural gas prices have dropped precipitously inthe middle of January and reached highs for the year in“shoulder months.” Eventually, prices come back to realitybut in the time they are moving large dollar gains orlosses can occur.In looking at gas prices, it is necessary to knowwhere the gas is sold as prices vary according to wherethe sale is made in the wellhead to consumer path. Unlikecrude oil, which is transformed into various commercialproducts, each with its own value, natural gas is essentiallythe same once it enters the transportation portion ofthe journey to the burner tip. Its value does increase as itmoves through the system going from the wellhead to theconsumer because of the added value of the transportationand services bringing the gas to market.The simplest place for pricing natural gas is gas soldat the wellhead. Gas priced on a Btu value at the wellheadwill accurately reflect the value of the gas further downthe chain even though wellhead gas might need to begathered, treated, and/or processed. Once the gas is pipelinequality, its price reflects where in the transportationline from sales point to ending sales point it is at the time.Anywhere in the chain, the wellhead price can be determinedby net backing the price to the wellhead by subtractingthe additional costs to get to the point of pricing.The value of the gas, since it is a commodity, is open tosupply/demand forces on the pricing. In addition, sinceat times pipeline capacity for moving the gas is susceptibleto supply/demand restraints for capacity, the pricedifferentials based on location can be affected also. Thisis how the basis differential of gas prices between differentlocations occurs. This is reflected in Figure 21.8. Gaspurchased at the wellhead is done so on a wellhead price.Gas purchased further downstream might be termed“into pipe price” or “hub price” if coming from a centralpoint where gas supplies come together for distributionto pipelines for long-haul transportation.The New York Mercantile in making a futures marketfor natural gas, has its main contract, at the HenryHub, in Louisiana because of the hub’s central localityand easy accessibility. The difference between prices formajor hubs or selling locations is termed the “basis” pricingand can vary much depending on the current supply/demand factors.The first of the typical major pricing points for naturalgas would be the wellhead or field price. This mightinclude actual sales at the wellhead, at a central locationafter the gas is gathered in the field, or at the tailgate of atreating and/or processing plant depending on the plant

NATURAL GAS PURCHASING 565location in relation to the transporting pipeline.The next major pricing point would be the “intopipe” price where the gas goes into a pipeline for transportationto the marketing area or to a “hub” for furtherredirection and transportation to the consuming area. Thehub has the ability to dispatch the gas coming in on onepipeline to another in the variety of pipelines coming intoand leaving the hub.From the hub pricing, gas would then be priced atthe city gate where it is transferred to a local distributioncompany for delivery to the consumer. The pricing forthe consumer would be based on the “sales point” price,which would be a total price for the gas including all thetransportation and services required to get the gas to theuser's receipt point.The individual price paid by the buyer is dependenton many factors starting from the wellhead pricing to theprice at the meter coming onto the buyer’s property. Ingeneralities, the government and other reporting servicesreport the prices at the major pricing points and at theconsumers’ location. The major consuming sectors whereprices are reported are the residential, commercial, industrial,and electric generating markets. Since the progressionfrom each of the stages from production to market carriesa cost factor, it is important to know where in the deliverychain the price quoted applies. Figure 21.9 is a comparisonof prices at each of these major market points for 1998.21.3.4 EnvironmentalEnvironmentally, natural gas is the preferred fuel.Even though it is a fossil fuel, the amount of carbon dioxidereleased is the lowest per unit of energy received ofthe major fossil fuels. Natural gas is ideal for its handlingand transportation qualities. Its environmental advantagesmakes it the most popular fuel and fuel of choice formany applications. It presents no unique environmentalconcerns to the user and as long as the supply is pipelinequality, the fuel source is of no concern in regard to environmentalpurposes.21.3.5 Regulatory ChangesTo the average gas buyer, the new natural gas industrypresents few regulatory problems or concerns otherthan those imposed by local or state authorities. The federalregulations from prior years on natural gas have beenreduced. While natural gas pricing is no longer underfederal regulations, it is still tied to some of the originalfederal natural gas laws. In today’s markets, these areessentially of no interference to commerce. It does meanthat under certain extreme conditions, federal regulationscould again be imposed on natural gas and certain usescould be restricted.For the current conditions, the buyer mainly has tobe concerned with local and state rules and regulations.Transportation, storage and handling regulations areagain local and state but here, federal agencies do play arole. The Department of Transportation and the EnvironmentalProtection Agency have jurisdiction in the areasof pipeline safety and environment, respectively. Buyersshould insure in their negotiations and contracts withsellers, transporters, and providers that all regulationsare covered and the responsibility for meeting these rulesare a part of the transaction. The contracts for buying andtransporting should speak directly to whose responsibilitymeeting the requirements will fall and which partieswill be responsible for the consequences if failure occurs.POWERINDUSTRIALCOMMERCIALRESIDENTIALCITYGATEWELLHEAD0 5 10 15 20Source: U.S. EIA & Natural Gas WEEKEXCLUDES ALASKA & HAWAII$MMBtuFigure 21.9 Natural Gas Prices by Sales Points for 1998

566 ENERGY MANAGEMENT HANDBOOKAgencies having responsibility for natural gas regulationsat the federal and state levels are easily accessible.Table 21.4 lists the major federal agencies including theweb sites. State Public Utility Commissions (PUC) caneasily be located if information is necessary. Further,many law firms and consultants specialize in the regulatoryaspects and should be contacted if necessary.21.4 BUYING NATURAL GASTo buy natural gas for either small or large operations,a thorough knowledge of the structure of the naturalgas marketing system is essential. Again, this is thebig change from the days when the industry was underprice regulation by the federal government. Gas sales toconsumers were through only one route—producers topipeline transporter and merchant to local distributorto consumer. In the beginning of the transition to openmarketing, this was referred to as "system gas." In localitieswhere LDC are still the merchants, this is still thecase. In a very small number of occasions, the chain wasshortened to producer to pipeline to major consumer.Now—even with states in general still having controlover the local distribution, the chain can be as short asproducers to consumer or more generally, producers tomarketers to consumers for relatively large users andproducers to marketers to distributors for residential andmost commercial and small industrial applications. Thisis the free market for natural gas. Buyers are free to pickany marketer or seller to supply gas. Open transportationis available to everyone—at least it should be!21.4.1 Physical and Financial MarketsSince natural gas is a commodity—it is fungible—and its supply is at times at the mercy of many factorsincluding weather, demand, economics, etc., there is amarket for buying gas supplies in the future. Commonly,this is called the “futures market” as opposed to thephysical market where the actual commodity goes to thebuyer either for resale or consumption. Many users ofnatural gas buy or “hedge” on the commodity market totake advantage of prices offered in the future. The NewYork Commodity Exchange (NYMEX) offers contracts forup to 36 months and several banks and operators do anover the counter market offering prices even further out.The consumers or sellers (producers, marketers, users,etc.) using the futures market are usually hedging as ameans of price risk protection.As an example, a fertilizer manufacturer is a largeuser of natural gas for making ammonia and derivativesfor use in fertilizers and industrial chemicals. If it takesthe ammonia manufacturer an average of 60 to 90 daysfrom the time he buys the raw material natural gas to beready to sell it as ammonia, he has to worry about theprice of both the ammonia and the price of the replacementnatural gas changing during the period. If he uses$2 per MMBtu gas for the ammonia and then after sellingthe ammonia has to buy $3 per MMBtu gas to makenew ammonia, he could be at a price disadvantage inthe ammonia market. To “hedge” against these kinds ofprice changes, the manufacturer can buy “futures” whenhe starts making ammonia with the $2 raw material. Hecan protect his future-buying price for the raw material,which represent 70-80 percent of the manufactured costof the ammonia, by hedging his future purchases.Since the prices on the futures market move constantly,almost daily for the near term market and lessas time goes out, the futures market makes an ideal mediumof wagering what the price will be in the future. The“speculators” who come into the market have no need forthe commodity nor will they most likely ever take actualphysical ownership of it. Their purpose is strictly to wageron where the price will be on a certain date. It can be eitherup or down from the price on the day they buy “futures.”This is not a small market but one in billions of dollars. In1999, it was estimated that for every billion cubic feet ofgas consumed in an average day, 10 to 12 billion cubic feetwere traded on the NYMEX exchange and other markets.Of course, some of this 12-fold excess of consumption wentto hedging, but roughly speculators traded 90 percent. Theaverage amount of gas consumed per day in 1999 withoutregard to seasonality was about 60 billion cubic feet. Usingthe wellhead price of around $3.00/MMBtu, about $180million was traded each average day for the consumptionof gas. In the financial trading markets, almost two billiondollars per day were traded!Other than to have mentioned the financial marketand show its significance in the natural gas industry, thischapter is devoted to physical gas buying. The buyersand sellers both need to know about the financial marketsand evaluate their own need to participate or not in thistype of gas transactions. There are many marketing companies,financial houses, and consultants well versed inthe financial markets and how trading in these can lowerthe over all purchase costs of the commodity. Many booksare written on this aspect. Buyers and sellers should becomefamiliar with all sources of information in this areain helping to either maximize the return for the productfor the sellers or minimize the purchase costs for the consumers.The comments on buying gas for use does notnegate the financial market but, leaves it to other sourcesfor the users to learn how to work within the financialframework including its benefits and risks. Knowledge

NATURAL GAS PURCHASING 567of the financial markets are necessary because of the impactthe financial market has on the physical market andprices for natural gas.21.4.2 Actually Buying the GasSo—how does the gas user get down to the basicsof buying natural gas? Do they call the local distributor,if the consuming facility is in an area served by the localdistributor, or does the buyer shop around for the bestprice and service? Again, information and knowledgeare the secret to success. The buyer must know what isneeded to determine what path to follow in buying naturalgas. If the buyer is looking for a source of gas for anew operations, one never before using natural gas as thefuel, then they must estimate the necessary parameters toknow how much is needed to fill the requirements of thatoperation. If the buyer is replacing an expiring contract orhaving to change vendors, then they have the historicalrecord to help in knowing what is needed to renew thesupply sources. They can use the existing informationand records to predict with greater accuracy what volumeof gas will be needed, the changes on a daily or othertime basis that will be needed and what were the priorcosts for the gas supply. With this information, the fuelbuyer can look for new sources to meet the needs moreefficiently and cheaply.The very first question to be answered is how muchnatural gas will be consumed on a daily basis and whatwill be the range of use on a daily, weekly, monthly andannual basis. The information could even be a question ofan hourly rate as to how large a swing does the user anticipate.These are the big questions to answer in making thefirst step in trying to select a supplier or seller. Knowingthe quantity and conditions of where the rate will varyare crucial to starting the buying process. Whether theconsumer is a large or small user of gas will play a majorrole in what selections are open to it for purchasing gas.The physical conditions prevailing in the area of the locationusing the gas will play a role because of regulationsof the area and the actual physical availability of pipes fortransporting the gas to the consumer.Typically, the break from a small user to a largeone is a rate of about one thousand cubic feet per dayor in energy units, about a million Btu per day. Mostlocal gas distribution companies will talk in “therms”and “dekatherms” rather than Btu or cubic feet. Thedekatherm is ten therms. Each therm is 100,000 Btu. Eachdekatherm is one million Btu. The line between large andsmall users is not rigid. Applications coming close to thisapproximation may still meet the criteria for going thelarge user route. If the user is on the small side, dependingon the state or location of the use, it still may have analternative of buying from the local distribution companyor using the LDC for transportation only and buying thecommodity from a marketing entity. Making contact withmarketing companies, which will be discussed later, andgetting information on the local regulatory rules will helpin making this decision. Many local distribution companieshave set up their own non-regulated marketing companiesto help consumers buy gas at the lowest price withthe required service criteria of their own operations. Oneshould not forget the potential of ECommerce, the newestway to buy and sell natural gas. A smart buyer will lookat all possible sources for meeting his requirements at thelowest price but with reliability and service.In buying a commodity like natural gas, price aloneshould not be the only criteria. Service (security of service,emergency additional supplies, etc.) equally impactsthe buyer’s bottom line as does price in meeting fuelrequirements. Having a cheap supply of gas where itsavailability is so uncertain as to disrupt plant or businessoperations is really an expensive supply when looked atin the total picture. Security of supply or additional supplies,etc. is a valuable consideration to be included inpricing natural gas sources.The large users—those over the thousand cubic feetlevel or close to it, should investigate all possible sourcesfor supply and transportation. Their sources may go allthe way back to the wellhead or producer marketingcompanies. Depending on how large a supply is neededat a given location, the buyer may include dealing withpipelines and distribution companies for transportationand delivery of the gas. Once the buyer knows in generalwhich direction to go, the big issues then become findinga marketer, transportation, and contracts for the servicesand commodity.21.4.3 Natural Gas MarketersMarketers come in varying forms, sizes, and descriptions.One can look at it much like purchasinggasoline at the local filling station—”Full-Service” or“Self-Serve.” To add a little more variety or confusion,gas buying and selling is moving to ECommerce and thebusiness-to-business Internet capabilities. When the startof marketing companies began in the m id-1980s to takethe place of the merchant function performed by the pipelines,it was almost anyone with a telephone and a pencilcould be called a natural gas marketer. Through the years,with a number of the marketing companies taking onadded scope and abilities, the ”fly-by-night,“ less reliablemarketers were pushed out of the business. Even someof the more reputable, better financed groups have gonebecause of the inability to be profitable in a fast moving,sometimes, irrational market place. With financial trad-

568 ENERGY MANAGEMENT HANDBOOKing exceeding high volumes of trading each day, risk becomesan even more important element of consideration.Marketing natural gas is more than just selling anddelivering gas to the consumer. The gas business is bigbusiness running into revenues of around $100 Billionper year depending on the exact price for the commoditythat year. The $100 Billion is only a measure of the actualcommodity trading on an idealized basis of direct tradesfrom producers to marketers to consumers. Actually, anaverage cubic foot of gas most likely gets traded three tofour times before coming to the consumer, the entity withthe burner tip that will consume the gas and put it out ofthe market. This is only for the physical side of the trading—theplace where the commodity actually is moved toa final destination for consumption. The total natural gasconsumption in 1999 was 22 trillion cubic feet (Tcf).This pales in comparison to the financial marketswhere 10 to 12 time the volume traded each day in thephysical market of consumed gas is traded in the financialsector. The money moved in this arena is beyond the$100 Billion discussed previously. At times, the marketis responding more to the financial than to the physicaldrives. The speculators are doing more to move the marketthan the actual users who need the natural gas for fuelor feedstock. Like all commodities, natural gas makes anideal medium for financial trading. There are those whoneed to make a play in the market for the protection orrisk-adverse properties the market gives. Those whoproduce the gas and those using large quantities can buysome protection of the future price by buying futures.This is “hedging.” The futures buyer is taking a positionfor a given month in the future where the price he payswill be the price for the quantity of gas he purchasedfutures for on that given month. He/she has locked inthe price for gas anywhere from a month forward to 36months forward. Whether buying or selling gas, hedgingis a tool to relieve some of the risk in buying or selling arelatively volatile commodity.The volatility of natural gas prices (no pun intended)makes it an ideal commodity for speculators to makea market in it for the sheer purpose of making money. Thespeculator is betting the price will be higher or lower on agiven date and is willing to take a position by buying thecommodity for trading at that time. Much of the tradingin natural gas is for speculation and this can only add tothe volatility of the market place. While most of the hedgersbring a relatively simple mentality to the market placebased on supply/demand parameters, the economy, andother pertinent factors, the speculators have a “statistics”of their own for playing the market. Basically, thespeculators are “market technicians” and play a statisticalanalysis of the market itself for buying and selling thecommodities. The mentality of the speculator is basically,“who needs to know all the details of the commodity, themarket place itself shows the results and following themarket place with its own statistical tools is the way togo.” Of course, many of these speculators are very largein the amount of money they control. When the signalsshow its time to buy or sell, very large sums of moneycan come into or leave the market. Easy to see how thiscan make the price of the commodity very volatile. Figure21.5 shows the prices for natural gas, coal, and crude oilfor the last five years to give a comparison among thesethree major fuels for electric generation as to the marketvolatility of each one.21.4.4 Finding the SellerNow, to whom one goes for buying natural gas isa question of the degree of service expected and needed.A large user wanting to hedge prices to insure a strongercontrol in the price paid for the commodity might go toa “full-service” marketer while someone needing a relativelysmall amount of gas at a reasonable price can callthe local distribution company or a more “self-service”marketing company. The selection is difficult becausethere are so many choices. There are roughly 30 majormarketing companies handling natural gas and anywhere up to a couple of hundred smaller groups. Thereare the local distribution companies in the area. Mostof them, in addition to selling “system gas” will havean affiliate or subsidiary selling market sensitive pricednatural gas also. System gas is natural gas the LDC haspurchased for resale to its local customers. Since this customerbase includes residential and commercial customersas well as the industrial sector, the average price willbe higher usually. Most local distribution companies havemade available open access transportation so that largeindustrial user can bring in its own gas supply and letthe local company transport it to the buyer’s facility. Aspart of its tariff, the LDC will set a minimum amount ofgas the buyer uses as a criteria for allowing the buyer topurchase its own gas and use the LDC for transportation.The tariff will set the cost for transportation by the LDC.In addition to the transportation costs, rate of return,and other pipeline costs in the charge, in many tariffs aprovision includes any local taxes or fees made by localgovernments for transporting natural gas.The LDC or pipeline affiliate will only sell the commodity.The buyer might also have a choice of buying fromother marketers and can “shop” its purchase needs to getthe best package of prices, services, and other options.The local distribution company would most likely be thetransporter for the customer. In some locations, this maynot be the case depending on the location of the buyer and

NATURAL GAS PURCHASING 569accessibility to other pipelines for transportation.Remember, price alone should not be the only considerationin purchasing natural gas. Dependability andservice have a definite value. One should always keep thisin mind when buying gas supplies. It might be wonderfulto buy the cheapest gas only to be unable to get it whenweather or other problems make the supply scarce!After the prospective buyer has determined whatits needs are and what alternatives it can live with, thenthe buyer should find the best source for buying the supplythat meets those necessities. Looking for a good suppliercan be a big part of the purchasing decision. But, itis an important element in the total economics that thebuyer succeeds in selecting the right source or sources.The buying could include the transportation of the gasto the consumer or, again, depending on the sophisticationof the buyers, they might purchase transportationseparately from buying the commodity. These are all partof the difficulties of purchasing a gas supply. Becauseof the many parameters to be covered and the need toknow the players and the system, many companies seekconsultants to help either initially or continually to makebetter decisions in gas purchases. The difficulty is unlessthe buyer is in the buying sector almost continuously, heor she will be at a distinct disadvantage in seeking theoptimum natural gas sources. The expense of using amarketing company or a consultant can be a very smallprice in finding the most effective and efficient source ofsupply. Table 21.5 lists the largest marketing companieswhether “full-service” or “self-service” in style.Looking at Table 21.5, one can see that many ofthe major marketing companies listed are a subsidiaryor partner of natural gas pipeline companies, either thelong-distance movers or the local distribution ones. Otherbig marketing companies are a subsidiary of the naturalgas producer. Companies like Shell Oil (subsidiaryCoral Energy), Texaco, and Exxon-Mobil all have marketingcompanies. Most of these are more to the “Self-Serve”type where selling the gas they produce or gas fromtheir associates or partners in the field is the purpose fortheir marketing operations. The buyer should sample alarge enough group of marketers to insure getting thebest price, reliability, and service. Selecting a gas supplysource is not an overnight task. The work needed shouldbe in proportion to the value of the gas to the operation.If large supplies of gas are needed, differences of a pennyor two or reliability are very important.The new area touched upon only lightly becauseof its newness is buying gas using ECommerce and thebusiness-to-business Internet. Many of the major gasTable 21.5 Natural Gas Marketing Companies

570 ENERGY MANAGEMENT HANDBOOKmarketers, either singularly or joining into a consortium,are making markets buying and selling gas through theInternet. Table 21.5 lists the marketing companies currently,summer of 2000, ready to trade using cyberspaceand Table 21.6 lists all ECommerce companies doingbusiness in the energy sector. The list will grow and playerswill change with time. The Internet marketers makeit easy for the knowledgeable trader to buy or sell gaswithout having to use a marketer or broker. How muchadditional effort the buyer will need to complete the saleand transportation will depend on how this system ofmarketing grows and prospers. After only a short time ofthis method of marketing being in existence, large enoughvolumes have been traded to see the value and potentialfor ECommerce business in the natural gas industry.21.4.5 Natural Gas PricingNatural gas is a commodity. There are many suppliersand the commodity is fungible. Its price is a functionof its availability. Simple, but true. When supply is perceivedsufficient, gas prices based on current conditionswould be in the $2-3/MMBtu range. Over-supply willsee the price drop significantly sometimes coming down40 to 50 percent of this price. T ight supplies can do thesame with the cap sometimes on a short-term basis, beingalmost unlimited. Early summer 2000 prices teeteredaround the $4/MMBtu range and went above $5/MMBtuby early fall because of fears of short supplies caused bythe summer electric generating high demand, gas storagefilling requirements, and winter gas supply. While themovement is based on supply/demand parameters, theproblem is two-fold: no one knows the supply/demandpicture with accuracy and secondly, fact and perceptionplay unequal roles. In the end, each buyer and seller mustmake its own decision on where the price will go in theshort and long term futures.Historically, natural gas prices in the beginningwere cents per thousand cubic feet. After crude oil pricesbecame market sensitive in the early 1970s, it was notuntil natural gas prices were decontrolled that gas ininterstate commerce came up to realistic prices rangingfrom over a dollar to $5-6/MMBtu. Gas prices during the1970s, before decontrol, for the intrastate market quicklycame to market sensitive levels of $3 to $6/MMBtu. TheNatural Gas Policy Act of 1978 ended the difference betweeninterstate and intrastate pricing. The high price fornatural gas after decontrol was an affect of the legislation,which set up about a dozen pricing categories. When thegas surpluses of the mid-1980s started, where the legislationhad set the “maximum lawful price,” it did nothingfor a minimum price. The gas merchants of that time, thepipelines, brought the prices down to the $2/MMBturange quickly. Since 1985, natural gas prices have variedfrom around a $1/MMBtu to current highs above $5/MMBtu. Figure 21.5 shows natural gas price history duringthe period 1995 through 1999.There are tools to help in price analysis and forecasting.In addition to the sources for tracking the current gasprices, there are tools for helping in projections of futureprices. Services that can supply forecasts based on theirinterpretation of the future are available. Many of the financialhouses making a market in natural gas and otherenergy futures have current material on their analysisof gas markets. The federal government has many publicationsand resources for tracking and estimating gassupply, demand, and pricing. And, of course, there aremany consultants offering pricing, supply, and demandforecasts. Many of the sources are free. Two things tokeep in mind regardless of the source of information.First, forecasting is an art. There are statistical methodsand programs to help in making predictions but many ofthe assumptions are based on the forecasters’ ability andexperience. It is still art not fact or science. Who can predictwith accuracy and precision the weather for a weekor six months out? Hurricanes come, blizzards come andsometimes little is known before hand. There is even adifference if the extremely cold weather comes during theweek or only on the weekend. During the week, schoolsand business facilities need gas for heating; weekendsthey are closed. The second point is simple. If the forecasterhas an ax to grind, be careful of the conclusions!Since it is an art, the forecaster who has a specific purposecan be prejudiced whether conscious of it or not. Someof the trade sources for natural gas price reporting andforecasting are listed in Table 21.7.Since gas is a commodity and depends more on thefactors of supply/demand for pricing than actual costs,gas prices vary significantly over a short time period.Each month, some of the gas trade publications (seeTable 21.7) give what is called the “gas price index.” Thisnumber is based on the price sellers and buyers are usingat the end of the month and it becomes the index for thenext month. The index changes each month and manycontracts use the index from a given publication as theprice gas will be bought or sold for that month. Sometimesthe contract will call for a penny or two per millionBtu above or below the index. The major index used atthis time is based on natural gas sold at the Henry Hub inLouisiana, a very common place for gas sales and trades.There are many more places where gas is traded and eachof these will have an index of its own or a “basis price,” amethod for converting from the Henry Hub price to thatlocation’s price. It is usually based on the value of the gasat that location versus Henry Hub and the added cost

NATURAL GAS PURCHASING 571Table 21.6 Commerce & Energy———————————————————————————————————————————————————Number Internet Address Activities & Purpose———————————————————————————————————————————————————1 icapenergy.com ICAP Energy, Inc. is an over-the-counter broker of natural gas, electricity and weatherderivatives.2 www.apx.com Is International. APX leverages the Internet to trade all electricity products—the electricitycommodity itself, the transmission rights needed for delivery, and the ancillaryservices that support and reliability. All integrated.3 www.buyenergyonline.com Buy and sell energy—Great Britain.4 www.capacitycenter.com Alerts to find capacity on natural gas pipelines.5 www.coralconnect.com Lots of information. Allows one to buy and trade gas and electricity with Coral Energy.6 www.energy.com Deregulation Status, Consumer Education, Links to energy suppliers, etc.7 www.energycrossroads.com “The e-partner of choice for small to mid-sized utilities.”8 www.energyon.com Buy energy on retail basis.9 www.greenmountain.com Green power focus.10 www.houstonstreet.com Electronic trading.11 www.oilexchange.com Oil property sales.12 www.onlinechoice.com. Gives customers access to buying pools. Also allows you to state your needs and receivea bid back.13 www.powerchoice.com Pepco offering to help in gas and electricity choice and to provide other electricity servicessuch as power surge protection.14 www. redmeteor.com Energy Trading system.15 www.scana.com SCANA Online is an Internet-based energy auction.16 www.itron.com Provides interactive energy e-business solutions for optimizing energy usage and energyprocurement processes.17 www.tradecapture.com TradeCapture.com is an innovation which will give you one stop shopping for multiplecommodities and locations in the physical and financial commodity markets. Has officesin Canada, Mexico and Great Britain.18 www.trueadvantage.com TrueAdvantage is a leading provider of private-label sales leads and RFP services to theonline B2B marketplace including energy.19 www.truequote.com Provides Price Discovery.20 www.utilisave.com Utility cost management, cost recovery, and e-commerce procurement solutions.of transportation between the two locations. The basisdoes not always vary as the value of gas transportationchanges. When gas prices are rising, the basis value canincrease and vice-versa. Examples of these differencescan be found in the trade publications listing natural gasprices, see Table 21.7.21.4.6 Contracts for Purchasing Natural GasAs has been said previously, the major change tobuying natural gas in the new millennium is the ability tobuy from many sources. This can mean buying from theactual producer regardless of where the consumer is located,to buying from local or national marketers or the localdistribution company. A consumer might buy from the localdistribution company in its area either directly from theutility or from a non-regulated marketing subsidiary of theutility. Another major difference today is the consumer canbuy the commodity and the transportation separately ortogether depending on the source of the gas, the quantity,the service required, and/or the location of the consumer

572 ENERGY MANAGEMENT HANDBOOKTable 21.7 Natural Gas Price Reporting Sources———————————————————————————————————————————————————SOURCE MEDIA TIMING TELEPHONE WEBSITE———————————————————————————————————————————————————1 ACEO/NGX Internet Same day & www.naturalgas.comnear month2 Bloomberg Energy Internet Spot Market A/C 609/279-4261 www.bloomberg.com/energyCurrent less 203 CNN The Financial Internet minutes www.CNNfn.comNetwork4 ENERFAX Internet Daily www.enerfax.com5 GAS DAILY Printed, Fax, Daily 800/424-2908 www.ftenergyusa.com/gasdaily& ElectronicCrutchfieldEnergy Data6 GASearch Internet Market 972/247-2968 www.gasearch.comintelligence7 Platts Printed, Fax, & Electronic Biweekly & Current A/C 800/752-8878 www.mhenergy.com8 Natural Gas Internet Various www.naturalgas.com9 NATURAL GAS Internet & Printed Monthly www.eia.doe.gov/oil_gasMONTH, EIA10 NATURAL GAS WEEK Printed, Fax, & Electronic Weekly & Current 800/621-0050 www.energyintel.com11 NATURAL GAS Printed, Fax, & Electronic Current futures A/C 703/318-8848 www.intelligencepress.comINTELLIGENCEWeekly & Current12 NYMEX Internet delayed 30 min www.nymex.com13 Reuters Fax & Electronic Current delayed A/C 800/438-6992 www.reuters.com———————————————————————————————————————————————————and time to insure proper legal resources are used innegotiating and drawing up the contracts for buying anddelivering natural gas are well worth the effort. Even invery short times of delivery or for very small quantities ofgas to be purchased and delivered, contracts must accuratelyand legally cover protection of all parties involvedin the transaction. This is where an “ounce of preventionis worth a pound of cure!”A contract or contracts between the two or moreparties will spell out the details of the transactionsneeded to purchase and deliver the gas from the sourceto the consumer. Many of today’s gas deals are doneover the telephone based upon agreed-to basic terms.Some are being done through computer and cyberspace.Whenever there is an on going relationship of supplyingnatural gas over a period of time regardless of the lengthof the time of delivery, there will be a contract or contractscovering buying and selling conditions includingtransportation, delivery, metering, payment, ownership,etc. There might be a contract to supply natural gas foras little as an hour or as long as a year or two and up toas long as 10 to 20 years. The long-term contracts of thecontrolled period when the transporting company marketedgas are no longer in use. Typically today, regardbothfor physical and/or regulatory reasons.These changes in how gas can be purchased havebrought changes in how contracts are written betweenthe supplier and the consumer. If the buyer is responsiblefor its own transportation, it would mean havingto contract for this transportation as well as contractingfor the gas supply. It also opens up new considerations.The buyer wants to make sure they are well protectedin getting the gas they pay for from the vendor and ifthe case is such, the transporter as well. In addition, thebuyer must be concerned he is protected from any liabilitythat might occur because of damage caused by the gasin the sale and delivery to the user. Contracts are legaldocuments covering these elements and need to be clearand accurate. After something happens—such as beingcharged for gas not received or for someone hurt in anaccident involving the gas in question—is not the time tostart looking at the contract. Who is responsible, or whatlimits there are for the difference between paying for avolume of gas and receiving a smaller amount, and anyother conditions and situations differing from what wasexpected should be stated in the contract. Recourse andresponsibilities should be spelled out in the contract.With contracts being legal documents, the expense

NATURAL GAS PURCHASING 573less of the term of the contract, provisions are includedfor price adjustments and for security of supply. Also,typically today, contracts longer than six months or ayear are considered long-term contracts. Contracts upto three to four years can be for a fixed price or a marketbased price depending on the whims of the buyer andseller. Most fixed priced contracts today will be based onthe financial market for futures contracts to protect thebuyer and seller from catastrophe due to sudden marketpeaks where the seller would be obligated to supply gasat a fixed price when prices are rising for the commodity.Further, the contract will protect the seller from the buyerending the contract prematurely. Likewise, the buyerwill want protection should prices drop significantly toreopen the pricing provisions.A contract is an allocation of risks between thebuyer and seller. It is the same between the party buyingthe transportation and the transporter. Every businessdeal involves risks and the contract sets the responsibilitiesof the parties so that there should be little argumentif something does not go according to plan. The seller istaking the risk of supplying gas and the risk of gettingits payment for the gas and services supplied. The buyeris taking the risk of having a reliable, secure source ofgas for its business needs. These are the major risks eachparty is taking and the contract is a written document toinsure both are protected as much as possible. Contractsare written documents to help in allocating these risks.But, even the best contract, written by the best lawyersand negotiators, is really, no better than the people offeringthe commodity and services and the people buyingthe services and commodity. No contract will help if theparty involved is not honorable, trustworthy and capableof doing what it claims it can do in the contract. Further,signing a contract and then planning on going to courtto enforce it is a waste of good time and assets of eitherparty. Contracts are like locks on doors—they are for thebenefit of good people to insure no one gets confused orforgets the details of the arrangement. Contracts do littleto protect from dishonest or untrustworthy business associates.Of course, even with good contracts and goodintentions of the parties, things go wrong and contractdisputes arise. These disputes can involve large loses ofmanagement time and company assets. Well-written andnegotiated contracts can keep the disputes to a minimumin happening and to minimum losses when the unexpecteddoes occur.Since one or more contracts may be needed to purchaseand deliver natural gas, the buyer should be carefulin his actions. Contracts for the purchase of natural gaswill usually have the following major areas of considerationas listed below. Many of these will apply to thetransportation contracts as well unless the purchase ofthe gas includes the transportation. Since today, sellersand buyers will vary considerably in their position in therespective industries, the contract needs to be tailoredspecifically between the two or more parties involvedin the transactions. A contract for buying natural gasfrom a local distribution company will be different inmany aspects from the contract between a marketingcompany and the buyer. The local distribution companyis a regulated entity and many elements that will be ina contract are part of the regulatory aspect. Most of thespecific items the utility will have to abide by are givenin its tariffs, which are filed with local or state regulatoryagencies. Some of the major elements of the gas purchasecontracts are outlined in the following:1. Purpose & Scope of the Agreement—What is tobe accomplished by the contract. Who will be supplyinggas, how will it be transported, and whowill receive the gas. Additional comments as to thepotential use, whether a sole supplier, etc. mightalso be included in this section.2. Definitions—Lists the standard and special termsused in the contract. Especially important in naturalgas contracts because of the uniqueness of thecommodity, the industry ways of doing business,and the specific parameters of the operations thegas is being purchased for by this contract.3. Term of the Agreement—A statement givinghow long the agreement will be in force and whatconditions will terminate the agreement. Somecontracts will include information on methods andoptions of extending the contract past the initialterms of the agreement.4. Quantity—Here, the details of the total quantity ofnatural gas to be sold and delivered by the sellerand received by the buyer will be stated. Informationon the daily contract quantity (DCQ) oreven hourly contract quantity will be stated. Anyspecific deviations from the regular amount suchas swing quantities needed during high productionor other causes are listed here. Penalties theseller is willing to accept for the buyer’s failure totake the quantity of gas set in the contract will belisted in this section. Also, the converse, penaltiesthe buyer is willing to accept for the seller’s nonperformanceaccording to the contract will bestated in this section. If there is any take-or-paylanguage, this is the section for it. Take-or-pay

574 ENERGY MANAGEMENT HANDBOOKis an agreement for the buyer to pay for gas ifcontracted but not taken. The Buyer usually has aperiod to make up the deficiency. The section willalso state whether the gas is being sold on a firmbasis with the buyer and seller obligated as statedto perform or if the gas is being made availableor will be taken on a “best efforts” basis. Veryimportant in this section could be the ways thebuyer “nominates” takes for certain periods. Thesection will include means for balancing the accountand give additional penalties for under- orover-quantities of gas taken by the Buyer. Othersubjects that play a role in the quantity of gasto be supplied such as well or reserve measurementsif buying directly from a producer or othersupply considerations if buying from a marketercan be in this section.5. Price—Price to be paid for the natural gas to bedelivered by this contract as well as any statementsregarding price escalators and/or meansto renegotiate the price will be stated in this section.Omissions of statements to this effect can beconstrued as a statement of the contract so caremust be exercised on what is said and what is leftout. Any language needed for agreement on priceindexing or other means of adjusting the price tocurrent market conditions will be included. Thewriting should include provisions for both priceincreases as well as decreases if this is the desiredpurpose of the statements regarding changes formarket or other conditions. The Price Section willcover any additional expenses or costs the buyeris willing to undertake in addition to the directcost of the gas. If the contract is with the produceror an interest owner in a gas well, the section willstate who is responsible for any gathering, treating,or processing costs. Again, for a contract withthe seller being a producer, provisions will be inthis section for who has responsibility for severanceand other taxes, royalties or other chargesthe producer might be liable for payment. Pricingunits most commonly used today will be energyunits such as British thermal units (Btu). The dollarvalue per thousand Btu is typical such as gasat $0.003/MBtu. Since a typical volume measurementof natural gas is a thousand cubic feet (Mcf)and typical pipeline quality gas of this volumewould have about a million Btu in energy units,the typical unit for gas sales is a million Btu orMMBtu. The above example would be $ 3.00/MMBtu. The Pricing Section will also includelanguage in today’s contracts protecting both thebuyer and the seller from the vagaries of the naturalgas markets today. While these in effect reducethe coverage of the contract and change some ofthe allocations of risks set by the contract, oftenboth parties are willing to have a contract withlegal means of changing the pricing conditions ofthe contract. The long-term, fixed price contractswent out with decontrol. There are still fixed pricecontracts but, the seller will protect its position bygoing to the financial market and buying futuresto protect his position of supplying long-term,fixed price natural gas. Since the seller is takingsteps to insure supplying the gas at a fixed priceby buying futures, the seller will protect his actionsby putting clauses in the contract to protectthis position should the buyer fail to take the gasas contracted.6. Transportation—Transportation details as to whohas responsibility, the transporter, costs, etc. todeliver the gas to the buyer must be included inthe contract. Crucial items are who is responsiblefor arranging the transportation, who will pay forthe transportation, and whether the transportationwill be on a firm or interruptible basis. The transportationmust cover the full course of bringingthe natural gas from the source to the buyer’s locationincluding bringing it to the accepted deliverypoint(s) as stated in the next paragraph. The buyermust insure they are covered in case transportationis unobtainable or ceases after delivery hasstarted. The contract will include any special conditionson either the buyer’s or seller’s part to takeinto account any special situations either partycould have to interfere with transporting the gasfrom the source to the delivery point(s). Further,any regulatory matters pertaining to the gas transportationshould be referred to in this section andin more detail in the regulatory section discussedfurther in this chapter. This section should coverwho has responsibility for overages and balancingof the account, measurement, disagreements onquantities, and payments of transportation and associatedcharges.7. Delivery Point(s)—Since the delivery point orpoints are different in each situation, the contractneeds to state delivery or alternate delivery pointsin very specific language. This clause can becomea very crucial one in times where there is a disputeover quantities of gas sold or received. There is

NATURAL GAS PURCHASING 575also a slightly different interpretation of this clausein light of the new sales methods where there is aseparate contract for the sale of the natural gas andone for the transportation. To the gas buyer, theonly real delivery point is when the natural gascrosses the meter and valve where the gas comesdirectly into the buyer’s system. The buyer wantsto be responsible only for the gas received in hissystem. What gas is presumed sold or dispatchedat some other location such as where the gas mightcome off of an interstate pipeline into the pipe ofthe local pipeline delivering the gas to the consumeris really not the concern of the buyer. Thisis an argument between the two pipelines or betweenthe pipeline and the seller depending on thecontract for transportation between the significantparties. Delivery point is also crucial in assessingresponsibility for problems that might arise fromthe gas in question. An explosion or fire resultingfrom the improper handling of the gas and theensuing legal action by one or more party couldbe influenced by the delivery point as to who hadresponsibility for the gas at the time and locationof the accident. At times, delivery points may needto be changed to meet requirements of either partiesand the need to change should be included inthis section to insure that changing the locationsaccording to the contract do not in any way negatethe contract or the terms in the contract.8. Measurement & Quality—Methods, conditions,timing, and authority for the measurement of thegas volume and quality are given in this section.Usually, a trade association or other organization'smethods and requirements for measurementare called for in this section to insure theproper measurement of the quantity of gas soldor bought and the quality of the gas under thecontract. Remedies or alternatives should be includedin this section for those cases where thegas fails to meet the quality requirements of thecontract whether on a short-time, unexpectedrare or single occurrence or a continuing failureto meet the specifications.9. Billing—Terms for the billing, who is responsiblefor payment, manner and methods of payment,etc. are included in this section.10. Force Majeure—This the clause in the contract toprotect both parties affected by a totally unforeseeableoccurrence which is beyond the control of theparty seeking protection from the responsibilitiesof the contract. Many times, this is referred to as an“act of God” and includes severe weather, acts ofwar or insurrection, strikes, etc.11. Warranty of Title—The clause guarantees theseller has title to the gas and can sell it. Includedare allowances for the buyer to recover damages ifthere is a failure of title should another party makeclaim to ownership of the gas.12. Regulatory—All necessary permits, licenses,etc. must be obtained according to FERC regulationsand any state or local authorities havingjurisdiction over the selling and transportation ofthe natural gas. The party or parties having theresponsibility for obtaining these and paymentsrequired should be fully covered.13. Assignments—Specifications for the transfer ofrights under the contract are covered in this section.This could be an important item in light ofthe various changes occurring in the gas industrytoday. The buyer should insure coverage includeschanges that might impact gas transportation aswell as the commodity if the seller was responsiblefor transportation as well as for the natural gas.21.5 NEW FRONTIERS FOR THE GAS INDUSTRYA number of challenges face the natural gas industrygoing into the new millennium. Each of these willhave an impact on the future buying and trading of thecommodity. A summary of these follows:1. Complete the natural gas decontrol to the finallevel—local markets in each state.2. Complete the development of an energy industrythat incorporates other energy sources like power,fuel oil, etc.3. Develop the delivery system to insure secure supplyof the larger quantities of natural gas demandforecast for future years.4. Develop new supply sources to meet the forecasteddemand through the 2015 period forward.Each one of these will play an important role in thegas industry of the future. More important, each one willrequire capital flows into the industry to make them possible.


CHAPTER 22CONTROL SYSTEMSSTEVE DOTY, PE, CEM22.1 INTRODUCTIONControl systems are an integral part of manyenergy related processes. Control systems can be assimple as a residential thermostat, to very complexcomputer controlled systems for multiple buildings,to industrial process control. Their diligence andrepeatability can also serve to maintain the savings ofthe project improvements for years, further justifyingtheir existence by providing economic return to thecustomer. This chapter will introduce the reader tosome concepts of automatic control theory, followed bypractical applications useful to the field of the EnergyProfessional. Upon completing this chapter the readerwill gain a basic understanding of common terms, controltechnology and control mode categories, basic input andoutput instrumentation, and the practical need to temper‘things possible’ with the skill level of the operators whowill inherit it. The importance of system controllabilityand user-friendliness, as primary design parameters, willbe stressed. Basic control strategies will be discussed.Estimating savings from the use of Automatic Controlswill be discussed. Finally, there will be an introductionto some complex optimization methods, and suggestedtopics for further study. Examples are used throughout.The intent of this chapter is to focus on theapplication of Automatic Controls as a tool to achieveenergy savings. Some background information oncontrols is obviously a necessary prerequisite; howeverthis chapter does not attempt to cover all aspects ofautomatic control theory or application or to make thereader a controls expert. Less emphasis will be placed onhardware and theory, and more emphasis will be placedon practical applications and tangible benefits. A basicbackground in energy-related subject matter, commonenergy units, common energy-consuming machinery andsystems, HVAC concepts, and the general field of energyengineering is assumed. The field of Automatic Controlsis a busy technology with lots of jargon, hardware andsoftware variations and details, and the sheer volume ofit can create an air of mystery and awe. If this chapter577is successful, the reader will be able to separate thefundamental control concepts from the technical details,and effectively apply Automatic Controls to achieveenergy savings.Many of the examples given are for commercialbuilding HVAC and lighting systems, since these examplesare common and should be familiar to the reader. Similarconcepts and considerations apply equally to other fieldsof endeavor where energy savings are a driving force.A Glossary of Terms provides clarification ofcommon terms used in the field of Automatic Controls.22.2 WHY AUTOMATIC CONTROL?• Regulation: Many things need attention andadjustment to compensate for changing conditions,or varying demands. Examples of this are commonin living organisms, such as body temperature, bloodpressure, etc. Process control regulation is really justemulating the concepts of such natural processes.The field of automatic control is similar in that we‘continually adjust some device to cause a particularmeasured variable to remain at a desired state.’Examples:The need to throttle heating and cooling equipmentsized for maximum load that is effectively oversizedat part load conditions.Varying occupancy, and systems attendant to theoccupants (lighting, ventilation).Varying product throughput rate throughmanufacturing facilities.Varying demands, and the need to maintain levelor full state for water or fuel reservoirs, feed orcoal bins, etc.• Coordination: Organizing or sequencing multipleprocesses in a logical and efficient manner is animportant aspect of automatic control applications.• Automation: Human beings can make very goodmanual controllers because we can think on ourfeet and consider many variables together, butmost control tasks are repetitive and suitable formechanization. Automatic operation allows peopleto provide oversight of system operations and more

578 ENERGY MANAGEMENT HANDBOOKeffectively utilize their time.• Consistency: Manual control by people can beeffective, although we are not all that repeatableand are sometimes forgetful. Using machineryfor automatic control adds the improvement ofconsistent, repeatable operations. The repeatabilityand consistency feature of automatic control is veryimportant in manufacturing.• Conservation: Supplemental enhancement controlroutines can be incorporated to reduce energy usewhile still maintaining good control. It is importantto note that control systems do not necessarilyreduce energy consumption, unless specificallyapplied and designed for that purpose.22.3 WHY OPTIMIZATION?The 80-20 Rule reminds us that we can usually hopeto achieve 80 percent of the measure’s potential with20 percent of the difficulty, but the remaining portionrequires much more effort. For this text, Optimizing refersto reducing energy consumption as much as possible,often approaching the barrier of diminishing returns.Optimization can be characterized as taking over wherethe basic controls left off and working on the remainingopportunities—the ones that aren’t as easy to attain.The appropriate use of optimization depends upon thecustomer’s priorities, and these should be tested beforethe decision to optimize is made. Of course, from anenergy conservation or ecology standpoint, we should allpress for that last 20 percent. But if maximum simplicitycontrols that require only basic skills are a main focus ofthe customer, optimization may not be a good application.Similarly, projects where reliability is priority #1 maybe better served with basic control routines, allowingthe extra 20 percent potential to slip away to gain theadvantages of simplicity. Economics always comes intoplay, and some optimization projects (chasing the last20 percent) may not have the attractive payback periodsof their 80 percent counterparts. Most projects representsome balance of these interests, depending upon theneeds of the customer. It is important to understandthat optimization for maximum benefit will not be foreveryone.A case in point for optimization is the subject offixed set points, which are often a matter of convenienceor approximation, and usually represent a compromisein optimal energy use. The more factors we can take intoaccount, the closer to optimal will be the result, as statedby Liptak: “…multivariable optimization is the approachof common sense. It is the control technique applied bynature, and frequently it is also the simplest and mostelegant method of control.” [1: pp xi]To summarize, the desires for maximum simplicity andmaximum efficiency are at odds with each other. A systemthat is perceived as being too complex will likely fallinto disrepair and be bypassed or unplugged. If thecustomer is committed to squeezing their energy coststhrough optimization, they will need to also embracethe technology and be willing to adapt and changealong with the process. It is almost a given that pushingthe envelope of optimization requires the operationspersonnel to accept additional complication and raise thebar of required operational skill. This concept should bediscussed in advance to be sure the project isn’t set up tofail by being unacceptably complex.22.4 TECHNOLOGY CLASSIFICATIONS22.4.1 IntroductionThe following is a very important first statementbefore any discussion about control hardware: “Thetype of hardware used in optimization is less importantthan the understanding of the process and of the controlconcepts that are to be implemented.” [1: pp 42]The main goal should be to become clear about theprocess fundamentals and what should happen—thenthe parts and pieces are just details. This discussionof different available hardware types is a familiar butsometimes laborious and dull part of any controls text.Remember that Automatic Controls are really nothingmore than machines that do for us what we would doourselves if we had nothing better to do; they do work forus like any other tool, and are only as clever as the peoplewho craft them.22.4.2 Conventional Electric(also On-Off or Two-State)Electricity is used as the power source. Control isdiscrete (on-off, high-low, cut-in/ cut-out, etc.). Contactclosures are used to implement control logic. Thisprinciple of control has widespread use, varying fromsimple and familiar control to complex interlocks andBoolean Logic.Examples:• A basic home heating thermostat cuts in at67 degrees F and cuts out at 69 degrees F,thereby maintaining a room temperature ofapproximately 68 degrees F.• A well pump controller cuts in when the storagetank pressure is 50 psig and cuts out whenthe storage tank pressure is 70 psig, thereby

CONTROL SYSTEMS 579maintaining a system pressure of approximately60 psig.• An interlock circuit that prevents an exhaust fanfrom starting until the associated make-up airdamper is first proven open.The differentiator between Conventional Electriccontrols and Analog Electronic Controls is the discrete(two-state) nature of the inputs and outputs; analogcontrols have varying rate inputs and outputs.22.4.2.1 Floating Control.A variation of this two-state control is “ floatingcontrol.” Technically still “on-off,” this unique controlmethod has a system control action similar to analog(modulating) control. With floating control, wheneverthe signal is sufficiently off set point, the motor actuator(moving a valve or damper) is energized, providingcorrection to the process. What makes floating controldifferent is that once the feedback measurement indicatesthe process has returned to set point, the actuator isde-energized but the device (valve or damper) holdsits last position—which may be 1/4 open, 1/2 open, orany midrange point. Because of this, floating controlcan achieve tighter control than simple on-off control,approaching that of true modulating control. Floatingcontrol is less expensive to implement than modulatingcontrol, and is used for small terminal HVAC controlvalves and dampers, as well as large machinery likechiller inlet vanes. See also Control Modes for a diagramthat illustrates floating control action.22.4.3 Analog ElectronicElectricity is used as the power source. The keydifference between analog and conventional electriccontrol is modulation. Analog controls have variableinputs and outputs, not just two states. Minor changesin output positioning of controlled devices make tightercontrol possible than with two-state (on-off) electriccontrols.The hardware for analog electronic controlsmay include resistors, rheostats, Wheatstone bridges,operational amplifiers, or may use solid state componentsto measure the process and modulate the output devices.Some considerations of analog electronic controls:• Unless a control dead band is built into the controller,it is likely that the controlled device will be activatedby even the most minute deviation from set point.The action may be nearly imperceptible, but ifthis occurs the repeated hovering about the idealset point may over work the actuators and causereduced device lifespan.• Analog electronic technology “connectibility”options are generally limited to remote adjustments,remote alarm panels, etc. This, and modest cost,make them a popular choice for basic HVACmanufactured systems that come with factoryinstalledcontrols.22.4.4 PneumaticThis is the general term for controls that usecompressed air as the motive force for control inputs andoutput, instead of electricity. Analog pneumatic pressuresare alternately coded and de-coded into control units;for example: 3-15 psig = 0-100 degF. Discrete pneumaticpressures are also coded; for example, 0 psig = off, 20 psig= on. Pneumatic devices can be two-state or modulating,but are most commonly modulating. Pneumatic controlsoften have interface devices that communicate pneumaticsignals to and from their electric counterparts, such asPressure-to-Electric switches (PE switches), and variousElectric-to-Pneumatic solenoids and transducers.Some considerations for pneumatic controls:• Air supply quality is critical. No oil or water allowedat the instruments!• Contamination in the main air system representsa potential single point of failure for many or allpneumatic controls.22.4.5 Digital Control (also calledDirect Digital Control or DDC)This control technology uses microprocessors toprovide control. A big advantage to DDC controls isthe fact that changes to the system are often made withsoftware and do not automatically require physicalchanges and cost like other technologies. Discrete (onoff)information is readily absorbed as a “1” or “0.” Foranalog processing, interface equipment called digital-toanalogconverters (D/A) and analog-to-digital converters(A/D) are used. The higher the resolution of the A/D andD/A conversion, the closer the digital signals resembletrue analog signals, allowing smoother control.One of the great enhancements of digital controlsover the last 30 years has been the concept of distributedcontrol. This technology shift occurred in response tocustomer complaints of excessive dependence on singlehardware or software points, and widespread loss ofcontrol after a single item failure. Current best practiceuses multiple smaller controllers at the points of control.These communicate upstream to a supervisory operatorworkstation, but each single failure point now affects amuch smaller area, increasing overall process reliability.• DDC Graphics: A Graphical User Interface (GUI) isan overlay onto most DDC control systems. While

580 ENERGY MANAGEMENT HANDBOOKlabor-intensive to create, these are useful for gainingacceptance of the control system, and can reducethe skill level needed for digital control systemnavigation. The GUI uses easily recognized iconsand symbols, color-coded messages and alarms, andother visual methods, to increase the user-friendlinessof the system. The GUI is analogous to the labeleddisplay boards of traditional pneumatics.• Field User Interface: For effective operations andmaintenance, the user interface should not belimited to operator workstation graphics. Like thepressure gages or test ports of a pneumatic system,some form of user interface should be provided atthe point of control to invite personnel to interact.This may be a fixed display-adjusting devicethat can scroll through a few points, to a portableoperator workstation that can be connected at will.Without some provision for user interface in thefield, the DDC system can introduce a frustration tothe operations personnel who cannot “see” what isgoing on at the equipment locations.• DDC Controller Hardware: Today’s digital controlscome in a variety of styles, to suit the application. Forcommon applications, mass produced and low costdefinite purpose controllers are available. These usefactory programming, also called ‘canned’ software,which speeds application and start-up, but maylimit the opportunities to create special instructionsor for optimization, unless sufficient programmingis available from the supervisory controller toachieve the intended result. Multi-purpose genericcontrollers are also available that utilize customprogramming. These generally come with a fixednumber of input and output points, but the pointcapacity can often be expanded with multiplexers orother point expander cards. Since the programmingis custom, these controllers are the natural choicefor optimization routines that require dynamiccalculations and adaptability. For control processeswith a large number of points, point expanderboards are usually preferable to multiple linkedcontrollers because having a single processor braineliminates some unpredictable operations modesthat come from partial failures, e.g. if just one of thecontrollers fails. Conversely, too many expanderboards can turn this application into a large-effectsingle point of failure.• Proprietary vs. Open DDC Controls: The topicof proprietary controls, captive customers, openprotocols, hybrid systems, gateways, etc. is a hottopic in the control industry today. While the conceptis not difficult to grasp, details and implications arevery numerous and complex, and beyond the scopeof this text. The basic argument FOR proprietarysystems is the one-stop-shopping ease of purchasing,and security of all-encompassing technical resourcesfor system maintenance and system integration; forlarge or complex systems this can be an importantbenefit. The basic argument AGAINST proprietarysystems is the proprietary lock on the customer, andhow the lack of competition almost certainly resultshigher replacement costs that escalate during theownership life cycle.Current state of the industry includes bothproprietary systems and “ open protocol” systems, andboth are viable, requiring a customer choice where newsystems are proposed. Gateways and translator offeringsare hardware go-between solutions that offer limitedconnectivity between proprietary systems. These arecommon solutions, but can create new problems evenas they solve others—and many retain the proprietarynature they are intended to solve.• DDC Information Technology (IT) SystemMaintenance: Systems that share communicationinfrastructure with other, more sensitive,systems require careful attention to informationconfiguration and maintenance. The ideal approachis to have computer Information Technology (IT)capabilities on staff; devoting their time to controlsystem integration activities as an in-house expense;however this may be cost prohibitive for all butthe largest systems. The more common ‘drop-in’approach is to select an industry partner to providethis technical service, although this can gravitateback toward proprietary solutions unless theperson maintains an independent status from anyequipment manufacturer.22.5 CONTROL MODES22.5.1 IntroductionDeciding which control mode to apply is important,regardless of the technology used. It is important tounderstand that these modes can be implemented usingmany of the available technology types. In many cases,simple on-off control is adequate, and very appropriate.In other cases, the desired effect can only be achievedwith modulating controls. The following are basic control

CONTROL SYSTEMS 581———————————————————————————————————————————————————TECHNOLOGY PROS CONS———————————————————————————————————————————————————Conventional • Simple • Definite purpose controls are not flexibleElectric • Proven without hardware change.• Easy to understand and troubleshoot • Limited capabilities for optimization.• Inexpensive• Accommodates “and-or” logic with simpleseries-parallel circuits (relay logic).• Floating control variation approachesanalog control quality with reduced cost.———————————————————————————————————————————————————Analog • Lowest cost stand-alone modulating • Long term drift unless properly maintained.Electronic control option. • Short-lived components.• Often standard equipment on packaged • Definite purpose controls are not flexibleHVAC equipment.without hardware change.• Often not user-friendly.• Limited capabilities for optimization.• Input/Output hardware items, especiallyactuators, are more expensive than conventionalon-off devices.———————————————————————————————————————————————————Pneumatic • Simple • Long term drift unless properly maintained.• Proven• Temperature dependent drift.• End devices extremely durable• Definite purpose controls are not flexible without• Can be long lived if designed and maintainedproperly.hardware change.• Limited capabilities for optimization.• Susceptible to system-wide failure if compressedair system is contaminated.• Complicated adjustments for pneumatic controllers• First cost includes infrastructure cost (tubing).• Future operator work force may have areduction in skill level for this technology.———————————————————————————————————————————————————Digital • Excellent flexibility, using software changes • Highest first cost.Control with existing hardware. • First cost includes infrastructure cost (cabling).( DDC) • Best option for optimization due to • Often outdated before physically worn out,computing power.due to rapid technology advances.• Opportunities for communication linkage to • Tendency for proprietary equipmentother, compatible equipment for largemanufacturers to create captive customersdata gathering/workstation benefit.and high life cycle costs for repair parts• Convenient method of recording keyand software upgrades.measurements, baseline information,historical data, trends, run-times, etc.• Convenient remote monitoring.———————————————————————————————————————————————————Figure 22.1 Pros and Cons of Different Control Technologiesmodes. The accompanying diagrams will illustrate typicalsystem performance.The term “ system capacitance” refers to the rateof response of a system to a stimulus. Systems with alarge capacitance tend to resist change and the effects ofcontrol are felt more slowly than systems with smallercapacitance. Comparing the effect to a flywheel orrelative mass is a good way to describe this concept.Another useful example to illustrate system capacitanceis an instantaneous electric water heater (small volumeof water) compared to a standard residential tank-typewater heater—upon energizing the heater elements, thewater temperature in the tank unit changes much moreslowly because it has more mass, and we say it has greatersystem capacitance.The term “ gain” is a control term synonymous withsensitivity, and is usually an adjustable value used to tunethe controls. If a quicker response is desired for a smallinput change, the gain is increased. This makes the inputchange more noticeable, and results in a stronger outputreaction from the controller.22.5.2 On-Off ControlAlso called “ two position control” this rudimentary

582 ENERGY MANAGEMENT HANDBOOKmode is used with equipment that is either on or off. Anominal set point exists but is rarely actually achieved,except in passing. A range of control values must betolerated to avoid short-cycling the equipment, andtemperature ranges are often fairly wide for this reason.In the case of equipment that cannot be modulated,this is often the only choice. Thesmoothness of control dependsstrongly upon the system capacitance;systems with very lowcapacitance can experience shortcycling problems using twoposition control.output is issued to regulate a process, and the magnitudeof the output is directly proportional to the size of theerror. This type of control is economical and reliable. Acharacteristic offset (residual error) is natural with thistype of controller, and the size of the offset will increasewith load. This offset is due to the inherent fixed-gain of22.5.3 Floating ControlA hybrid combination ofon-off control and modulatingcontrol, also called incrementalcontrol. As with on-off control,there is a control range (cut-in/cut-out). However, unlike on-offcontrol, floating control systemshave the ability to maintain amid-position of the controlleddevice, instead of full-on or fulloff.Between the cut-in and cut-outthresholds the controlled devicemerely holds its last position. Theprocess variable is not actuallyunder control within this range,and is seen to float with the loaduntil it crosses a threshold to getanother incremental nudge in thecorrecting direction. This controlis not as tight as true modulatingcontrol, but is inexpensive andreliable. Equipment items fromterminal units to 1000 HP WaterChillers are controlled in thismanner with good success. Notethat floating control is limited toprocesses that change slowly anddo not require very tight control.22.5.4 Proportional-Only Control (P)This is the basic modulatingcontrol, and what most commercialpneumatic and analog electronicsystems utilize. It is essentially anerror-sensing device with a fixedgain or amplification. A controlFigure 22.2 On-Off Control Mode DiagramFigure 22.3 Floating Control Mode Diagram

CONTROL SYSTEMS 583the controller.If the proportional control action is too sensitive (gainset too high), the controller’s response will be excessive,and oscillation or hunting will occur. When this occurs,the controller output (and the equipment connected to it)oscillate up-and-down, open-and-closed, etc. and don’tsettle out.22.5.5 Proportional-Plus-IntegralControl (PI)An integral functionis added to aproportional-only controllerto eliminate theresidual error. Thiscontrol action adjuststhe gain to a strongerand stronger value untilthe error is eliminated.In theory, the integralcontroller will not rest aslong as any error exists,however it is common toallow a small acceptableerror band aroundthe nominal set pointto prevent incessantlow-level hunting asthe controller seeksthe perfect “zero” errorcondition. In practice thishas the appearance of slowly butsurely building up the outputto taper off the error. Since the“ wind up” effect is slow, butalso relentless, integral controlproblems can occur in processesthat change rapidly. Also, if thecontroller is left active for longperiods while the controlledequipment has been turned off,the integral controller will “windup” to a 100% output. Upon startupafter a long period of windup,the integral function can bestrong enough to “stick” at fulloutput for long periods of timewith complete loss of control.Therefore, whenever integralcontrol is used, some form ofhardware or software interlockmust be used as an anti-wind-up measure. The mostcommon approach to preventing wind-up is to simplyturn off the controller when the process is stopped.22.5.6 Proportional-Integral–Derivative Control (PID)PID Control is used to accommodate rapid changes inprocess and minimize overshoot. This is done by reactingto the rate of change of error (derivative) instead of theFigure 22.4 Proportional Control Mode DiagramFigure 22.5 Proportional-integral Control Mode Diagram

584 ENERGY MANAGEMENT HANDBOOKmagnitude of the error (proportional) or the durationof the error (integral). In reviewing the characteristicresponse curves, PID control looks like the absolute bestand, in fact, does provide the tightest control of all thecontrol modes. However, the derivative gain is verytouchy to set up properly and can easily cause instabilityof control, especially at the beginning of a batch processor after a large step change. In HVAC work, the derivativeterm is seldom used to avoid the potential for instability,and since most HVAC processes are relatively slow actingand are tolerant of temporary overshoot of PI control. Inmany process control applications, PID control is essentialsince close control is often tied to product quality.• Normal offset from proportional control.• Normal time lag from Integral control.• Independent, but sequenced, modulating controllers.• Adjacent processes, such as two comfort zones in anopen bay with widely different user set points.Where the process does not require absolute tightcontrol, building in some blank spaces or dead bandbetween sequenced elements is a simple way to achievingand sustaining energy savings.22.6 INPUT/OUTPUT DEVICES22.5.7 SequencingWhile not actually a separate control mode, thistopic deserves special attentionsince it has great potential forenergy savings. A significantblight in many industry processesis the overlapping of adjacent andopposing processes. A commonexample is HVAC reheat, whereair that has just been cooled withan energy source is now beingheated. This is analogous todriving around with the acceleratorfixed and controlling the vehiclespeed with the brakes; even if thedetriment to the brake system isneglected, the effect upon vehiclefuel economy is obvious. There aremany examples of heating/coolingoverlap, some deliberate but mostinadvertent—but all make overallenergy consumption higher thannecessary. One very effective toolin combating or correcting this iseffective equipment sequencing.Ignoring some of the imperfectionsof real-world control systems,some control systems that appearsequenced will actually overlapoccasionally, or over time. Somethings that contribute to inadvertentoverlap are:22.6.1 IntroductionThere are many available input and output devices,Figure 22.6 Proportional-Integral-Derivative Control Mode Diagram• Instrument drift.• System influence upon actuatorrange of motion.• Normal overshoot from on-offcontrols.Figure 22.7 Sequencing with Dead Band

CONTROL SYSTEMS 585serving many basic and specific needs, along withranges of quality and other features as required for thejob, and the Controls Application Engineer quicklybecomes familiar with many of these in great detail. TheEnergy Professional may choose to delve into the sea ofproducts, but can also be very effective by providing onlyperformance-based generic requirements and managingthe project from a more macro view.For input and output devices, there are basic distinctionsbetween Transducers, Switches, Sensors, and Transmitterswhich are useful to understand.• Transducers: These are the core of any instrument,and are used to convert the basic physicalphenomena of interest into a form more useful tothe instrument.Examples:— Temperature: bimetal coil, two-phase gasbellows, thermistor, RTD, etc.— Pressure: Bourdon tube, diaphragm, strain gage,etc.The term ‘transducer’ is also commonly appliedto output signal form changing equipment, suchas converting an analog electronic signal to acorresponding pneumatic signal for use by apneumatic actuator.• Switches: Devices that can have two states (on-off,open-close) etc., used for regulating on-off electricalcircuits as inputs or to actuate other equipment ordevices in a two-position manner. Solenoid valves,relays, etc. are in this class of instruments.• Sensors: These are “passive” input devices that canbe read directly by the controller with no signalconditioning. They are usually limited to short cablelengths and further limited to transducers withinherent linear outputs.Examples:— RTD (resistance temperature detector) that canbe wired and read directly to the controller.• Transmitters: These are the typical analog inputinstrument. A transducer is coupled with some formof pneumatic or electronic signal conditioning. Thetransmitter output is a linear, standard value easilydecoded as an input. Often, the signal conditioningallows for stable signal over a large distance toallow remote location of the device, hence the term‘transmitter.’Examples:— 0-100 degF ‡ 3-15 psig— 0-100 degF ‡ 4-20 mA (milliamps)— 0-100 degF ‡ 0-10 VDC (Volts DC)— 4-20mA output ‡ 3-15 psig22.6.2 Conventional Devices and WiringConventional wiring architecture consists of devicesin the field; each wired back to the input or outputterminals of the controller, with a dedicated set of wires(often a pair of wires). Thus, for 100 instruments therewould be 100 sets of wires finding their way home to thecontroller. This type of wiring is very straight forward andis referred to as home run wiring. At the control panel thenumber of wires is the greatest and the system of wiresfans out and disperses, the further it gets from the centralpoint.22.6.3 Addressable (‘Smart’) Devices and WiringAlso called ‘smart’ devices, these devices can becontrolled directly by a digital control system, by givingthe device a unique identification code or “address” todifferentiate it from all other similar devices. Usuallyeach addressable device has a means of setting up itsown unique address or name, to permit uniquenessin communication. Addressable devices relay theirinformation or accept their commands digitally, and donot rely on transmitters or the like. Addressable deviceshave the distinct advantage of reduced wiring, since asingle loop of communication trunk wiring can be sharedby multiple devices; on systems with large numberof points, the difference can be remarkable. However,addressable devices cost more than conventional (nonaddressable)devices, since they include on-boardcommunication hardware, as well as A/D and D/Aconverters for addressable analog devices. Since they costmore, their application requires balancing the cost vs. thebenefits. Of the devices listed below, the ones currentlymost popular are the simple contact closure input/outputdevices used in large quantities for security, fire alarm,and lighting control systems, where the economies ofscale show an advantage. The single loop communicationcan present new failure modes compared to conventionalhome run wiring methods; i.e. if a single cable is cut howmany devices are affected?Common Addressable Devices:• Fire alarm devices—input and output• Lighting control devices—input and output• Security devices—input and outputOther Addressable Devices:• Lighting ballast

586 ENERGY MANAGEMENT HANDBOOK• Actuators• Transmitters and sensors• Thermostats• Relays22.6.4 Linearization“For simplicity of design, a linear relationshipbetween input and output is highly desirable.” [3: pp28]Linearity is a high priority in instrumentation, forboth inputs and outputs. This is because the controller’salgorithms or ratios are arithmetic in nature, comparingthe input to a standard and producing a derived output.Linearization makes the effect of the controller’s decisionspredictable and manageable. Understanding howlinearization can affect the control system is important toassure project success.Input/Output instruments are considered linearwhen an incremental change of input value producesan equal increment of output value regardless of thevalue of the input or location of the device’s range. Forexample, a change of 1 degree F might produce a changeof 0.16mA in transmitter output. If truly linear, the devicewould produce this 0.16mA change in current if reading0 to 1 degF, or if reading 100 to 101 degF. To the extentthat non-linearities exist in the control loop, errorsand unpredictability will also exist and so should beidentified and minimized.Many natural phenomena are linear, and many arenot. Linear Examples include metal resistance with respectto temperature changes, static pressure with respect todepth of liquid, and volume of a vertical cylinder withrespect to level. Non-Linear Examples include the volumeof a conical container or a horizontal cylinder with respectto level, the change in flow with respect to butterfly valveor single blade damper position, and a heat exchangerheat transfer rate with respect to flow rate. Some naturalphenomena behave in predictable, but non-linear fashion.These include thermocouples and differential pressureflow meters (head loss devices). Thermocouples arelinearized by a look-up table or mathematical expressionthat defines the non-linearity, while head-loss meters arelinearized with a square root extractor. Most commoninput measurements have already been linearized bythe instrument manufacturers. Instead of hardwarecharacterization, it is possible to linearize inputs andoutputs through software. This “software linearization”is sometimes used with industrial controls, but seldomwith commercial controls.A common control issue with linearization is controlvalves and dampers. These are notorious for having nonlinearresponse with respect to travel position. Controlvalves are generally characterized as either linear,quick opening or equal percentage. For valves, the flow“character” is achieved by specially contouring the valveplug, to influence the flow rates at different valve stempositions. Characterized ball valves are also available, togreatly improve the inherent quick-opening flow patternof these valve trim shapes. Control dampers and butterflyvalves are either flat blade or flat disk shapes and do nothave selectable characterized flow patterns like controlvalves do. In the case of control dampers, about allthat can be done to reduce non-linear air flow throughthe damper is to down-size it to create a high pressuredrop, which may create other complications or costs,especially in outside air ducts and large ducts withoutroom for transitions. Where linear control of an air streamis important, air valves can be used which are availablewith characterized flow patterns.Modulation of a heat exchange process is a commonautomatic control application. The following exampleapplies for most types of heat exchanges including shelland tube, tube and fin, etc., and includes all HVAC air coils.The heat transfer characteristics of a heat exchanger canbe likened to a ‘quick opening valve’ since the incrementalchange in heat transfer for the first fraction of fluid inputis much higher than the last fraction. Controlling flowlinearly through a heat exchanger will yield a non-linearoutput with associated control problems, especially tuningissues. In this case, the non-linear flow characteristic ofa control valve is deliberately used to improve processcontrol. The standard approach to correct this is to usea control valve with a flow characteristic that is a mirrorimage of the inherent heat exchanger performance (equalpercentage type), canceling this inherent phenomenon sothe overall control effect is nearly linear. This examplenot only illustrates how linearity is important to a controlsystem, but also points out that for heat exchangerapplications, the control valve selected should almostalways be the equal percentage type.22.7 VALVES AND DAMPERS22.7.1 Valve and Damper SelectionSelection criteria usually includes a minimumpressure drop for proper authority and best linearityover the process.22.7.1.1 Valve SizingThe typical sizing procedure for a hydronic controlvalve is 5 psig wide open pressure drop. The followingexplains the reason for this. The underlying principle forreasonable control of a heat transfer coil is for the control

CONTROL SYSTEMS 587Figure 22.8 Control Valve Characteristicsvalve wide open pressure drop to be at least as high asthe coil it controls. So, if an air handler coil is sized for a5 psig pressure drop at full flow, then the control valve at5 psig full flow pressure drop would be appropriate. It iscommon practice to select HVAC water coils at 10 ft. w.c.or so wide open pressure drop, which equates to 10/2.31= 4.3 psig. Thus the 5-psi valve pressure drop conventionis a reflection of the coil sizing convention. An extensionof this logic would be that if the coil were selected at 1psig wide open pressure drop, then the control valvesizing criteria could also be reduced. This is in fact thecase, although seldom done in practice since coils selectedat extremely low pressure drop have other, new issues,namely reduced velocity and laminar flow heat transferdegradation at reduced flows.Editorial Comment: The practice of adding systemresistance to achieve good control is counterintuitiveand definitely an opportunity for improvement in theindustry, because adding circuit resistance to anyfluid handling system increases the system energyrequirements.22.7.1.2 Leak-By and Close OffWhen specifying control actuators, specificationsand close attention are important to achieve reliable closeoffperformance. This is true for both valves and dampers,especially those with marginal actuator close-off ratings,excessive system pressure, large damper sections, metalseatedvalves, undercut butterflyvalves, and actuators without apositive seating mechanism toimpart a residual tight seating force.Spring-return pneumatic actuatorsare forgiving in this sense becausethey inherently provide residualseating force. Some electronicactuators have mechanisms toprovide ample seating force,but others rely on simple traveladjustments that define the openand closed positions—these are veryundesirable since the opportunityfor internal leak by, with subsequentheat/cool overlap, is high. Ballvalves can be used for modulatingservice if they have characterizedseats, and are inherently tightseating.Actuators that are onlymarginally strong enough to closeoff against system flow can rob thesystem of efficiency over time as system pressures change,valve stems bind, damper axles stiffen, etc. Determiningsystem needs and requiring close off ratings well inexcess of this (e.g. at least 50% more) is good practice forsustainable operation.22.7.1.3 Damper SizingJust like valves, down-sizing dampers will improvecontrol at the expense of raising system pressure. Inpractice, dampers are often left duct-sized, even thoughresulting control is poor. There are several reasons forthis:• Dampers are relatively cheap, compared to thetransition costs for a duct fabricator.• Dampers are often large, and transitions take upspace that may not be available.• At outside air intakes, a down-sized damper cancause rain or snow entrainment.For air flows control by dampers, other thanHVAC, proper sizing will yield more linear control and isrecommended where practical. For control purposes, theopposed blade damper is normally used, since its aperturesize and overall resistance varies the most directly withtravel. Conversely, “parallel blade” or round dampersare highly non-linear in nature and hard to control unlessdrastically down-sized, or used for two position controlonly.

588 ENERGY MANAGEMENT HANDBOOK22.7.1.4 Other Damper ConsiderationsLarge damper sections are often problematic. Forcost reasons, there is often a desire to use fewer, largeractuators and link the dampers together so an adjacentactuator is driven by another damper, not an actuator. Inpractice, this can easily result in one end of a long sectionsubstantially open even as the actuator-end is closed. Thisis due to the fact that the damper blades and axles willtwist and stretch. Methods to prevent this undesirablecondition include multiple actuators or jack-shafting.22.7.2 Valve and Damper ActuatorsLike other instruments, actuators come in a varietyof styles and quality levels, and each has its pros andcons. Significant differences exist between manufacturersthat make generalizations and rules of thumb difficult.One thing is certain about actuators: they are a movingpart, with mechanical components, and will requiremaintenance; therefore consideration should be givento life cycle cost and maintainability. Some of the lowercost actuators are not intended to be serviced. For eachof the types listed there are serviceable and throwawayvariations, as well as spring-return/non-spring returntypes. Types of actuators include:• Pneumatic spring return• Pneumatic air-to-open/air-to-close• Electric motor/gear reduction in oil bath• Electric motor/gear reduction-open air• Electric hysterisis/stalling motor• Hydraulic• Wax motors (thermal expansion)• System powered actuators/using air or watersystem pressure as the motive force• Self-powered actuators/using a capillary bulb andbellows22.7.2.1 Actuator “Normal” (spring return) PositionFor valves and dampers, the phrases normally openor (N.O.) and normally closed (N.C.) refer to the deviceposition with no power applied, where a spring-returningmechanism exists to drive it to one position or another.Valves required to have a “fail position” are necessary inmany applications to provide an increased measure ofreliability if control power is lost. In the case of comfortheating and cooling, the choice is made by asking “upona loss of power, control signal or air pressure, would Irather have full heat, full cool, or don’t care?” In othercases, there are other operational issues like overheating,over-humidifying, etc. that should be considered. Withoutthe spring return feature, the actuator will simply remainat its last position prior to the power interruption. Springreturnactuators are more costly, so should be usedprudently, where the added cost is justified.For large pneumatic cylinder actuators, a measureof fail-safe control can be provided without theexpense of a spring system; using an air-to-open/airto-closeactuator (no spring) and a small spring returnair solenoid valve, the position of the actuator can berelatively assured on power loss, as long as compressedair remains available.22.8 INSTRUMENT ACCURACY,REPEATABILITY, AND DRIFT22.8.1 IntroductionLike anything else, there are different grades ofinstruments with corresponding costs, so the task of thespecifier is to separate the needs from the wants, and tobalance the performance with the costs. With an awarenessof some of the basic considerations and of instrumentgrades and selection criteria, good decisions are usuallyevident. Leaving the instrument selection entirely up tothe vendor may or may not be the best approach. To theextent that the ‘standard offering’ instrument portfolio hasgood performance, this can save money, however a reviewof the proposed instruments is advised just to be sure.When reviewing product literature for instrumentation,like any other equipment, it is often as important whatis not said, as what is said on a component specificationsheet. For example, if long term drift is not mentioned,ask yourself “why is that?”22.8.2 AccuracyIn the realm of instrument calibration and accuracyspecifications, there are two important terms: “percentof reading” and “percent of span.” It is not enough tosay “+/- 5%” when specifying a calibration toleranceor instrument accuracy rating. The tolerance should bestated as either +/- xx degF, +/- xx psi, etc., +/- xx% ofreading, or +/- xx% of span. Usually the instrumentsthat are rated in terms of +/-% of reading are the higherquality instruments.The following example illustrates the importantdifference between “% of reading” and “% of span”concepts. Consider a 0-100 psig (span) transmitter that isindicating 10 psig (reading).Accuracy Spec “+/- 5% of reading,”Acceptable Range 9.5-10.5 psigAccuracy Spec “+/- 5% of span,”Acceptable Range 5-15 psig

CONTROL SYSTEMS 589Figure 22.9. Suggested Tolerances For Commercial Instruments.22.8.3 RepeatabilityRepeatability is self-defining; it is the ability ofan instrument or process to faithfully repeat itself,given identical conditions. For instrumentation, thisis synonymous with precision, and the mark of bettergrade instruments. Regarding instrumentation, ageneral statement is that accuracy can be adjusted, butrepeatability is a function of the instrument qualityand cannot be changed. Repeatability is determined, inlarge part, by the stability of the output in the face ofenvironmental changes, such as ambient temperatureeffect, voltage fluctuation, pressure fluctuation, etc.Instruments whose readings are susceptible to changes inambient temperature can be very problematic if locatedin areas where the temperature is expected to change,however this specification (and the cost to mitigate itwith higher quality instruments) is of much less concernif located in areas of constant temperature. Commercialgrade pneumatic instruments are usually susceptibleto problems from ambient temperature changes, sincethermal expansion changes the volume of tubing, size oforifices, etc. Some lower grade electronic components arealso affected by ambient temperature changes.22.8.4 DriftDrift is an undesirable but inevitable quality forany instrument. Long-term drift is attributable to manyfactors including normal degradation from age. A generalrule of thumb is that the long term drift, from all sourcescombined, leave the instrument reading within reasonfor a period of five (5) years, to reduce the need for O/Mactivities and constant maintenance of the device, leavingit to serve instead of being served by the facility.Auto-Zero Feature: Some devices, such as stackgas sensors and very-low range differential pressuretransmitters, periodically re-establish the zero point oftheir output span by simultaneously providing a zeroinput condition and automatically adjusting the outputto a zero value.22.9 BASIC CONTROL BLOCK DIAGRAMS22.9.1 IntroductionBesides control technology and control mode choices,a basic consideration of control strategy is whether it isopen or closed loop.22.9.2 Closed Loop ControlThe controlled system impacts the measuredvariable, and process measurement provides feedback tothe controller.Examples of Closed Loop Control:• Room thermostat controls a heating water valve, toregulate heat to that room.• Leaving water temperature sensor for a heatexchanger controls the steam inlet valve to that heatexchanger.22.9.3 Open Loop ControlAn open-loop control system is characterized as onewhose output has no impact on the measured variable,and so any form of process feedback is impossible.Examples of Open Loop Control:• Turning off the building heating boilers in Apriland leaving them off until September. A commoncontrol strategy, this follows the common sensenotion that boilers should not be needed in summer,but does not address spring and fall vary well andcan lead to both discomfort and energy loss whenweather is unseasonably cold or warm, usually nearthe calendar cutoff dates. This is open-loop becausethe outside air temperature is unaffected by whetheror not the boiler is on. This is also open-loop to thebuilding itself, because call for heat feedback willnot be heard.

590 ENERGY MANAGEMENT HANDBOOKFigure 22.10 Closed Loop Control• Automatic reset of hot watertemperature from outside air.A common control strategy, thisprovides general compensationbased on the common sense notionthat ‘the colder it gets outside themore heat we’ll need,’ howeveris open-loop since the outside airtemperature is unaffected by watertemperature.• Starting the building HVACsystem one hour before occupancy.A common strategy, this followsthe common sense notion thatthe building will need some timeto warm up (or cool down) afterbeing off all night, or all weekend.It provides general compensationfor the thermal lag in the buildingmass, but is open-loop becausevariations in actual time requiredin different seasons is not considered. For exampleit may take (4) hours after a long and cold weekendbut only a half-hour after a single night in spring,but the controller is blind to these facts.• Thermostat in room 1 controls the hot water controlvalve serving room 2. In this example, closed loopfeedback control is changed into open loop controldue to a design or installation error.Figure 22.11 Open Loop Control22.10 KEY ELEMENTS OF SUCCESSFULLYAPPLIED AUTOMATIC CONTROLS22.10.1 Examples of Good andBad Control ApplicationsExample.• BAD. Two thermostats in an exterior room; oneserving the perimeter heat and the other serving

CONTROL SYSTEMS 591Figure 22.12 Controls Application Checklist

592 ENERGY MANAGEMENT HANDBOOKthe VAV cooling box overhead air distribution.• GOOD. One thermostat in a perimeter room;sequencing the VAV box and perimeter heat,with a dead band in between to preventoverlap.22.10.2 Examples of Good and BadSystem ControllabilityExample.• BAD. A single step controller on a 50-tonHVAC package unit that has one compressorattempting to maintain a constant 55 degreedischarge temperature. Excessive equipmentcycling will result, regardless of controls, andthe consequence of poor control and likelypremature equipment failure.• GOOD. A four-step controller on a 50-ton HVACpackage unit that has four compressors, where12.5 tons represents a typical part load day forthe building. Cut-in and cut-out set wider thanthe temperature change of each stage, to preventshort-cycling. Minimum on-off times furthersafeguard equipment from short-cycle damage.22.10.3 Examples of User-FriendlyControl Design Features• Match the system complexity to the user’s levelof sophistication, or their willingness to learn.If it’s too complex, they’ll probably just unplugit 6-months after start-up. Training will help toraise skill levels and avoid over-simplification.Be patient, especially for anything complex.Any type of measurement that allows the userto see the savings created by the new controlsystem will help spark interest and encourageownership and buy-in.• Control diagrams at the control panel as a handyresource.• User adjustment for comfort control applications.• Devices located in accessible locations forrepairs and calibration.• Redundant Visual Indicators by key inputs andoutputs.• Valve position indicators that are visible fromthe floor.• Temperature and pressure gages adjacent to keymeasurement points.• Man machine interface provision at the processlocation.• Override Provisions for key output points toaccommodate emergencies without disconnectingthings.• Standard parts that are readily available.• Sensors and Transmitters with long-termstability for minimal need for calibration.• Arrange control wiring/tubing with troubleshootingin mind.• Power on-off switches and air shut-off valves,in key locations to assist the service person.• 3-valve manifolds at flow meters to allow zerocalibration.• Drain and Vent valves at fluid transmitters forroutine calibration.• Minimum on-times and off-times to preventshort cycling.• Make normal-use operator parametersadjustable, and critical parameters non adjustable,to allow user interaction and reduceinadvertent changes for the worse.• Plug-In relays for easy replacement.• Relays with LED lights to indicate at a glancewhen the relay is on or off.• Separate control loops that are interrelatedin a process designed with coordinated andlinked set points, to prevent adjustment of onecontroller setting from inadvertently impactinganother one.• Control system logic design to identify faultysensor readings, to prevent faulty controlaction.• In general, try to provide a control design thatwill be accepted, and will last.22.11 OPERATIONS AND MAINTENANCEFor the control design to stand the test of time (e.g.to be sustainable), it must be a good fit for the Owner.22.12 EXPECTED LIFE OF CONTROL EQUIPMENTBe realistic about how long things will last. Nothinglasts forever. For example, a control project based on a 10-year life will need to include cost of repairs to be realistic.Note for the table below that life spans shown vary bymanufacturer and grade. This information is offeredas a prompt for realistic life cycle cost estimating whencontrols are applied.

CONTROL SYSTEMS 593Figure 22.13 User-Friendly Control System Checklist

594 ENERGY MANAGEMENT HANDBOOKFigure 22.14 Typical Control Equipment Life Spans22.13 BASIC ENERGY-SAVINGCONTROL APPLICATIONSCommon themes throughout the following list ofapplications:• Strive to satisfy most of the people most of thetime, but not all the people all the time (80-20Rule)• Try not to run any equipment continuously, oroff-season• Avoid heating, cooling, and lighting areas thatare unoccupied• “Just enough” air pressure, water pressure, etc.to satisfy the point of use• “Just enough” heating and cooling, asdetermined by the point of use• “Just enough” ventilation air for the people andto make up the exhaust needs• Eliminate simultaneous heating andcooling wherever possible; minimize whereunavoidable22.14 ADVANCED ENERGY-SAVINGCONTROL APPLICATIONS(See Figures 22.17 and 22.18.)22.15 FACILITIES OPERATIONSCONTROL APPLICATIONSAutomatic Controls, especially DDC controls, can bea very valuable tool for facilities personnel. Computerizedmaintenance management, early detection, remoteservicing, automatic notification, trends and logs, andother features can be used to improve facility operationalquality.Figure 22.15 Basic Lighting Control Applications

CONTROL SYSTEMS 595Figure 22.16 Basic HVAC Control Applications

596 ENERGY MANAGEMENT HANDBOOKFigure 22.16 Basic HVAC Control Applications (Concluded)Figure 22.17 Advanced Lighting Control Applications

CONTROL SYSTEMS 597Figure 22.18 Advanced HVAC Control Applications(Continued)

598 ENERGY MANAGEMENT HANDBOOKFigure 22.18 Advanced HVAC Control Applications(Continued)

CONTROL SYSTEMS 599Figure 22.18 Advanced HVAC Control Applications(Concluded)

600 ENERGY MANAGEMENT HANDBOOKFigure 22.19 Facilities Operations Control Applications

CONTROL SYSTEMS 60122.16 CONTROL SYSTEM APPLICATIONPITFALLS TO AVOIDThe following are some common pitfalls to avoid inapplying Automatic Controls.22.17 COSTS AND BENEFITS OFAUTOMATIC CONTROLAutomatic Controls are unique in that they oftenprovide both tangible and intangible costs and benefits.Tangible Benefits, like other energy-related projects,include energy savings and demand savings. Intangiblebenefits are those that are difficult to quantify or predict.Some of these only apply to Digital Control systems.22.18 ESTIMATING SAVINGS FROM APPLIEDAUTOMATIC CONTROL SYSTEMS22.18.1 IntroductionEconomic barriers are among the greatest obstaclefor the Energy Professional, and Automatic Controls areno exception. Acceptance of the cost of new technologyalmost always requires economic justification, and rightlyso. Finding ways to predict and then demonstrate thesavings is necessary should be the goal of any AutomaticControls project, since this will build momentum forsubsequent projects.While it is often easy to visualize that savings willoccur from Automatic Control improvements, estimatingthem can be a very daunting and intimidating challenge.This presents a dilemma to the Energy Professionalsince cost justification is almost always an expectation.Projects that have merit but defy quantification may beoverlooked.Done accurately, the cost saving calculations can belaborious and expensive, posing a barrier to otherwiseviable projects. One approach is to produce reasonableestimates using abbreviated estimating methods or‘rules of thumb’ where possible. As with all estimates,being conservative is a key to success so that projectperformance is seen to under-sell and over-deliver. Evenwhen not a contract requirement to guarantee savings,there is always an expectation that the estimated costsand savings come to pass; further, the credibility of theenergy professional is built largely on the accuracy ofthese estimates. Bearing in mind that there will alwaysbe uncertainties and uncontrolled variables at work, it isusually good practice to de-rate the calculated savings.By artificially reducing savings (de-rating) and artificiallyincreasing projected costs (contingency allowance), twothings happen.Effect of de-rating project estimated savings:• The odds of project performance exceedingexpectations increases.• Projects with marginal Return on Investment(ROI) look worse and may be eliminated.The applied de-rate will depend on the level ofuncertainty. A high degree of uncertainty suggests theneed for a higher de-rate. A value between 20-30% isFigure 22.20 Control System Pitfalls

602 ENERGY MANAGEMENT HANDBOOKFigure 22.21 Control System Costs and Benefitssuggested for most applications, although there are caseswhere no de-rate is needed.Examples:• Easy to Quantify. A lighting replacement projectwith 24-7 operation need not be de-rated at allsince there are few, if any, uncertainties. This iseasy to quantify, since only one parameter haschanged (light fixture energy efficiency).• Hard to Quantify. An Automatic Controlproject that includes multiple control systemimprovements implemented at the same time,such as variable pumping, free cooling modes,supply air reset, boiler lock-out, morning warmup,and new quarter-turn (no leak by) terminalunit control valves. This adds uncertainty forwhat savings comes from individual measures.

CONTROL SYSTEMS 603From a pure simplicity standpoint, a projectconsisting of several control improvements couldbe implemented one at a time, with six monthsof post-project Measurement and Verification(M&V), therefore changing only one thing ata time. This would have the clear advantageof knowing what savings occurred from whatmeasure, but would probably not be done inpractice due to the protracted length of time forthe entire project to be implemented, as well aslost savings from delayed implementation.22.18.2 Replacement CostsA common practice for existing facilities is to proposeequipment replacement using energy efficiency asjustification. In general, it is easy to justify the differentialcost of upgrades to higher efficiency equipment, butoften impossible to justify the entire replacement projecton energy savings alone. Burdening the project cost withunrelated expenses, such as equipment replacement thatwas due anyway, makes the payback look worse andcreates an unfair perception of long paybacks. Wheneverpossible, energy improvement expenses should befairly separated from normal replacement project costs.Equipment that is near the end of its useful life should bea planned replacement expense, regardless of the desireto reduce energy costs. If replaced early, the remainingvalue of the equipment may be appropriately ‘charged’ tothe energy project, but not the entire project cost since thiswould need to be done anyway.22.18.3 Barriers to quantifying savingsfrom controls changes• Control parameters are almost always useradjustableand will usually be fine-tuned duringthe life of the project, including the post-projectmeasurement period.• Control algorithms often include multiplevariables which do not act independently.Consequently the effects on energy consumptionoften defy isolating and quantifying separately.An uncertainty is whether one measure willinteract with another. In many cases, theaggregate savings may be less than the sum ofthe parts, in which case the savings would beover-stated if calculated independently.• Control system improvements that improvequality, comfort, ventilation, or other featuresfrom some deficient state may actually increaseenergy use in some ways, eroding the overallsavings of the measure itself.22.18.4 MethodologyEstablish realistic baselines. Identify variables andtheir interactions. Identify competing or complementingprocesses that will subtract from measure savings.Reduce the number variables to simplify analysis. Treatuncertainties with contingencies, by either inflatingthe baseline or de-rating the savings. With the baselineestablished, use experience to evaluate calculated savingsas percent of total expenses. Use all these indicators assanity checks to avoid over-stating savings. After projectcompletion, get after-the-fact measurements wherepossible to compare actual to estimated savings; collectingthis real data will allow improved estimating and reducedcontingencies over time.22.18.5 Quantified Savings Examples.The following are some examples of how to generallyquantify savings from Automatic Controls measures.Many of the methods use rules of thumb, and assumptionsto simplify the work. More accurate results can be had byusing a rigorous computer model, but since time is moneythe luxury of detailed models is not always available.Without an excessive investment in time, the methodsshown below will yield results close enough to identifyprobable savings and to tell if the measure is worthwhileor not. While this is by no means a complete listing, itis hoped to convey the general method of abbreviatedenergy accounting for multi-faceted processes whenimprovements are proposed. Some of the examples showstraight forward benefits, and others show benefits withparasitic losses or competing processes. The first examplebelow is Condenser Water Reset, and is discussed indetail. Other abbreviated solutions are provided in thetable that follows.22.18.6 Detailed Example: Condenser Water Reset.Note: The cooling tower term “approach” is thedifference between leaving condenser water temperatureand ambient wet bulb temperature.The variables:The chiller kW/ton varies according to load, andthe cooling tower kW/ton varies according to wet bulbtemperature and approach. The cooling tower fan energyvaries according to chiller ton hours, tower capacity, andcondenser temperature set point with respect to ambientwet bulb. Overall savings varies with annual ton-hours,chiller efficiency, the ratio of chiller-to-tower efficiency(kW/ton), chiller low limit in accepting colder condenserwater, and coincident wet bulb temperatures. The wetbulb variable means that any rules of thumb developedfor annual savings would be area/climate specific.

604 ENERGY MANAGEMENT HANDBOOKAlthough the chiller energy is reduced directly asthe condenser temperature is lowered (1-1.5% per degree),the cooling tower efficiency decreases at lower approachtemperatures, and eventually a point is reached wherethe savings in chiller energy are met by the added coolingtower horsepower. This break-even point varies with thecooling tower sizing, and towers with high hp/ton (smallbox, big fan) will hit the wall sooner than more generouslysized cooling towers. The cooling tower energy penaltyvaries depending upon the approach temperature andrises sharply for each degree below 7 degrees F approach.Some common values of the energy penalty, in percentincrease per degree lowered, are as follows:12 degF Approach 7.3%11 degF Approach 7.8%10 degF Approach 8.3%9 degF Approach 8.9%8 degF Approach 9.7%7 degF Approach 10.6%8.9% Average cooling tower fanenergy penalty per degree lowered.The chillers place a physical limit on this controlprocess; some can take colder condenser water thanothers. Centrifugal chillers can accept from 55 to 70degrees F entering condenser water, depending uponthe manufacturer. Screw chillers are generally limited to70 degF entering condenser water. Many reciprocatingchillers are limited to 70 degrees F, but some can operate atlower temperatures. IMPORTANT: For each application,the limits of the machinery need to be identified and thecontrol system must stay within those limits to assureno damage or detriment is done to the chillers.Refer to the table below. For a chiller with 0.5kW/ton efficiency, paired with a 0.07 kW/ton cooling tower,a proposed 5 degree reduction (from 12 to 7 degrees Fapproach), would yield worse energy consumption thanleaving it at 12 degrees F, due to the high cooling tower fanenergy penalty. This same example, with all things equalexcept a 0.04 kW/ton cooling tower, would save 0.35%per degree lower overall cooling energy. This exampleshows that during operation at or near design wet bulbconditions, most or all of the theoretical chiller savingsfrom condenser water reset will usually be negated bythe added cooling tower energy use, unless the coolingtower is very efficient (0.04 kW/ton or less).While the energy savings near design conditionsmay be marginal, most of the chiller hours will be atwet bulb conditions that are more favorable, and thisis where the energy savings are attained. The controlstrategy capitalizes on this by continuously adjustingthe operating set point based on ambient wet bulb.For those chiller operating hours when the wet bulb issignificantly lower than design and the cooling towercan easily produce colder water with little tower energypenalty, the savings will be much more pronounced andcloser to the theoretical 1-1.5% per degree. Of course, thechiller load during these shoulder seasons or overnightperiods are usually lower than maximum, so knowingthe chiller load profile and coincident wet bulb is key toquantifying savings.Chillers that can accept very cold condenserwater and have cooling needs coincident with reducedwet bulb temperatures such as in the SouthwesternUS, when it is easy to achieve colder condenser watertemperatures, can achieve 10-15% annual overall coolingenergy savings using condenser water reset, comparedto a fixed temperature set point of 70 degrees F.The following diagram illustrates the pronouncedeffect the system cooling tower has upon the overallcooling savings of the condenser water reset controlroutine. Remember, without the cooling tower fan energyexpense, the savings from the chiller would be 1-1.5% perdegree lowered.The following diagram shows how the relative hprating of the cooling tower (kW/ton) affects the optimalcooling tower leaving water temperature set point.The “approach” value is a parameter of the optimizedcondenser temperature calculation, and should be selectedbased on the cooling tower in use for best economy.The “approach” value is a parameter of theoptimized condenser temperature calculation, and shouldbe selected based on the cooling tower in use for besteconomy. Suggested values of cooling tower approach forbest overall cooling efficiency (kW/ton) are shown below.By using the “Ratio” instead of specific combinations,Figure 22.22 Cooling Tower Effect on Condenser WaterReset Savings

CONTROL SYSTEMS 605Figure 22.23 Cooling Tower Effect on OptimumCondenser Water Temperaturethis information can be applied to any combinationof chiller and cooling tower. This is the value insertedin the sequence “…optimum cooling tower set pointshall be equal to the calculated wet bulb temperatureplus approach…,” and provides further evidence of theimportance of amply sized, low HP cooling towers.EXAMPLES OF QUANTIFYING SAVINGS FROMAUTOMATIC CONTROL MEASURES(See Figure 22.24.)22.19 CONCLUSION AND FURTHER STUDYAutomatic controls are useful for basic regulation,and quality control of processes and environments.They can also be leveraged for energy savings throughoptimization. Properly applied, these systems are reliableand cost effective.Let us now return to the Chapter Intent, so see if theobjectives were met. The stated purpose of this chapterwas to focus on the application of Automatic Controlsas a tool to achieve energy savings. The reader shouldreview the titles of each section, reflect on the key topicsthey have taken away, and decide if the stated objectivewas met. If you have gained insight into how automaticcontrols can help you achieve your energy goals, haveincreased your comfort level with the subject matter, knowimportant questions to ask the vendors, and are eager toput these systems to work for you, then this chapter hasbeen worthwhile. If you do not feel the objective wasmet for you, consider re-reading the material, seekingout supplemental material, or contacting the author forfurther discussion. Very often a second text on the samesubject will amplify the effect of the first text, by reinforcingconcepts with different examples and descriptions.This chapter was necessarily made brief, and doesnot pretend to be a complete treatment of the subject.The following advanced topics were not addressed, andare listed for the interested reader to pursue throughadditional study.• Addressable I/O devices, network communications,and equipment interfacing technologies.• Boolean Logic.• Cascade Control.• Computerized Maintenance Management Systems(CMMS).• Control Loop Interaction.• Ergonomic Considerations. Trends, logs, reports,alarms, graphics.• Failure Modes, Mitigation, and Fault Tolerance.• Feed Forward Control.• Fuzzy Logic.• IT Security.• Ladder Logic.• Loop Tuning. Step changes, dead time, stability,response time.• Measurement and Verification.• Open Protocols, XML, Gateways, and Translatorsfor DDC networking.• SAMA logic diagrams, multi-element industrialcontrol methods.• SCADA systems.• Self-Powered Controls.• Wireless Controls.

606 ENERGY MANAGEMENT HANDBOOKFigure 22.24 Quantifying Control System Savings

CONTROL SYSTEMS 607Figure 22.24 Quantifying Control System Savings (Continued)

608 ENERGY MANAGEMENT HANDBOOKFigure 22.24 Quantifying Control System Savings (Continued)







CONTROL SYSTEMS 615Figure 22.24 Quantifying Control System Savings (Concluded)

616 ENERGY MANAGEMENT HANDBOOK22.20 GLOSSARY OF TERMSAccuracy: A performance measurement of an instrumentto produce a value equal to the actual value.Regarding instrumentation, a general statement isthat accuracy can be adjusted, but repeatability is afunction of the instrument.Addressable Devices: Also called smart devices. A termused to describe equipment that can be given aunique identifying address and controlled directlyby a digital control system, often in reference tocommodity-type equipment that exists in quantityin a facility. By having unique addresses, multipledevices can share a common communication busand reduce the volume of wiring.Analog: Variable (input or output), contrasting todiscrete.Bi-Metal: A basic temperature transducer formed byjoining two metals with different thermal expansionproperties. Common shapes are coiled ribbons (thatturn upon a temperature change), bars (that bendupon a temperature change), and disks (that warpand ‘snap’ upon a temperature change). Movementis predictable and repeatable, and is integral tomany temperature control instruments.Binary: Synonymous with on-off or discrete (input oroutput), contrasting to analog.Boolean Logic: A logical technique used to formulateprecise queries using true-false connectors or‘operators’ between concepts. The primary operatorsare AND, OR, and NOT. Words or concepts joinedwith these operators, and parentheses are used toorganize the sequence groups of concepts. The truefalsenature of Boolean Logic makes it compatiblewith binary logic used in digital computers, since aTRUE or FALSE result can be easily represented bya 0 or 1. Named after George Boole.Building Automation System (BAS):BAS is the term given to a computerized automaticcontrol system when the primary focus: “…is onautomating as much as possible to save labor.” [7:pp 38.10]. Building Automation System (BAS) iscurrently the most common term used to describecomputerized control in buildings that provide oneor more of the functions:Energy Management System (EMS)Facility Management System (FMS)Energy Monitoring and Control (EMCS)Cascade Control: Also called Master-Sub Master. This is acombination of two controllers in series with a relatedmeasured variable. The output of the first controllerbecomes an input to the second controller which canthen amplify it. Often, the output of the first (master)becomes the set point of the second controller (submaster).Useful in improving modulating control ofsystems with very slow process response.Cavitation: A problem phenomenon with fluid flow,often with control valves or pumps, where the localpressure drop is sufficiently high to cause temporaryboiling of the fluid. When the pressure is sufficientlyregained, the vapor bubbles created by boilingcollapse and create a shock. If this occurs within thevicinity of fluid handling apparatus it can damagethe equipment.Closed Loop: A control system which includes relatedprocess feedback input(s) and controlled deviceoutput(s), collectively forming a regulating process.The defining characteristic is the feedback inputpoint that senses changes in the process and theeffect of the controlled device, to close the loop ofcommunication.Control Loop: The general term for a collection of controlsystem components used to regulate a process thatincludes as a minimum a Controller, Set Point, ControlElement (Controlled Device), Associated Process,Process Measurement (Measured Variable).Controlled Device: The manipulated device respondingto the controller output, that then impacts theprocess itself. The item being manipulated by thecontroller.Controller: A device that compares set point to measuredvalues, and determines an appropriate outputresponse.Cut In: A parameter of two-position control. The cut-invalue is where the controlled variable is sufficientlybeyond the set point for the controlled equipment tobe turned on (cut-in).Cut Out: A parameter of two-position control. The cut-outvalue is where the controlled variable is sufficientlybeyond the set point for the controlled equipment tobe turned off (cut-out).Dead Band: Also called zero energy dead band. Refers toa deliberate gap in control span between sequenced,usually conflicting, processes, to avoid controloverlap and subsequent operational or energyconsumption issues.Example:A single temperature controller that sequences bothheating and cooling equipment to a common duct orspace may use a Dead Band to prevent simultaneousheating and cooling.Dead Time: The time between a change in the processinput and when that change is felt in the downstreamprocess measurement. Often a function of the time it

CONTROL SYSTEMS 617takes material to flow from one point to another.De-Bouncing: A type of discrete input signal conditioningused for digital control systems. Mechanical snapactingcontacts actually open and close many timesand can trigger control output problems whenmonitored by a high speed digital circuit. The debouncingsignal conditioning has, as its purpose, tofilter out the bouncing ‘noise’ so the controller seesa simple open-or-closed state.Demand Limiting: A control strategy with the purpose ofreducing electrical demand, not necessarily energy.This is applied to facilities with utility demandcharges, and usually is designed to reduce peak (max)demand thereby reducing utility demand charges.Derivative Control Mode: Also called rate or anticipatorycontrol, and is used to increase response time aftera disturbance or step change. For a given rate ofchange of error, there is a unique value of controlleroutput. This control mode reacts to the rate ofchange of error, not the magnitude of error. Thismode cannot be used alone since there is no outputwhen the error is zero.Direct Acting: Control Action that increases its output asthe measured variable increases above set point.Discrete: Synonymous with on-off or binary (input oroutput). Characterized by having two possiblestates. Typical example is an open-close contactinput, or relay output.Dry Contact: A discrete signal, output or input, that ismade with a contact closure such as a mechanicalrelay or switch, that has no voltage at the terminalsand depends upon voltage from the initiatingcircuit to read its two-state resistance. Contrasting,when the device has a voltage at the terminals (asif to light a light bulb when the contacts close) it istermed the wetting voltage.Duty Cycling: A demand-limiting strategy wheremultiple electric points of use, often motors, whichotherwise run continuously, are controlled to runsome fraction of time, e.g. 40 minutes on and 20minutes off. By coordinating the run and off times ofmultiple items, the aggregate utility electric demandcan be reduced, to generate demand savings.Energy Monitoring and Control System (EMCS): SeeBuilding Automation SystemEnergy Management System (EMS): EMS is the termgiven to a computerized automatic control systemwhen the primary focus: “…is on saving energyby specific automatic control programs.“ [7: pp38.10,11]. See also Building Automation System.Facility Management System (FMS): FMS is the termgiven to a computerized automatic control systemwhen the primary focus: “…goes beyond HVACcontrols and/or beyond a single building, such asincluding fire, security, or manufacturing systems.”[7: pp 38.11]. See also See Building AutomationSystem.Feedback: The measurement of the process output thatis returned to the process controller that is alsoinfluenced by the controlled device. Feedback isthe differentiating factor for a closed-loop controlsystem, contrasted to an open-loop control system.Floating Control Mode: A form of discrete (on-off) controlwith a null position whereby the controller holds thelast output when set point is achieved, rather thanreturning to zero output.Gain: A tuning constant that multiplies an input oroutput parameter, usually to increase sensitivityand improve control.GUI: Graphical User Interface. The software that overlaysthe machine coding, adding the user-friendlinesslook and feel to a digital control system workstation.Hunting: Chronic, repeating oscillation (overshoot andundershoot) of a modulating control system, usuallyindicating a poorly tuned control loop.Hysteresis: A physical phenomenon best explained asinertia or a body’s tendency to stay at rest. In controlsystems this can affect input and output instruments,and controlled devices (valves and dampers). Thehysteresis effect will require a different value ofcontrol to cause a change or movement, dependingupon whether approaching the value from higheror lower values. Generally, hysteresis is an issuefor heavier controlled devices or with minute I/Ochanges.Indicating Transmitter: A transmitter with an auxiliarydisplay or meter to provide local indication of themeasurement and/or output signal.Input/Output Points (I/O): The general term for theidentifiable instruments (points) used to relay thecontrol system information into (input points) andout from (output points) the controller, linking thecontroller to the actual process.Integral Control Mode: Also called reset control, and isused to return the process to zero error by producingan output any time the error is other than zero.Significant is that the output increases over time andis useful to return the process to a zero error state.Interlock: A control strategy that requires one process(discrete or analog) to depend upon the on-off stateof a separate but related process.Examples:— A boiler firing interlocked with combustion air

618 ENERGY MANAGEMENT HANDBOOKdamper requires the damper to be proven openprior to firing. This interlock would be for safetyand is normally “hard-wired,” so as not to dependupon any software intervention.— A cooling coil control routine interlocked withsupply fan operation requires the supply fan tobe proven running prior to engaging in controlof the cooling coil. This interlock would be fornormal control since the feedback measurement isdownstream of the cooling coil and would not besensed without air flow, and is usually a softwareinterlock.Ladder Diagram: Also called an elementary wiringdiagram. This is an electrical circuit representationwhere the high and low voltage terminals (120VAC,24VAC, 24VDC, etc) are shown as vertical lines onopposite sides of the page. Each circuit sharing thispower source, with its control contacts and load,becomes a “rung” on the ladder.M&V: Measurement and VerificationMaster-Sub Master Control: See “Cascade Control”Measured Variable: The measurement representative ofthe process that is the basis for controller action,relative to set point.Measurement and Verification (M&V): The after-thefactactivities that are used to verify whether and towhat extent energy savings have occurred, to gagethe actual energy savings and cost savings benefitof a project. Often, these are compared to estimatedsavings. In some cases, M&V is part of a contractstipulation for guaranteed savings and may formthe basis of payment from one party to another.Depth and rigor of these activities varies dependingupon need for accuracy and available funds.Minimum Position: An adjustment parameter ofcontrolled devices, referring to the lowest levelof control action allowed, regardless of furtherreduction in controller signal.Examples:Minimum fan speedMinimum damper positionModulating: See also “Analog.” Varying position inresponse to need, contrasting to two-position.Most-open Valve: A control algorithm that senses damperor valve positions at the various points of use todetermine demand. This strategy serves the mostdemanding area, but strives to provide just enoughprocess media to satisfy it, thereby reducing overallsystem energy use. Similar control routine appliesto “most open damper.”Open Loop Control: The controlled system has noimpact on the measured variable. No amount ofchange in the controlled variable or in the output ofthe controller will cause a change in the measuredvariable.Optimization: Actively monitoring and controlling eachof the pertinent parameters of a process to maximizeproductivity and/or to minimize energy use. Oftenrequires control of multiple dynamic variables.Optimization can apply to single equipment itemsor large systems.Oscillation: Repeated over and over-shooting about thedesired set point. For two-position control, this isnormal. For modulating control, this is abnormal.See “Hunting.”Overshoot: A measure of control system response tochanging loads. After a corrective action is taken, andbefore settling out at the new level, the controlledvariable is usually driven beyond the set point as ittries to recover. The extent to which it temporarilygoes beyond the mark is the overshoot.Point - Controls context: Usually used with DigitalControl systems, each unique identifiable inputor output item. The number of system pointsrepresents the size of the system, e.g. the numberof connected instruments that can be individuallycontrolled. Most commonly used in reference tohardware items, but also applies to software points.Proportional Band: The range of error (+/-) that willcause the proportional controller output to varybetween 0-100%. As the proportional gain increases,the proportional band decreases.Proportional Control Mode: Produces a linearproportional output that is a direct response to themeasured error, configured to counteract the errorand provide basic regulation. A residual offseterror is characteristic of this control mode, e.g. atrue zero-error state can never be achieved usingproportional-only control.Proportional Offset: The amount of residual errorremaining in the process using proportional-onlycontrol.Pulse Width Modulation: An output signal conditioningstrategy that converts a modulating 0-100% outputto a percentage of on time output, to impose theproportional control onto a discrete device. Oftenused in the control of electric resistance heating orincremental control electric motor actuators.Real Time: (general context): A computer system thatresponds to inputs without delay. (Computer sciencecontext): A computer system that updates informationat the same rate it receives information.Relay: A discrete output interface device, used tocontrol a large current (via heavy contacts) with

CONTROL SYSTEMS 619a small current (the relay coil current) and/or toisolate controller low voltage power (DC) from thecontrolled equipment.Relay Logic: The term given to discrete logical control(if-then, and, or) accomplished by arranging relaycontact in series and/or parallel electrical circuits.Repeatability: A performance parameter that measuresthe ability of a process or instrument to producethe same value of output during repeated trials.Instrument repeatability is synonymous withinstrument precision.Reset or Reset Schedule: A control algorithm wherebyone set point is varied, usually linearly, betweentwo values, as a function of another analog value.Example: Resetting hot water temperature set point(HW) from outside air temperature (OA):Reset ScheduleOA HW70 12030 180Resolution: The minimum measurable value of an inputvariable or the minimum incremental change of theoutput variable. In digital control systems, this isoften a measurement criterion of the A/D or D/Atransformation, e.g. the number of steps providedand how closely it emulates true analog control.Reverse Acting: Control Action that decreases its outputas the measured variable increases above set point.RTD: Resistance Temperature Detector: A transducer thatleverages the physical properties of certain materialsthat predictably change resistance as a function oftemperature. RTD response is characterized as beingnearly linear and stable over time.SCADA: Supervisory Control and Data Acquisition.These are industrial control systems with very largepoint counts, lots going on, often widespread andremote, and often critical in nature. Thousands ofpoint systems are common. Example: Utility energyand water metering and control.SCR: Silicon Controlled Rectifier. A device used toproportionally control an electric load, often aresistance heater or motor. The rectifier is anelectronic switch so is technically an on-off device.However, SCR’s are normally controlled at veryhigh speeds in time-proportioning fashion to createa modulating effect.Sensor: The general name for a device whose purpose isto sense some media and translate the measurementinto a convenient and predictable form, oftenelectrical, pneumatic.Set Point: The desired steady state of a controlled process.The goal of the control activity.Settling Time: “…the time required to for the processcontrolloop to bring the dynamic variable back towithin the allowable range,…” [3: pp 10]Short Cycling: A dysfunction of some two-positioncontrol applications where the cut-in and cut-outsettings are too narrow or the process rate is to fast,with the result being in rapid cycling of equipment,often to the detriment of the equipment.Soft Start: Any of several methods or reducing start-updemand upon a motor, usually by reduced voltageor mechanically unloading the driven load.Step Change: Also known as a disturbance or bump, thisdescribes a sudden and significant event processoutput change. In the context of control loop tuning,a step change is a valuable testing function that willdemonstrate the ability of the tuned components toreact properly to sudden process changes withoutundue loss of control, excess drift from set point,hunting, or other control anomalies.System Capacitance: A general term that describesthe relative strength or capacity of the controlledequipment to effect a change to the processmeasured variable, and a good indicator of systemcontrollability and control mode choice. In lay terms,a low system capacitance acts like the equipmentis over-sized with the tendency to short-cycle as aresult. In technical terms this is measured by thepct process change resulting from a step change inoutput. Systems with high capacitance can usuallybe controlled with any control mode. Systems withlow capacitance are often troublesome. Systems withlow capacitance may experience short-cycling andinstability using on-off control unless excessivelywide range is used. Modulated systems with lowsystem capacitance may require careful tuning andcomplex proportional control (PI, PID) to maintainstable control.Thermistor: A semi-conductor device used to measuretemperature. Lightweight and inexpensive,thermistors are popular for low cost or non-criticalapplications such as residential or light commercialtemperature control, where minor errors and longterm drift are acceptable. Thermistor outputs are notlinear with respect to changes in temperature, andusually require some form of signal conditioning toavoid errors.Thermocouple: A basic transducer for temperaturemeasurement that uses the physical principal ofgalvanic action between dissimilar materials. Thegalvanic response is a function of temperature andso, with proper signal conditioning, can be used tomeasure temperature.

620 ENERGY MANAGEMENT HANDBOOKThrottling Range: “…the amount of change in thecontrolled variable that causes the controlled deviceto move from one extreme to the other, from fullopento full-closed.” [2: pp 1:21]Transducer: A device that performs the initial inputor output conversion of a dynamic variable intoproportional electrical or pneumatic information,often a very low level signal requiring further signalconditioning.Examples include: Thermocouple, thermistor, RTD,variable capacitance dP cell, strain gage vibrationtransducer, flow element (orifice plate, etc)Transmitter: The general name for an input device thatprovides signal conditioning from a transducer orother basic measurement signal, transforming itinto some proportional information in a useful form,often with the ability to send the information overlong distance without loss of accuracy (transmit).Tuning: Adjustment of gains and various parameters ofa dynamic control system to achieve acceptable andstable control.Two-position Control Mode: Also called On-Off control,this is the most elementary control mode. Output iseither 100% if sufficiently below set point, or 0% ifsufficiently above set point. Over and undershootis normal for this control mode. The least costly ofall control methods. Regarding two position control:“Generally, the two-position control mode is bestadapted to large scale systems with relatively slowprocess rates.” [3: pp 290].Wax Motor: A self-powered actuator whose motiveforce is thermal expansion of a wax pellet/bellowsassembly. The expansion is translated into a linearmovement to vary the position of another device.Common uses include small thermally compensateddevices such as shower mixing valves, and thermallycompensated air diffusers.Sources1. Optimization of Unit Operations, Bela Liptak, pub. Chilton BookCompany, 1987.2. “Fundamentals of HVAC Control Systems,” Steven Taylor, PE,ASHRAE, Atlanta Georgia, 2004.3. Process Control Instrumentation Technology, second ed, Curtis D.Johnson, pub. John Wiley & Sons, 1982.4. “Hierarchy of HVAC Design Needs,” David Schwaller, PE,ASHRAE Journal August 2003.5. “Productivity Benefits Due to Improved Indoor Air Quality,”NEMI, National Energy Management Institute, August 1995.6. “Quantifying The Energy Benefits Of HVAC MaintenanceTraining and Preventive Maintenance,“ AEE, Energy Engineering;Vol. 96; Issue 2; 1999.7. “Computer Applications,” 1999 ASHRAE Applications Handbook,ASHRAE, Atlanta Georgia, 1999.

CHAPTER 23ENERGY SECURITY AND RELIABILITYBRADLEY L. BRACHER, P.E., C.E.M.Great Plains Energy Consulting, Inc.Oklahoma City, OK23.1 INTRODUCTIONReliable utility services are vital to all industrial,commercial and military installations. Loss of electricity,thermal fuels, water, environmental control, or communicationssystems can bring many operations to an immediatehalt resulting in significant economic loss due tounscheduled downtime, loss of life, or threat to nationalsecurity. These services are delivered by vast, complexnetworks with many components. A small number ofdamaged components is often sufficient to disable portionsof these networks or halt operation of the entire system.These component failures can be caused by equipmentfailure, natural disaster, accidents, and sabotage.The need for security continues to grow in all areas.It has become increasingly apparent in recent years thatlaw enforcement agencies cannot provide the neededresources and personnel to protect citizens, corporations,and private property. Theft and vandalism are ever present.Terrorism, shootings, and bombings continue at alevel pace and increase in intensity. The transition froman industrial economy to an information economy isbound to be accompanied by political, social, and economicturmoil. Law enforcement does what it can to detectand prevent crime in advance but will of necessity beforever relegated to a primary role of investigating afterthe fact. Responsibility for security now, as it always has,falls upon individuals and private corporations to securetheir own well-being.Many companies raised their awareness of energysecurity issues while preparing for anticipated problemsassociated with Year 2000 (Y2K) computer problems.Managers feared widespread utility outages initiated bycomputer malfunctions. Fortunately, business and industrytook this threat seriously and acted in advance toprevent major problems. Not knowing the extent of problemsthat might occur on January 1, 2000, many facilitymanagers critically examined the utility supply systemsthey rely on for the first time, installed back-up systems,and developed contingency plans. With Y2K behind themwithout major incident, many managers have now directedtheir attention elsewhere. However, the reliabilityof utility systems is not a dead issue.Many analysts are watching the pending deregulationof the electric utility industry to see how that willaffect system reliability. Some utilities have curtailed routinemaintenance in anticipation of mandatory divestitureof assets. Work like tree trimming, system expansion,and replacement of aging equipment is deferred. Thispermits an increase of current profit and reduces futurefinancial risk that could occur if regulators do not permitutilities to fully recover stranded system costs when theyunbundle services. Wise managers are anticipating therisks now and preparing to take action to improve theenergy security of their facilities. Only time will reveal thetrue extent of these threats.Energy security is the process of assessing the riskof loss from unscheduled utility outages and developingcost effective solutions to mitigate or minimize thatrisk. This chapter will explore the vulnerable networknature of utility systems and the events or threats thatdisrupt utility services. Methods will be presented to assessthese threats and identify those most likely to resultin serious problems. Actions to counter these threats willbe introduced along with a methodology to evaluate thecost effectiveness of proposed actions. Finally, the linksbetween energy security and energy management will bediscussed.23.1.1 Principles of SecurityGeneral physical security involves four areas ofconcern: deterrence, delay, detection, and intervention.Deterrence is a way of making a potential target unattractiveto those wishing to engage in mischief or mayhem.Properly implemented deterrence leads the wouldbe attacker to believe their actions would not result in thedesired outcome, require a level of effort not commensuratewith the objective, or entail a high likelihood ofcapture. Examples of physical security include area lighting,physical barriers, visible intrusion detection systems,and the presence of guards. It is not cost effective to postguards at most utility installations like substations due tothe large number of sites. Roving security patrols on an621

622 ENERGY MANAGEMENT HANDBOOKunpredictable schedule can be effective.Delay mechanisms increase the amount of timeand effort required to accomplish an unauthorized entryand execute a criminal task. The increased effort requiresadditional planning and manpower on the part of thecriminal and thus serves as a deterrent. Examples includefences, barbed wire, razor wire, secure doors and windows,locks, and channels that restrict the flow of peopleand vehicles. A delay mechanism must add enough timeto the criminal’s task to permit security forces to arriveand intercept the intruder and is, therefore, most effectivewhen combined with detection systems.Detection systems are intrusion alarms and videomonitoring in both the visible and infrared spectrums.These systems alert security forces to the presence ofintruders and allow them to respond in a manner thatbrings the intrusion to an end with a minimum of lossor damage. The time required to penetrate the facility,carry out the theft or assault, and exit the facility mustbe greater than the longest probable response time ofsecurity forces. The response time dictates the design andselection of delay mechanisms.Intervention is the final line of defense in physicalsecurity. It consists primarily of guards or security forces.Intruders must ultimately be confronted face-to-face ifaggression is to be halted. Their physical presence servesas a deterrent. They can be permanently placed for criticalfacilities. They can be deployed on a full time basis to lesscritical facilities when intelligence information indicates ahigh threat environment. Roving patrols can be used formultiple facilities of a less critical nature. Security forcesshould be able to respond in mass to a detected intrusionto abort crimes in progress.All of these physical security fundamentals can beused to reduce the vulnerability of utility systems. Additionalcountermeasures specific to utility systems shouldbe implemented. Actions such as improving componentreliability, installing redundant systems, preparing forrapid recovery, and contingency planning are discussedat length later.23.1.2 Utility Systems As Interdependent NetworksModern industrial, institutional, and commercial facilitiesare dependent on many utility systems. The utilityservices supporting modern facilities include electricity,thermal fuels (natural gas, fuel oil, coal), water, steam,chilled water, compressed air, sanitary sewer, industrialwaste, communication systems, and transportation systems.Some of these systems are restricted to the site.Others are a maze of distribution paths and processingstations connecting the site to distant generation or productionfacilities. The individual components of thesesystems form networks.Any failed component in a network has the potentialto disrupt or degrade the performance of the entiresystem. Some networks contain only a single path linkingthe site to the source of its utility supply. The loss of thatsingle path results in total disruption at the site. Othernetworks have redundant paths. Alternate routes areavailable to serve the site at full or partial capacity shouldone of the routes be damaged. In a redundant system,loads normally carried by the damage portion of the systemwill be shifted to an operational part of the network.This places extra stress on the remaining portion of thesystem. Marginal components sometimes fail under thisadditional load, causing additional segments of the networkto fail.Utility networks do not operate in isolation; they arelinked to each other in a web of dependence. Each providesa vital service to the other. The impact of one failedutility network is felt in many other systems. The effectsspread like ripples on a pond. More systems will fail ifredundant or back-up systems are not in place. Often, theeasiest way for a saboteur to disable a utility system is todamage another system on which it depends.Example. Figure 23.1 illustrates the mutual dependenceutility systems share with each other at one facility.This facility generates some of its own electricity withnatural gas engines, pumps a portion of its water fromwells using electric pumps, utilizes steam absorptionchillers, and produces compressed air from a mixtureof electric and gas engine compressors. A disruption ofnatural gas will immediately halt the on-site generationof electricity and steam and curtail the production ofcompressed air since these systems depend directly onnatural gas as their prime mover. Failure of one utilitywill cascade to other systems. The absorption chillerswould also experience a forced outage since they arepowered by steam which is experiencing a forced outagedue to the lack of natural gas. All process equipment thatrequires any of these disrupted utilities will be idle untilfull utility service is restored.23.1.3 Threats to Utility SystemsThe individual components of a utility system canbe damaged or disabled in many ways. The network itselfmay be wholly or partially disrupted depending onthe criticality of the component or components damaged.The threats to network components can be segregatedinto four categories: equipment failure, natural disaster,accidents, and sabotage.Equipment failure is the normal loss of system componentsas they reach the end of their expected life. Thereliability of most electrical and mechanical components

ENERGY SECURITY AND RELIABILITY 623Figure 23.1 Interdependence of utilitynetworks.follows a “ bath-tub curve” illustrated in Figures 23.2 and23.3. These curves begin with a high rate of failure earlyin the life cycle. This failure rate rapidly decreases until itreaches a minimum value that remains relatively stablethroughout the normal operating period. The failure rateincreases again once the component enters the wear-outphase.Natural disasters are the most common cause ofwidespread network failure. They include earthquake,hurricane, tornado, wind, lightning, fire, flood, ice, andanimal damage. Utility companies are well versed indealing with these situations and generally have themeans to effect repairs rapidly. However, widespreaddamage can leave some customers without service fordays resulting in substantial economic loss.Accidents are unintentional human actions such astraffic accidents, operator error, fires, and improper designor modification of the system. Operator error playeda significant role in the failure of the Chernoble and ThreeMile Island nuclear plants. It also contributed to both ofthe New York City blackouts. Once a failure sequence hasbegun, the system is in an abnormal state. Operators maylack the experience to know how the system will respondto a given corrective action or might fail to respond quicklyenough to a dynamic and rapidly changing situation.Sabotage covers the realm of intentional humanaction. It includes the entire spectrum from a single disgruntledemployee to acts of terrorism to military action.Terrorism has been a fact of life in many countries for along time. Its use in the United States has increased great-Figure 23.2 Typical “bath-tub curve” for electronicequipment showing failure rate per unit time.Figure 23.3 Typical “bath-tub curve” for mechanicalequipment showing failure rate per unit time.

624 ENERGY MANAGEMENT HANDBOOKly in recent years. Most terrorist organizations have socialor political motivations that could lead them to targetindustrial or commercial operations. Military actions arebeyond the scope of what most industrial or commercialfacilities can handle. These types of threats are of interestonly to the military itself.23.2 RISK ANALYSIS METHODSRisk analysis is a necessary first step in the processof minimizing the losses associated with unscheduledutility outages. The purpose of risk analysis is to understandthe system under study and establish a knowledgebase from which resources can be optimally allocatedto counter known threats. This understanding comesfrom a systematic evaluation that collects and organizesinformation about the failure modes of a utility system.Once these failure modes are understood decisions canbe made that reduce the likelihood of a failure, facilitateoperation under duress conditions, and facilitate rapidrestoration to normal operating conditions.Two broad categories of risk analysis methods areavailable. The inductive methods make assumptionsabout the state of specific system components or someinitiating event and then determine the impact on theentire system. They can be used to examine a system toany level of detail desired, but are generally only used toprovide an overview. It is impractical and often unnecessaryto examine every possible failure or combination offailures in a system. When the complexity or importanceof a system merits more detailed analysis a deductivemethod is used.Deductive analysis makes an assumption about thecondition of the entire system and then determines thestate of specific components that lead to the assumedcondition. Fault tree analysis is the most common andmost useful deductive technique. The deductive methodis preferred because it imposes a framework of order andobjectivity in place of what is often a subjective and haphazardprocess.Probabilities can be incorporated into all of thesemethods to estimate the overall reliability of the system.Probabilistic techniques are best used when analyzingequipment failures but have also been used with somesuccess in the evaluation of human error or accident.They are of less value when evaluating natural disastersand meaningless when applied to sabotage or other formsof organized hostility. The results would not only dependon the probability of a sabotage attempt occurring butalso on the probability that all actions necessary to disablethe system were successfully accomplished during theattack. This type of data is extremely difficult to obtainand is highly speculative. Therefore, the analysis is bestconducted under the assumption that an adverse situationwill definitely occur and proceed to determine whatspecific scenarios will result in system failure 1 .23.2.1 Inductive MethodsA number of analysis tools are available to examinethe effects of single component failures of a system. Threecommon and well developed techniques are FailureMode and Effect Analysis (FMEA), Failure Mode Effectand Criticality Analysis (FMECA) and Fault HazardAnalysis (FHA). These methods are very similar andbuild upon the previous technique by gradually increasingin scope. Most utilize failure probabilities of individualcomponents to estimate the reliability of the entiresystem. Preprinted forms are generally used to help collectand organize the information.FMEA recognizes that components can fail in morethan one way. All of the failure modes for each componentare listed along with the probability of failure. Thesefailure modes are then sorted into critical and non-criticalfailures. The non-critical failures are typically ignored inthe name of economy. Any failure modes with unknownconsequences should be considered critical. Figure 23.4 isa typical data sheet used in FMEA.Failure Mode Effect and Criticality Analysis (FME-CA) is very similar to FMEA except that the criticality ofthe failure is analyzed in greater detail and assurancesand controls are described for limiting the likelihood ofsuch failures 2 . FMECA has four steps. The first identifiesthe faulted conditions. The second step explores the potentialeffects of the fault. Next, existing corrective actionsor countermeasures are listed that minimize the risk offailure or mitigate the impact of a failure. Finally, the situationis evaluated to determine if adequate precautionshave been made and if not to identify additional actionitems. This technique is of particular value when workingwith the system operators. It can lead to an excellentunderstanding of how a system is actually operated asopposed to how the designer intended for it to be operated.Figure 23.5 is a data sheet used with FMECA.Fault Hazard Analysis (FHA) is another permutationof Failure Mode and Effect Analysis (FMEA). Itsvalue lies in an ability to detect faults that cross organizationalboundaries. Figure 23.6 is a data collection form foruse with FHA. It is the same data form used with FMEAwith three addition columns. Faults in column five aretraced up to an organizational boundary. Column six isadded to list upstream components that could cause thefault in question. Factors that cause secondary failures arelisted in column seven. These are things like operational

ENERGY SECURITY AND RELIABILITY 625conditions or environmental variables known to affectthe component. A remarks column is generally includedto summarize the situation. FHA is an excellent startingpoint if more detailed examination is anticipated sincethe data are collected in a format that is readily used inFault Tree Analysis.These techniques concern themselves with the effectsof single failures. Systems with single point failurestend to be highly vulnerable if special precautions are nottaken and these methods will highlight those vital components.However, critical utility systems are normallydesigned to be redundant. The majority of forced outagesin a redundant system will be caused by the simultaneousfailure of multiple components. Techniques that only considersingle point failures will significantly underestimatethe vulnerability of a system. No matter how unlikely anevent or combination of events may seem, experience hasproven that improbable events do occur.The Double Failure Matrix (DFM) examines the effectsof two simultaneous failures. A square grid is laidout with every failure mode of every component listedalong the columns and also along the rows. The intersectionof any two failure modes in the matrix represents adouble failure in the system. The criticality of that doublefailure is listed at the intersection. Critical and catastrophicfailures are explored further to identify corrective actionsor alternative designs. The diagonal along the gridis the intersection of a component failure with itself. Thisis the set of single mode failures. It can be used as a firstcut analysis and later expanded to encompass the entireset of two component failures if desired. This methodbecomes quite cumbersome or prohibitively difficult withlarge or complex systems.Event Tree Analysis is an exhaustive methodologythat considers every possible combination of failedcomponents. It is known as a tree because the pictorialillustration branches like a tree every time anothercomponent is included. The tree begins with a fullyoperational system that represents the trunk of the tree.The first component is added and the tree branches intwo directions. One branch represents the normal operationalstate of the component and the other represents thefailed state. The second component is considered next. Abranch representing the operational and failed states isFigure 23.4 Data collection sheet for Failure Mode andEffect Analysis (FMEA).Figure 23.5 Data collection sheet for Failure Mode Effectand Criticality Analysis (FMECA).

626 ENERGY MANAGEMENT HANDBOOKFigure 23.6 Data collection sheet for Failure Hazard Analysis (FHA).added to each of the existing branches, resulting in a totalof four branches. Additional components are added in alike manner until the entire system has been included.The paths from the trunk to the tip of each branch arethen evaluated to determine the state of the entire systemfor every combination of failed components. Complexsystems can be in a state of total success, total failure, orsome variant of partial success or failure.Example. Figure 23.7 is a simplified one-line diagramof an electric distribution system for a facility with criticalloads. The main switch gear at the facility is a double-endedsubstation fed from two different commercial powersources. All critical loads are isolated on a single busswhich can receive power from either commercial source oran on-site emergency generator. The facility engineer conducteda Failure Mode Effect and Criticality Analysis. Hisfindings are listed in Figure 23.8. Three single point failureswere identified that result in a forced outage of the criticalloads if any one of these components fails: automatictransfer switch, main breaker at the critical load switchgear, and the buss in the critical load switch gear. Actionitems were listed for all components to improve systemreliability. Most of the breakers require on-site spares formaximum reliability, however, a number of these can beshared to minimize the expense. For example, the mainbreakers at switch gear “A” and “B” can share a spare sincethey are of the same size. The most important action itemsare those for the single point failures.FMECA identifies only single point failures and cansignificantly underestimate the risk of a forced outagesince improbable events like double failures do occur. ADouble Failure Matrix (DFM) was constructed in Figure23.9 to identify the combinations of two failed componentsthat would deprive the critical loads of electricity.Rules were defined to gauge the criticality of each doublefailure. Loss of any one of the three sources of electricityis a level 2 failure with marginal consequences. Loss ofany two of the three sources is a level 3 failure with criticalconsequences and loss of all three sources is a level 4failure with catastrophic consequences since all critical

ENERGY SECURITY AND RELIABILITY 627Figure 23.7 Simplified electric distribution system forexample.loads are in an unscheduled forced outage. The upperportion of the matrix is left blank in this example since itis a mirror image of the lower portion.The highlighted diagonal is the intersection of acomponent with itself and represents the single pointfailures. The results of the FMECA are verified since thethree components identified as single point failures arenow shown to be level 4 failures along the diagonal. Sincethese components are single point failures their rows andcolumns are filled with 4’s denoting a catastrophic loss ofpower when they and any other component are simultaneouslyfailed. Six additional level 4 failures appear on thechart that represent true double component failures. All ofthese involve the inability to transfer commercial powerto the critical load buss and a simultaneous failure of theemergency generator or its support equipment. This couldlead the facility manager to investigate alternate systemsconfigurations that have more than one way of transferringcommercial power to the critical loads. Additionally,the threats that can reasonably be expected to damage thecomponents involved in level 4 failures should be carefullyexplored and countermeasures implemented to reduce thelikelihood of damage. The level 3 failures that occur alongthe diagonal also merit special attention.23.2.2 Deductive MethodFault Tree Analysis is the most useful of the deductivemethods and is preferred by the author above all theinductive methods. Benefits include an understandingof all system failure modes, identification of the mostcritical components in a complex network, and the abilityto objectively compare alternate system configurations.Fault trees use a logic that is essentially the reverse of thatused in event trees. In this method a particular failurecondition is considered and a logic tree is constructedthat identifies the various combinations and sequence ofother failures that lead to the failure being considered.This method is frequently used as a qualitative evaluationmethod in order to assist the designer, planner or operatorin deciding how a system may fail and what remediesmay be used to overcome the cause of failure 3 .The fault tree is a graphical model of the variouscombinations of faults that will result in a predefinedundesired condition. Examples of this undesired condi-

628 ENERGY MANAGEMENT HANDBOOKFigure 23.8 Failure Mode Effect and Criticality Analysis (FMECA) for example.

ENERGY SECURITY AND RELIABILITY 629Figure 23.9 Double Failure Matrix (DFM) for electric distribution system in example.tion are:a) Total loss of electricity to surgical suite.b) Total loss of chilled water to computer facility.c) Steam boiler unable to generate steam.d) Loss of natural gas feedstock to fertilizer plant.e) Water supply to major metropolitan area curtailed tohalf of minimum requirement.f) Environmental conditioning of controlled experimentis interrupted for more than thirty minutes.The faults can be initiated by sabotage actions, softwarefailures, component hardware failures, human errors, orother pertinent events. The relationship between theseevents is depicted with logic gates.The most important event and logic symbols areshown in Figure 23.10. A basic event is an initiating faultthat requires no further development or explanation. Thebasic event is normally associated with a specific componentor subsystem failure. The undeveloped event is afailure that is not considered in further detail because it isnot significant or sufficient information is not available.Logic gates are used to depict the relationship betweentwo or more events and some higher level failure ofthe system. This higher level failure is known as the outputof the gate. These higher level failures are combinedusing logic gates until they culminate in the top event of

630 ENERGY MANAGEMENT HANDBOOKthe tree, which is the previously defined undesired event.A simple fault tree is illustrated in Figure 23.11.The “OR” gate shows that the higher level failurewill occur if at least one of the input events occurs. An“OR” gate could be used to model two circuit breakersin series on a radial underground feeder. If either circuitbreaker is opened the circuit path will be broken and allloads served by that feeder will be deprived of electricity.The output event of an “AND” gate occurs onlyif all of the input events occur simultaneously. It wouldbe used to model two redundant pumps in parallel. Ifone pump fails the other will continue to circulate fluidthrough the system and no higher order failure will occur.If both pumps fail at the same time for any reason theentire pumping subsystem fails.The logic of the fault tree is analyzed using booleanalgebra to identify the minimal cut sets. A minimal cut setis a collection of system components which, when failed,cause the failure of the system. The system is not in afailed state if any one of the components in this set hasnot failed or is restored to operation. In fault tree terminology,a cut set is a combination of basic events that willresult in the undesired top event of the tree. A computeris normally used to automate this tedious and error pronemathematical procedure.Cut sets are utilized because they directly correspondto the modes of system failure. In a simple case, thecut sets do not provide any insights that are not alreadyquite obvious. In more complex systems, where the systemfailure modes are not so obvious, the minimal cut setcomputation provides the analyst with a thorough andsystematic method to identify the combinations of componentfailures which culminate in the top event. Oncean exhaustive list of cut sets is assembled, they can beanalyzed to determine which components occur in failuremodes with the highest frequency. These, along with thesingle point failures, are the most critical components ofthe entire system and merit special attention to keep themout of harm’s way.23.3 COUNTERMEASURESCorrective actions, or countermeasures, are implementedto reduce the risk of an unscheduled outage.These measures should initially be focused on singlecomponent failures and those with the most catastrophicconsequences when failed. The second priority are thosecomponents that occur in the largest number of cut sets.Countermeasures fall into three broad categories: protectivemeasures, redundant systems, and rapid recovery.Basic Undeveloped OR AndEvent Event Gate Gatefailure ofUPSfailure oftransferswitchEVENT AND LOGIC SYMBOLSFigure 23.10 Event and logic symbolsloss of powertocritical loadcircuitbreakermanuallyopenedloss of powerto main bussfailure ofemergencygeneratorloss of powerto UPSloss of power totransfer switchfailure ofmain bussloss ofcommercialpowerFigure 23.11 Sample fault tree for electric distributionsystem with uninterruptible power supply and emergencygenerator.Risk analysis contributes vital information to thecountermeasure process by imparting an understandingof the system failure modes. By knowing how the systemcan fail and what components contribute the those failures,the facility manager can make informed decisionsthat allocate resources to those countermeasures thatmitigate the most significant risks.Facilities and the utility systems that support themare complex. No single countermeasure is sufficient tomitigate all risk. A multi-faceted approach is required.Just as a three-legged stool will not stand on one leg, nei-

ENERGY SECURITY AND RELIABILITY 631ther will a facility be secure against disruption of utilitysupport without a comprehensive approach.23.3.1 Physical SecurityProtective countermeasures are actions taken beforea crisis occurs to safeguard the components of autility system against mechanical or electrical damage.Systems are normally constructed to withstand vandalism,severe weather, and other foreseeable events. Criticalsystems are normally designed with high reliabilitycomponents.Physical barriers are the first protective action thatshould be considered. They establish a physical and psychologicaldeterrent to unauthorized access. Their purposeis to define boundaries for both security and safety,detect entry, and delay and impede unauthorized entry.Fences are the most common barrier. They are used tosecure the perimeter of a site. In high value sites, fencingshould be supplemented with barbed wire or razor wire.Walls are another excellent barrier. They are routinelyused to isolate electrical vaults and mechanical roomswithin a building for safety reasons. Proper access controlmakes these areas more secure as well as safe. It shouldbe remembered that physical barriers are not sufficientto deter a determined adversary intent on causing damage.When the situation warrants, physical barriers mustbe supplemented with surveillance systems and guards.Area lighting is used in combination with physical barriersto further deter intrusion and aid security personnelin the detection of unauthorized entry.Many components of utility systems are built inexposed locations that make them vulnerable to traffic accidentsor tampering. Pipeline components such as valvesand regulators can be buried in pits. Berms can protectmany components like metering stations and regulatorsby sheltering them from vehicular traffic, deflectingblasts, and obscuring the line of sight necessary for firearmdamage. Bollards offer excellent protection againstdelivery vehicles and lawn mowers. Creek and rivercrossing should be designed to withstand the full forceof flooding. This includes any large debris that might becarried by the current.Site planning is an important aspect of energy security.Vital components should be located away fromperimeter fences and shielded from view when possible.Dispersal of resources is another excellent strategy. Whenredundant system components like electrical transformersare co-located they can all be disabled by a singleevent. Hardening of components and the structures thathouse them is appropriate in some instances. Hardeningmakes a system tolerant to the blast, shock, and vibrationcaused by explosions or earthquakes.23.3.2 Component ReliabilityReliability improvement of individual componentscan significantly reduce outages related to equipmentfailures. This is especially applicable to single point failures.Equipment brands and models known to have higha mean time between failure (MTBF) should be specified.Newly constructed systems should be thoroughlytested at load to insure they are beyond the known infantmortality period. An ongoing preventive maintenanceprogram should be implemented for existing facilities tokeep them in a state of high reliability. Such a programshould include through inspection, adjusting, lubrication,and replacement of failed redundant components.Aging components should be replaced before they enterthe wear out region. Standby systems, such as emergencyelectric generators, should be periodically tested at fullload. These procedures require trained personnel, testequipment, and meticulous record keeping.23.3.3 Redundant SystemsMany actions can be taken to facilitate continuedoperation when a portion of the utility infrastructure iscrippled or otherwise unavailable to the facility. The mostimportant of these is redundancy. Critical facilities shouldbe supported by multiple independent supply routes.Electrical feeders should follow different geographicroutes and, ideally, come from different substations.Telephones lines can sometimes to routed to differentswitches. Backup generators and uninterruptible powersupplies guard against disturbances and disruptionsthat occur off-site in the electric grid. Fuels subject tocurtailment, such as natural gas, can be supplementedwith alternate fuels such as fuel oil or propane-air mixingsystems that can be stored on-site. Sites that burn coalshould maintain a stockpile to guard against strikes ortransportation uncertainties. Redundant systems shouldbe sized with sufficient excess capacity to carry all criticalloads and all support functions, such as lighting and environmentalconditioning, that are required for continuousoperation.23.3.4 Rapid RecoveryOnce a crisis has developed, the overriding goal isto get the system operating in a normal state as rapidlyas possible and resume full scale operations. Stockpilingcritical or hard to get parts will reduce the recovery timemore than any other action. These components shouldhave been identified during the risk analysis. They mustbe stored separately from the components they are intendedto replace to preclude the possibility of a singlethreat damaging both the primary component and thespare. Emergency response teams must be available to

632 ENERGY MANAGEMENT HANDBOOKeffect the repairs. If damage is expected to exceed thecapabilities of in-house personnel, additional parts suppliersand alternate repair crews should be identified andcontracts in place to facilitate their rapid deployment.Simulation of recovery actions is an excellent training toolto prepare crews for operation under adverse conditions.Realistic simulation also helps identify unexpected obstacleslike limited communication capability, electroniclocks that won’t open without electricity, and vehiclesthat can’t be refueled.Any number of things can and will go wrong oncean emergency has been initiated. Operating with thelights off, both literally and figuratively, is a demandingtask even for the prepared. Portable generators and lightsare needed for twenty-four hour repair operations. Selfcontained battery-powered light carts should be availablefor use inside buildings. Clear lines of authority must bedefined in advance. Crisis coordinators must be able tocontact response teams even if the telephones don’t work.Crew members need to know where to report and towhom. These things must be thought out in advance anddocumented in contingency plans.23.3.5 Contingency PlanningContingency planning is a vital follow-up to thecountermeasure process. Plans give order to the chaossurrounding a catastrophic event. A manager’s thinkingis not always clear in the fog of a crisis. Lines of communicationbreak down. A plan provides the necessaryframework of authority to implement restoration actions.The plan identifies the resources necessary to effectrepairs and sets priorities to follow if the damage is extensive.Occasionally the planning process identifies criticalcomponents not previously identified or highlightsbottlenecks such as communication systems that maybecome overloaded during the crisis.The goal of contingency planning is rapid recoveryto a normal operating state. Risk analysis plays an importantrole in planning. It identifies those high-risk componentsor sub-systems that are most vulnerable or result incatastrophic situations when failed. Contingency plansmust be developed to complement the implementationof countermeasures that reduce the likelihood of damageoccurring. Risk analysis also identifies those componentsthat only yield moderate consequences when damaged.These components are frequently not vital enough tomerit the expense of constructing physical barriers orredundant systems. A solid contingency plan identifyingsources of parts and labor to effect repairs may be theleast cost option.The most important task performed in the plan is thedesignation of an emergency coordinator and delineationof authorities and responsibilities. A single coordinator isnecessary to insure priorities are followed and to resolveconflicts that may develop. Each member of the responseteam should have a well defined role. Proper planningwill insure no vital tasks “fall through the cracks” unnoticed.No plan can anticipate every contingency that mayarise, but a well written plan provides a framework thatcan be adapted to any situation. An energy contingencyplan should, as a minimum, contain the following items:• Definition of specific authorities and responsibilities• Priorities for plant functions and customers• Priorities for protection and restoration of resources• Curtailment actions• Recall of personnel (in-house, contract, and mutualaid)• Location of spares• Location and contacts for repair equipment• Equipment suppliersEach critical component identified during risk analysisshould be addressed in detail.Contingency plans should be exercised on a regularbasis. This familiarizes the staff with the specific rolesthey should assume during an emergency. Practicingthese tasks allows them to become proficient and providesan opportunity to identify deficiencies in the plan.Exercising also creates an awareness that a crisis canoccur. This gets personnel thinking about actions thatreduce risk in existing and new systems. Exercises canrange from a paper game or role playing to a full scalesimulation that actually interrupts the power to portionsof the facility. While full scale simulation is expensive interms of both labor and productivity, many valuable lessonscan be learned that would not otherwise be apparentuntil an actual disruption occurred.The most valuable product of repetitive gaming, forall of the people that truly participate, is the unfoldingof a process of how to react to a real energy emergency,a process including a communications net, recall procedures,and information systems of all sorts. Repetitivegaming tends to fine-tune the contingency plan, exposingshort comings, honing it down to an effective and workingdocument. It is impossible to learn true energy emergencymanagement in a totally calm, non-emergencyatmosphere. 423.4 ECONOMICS OF ENERGYSECURITY AND RELIABILITYThe consequences of utility outages cover the spectrumfrom minor annoyances to catastrophic loss of life

ENERGY SECURITY AND RELIABILITY 633or revenue. Financial losses can be grouped in severalconvenient categories. Loss of work in progress coversmany things. A few examples include parts damagedduring machining operations or by being stranded incleaning vats, spoiled meats or produces in cold storage,interruption of time sensitive chemical process, and lostor corrupted data. Lost business opportunities are commonin retail outlets that depend on computerized registers.Businesses that require continuous contact withcustomers such as reservation centers, data processingfacilities, and communications infrastructure also experiencelost business opportunities when utilities are disrupted.Many data processing and reservation centersestimate the cost of interruptions in the six figure rangeper hour of downtime.The cost of implementing countermeasures shouldbe commensurate with the cost of an unscheduled outage.This requires a knowledge of the financial impact ofan outage. If a probabilistic risk analysis technique wasutilized, the frequency and duration of outages can beestimated using conventional reliability theory. Costsassociated with recovery operations, damaged work inprogress, and lost business opportunities are calculatedfor each type of outage under analysis. The expectedannual loss can then be estimated for each scenario bymultiplying the duration by the hourly cost and addingthe cost of repairs and recovery operations. Countermeasurefunding can be prioritized on the basis of expectedfinancial loss, probability of occurrence, paybacktime, or other relevant measure.For non-probabilistic techniques, a more subjectiveapproach must be utilized. The duration of most outagescan be estimated based on known repair times. Theaccuracy of these estimates becomes questionable whenwidespread damage requires repair crews to servicemultiple sites or stocks of spare parts are exhausted.The absence of hard data on the frequency of such outagesincreases the subjective nature of the estimate. Onetechnique is to simply assume that a crisis will occurand allocate resources to mitigate the risk on the basis offinancial loss per occurrence or a cost/benefit ratio.Payback period and cost/benefit ratio are the mostcommon economic analysis tools used to rank countermeasurescompeting for budget dollars. They can alsobe used to determine if a particular measure shouldbe funded at all. Payback is calculated by dividing thecost to implement a countermeasure by the expectedannual loss derived from a probabilistic risk analysis.The payback period can be evaluated using normal corporatepolicy. A payback of two years or less is usuallysufficient to justify the expenditure. The cost/benefitratio is calculated by dividing the cost to implement acountermeasure by the losses associated with a singleforced outage. If the ratio is about one the expenditurewill be paid back after a single incident. If the ratio isabout one-half, the expenditure will be paid back aftertwo incidents.Insurance should be considered to shift the riskof financial loss to another party. Business interruptionpolicies are available to compensate firms for lost opportunitiesand help cover the cost of recovery operations.The cost of insurance policies should be treated as analternate countermeasure and compared using paybackor cost/benefit ratio.23.5 LINKS TO ENERGY MANAGEMENTMany energy security projects can piggy-back onenergy conservation projects at a nominal incrementalcost. Purchased utility costs can be reduced whilereliability is improved. This can dramatically alter theeconomics of implementing certain countermeasures.Fuel switching strategies let installations take advantageof interruptible natural gas tariffs or transportationcontracts. Peak shaving with existing or proposed generatorswill reduce electric demand charges. Time-of-userates and real-time pricing make self-generation veryattractive during certain seasons or at particular timesof the day.Many industrial facilities purchase utilities on interruptiblesupply contracts to reduce purchased utilitycosts. Curtailments are usually of short duration andare often contractually limited to a specified numberof hours per year. Alternate fuel sources are normallyinstalled prior to entering into an interruptible supplycontract. An example would be to replace the naturalgas supply to a steam plant with fuel oil or propane.This gives the facility the option of burning the leastexpensive fuel. Caution must be exercised when designinga dual-fuel system. Utility service is often curtailedat a time when it is most needed. Natural gas is typicallycurtailed during the coldest days of winter whendemand is highest and supplies become limited due tofreezing of wells. An extended curtailment could outlastthe on-site supply of an alternate fuel. The same coldweather which caused the primary fuel to be curtailedmay also result in shortages of the alternate fuel. Anunscheduled plant shutdown could occur if the storagesystem is undersized or if proper alternate arrangementwere not made in advance.Thermal energy storage systems are installedto reduce on-peak electric demand charges. They canalso play a key role in facility reliability. Chillers are

634 ENERGY MANAGEMENT HANDBOOKfrequently considered non-critical loads and are notconnected to emergency generators. When a disruptionoccurs, critical equipment that requires cooling must beshut-down until electricity is restored and the chilledwater system is returned to operation. The time requiredto restart the chilled water system and critical equipmentcan last many hours longer than the electrical disruptionthat initiated the event. When a storage system is inplace, only the chilled water circulating pumps need tobe treated as critical loads powered by the generator.Chilled water can be continuously circulated from thestorage system to the equipment, thus precluding theneed for a shut-down in all but the longest of outages.When the chillers are removed from the emergency generatoradditional process equipment can be connected inits place and kept operating during outages or a smallergenerator can be installed at a lower first cost.Energy conservation also has a direct impact on energysecurity. Conservation projects reduce the amountof energy required to perform at full capacity. By consumingless energy, an installation with a fixed quantityof alternate fuels stored on-site can remain in operationlonger under adverse conditions. More efficient operationscan also reduce the size of stand-by generators,uninterruptible power supplies, and similar systems.Smaller equipment usually means reduced constructioncosts and improved economics.23.6 IMPACT OF UTILITY DEREGULATIONDeregulation of the utility industry has the potentialto impact energy security and reliability in a waythe could greatly exceed the consequences of Year 2000problems. Y2K problems such as forced outages of powerplants or widespread blackouts failed to materializeeven in countries that did little to prepare for the muchanticipated century date rollover. This absence of Y2Kproblems has led many to conclude the threats neverexisted at all, which is not a correct perception. The potentialfor disruption did exist, it was just of unknownmagnitude. Many energy managers will conclude thatderegulation also carries with it no risk and will donothing to prepare.The risk associated with utility deregulation is alsoof unknown magnitude but is of a different nature thanY2K. Deregulation is not a one-time event. It is a permanentstructural change in an entire industry that willhave long term effects which evolve over time. Some ofthe issues that can be predicted now include reducedreliability of transmission and distribution systems, reducedreliability of generation systems, higher potentialfor contract default, and market price risk.Many utilities have reduced scheduled maintenanceof transmission and distribution systems andgeneration systems. This decision has two root causes.The first is the wave of downsizing that swept the industryduring the 1990s. These attempts to reduce operatingcosts and improve the bottom line have resultedin extended maintenance cycles, fewer spare parts, andsmaller crews responding to forced outages and naturaldisasters. While many of these companies are showingtemporary profit increases, they are also reducing systemreliability and extending mean time to repair. Thelong term consequences of these decisions are yet to befully felt.A secondary motivation for deferred maintenanceis the uncertainty associated with deregulation. Regulatorswill require utilities to unbundle the services offered.Integrated companies offering generation, transmissionand distribution, billing, and other customer serviceswill evolve into multiple, sometimes competing, companiesin much the same way the telephone industrywas broken up. Regulators may not permit full recoveryof capital investments in infrastructure. The fear of notrecovering these “stranded costs” has lead some utilitiesto reduce their investment in replacement or upgradedequipment. The consequences are the same: reducedreliability.While transmission and distribution is expected toremain a monopoly industry, many questions remain tobe answered regarding system operation and reliability.Generation will become a competitive industry. Old,inefficient plants with high heat rates will not be able tocompete on a cost basis in the new order but must continueto operate and fill the demand for electricity. Newmerchant plants and distributed generators with highefficiencies are being planned and constructed now.These, along with the old utility plants, will be unregulatedplants competing for customers and operated forprofit.Under the old, regulated structure, utilities hadmultiple generators that improved reliability. If oneplant failed, the system had sufficient spinning reserveto immediately compensate. Many unregulated operatorshave only a single plant in the region with no spinningreserve to insure reliability. If a generation providerexperiences a forced outage the customer may have nochoice but to purchase power from a default provider oron the short term spot market at greatly increased costs.The only alternative would be an immediate curtailmentof all operations. Recent system disruptions in the midwestand pacific northwest have seen spot market pricesrise several orders of magnitude until failed plants were

ENERGY SECURITY AND RELIABILITY 635returned to service.Experience with deregulated natural gas providesan indicator of future problems that might occur withelectricity. Gas producers contract to provide gas. If thespot market price of gas increases before the completionof the contract term, then the producer’s profit isreduced. Unethical operators have defaulted on contractobligations by refusing to allocate the gas to the originalcontracted customer. They instead sell the gas on thespot market to the highest bidder. This forces the originalcustomer to curtail load or obtain supplies on thespot market at current prices. The customer’s attemptto obtain stable commodity prices and minimize marketrisk through contractual instruments is nullified by theunethical acts of a supplier. Similar situations are boundto occur as the electric industry deregulates.23.7 SUMMARYAll facilities require a continuous and adequatesupply of utility support to function. Energy securityis the process of evaluating utility systems and implementingactions that minimize the impact of unscheduledoutages that prevent a facility from operating atfull capacity. Utility systems are networks with manycomponents. Loss of a few, or in some cases one, criticalcomponents is sufficient to disable a network or leaveit operating at partial capacity. Additionally, utility networksare not independent. They support each other in asymbiotic manner. The collapse of one network can leadto a domino effect that causes other networks to fail.The critical components of a utility network can beidentified by using risk analysis techniques. The inductivemethods make assumptions about the status of individualcomponents and then determine what impactis felt on the entire system. Deductive methods, notablyfault tree analysis, use an opposite approach. Fault treeanalysis assumes the system is in some undesired conditionand proceeds to determine what combinationsof failed components will result in that condition. Thecombinations of failed components are called cut sets.The component failures can be caused by equipmentfailure, natural disaster, accident, or sabotage.Countermeasures are actions that prevent or minimizethe impact of utility disruptions. Three countermeasuresthat should not be neglected by any facilitymanager are physical protection, redundancy, and stockpilingof critical spare parts. Coupled with contingencyplanning, these counter measures will greatly enhancethe energy security of any installation. Contingencyplans should be exercised for maximum effectiveness.Utility disruptions can cause a wide range of impactson the affected facility. In most cases, this impactcan be expressed as a dollar value. This financial impactis useful in evaluating the economics of implementingcountermeasures. The most common economic analysistools are payback and cost/benefit ratio. Many countermeasurescan be made more cost effective by linkingthem to energy conservation projects.References1. Bracher, Bradley L., Utility Risk Analysis, Proceedings of the 8thWorld Energy Engineering Congress, 1989, p. 601.2. U.S. Nuclear Regulatory Commission, Fault Tree Handbook, p. II-4.3. Billinton and Allan, Reliability Evaluation of Engineering Systems:Concepts and Techniques, p. 113.4. Broussard, Peter A., Energy Security for Industrial Facilities - ContingencyPlanning for Power Disruption, PennWell Publishing Company,1994, pg. 111.


CHAPTER 24UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCINGGEORGE R. OWENS, P.E. C.E.M.Energy and Engineering Solutions, Inc.24.0 INTRODUCTION“ Utility Deregulation,” “Customer Choice,” “ UnbundledRates,” “Re-regulation,” “Universal ServiceCharge,” “Off Tariff Gas,” “ Stranded Costs,” “CompetitiveTransition Charge (CTC),” “Caps and Floors,” “ LoadProfiles” and on and on are the new energy buzzwords.They are all the jargon are being used as customers,utilities and the new energy service suppliers becomeproficient in doing the business of utility deregulation.Add to that the California energy shortages androlling blackouts, the Northeast and Midwest outages of2003, scandal, rising energy prices, loss of price protectionin deregulated states and you can see why utilityderegulation is increasingly on the mind of utility customersthroughout the United States and abroad.With individual state actions on deregulatingnatural gas in the late 80’s and then the passage of theEnergy Policy Act (EPACT) of 1992, the process of deregulatingthe gas and electric industry was begun. Becauseof this historic change toward a competitive arena,the utilities, their customers, and the new energy serviceproviders have begun to reexamine their relationships.How will utility customers, each with varyingdegrees of sophistication, choose their suppliers ofthese services? Who will supply them? What will itcost? How will it impact comfort, production, tenantsand occupants? How will the successful new playersbring forward the right product to the marketplace tostay profitable? And how will more and better energypurchases improve the bottom line?This chapter reviews the historic relationshipsbetween utilities, their customers, and the new energyservice providers, and the tremendous possibilities fordoing business in new and different ways.The following figure portrays how power is generatedand how it is ultimately delivered to the endcustomer.1. Generator – Undergoing deregulation2. Generator Substation – See 1The Power Flow Diagram3. Transmission System – Continues to be regulatedby the Federal Energy Regulatory Commission(FERC) for interstate and by the individual statesfor in-state systems4. Distribution Substation – Continues to be regulatedby individual states5. Distribution Lines – See 46. End Use Customer – As a result of deregulation,will be able to purchase power from a numberof generators. Will still be served by the local“wires” distribution utility which is regulated bythe state.24.1 AN HISTORICAL PERSPECTIVE OFTHE ELECTRIC POWER INDUSTRYAt the turn of the century, vertically integratedelectric utilities produced approximately two-fifths ofthe nation’s electricity. At the time, many businesses(nonutilities) generated their own electricity. When utilitiesbegan to install larger and more efficient generatorsand more transmission lines, the associated increase inconvenience and economical service prompted manyindustrial consumers to shift to the utilities for theirelectricity needs. With the invention of the electric motorcame the inevitable use of more and more home appliances.Consumption of electricity skyrocketed alongwith the utility share of the nation’s generation.637

638 ENERGY MANAGEMENT HANDBOOKThe early structure of the electric utility industrywas predicated on the concept that a central source ofpower supplied by efficient, low-cost utility generation,transmission, and distribution was a natural monopoly.In addition to its intrinsic design to protect consumers,regulation generally provided reliability and a fair rateof return to the utility. The result was traditional ratebase regulation.For decades, utilities were able to meet increasingdemand at decreasing prices. Economies of scale wereachieved through capacity additions, technological advances,and declining costs, even during periods whenthe economy was suffering. Of course, the monopolisticenvironment in which they operated left them virtuallyunhindered by the worries that would have been createdby competitors. This overall trend continued until thelate 1960s, when the electric utility industry saw decreasingunit costs and rapid growth give way to increasingunit costs and slower growth.The passage of EPACT-1992 began the process ofdrastically changing the way that utilities, their customers,and the energy services sector deal (or do not deal)with each other. Regulated monopolies are out and customerchoice is in. The future will require knowledge,flexibility, and maybe even size to parlay this changingenvironment into profit and cost saving opportunities.One of the provisions of EPACT-1992 mandatesopen access on the transmission system to “wholesale”customers. It also provides for open access to “exemptwholesale generators” to provide power in direct competitionwith the regulated utilities. This provision fosteredbilateral contracts (those directly between a generatorand a customer) in the wholesale power market. Theregulated utilities then continue to transport the powerover the transmission grid and ultimately, through thedistribution grid, directly to the customer.What EPACT-1992 did not do was to allow for “retail”open access. Unless you are a wholesale customer,power can only be purchased from the regulated utility.However, EPACT-1992 made provisions for the statesto investigate retail wheeling (“wheeling” and “openaccess” are other terms used to describe deregulation).Many states have held or are currently holding hearings.Several states either have or will soon have pilotprograms for retail wheeling. The model being used isthat the electric generation component (typically 60-70%of the total bill), will be deregulated and subject to fullcompetition. The transmission and distribution systemswill remain regulated and subject to FERC and statePublic Service Commission (PSC) control.A new comprehensive energy bill, EPACT-2005, wassigned into law in 2005, just as this edition was beingfinalized. Look for expanded discussion of EPACT-2005in future editions of this chapter. This bill affects energyproduction, including renewables, energy conservation,regulations on the country’s transmission grids, utilityderegulation as well as other energy sectors. Tax incentivesto spur change are key facets of EPACT-2005.ELECTRIC INDUSTRY DEREGULATION TIME LINE1992 - Passage of EPACT and the start of the debate.1995 & 1996 - The first pilot projects and the start ofspecial deals. Examples are: The automakers inDetroit, New Hampshire programs for directpurchase including industrial, commercial andresidential, and large user pilots in Illinois andMassachusetts.1997 - Continuation of more pilots in many states andalmost every state has deregulation on the legislativeand regulatory commission agenda.1998 - Full deregulation in a few states for large users(i.e., California and Massachusetts). Many stateshave converged upon 1/1/98 as the start oftheir deregulation efforts with more pilots andthe first 5% roll-in of users, such as Pennsylvaniaand New York.2000 - Deregulation of electricity became common formost industrial and commercial users and beganto penetrate the residential market in severalstates. These included Maryland, New Jersey,New York, and Pennsylvania among others. Seefigure 24.1.2002/3- Customers have always had a “backstop” ofregulated pricing. Now that the transition periodsare nearing their end, customers are facedwith the option of buying electricity on the openmarket without a regulated default price.2003 - During the summer, parts of the northeast andupper Midwest experience a massive blackoutthat shuts down businesses and residentialcustomers. The adequacy of the transmissionsystem is blamed.2005 - EPACT-2005 becomes law24.2 THE TRANSMISSION SYSTEM AND THEFEDERAL ENERGY REGULATORY COMMISSION’S(FERC) ROLE IN PROMOTING COMPETITION INWHOLESALE POWEREven before the passage of EPACT in 1992, FERCplayed a critical role in the competitive transformationof wholesale power generation in the electric powerindustry. Specific initiatives include notices of proposed

UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCING 639rulemaking that proposed steps toward the expansionof competitive wholesale electricity markets. FERC’sOrder 888, which was issued in 1996, required publicutilities that own, operate, or control transmission linesto file tariffs that were non-discriminatory at rates thatare no higher than what the utility charges itself. Theseactions essentially opened up the national transmissiongrid to non-discretionary access on the wholesale level(public utilities, municipalities and rural cooperatives).This order did not give access to the transmission gridto retail customers.In an effort to ensure that the transmission gridis opened to competition on a non-discriminatory basis,Independent System Operators (ISO’s) are beingformed in many regions of the country. An ISO is anindependent operator of the transmission grid and isprimarily responsible for reliability, maintenance (even ifthe day-to-day maintenance is performed by others) andsecurity. In addition, ISO’s generally provide the followingfunctions: congestion management, administeringtransmission and ancillary pricing, making transmissioninformation publicly available, etc.24.3 STRANDED COSTSStranded costs are generally described as legitimate,prudent and verifiable costs incurred by a public utility ora transmitting utility to provide a service to a customerthat subsequently are no longer used. Since the asset orcapacity is generally paid for through rates, ceasing to usethe service leaves the asset, and its cost, stranded. In thecase of de-regulation, stranded costs are created when theutility service or asset is provided, in whole or in part,to a deregulated customer of another public utility ortransmitting utility. Stranded costs emerge because newgenerating capacity can currently be built and operatedat costs that are lower than many utilities’ embeddedcosts. Wholesale and retail customers have, therefore, anincentive to turn to lower cost producers. Such actionsmake it difficult for utilities to recover all their prudentlyincurred costs in generating facilities.Stranded costs can occur during the transition toa fully competitive wholesale power market as somewholesale customers leave a utility’s system to buypower from other sources. This may idle the utility’sexisting generating plants, imperil its fuel contracts,and inhibit its capability to undertake planned systemexpansion leading to the creation of “stranded costs.”During the transition to a fully competitive wholesalepower market, some utilities may incur stranded costsas customers switch to other suppliers. If power previouslysold to a departing customer cannot be sold toan alternative buyer, or if other means of mitigating thestranded costs cannot be found, the options for recoveringstranded costs are limited.The issue of stranded costs has become contentiousin the state proceedings on electric deregulation. UtilitiesRetail access is either currently available to all or some customers or will soon beavailable. Those states are Arizona, Connecticut, Delaware, District of Columbia, Illinois,Maine, Maryland, Massachusetts, Michigan, New Hampshire, New Jersey, New York, Ohio,Oregon, Pennsylvania, Rhode Island, Texas, and Virginia.In Oregon, no customers are currently participating inthe State’s retail access program, but the law allowsnonresidential customers access. Yellow colored states arenot actively pursuing restructuring. Those states are Alabama,Alaska, Colorado, Florida, Georgia, Hawaii, Idaho,Indiana, Iowa, Kansas, Kentucky, Louisiana, Minnesota,Mississippi, Missouri, Nebraska, North Carolina, NorthDakota, South Carolina, South Dakota, Tennessee, Utah,Vermont, Washington, West Virginia, Wisconsin, andWyoming. In West Virginia, the Legislature and Governorhave not approved the Public Service Commission’s restructuringplan, authorized by HB 4277. The Legislaturehas not passed a resolution resolving the tax issues ofthe PSC’s plan, and no activity has occurred since earlyin 2001. A green colored state signifies a delay in therestructuring process or the implementation of retail access.Those states are Arkansas, Montana, Nevada, NewMexico, and Oklahoma. California is the only blue coloredstate because direct retail access has been suspended.*As of January 30, 2003, Department of Energy, EnergyInformation AdministrationFigure 24.1 Status of State Electric Industry Restructuring Activity*

640 ENERGY MANAGEMENT HANDBOOKhave argued vehemently that they are justified in recoveringtheir stranded costs. Customer advocacy groups, onthe other hand, have argued that the stranded costs proposedby the utilities are excessive. This is being workedout in the state utility commissions. Often, in exchangefor recovering stranded costs, utilities are joining in settlementagreements that offer guaranteed rate reductionsand opening up their territories to deregulation.24.4 STATUS OF STATE ELECTRICINDUSTRY RESTRUCTURING ACTIVITYElectric deregulation on the retail level is determinedby state activity. Many states have or are in theprocess of enacting legislation and/or conducting proceedings.See Figure 24.1.24.5 TRADING ENERGY -MARKETERS AND BROKERSWith the opening of retail electricity markets inseveral states, new suppliers of electricity have developedbeyond the traditional vertically integrated electricutility. Energy marketers and brokers are the new companiesthat are being formed to fill this need. An energymarketer is one that buys electricity or gas commodityand transmission services from traditional utilities orother suppliers, then resells these products. An energybroker, like a real estate broker, arranges for sales butdoes not take title to the product. There are independentenergy marketers and brokers as well as unregulatedsubsidiaries of the regulated utility.According to The Edison Electric Institute, theenergy and energy services market was $360 billion in1996 and was expected to grow to $425 billion in 2000.To help put these numbers in perspective, this market isover six times the telecommunications marketplace. Asmore states open for competition, the energy marketersand brokers are anticipating strong growth. Energy suppliershave been in a merger and consolidation mode forthe past few years. This will probably continue at thesame pace as the energy industry redefines itself evenfurther. Guidance on how to choose the right supplierfor your business or clients will be offered later on inthis chapterThe trading of electricity on the commoditiesmarket is a rather new phenomenon. It has been recognizedthat the marketers, brokers, utilities and endusers need to have vehicles that are available for themanaging of risk in the sometimes-volatile electricitymarket. The New York Mercantile Exchange (NYMEX)has instituted the trading of electricity along with itsmore traditional commodities. A standard model for anelectricity futures contract has been established and istraded for delivery at several points around the country.As these contracts become more actively traded, theirusefulness will increase as a means to mitigate risk. Anexample of a risk management play would be when apower supplier locks in a future price via a futures oroptions contract to protect its position at that point intime. Then if the prices rise dramatically, the supplier’sprice will be protected.24.6 THE IMPACT OF DEREGULATIONHistorically, electricity prices have varied by afactor of two to one or greater, depending upon wherein the county the power is purchased. See Figure 24.2.These major differences even occur in utility jurisdictionsthat are joined. The cost of power has variedbecause of several factors, some of which are under theutilities control and some that are not, such as:• Decisions on projected load growth• The type of generation• Fuel selections• Cost of labor and taxes• The regulatory climateAll of these factors contribute to the range of pricing.Customers have been clamoring for the right to choosethe supplier and gain access to cheaper power for quitesome time. This has driven regulators to impose utilityderegulation, often with opposition from the incumbentutilities.Many believe that electric deregulation will evenout this difference and bring down the total averageprice through competition. There are others that donot share that opinion. Most utilities are already takingactions to reduce costs. Consolidations, layoffs, andmergers are occurring with increased frequency. Aspart of the transition to deregulation, many utilities arerequesting and receiving rate freezes and reductions inexchange for stranded costs.One factor has remained a constant until the early2000’s. Customers have always had a “backstop” ofregulated pricing until recently. Now that the transitionperiods are nearing their end, customers are faced withthe option of buying electricity on the open marketwithout a regulated default price. The risks to customershave increased dramatically. And, energy consultants

UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCING 641Figure 24.2 Electricity Cost by StateAverage Revenue from Electric Sales to Industrial Consumers by State, 1995 (Cents per Kilowatt-hour)and ESCOs are having a difficult time predicting thedirection of electricity costs.All of this provides for interesting background andstatistics, but what does it mean to energy managersinterested in providing and procuring utilities, commissioning,O&M (operations and maintenance), andthe other energy services required to build and operatebuildings effectively? Just as almost every business enterprisehas experienced changes in the way that theyoperate in the 90’s and 2000 and beyond, the electricutilities, their customers and the energy service sectormust also transform. Only well-prepared companies willbe in a position to take advantage of the opportunitiesthat will present themselves after deregulation. Buildingowners and managers need to be in a position to activelyparticipate in the early opening states. The followingquestions will have to be answered by each and everycompany if they are to be prepared:• Will they participate in the deregulated electricmarket?• Is it better to do a national account style supplyarrangement or divide the properties by regionand/or by building type?• How will electric deregulation affect their relationshipswith tenants in commercial, governmentaland institutional properties?• Would there be a benefit for multi-site facilities topartake in purchasing power on their own?• Should the analysis and operation of electric deregulationefforts be performed in-house or byconsultants or a combination?• What criteria should be used to select the energysuppliers when the future is uncertain?24.7 THE TEN-STEP PROGRAM TOSUCCESSFUL UTILITY DEREGULATIONIn order for the building sector to get ready forthe new order and answer the questions raised above,this ten-step program has been developed to ease thetransition and take advantage of the new opportunities.This Ten Step program is ideally suited to building ownersand managers as well as energy engineers that arein the process of developing their utility deregulationprogram.Step #1 - Know Thyself• When do you use the power• Distinguish between summer vs. winter, night vs.day

642 ENERGY MANAGEMENT HANDBOOK• What load can you control/change• What $$$ goal does your business have• What is your 24 hr. load profile• What are your in-house engineering, monitoringand financial strengthsStep #2 - Keep Informed• Read, read, read—network, network, network• Interact with your professional organizations• Talk to vendors, consultants, and contractors• Subscribe to trade publications• Attend seminars and conferences• Utilize internet resources—news groups, WWW,E-mail• Investigate buyer’s groupsStep #3 - Talk to Your Utilities (all energy types)• Recognize customer relations are improving• Discuss alternate contract terms or other energyservices• Find out if they are “for” or “agin” deregulation• Obtain improved service items (i.e., reliability)• Tell them your position and what you want. Nowis not the time to be bashful• Renegotiate existing contractsStep #4 - Talk to Your Future Utility(ies)• See Step #3• Find out who is actively pursuing your market• Check the neighborhood, check the region, looknationally• Develop your future relationships• Partner with Energy Service Companies (ESCOs),power marketing, financial, vendor and other partnersfor your energy services needsStep #5 - Explore Energy Services Now(Why wait for deregulation?)• Implement “standard” energy projects such aslighting, HVAC, etc.• Investigate district cooling/heating• Explore selling your central plant• Calculate square foot pricing• Buy comfort, Btus or GPMs; not kWhs• Outsource your Operations and Maintenance• Consider other work on the customer side of themeterStep #6 - Understand the Risks• Realize that times will be more complicated in thefuture• Consider the length of a contract term in uncertaintimes• Identify whether you want immediate reductionsnow, larger reductions later or prices tied to someother index• Determine the value of a flat price for utilities• Be wary of losing control of your destiny-turningover some of the operational controls of your energysystems• Realize the possibility some companies will not bearound in a few years• Determine how much risk you are willing to takein order to achieve higher rewardsStep #7 - Solicit Proposals• Meet with the bidders prior to issuing the RequestFor Proposal (RFP)• Prepare the RFP for the services you need• Identify qualified players• Make commissioning a requirement to achieve theresultsStep #8 - Evaluate Options• Enlist the aid of internal resources and outsideconsultants• Narrow the playing field and interview the finalistsprior to awarding• Prepare a financial analysis of the results over thelife of the project—Return on Investment (ROI) andNet Present Value (NPV)• Remember that the least first cost may or may notbe the best value• Pick someone that has the financial and technicalstrengths for the long term• Evaluate financial options such as leasing orsharedStep #9 - Negotiate ContractsRemember the following guidelines when negotiatinga contract:• The longer the contract, the more important theescalation clauses due to compounding• Since you may be losing some control, the contractdocument is your only protection• The supplying of energy is not regulated like thesupplying of kWhs are now• The clauses that identify the party taking responsibilityfor an action, or “Who Struck John” clauses,are often the most difficult to negotiate• Include monitoring and evaluation of results• Understand how the contract can be terminatedand what the penalties for early termination areStep #10 - Sit Back and Reap the Rewards• Monitor, measure, and compare

UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCING 643• Don’t forget Operations & Maintenance for thelong term• Keep looking, there are more opportunities outthere• Get off your duff and go to Step #1 for the nextround of reductions24.8 AGGREGATIONAggregation is the grouping of utility customersto jointly purchase commodities and/or other energyservices. There are many aggregators already formed orbeing formed in the states where utility deregulation isoccurring. There are two basic forms of aggregation:1. Similar Customers with Similar NeedsSimilar customers may be better served via aggregationeven if they have the same load profiles• Pricing and risk can be tailored to similar customersneeds• Similar billing needs can be met• Cross subsidization would be eliminated• Trust in the aggregator; i.e. BOMA for officebuilding managers membership2. Complementary Customers that May Enhance theTotalDifferent load profiles can benefit the aggregatedgroup by combining different load profiles.• Match a manufacturing facility with a flat orinverted load profile to an office building thathas a peaky load profile, etc.• Combining of load profiles is more attractive toa supplier than either would be individuallyWhy Aggregate?Some potential advantages to aggregating are:• Reduction of internal administration expense• Shared consulting expenses• More supplier attention resulting from a largerbid• Lower rates may be the result of a larger bid• Lower average rates resulting from combining dissimilaruser profilesWhy Not Aggregate?Some potential disadvantages from aggregatingare:• If you are big enough, you are your own aggregation• Good load factor customers may subsidize poorload factor customers• The average price of an aggregation may be lowerthan your unique price• An aggregation cannot meet “unique” customerrequirementsFactors that affect the decision on joining an aggregationDetermine if an aggregation is right for yoursituation by considering the following factors. Anunderstanding of how these factors apply to youroperation will result in an informed decision.• Size of load• Load profile• Risk tolerance• Internal abilities (or via consulting)• Contract length flexibility• Contract terms and conditions flexibility• Regulatory restrictions24.9 IN-HOUSE VS. OUTSOURCINGENERGY SERVICESThe end user sector has always used a combinationof in-house and outsourced energy services. Many largemanagers and owners have a talented and capable staffto analyze energy costs, develop capital programs, andoperate and maintain the in-place energy systems. Others(particularly the smaller players who cannot justifyan in-house staff) have outsourced these functions toa team of consultants, contractors, and utilities. Theserelationships have evolved recently due to downsizingand returning to the core businesses. In the new era ofderegulation, the complexion of how energy services aredelivered will evolve further.Customers and energy services companies are alreadygetting into the utility business of generating anddelivering power. Utilities are also getting into the actby going beyond the meter and supplying chilled/hotwater, conditioned air, and comfort. In doing so, manyutilities are setting up unregulated subsidiaries to providecommissioning, O&M, and many other energyservices to customers located within their territory, andnationwide as well.A variety of terms are often used: PerformanceContracting, Energy System Outsourcing, Utility PlantOutsourcing, Guaranteed Savings, Shared Savings,Sell/Leaseback of the central plant, Chauffage (used inEurope), Energy Services Performance Contract (ESPC),etc. Definitions are as follows:• Performance ContractingIs the process of providing a specific improvement

644 ENERGY MANAGEMENT HANDBOOKsuch as a lighting retrofit or a chiller change-out,usually using the contractor’s capital and then payingfor the project via the savings over a specificperiod of time. Often the contractor guarantees alevel of savings. The contractor supplies capital,engineering, equipment, installation, commissioningand often the maintenance and repair.• Energy System OutsourcingIs the process of divesting of the responsibilitiesand often the assets of the energy systems to athird party. The third party then supplies thecommodity, whether it be chilled water, steam,hot water, electricity, etc., at a per unit cost. Thethird party supplier then is responsible for theimprovement capital and operations and maintenanceof the energy system for the duration of thecontract.AdvantagesThe advantages of a performance contract or anenergy system outsourcing project revolves around fourmajor areas:1. Core Business IssuesMany industries and corporations have been reexaminingall of their non-core functions to determineif they would be better served by outsourcingthese functions. Performance contracting or outsourcingcan make sense if someone can be foundthat can do it better and cheaper than what can bemanaged by an in-house staff. Then the buildingmanagers can oversee the contractor and not thecomplete operation. This may allow the buildingto devote additional time and resources to othercore business issues such as increasing revenuesand reducing health care costs.2. MonetizationOne of the unique features of a performance contractor an energy system outsourcing project isthe opportunity to obtain an up front payment.There is an extreme amount of flexibility availabledepending upon the needs. The amount availablecan range from zero dollars to the approximatecurrent value of the installation. The more valueplaced on the up front payment will necessarilycause the monthly payments to increase as well asthe total amount of interest paid.3. Deferred Capital CostsMany electrical and HVAC energy systems are atan age or state of repair that would necessitatethe infusion of a major capital investment in thenear future. These investments are often requiredto address end-of-life, regulatory and efficiency issues.Either the building owner or manager couldprovide the capital or a third party could supplyit and then include the repayment in a commoditycharge plus interest; (“there are no free lunches”).4. Operating CostsThe biggest incentive to a performance contractor an energy system outsourcing project is that ifthe right supplier is chosen with the right incentives,then the total cost to own and operate thecentral plant can be less. The supplier, havingexpertise and volume in their core area of energyservices, brings this to reality. With this expertiseand volume, the supplier should be able to purchasesupplies at less cost, provide better-trainedpersonnel and implement energy and maintenancesaving programs. These programs can range fromcapital investment of energy saving equipment tooptimizing operations, maintenance and controlprograms.DisadvantagesPotentially, there are several disadvantages toundertaking a performance contract or an outsourcingproject. The items identified in this section need to berecognized and mitigated as indicated here and in theRisk Management section.1. Loss of ControlAs with any service, if it is outsourced, the serviceis more difficult to control. The building is left withdepending upon the skill, reliability and dedicationof the service supplier and the contract to obtainsatisfactory results. Even with a solid contract; ifthe supplier does not perform or goes out of business,the customer will suffer (see the Risk Managementsection). Close coordination between thebuilding and the supplier will be necessary overthe long term of the contract to adjust to changingconditions.2. Loss of FlexibilityUnless addressed adequately in the contract,changes that the building wants or needs to makecan cause the economics of the project to be adverselyaffected. Some examples are:• Changes in hours of operation• New systems that require additional coolingor heating, such as an expansion or renova-

UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCING 645tion, conversion of office or storage space toother uses, additional equipment requiringadditional cooling, etc.• Scheduling outages for maintenance or repairs• Using in house technicians for other servicesthroughout the building. If this situation occursin current operation, provisions for additionalbuilding staff or having the suppliermake the technician available needs to be arranged.If additional costs are indicated, theyshould be included in the financial analysis.3. Cost IncreasesThis only becomes a disadvantage if the contractdoes not adequately foresee and cover every contingencyand changing situation adequately. Toprotect themselves, the suppliers will try to putas much cost risk onto the customer as possible.It is the customer and the customer’s consultantsand attorneys responsibility to define the risks andinclude provisions in the contract.Financial IssuesThe basis for success of a performance contractor an energy system outsourcing project is dividedbetween the technical issues, contract terms, supplier’sperformance and how the project will be financed. Thesetypes of projects are as much (if not more) about thefinancial deal than the actual supplying of a commodityor a service. (See Chapter 4 -Economic Analysis and LifeCycle Costing) The answers to some basic questions willhelp guide the decision making process.• Is capital required during the term of the project?The question of the need for capital is one of themajor driving factors of a performance contractor an energy outsourcing project. Capital investedinto the HVAC and electrical systems for efficiencyupgrades, end of life replacements, increased reliabilityor capacity and environmental improvementscan be financed through the program.• Who will supply the capital and at what rate?The answer to the question of who will be supplyingthe capital should be made based upon yourability to supply capital from internal operations,capital improvement funds, borrowing ability andany special financing options such as tax freebonds or other low interest sources. If capital isneeded for other uses such as expansions and otherrevenue generating or cost reduction measures,then energy system outsourcing may be a goodchoice.• Is there a desire to obtain a payment up front?As stated previously, a performance contract orenergy system outsourcing project presents theopportunity to obtain a payment up front for theassets of the HVAC and electrical systems. However,any up-front payment increases the monthlypayment over the term of the contract and shouldbe considered similar to a loan.• Does the capital infusion and better operations generateenough cash flow to pay the debt?This is the sixty-four dollar question. Only byperforming a long-term evaluation of the economicsof the project with a comparison to thein house plan can the financial benefits be fairlycompared. A Net Present Value and Cash Flowanalysis should be used for the evaluation of aperformance contract or energy system outsourcingproject. It shows the capital and operating impactof the owner continuing to own and operatea HVAC and electrical systems. This is comparedto a third party outsourced option. The analysisshould be for a long enough period to incorporatethe effect of a major capital investment.This is often done for a 20-year period. This typeof analysis would allow the building owner ormanager to evaluate the financial impact of theproject over the term of the contract. Included inthe analysis should be a risk sensitivity assessmentthat would bracket and define the range ofresults based upon changing assumptions.Other Issues1. Management and Personnel Issues• Management - Usually, an in-house manager willneed to be assigned to manage the supplier and thecontract and to verify the accuracy of the billing.An in-house technical person or an outside consultantshould have the responsibility to periodicallyreview the condition of the equipment to protectthe long-term value of the central plant.• Personnel - Existing employees need to be considered.This may or may not have a monetary consequencedue to severance or other policies. If thereis an impact, it needs to be reflected in the analysis.It would usually be to the building’s benefit if theyears of knowledge and experience represented bythe current engineers could be transferred to the

646 ENERGY MANAGEMENT HANDBOOKnew supplier. Another personnel concern is theeffect on the moral of the employees due to theirfear of losing their jobs.2. Which services to outsource?Where there are other services located in the centralplant that are not outsourced, these need to beidentified in the documents. These could includecompressed air for controls, domestic water, hotwater, etc. A method of allocating costs for sharedservices will need to be established and managedthrough the duration of the contract.3. Product specificationsThe properties of the supplied service need to beadequately described to judge if the supplier ismeeting the terms of the contract. Quantities liketemperature, water treatment values, pressure, etc.needs to be well defined.4. Early TerminationThere should be several options in the contract forearly termination. The most obvious is for lack ofperformance. In this case, lack of performance canrange from total disruption of service to not meetingthe defined values of the commodity to lettingthe equipment deteriorate. There should also bethe ability to have the building owner terminatethe contract if the building owner decides that theywant to take the central plant in-house or find anothercontractor. If the supplier is in default, thena “make whole” payment would be required of thebuilding to terminate the contract in this case.Risk ManagementAs with any long-term commitment, the mostimportant task is to identify all of the potential risks,evaluate their consequences and probability and thento formulate strategies that will mitigate the risks. Thiscould be in the form of the contract document languageor other financial instruments for protection. One of themost important areas of risk management mitigation isto choose a supplier that will deliver what is promisedover the entire contract period.1. How to Choose a SupplierIn addition to price, the following factors are importantto the success of a project and should beevaluated before selecting a supplier.• Track record• Knowledge of your business, priorities andrisk tolerance• Size• Financial backing• Customer service and reporting• “Staying Power”2. Long Term ContractsBecause the potential supplier will be investingcapital for increased life, reliability and efficiency,the contract needs to be long enough to recoverthe costs and provide a positive cash flow. Thelength of the project can vary from three to fiveyears for a simple, small-scale project up to ten totwenty years for one of increased complexity. Costimpacts at the termination of the contract needs tobe adequately addressed, such as:• Renewals• Buyouts• Equipment leases• Equipment condition at the end of the contract3. Changing Assumptions• Interest rates• Utility rates• Maintenance and repair costs• Areas served (i.e., expansions/renovations/contractions)• Regulations; building specific, environmental,OSHA, local codes, etc.• Utility deregulation4. Other Risks• The impact of planned or unplanned outagesof the central plant• The consequences of the supplier not beingable to maintain chilled water temperature orsteam pressure• “Take or Pay” This provision of a contractrequires the customer to pay a certain amounteven if they do not use the commodity• Defaults and Remedies24.10 SUMMARYThis chapter presented information on the changingworld of the utility industry in the new millennium.Starting in the 80’s with gas deregulation and the passageof the Energy Policy Act of 1992 for electricity,the method of providing and purchasing energy waschanged forever. Utilities began a slow change fromvertically integrated monopolies to providers of regulatedwires and transmission services. Some utilities

UTILITY DEREGULATION AND ENERGY SYSTEM OUTSOURCING 647continued to supply generation services, through theirunregulated enterprises and by independent powerproducers in the deregulated markets while others soldtheir generation assets and became “wires” companies.Customers became confused in the early stages of deregulation,but by the end of the 1990’s some becamemore knowledgeable and successful in buying deregulatednatural gas and electricity.In the early 2000’s, difficulties have developed inthe deregulated utility arena. California rescinded deregulation(except for existing contracts) after shortages,rolling blackouts and price increases sent the utilitiesinto a tailspin. The great blackout of 2003 raises concernsabout the reliability of the transmission system. And theloss of regulated rates provides more challenges to customersand their consultants. However, many customerscontinue to participate in the deregulated marketsto obtain reduced (or stable) prices, reduce their riskof big price swings and incorporate energy reductionprograms with energy procurement programs.Another result of deregulation has been a re-examinationby customers of outsourcing their energy needs.Some customers have “sold” their energy systems toenergy suppliers and are now purchasing Btus insteadof kWhs. The energy industry responded with energyservice business units to meet this new demand for outsourcing.Performance contracting and energy systemoutsourcing can be advantageous when the organizationdoes not have internal expertise to execute these projectsand when other sources of capital are needed. However,performance contracting and energy system outsourcingis not without peril if the risks are not understood andmitigated. Before undertaking a performance contract orenergy system outsourcing project, the owner or managerfirst needs to define the financial, technical, legaland operational issues of importance. Next, the properresources, whether internal or outsourced, need to bemarshaled to define the project, prepare the Requestfor Proposal, evaluate the suppliers and bids, negotiatea contract and monitor the results, often over a longperiod. If these factors are properly considered andexecuted, the performance contract or energy systemoutsourcing often produce results that could not beobtained via other project methods.BIBLIOGRAPHYPower Shopping and Power Shopping II, A publication of theBuilding Owners and Managers Association (BOMA)International, 1201 New York Avenue, N.W., N.W., Suite300, Washington, DC 20005.The Changing Structure of the Electric Power Industry: HistoricalOverview, United States Department of Energy, EnergyInformation Administration, Washington, DC.The Ten Step Program to Successful Utility Deregulation for BuildingOwners and Managers, George R Owens PE CEM,President Energy and Engineering Solutions, Inc. (EESI),9449 Penfield Ct., Columbia, MD 21045.Performance Contracting and Energy System Outsourcing, GeorgeR Owens PE CEM, President Energy and EngineeringSolutions, Inc. (EESI), 9449 Penfield Ct., Columbia, MD21045.Generating Power and Getting It to The Consumer, Edison ElectricInstitute, 701 Pennsylvania Ave NW, Washington, DC,20004.The Changing Structure of the Electric Power Industry: An Update,US Department of Energy, Energy Information Administration,DOE/EIA-0562(96)PJM Electricity Futures, New York Mercantile Exchange (NY-MEX) web page, www.nymex.comSOME USEFUL INTERNET RESOURCES10 Step paper - www.eesienergy.comState activities - www.eia.doe.gov/cneaf/electricity/chg_str/State regulatory commissions www.naruc.orgUtilities - www.utilityconnection.comMaillist - AESP-NET@AESP.org


CHAPTER 25FINANCING ENERGY MANAGEMENT PROJECTSERIC A. WOODRUFF, PH.D., CEM, CEP, CLEPJohnson Controls, Inc.25.1 INTRODUCTIONFinancing can be a key success factor for projects.This chapter’s purpose is to help facility managersunderstand and apply the financial arrangements availableto them. Hopefully, this approach will increasethe implementation rate of good energy managementprojects, which would have otherwise been cancelled orpostponed due to lack of funds.Most facility managers agree that energy managementprojects (EMPs) are good investments. Generally,EMPs reduce operational costs, have a low risk/rewardratio, usually improve productivity and even have beenshown to improve a firm’s stock price. 1 Despite thesebenefits, many cost-effective EMPs are not implementeddue to financial constraints. A study of manufacturingfacilities revealed that first-cost and capital constraintsrepresented over 35% of the reasons cost-effective EMPswere not implemented. 2 Often, the facility manager doesnot have enough cash to allocate funding, or can not getbudget approval to cover initial costs. Financial arrangementscan mitigate a facility’s funding constraints, 3 allowingadditional energy savings to be reaped.Alternative finance arrangements can overcomethe “initial cost” obstacle, allowing firms to implementmore EMPs. However, many facility managers are eitherunaware or have difficulty understanding the varietyof financial arrangements available to them. Most facilitymanagers use simple payback analyses to evaluateprojects, which do not reveal the added value ofafter-tax benefits. 4 Sometimes facility managers do notimplement an EMP because financial terminology andcontractual details intimidate them. 5To meet the growing demand, there has been adramatic increase in the number of finance companiesspecializing in EMPs. At a recent World Energy EngineeringCongress, finance companies represented themost common exhibitor type. These financiers are introducingnew payment arrangements to implement EMPs.Often, the financier’s innovation will satisfy the uniquecustomer needs of a large facility. This is a great servicehowever, most financiers are not attracted to small facilitieswith EMPs requiring less than $100,000. Thus, manyfacility managers remain unaware or confused about thecommon financial arrangements that could help themimplement EMPs.Numerous papers and government programs havebeen developed to show facility managers how to usequantitative (economic) analysis to evaluate financialarrangements. 4,5,6 (Refer to Chapter 4 of this book.)Quantitative analysis includes computing the simple payback,net present value (NPV), internal rate of return (IRR), or lifecyclecost of a project with or without financing. Althoughthese books and programs show how to evaluate theeconomic aspects of projects, they do not incorporatequalitative factors like strategic company objectives,(which can impact the financial arrangement selection).Without incorporating a facility manager’s qualitativeobjectives, it is hard to select an arrangement that meetsall of the facility’s needs. A recent paper showed thatqualitative objectives can be at least as important asquantitative objectives. 9This chapter hopes to provide some valuable information,which can be used to overcome the previouslymentioned issues. The chapter is divided into severalsections to accomplish three objectives. Sections 2 and3 introduce the basic financial arrangements via a simpleexample. In sections 4 and 5, financial terminology isdefined and each arrangement is explained in greaterdetail while applied to a case study. The remaining sectionsshow how to match financial arrangements to differentprojects and facilities.25.2 FINANCIAL ARRANGEMENTS:A SIMPLE EXAMPLEConsider a small company “PizzaCo” that makesfrozen pizzas, and distributes them regionally. PizzaCouses an old delivery truck that breaks down frequentlyand is inefficient. Assume the old truck has no salvagevalue and is fully depreciated. PizzaCo’s managementwould like to obtain a new and more efficient truck toreduce expenses and improve reliability. However, theydo not have the cash on hand to purchase the truck.Thus, they consider their financing options.649

650 ENERGY MANAGEMENT HANDBOOK25.2.1 Purchase the Truck with a Loan or BondJust like most car purchases, PizzaCo borrowsmoney from a lender (a bank) and agrees to a monthlyre-payment plan. Figure 25.1 shows PizzaCo’s annualcash flows for a loan. The solid arrows represent thefinancing cash flows between PizzaCo and the bank.Each year, PizzaCo makes payments (on the principal,plus interest based on the unpaid balance), until thebalance owed is zero. The payments are the negativecash flows. Thus, at time zero when PizzaCo borrowsthe money, they receive a large sum of money from thebank, which is a positive cash flow (which will be usedto purchase the truck).The dashed arrows represent the truck purchase aswell as savings cash flows. Thus, at time zero, PizzaCopurchases the truck (a negative cash flow) with themoney from the bank. Due to the new truck’s greater efficiency,PizzaCo’s annual expenses are reduced (whichis a savings). The annual savings are the positive cashflows. The remaining cash flow diagrams in this chapterutilize the same format.PizzaCo could also purchase the truck by selling abond. This arrangement is similar to a loan, except investors(not a bank) give PizzaCo a large sum of money(called the bond’s “par value”). Periodically, PizzaCowould pay the investors only the interest accumulated.As Figure 25.2 shows, when the bond reaches maturity,PizzaCo returns the par value to the investors. Theequipment purchase and savings cash flows are thesame as with the loan.25.2.2 Sell Stock to Purchase the TruckIn this arrangement, PizzaCo sells its stock to raisemoney to purchase the truck. In return, PizzaCo is expectedto pay dividends back to shareholders. Sellingstock has a similar cash flow pattern as a bond, with afew subtle differences. Instead of interest payments tobondholders, PizzaCo would pay dividends to shareholdersuntil some future date when PizzaCo could buythe stock back. However, these dividend payments arenot mandatory, and if PizzaCo is experiencing financialstrain, it does not need to distribute dividends. On theother hand, if PizzaCo’s profits increase, this wealth willbe shared with the new stockholders, because they nowown a part of the company.25.2.3 Rent the TruckJust like renting a car, PizzaCo could rent a truckfor an annual fee. This would be equivalent to a truelease. The rental company (lessor) owns and maintainsthe truck for PizzaCo (the lessee). PizzaCo pays therental fees (lease payments) which are considered taxdeductiblebusiness expenses.Figure 25.3 shows that the lease payments (solidarrows) start as soon as the equipment is leased (yearzero) to account for lease payments paid in advance.Lease payments “in arrears” (starting at the end of thefirst year) could also be arranged. However, the leasingcompany may require a security deposit as collateral.Notice that the savings cash flows are essentially thesame as the previous arrangements, except there is noequipment purchase, which is a large negative cash flowat year zero.Figure 25.2 PizzaCo’s Cash Flows for a Bond.Figure 25.1 PizzaCo’s Cash Flows for a Loan.Figure 25.3. PizzaCo’s Cash Flows for a True Lease.

FINANCING ENERGY MANAGEMENT PROJECTS 651In a true lease, the contract period should be shorterthan the equipment’s useful life. The lease is cancelablebecause the truck can be leased easily to someoneelse. At the end of the lease, PizzaCo can either returnthe truck or renew the lease. In a separate transaction,PizzaCo could also negotiate to buy the truck at the fairmarket value.If PizzaCo wanted to secure the option to buy thetruck (for a bargain price) at the end of the lease, thenthey would use a capital lease. A capital lease can bestructured like an installment loan, however ownershipis not transferred until the end of the lease. Thelessor retains ownership as security in case the lessee(PizzaCo) defaults on payments. Because the entire costof the truck is eventually paid, the lease payments arelarger than the payments in a true lease, (assumingsimilar lease periods). Figure 25.4 shows the cash flowsfor a capital lease with advance payments and a bargainpurchase option at the end of year five.There are some additional scenarios for lease arrangements.A “vendor-financed” agreement is whenthe lessor (or lender) is the equipment manufacturer.Alternatively, a third party could serve as a financingsource. With “third party financing,” a finance companywould purchase a new truck and lease it to PizzaCo.In either case, there are two primary ways to repay thelessor.1. With a “fixed payment plan”; where payments aredue whether or not the new truck actually savesmoney.2. With a “flexible payment plan”; where the savingsfrom the new truck are shared with the thirdparty, until the truck’s purchase cost is recoupedwith interest. This is basically a “shared savings”arrangement.25.2.4 Subcontract Pizza Delivery to a Third PartySince PizzaCo’s primary business is not delivery,it could subcontract that responsibility to another company.Let’s say that a delivery service company wouldprovide a truck and deliver the pizzas at a reduced cost.Each month, PizzaCo would pay the delivery servicecompany a fee. However, this fee is guaranteed to beless than what PizzaCo would have spent on delivery.Thus, PizzaCo would obtain savings without investingany money or risk in a new truck. This arrangement isanalogous to a performance contract.This arrangement is very similar to a third-partylease and a shared savings agreement. However with aperformance contract, the contractor assumes most ofthe risk, (because they supply the equipment, with littleor no investment from PizzaCo). The contractor alsois responsible for ensuring that the delivery fee is lessthan what PizzaCo would have spent. For the PizzaCoexample, the arrangement would designed under theconditions below.• The delivery company owns and maintains thetruck. It also is responsible for all operations relatedto delivering the pizzas.• The monthly fee is related to the number of pizzasdelivered. This is the performance aspect of thecontract; if PizzaCo doesn’t sell many pizzas, thefee is reduced. A minimum amount of pizzas may berequired by the delivery company (performance contractor)to cover costs. Thus, the delivery companyassumes these risks:1. PizzaCo will remain solvent, and2. PizzaCo will sell enough pizzas to covercosts, and3. the new truck will operate as expected andwill actually reduce expenses per pizza, and4. the external financial risk, such as inflationand interest rate changes, are acceptable.• Because the delivery company is financially strongand experienced, it can usually obtain loans at lowinterest rates.• The delivery company is an expert in delivery; ithas specially skilled personnel and uses efficientequipment. Thus, the delivery company can deliverthe pizzas at a lower cost (even after addinga profit) than PizzaCo.Figure 25.4 PizzaCo’s Cash Flows for a Capital Lease.Figure 25.5 shows the net cash flows according toPizzaCo. Since the delivery company simply reduces

652 ENERGY MANAGEMENT HANDBOOKPizzaCo’s operational expenses, there is only a net savings.There are no negative financing cash flows. Unlikethe other arrangements, the delivery company’s fee is aless expensive substitute for PizzaCo’s in-house deliveryexpenses. With the other arrangements, PizzaCo had topay a specific financing cost (loan, bond or lease payments,or dividends) associated with the truck, whetheror not the truck actually saved money. In addition, PizzaCowould have to spend time maintaining the truck,which would detract from its core focus: making pizzas.With a performance contract, the delivery company ispaid from the operational savings it generates. Becausethe savings are greater than the fee, there is a net savings.Often, the contractor guarantees the savings.the benefits provided by the equipment. In the SimpleExample, the benefit was the pizza delivery. PizzaCo was notconcerned with what type of truck was used.The decision to purchase or utilize equipment ispartly dependent on the company’s strategic focus. If acompany wants to delegate some or all of the responsibilityof managing a project, it should use a true lease,or a performance contact. 10 However, if the companywants to be intricately involved with the EMP, purchasingand self-managing the equipment could yield thegreatest profits. When the building owner purchasesequipment, he/she usually maintains the equipment,and lists it as an asset on the balance sheet so it can bedepreciated.Financing for purchases has two categories:1. Debt Financing, which is borrowing money fromsomeone else, or another firm. (using loans, bondsand capital leases)2. Equity Financing, which is using money from yourcompany, or your stockholders. (using retainedearnings, or issuing common stock)Figure 25.5 PizzaCo’s Cash Flows for a PerformanceContract.Supplementary Note: Combinations of the basic financearrangements are possible. For example, a shared savings arrangementcan be structured within a performance contract.Also, performance contracts are often designed so that thefacility owner (PizzaCo) would own the asset at the end ofthe contract.25.3 FINANCIAL ARRANGEMENTS:DETAILS AND TERMINOLOGYTo explain the basic financial arrangements inmore detail, each one is applied to an energy management-relatedcase study. To understand the economicsbehind each arrangement, some finance terminology ispresented below.25.3.1 Finance TerminologyEquipment can be purchased with cash on-hand(officially labeled “retained earnings”), a loan, a bond,a capital lease or by selling stock. Alternatively, equipmentcan be utilized with a true lease or with a performancecontract.Note that with performance contracting, the buildingowner is not paying for the equipment itself, butIn all cases, the borrower will pay an interestcharge to borrow money. The interest rate is called the“cost of capital.” The cost of capital is essentially dependenton three factors: (1) the borrower’s credit rating, (2)project risk and (3) external risk. External risk can includeenergy price volatility, industry-specific economicperformance as well as global economic conditions andtrends. The cost of capital (or “cost of borrowing”) influencesthe return on investment. If the cost of capitalincreases, then the return on investment decreases.The “minimum attractive rate of return” (MARR)is a company’s “hurdle rate” for projects. Because manyorganizations have numerous projects “competing” for funding,the MARR can be much higher than interest earned froma bank, or other risk-free investment. Only projects with areturn on investment greater than the MARR should beaccepted. The MARR is also used as the discount rateto determine the “net present value” (NPV).25.3.2 Explanation of Figures and TablesThroughout this chapter’s case study, figures arepresented to illustrate the transactions of each arrangement.Tables are also presented to show how to performthe economic analyses of the different arrangements.The NPV is calculated for each arrangement.It is important to note that the NPV of a particulararrangement can change significantly if the cost of capital,MARR, equipment residual value, or project life is ad-

FINANCING ENERGY MANAGEMENT PROJECTS 653justed. Thus, the examples within this chapter are providedonly to illustrate how to perform the analyses. The cashflows and interest rates are estimates, which can vary fromproject to project. To keep the calculations simple, end-ofyearcash flows are used throughout this chapter.Within the tables, the following abbreviations andequations are used:EOY = End of YearSavings = pre-Tax Cash FlowDepr. = DepreciationTaxable Income = Savings - Depreciation - InterestPaymentTax = (Taxable Income)*(Tax Rate)ATCF = After Tax Cash Flow =Savings – Total Payments – TaxesTable 25.1 shows the basic equations that are usedto calculate the values under each column headingwithin the economic analysis tables.been fully depreciated, he/she can claim the book value as atax-deduction.*25.4 APPLYING FINANCIAL ARRANGEMENTS:A CASE STUDYSuppose PizzaCo (the “host” facility) needs a newchilled water system for a specific process in its manufacturingplant. The installed cost of the new system is$2.5 million. The expected equipment life is 15 years,however the process will only be needed for 5 years,after which the chilled water system will be sold at anestimated market value of $1,200,000 (book value atyear five = $669,375). The chilled water system shouldsave PizzaCo about $1 million/year in energy savings.PizzaCo’s tax rate is 34%. The equipment’s annual maintenanceand insurance cost is $50,000. PizzaCo’s MARRis 18%. Since at the end of year 5, PizzaCo expects to sellthe asset for an amount greater than its book value, theTable 25.1 Table of Sample Equations used in Economic Analyses.———————————————————————————————————————————————————A B C D E F G H I J———————————————————————————————————————————————————Payments Principal TaxableEOY Savings Depreciation Principal Interest Total Outstanding Income Tax ATCF———————————————————————————————————————————————————nn+1 = (MACRS %)* =(D) +(E) =(G at year n) =(B)–(C)–(E) =(H)*(tax rate) =(B)–(F)–(I)n+2 (Purchase Price) –(D at year n+1)———————————————————————————————————————————————————Regarding depreciation, the “modified acceleratedcost recovery system” (MACRS) is used in the economicanalyses. This system indicates the percent depreciationclaimable year-by-year after the equipmentis purchased. Table 25.2 shows the MACRS percentagesfor seven-year property. For example, after the firstyear, an owner could depreciate 14.29% of an equipment’svalue. The equipment’s “book value” equals the remainingunrecovered depreciation. Thus, after the first year, the bookvalue would be 100%-14.29%, which equals 85.71% of theoriginal value. If the owner sells the property before it has*To be precise, the IRS uses a “half-year convention” for equipmentthat is sold before it has been completely depreciated. In the tax yearthat the equipment is sold, (say year “x”) the owner claims only Ωof the MACRS depreciation percent for that year. (This is becausethe owner has only used the equipment for a fraction of the finalyear.) Then on a separate line entry, (in the year “x*”), the remainingunclaimed depreciation is claimed as “book value.” The x* year ispresented as a separate line item to show the book value treatment,however x* entries occur in the same tax year as “x.”Table 25.2 MACRS Depreciation Percentages.—————————————————————————EOY MACRS Depreciation Percentagesfor 7-Year Property—————————————————————————0 0—————————————————————————1 14.29%—————————————————————————2 24.49%—————————————————————————3 17.49%—————————————————————————4 12.49%—————————————————————————5 8.93%—————————————————————————6 8.92%—————————————————————————7 8.93%—————————————————————————8 4.46%—————————————————————————

654 ENERGY MANAGEMENT HANDBOOKadditional revenues are called a “capital gain,” (whichequals the market value – book value) and are taxed.If PizzaCo sells the asset for less than its book value,PizzaCo incurs a “capital loss.”PizzaCo does not have $2.5 million to pay for thenew system, thus it considers its finance options. PizzaCois a small company with an average credit rating,which means that it will pay a higher cost of capital thana larger company with an excellent credit rating. As withany borrowing arrangement, if investors believe that aninvestment is risky, they will demand a higher interestrate.25.4.1 Purchase Equipment withRetained Earnings (Cash)If PizzaCo did have enough retained earnings(cash on-hand) available, it could purchase the equipmentwithout external financing. Although externalfinance expenses would be zero, the benefit of tax-deductions(from interest expenses) is also zero. Also, anycash used to purchase the equipment would carry an“opportunity cost,” because that cash could have beenused to earn a return somewhere else. This opportunitycost rate is usually set equal to the MARR. In otherwords, the company lost the opportunity to invest thecash and gain at least the MARR from another investment.Of all the arrangements described in this chapter,purchasing equipment with retained earnings is probablythe simplest to understand. For this reason, itwill serve as a brief example and introduction to theeconomic analysis tables that are used throughout thischapter.25.4.1.1 Application to the Case StudyFigure 25.6 illustrates the resource flows betweenthe parties. In this arrangement, PizzaCo purchases thechilled water system directly from the equipment manufacturer.Once the equipment is installed, PizzaCo recoversthe full $1 million/year in savings for the entire fiveyears, but must spend $50,000/year on maintenance andinsurance. At the end of the five-year project, PizzaCoexpects to sell the equipment for its market value of$1,200,000. Assume MARR is 18%, and the equipmentis classified as 7-year property for MACRS depreciation.Table 25.3 shows the economic analysis for purchasingthe equipment with retained earnings.Reading Table 25.3 from left to right, and top tobottom, at EOY 0, the single payment is entered intothe table. Each year thereafter, the savings as well asthe depreciation (which equals the equipment purchaseprice multiplied by the appropriate MACRS % for eachyear) are entered into the table. Year by year, the taxableincome = savings – depreciation. The taxable income isthen taxed at 34% to obtain the tax for each year. Theafter-tax cash flow = savings - tax for each year.At EOY 5, the equipment is sold before the entirevalue was depreciated. EOY 5* shows how the equipmentsale and book value are claimed. In summary, theNPV of all the ATCFs would be $320,675.25.4.2 LoansLoans have been the traditional financial arrangementfor many types of equipment purchases. A bank’swillingness to loan depends on the borrower’s financialhealth, experience in energy management and numberof years in business. Obtaining a bank loan can be difficultif the loan officer is unfamiliar with EMPs. Loanofficers and financiers may not understand energyrelatedterminology (demand charges, kVAR, etc.). Inaddition, facility managers may not be comfortable withthe financier’s language. Thus, to save time, a bank thatcan understand EMPs should be chosen.Most banks will require a down payment and collateralto secure a loan. However, securing assets canbe difficult with EMPs because the equipment often becomespart of the real estate of the plant. For example, itwould be very difficult for a bank to repossess lighting fixturesfrom a retrofit. In these scenarios, lenders may be willingto secure other assets as collateral.BankChilled WaterSystem ManufacturerPurchaseAmountEquipmentPizzaCoChilled WaterSystem ManufacturerLoan PrincipalPurchaseAmountEquipmentPizzaCoFigure 25.6 Resource Flows for Using Retained Earnings25.7 Resource Flow Diagram for a Loan.

FINANCING ENERGY MANAGEMENT PROJECTS 655Table 25.3 Economic Analysis for Using Retained Earnings.——————————————————————————————————————————————EOY Savings Depr. Payments Principal Taxable Tax ATCFPrincipal Interest Total Outstanding Income——————————————————————————————————————————————0 2,500,000 -2,500,0001 950,000 357,250 592,750 201,535 748,4652 950,000 612,250 337,750 114,835 835,1653 950,000 437,250 512,750 174,335 775,6654 950,000 312,250 637,750 216,835 733,1655 950,000 111,625 838,375 285,048 664,9535* 1,200,000 669,375 530,625 180,413 1,019,588——————————————————————————————————————————————2,500,000Net Present Value at 18%: $320,675——————————————————————————————————————————————Notes: Loan Amount: 0Loan Finance Rate: 0% MARR 18%Tax Rate 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: =(Market Value - Book Value)=(1,200,000 - 669,375) = $530,625——————————————————————————————————————————————25.4.2.1 Application to the Case StudyFigure 25.7 illustrates the resource flows betweenthe parties. In this arrangement, PizzaCo purchases thechilled water system with a loan from a bank. PizzaComakes equal payments (principal + interest) to the bankfor five years to retire the debt. Due to PizzaCo’s smallsize, credibility, and inexperience in managing chilledwater systems, PizzaCo is likely to pay a relatively highcost of capital. For example, let’s assume 15%.PizzaCo recovers the full $1 million/year in savingsfor the entire five years, but must spend $50,000/year on maintenance and insurance. At the end of thefive-year project, PizzaCo expects to sell the equipmentfor its market value of $1,200,000. Tables 25.4 and 25.5show the economic analysis for loans with a zero downpayment and a 20% down payment, respectively. Assumethat the bank reduces the interest rate to 14% forthe loan with the 20% down payment. Since the assetis listed on PizzaCo’s balance sheet, PizzaCo can usedepreciation benefits to reduce the after-tax cost. In addition,all loan interest expenses are tax-deductible.25.4.3 BondsBonds are very similar to loans; a sum of money isborrowed and repaid with interest over a period of time.The primary difference is that with a bond, the issuer(PizzaCo) periodically pays the investors only the interestearned. This periodic payment is called the “couponinterest payment.” For example, a $1,000 bond with a 10%coupon will pay $100 per year. When the bond matures, theissuer returns the face value ($1,000) to the investors.Bonds are issued by corporations and governmententities. Government bonds generate tax-free income forinvestors, thus these bonds can be issued at lower ratesthan corporate bonds. This benefit provides governmentfacilities an economic advantage to use bonds to financeprojects.25.4.3.1 Application to the Case StudyAlthough PizzaCo (a private company) would notbe able to obtain the low rates of a government bond,they could issue bonds with coupon interest rates competitivewith the loan interest rate of 15%.In this arrangement, PizzaCo receives the investors’cash (bond par value) and purchases the equipment.PizzaCo uses part of the energy savings to paythe coupon interest payments to the investors. When thebond matures, PizzaCo must then return the par valueto the investors. See Figure 25.8.As with a loan, PizzaCo owns, maintains and depreciatesthe equipment throughout the project’s life. Allcoupon interest payments are tax-deductible. At the end

656 ENERGY MANAGEMENT HANDBOOKTable 25.4 Economic Analysis for a Loan with No Down Payment.——————————————————————————————————————————————EOY Savings Depr. Payments Principal Taxable Tax ATCFPrincipal Interest Total Outstanding Income——————————————————————————————————————————————0 2,500,0001 950,000 357,250 370,789 375,000 745,789 2,129,211 217,750 74,035 130,1762 950,000 612,250 426,407 319,382 745,789 1,702,804 18,368 6,245 197,9663 950,000 437,250 490,368 255,421 745,789 1,212,435 257,329 187,492 116,7194 950,000 312,250 563,924 181,865 745,789 648,511 455,885 55,001 49,2105 950,000 111,625 648,511 97,277 745,789 0 741,098 251,973 -47,7615* 1,200,000 669,375 530,625 180,413 1,019,5882,500,000Net Present Value at 18%: $757,121——————————————————————————————————————————————Notes: Loan Amount: 2,500,000 (used to purchase equipment at year 0)Loan Finance Rate: 15% MARR 18%Tax Rate 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: =(Market Value - Book Value)=(1,200,000 - 669,375) = $530,625——————————————————————————————————————————————Table 25.5 Economic Analysis for a Loan with a 20% Down-Payment,——————————————————————————————————————————————EOY Savings Depr. Payments Principal Taxable Tax ATCFPrincipal Interest Total Outstanding Income——————————————————————————————————————————————0 500,000 2,000,000 –500,0001 950,000 357,250 302,567 280,000 582,567 1,697,433 312,750 106,335 261,0982 950,000 612,250 344,926 237,641 582,567 1,352,507 100,109 34,037 333,3963 950,000 437,250 393,216 189,351 582,567 959,291 323,399 109,956 257,4774 950,000 312,250 448,266 134,301 582,567 511,0241 503,449 171,173 196,2605 950,000 111,625 511,024 71,543 582,567 0 766,832 260,723 106,7105* 1,200,000 669,375 530,625 180,413 1,019,588——————————————————————————————————————————————2,500,000Net Present Value at 18%: $710,962——————————————————————————————————————————————Notes: Loan Amount: 2,000,000 (used to purchase equipment at year 0)Loan Finance Rate: 14% MARR 18%500,000 Tax Rate 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: =(Market Value - Book Value)=(1,200,000 - 669,375) = $530,625——————————————————————————————————————————————

FINANCING ENERGY MANAGEMENT PROJECTS 657Chilled WaterSystem ManufacturerPurchaseAmountEquipmentBondInvestorsPizzaCoPaymentsFigure 25.8 Resource Flow Diagram for a Bond.of the five-year project, PizzaCo expects to sell the equipmentfor its market value of $1,200,000. Table 25.6 showsthe economic analysis of this finance arrangement.25.4.4 Selling StockAlthough less popular, selling company stock isan equity financing option which can raise capital forprojects. For the host, selling stock offers a flexible repaymentschedule, because dividend payments to shareholdersaren’t absolutely mandatory. Selling stock is alsooften used to help a company attain its desired capitalstructure. However, selling new shares of stock dilutesthe power of existing shares and may send an inaccurate“signal” to investors about the company’s financialstrength. If the company is selling stock, investors maythink that it is desperate for cash and in a poor financialcondition. Under this belief, the company’s stock pricecould decrease. However, recent research indicates thatwhen a firm announces an EMP, investors react favorably.11 On average, stock prices were shown to increaseabnormally by 21.33%.By definition, the cost of capital (rate) for sellingstock is:cost of capital selling stock = D/Pwhere D = annual dividend paymentP = company stock priceHowever, in most cases, the after-tax cost of capitalfor selling stock is higher than the after-tax cost of debtfinancing (using loans, bonds and capital leases). This isbecause interest expenses (on debt) are tax deductible,but dividend payments to shareholders are not.In addition to tax considerations, there are otherreasons why the cost of debt financing is less than thefinancing cost of selling stock. Lenders and bond buyers(creditors) will accept a lower rate of return because theyare in a less risky position due to the reasons below.• Creditors have a contract to receive money at acertain time and future value (stockholders haveno such guarantee with dividends).• Creditors have first claim on earnings (interest ispaid before shareholder dividends are allocated).Table 25.6 Economic Analysis for a Bond.——————————————————————————————————————————————EOY Savings Depr. Payments Principal Taxable Tax ATCFPrincipal Interest Total Outstanding Income——————————————————————————————————————————————0 2,500,0001 950,000 357,250 375,000 375,000 2,500,000 217,750 74,035 500,9652 950,000 612,250 375,000 375,000 2,500,000 -37,250 -12,665 587,6653 950,000 437,250 375,000 375,000 2,500,000 137,750 46,835 528,1654 950,0 0 312,250 375,000 375,000 2,500,000 262,750 89,335 485,6655 950,000 111,625 2,500,000 375,000 2,875,000 0 463,375 157,548 -2,082,5485* 1,200,000 669,375 530,625 180,413 1,019,5882,500,000Net Present Value at 18%: 953,927——————————————————————————————————————————————Notes: Loan Amount: 2,500,000 (used to purchase equipment at year 0)Loan Finance Rate: 0% MARR 18%Tax Rate 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: =(Market Value - Book Value)=(1,200,000 - 669,375) = $530,625——————————————————————————————————————————————

658 ENERGY MANAGEMENT HANDBOOK• Creditors usually have secured assets as collateraland have first claim on assets in the event of bankruptcy.Despite the high cost of capital, selling stock doeshave some advantages. This arrangement does not bindthe host to a rigid payment plan (like debt financingagreements) because dividend payments are not mandatory.The host has control over when it will pay dividends.Thus, when selling stock, the host receives greaterpayment flexibility, but at a higher cost of capital.25.4.4.1 Application to the Case StudyAs Figure 25.9 shows, the financial arrangement isvery similar to a bond, at year zero the firm receives $2.5million, except the funds come from the sale of stock.Instead of coupon interest payments, the firm distributesdividends. At the end of year five, PizzaCo repurchasesthe stock. Alternatively, PizzaCo could capitalize thedividend payments, which means setting aside enoughmoney so that the dividends could be paid with theinterest generated.Table 25.7 shows the economic analysis for issuingstock at a 16% cost of equity capital, and repurchasingthe stock at the end of year five. (For consistency ofcomparison to the other arrangements, the stock pricedoes not change during the contract.) Like a loan orbond, PizzaCo owns and maintains the asset. Thus, theannual savings are only $950,000. PizzaCo pays annualdividends worth $400,000. At the end of year 5, PizzaCoexpects to sell the asset for $1,200,000.Note that Table 25.7 is slightly different from theother tables in this chapter:Taxable Income = Savings – Depreciation, andATCF = Savings – Stock Repurchases - Dividends- Tax25.4.5 LeasesFirms generally own assets, however it is the use ofthese assets that is important, not the ownership. Leasingis another way of obtaining the use of assets. Thereare numerous types of leasing arrangements, rangingChilled WaterSystem ManufacturerPurchaseAmountEquipmentCashInvestorsPizzaCoSellStockFigure 25.9 Resource Flow Diagram for Selling Stock.Table 25.7 Economic Analysis of Selling Stock.——————————————————————————————————————————————EOY Savings Depr. Stock Transactions Dividend Taxable Tax ATCFSale of Stock Repurchase Payments Income——————————————————————————————————————————————0 $2,500,000 from Stock Sale is used to purchase equipment, thur ATCF = 01 950,000 357,250 400,000 592,750 201,535 348,4652 950,000 612,250 400,000 337,750 114,835 435,1653 950,000 437,250 400,000 512,750 174,335 375,6654 950,000 312,250 400,000 637,750 216,835 333,1655 950,000 111,625 2,500,000 400,000 838,375 285,048 2,235,0485* 1,200,000 669,375 530,625 180,413 1,019,5882,500,000——————————————————————————————————————————————Net Present Value at 18%: 477,033——————————————————————————————————————————————Notes: Value of Stock Sold (which is repurchased after year 5 2,500,000 (used to purchase equipment at year 0)Cost of Capital = Annual Dividend Rate: 16% MARR = 18%Tax Rate = 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: = (Market Value - Book Value)= (1,200,000 - 669,375) = $530,625——————————————————————————————————————————————

FINANCING ENERGY MANAGEMENT PROJECTS 659from basic rental agreements to extended payment plansfor purchases. Leasing is used for nearly one-third of allequipment utilization. 12 Leases can be structured andapproved very quickly, even within 48 hours. Table 25.8lists some additional reasons why leasing can be an attractivearrangement for the lessee.Table 25.8 Good Reasons to Lease.—————————————————————————Financial Reasons• With some leases, the entire lease payment is taxdeductible.• Some leases allow “off-balance sheet” financing,preserving credit linesRisk Sharing• Leasing is good for short-term asset use, and reducesthe risk of getting stuck with obsolete equipment• Leasing offers less risk and responsibility—————————————————————————Basically, there are two types of leases; the “truelease” (a.k.a. “operating” or “guideline lease”) and the“capital lease.” One of the primary differences betweena true lease and a capital lease is the tax treatment. In atrue lease, the lessor owns the equipment and receivesthe depreciation benefits. However, the lessee can claimthe entire lease payment as a tax-deductible businessexpense. In a capital lease, the lessee (PizzaCo) ownsand depreciates the equipment. However, only the interestportion of the lease payment is tax-deductible. Ingeneral, a true lease is effective for a short-term project,where the company does not plan to use the equipmentwhen the project ends. A capital lease is effective forlong-term equipment.25.4.5.1 The True LeaseFigure 25.10 illustrates the legal differences betweena true lease and a capital lease. 13 A true lease (oroperating lease) is strictly a rental agreement. The word“strict” is appropriate because the Internal RevenueService will only recognize a true lease if it satisfies thefollowing criteria:1. the lease period must be less than 80% of theequipment’s life, and2. the equipment’s estimated residual value must be≥20% of its value at the beginning of the lease,and3. there is no “bargain purchase option,” andDoes the lessor have:▼≥ 20% investment in asset at all times?yes▼≥20% residual value?yes▼lease period ≤ 80% asset’s life?yes▼Does lessee have:▼a loan to the lessor?no▼a bargain purchase option?no▼True LeaseyesCapital LeaseFigure 25.10 Classification for a True Lease.▼ ▼ ▼ ▼ ▼ ▼4. there is no planned transfer of ownership, and5. the equipment must not be custom-made and onlyuseful in a particular facility.25.4.5.2 Application to the Case StudyIt is unlikely that PizzaCo could find a lessor thatwould be willing to lease a sophisticated chilled watersystem and after five years, move the system to anotherfacility. Thus, obtaining a true lease would be unlikely.However, Figure 25.11 shows the basic relationship betweenthe lessor and lessee in a true lease. A third-partyleasing company could also be involved by purchasingnonono▼▼

660 ENERGY MANAGEMENT HANDBOOKChilled WaterSystem Manufacturer(Lessor)Lease PaymentsLeased EquipmentPizzaCo(Lessee)Figure 25.11 Resource Flow Diagram for a TrueLease.the equipment and leasing to PizzaCo. Such a resourceflow diagram is shown for the capital lease.Table 25.9 shows the economic analysis for a truelease. Notice that the lessor pays the maintenance andinsurance costs, so PizzaCo saves the full $1 million peryear. PizzaCo can deduct the entire lease payment of$400,000 as a business expense. However PizzaCo doesnot obtain ownership, so it can’t depreciate the asset.25.4.5.3 The Capital LeaseThe capital lease has a much broader definitionthan a true lease. A capital lease fulfills any one of thefollowing criteria:1. the lease term ≥80% of the equipment’s life;2. the present value of the lease payments ≥80% ofthe initial value of the equipment;3. the lease transfers ownership;4. the lease contains a “bargain purchase option,”which is negotiated at the inception of the lease.Most capital leases are basically extended paymentplans, except ownership is usually not transferreduntil the end of the contract. This arrangement iscommon for large EMPs because the equipment (suchas a chilled water system) is usually difficult to reuseat another facility. With this arrangement, the lesseeeventually pays for the entire asset (plus interest). Inmost capital leases, the lessee pays the maintenanceand insurance costs.The capital lease has some interesting tax implicationsbecause the lessee must list the asset on its balancesheet from the beginning of the contract. Thus,like a loan, the lessee gets to depreciate the asset andonly the interest portion of the lease payment is taxdeductible.25.4.5.4 Application to the Case StudyFigure 25.12 shows the basic third-party financingrelationship between the equipment manufacturer, lessorand lessee in a capital lease. The finance company(lessor) is shown as a third party, although it also couldbe a division of the equipment manufacturer. Becausethe finance company (with excellent credit) is involved,a lower cost of capital (12%) is possible due to reducedrisk of payment default.Like an installment loan, PizzaCo’s lease paymentscover the entire equipment cost. However, the leasepayments are made in advance. Because PizzaCo isconsidered the owner, it pays the $50,000 annual maintenanceexpenses, which reduces the annual savings to$950,000. PizzaCo receives the benefits of depreciationand tax-deductible interest payments. To be consistentwith the analyses of the other arrangements, PizzaCowould sell the equipment at the end of the lease for itsmarket value. Table 25.10 shows the economic analysisfor a capital lease.Table 25-9 Economic Analysis for a True Lease——————————————————————————————————————————————EOY Savings Depr. Lease Principal Taxable Tax ATCFPayments Total Outstanding Income——————————————————————————————————————————————0 400,000 400,000 -400,000 -400,0001 1,000,000 0 400,000 400,000 600,000 204,000 396,0002 1,000,000 0 400,000 400,000 600,000 204,000 396,0003 1,000,000 0 400,000 400,000 600,000 204,000 396,0004 1,000,000 0 400,000 400,000 600,000 204,000 396,0005 1,000,000 0 1,000,000 340,000 660,000——————————————————————————————————————————————Net Present Value at 18%: $953,757——————————————————————————————————————————————Notes: Annual Lease Payment: 400,000MARR = 18%Tax Rate 34%——————————————————————————————————————————————

FINANCING ENERGY MANAGEMENT PROJECTS 661Chilled WaterSystem ManufacturerLease PaymentsPurchaseAmountEquipmentPizzaCoFinanceCompanyLeasedEquipmentFigure 25.12 Resource Flow Diagram for a CapitalLease.25.4.5.5 The Synthetic LeaseA synthetic lease is a “hybrid” lease that combinesaspects of a true lease and a capital lease. Through carefulstructuring and planning, the synthetic lease appearsas an operating lease for accounting purposes (enablesthe Host to have off-balance sheet financing), yet alsoappears as a capital lease for tax purposes (to obtain depreciationfor tax benefits). Consult your local financingexpert to learn more about synthetic leases; they mustbe carefully structured to maintain compliance with theassociated tax laws.With most types of leases, loans and bonds themonthly payments are fixed, regardless of the equipment’sutilization, or performance. However, shared savingsagreements can be incorporated into certain typesof leases.25.4.6 Performance ContractingPerformance contracting is a unique arrangementthat allows the building owner to make necessaryimprovements while investing very little money upfront.The contractor usually assumes responsibility forpurchasing and installing the equipment, as well asmaintenance throughout the contract. But the uniqueaspect of performance contracting is that the contractoris paid based on the performance of the installedequipment. Only after the installed equipment actuallyreduces expenses does the contractor get paid. Energyservice companies (ESCOs) typically serve as contractorswithin this line of business.Unlike most loans, leases and other fixed paymentarrangements, the ESCO is paid based on the performanceof the equipment. In other words, if the finishedproduct doesn’t save energy or operational costs, thehost doesn’t pay. This aspect removes the incentive to“cut corners” on construction or other phases of the project,as with bid/spec contracting. In fact, often there isan incentive to exceed savings estimates. For this reason,performance contracting usually entails a more “facility-Table 25.10 Economic Analysis for a Capital Lease.——————————————————————————————————————————————EOY Savings Depr. Payments Principal Taxable Tax ATCFPrincipal Interest Total Outstanding Income——————————————————————————————————————————————0 619,218 0 619,218 1,880,782 -619,2181 950,000 357,250 393,524 225,694 619,218 1,487,258 367,056 124,799 205,9832 950,000 612,250 440,747 178,471 619,218 1,046,511 159,279 54,155 276,6273 950,000 437,250 493,637 125,581 619,218 552,874 387,169 131,637 199,1454 950,000 312,250 552,874 66,345 619,218 0 571,405 194,278 136,5035 950,000 111,625 838,375 285,048 664,9535* 1,200,000 669,375 530,625 180,413 1,019,588——————————————————————————————————————————————2,500,000Net Present Value at 18%: $681,953——————————————————————————————————————————————Notes: Total Lease Amount: 2,500,000However, Since the payments are in advance, the first payment is analogous to a Down-PaymentThus the actual amount borrowed is only = 2,500,000 - 619,218 = 1,880,782Lease Finance Rate: 12% MARR 18%Tax Rate 34%MACRS Depreciation for 7-Year Property, with half-year convention at EOY 5Accounting Book Value at end of year 5: 669,375Estimated Market Value at end of year 5: 1,200,000EOY 5* illustrates the Equipment Sale and Book ValueTaxable Income: = (Market Value - Book Value)= (1,200,000 - 669,375) = $530,625——————————————————————————————————————————————

662 ENERGY MANAGEMENT HANDBOOKwide” scope of work (to find extra energy savings), thanloans or leases on particular pieces of equipment.With a facility-wide scope, many improvementscan occur at the same time. For example, lighting and airconditioning systems can be upgraded at the same time.In addition, the indoor air quality can be improved.With a comprehensive facility management approach,a “domino-effect” on cost reduction is possible. For example,if facility improvements create a safer and higherquality environment for workers, productivity couldincrease. As a result of decreased employee absenteeism,the workman’s compensation cost could also bereduced. These are additional benefits to the facility.Depending on the host’s capability to manage therisks (equipment performance, financing, etc.) the hostwill delegate some of these responsibilities to the ESCO.In general, the amount of risk assigned to the ESCOis directly related to the percent savings that must beshared with the ESCO.For facilities that are not in a good position to managethe risks of an energy project, performance contractingmay be the only economically feasible implementationmethod. For example, the US Federal Government usedperformance contracting to upgrade facilities when budgetswere being dramatically cut. In essence, they “sold” some oftheir future energy savings to an ESCO, in return for receivingnew equipment and efficiency benefits.In general, performance contracting may be thebest option for facilities that:• are severely constrained by their cash flows;• have a high cost of capital;• don’t have sufficient resources, such as a lack ofin-house energy management expertise or an inadequatemaintenance capacity*;• are seeking to reduce in-house responsibilities andfocus more on their core business objectives; or• are attempting a complex project with uncertainreliability or if the host is not fully capable ofmanaging the project. For example, a lighting retrofithas a high probability of producing the expected cashflows, whereas a completely new process does not have*Maintenance capacity represents the ability that the maintenancepersonnel will be able to maintain the new system. It has been shownthat systems fail and are replaced when maintenance concerns arenot incorporated into the planning process. See Woodroof, E. (1997)“Lighting Retrofits: Don’t Forget About Maintenance,” Energy Engineering,94(1) pp. 59-68.the same “time-tested” reliability. If the in-house energymanagement team cannot manage this risk, performancecontracting may be an attractive alternative.Performance contracting does have some drawbacks.In addition to sharing the savings with an ESCO,the tax benefits of depreciation and other economic benefitsmust be negotiated. Whenever large contracts areinvolved, there is reason for concern. One study foundthat 11% of customers who were considering EMPsfelt that dealing with an ESCO was too confusing orcomplicated. 14 Another reference claims, “with complexcontracts, there may be more options and more room forerror.” 15 Therefore, it is critical to choose an ESCO witha good reputation and experience within the types offacilities that are involved.There are a few common types of contracts. TheESCO will usually offer the following options:• guaranteed fixed dollar savings;• guaranteed fixed energy (MMBtu) savings;• a percent of energy savings; or• a combination of the above.Obviously, facility managers would prefer the optionswith “guaranteed savings.” However this extrasecurity (and risk to the ESCO) usually costs more. Theprimary difference between the two guaranteed optionsis that guaranteed fixed dollar savings contracts ensuredollar savings, even if energy prices fall. For example, ifenergy prices drop and the equipment does not save as muchmoney as predicted, the ESCO must pay (out of its ownpocket) the contracted savings to the host.Percent energy savings contracts are agreementsthat basically share energy savings between the host andthe ESCO. The more energy saved, the higher the revenuesto both parties. However, the host has less predictablesavings and must also periodically negotiate withthe ESCO to determine “who saved what” when sharingsavings. There are numerous hybrid contracts availablethat combine the positive aspects of the above options.25.4.6.1 Application to the Case StudyPizzaCo would enter into a hybrid contract; percentenergy savings/guaranteed arrangement. The ESCO wouldpurchase, install and operate a highly efficient chilledwater system. The ESCO would guarantee that PizzaCowould save the $1,000,000 per year, but PizzaCo wouldpay the ESCO 80% of the savings. In this way, PizzaCowould not need to invest any money, and would simplycollect the net savings of $200,000 each year. To avoidperiodic negotiations associated with shared savings

FINANCING ENERGY MANAGEMENT PROJECTS 663agreements, the contract could be worded such that theESCO will provide guaranteed energy savings worth$200,000 each year.With this arrangement, there are no depreciation,interest payments or tax-benefits for PizzaCo. However,PizzaCo receives a positive cash flow with no investmentand little risk. At the end of the contract, the ESCOremoves the equipment. At the end of most performancecontracts, the host usually acquires or purchases theequipment for fair market value. However, for this casestudy, the equipment was removed to make a consistentcomparison with the other financial arrangements.Figure 25.13 illustrates the transactions betweenthe parties. Table 25.11 presents the economic analysisfor performance contracting.Note that Table 25.11 is slightly different from theother tables in this chapter: Taxable Income = Savings– Depreciation – ESCO Payments.Chilled WaterSystem ManufacturerEquipmentESCOPaymentsPurchaseAmountLoanPizzaCoESCOBank/FinanceCo.InstallsEquipment,GuaranteesSavingsPaymentsFigure 25.13 Transactions for a Performance Contract.25.4.7 Summary Of Tax BenefitsTable 25.12 summarizes the tax benefits of eachfinancial arrangement presented in this chapter.25.4.8 Additional OptionsCombinations of the basic financial arrangementscan be created to enhance the value of a project. A sampleof the possible combinations are described below.• Third party financiers often cooperate with performancecontracting firms to implement EMPs.• Utility rebates and government programs mayprovide additional benefits for particular projects.• Tax-exempt leases are available to governmentfacilities.• Insurance can be purchased to protect against risksrelating to equipment performance, energy savings,etc.• Some financial arrangements can be structured asnon-recourse to the host. Thus, the ESCO or lessorwould assume the risks of payment default. However,as mentioned before, profit sharing increaseswith risk sharing.Attempting to identify the absolute best financialarrangement is a rewarding goal, unless it takes toolong. As every minute passes, potential dollar savingsare lost forever. When considering special grant funds,rebate programs or other unique opportunities, it isimportant to consider the lost savings due to delay.Table 25.11 Economic Analysis of a Performance Contract.——————————————————————————————————————————————ESCO Principal TaxableEOY Savings Depr. Payments Total Outstanding Income Tax ATCF——————————————————————————————————————————————01 1,000,000 0 800,000 800,000 200,000 68,000 132,0002 1,000,000 0 800,000 800,000 200,000 68,000 132,0003 1,000,000 0 800,000 800,000 200,000 68,000 132,0004 1,000,000 0 800,000 800,000 200,000 68,000 132,0005 1,000,000 0 800,000 800,000 200,000 68,000 132,000Net Present Value at 18%: $412,787——————————————————————————————————————————————Notes: ESCO purchases/operates equipment. Host pays ESCO 80% of the savings = $800,000.The contract could also be designed so that PizzaCo can buy the equipment at the end of year 5.——————————————————————————————————————————————

664 ENERGY MANAGEMENT HANDBOOKTable 25.12 Host’s Tax Benefits for each Arrangement.—————————————————————————————————————————Depreciation Interest Payments are Total Payments areARRANGEMENT Benefits Tax-Deductible Tax-Deductible—————————————————————————————————————————Retained EarningsXLoan X XBond X XSell StockXCapital Lease X XTrue LeaseXPerformance ContractX—————————————————————————————————————————25.5 “PROS” & “CONS” OF EACHFINANCIAL ARRANGEMENTThis section presents a brief summary of the “Pros”and “Cons” of each financial arrangement from thehost’s perspective.Loan“Pros”:• host keeps all savings,• depreciation & interest payments are tax-deductible,• host owns the equipment, and• the arrangement is good for long-term use ofequipment“Cons”:• host takes all the risk, and must install and manageprojectBondHas the same Pros/Cons as loan, and“Pro”:• good for government facilities, because they canoffer a tax-free rate (that is lower, but consideredfavorable by investors)Sell StockHas the same Pros/Cons as loan, and“Pro”:• selling stock could help the host achieve its targetcapital structure“Con”:• dividend payments (unlike interest payments) arenot tax-deductible, and• dilutes company controlUse Retained EarningsHas the same Pros/Cons as loan, and“Pro”:• host pays no external interest charges. Howeverretained earnings do carry an opportunity cost,because such funds could be invested somewhereat the MARR.“Con”:• host loses tax-deductible benefits of interest chargesCapital LeaseHas the same Pros/Cons as loan, and“Pro”:• Greater flexibility in financing, possible lower costof capital with third-party participationTrue Lease“Pros”:• allows use of equipment, without ownershiprisks,• reduced risk of poor performance, service, equipmentobsolescence, etc.,• good for short-term use of equipment, an• entire lease payment is tax-deductible“Cons”:• no ownership at end of lease contract, and• no depreciation tax benefitsPerformance Contract“Pros”:• allows use of equipment, with reduced installment/operationalrisks, and• reduced risk of poor performance, service, equipmentobsolescence, etc., and

FINANCING ENERGY MANAGEMENT PROJECTS 665• allows host to focus on its core business objectives“Cons”:• potentially binding contracts, legal expenses, andincreased administrative costs, and• host must share project savings25.5.1 Rules of ThumbWhen investigating financing options, consider thefollowing generalities:Loans, bonds and other host-managed arrangementsshould be used when a customer has the resources(experience, financial support, and time) tohandle the risks. Performance contracting (ESCOassumes most of the risk) is usually best when acustomer doesn’t have the resources to properlymanage the project. Remember that with any arrangementwhere the host delegates risk to anotherfirm, the host must also share the savings.Leases are the “middle ground” between owningand delegating risks. Leases are very popular dueto their tax benefits.True leases tend to be preferred when:• the equipment is needed on a short-term basis;• the equipment has unusual service problems thatcannot be handled by the host;• technological advances cause equipment to becomeobsolete quickly; or• depreciation benefits are not useful to the lessee.Capital Leases are preferred when:• the installation and removal of equipment iscostly;• the equipment is needed for a long time; or• the equipment user desires to secure a “bargainpurchase option.”25.6 CHARACTERISTICS THAT INFLUENCEWHICH FINANCIAL ARRANGEMENT IS BESTThere are at least three types of characteristicsthat can influence which financial arrangement shouldbe used for a particular EMP. These include facilitycharacteristics, project characteristics and financial arrangementcharacteristics. In this section, quantitativecharacteristics are bulleted with this symbol: $. Thequalitative characteristics are bulleted with this symbol:‹ Note that qualitative characteristics are generally“strategic” and are not associated with an exactdollar value.A few of the Facility Characteristics include:‹ The long-term plans of facility. For example, is thefacility trying to focus on core business objectivesand outsourcing other tasks, such as EMPs?$ The facility’s current financial condition. Creditratings and ability to obtain loans can determinewhether certain financial arrangements are feasible.‹ The experience and technical capabilities of inhousepersonnel. Will additional resources (personnel,consultants, technologies, etc.) be needed tosuccessfully implement the project?$ The facility’s ability to obtain rebates from thegovernment, utilities, or other organizations. Forexample, there are Dept. of Energy subsidies availablefor DOE facilities.$ The facility’s ability to obtain tax benefits. For example,government facilities can offer tax-exemptinterest rates on bonds.A few of the Project Characteristics include:$ The project’s economic benefits. Net Present Value,Internal Rate of Return and Simple Payback.‹ The project’s complexity and overall risk. For example,a complex project that has never been donebefore has a different level of risk than a standardlighting retrofit.‹ The project’s alignment with the facility’s longtermobjectives. Will this project’s equipment beneeded for long-term goals?‹ The project’s cash flow schedule and the variancebetween cash flows. For example, there may be significantdifferences in the acceptability of a projectbased on when revenues are received.A few of the Financial Arrangement Characteristicsinclude:$ The economic benefit of a project using a particularfinancial arrangement. The Net Present Value andInternal Rate of Return can be influenced by thefinancial arrangement selected.

666 ENERGY MANAGEMENT HANDBOOK‹ The impact on the corporate capital structure. Forexample, will additional debt be required to financethe project? Will additional liabilities appearon the firm’s balance sheet and impact the imageof the company to investors?‹ The flexibility of the financial arrangement. Forexample, can the facility manager alter the contractand payment terms in the event of revenue shortfallor changes in operational hours?25.7 INCORPORATING STRATEGICISSUES WHEN SELECTING FINANCIALARRANGEMENTSBecause strategic issues can be important whenselecting financial arrangements, the facility managershould include them in the selection process. The followingquestions can help assess a facility manager’sneeds.• Does the facility manager want to manage projectsor outsource?• Are net positive cash flows required?• Will the equipment be needed for long-termneeds?• Is the facility government or private?• If private, does the facility manager want theproject’s assets on or off the balance sheet?• Will operations be changing?From the research experience, a Strategic IssuesFinancing Decision Tree was developed to guide facilitymanagers to the financial arrangement which is mostlikely optimal. Figure 25.14 illustrates the decision tree,which is by no means a rule, but it embodies somegeneral observations from the industry.Working the tree from the top to bottom, thefacility manager should assess the project and facilitycharacteristics to decide whether it is strategic to managethe project or outsource. If outsourced, the “performancecontract” would be the logical choice.* If thefacility manager wants to manage the project, the nextstep (moving down the tree) is to evaluate whether theproject’s equipment will be needed for long or shorttermpurposes. If short-term, the “true lease” is logical.If it is a long-term project, in a government facility, the“bond” is likely to be the best option. If the facility is*It should be noted that a performance contract could be structuredusing leases and bonds.in the private sector, the facility manager should decidewhether the project should be on or off the balancesheet. An off-balance sheet preference would lead backto the “true lease.” If the facility manager wants theproject’s assets on the balance sheet, the Net PresentValue (or other economic benefit indicator) can helpdetermine which “host-managed” arrangement (loan,capital lease or cash) would be most lucrative.25.8 CHAPTER SUMMARYIt is clear that knowing the strategic needs of thefacility manager is critical to selecting the best arrangement.There are practically an infinite number of financialalternatives to consider. This chapter has providedsome information on the basic financial arrangements.Combining these arrangements to construct the best contractfor your facility is only limited by your creativity.25.9 GLOSSARYCapitalizeTo convert a schedule of cash flows into a principalamount, called capitalized value, by dividing by a rateof interest. In other words, to set aside an amount largeenough to generate (via interest) the desired cash flowsforever.Capital or Financial LeaseLease that under Statement 13 of the Financial AccountingStandards Board must be reflected on a company’sbalance sheet as an asset and corresponding liability.Generally, this applies to leases where the lessee acquiresessentially all of the economic benefits and risksor the leased property.DepreciationThe amortization of fixed assets, such as plant andequipment, so as to allocate the cost over their depreciablelife. Depreciation reduces taxable income, but isnot an actual cash flow.Energy Service Company (ESCO)Company that provides energy services (and possiblyfinancial services) to an energy consumer.HostThe building owner or facility that uses the equipment.LenderIndividual or firm that extends money to a borrower

FINANCING ENERGY MANAGEMENT PROJECTS 667EMP CHARACTERISTICSFACILITY CHARACTERISTICSProj. ComplexityProj. ReliabilityLong-termEquip.Cash FlowScheduleMManage orOutsourceOIn-houseStaffExperienceMgmt.’sStrategicFocusCapital Willingto CommitHost-managedArrangementsPerf.Conf.TimeFrameGovt. orPrivatelongshortTrueLeaseGovt.PrivateOff Balance SheetBondOn or OffBalanceSheetOn Balance SheetNPVQUANTITATIVE FACTORSInt. RateTaxesCash FlowTimingLoan,Cap. LeaseFigure 25.14 Strategic Issues Financing Decision Tree.with the expectation of being repaid, usually with interest.Lenders create debt in the form of loans or bonds. Ifthe borrower is liquidated, the lender is paid off beforestockholders receive distributions.LesseeThe renter. The party that buys the right to use equipmentby making lease payments to the lessor.LessorThe owner of the leased equipment.Line of CreditAn informal agreement between a bank and a borrowerindicating the maximum credit the bank will extend.A line of credit is popular because it allows numerousborrowing transactions to be approved without the reapplicationpaperwork.LiquidityAbility of a company to convert assets into cash or cashequivalents without significant loss. For example, investmentsin money market funds are much more liquidthan investments in real estate.Leveraged LeaseLease that involves a lender in addition to the lessor andlessee. The lender, usually a bank or insurance company,puts up a percentage of the cash required to purchasethe asset, usually more than half. The balance is put up

668 ENERGY MANAGEMENT HANDBOOKby the lessor, who is both the equity participant andthe borrower. With the cash the lessor acquires the asset,giving the lender (1) a mortgage on the asset and(2) an assignment of the lease and lease payments. Thelessee then makes periodic payments to the lessor, whoin turn pays the lender. As owner of the asset, the lessoris entitled to tax deductions for depreciation on the assetand interest on the loan.MARR (Minimum Attractive Rate of Return)MARR is the “hurdle rate” for projects within a company.MARR is used to determine the NPV; the annualafter-tax cash flow is discounted at MARR (which representsthe rate the company could have received witha different project).Net Present Value (NPV)As the saying goes, “a dollar received next year is not worthas much as a dollar today.” The NPV converts the worth ofthat future dollar into what is worth today. NPV convertsfuture cash flows by using a given discount rate. For example,at 10%, $1,000 dollars received one year from now is worthonly $909.09 dollars today. In other words, if you invested$909.09 dollars today at 10%, in one year it would be worth$1,000.NPV is useful because you can convert future savingscash flows back to “time zero” (present), and then compareto the cost of a project. If the NPV is positive, theinvestment is acceptable. In capital budgeting, the discountrate used is called the hurdle rate and is usuallyequal to the incremental cost of capital.“Off-Balance Sheet” FinancingTypically refers to a True Lease, because the assets arenot listed on the balance sheet. Because the liabilityis not on the balance sheet, the Host appears to befinancially stronger. However, most large leases mustbe listed in the footnotes of financial statements, whichreveals the “hidden assets.”Par Value or Face ValueEquals the value of the bond at maturity. For example,a bond with a $1,000 dollar par value will pay $1,000to the issuer at the maturity date.Preferred StockA hybrid type of stock that pays dividends at a specifiedrate (like a bond), and has preference over commonstock in the payment of dividends and liquidation ofassets. However, if the firm is financially strained, itcan avoid paying the preferred dividend as it wouldthe common stock dividends. Preferred stock doesn’tordinarily carry voting rights.Project FinancingA type of arrangement, typically meaning that a SinglePurpose Entity (SPE) is constructed. The SPE serves as aspecial bank account. All funds are sent to the SPE, fromwhich all construction costs are paid. Then all savingscash flows are also distributed from the SPE. The SPEis essentially a mini-company, with the sole purpose offunding a project.Secured LoanLoan that pledges assets as collateral. Thus, in the eventthat the borrower defaults on payments, the lender hasthe legal right to seize the collateral and sell it to payoff the loan.True Lease or Operating Lease or Tax-Oriented LeaseType of lease, normally involving equipment, wherebythe contract is written for considerably less time thanthe equipment’s life and the lessor handles all maintenanceand servicing; also called service lease. Operatingleases are the opposite of capital leases, where thelessee acquires essentially all the economic benefits andrisks of ownership. Common examples of equipmentfinanced with operating leases are office copiers, computers,automobiles and trucks. Most operating leasesare cancelable.WACC (Weighted Average Cost of Capital)The firm’s average cost of capital, as a function of theproportion of different sources of capital: Equity, Debt,Preferred Stock, etc. For example, a firm’s target capitalstructure is:Capital Source Weight (w i )Debt 30%Common Equity 60%Preferred Stock 10%and the firm’s costs of capital are:before tax cost of debt = k d = 10%cost of common equity = k s = 15%cost of preferred stock = k ps = 12%Then the weighted average cost of capital will be:WACC= w d k d (1-T) + w s k s +w ps k pswhere w i = weight of Capital Source iT = tax rate = 34%After-tax cost of debt = k d (1-T)

FINANCING ENERGY MANAGEMENT PROJECTS 669Thus,WACC= (.3)(.1)(1-.34) +(.6)(.15) + (.1)(.12)WACC= 12.18%References1. Wingender, J. and Woodroof, E., (1997) “When Firms PublicizeEnergy Management Projects Their Stock Prices Go Up: HowHigh?—As Much as 21.33% within 150 days of an Announcement,”Strategic Planning for Energy and the Environment, Vol.17(1), pp. 38-51.2. U.S. Department of Energy, (1996) “Analysis of Energy-EfficiencyInvestment Decisions by Small and Medium-SizedManufacturers,” U.S. DOE, Office of Policy and Office ofEnergy Efficiency and Renewable Energy, pp. 37-38.3. Woodroof, E. and Turner, W. (1998), “Financial Arrangementsfor Energy Management Projects,” Energy Engineering 95(3)pp. 23-71.4. Sullivan, A. and Smith, K. (1993) “Investment Justificationfor U.S. Factory Automation Projects,” Journal of the MidwestFinance Association, Vol. 22, p. 24.5. Fretty, J. (1996), “Financing Energy-Efficient Upgraded Equipment,”Proceedings of the 1996 International Energy andEnvironmental Congress, Chapter 10, Association of EnergyEngineers.6. Pennsylvania Energy Office, (1987) The Pennsylvania LifeCycle Costing Manual.7. United States Environmental Protection Agency (1994). ProjectKalc,Green Lights Program, Washington DC8. Tellus Institute, (1996), P2/Finance version 3.0 for MicrosoftExcel Version 5, Boston MA.9. Woodroof, E. And Turner, W. (1999) “Best Ways to FinanceYour Energy Management Projects,” Strategic Planning for Energyand the Environment, Summer 1999, Vol. 19(1) pp. 65-79.10. Cooke, G.W., and Bomeli, E.C., (1967), Business Financial Management,Houghton Mifflin Co., New York.11. Wingender, J. and Woodroof, E., (1997) “When Firms PublicizeEnergy Management Projects: Their Stock Prices Go Up,”Strategic Planning for Energy and the Environment, 17 (1) pp.38-51.12. Sharpe, S. and Nguyen, H. (1995) “Capital Market Imperfectionsand the Incentive to Lease,” Journal of Financial Economics,39(2), p. 271-294.13. Schallheim, J. (1994), Lease or Buy?, Harvard Business SchoolPress, Boston, p. 45.14. Hines, V. (1996),”EUN Survey: 32% of Users Have SignedESCO Contracts,” Energy User News 21(11), p.26.15. Coates, D.F. and DelPonti, J.D. (1996), “Performance Contracting:a Financial Perspective” Energy Business and TechnologySourcebook, Proceedings of the 1996 World Energy EngineeringCongress, Atlanta. p. 539-543.


CHAPTER 26COMMISSIONING FOR ENERGY MANAGEMENTDAVID E. CLARIDGEProfessorMechanical Engineering DepartmentTexas A&M UniversityMINGSHENG LIUAssociate ProfessorArchitectural EngineeringUniversity of Nebraska - LincolnW.D. TURNERProfessorMechanical Engineering DepartmentTexas A&M University26.1 INTRODUCTION TO COMMISSIONINGFOR ENERGY MANAGEMENTCommissioning an existing building has beenshown to be a key energy management activity overthe last decade, often resulting in energy savings of10%, 20% or sometimes 30% without significant capitalinvestment. Commissioning is more often applied tonew buildings today than to existing buildings, but theenergy manager will have more opportunities to applythe process to an existing building as part of the overallenergy management program. Hence, this chapter emphasizescommissioning applied to existing buildings,but also provides some commissioning guidance forthe energy manager who is involved in a constructionproject.Commissioning an existing building providesseveral benefits in addition to being an extremely effectiveenergy management strategy. It typically providesan energy payback of one to three years. In addition,building comfort is improved, systems operate betterand maintenance cost is reduced. Commissioning measurestypically require no capital investment, thoughthe process often identifies maintenance that is requiredbefore the commissioning can be completed. Potentialcapital upgrades or retrofits are often identified duringthe commissioning activities, and knowledge gainedduring the process permits more accurate quantificationof benefits than is possible with a typical audit. Involvementof facilities personnel in the process can also leadto improved staff technical skills.This chapter is intended to provide the energymanager with the information needed to make the decisionto conduct an in-house commissioning programor to select and work with an outside commissioningprovider. There is no single definition of commissioningfor an existing building, or for new buildings, so severalwidely used commissioning definitions are given. Thecommissioning process used by the authors in existingbuildings is described in some detail, and commoncommissioning measures and commissioning resourcesare described so the energy manager can choose how toimplement a commissioning program. Monitoring andverification is very important to a successful commissioningprogram. Some commissioning specific M&Vissues are discussed, particularly the role of M&V inidentifying the need for follow-up commissioning activities.Commissioning a new building is described fromthe perspective of the energy manager. Three case studiesillustrate different applications of the commissioningprocess as part of the overall energy management program.26.2 COMMISSIONING DEFINITIONSTo commission a navy ship refers to the order orprocess that makes it completely ready for active duty.Over the last two decades, the term has come to refer tothe process that makes a building or some of its systemscompletely ready for use. In the case of existing buildings,it generally refers to a restoration or improvementin the operation or function of the building systems.26.2.1 New Building CommissioningASHRAE defines building commissioning as: “theprocess of ensuring systems are designed, installed,functionally tested, and operated in conformance withthe design intent. Commissioning begins with planningand includes design, construction, start-up, acceptance,and training and can be applied throughout the life ofthe building. Furthermore, the commissioning processencompasses and coordinates the traditionally separatefunctions of systems documentation, equipment start-671

672 ENERGY MANAGEMENT HANDBOOKup, control system calibration, testing and balancing,and performance testing.” 1This guideline was restricted to new buildings, butit later became evident that while initial start-up problemswere not an issue in older buildings, most of theother problems that commissioning resolved were evenmore prevalent in older systems.26.2.2 RecommissioningRecommissioning refers to commissioning a buildingthat has already been commissioned at least once.After a building has been commissioned during theconstruction process, recommissioning ensures that thebuilding continues to operate effectively and efficiently.Buildings, even if perfectly commissioned, will normallydrift away from optimum performance over time, dueto system degradation, usage changes, or failure to correctlydiagnose the root cause of comfort complaints.Therefore, recommissioning normally reapplies theoriginal commissioning procedures in order to keep thebuilding operating according to design intent or it maymodify them for current operating needs.Optimally, recommissioning becomes part of afacility’s continuing O&M program. There is not yeta consensus on recommissioning frequency, but someconsider that it should occur every 3 to 5 years. If thereare frequent build-outs or changes in building use, recommissioningshould be applied more often. 226.2.3 RetrocommissioningRetrocommissioning is the first-time commissioningof an existing building. Many of the steps in theretrocommissioning process are similar to those for commissioning.Retrocommissioning, however, occurs afterconstruction, as an independent process, and its focus isusually on energy-using equipment such as mechanicalequipment and related controls. Retrocommissioningmay or may not bring the building back to its originaldesign intent, since the usage may have changed or theoriginal design documentation may no longer exist. 326.2.4 Continuous Commissioning ®45Continuous Commissioning (CC SM ) is an ongoingprocess to resolve operating problems, improve comfort,optimize energy use, and identify retrofits for existingcommercial and institutional buildings and centralplant facilities. CC focuses on improving overall systemcontrol and operations for the building, as it is currentlyutilized, and on meeting existing facility needs.CC is much more than an operations and maintenanceprogram. It is not intended to ensure that a building’ssystems function as originally designed, but it ensuresthat the building and its systems operate optimally tomeet the current uses of the building. As part of theCC process, a comprehensive engineering evaluation isconducted for both building functionality and systemfunctions. Optimal operational parameters and schedulesare developed based on actual building conditionsand current occupancy requirements.26.3 THE COMMISSIONING PROCESSIN EXISTING BUILDINGSThere are multiple terms that describe the commissioningprocess for existing buildings as noted in theprevious section. Likewise, there are many adaptationsof the process itself. The same practitioner will implementthe process differently in different buildings, basedon the budget and the owner requirements. The processdescribed here is the process used by the chapter authorswhen the owner wants a thorough commissioningjob. The terminology used will refer to the continuouscommissioning process, but many of the steps are thesame for retrocommissioning or recommissioning. Themodel described assumes that a commissioning provideris involved, since that is normally the case. Some(or all) of the steps may be implemented by the facilitystaff if they have the expertise and adequate staffinglevels to take on the work.CC focuses on improving overall system controland operations for the building, as it is currently utilized,and on meeting existing facility needs. It does not ensurethat the systems function as originally designed, but ensuresthat the building and systems operate optimally tomeet the current requirements. During the CC process, acomprehensive engineering evaluation is conducted forboth building functionality and system functions. Theoptimal operational parameters and schedules are developedbased on actual building conditions and currentoccupancy requirements. An integrated approach is usedto implement these optimal schedules to ensure practicallocal and global system optimization and persistence ofthe improved operation schedules.26.3.1 Commissioning TeamThe CC team consists of a project manager, oneor more CC engineers and CC technicians, and one ormore designated members of the facility operating team.The primary responsibilities of the team members areshown in Table 26.1. The project manager can be anowner representative or a CC provider representative.It is essential that the engineers have the qualificationsand experience to perform the work specified in the

COMMISSIONING FOR ENERGY MANAGEMENT 673table. The designated facility team members generallyinclude at least one lead HVAC technician and an EMCSoperator or engineer. It is essential that the designatedmembers of the facility operating team actively participatein the process and be convinced of the value of themeasures proposed and implemented, or operation willrapidly revert to old practices.26.3.2 CC ProcessThe CC process consists of two phases. The firstphase is the project development phase that identifiesthe buildings and facilities to be included in the projectand develops the project scope. At the end of thisphase, the CC scope is clearly defined and a CC contractis signed as described in Section 26.3.2.1. The secondphase implements CC and verifies project performancethrough the six steps outlined in Figure 26.1 and describedin Section 26.3.2.2.26.3.2.1 Phase 1: Project DevelopmentStep 1: Identify Buildings or FacilitiesObjective: Screen potential CC candidates with minimaleffort to identify buildings or facilities that will receivea CC audit. The CC candidate can be a building, anentire facility, or a piece of equipment. If the building ispart of a complex or campus, it is desirable to select theentire facility as the CC candidate since one mechanicalproblem may be rooted in another part of the buildingor facility.Approach: The CC candidates can be selected based onone or more of the following criteria:• The candidate provides poor thermal comfort• The candidate consumes excessive energy, and/or• The design features of the facility HVAC systemsare not fully used.If one or more of the above criteria fits the descriptionof the facility, it is likely to be a good candidate forCC. CC can be effectively implemented in buildings thathave received energy efficiency retrofits, in newer buildings,and in existing buildings that have not receivedenergy efficiency upgrades. In other words, virtuallyany building can be a potential CC candidate.The CC candidates can be selected by the buildingowner or the CC provider. However, the buildingowner is usually in the best position to select the mostpromising candidates because of his or her knowledgeof the facility operation and costs. The CC providerTable 26.1 Commissioning team members and their primary responsibilities.——————————————————————————————————————————————Team Member(s)Primary Responsibilities——————————————————————————————————————————————Project Manager 1. Coordinate the activities of building personnel and the commissioning team2. Schedule project activities——————————————————————————————————————————————CC Engineer(s) 1. Develop metering and field measurement plans2. Develop improved operational and control schedules3. Work with building staff to develop mutually acceptable implementation plans4. Make necessary programming changes to the building automation system5. Supervise technicians implementing mechanical systems changes6. Project potential performance changes and energy savings7. Conduct an engineering analysis of the system changes8. Write the project report——————————————————————————————————————————————Designated Facility Staff 1. Participate in the initial facility survey2. Provide information about problems with facility operation3. Suggest commissioning measures for evaluation4. Approve all CC measures before implementation5. Actively participate in the implementation process——————————————————————————————————————————————CC Technicians 1. Conduct field measurements2. Implement mechanical, electrical, and control system program modificationsand changes, under the direction of the project engineer——————————————————————————————————————————————

674 ENERGY MANAGEMENT HANDBOOKAn experienced engineer should review this informationand determine the potential of the CC processto improve comfort and reduce energy cost. The CCprojects often improve building comfort and reducebuilding energy consumption at the same time. However,some of the CC measures may increase buildingenergy consumption in order to satisfy room comfortand indoor air requirements. For example, providingbuilding minimum outside air will certainly increase thecooling energy consumption during summer and heatingconsumption during winter compared to operatingthe building with no outside air. If the potential justifiesa CC audit, a list of preliminary commissioning measuresfor evaluation in a CC audit should be developed.If the owner is interested in proceeding at this point, aCC audit may be performed.Step 2: Perform CC Audit and Develop Project ScopeObjectives: The objectives of this step are to:• Define owner’s requirements• Check the availability of in-house technical supportsuch as CC technicians• Identify major CC measures• Estimate potential savings from CC measures andcost to implementFigure 26.1 Outline of Phase II of the CC Process:Implementation & Verificationshould then perform a preliminary analysis to check thefeasibility of using the CC process on candidate facilitiesbefore performing a CC audit.The following information is needed for the preliminaryassessment:• Monthly utility bills (both electricity and gas) forat least 12 months (actual bills preferable to a tableof historic energy and demand data because meterreading dates are needed)• General building information: size, function, majorequipment, and occupancy schedules• O&M records, if available• Description of any problems in the building, suchas thermal comfort, indoor air quality, moisture, ormildew.Approach: The owner’s representative, the CC projectmanager and the CC project engineer will meet. The expectationsand interest of the building owner in comfortimprovements, utility cost reductions, and maintenancecost reductions will be discussed and documented. Theavailability and technical skills of in-house technicianswill be discussed. After this discussion, a walkthroughmust be conducted to identify the feasibility of theowner expectations for comfort performance and improvedenergy performance. During the walkthrough,the CC engineer and project manager will identify majorCC measures applicable to the building. An in-housetechnician should participate in this walk-through toprovide a local operational perspective and input. Theproject engineer estimates the potential savings and thecommissioning cost and together with the project managerprepares the CC audit report that documents thesefindings as well as the owner expectations.Special Considerations:• A complete set of mechanical and control systemdesign documentation is needed• The CC engineer and technician will make preliminarymeasurements of key equipment operatingparameters during the walk-through

COMMISSIONING FOR ENERGY MANAGEMENT 675• Any available measured whole building level orsub-metered energy consumption data from standalonemeters or the building automation systemshould be utilized while preparing the report.A CC audit report must be completed that listsand describes preliminary CC measures, the estimatedenergy savings from implementation, and the cost ofcarrying out the CC process on the building(s) evaluatedin the CC audit.There may be more than one iteration or variationat each step described here, but once a contract is signed,the process moves to Phase 2 as detailed below.26.3.2.2 Phase 2: CC Implementation and VerificationStep 1: Develop CC plan and form the project teamObjectives:• Develop a detailed work plan• Identify the entire project team• Clarify the duties of each team memberApproach: The CC project manager and project engineerdevelop a detailed work plan for the project, thatincludes major tasks, their sequence, time requirements,and technical requirements. The work plan is then presentedto the building owner or representative(s) at ameeting attended by any additional CC engineers and/or technicians on the project team. During the meeting,the owner contact personnel and in-house technicianswho will work on the project should be identified. Ifin-house technicians are going to conduct measurementsand system adjustments, additional time shouldbe included in the schedule unless they are going to bededicated full time to the CC project. Typically, in-housetechnicians must continue their existing duties and cannotdevote full time to the CC effort, which results inproject delays. In-house staff may also require additionaltraining. The work plan may need to be modified, dependingon the availability and skill levels of in-housestaff utilized.Special Issues:• Availability of funding to replace/repair partsfound broken• Time commitment of in-house staff• Training needs of in-house staffDeliverable: CC Report Part I—CC Plan that includesproject scope and schedule, project team, and task dutiesof each team member.Step 2: Develop performance baselinesObjectives:• Document existing comfort conditions• Document existing system conditions• Document existing energy performanceApproach: Document all known comfort problems inindividual rooms resulting from too much heating,cooling, noise, humidity, odors (especially from mold ormildew), or lack of outside air. Also, identify and documentany HVAC system problems including:• Valve and damper hunting• Disabled systems or components• Operational problems• Frequently replaced partsAn interview and walk-through may be requiredalthough most of this information is collected during theCC audit and Step 1. Room comfort problems shouldbe quantified using hand held meters or portable dataloggers. System and/or component problems should bedocumented based on interviews with occupants andtechnical staff in combination with field observationsand measurements.Baseline energy models of building performance arenecessary to document the energy savings after commissioning.The baseline energy models can be developedusing one or more of the following types of data:• Short-term measured data obtained from data loggersor the EMCS system• Long-term hourly or 15-minute whole buildingenergy data, such as whole-building electricity,cooling and heating consumption, and/or• Utility bills for electricity, gas, and/or chilled orhot waterThe whole-building energy baseline models normallyinclude whole building electricity, cooling energy,and heating energy models. These models are generallyexpressed as functions of outside air temperature sinceboth heating and cooling energy are normally weatherdependent.Any component baseline models should be representedusing the most relevant physical parameter(s) asthe independent variable(s). For example, the fan motorpower should be correlated with the fan airflow andpump motor energy consumption should be correlatedwith water flow.Short-term measured data are often the most costeffective and accurate if the potential savings of CC measuresare independent of the weather. For example, a

676 ENERGY MANAGEMENT HANDBOOKsingle true power measurement can be used to developthe baseline fan energy consumption if the pulley is tobe changed in a constant air volume system. Short-termdata are useful for determining the baseline for specificpieces of equipment, but are not reliable for baseliningoverall building energy use. They may be used withcalibrated simulation to obtain plausible baselines whenno longer term data is available.Long-term measurements are normally requiredsince potential savings of CC measures are weatherdependent. These measurements provide the most convincingevidence of the impact of CC projects. Longtermdata also help in continuing to diagnose systemfaults during ongoing CC. Although more costly thanshort-term measured data, long-term data often producesadditional savings making them the preferreddata type. For example, unusual energy consumptionpatterns can be easily identified using long-term shortintervalmeasured data. “Fixing” these unusual patternscan result in significant energy savings. Generallyspeaking, long-term interval data for electricity, gas, andthermal usage are preferred.Utility bills may be used to develop the energy usebaselines if the CC process will result in energy savingsthat are a significant fraction (>15%) of baseline use andif the building functions and use patterns will remainthe same throughout the project.The CC engineers should provide the meteringoptions(s) that meet the project requirements to thebuilding owner or representative. A metering methodshould be selected from the options presented by theCC engineer, and a detailed metering implementationplan developed. It may be necessary to hire a meteringsubcontractor if an energy information system isinstalled prior to implementation of the CC measures.More detailed information on savings determination isin contained the measurement and verification chapterof this handbook (Chapter 27).Special Considerations:• Use the maintenance log to help identify majorsystem problems• Select a metering plan that suits the CC goals andthe facility needs• Always consider and measure or obtain weatherdata as part of the metering plan• Keep meters calibrated. When the EMCS system isused for metering, both sensors and transmittersshould be calibrated using field measurementsDeliverables: CC Report Part II: Report on CurrentBuilding Performance, that includes current energyperformance, current comfort and system problems,and metering plans if new meters are to be installed.Alternatively, if utility bills are used to develop the baselineenergy models, the report should include baselineenergy models.Step 3: Conduct system measurements and developproposed CC measuresObjectives:• Identify current operational schedules and problems• Develop solutions to existing problems• Develop improved operation and control schedulesand setpoints• Identify potential cost effective energy retrofit measuresApproach: The CC engineer should develop a detailedmeasurement cut-sheet for each major system. The cutsheetshould list all the parameters to be measured, andall mechanical and electrical parts to be checked. TheCC engineer should also provide measurement trainingto the technician if a local technician is used to performsystem measurements. The CC technicians shouldfollow the procedures on the cut-sheets to obtain themeasurements using appropriate equipment.The CC engineer conducts an engineering analysis todevelop solutions for the existing problems; and developsimproved operation and control schedules andsetpoints for terminal boxes, air handling units (AHUs),exhaust systems, water and steam distribution systems,heat exchangers, chillers, boilers, and other componentsor systems as appropriate. Cost effective energy retrofitmeasures can also be identified and documented duringthis step, if desired by the building owner.Special Considerations:• Trend main operational parameters using theEMCS and compare with the measurements fromhand meters• Print out EMCS control sequences and schedules• Verify system operation in the building and compareto EMCS schedulesDeliverable: CC Report Part III: Existing System Conditions.This report includes:• Existing control sequences and setpoints for allmajor equipment, such as AHU supply air temperature,AHU supply static pressures, terminalbox minimum airflow and maximum airflow values,water loop differential pressure setpoints, and

COMMISSIONING FOR ENERGY MANAGEMENT 677equipment on/off schedules• List of disabled control sequences and schedules• List of malfunctioning equipment and control devices• Engineering solutions to the existing problems anda list of repairs required• Proposed improved control and operation sequencesand schedulesStep 4: Implement CC measuresObjectives:• Obtain approval for each CC measure from thebuilding owner’s representative prior to implementation• Implement solutions to existing operational andcomfort problems• Implement and refine improved operation andcontrol schedulesApproach: The CC project manager and project engineershould present the engineering solutions to existingproblems and the improved operational and controlschedules to the building owner’s representative in oneor more meetings. The in-house operating staff shouldbe invited to this meeting(s). All critical questionsshould be answered. It is important at this point to get“buy-in” and approval from both the building owner’srepresentative and the operating staff. The meeting(s)will decide the following issues:• Approval, modification or disapproval of each CCmeasure• Implementation sequence of CC measures• Implementation schedulesCC implementation should start by solving existingproblems. The existing comfort and difficult controlproblems are the first priority of the occupants, operatingstaff, and facility owner. Solving these problemsimproves occupant comfort and increases productivity.The economic benefits from comfort improvements aresometimes higher than the energy cost savings, thoughless easily quantified. The successful resolution of existingproblems can also earn trust in the CC engineerfrom facility operating staff, facility management andthe occupants.Implementation of the improved operation andcontrol schedules should start at the end of the comfortdelivery system, such as at the terminal boxes, and endwith the central plant. This procedure provides benefitsto the building occupants as quickly as possible. It alsoreduces the overall load on the system. If the processis reversed, the chiller plant is commissioned first. Thechiller sequences are developed based on the currentload. After the rest of the commissioning is complete, thechiller load may decrease by 30%, resulting in a need torevise the chiller operating schedules.The CC engineers should develop a detailedimplementation plan that lists each major activity. TheCC technician should follow this plan in implementingthe measures. The CC engineers should closely supervisethe implementation and refine the operational andcontrol schedules as necessary. The CC engineers shouldalso be responsible for the key software changes as necessary.Following implementation, the new operation andcontrol sequences must be documented in a way thathelps the building staff understand why they wereimplemented. Maintenance procedures for these measuresshould be provided. If any measures have notbeen implemented due to temporary impediments suchas an out of stock part, recommendations for their futureimplementation should be included.Special Considerations:• Ensure that the owner’s technical representativeunderstands each major measure• Encourage in-house technician involvement in theimplementation and/or have them implement asmany measures as possible• Document improvements in a timely manner.Deliverable: CC Report Part IV: CC Implementation.This report includes detailed documentation of implementedoperation and control sequences, maintenanceprocedures for these measures, and recommendationsfor measures to be implemented in the future.Step 5: Document comfort improvements and preliminaryenergy savingsObjectives:• Document improved comfort conditions• Document improved system conditions• Document preliminary energy savingsApproach: The comfort measurements taken in Step2 (Phase 2) should be repeated at the same locationsunder comparable conditions to determine impact onroom conditions. The measured parameters, such astemperature and humidity, should be compared withthe measurements from Step 2.The M&V procedures adopted in Step 2 should beused to determine the early post-CC energy performance.

678 ENERGY MANAGEMENT HANDBOOKEnergy performance should be compared under the sameoccupancy conditions and weather normalized.Special Considerations:• Savings analyses should follow accepted measurementand verification protocols such as theIPMVP• Comfort conditions should conform to appropriateguidelines/design documents such as ASHRAEStandard 55.Deliverable: CC Report, Part 5: Preliminary Measurementand Verification Report. This report includesresults of detailed measurements of room conditionsand energy consumption after CC activities, and anyretrofit recommendations that may be provided. Theroom conditions should be compared with those fromthe pre-CC period. The projected annual energy savingsshould be determined according to the M&V approachadopted in Step 2.Step 6: Keep the commissioning continuousObjectives:• Maintain improved comfort and energy performance• Provide measured annual energy savings.Approach: The CC engineers should review the systemoperation periodically to identify any operatingproblems and develop improved operation and controlschedules as described below.The CC engineers should provide follow-up phoneconsultation to the operating staff as needed, supplementedby site visits. This will allow the operating staffto make wise decisions and maintain the savings andcomfort in years to come. If long term measured data areavailable, the CC engineers should review the energydata quarterly to evaluate the need for a site visit. Ifthe building energy consumption has increased, the CCengineers determine possible reasons and verify withfacility operating staff. Once the problem(s) is identified,the CC engineer should visit the site, develop measuresto restore the building performance, and supervise thefacility staff in implementing the measures. If the CCengineer can remotely log onto the EMCS system, theCC engineer can check the existing system operationquarterly using the EMCS system. When a large numberof operation and control measures are disabled, a sitevisit is necessary. If the CC engineer cannot evaluatethe facility using long-term measured energy data andEMCS system information, the CC engineer should visitthe facility semi-annually.One year after CC implementation is complete,the CC engineer should write a project follow-up reportthat documents the first-year savings, recommendationsor changes resulting from any consultation or site visitsprovided, and any recommendations to further improvebuilding operations.Special Considerations: Operating personnel oftenhave a high turnover rate, and it is important to trainnew staff members in the CC process and make surethey are aware of the reasons the CC measures wereimplementedOngoing follow-up is essential if the savings are to bemaintained at high levels over time.Deliverable: Special CC Report, that documents measuredfirst-year energy savings, results from first-yearfollow-up, recommendations for ongoing staff training,and a schedule of follow-up CC activities.26.3.3 Uses of Commissioning in theEnergy Management ProcessCommissioning can be used as a part of the energymanagement program in several different ways. It canbe used;• As a stand-alone measure• As a follow-up to the retrofit process• As an Energy Conservation Measure (ECM) in aretrofit program• To ensure that a new building meets or exceeds itsenergy performance goals26.3.3.1 A stand-alone measureCommissioning is probably most often implementedin existing buildings because it is the most cost-effectivestep the owner can take to increase the energyefficiency of the building, generally offering a pay-backunder three years, and often 1-2 years 6 . The CC processalso provides a high level of understanding of the buildingand its operation, enabling retrofit recommendationsdeveloped as part of the CC process to be made witha high level of certainty. The load reductions resultingfrom implementation of the CC process may also, forexample, enable use of a smaller high efficiency chiller.26.3.3.2 A follow-up to the retrofit processCC has often been used to provide additional savingsafter a successful retrofit 7,8 and an illustrative casestudy is provided in Section 26.4.3. It has also been usednumerous times to make an under-performing retrofitmeet or exceed the original expectations. The process

COMMISSIONING FOR ENERGY MANAGEMENT 679was initially developed for these purposes as part of theTexas LoanSTAR program.26.3.3.3 As an Energy ConservationMeasure (ECM) in a retrofit programThe rapid pay-back that generally results fromCC may be used to lower the pay-back of a packageof measures to enable inclusion of a desired equipmentreplacement that has a longer pay-back in a retrofitpackage 9 . This is illustrated by a case study in Section26.3.4. In this approach the CC engineers conduct the CCaudit in parallel with the retrofit audit conducted by thedesign engineering firm. Because the two approachesare different and look at different opportunities, it isvery important to closely coordinate these two audits.For example, the CC engineer may determine a needfor a variable frequency drive on a chilled water pump.This is a retrofit opportunity for the audit engineer andshould be written up as a retrofit ECM. Similarly, theaudit engineer may encounter a CC opportunity duringthe building walk-through audit, which should bereported to the CC engineer.26.3.3.4 To ensure that a new building meetsor exceeds its energy performance goalsCommissioning is generally used for a new buildingto ensure that the systems work and provide comfortfor the occupants with minimum start-up problems. Italso has been found to reduce expensive change ordersand other construction problems. It may also be used tosignificantly improve the efficiency of a new building byoptimizing operation to meet its actual loads and usesinstead of working to design assumptions 10,11 .The commissioning process has been describedusing an outside provider. It is certainly possible to performcommissioning using internal personnel when theneeded skills are available on staff and these engineersand technicians can be assigned to the commissioningprocess. This is directly analogous to the retrofit process.Most energy audits and retrofit designs are performedby external consultants, but they can and are providedby internal personnel on occasion.26.3.4 Case Study With CC As An ECM 12Prairie View A&M University is a 1.7 millionsquare foot campus, with most buildings served by acentral thermal plant. Electricity is purchased from alocal electric co-op.University staff identified the need for a major plantequipment replacements on campus. They wished to usethe Texas LoanSTAR program to finance the project. TheLoanSTAR program finances energy efficiency upgradesfor public buildings, requiring that the aggregate energypayback of all energy conservation measures (ECMs)financed be ten years or less. The program requires thatparticipating state agencies meter all buildings/plantsreceiving the ECMs and implement a comprehensiveM&V program. The cost of the detailed investment gradeaudit and the mandatory M&V can be rolled into theloan, but the simple payback must still meet the ten-yearcriterion. This typically means that the aggregate paybackof the ECMs must be 8 to 8-1/2 years (without the auditand M&V costs included). Replacement of items such aschillers, cooling towers, and building automation systemstypically have paybacks of considerably more than 10years. Hence, they can only be included in a loan if packagedwith low payback measures that bring the aggregatepayback below 10 years.The university administration wanted to maximizethe loan amount to get as much equipment replacementas possible. They also wanted to ensure that the retrofitswork properly after they are installed. To maximize theirloan dollars, they chose to include CC as an ECM. Theyalso chose to include the audit and M&V costs in theloan to minimize up front costs.The LoanSTAR Program provides a brief walkthroughaudit of the candidate buildings and plants.This audit is performed to determine whether there issufficient retrofit potential to justify a more thoroughinvestment grade audit.The CC audit is conducted in parallel with theretrofit audit conducted by the engineering designfirm, when CC is to be included as an ECM. The twoapproaches look at different opportunities, but therecan be some overlap, so it is very important to closelycoordinate both audits. For example, the CC engineermay determine a need for a variable frequency drive ona chilled water pump. This is a retrofit opportunity forthe audit engineer and should be written up as a retrofitECM. Similarly, the audit engineer may encounter a CCopportunity during the building audit, which should bereported to the CC engineer. It is particularly importantthat the savings estimated by the audit team are not“double counted.” The area of greatest overlap in thiscase was the building automation system. Considerablecare was taken not to mix improved EMCS operationwith operational improvements determined by the CCengineer, so both measures received proper credit.The same design engineering firm conducted boththe initial walk-through audit and the detailed, investmentgrade audit. ESL CC engineers likewise conductedan initial CC walk-through audit as well as the detailedCC audit.The CC measures identified included:

680 ENERGY MANAGEMENT HANDBOOK• hot and cold deck temperature resets• extensive EMCS programming to avoid simultaneousheating and cooling• air and water balancing• duct static pressure resets• sensor calibration/repair• improved start/stop/warm-up/shutdown schedulesThe CC engineers took the measurements requiredand collected data on building operation during the CCaudit to perform a calibrated simulation on the majorbuildings. Available metered data and building EMCSdata were also used. The CC energy savings werethen written as an ECM and discussed with the designengineer. Any potential overlaps were removed. Thecombined ECMs were then listed, and the total savingsdetermined.Table 26.2 summarizes the ECMs identified fromthe two audits.The CC savings were calculated to be $204,563,as determined by conducting calibrated simulation of16 campus buildings and by engineering calculationsof savings from improved loop pumping. No CC savingswere claimed for central plant optimization. Thosesavings were all applied to ECM #7, although it seemslikely that additional CC savings will accrue from thismeasure. The simple payback from CC is slightly underthree years, making it by far the most cost effective ofthe ECMs to be implemented. The CC savings representnearly 30% of the total project savings.Perhaps more importantly, CC accounted for 2/3of the “surplus” savings dollars available to buy downthe payback of the chillers and EMCS upgrade. WithoutCC as an ECM, the University would have had tochoose which ECMs to delete, one chiller and the EMCSupgrades; or some combination of chillers and limitedbuilding EMCS upgrades to meet the ten-year paybackcriteria. With CC, however, the university was able toinclude all these hardware items, and still meet the tenyearpayback.26.4. COMMISSIONING MEASURESCC measures can be placed in two basic categories.The first category includes a number of long-timeenergy management measures that eliminate operationwhen it isn’t needed, or simply “shut it off if it isn’tneeded.” A number of these measures are a bit morecomplex than simply turning it off, but all are widelyrecognized and practiced. However, opportunities toimplement some of these measures are often found,even in well-run facilities. These are discussed in somedetail since most facility personnel can implement thesemeasures. The second category of measures can broadlybe categorized as implementing control practices thatare optimized to the facility. These measures require arelatively high level of knowledge and skill to analyzethe operation of a building, develop the optimal controlsequences, and then implement them. These measuresare presented in less detail, but references are providedfor the reader who wishes to learn more. Some measuresinclude the implementation of retrofits or newhardware in ways that are relatively new and innovativeto provide rapid pay-back comparable to the other CCmeasures. These measures are sometimes considered asa separate category, but are not discussed to that levelof detail in this chapter.26.4.1 Eliminating Unnecessary OperationCommissioning begins with simple measures thatare included in any good energy management program.Simple rules like shut off any system that isn’t neededare the beginning point of a good commissioning programas well as a good energy management program.26.4.1.1 Remove Foot Heaters and Turn Off Desk FansThe presence of foot heaters and desk fans indicatesan unsuitable working environment and wastesenergy as well. To turn off foot heaters and desk fans,the following actions should be taken:• Adjust the individual zone temperature setpointaccording to the occupant’s desires;• Balance zone airflow if foot heaters are used in aportion of the zone;• Adjust AHU supply air temperature and static pressureif the entire building is too cold or too hot• Repair existing mechanical and control problems,such as replacing diffusers of the wrong type andrelocate return air grilles, to maintain a comfortablezone temperatureDifferent people require different temperatures tofeel comfortable. Some organizations, however, mandatethe zone temperature setpoint for both summerand winter. This often leads to comfort complaints andnegatively impacts productivity. The operating staffmust place comfort as a priority and adjust the roomtemperature setpoint as necessary. Workers should beasked to dress appropriately during the summer andwinter to maintain their individual comfort if setpointsare centrally mandated for a facility. Most complaintscan be eliminated when the room temperature is withinthe range of ASHRAE’s recommended comfort zone.

COMMISSIONING FOR ENERGY MANAGEMENT 681Table 26.2: Summary of Energy Cost Measures (ECMs) 13——————————————————————————————————————————————Annual SavingsElectric ElectricECM kWh/yr Demand Gas Cost Cost to Simple# ECM kW/yr MCF/yr Savings Implement Payback——————————————————————————————————————————————#1 Lighting 1,565,342 5,221 (820) $94,669 $ 561,301 6.0——————————————————————————————————————————————#2 Replace Chiller #3 596,891 1,250 -0- $33,707 $ 668,549 19.8——————————————————————————————————————————————#3 Repair Steam System -0- -0- 13,251 $58,616 $ 422,693 7.2——————————————————————————————————————————————#4 Install Motion Sensors 81,616 -0- (44.6) $ 3,567 $ 26,087 7.3——————————————————————————————————————————————#5 Add 2 Bldgs. to CW Loop 557,676 7,050 -0- $ 60,903 $ 508,565 8.4——————————————————————————————————————————————#6 Add Chiller #4 599,891 1,250 -0- $33,707 $ 668,549 19.8——————————————————————————————————————————————#7 Primary/Secondary 1,070,207 -0- -0- $49,230 $ 441,880 9.0Pumping——————————————————————————————————————————————#8 Replace DX Systems 38,237 233 -0- $ 2,923 $ 37,929 13.0——————————————————————————————————————————————#9 Replace DDC/EMCS 2,969,962 670 2,736 $151,488 $2,071,932 13.7——————————————————————————————————————————————#10 Continuous 2,129,855 -0- 25,318 $204,563 $ 605,000 3.0Commissioning——————————————————————————————————————————————Assessment Reports $ 102,775——————————————————————————————————————————————Metering $ 157,700——————————————————————————————————————————————M&V $ 197,500——————————————————————————————————————————————9,606,677 15,674 40,440 $693,373 $6,470,460 9.3——————————————————————————————————————————————26.4.1.2 Turn off Heating Systems During SummerHeating is not needed for most buildings duringthe summer. When the heating system is on, hot wateror steam often leaks through control valves, causes thermalcomfort problems, and consumes excessive coolingand heating energy. To improve building comfort anddecrease heating and cooling energy consumption, thefollowing actions should be taken:• Turn off boilers or heat exchangers if the entirebuilding does not need heating• Manually valve off heating and preheating coilsif the heating system has to be on for other systems• Reset differential pressure of the hot water loopto a lower value to prevent excessive pressure oncontrol valves during the summer• Trouble-shoot individual zones or systems, thathave too many cold complaints• Do not turn heating off too early in the spring toavoid having to turn the system back on repeatedlyThis measure may be applied in constant air volumesystems in dry climates. When the reheat system isshut off, room comfort may be maintained by increasingsupply air temperature. This measure is not suitablefor other climates where the cooling coil leaving airtemperature has to be controlled below 57°F to controlroom humidity levels.This simple measure results in significant energysavings as well as improved comfort in most buildings.Figure 26.2 compares the measured heating energy consumptionbefore and after manually shutting off AHUheating valves in a building in Austin, Texas.This building has a floor area of 147,000 squarefeet with two dual duct VAV systems. Before closingthe heating coil manual valves, the average daily steamconsumption varied from a low of 0.2 MMBtu/hr to ahigh of about 0.28 MMBtu/hr. After the manual valveswere closed, steam leakage was eliminated throughthe heating coil. The steam consumption immediatelydropped to slightly above 0.10 MMBtu/hr. Since themanual valves in this building can stay closed for

682 ENERGY MANAGEMENT HANDBOOKFigure 26.2 Comparison of measureddaily average steam consumptionbefore and after manually shuttingoff heating coil valves in the BusinessAdministration Building at theUniversity of Texas at Austin 14 .more than seven months, the annual steam savingsare 756 MMBtu/yr. The same amount of chilled waterwill also be saved if the building remains at the sametemperature, so the cooling energy savings will be 756MMBtu/yr as well. The annual energy cost savings is$7,560 at an energy price of $5/MMBtu. This savings isnot particularly large, but the only action required wasshutting two manual valves.26.4.1.3 Turn Off Systems During Unoccupied HoursIf a building is not occupied at nights or on weekends,the HVAC system may often be turned off completelyduring these periods. With a properly designedwarm-up/cool-down, building comfort can be maintainedwith significant energy savings. In a commercialor institutional building, office equipment and lightingmake up a large portion (often 50% or more) of the electricalrequirements. However, a significant portion of abuilding (15% or more) is normally unoccupied duringoffice hours due to travel, meetings, vacations, and sickleave. Turning off systems during unoccupied hoursresults in significant energy savings without degradingoccupant comfort. This measure can be achieved by thefollowing actions:• Turn off lights, computers, printers, fax machines,desk fans, and other office equipment when leavingthe office• Turn off lights and set back room thermostats aftercleaning• Turn off AHUs at nights and on weekends. Aschedule needs to be developed for each zone orair handling unit. Turning off the system too earlyin the evening or turning the system on too late inthe morning may cause comfort problems. Converselyturning off a system too late in the evening and turningthe system on too early in the morning may loseconsiderable savings• Turn off the boiler hot water pump at night duringthe summer when AHUs are turned off• Turn off chillers and chilled water pumps whenfree cooling is available or when AHUs are turnedoffFigure 26.3 presents the measured building electricityconsumption, excluding chiller consumption, beforeand after implementation of AHU and office equipmentturn-off on nights and weekends in the Stephen F. AustinBuilding (SFA) in Austin, Texas.The Stephen F. Austin Building has 470,000 squarefeet of floor area with 22 dual duct AHUs. During thefirst phase of implementation, 16 AHUs were turned offfrom midnight to 4 a.m. weekdays and weekends. Duringthe second phase, 22 AHUs were turned off from11:00 p.m. to 5 a.m. during weekdays and weekends. Inaddition, during the second phase, all occupants wereasked to turn off office equipment when they leave theiroffice.The measured results show that the nighttimewhole building electricity use decreased from 1,250 kWto 900 kW during the first phase. During the secondphase, the nighttime minimum electricity decreased to800 kW.It was observed that the daily peak electricity consumptionafter night shutdowns began is significantlylower than the base peak. For example, the lowest peakduring the second phase is 1,833 kW, which is 8% lower

COMMISSIONING FOR ENERGY MANAGEMENT 683Figure 26.3 Hourly whole buildingelectricity consumption atthe SFA Building before andafter night shut down of AHUswas initiated 15 .than the base peak. The lower electricity peak indicatesthat some office equipment remained off during the daytimeor the employees were more conscientious in turningoff lights and equipment when they left the office.The annual energy cost savings, including electricity,heating, and cooling, were determined to be $100,000/yrusing measured hourly data.26.4.1.4 Slow Down Systems DuringUnoccupied/Lightly-Occupied HoursMost large buildings are never completely unoccupied.It is not uncommon to find a few people workingregardless of the time of day. The zones that may beused during the weekends or at night, are also unpredictable.System shut down often results in complaints.Substantial savings can be achieved while maintainingcomfort conditions in a building by an appropriate combinationof the following actions:• Reset outside air intake to a lower level (0.05 cfm/ft 2 ) during these hours during hot summer andcold winter weather. Outside air can be reducedsince there will be very few people in the building.Check outside and exhaust air balance to maintainpositive building pressure• Reset the minimum airflow to a lower value, possiblyzero, for VAV terminal boxes• Program constant volume terminal boxes as VAVboxes, and reset the minimum flow from themaximum to a lower value, possibly zero duringunoccupied hours• Reset AHU static pressure and water loop differentialpressure to lower values• Set supply air fan at a lower speedThese measures maintain building comfort whileminimizing energy consumption. The savings are oftencomparable with the shut down option. Figure 26.4presents the measured hourly fan energy consumptionin the Education Building at the University of Texas atAustin.The Education Building has 251,000 ft 2 of floor areawith eight 50-hp AHUs that are operated on VFDs. Priorto the introduction of this measure, the motor controlcenter (MCC) energy consumption was almost constant.The CC measure implemented was to set the fan speedat 30% at night and on weekends. The nighttime slowroll decreased the fan power from approximately 50kW to approximately 25 kW while maintaining buildingcomfort.26.4.1.5 Limit Fan Speed During Warm Up and CoolDown PeriodsIf nighttime shut down is implemented, warm-upis necessary during the winter and cool down is requiredduring the summer. During warm-up and cooldown periods, fan systems are often run at maximumspeed since all terminal boxes require either maximumheating or maximum cooling. A simple fan speed limitcan reduce fan power significantly. This principle mayalso be used in other systems, such as pumps. The followingactions should be taken to achieve the fan energysavings:• Determine the optimal start up time using 80%(adjustable) fan capacity if automatic optimal startup is used• Set the fan speed limit at 80% (adjustable) manuallyand extend the warm up or cool down period

684 ENERGY MANAGEMENT HANDBOOKFigure 26.4 Measured Post-CC hourlysupply fan electricity consumptionin the Education Building 16 .by 25%. If the speed limit is set at another fractionalvalue (x), determine the warm up periodusing the following equation:T existT n = ———x• Keep outside air damper closed during warm-upand cool-down periodsThe fan energy savings increase significantly as thefan speed limit decreases. Figure 26.5 shows the theoreticalfan power savings. When the fan speed limit is 50%of design fan speed, the potential fan energy savings are75% of the fan energy even if the fan runs twice as long.The theoretical model did not consider the variable speeddrive loss. The actual energy savings will normally besomewhat lower than the model projected value.Note that if the outside air damper cannot beclosed tightly, extra thermal energy may be requiredto cool or warm up outside air that leaks through thedamper. This factor should be considered when thismeasure is used.26.4.2 Operational Efficiency Measures forAHU SystemsAir handler systems normally condition and distributeair inside buildings. A typical AHU system consistsof some combination of heating and cooling coils,supply and return air fans, filters, humidifiers, dampers,ductwork, terminal boxes, and associated safety andcontrol devices, and may include an economizer. Asthe building load changes, AHUs change one or moreof the following parameters to maintain building comfort:outside air intake, total airflow, static pressure, andsupply air temperature and humidity. Both operatingschedules and initial system set up, such as total airflowand outside airflow, significantly impact building energyconsumption and comfort. The following ten major CCmeasures should be used to optimize AHU operationand control schedules:• Adjust total airflow for constant air volume systems• Set minimum outside air intake correctly• Improve static pressure set-point and schedule• Optimize supply air temperatures• Improve economizer operation and control• Improve coupled control AHU operation• Valve off hot air flow for dual duct AHUs duringsummer• Install VFD on constant air volume systems• Install airflow control for VAV systems• Improve terminal box operation26.4.2.1. Adjust Total Air Flow and Fan Head for ConstantAir Volume SystemsAir flow rates are significantly higher than requiredin most buildings primarily due to system oversizing.In some large systems, an oversized fan causesover-pressurization in terminal boxes. This excessivepressurization is the primary cause of room noise. Theexcessive airflow often causes excessive fan energyconsumption, excessive heating and cooling energyconsumption, humidity control problems, and excessivenoise in terminal boxes 18 .

COMMISSIONING FOR ENERGY MANAGEMENT 685Figure 26.5 Potential fan energysavings using fan speed limiting16 .26.4.2.2. Set Minimum Outside Air Intake CorrectlyOutside air intake rates are often significantlyhigher than design values in existing buildings due tolack of accurate measurements, incorrect design calculationsand balancing, and operation and maintenanceproblems. Excessive outside air intake is caused by themixed air chamber pressure being lower than the designvalue, by significant outside air leakage throughthe maximum outside air damper on systems with aneconomizer, by the minimum outside air intake beingset to use minimum total airflow for a VAV system, orby lower than expected/designed occupancy.26.4.2.3. Improve Static Pressure Setpoint and ScheduleThe supply air static pressure is often used to controlfan speed and ensure adequate airflow to each zone.If the static pressure setpoint is lower than required,some zones may experience comfort problems due tolack of airflow. If the static pressure setpoint is too high,fan power will be excessive. In most existing terminalboxes, proportional controllers are used to maintain theairflow setpoint. When the static pressure is too high, theactual airflow is higher than its setpoint. The additionalairflow depends on the setting of the control band. Fieldmeasurements 19 have found that the excessive airflowcan be as high as 20%. Excessive airflow can also occurwhen terminal box controllers are malfunctioning. Forpressure dependent terminal boxes, high static pressurecauses significant excessive airflow. Consequently, highstatic pressure often causes unnecessary heating andcooling energy consumption. A higher than necessarystatic pressure setpoint is also the primary reason fornoise problems in buildings.26.4.2.4. Optimize Supply Air TemperaturesSupply air temperatures (cooling coil dischargeair temperature for single duct systems; cold deck andhot deck temperatures for dual duct systems) are themost important operation and control parameters forAHUs. If the cold air supply temperature is too low,the AHU may remove excessive moisture during thesummer using mechanical cooling. The terminal boxesmust then warm the over-cooled air before sendingit to each individual diffuser for a single duct AHU.More hot air is required in dual duct air handlers. Thelower air temperature consumes more thermal energyin either system. If the cold air supply temperature istoo high, the building may lose comfort control. The fanmust supply more air to the building during the coolingseason, so fan power will be higher than necessary.The goal of optimal supply air temperature schedulesis to minimize combined fan power and thermal energyconsumption or cost. Although developing optimal resetschedules requires a comprehensive engineering analysis,improved (near optimal) schedules can be developedbased on several simple rules. Guidelines for developingimproved supply air temperature reset schedules areavailable for four major types of AHU systems 20 .26.4.2.5. Improve Economizer Operation and ControlAn economizer is designed to eliminate mechanicalcooling when the outside air temperature is lowerthan the supply air temperature setpoint and decreasemechanical cooling when the outside air temperature isbetween the cold deck temperature and a high temperaturelimit, which is typically less than 70°F. Economizercontrol is often implemented so it controls mixed air

686 ENERGY MANAGEMENT HANDBOOKtemperature at the cold deck temperature or simply55°F. This control algorithm is far from optimum. It may,in fact, actually increase the building energy consumption.The economizer operation can be improved usingthe following steps:1. Integrate economizer control with optimal colddeck temperature reset. It is tempting to ignorecold deck reset when the economizer is operating,since the cooling is free. However, cold deck resetnormally saves significant heating.2. For a draw-through AHU, set the mixed air temperature1°F lower than the cold deck temperaturesetpoint. For a blow-through unit, set the mixed airtemperature at least 2°F lower than the supply airtemperature setpoint. This will eliminate chilledwater valve hunting and unnecessary mechanicalcooling.3. For a dual duct AHU, the economizer should bedisabled if the hot air flow is higher than the coldair flow since the heating energy penalty is thentypically higher than cooling energy savings.4. Set the economizer operating range as wide aspossible. For dry climates, the economizer shouldbe activated when the outside air temperature isbetween 30°F and 75°F, between 30°F and 65°F fornormal climates, and between 30°F and 60°F forhumid climates. When proper return and outsideair mixing can be achieved, the economizer can beactivated even when the outside air temperature isbelow 30°F.5. Measure the true mixed air temperature. Mostmixing chambers do not achieve complete mixingof the return air and outside air before reachingthe cooling coil. It is particularly important thatmixed air temperature be measured accuratelywhen an economizer is being used. An averagingtemperature sensor should be used for the mixedair temperature measurement.26.4.2.6. Improve Coupled Control AHU OperationCoupled control is often used in single-zone singleduct,constant volume systems. Conceptually, this systemprovides cooling or heating as needed to maintainthe setpoint temperature in the zone and uses simultaneousheating and cooling only when the humidistatindicates that additional cooling (followed by reheat)is needed to provide humidity control. However, thehumidistat is often disabled for a number of reasons. Tocontrol room relative humidity level, the control signalsor spring ranges are overlapped. Simultaneous heatingand cooling often occurs almost continuously.26.4.2.7. Valve Off Hot Air Flow for Dual Duct AHUsDuring SummerDuring the summer, most commercial buildings donot need heating. Theoretically, hot air should be zerofor dual duct VAV systems. However, hot air leakagethrough terminal boxes is often significant due to excessivestatic pressure on the hot air damper. For constantair volume systems, hot air flow is often up to 30% ofthe total airflow. During summer months, hot air temperaturesas high as 140°F have been observed due tohot water leakage through valves. The excessively highhot air temperature often causes hot complaints in somelocations. Eliminating this hot air flow can improvebuilding thermal comfort, reduce fan power, coolingconsumption, and heating consumption.26.4.2.8. Install VFD on Constant Air Volume SystemsThe building heating load and cooling load variessignificantly with both weather and internal occupancyconditions. In constant air volume systems, a significantamount of energy is consumed unnecessarily dueto humidity control requirements. Most of this energywaste can be avoided by simply installing a VFD on thefan without a major retrofit effort. Guidelines for VFDinstallation are available for dual duct, multi-zone, andsingle duct systems 21 .26.4.2.9. Airflow Control for VAV SystemsAirflow control of VAV systems has been an importantdesign and research subject since the VAV systemwas introduced. An airflow control method should:(1) ensure sufficient airflow to each space or zone; (2)control outside air intake properly; and (3) maintain apositive building pressure. These goals can be achievedusing the variable speed drive volume tracking (VS-DVT) method 22,23 .26.4.2.10. Improve Terminal Box OperationThe terminal box is the end device of the AHU system.It directly controls room temperature and airflow.Improving the set up and operation are critical for roomcomfort and energy efficiency. The following measuresare suggested:• Set minimum air damper position properly forpressure dependent terminal boxes.• Use VAV control algorithm for constant air volumeterminal boxes.

COMMISSIONING FOR ENERGY MANAGEMENT 687• Use airflow reset.• Integrate lighting and terminal box control.• Integrate airflow and temperature reset.• Improve terminal box control performance.26.4.3 Case Study—AHU CCin a Previously Retrofit Building 2426.4.3.1 Case Study Building DescriptionThe building studied in this paper is the 295,000gross ft 2 (226,000 ft 2 net) Zachry Engineering Center(ZEC), located on the Texas A&M University campus(30°N, 96°W) where the average January temperatureis 50°F and average July temperature is 84°F, and picturedin Figure 26.6. The building has four-floors plusan unconditioned basement parking level. It was constructedin the early 1970s and is a heavy structure with6 in. concrete floors and insulated exterior walls madeof precast concrete and porcelain-plated steel panels.About 12% of the exterior wall area is covered withsingle-pane, bronze-tinted glazing. The windows arerecessed approximately 24 in. from the exterior walls,which provides some shading. Approximately 2835 ft 2of northeast-facing clerestory windows admit daylightinto the core of the building.The ZEC includes offices, classrooms, laboratoriesand computer rooms and is open 24 hours per day, 365days per year with heaviest occupancy during normalworking hours between 8 a.m. and 6 p.m. on weekdays.Occupancy, electrical consumption and chilled waterconsumption show marked weekday/weekend differenceswith peak weekend electrical consumption lessthan 10% above the nightly minimum; weekday holidayoccupancy is similar to weekend usage with intermediateusage on weekdays between semesters when classFigure 26.6 The Zachry Engineering Center on theTexas A&M campus 25 .rooms are not in use, but laboratories and offices areoccupied.HVAC Systems. Twelve identical dual-duct systemswith 40 hp fans rated at 35,000 cfm and eight smaller airhandlers (3 hp average) supply air to the zones in thebuilding. Supply and return air ducts are located aroundthe perimeter of the building. These were operatedwith a constant outdoor air intake at a nominal valueof 10% of design flow. The large dual-duct constant airvolume (DDCAV) systems were converted to dual-ductvariable-air volume (DDVAV) systems accompaniedby connection to the campus energy management andcontrol system (EMCS) in 1991. This retrofit successfullyreduced fan power consumption by 44%, cooling consumptionby 23%, and heating consumption by 84%.Monitoring of energy use. In the engineering centerabout 50 channels of hourly data have been recordedand collected each week since May 1989. The sensors arescanned every 4 seconds and the values are integratedto give hourly totals or averages as appropriate. The importantchannels for savings measurement are those forair handler electricity consumption and whole-buildingheating and cooling energy use. Air handler electricityconsumption is measured at the building’s motor controlcenter (MCC) and represents all of the air-handling unitsand most of the heating, ventilating, and air-conditioning(HVAC)-related pumps in the building. Cooling andheating energy use are determined by a Btu meter whichintegrates the monitored fluid flow rate and temperaturedifference across the supply and return lines of thechilled- and hot-water supply to the building. The majorityof the 50 channels of monitored information comefrom one air handler that was highly instrumented.4.3.2 Continuous Commissioning of the Zachry EngineeringCenterThe Continuous Commissioning process was appliedto the Zachry Engineering Center in 1996-97. Inthis case, the initial survey and specification of monitoringportions of the CC process were not performed sincethe university president had decided to implement CCon campus based on its success in numerous other locationsrather than on the results of individual buildingsurveys. Metering had been installed much earlier in theZachry Engineering Center as part of the retrofit process,so performance baselines were already available.Conduct system measurements and develop proposedCC measuresThe facility survey found that the building controlsystem set-up was far from optimum and found numerousother problems in the building as well. The basic

688 ENERGY MANAGEMENT HANDBOOKFigure 26.7 ZEC daily motor control center (MCC) electric consumption in 1990 beforethe retrofit, in 1994 after the retrofit, and in 1997 after CC 26 .control strategies found in the building are summarizedin Table 26.3 in the column labeled “Pre-CC ControlPractice.” The ranges shown for constant parametersreflect different constant values for different individualair handlers.The control practices shown in the table are allwidely used, but none are optimal for this building. Thecampus controls engineer worked closely with the CCengineers during the survey. The items shown in Table26.3 could all be determined by examination of the controlsystem in the building, but the facility survey alsoexamines a great deal of the equipment throughout thebuilding and found numerous cases of valves that let toomuch hot or cold water flow, control settings that causedcontinuous motion and unnecessary wear on valves, airducts that had blown off of the terminal boxes, kinks inair ducts that led to rooms that could not be properlyheated or cooled, etc. Following the survey, the buildingperformance was analyzed and CC measures includingoptimum control schedules were developed for thebuilding in cooperation with the campus controls engineer.Implement CC measuresFollowing the survey, the building performancewas analyzed and optimum control schedules were developedfor the building in cooperation with the campuscontrol engineer. The air handlers, pumps and terminalboxes had major control parameters changed to valuesshown in the “Post-CC” column of Table 26.3. Most ofthe control parameters were optimized to vary as a functionof outside air temperature, T oa , as indicated.In addition to optimizing the control settings for theheating and cooling systems in the building, numerousproblems specific to individual rooms, ducts, or terminalboxes were diagnosed and resolved. These includeditems like damper motors that were disconnected, bentair ducts that could not supply enough air to properlycontrol room temperature, leaking air dampers, dampersthat indicated open when only partly open, etc.Problems of this sort often had led to occupantcomplaints that were partially resolved without fixingthe real problem. For example, if a duct was constrictedso inadequate flow reached a room, the pressure in theair handler might be increased to get additional flowinto the room. “Fixes” like this typically improve roomcomfort, but sometimes lead to additional heating andcooling consumption in every other room on the sameair handler.Document energy savingsImplementation of these measures resulted in significantadditional savings beyond the original savingsfrom the VAV retrofit and controls upgrade as shownin Figures 26.7, 26.8 and 26.9. Figure 26.7 shows themotor control center power consumption as a func-

COMMISSIONING FOR ENERGY MANAGEMENT 689Table 26.3. Major control settings in theZachry Engineering Center before and after implementation of CC 27 .——————————————————————————————————————————————Parameter Pre-CC Control Practice Post-CC Control Practice——————————————————————————————————————————————Pressure in air ducts Constant at 2.5 - 3.5 in H 2 O 1- 2 inH 2 O as T oa increases——————————————————————————————————————————————Cold air temperature Constant at 50°F - 55°F 60°F - 55°F as T oa increases——————————————————————————————————————————————Hot air temperature Constant at 110°F - 120°F 90°F - 70°F as T oa increases——————————————————————————————————————————————Air flow to rooms Variable - but inefficient Optimized min/max flow anddamper operation——————————————————————————————————————————————Heating pump control Operated continuously On when T oa >55°F——————————————————————————————————————————————Cooling pump control Variable speed with shut-off Pressure depends on flow——————————————————————————————————————————————tion of ambient temperature for 1990, 1994 and 1997.It is evident that the minimum fan power has beencut in half and there has been some reduction even atsummer design conditions. Figure 26.9 shows the hotwater consumption for 1990, 1994 and 1997, again asa function of daily average temperature. The retrofitreduced the annual hot water (HW) consumption forheating to only 16% of the baseline, so there is littleroom for further reduction. However, it can be seenthat the CC measures further reduced HW consumption,particularly at low temperatures. The largestsavings from the CC measures are seen in the chilledwater consumption as shown in Figure 26.8. The largestfractional savings occur at low ambient temperatures,but the largest absolute savings occur at the highestambient temperatures.The annualized consumption values for the baseline,post-retrofit and post-CC conditions are shownin Table 26.4. The MCC consumption for 1997 was1,209,918 kWh, 74% of the 1994 consumption and only41% of the 1990 consumption. On an annual basis, thepost-CC HW consumption normalized to 1994 weatherwas 1940 MMBtu, a reduction to only 10% of baselineconsumption and a reduction of 34% from the 1994consumption. The CC measures reduced the post-CCchilled water (CHW) consumption to 17,400 MMBtu, areduction of 17,800 MMBtu which is noticeably largerthan the 13,900 MMBtu savings produced by the retrofit.The CHW savings accounted for the largest portion ofthe CC savings in this cooling-dominated climate.Generalized application of case study and conclusionsThe major energy savings from the CC activities inthe case study building resulted from five items.1. Optimization of duct static pressures at lowerlevels. This reduces fan power and also reducesdamper leakage that increases both heating andcooling consumption.2. Optimization of cold air temperatures. Most buildingsin our experience use a constant set-point forTable 26.4. Consumption at the Zachry Engineering Center before and after retrofit and after implementationof CC measures 28 .——————————————————————————————————————————————Baseline Consumption Post-retrofit Post-Retrofit Post-CC Post-CC——————————————————————————————————————————————Fan Power 2,950,000 kWh 1,640,000 kWh 56% 1,210,000 kWh 41%——————————————————————————————————————————————Chilled Water 45,800 MMBtu 35,300 MMBtu 77% 17,400 MMBtu 37%——————————————————————————————————————————————Heating Water 18,800 MMBtu 2940 MMBtu 16% 1940 MMBtu 10%——————————————————————————————————————————————

690 ENERGY MANAGEMENT HANDBOOKthe cold air temperature which is very inefficient.Even if this set-point is changed, it is seldom optimized.3. Optimization of hot air temperatures. We findthat most buildings modulate hot air temperatureaccording to outside air temperature, but it is normallysignificantly higher than necessary.4. Optimize settings on VAV boxes. Most VAV terminalboxes have minimum flow set-points at nightthat are too high.5. Optimize pump control. Static pressure set-pointson chilled water and hot water pumps are generallyset higher than necessary.The CC process has been illustrated by applicationto a large building which had earlier had a major retrofitperformed. The post-CC consumption values represent41% of the pre-retrofit fan and pump consumption, 10%of the pre-retrofit hot water consumption, and 38% ofthe pre-retrofit chilled water consumption. Using thebaseline energy prices, the post-CC consumption reflectsan HVAC energy cost that is only 36% of the baselineHVAC cost and is only 65% of the HVAC cost after theretrofit. The measures implemented in the case studybuilding are quite typical of CC measures implementedin other buildings. These results are better than averagefor the process, but are not one-of-a-kind.26.4.4 CC Measures for Water/SteamDistribution SystemsDistribution systems include central chilled water,hot water, and steam systems, that deliver thermal energyfrom central plants to buildings. In turn, the systemdistributes the chilled water, hot water and steamto AHU coils and terminal boxes. Distribution systemsmainly consist of pumps, pipes, control valves, and variablespeed pumping devices. This section focuses on theCC measures for optimal pressure control, water flowcontrol, and general optimization.26.4.4.1 Improve Building ChilledWater Pump OperationMost building chilled water pumping systemsare equipped with variable speed devices (VSDs). If aVSD is not installed, retrofit of a VSD is generally recommended.The discussion here is limited to systemswhere a VSD is installed. The goal of pumping optimizationis to avoid excessive differential pressures acrossthe control valves, while providing enough water toFigure 26.8. ZEC daily chilled water consumption for 1990 before the retrofit, 1994after the retrofit, and 1997 after CC 29 .

COMMISSIONING FOR ENERGY MANAGEMENT 691Figure 26.9. ZEC daily hot water consumption for 1990 before the retrofit, 1994 afterthe retrofit, and 1997 after CC 30 .each building, coil, or other end use. An optimal pumpdifferential pressure schedule should be developed thatprovides adequate pressure across the hydraulicallymost remote coil in the system under all operating conditions,but does not provide excess head.26.4.4.2 Improve Secondary Loop OperationFor buildings supplied by a secondary loop froma central plant, building loop optimization should beperformed before the secondary loop optimization.Source Distributed Systems: If there are no buildingpumps, the secondary pumps must provide the pressurehead required to overcome both the secondary loop andthe building loop pressure losses. In this case, the secondaryloop is called a source distributed system. Thesecondary loop pumps should be controlled to provideenough pressure head for the most remote coil. If VFDsare installed, the differential pressure can be controlledby modulating pump speed. Otherwise, the differentialpressure can be modulated by changing the number ofpumps in operation.Source Distributed Systems With Building Pumps: Inmost campus settings, each building has a pump. Theoptimal differential pressure setpoint should then bedetermined by optimizing the secondary loop pressuresetpoint so the combined secondary pump and buildingpumping power is minimized. This can be doneby developing a pressure reset schedule that requiresmaximum building pump power at the most hydraulicallyremote building on the secondary loop. This mayoccur with a negative differential pressure across themost remote building.26.4.4.3 Improving Central Plant Water Loop OperationThe central plant loop optimization should be performedafter secondary loop optimization.Single Loop Systems: For most heating distribution systemsand some chilled water systems, a single loop isused instead of primary and secondary systems. Underpartial load conditions, fewer pumps can be used forboth chillers and heat exchangers. This can result in lesspump power consumption.Primary-secondary Loop Systems: Primary-secondarysystems are the most common chilled water distributionsystems used with central chiller plants. This design isbased on the assumption that the chilled water flowthrough the chiller must be maintained at the designlevel. This is seldom needed. Due to this incorrect assumption,a significant amount of pumping power iswasted in numerous central plants. Design engineers

692 ENERGY MANAGEMENT HANDBOOKmay or may not include an isolation valve on the bypassline of the primary loop. Procedures are availableto optimize pump operation for both cases 31,32,33,34 .26.4.5 CC Measures for Central Chiller PlantsThe central chiller plant includes chillers, coolingtowers, a chilled water distribution system, andthe condenser water distribution system. Although asecondary pumping system may be physically locatedinside the central plant, commissioning issues dealingwith secondary loops are discussed in the previoussection. The central chiller plant produces chilled waterusing electricity, steam, hot water, or gas. The detailedcommissioning measures vary with the type of chillerand this section gives general commissioning measuresthat apply to a typical central cooling plant and that canproduce significant energy savings.Use the Most Efficient Chillers: Most central chillerplants have several chillers with different performancefactors or efficiencies. The differences in performancemay be due to the design, to performance degradation,age, or operational problems. One chiller may have ahigher efficiency at a high load ratio while another mayhave a higher efficiency at a lower load ratio. Runningchillers with the highest performance can result in significantenergy savings because you will be providingthe greatest output for the least input.Reset the Supply Water Temperature: Increasing thechilled water supply temperature can decrease chillerelectricity consumption significantly. The general ruleof-thumbis that a one degree Fahrenheit increasecorresponds to a decrease in compressor electricity consumptionof 1.7%. The chilled water supply temperaturecan be reset based on either cooling load or ambientconditions.Reset Condenser Return Water Temperature: Decreasingcooling tower return water temperature has the same effectas increasing the chilled water supply temperature.The cooling tower return temperature should be resetbased on weather conditions. The following providesgeneral guidelines:• The cooling tower return water temperaturesetpoint should be at least 5°F (adjustable basedon tower design “approach”) higher than the ambientwet bulb temperature. This prevents excessivecooling tower fan power consumption.• The cooling tower water return temperatureshould not be lower than 65°F for chillers madebefore 1999, and should not be lower than 55°Ffor newer chillers. It is also recommended that youconsult the chiller manufacturer for specific limitson allowable condenser water temperature.The cooling tower return water temperature resetcan be implemented using the BAS. If it cannot be implementedusing the BAS, operators can reset the setpointdaily using the daily maximum wet bulb or dry bulbtemperature.Decreasing the cooling tower return temperaturemay increase fan power consumption. However, fan powermay not necessarily increase with lower cooling towerreturn water temperature. The following tips can help.• Use all towers. For example, use all three towerswhen one of the three chillers is used. This mayeliminate fan power consumption entirely. Thepump power may actually stay the same. Be sureto keep the other two tower pumps off.• Never turn on the cooling tower fan before the bypassvalve is totally closed. If the by-pass valve isnot totally closed, the additional cooling providedby the fan is not needed and will not be used. Savethe fan power!• Balance the water distribution to the towers andwithin the towers. Towers are often seen wherewater is flowing down only one side of the tower,or one tower may have twice the flow of another.This significantly increases the water return temperaturefrom the towers.Increase Chilled Water Return Temperature: Increasingchilled water return temperature has the same effect asincreasing chilled water supply temperature. It can alsosignificantly decrease the secondary pump power sincethe higher the return water temperature (for a given supplytemperature), the lower the differential temperature,and the lower the chilled water flow. Maximizing chilledwater return temperature is much more important thanoptimizing supply water temperature since it often providesmuch more savings potential. It is hard to increasesupply temperature 5°F above the design setpoint. It isoften easy to increase the return water temperature asmuch as 7°F by conducting water balance and shuttingoff by-pass of three-way valves.Use Variable Flow under Partial Load Conditions:Typical central plants use primary-secondary loops. Aconstant speed primary pump is often dedicated to aparticular chiller. When the chiller is turned on, thepump is on. Chilled water flow through each chilleris maintained at the design flow rate by this operatingschedule. When the building-loop flow is less than the

COMMISSIONING FOR ENERGY MANAGEMENT 693chiller loop flow, part of the chiller flow by-passes thebuilding and returns to the chiller. This practice causesexcessive primary pump power consumption and excessivelylow entering water tem perature to the chiller,which i ncreases th e compressor power consumption.It is generally perceived that chilled water flowsmust remain constant for chiller operational safety.Actually, most new chillers allow chilled water flow aslow as 30% of the design value. The chilled water flowcan be decreased to be as low as 50% for most existingchillers if the proper procedures are followed 35 .Varying chilled water flow through a chiller canresult in significant pump power savings. Although theprimary pumps are kept on all the time, the secondarypump power consumption is decreased significantlywhen compared to the conventional primary and secondarysystem operation. Varying chilled water flowthrough the chillers will also increase the chiller efficiencywhen compared to constant water flow withchilled water by-pass. More information can be foundin a paper by Liu 36 .Optimize Chiller Staging: For most chillers, the kW/tondecreases (COP increases) as the load ratio increasesfrom 40% to 80%. When the load ratio is too low, the capacitymodulation device in the chiller lowers the chillerefficiency. When the chiller has a moderate load, thecapacity modulation device has reasonable efficiency.The condenser and evaporator are oversized for the loadunder this condition so the chiller efficiency is higher.When the chiller is at maximum load, the evaporatorand condenser are marginally sized, reducing the chillerefficiency below its maximum value. Running chillersin the high efficiency range can result in significantelectrical energy savings and can improve the reliabilityof plant operation. Optimal chiller staging should bedeveloped using the following procedures:• Determine and understand the optimal load rangefor each chiller. This information should be availablefrom the chiller manufacturer. For example,chiller kW/ton typically has a minimum value(best efficiency) when the chiller load is somewherebetween 50% and 70% of the design value.• Turn on the most efficient chiller first. Optimize thepump and fan operation accordingly.• Turn on more chillers to maintain the load ratio(chiller load over the design load) within the optimalefficiency range for each chiller. This assumesthat the building by-pass is closed. If the buildingby-pass cannot be closed, the minimum chillerload ratio should be maintained at 50% or higherto limit primary pumping power increasesMaintain Good Operating Practices: The operatingprocedures recommended by the manufacturer shouldbe followed. It is important to calibrate the temperature,pressure, and current sensors and the flow switchesperiodically. The temperature sensors are especiallyimportant for maintaining efficient operation. Controlparameters must be set properly, particularly the timedelay relay.26.4.6 CC Measures for Central Heating PlantsCentral heating plants produce hot water, steam,or both, typically using either natural gas, coal or oil asfuel. Steam, hot water, or both are distributed to buildingsfor HVAC systems and other end uses, such ascooking, cleaning, sterilization and experiments. Boilerplant operation involves complex chemical, mechanicaland control processes. Energy performance and operationalreliability can be improved through numerousmeasures. However, the CC measures discussed in thissection are limited to those that can be implemented byan operating technician, operating engineers, and CCengineers.26.4.6.1 Optimize Supply Water Temperature and SteamPressureSteam pressure and hot water temperature are themost important safety parameters for a central heatingplant. Reducing the boiler steam pressure and hot watertemperature has the following advantages:• Improves plant safety• Increases boiler efficiency and decreases fuel consumption• Increases condensate return from buildings andimproves building automation system performance• Reduces hot water and steam leakage throughmalfunctioning valves26.4.6.2 Optimize Feed Water Pump OperationThe feed water pump is sized based on boilerdesign pressure. Since most boilers operate below thedesign pressure, the feed water pump head is oftensignificantly higher than required. This excessive pumphead is often dropped across pressure reducing valvesand manual valves. Installing a VSD on the feed waterpump in such cases can decrease pump power consumptionand improve control performance. Trimming theimpeller or changing feed water pumps may also befeasible, and the cost may be lower. However, the VSDprovides more flexibility, and it can be adjusted to anylevel. Consequently, it maximizes the savings and canbe adjusted to future changes as well.

694 ENERGY MANAGEMENT HANDBOOK26.4.6.3 Optimize Airside OperationThe key issues are excessive airflow and flu gastemperature control. Some excess airflow is requiredto improve the combustion efficiency and avoid havinginsufficient combustion air during fluctuations inairflow. However, excessive airflow will consume morethermal energy since it has to be heated from the outsideair temperature to the flue gas temperature. The boilerefficiency goes down as excessive airflow increases. Theflue gas temperature should be controlled properly. Ifthe flue gas temperature is too low, acid condensationcan occur in the flue. If the flue gas temperature is toohigh, it carries out too much thermal energy. The airsideoptimization starts with a combustion analysis, thatdetermines the combustion efficiency based on the flugas composition, flu gas temperature, and fuel composition.The typical combustion efficiency should be higherthan 80%. If the combustion efficiency is lower thanthis value, available procedures 37,38 should be used todetermine the reasons.26.4.6.4 Optimize Boiler StagingMost central plants have more than one boiler. Usingoptimal staging can improve plant energy efficiencyand reduce maintenance cost. The optimal stagingshould be developed using the following guidelines:• Measure boiler efficiency.• Run the higher efficiency boiler as the primarysystem and run the lower efficiency boiler as theback up system.• Avoid running any boiler at a load ratio less than40% or higher than 90%.• If two boilers are running at average load ratiosless than 60%, no stand-by boiler is necessary. Ifthree boilers are running at loads of less than 80%,no stand by boiler is necessary.Boiler staging involves boiler shut off, start up,and standby. Realizing the large thermal inertial andthe temperature changes between shut off, standby,and normal operation, precautions must be taken toprevent corrosion damage and expansion damage.Generally speaking, short-term (monthly) turn on/offshould be avoided for steam boilers. Hot water boilersare sometimes operated to provide water temperaturesas low as 80°F. This improves distribution efficiency, butmay lead to acid condensate in the flue. The hot watertemperature must be kept high enough to prevent thiscondensation.26.4.6.5 Improve Multiple Heat Exchanger OperationHeat exchangers are often used in central plants orbuildings to convert steam to hot water or high temperaturehot water to lower temperature hot water. If morethan one heat exchanger is installed, use as many heatexchangers as possible provided the average load ratiois 30% or higher. This approach provides the followingbenefits:• Lower pumping power. For example, if two heatexchangers are used instead of one under 100%load, the pressure loss through the heat exchangersystem will be decreased by 75%. The pumpingpower will also be decreased by 75%.• Lower leaving temperature on the heat source. Thecondensate should be super-cooled when the heatexchangers are operated at low load ratio. The exithot water temperature will be lower than the designvalue under the partial load condition. This willresult in less water or steam flow and more energyextracted from each pound of water or steam. Forexample, the condensate water may be sub-cooledfrom 215°F to 150°F under low heat exchangerloads. Compared with leaving the heat exchangerat 215°F, each pound of steam delivers 65 Btu morethermal energy to the heat exchanger.Using more heat exchangers will result in moresurface heat loss. If the load ratio is higher than 30%,the benefits mentioned above normally outweigh theheat loss. More information can be found in a paper byLiu et al. 3926.4.6.6 Maintain Good Operating Practices:Central plant operation involves both energy efficiencyand safety issues. Proper safety and maintenanceguidelines should be followed. The following maintenanceissues should be carefully addressed:• Blowdown: check blowdown setup if a boiler isoperating at partial load most of the time. Thepurpose of blowdown is to remove the mineral depositsin the drum. This deposit is proportional tothe cumulative make-up water flow, which is thenproportional to the steam or hot water production.The blowdown can often be set back significantly.If the load ratio is 40% or higher, the blowdowncan be reset proportional to the load ratio. If theload ratio is less than 40%, keep the blowdown rateat 40% of the design blowdown rate.• Steam traps: check steam traps frequently. Steamtraps have a tendency to fail, failed position is usuallyopen, and leakage costs can be significant. Asteam trap maintenance program is recommended.Consult the manufacturer and other manuals forproper procedures and methods.

COMMISSIONING FOR ENERGY MANAGEMENT 695• Condensate return: inspect the condensate returnfrequently. Make sure you are returning as muchcondensate as possible. This is very expensive water.It has high energy content and is treated water.When you lose condensate, you have to pay for themake-up water, chemicals, fuel, and, in some cases,sewage costs.26.4.7 Continuous Commissioning GuidelinesGuidelines can be used to assist in carrying theCC process out in an orderly manner. An abbreviatedsample set is provided in Appendix A that provides abasic check list, procedures, and documentation requirements.26.5 ENSURING OPTIMUMBUILDING PERFORMANCEThe CC activities described in the previous sectionswill optimize building system operation and reduce energyconsumption. To ensure excellent long-term performance,the following activities should be conducted.1. Document CC activities,2. Measure energy and maintenance cost savings,3. Train operating and maintenance staff,4. Measure energy data, and continuously measureenergy performance, and5. Obtain on-going assistance from CC engineers.This section discusses guidelines to perform thesetasks.26.5.1 Document the CC ProjectThe documentation should be brief and accurate.The operating sequences should be documented accuratelyand carefully. This documentation should not repeat theexisting building documentation. It should describe theprocedures implemented, including control algorithms,and briefly give the reasons behind these procedures.The emphasis is on accurate and usable documentation.The documentation should be easily used by operatingstaff. For example, operating staff should be ableto create operating manuals and procedures from thedocument.The CC project report should include accuratedocumentation of current energy performance, buildingdata, AHUs and terminal boxes, water loops and pumps,control system, and performance improvements.26.5.2 Measure Energy SavingsMost building owners expect the CC project to payfor itself through energy savings. Measurement of energysavings is one of the most important issues for CCprojects. The measurements should follow the proceduresdescribed in Chapter 27 (Measurement and Verification)of this handbook. Chapter 27 describes procedures fromthe International Performance Measurement and VerificationProtocol 40 (IPMVP). This section will provide a very briefdescription of these procedures, emphasizing issues thatare important in M&V for CC projects.The process for determining savings as adopted inthe IPMVP defines energy savings, E save , as:E Save = E base – E postwhere E base is the “baseline” energy consumption beforethe CC measures were implemented, and E post is themeasured consumption following implementation of theCC measures.Figure 26.10 shows the daily electricity consumptionof the air handlers in a large building in which theHVAC systems were converted from constant volumesystems to VAV systems using variable frequency drives.Consumption is shown for slightly over a year beforethe VFDs were installed (Pre), for about three monthsof construction and for about two years after installation(Post). In this case, the base daily electricity consumptionis 8,300 kWh/day. The post-retrofit electricity consumptionis 4,000 kWh/day, corresponding to electricitysavings of 4,300 kWh/day. During the construction period,the savings are slightly lower.However, in most cases, consumption shows morevariation from day to day and month to month than thatshown by the fan power for these constant speed fans.Hence, determination of the baseline must consider anumber of factors, including weather changes, changesin occupancy schedule, changes in number of occupants,remodeling of the spaces, equipment changes, etc.In the IPMVP, the baseline energy use, E base , is determinedeither from direct measurements, utility bills, orfrom a model of the building operation before the retrofit(or commissioning). The post-installation energy use isgenerally simply the measured energy use, but it may bedetermined from a model if measured data are not available.Before and after comparisons are often normalizedwith factors such as weather, occupancy, etc.The IPMVP includes four different M&V techniquesor options. These options, may be summarized asOption A—some measurements, but mostly stipulatedsavings, Option B: measurement at the system or devicelevel, Option C—measurement at the whole-buildingor facility level, and Option D—determination fromcalibrated simulation. Each option has its advantagesfor some special applications.The cost savings must also consider changes in

696 ENERGY MANAGEMENT HANDBOOKFigure 26.10 Daily electricity consumption for approximately one year before a retrofitand two years after the retrofit 41 .utility rates. It is generally recommended that the utilityrates in place before the retrofit or CC measures wereimplemented be used if any savings projections weremade, since those projections were made based on therates in effect at that time.In general, the least expensive M&V method thatwill provide the accuracy required should be used. Utilitybilling data will be the least expensive data wheneverit is available. It will sometimes provide the requiredaccuracy for CC projects, but has the disadvantagethat it may take considerable time before the improvedperformance is clearly evident. The discussion belowgenerally assumes that higher frequency data is beingused.26.5.2.1 Option A - Stipulated Savings(Involving some measurements)The stipulated option determines savings by measuringthe capacity or the efficiency of a system beforeand after retrofit or commissioning, and then multipliesthe difference by an agreed upon or “stipulated” factorsuch as the hours of operation, or the load on the system.This option focuses on a physical determination ofequipment changes to ensure that the installation meetscontract specifications. Key performance factors (e.g.lighting wattage) are determined with spot or shorttermmeasurements and operational factors (e.g. lightingoperating hours) are stipulated based on historical dataor spot measurement. Performance factors are measuredor checked yearly. This method provides reliable savingsestimation for applications where the energy savings areindependent of weather and occupancy conditions.For example, during the CC process, the fan pulleywas decreased from 18” to 16” for a constant volumeAHU. The fan power savings can be determined usingthe following method:• Measure the fan power consumption before changingthe pulley and the power consumption afterchanging the pulley.• Determine the number of hours in operation.• Determine the fan power savings as the productof the hourly fan power energy savings and thenumber of hours in operation.If the energy consumption varies with occupancyand weather conditions, this option should not be used.For example, the minimum airflow setting was adjustedfrom 50% to 0% for 100 VAV terminal boxes at night andduring weekends. Since the airflow depends on bothinternal and external loads, and the airflow may not be0% even if the minimum flow setting is 0%, this methodcannot be used to determine savings.

COMMISSIONING FOR ENERGY MANAGEMENT 697If Option A can provide the required accuracy, itwill generally be the second least expensive method,after utility data.26.5.2.2 Option B - Device/System Level MeasurementWithin Option B, savings are determined by continuousmeasurements taken throughout the projectterm at the device or system level. Individual loads orend-uses are monitored continuously to determine performanceand the long-term persistence of the measuresinstalled. The base line model can be developed usingthe measured energy consumption and other parameters.The energy savings can be determined as the differencebetween baseline energy consumption and themeasured energy consumption. This method providesthe best saving estimation for an individual device orsystem.The data collected can also be used to improve oroptimize the system operation, and as such is particularlyvaluable for continuous commissioning projects.Since measurements are taken throughout the projectterm, the cost is higher than option A.26.5.2.3 Option C - Whole Building Level MeasurementDetermines savings by analyzing “whole-building”or facility level data measured during the baselineperiod and the post-installation period. This option isrequired when it is desired to measure interaction effects,e.g. the impact of a lighting retrofit on the coolingconsumption as well as savings in lighting energy. Thedata used may be utility data, or sub-metered data.The minimum number of measurement channelsrecommended for performance assurance or savingsmeasurement will be the number needed to separateheating, cooling and other electric uses. The actual numberof channels will vary, depending on whether pulsesare taken from utility meters, or if two or three currenttransformers are installed to measure the three phasepower going into a chiller. Other channels may need tobe added, depending on the specific measures that arebeing evaluated.Option C requires that installation of the propersystems/equipment and proper operating practicesmust be confirmed. It determines savings from metereddata taken throughout the project term. The major limitationin the use of Option C for savings determinationis that the size of the savings must be larger than theerror in the baseline model. The major challenge is accountingfor changes other than those associated withthe ECMs or the commissioning changes implemented.Accurate determination of savings using Option Cnormally requires 12 months of continuous data beforea retrofit and continuous data after retrofit. However,for commissioning applications, a shorter period of dataduring which daily average ambient conditions covera large fraction of normal yearly variation is often adequate.26.5.2.4 Option D - Calibrated SimulationSavings are determined through simulation ofthe facility components and/or the whole facility. Themost powerful application of this approach calibratesa simulation model to baseline consumption data. Forcommissioning applications, it is recommended thatcalibration be to daily or hourly data. This type ofcalibration may be carried out most rapidly if simulateddata is compared to measured data as a functionof ambient temperature. Wei et al. 42 have developed“energy signatures” that greatly aid this process. Moreinformation can be found in Liu and Claridge 43 and amanual providing instructions on use of the method isavailable 44 .Just as for the other options, the implementationof proper operating practices should be confirmed. Itis particularly important that personnel experiencedin the use of the particular simulation tool conductthe analysis. The simulation analysis needs to be welldocumented, with electronic and hard copies of thesimulation program input and output preserved.26.5.2.5 Data Used to Determine SavingsNote that monthly bills may be used to estimatethe energy savings. This method is one version of OptionC described above. It is typically the least expensivemethod of verification. It will work fine if the followingconditions are met:1. Significant savings are expected at the utility meterlevel2. Savings are too small to cost-justify more data3. There will be no changes ina. Equipmentb. Schedulesc. Occupancyd. Space utilizationThe case shown in Figure 26.11 is an examplewhere monthly bills clearly show the savings. The savingswere large and consistent following the retrofit untilJune. At this point, a major deviation occurred. The presenceof other metering at this site showed that the utilitybill was incorrect. Further investigation showed that theutility meter had been changed and this had not beenconsidered in the bill sent. The consumption includedin this bill was greater than the site would have usedif it had used the peak demand recorded on the utility

698 ENERGY MANAGEMENT HANDBOOKFigure 26.11 Comparison of monthly utility bills before (top line) and after (bottomline) a retrofit 45 .meter for every hour of the billing period!However, daily or hourly data will show the resultsof commissioning measures much more quicklyand are an extremely valuable diagnostic tool whenproblems arise as described in Section 26.5.4. Hence itis recommended that such data be used for savings determinationand follow-up tracking whenever possible.26.5.3 Trained Operating and Maintenance StaffEfficient building operation begins with a qualifiedand committed staff. Since the CC process generallymakes changes in the way a building is operated to improvecomfort and efficiency, it is essential that the operatorsbe a part of the commissioning team. They needto work with the CC engineers, propose CC measuresand implement or help implement them. In addition toactively participating in the CC process, formal technicaltraining should be provided to ensure that the operatingstaff understands the procedures implemented so theycan perform trouble-shooting properly.26.5.4 Continuously Measure Energy PerformanceThe measurement of energy consumption data isvery important to maintain building performance andmaintain CC savings. The metered data can be used to:1. Identify and solve problems. Metered consumptiondata is needed to be sure that the building is stilloperating properly. If there is a component failureor an operating change that makes such a smallchange in comfort or operating efficiency that it isnot visible in metered consumption data, it generallyisn’t worth worrying about. If it does showup as even a marginal increase in consumption,trouble-shooting should be initiated.2. Trend/measure energy consumption data. This continuingactivity is the first line of defense againstdeclining performance. The same procedures usedto establish a pre-CC baseline can be used to establisha baseline for post-CC performance, andthis post-CC baseline can be used as a standardagainst which future performance is compared.Consumption that exceeds this baseline for a fewdays, or even a month may not be significant, butif it persists much more than a month, troubleshootingshould be used to find out what has ledto the increase. If it is the result of a malfunctioningvalve, you can fix it. If it is the result of 100 newcomputers added to the building, you will adjustthe baseline accordingly.3. Trend and check major operating parameters. Parameterssuch as cold-deck temperatures, zone supplytemperatures, etc. should be trended periodicallyfor comparison with historic levels. This can beextremely valuable when trouble-shooting and

COMMISSIONING FOR ENERGY MANAGEMENT 699when investigating consumption above the post-CC baseline.4. Find the real problems when the system needs to berepaired or fixed. It is essential that the same fundamentalapproach used to find and fix problemswhile the CC process is initiated be used whenevernew hot calls or cold calls are received.26.5.5 Utilize Expert Support as NeededIt is inevitable that a problem will come up which,even after careful trouble-shooting points toward a problemwith one or more of the CC measures that havebeen implemented. Ask the CC providers for help insolving such problems before undoing an implementedCC measure. Sometimes it will be necessary to modify ameasure that has been implemented. The CC engineerswill often be able to help with finding the most efficientsolution, and they will sometimes be able to help youfind another explanation, so the problem can be remediedwithout changing the measure.Ask help from the CC providers when you run intoa new problem or situation. Problems occasionally cropup that defy logical explanation. These are the problemsthat generally get resolved by trying one of threethings that seem like possible solutions, and playingwith system settings until the problem goes away. Thisis one of the most important situations in which experthelp is needed. Without the expert’s help, these kinds ofproblems—and the trial and error solutions—can undothe measure’s savings potential.26.5.6 How Well Do Commissioning Savings Persist?The Energy Systems Laboratory at Texas A&M hasconducted a study of 10 buildings on the Texas A&Mcampus that had CC measures implemented in 1996-97 46,47 . Table 26.5 shows the baseline cost of combinedheating, cooling and electricity use of each buildingand the commissioning savings for 1998 and 2000. Thebaseline consumption and savings for each year werenormalized to remove any differences due to weather.Looking at the totals for the group of 10 buildings,heating and cooling consumption increased by $207,258(12.1%) from 1998 to 2000, but savings from the earliercommissioning work were still $985,626. However, itmay also be observed that almost three-fourths of thisconsumption increase occurred in two buildings, theKleberg Building, and G. Rollie White Coliseum. The increasedconsumption of the Kleberg Building was due toa combination of component failures and control problemsas described in the case study in Section 26.5.6.1.The increased consumption in G. Rollie White ColiseumTable 26.5. Commissioning savings in 1998 and 2000 for 10 buildings on the Texas A&M campus48 .———————————————————————————————————————————Building Baseline Use 1998 Savings 2000($/Yr) ($/Yr) Savings($/yr)———————————————————————————————————————————Kleberg Building $484,899 $313,958 $247,415———————————————————————————————————————————G.R. White Coliseum $229,881 $154,973 $71,809———————————————————————————————————————————Blocker Building $283,407 $76,003 $56,738———————————————————————————————————————————Eller O&M Building $315,404 $120,339 $89,934———————————————————————————————————————————Harrington Tower $ 145,420 $64,498 $48,816———————————————————————————————————————————Koldus Building $ 192,019 $57,076 $61,540———————————————————————————————————————————Richardson Petroleum Building $273,687 $120,745 $120,666———————————————————————————————————————————Veterinary Medical Center $324,624 $87,059 $92,942Addition———————————————————————————————————————————Welmer Business Building $224,481 $47,834 $68,145———————————————————————————————————————————Zachry Engineering Center $436,265 $150,400 $127,620———————————————————————————————————————————Totals $2,910,087 $ 1,192,884 $985,626———————————————————————————————————————————

700 ENERGY MANAGEMENT HANDBOOKwas due to different specific failures and changes, butwas qualitatively similar to Kleberg since it resultedfrom a combination of component failures and controlchanges. The five buildings that showed consumptionchanges of more than 5% from 1998 to 2000 were allfound to have different control settings that appear toaccount for the changed consumption (including thedecrease in the Wehner Business Building).These data suggest that commissioning savingsgenerally persist, but tracking can subsequently uncoverproblems that did not cause comfort problems, but haveincreased consumption by $10,000-$100,000 per year inlarge buildings.26.5.6.1 Commissioning persistence case study - KlebergBuilding 49The Kleberg Building is a teaching/research facilityon the Texas A&M campus consisting of classrooms,offices and laboratories, with a total floor area of approximately165,030 ft 2 . Ninety percent of the buildingis heated and cooled by two (2) single duct variable airvolume (VAV) air handling units (AHU) each havinga pre-heat coil, a cooling coil, one supply air fan (100hp), and a return air fan (25 hp). Two smaller constantvolume units handle the teaching/lecture rooms in thebuilding. The campus plant provides chilled water andhot water to the building. The two (2) parallel chilledwater pumps (2 × 20 hp) have variable frequency drivecontrol. There are 120 fan-powered VAV boxes withterminal reheat in 12 laboratory zones and 100 fan-poweredVAV boxes with terminal reheat in the offices. Thereare six (6) exhaust fans (10-20 hp, total 90 hp) for fumehoods and laboratory general exhaust. The air handlingunits, chilled water pumps and 12 laboratory zones arecontrolled by a direct digital control (DDC) system. DDCcontrollers modulate dampers to control exhaust airflowfrom fume hoods and laboratory general exhaust.A CC investigation was initiated in the summerof 1996 due to the extremely high level of simultaneousheating and cooling observed in the building (Abbas,1996). Figures 26.12 and 26.13 show daily heating andcooling consumption (expressed in average kBtu/hr)as functions of daily average temperature. The Pre-CCheating consumption data given in Figure 26.13 showsvery little temperature dependence as indicted by theregression line derived from the data. Data values weretypically between 5 and 6 MMBtu/hr with occasionallower values. The cooling data (Figure 26.12) showsmore temperature dependence and the regression lineindicates that average consumption on a design daywould exceed 10 MMBtu/hr. This corresponds to only198 sq.ft./ton based on average load, which is a highcooling density.It was soon found that the preheat was operatingcontinuously, heating the mixed air entering the coolingcoil to approximately 105˚F, instituted in response toa humidity problem in the building. The preheat wasturned off and heating and cooling consumption bothdropped by about 2 MMBtu/hour as shown by themiddle clouds of data in Figures 26.12 and 26.13. Subsequently,the building was thoroughly examined anda comprehensive list of commissioning measures wasdeveloped and implemented. The principal measuresimplemented that led to reduced heating and coolingconsumption were:• Preheat to 105˚F was changed to preheat to 40˚F• Cold deck schedule changed from 55˚F fixed tovary from 62˚F to 57˚F as ambient temperaturevaries from 40˚F to 60˚FFigure 26.12 Pre-CC and post-CC heatingwater consumption at the KlebergBuilding vs. daily average outdoortemperature 50 .

COMMISSIONING FOR ENERGY MANAGEMENT 701Figure 26.13 Pre-CC and post-CCchilled water consumption at theKleberg Building vs. daily averageoutdoor temperature 51 .• Economizer—set to maintain mixed air at 57˚Fwhenever outside air below 60˚F• Static pressure control—reduced from 1.5 inH 2 O to1.0 inH 2 O and implemented night-time set back to0.5 inH 2 O• Replaced or repaired a number of broken VFDboxes• Chilled water pump VFDs were turned on.Additional measures implemented includedchanges in CHW pump control—changed so one pumpmodulates to full speed before the second pump comeson instead of operating both pumps in parallel at alltimes, building static pressure was reduced from 0.05 in.w.c. to 0.02 in. w.c., and control changes were made toeliminate hunting in several valves. It was also observedthat there was a vibration at a particular frequency in thepump VFDs that influenced the operators to place theseVFDs in the manual mode, so it was recommended thatthe mountings be modified to solve this problem.These changes further reduced chilled water andheating hot water use as shown in Figures 26.12 and26.13 for a total annualized reduction of 63% in chilledwater use and 84% in hot water use. Additional followupconducted from June 1998 through April 1999 focusedon air balance in the 12 laboratory zones, general exhaustsystem rescheduling, VAV terminal box calibration, adjustingthe actuators and dampers, and calibrating fumehoods and return bypass devices to remote DDC control 52(Lewis, et al. 1999). These changes reduced electricityconsumption by about 7% or 30,000 kWh/mo.In 2001 it was observed that chilled water savingsfor 2000 had declined to 38% and hot water savings to62% as shown in Table 26.6. Chilled water data for 2001and the first three months of 2002 are shown in Figure26.14. The two lines shown are the regression fits to thechilled water data before CC implementation and afterimplementation of CC measures in 1996 as shown inFigure 26.13. It is evident that consumption during 2001is generally appreciably higher than immediately followingimplementation of CC measures. The CC groupperformed field tests and analyses that soon focused ontwo single-duct VAV AHU systems, two chilled waterpumps, and the Energy Management Control System(EMCS) control algorithms as described in Chen et al. 53Several problems were observed as noted below.Problems Identified• The majority of the VFDs were running at a constantspeed near 100% speed.• VFD control on two chilled water pumps wasagain bypassed to run at full speed.• Two chilled water control valves were leaking badly.Combined with a failed electronic to pneumaticswitch and the high water pressure noted above,this resulted in discharge air temperatures of 50°Fand lower and activated preheat continuously.• A failed pressure sensor and two failed CO 2 sensorsput all outside air dampers to the full openposition.• The damper actuators were leaking and unable tomaintain pressure in some of the VAV boxes. This

702 ENERGY MANAGEMENT HANDBOOKTable 26.6. Chilled water and heating water usage and saving in the Kleberg Building for threedifferent years normalized to 1995 weather 54.——————————————————————————————————————————————Type Pre-CC Post-CC Use/Savings 2000 Use/SavingsBaseline Use Savings % Use Savings (%)(MMBtu/yr) (MMBtu/yr) (MMBtu/yr)——————————————————————————————————————————————CHW 72935 26537 63.6% 45431 37.7%——————————————————————————————————————————————HW 43296 6841 84.2% 16351 62.2%——————————————————————————————————————————————caused cold air to flow through the boxes evenwhen they were in the heating mode, resulting insimultaneous heating and cooling. Furthermoresome of the reheat valves were malfunctioning.This caused the reheat to remain on continuouslyin some cases.• Additional problems identified from the field surveyincluded the following: 1) high air resistancefrom the filters and coils, 2) errors in a temperaturesensor and static pressure sensor, 3) high staticpressure setpoints in AHU1 and AHU2.This combination of equipment failure compoundedby control changes that returned several pumps andfans to constant speed operation had the consequenceof increasing chilled water use by 18,894 MMBtu andhot water use by 9,510 MMBtu. This amounted to anincrease of 71% in chilled water use and more thandoubled hot water use from two years earlierThese problems have now been corrected and buildingperformance has returned to previously low levels asillustrated by the data for April-June 2002 in Figure 26.4.These data are all below the lower of the two regressionlines and is comparable to the level achieved after additionalCC measures were implemented in 1998-99.26.6 COMMISSIONING NEW BUILDINGSFOR ENERGY MANAGEMENTThe energy manager’s effort is generally directedtoward the improving the efficiency of existing buildings.However, whenever the organization initiatesdesign and construction of a new building that willbecome part of the energy manager’s portfolio of buildings,it is extremely important that the energy managerbecome an active part of the design and constructionteam to ensure that the building incorporates all appropriateenergy efficiency technologies. It is just asFigure 26.14 ChW data for the Kleberg Building forJanuary 2001- June 2002 55 .important that the perspective of operational personnelbe included in the design process so it will be possibleto effectively and efficiently operate the building. Oneof the best ways to accomplish these objectives is tocommission the building as it is designed and built.The primary motivation for commissioning HVACsystems is generally to achieve HVAC systems that workproperly to provide comfort to building occupants atlow cost. In principle, all building systems should bedesigned, installed, documented, tested, and buildingstaff trained in their use. In practice, competitivepressures and fee structures, and financial pressures tooccupy new buildings as quickly as possible generallyresult in buildings that are handed over to the ownerswith minimal contact between designers and operators,minimal functional testing of systems, documentationthat largely consists of manufacturers system componentmanuals, and little or no training for operators.This in turn leads to problems such as mold growth inwalls of new buildings, rooms that never cool properly,air quality problems, etc. Such experiences were doubtlessthe motivation for the facility manager for a largeuniversity medical center who stated a few years agothat he didn’t want to get any new buildings. He only

COMMISSIONING FOR ENERGY MANAGEMENT 703wanted three-year old buildings in which the problemshad been fixed.Although commissioning provides higher qualitybuildings and results in fewer initial and subsequentoperational problems, most owners will include commissioningin the design and construction process onlyif they believe they will benefit financially from commissioning.It is much more difficult to document theenergy cost savings from commissioning a new buildingthan an existing building. There is no historical use patternto use as a baseline. However, it has been estimatedthat new building commissioning will save 8% in energycost alone compared with the average building which isnot commissioned 56 . This offers a payback for the cost ofcommissioning in just over four years from the energysavings alone and also provides improved comfort andair quality.Commissioning is often considered to be a punchlistprocess that ensures that the systems in a buildingfunction before the building is turned over to the owner.However, the process outlined in Table 26.7 shows theprocess beginning in the pre-design phase. It is most effectiveif allowed to influence both design and construction.It is essential that the energy manager be involvedin the commissioning process on the owner’s team nolater than the design phase of the construction process.This permits input into the design process that can havemajor impact on the efficiency of the building as builtand can lead to a building that has far fewer operationalproblems.26.7 SUMMARYCommissioning of existing buildings is emergingas one of the most cost effective ways for an energymanager to lower operating costs, and typically doesso with no capital investment, or with a very minimalamount. It has been successfully implemented in severalhundred buildings and provides typical paybacks of oneto three years.It is much more than the typical O&M program. Itdoes not ensure that the systems function as originallydesigned, but focuses on improving overall system controland operations for the building as it is currentlyutilized and on meeting existing facility needs. Duringthe CC process, a comprehensive engineering evaluationis conducted for both building functionality and systemfunctions. The optimal operational parameters andschedules are developed based on actual building conditions.An integrated approach is used to implementthese optimal schedules to ensure practical local andTable 26.7. The commissioning process for new buildings57 .—————————————————————————1. Conception or pre-design phase(a) Develop commissioning objectives(b) Hire commissioning provider(c) Develop design phase commissioning requirements(d) Choose the design team—————————————————————————2. Design phase(a) Commissioning review of design intent(b) Write commissioning specifications for biddocuments(c) Award job to contractor(d) Develop commissioning plan—————————————————————————3. Construction/Installation phase(a) Gather and review documentation(b) Hold commissioning scoping meeting and finalizeplan(c) Develop pre-test checklists(d) Start up equipment or perform pre-test checkliststo ensure readiness for functional testingduring acceptance—————————————————————————4. Acceptance phase(a) Execute functional test and diagnostics(b) Fix deficiencies(c) Retest and monitor as needed(d) Verify operator training(e) Review O&M manuals(f) Building accepted by owner—————————————————————————5. Post-acceptance phase(a) Prepare and submit final report(b) Perform deferred tests (if needed)(c) Develop recommissioning plan/schedule—————————————————————————global system optimization and to ensure persistence ofthe improved operational schedules.The approach presented in this chapter begins byconducting a thorough examination of all problem areasor operating problems in the building, diagnoses theseproblems, and develops solutions that solve these problemswhile almost always reducing operating costs atthe same time. Equipment upgrades or retrofits may beimplemented as well, but have not been a factor in thecase studies presented, except where the commissioningwas used to finance equipment upgrades. This is insharp contrast to the more usual approach to improving

704 ENERGY MANAGEMENT HANDBOOKthe efficiency of HVAC systems and cutting operatingcosts that primarily emphasizes system upgrades orretrofits to improve efficiency.Commissioning of new buildings is also an importantoption for the energy manager, offering an opportunityto help ensure that new buildings have theenergy efficiency and operational features that are mostneeded.26.8. FOR ADDITIONAL INFORMATIONTwo major sources of information on commissioningexisting buildings are the Continuous CommissioningSM Guidebook: Maximizing Building EnergyEfficiency and Comfort (Liu, M., Claridge, D.E. andTurner, W.D., Federal Energy Management Program,U.S. Dept. of Energy, 144 pp., 2002) and A Practical Guidefor Commissioning Existing Buildings (Haasl, T. and Sharp,T., Portland Energy Conservation, Inc. and Oak RidgeNational Laboratory for U.S. DOE, ORNL/TM-1999/34,69 pp. + App., 1999). Much of this chapter has beenabridged and adapted from the CC Guidebook.There are a much wider range of materials availablethat treat commissioning of new buildings. Twodocuments that provide a good starting point areASHRAE Guideline 1-1996: The HVAC CommissioningProcess (American Society of Heating, Refrigerating andAir-Conditioning Engineers, Atlanta, GA, 1996) and theBuilding Commissioning Guide—Version 2.2 (U.S. GSAand U.S. DOE, 1998 by McNeil Technologies, Inc. andEnviro-Management & Research, Inc. available at www.emrinc.com).The case studies in this chapter have been largelyabridged and adapted from the following three papers:Claridge, D.E., Liu, M., Deng, S., Turner, W.D.,Haberl, J.S., Lee, S.U., Abbas, M., Bruner, H., and Veteto,B., “Cutting Heating and Cooling Use Almost in HalfWithout Capital Expenditure in a Previously RetrofitBuilding,” Proc. of 2001 ECEEE Summer Study, Mandeliu,France, Vol. 2, pp. 74-85, June 11-16, 2001.Turner, W.D., Claridge, D.E., Deng, S. and Wei, G.,“The Use of Continuous Commissioning SM As An EnergyConservation Measure (ECM) for Energy Efficiency Retrofits,”Proc. of 11th National Conference on Building Commissioning,Palm Springs, CA, CD, May 20-22, 2003.Claridge, D.E., Turner, W.D., Liu, M., Deng, S., Wei,G., Culp, C., Chen, H. and Cho, S.Y., “Is CommissioningOnce Enough?,” Solutions for Energy Security & FacilityManagement Challenges: Proc. of the 25th WEEC, Atlanta,GA, pp. 29-36, Oct. 9-11, 2002.References1. ASHRAE, ASHRAE Guideline 1-1996: The HVAC CommissioningProcess, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, GA, 1996.2. U.S. Department of Energy, Building Commissioning: The Keyto Quality Assurance, Washington, DC. 1999.3. ibid.4. Continuous Commissioning is a registered trademark and CCis a service mark of the Texas Engineering Experiment Station(TEES). Contact TEES for further information.5. Liu, M., Claridge, D.E. and Turner, W.D., Continuous CommissioningSM Guidebook: Maximizing Building Energy Efficiency andComfort, Federal Energy Management Program, U.S. Dept. ofEnergy, 144 pp., 2002.6. Claridge, D.E., Haberl, J., Liu, M., Houcek, J., and Athar, A.,“Can You Achieve 150% of Predicted Retrofit Savings: Is It Timefor Recommissioning?” A CEEE 1994 Summer Study on EnergyEfficiency In Buildings Proceedings: Commissioning, Operation andMaintenance, Vol. 5, American Council for an Energy EfficientEconomy, Washington, D.C., pp. 73-87, 1994.7. Claridge, D.E., Liu, M., Deng, S., Turner, W.D., Haberl, J.S., Lee,S.U., Abbas, M., Bruner, H., and Veteto, B., “Cutting Heatingand Cooling Use Almost in Half Without Capital Expenditurein a Previously Retrofit Building,” Proc. of 2001 ECEEE SummerStudy, Mandeliu, France, Vol. 2, pp. 74-85, June 11-16, 2001.8. Turner, W. D., Claridge, D. E., Deng, S. and Wei, G., “The Useof Continuous Commissioning SM As An Energy ConservationMeasure (ECM) for Energy Efficiency Retrofits,” Proc. of11th National Conference on Building Commissioning, PalmSprings, CA, CD, May 20-22, 2003.9. Zhu, Y., Liu, M., Claridge, D.E., Feary, D. and Smith, T., “AContinuous Commissioning Case Study of a State-of-the-ArtBuilding,” Proceedings of the 5th National Commissioning Conference,Huntington Beach, CA, pp. 13.1 - 13. 10, April, 1997.10. Liu, M., Zhu, Y., Powell, T., and Claridge, D.E., “System OptimizationSaves $195,000/yr. in a New Medical Facility,” Proceedingsof the 6th National Conference on Building Commissioning,Lake Buena Vista, FL, pp. 14.2.1-14.2.11, May 18-20, 1998.11. Turner, et al., op. cit.12. Source: Adapted from Turner et al. op. cit.13. Source: Liu, Claridge, and Turner, op. cit.14. Source: ibid.15. Source: ibid.16. Source: ibid.17. Liu, M., Zhu, Y., Park, B.Y., Claridge, D.E., Feary, D.K. and Gain,J., “Airflow Reduction to Improve Building Comfort and ReduceBuilding Energy Consumption - A Case Study,” ASHRAETransactions-Research, Vol. 105, Part 1, pp. 3 84 - 3 90, 1999.18. Liu, M., Zhu, Y., Claridge, D. and White, E., “Impacts of StaticPressure Set Level on the HVAC Energy Consumption andIndoor Conditions,” ASHRAE Transactions-Research. Volume 103,Part 2, pp. 221-228, 1997.19. Liu, Claridge, and Turner, op. cit.20. ibid.21. ibid.22. Liu, M., “Variable Speed Drive Volumetric Tracking (VSDVT)for Airflow Control in Variable Air Volume (VAV) Systems,”Proceedings of Thirteenth Symposium on Improving BuildingSystems in Hot and Humid Climates, San Antonio, TX, pp.195-198, May 15-16, 2002.23. Claridge, et al. 2001, op. cit.24. Source: ibid.25. Source: ibid.26. Source: ibid.27. Source: ibid.28. Source: ibid.29. Source: ibid.

COMMISSIONING FOR ENERGY MANAGEMENT 70530. Liu, Claridge, and Turner, op. cit.31. Liu, M., “Variable Water Flow Pumping for Central ChilledWater Systems,” ASME Journal of Solar Energy Engineering, Vol.124, pp. 300-304, 2002.32. Deng, S., Turner, W.D., Batten, T., and Liu, M., “ContinuousConumriissioning SM of a Central Chilled Water and HeatingHot Water System,” Proc. Twelfth Symposium on ImprovingBuilding Systems in Hot and Humid Climates, San Antonio,TX, pp. 199-206, May 15-16, 2000.33. Deng, S. Turner, W. D., Claridge, D. E., Liu, M., Bruner, H.,Chen, H. and Wei, G. “Retrocommissioning of Central Chilled/Hot Water Systems,” ASHRAE Transactions-Research, Vol. 108,Part 2, pp. 75-81, 2002.34. Liu, Claridge, and Turner, op. cit.35. Liu, M., op. cit.36. Liu, Claridge, and Turner, op. cit.37. Wei, G., Liu, M., and Claridge, D.E., “In-situ Calibration ofBoiler Instrumentation Using Analytic Redundancy,” InternationalJournal of Energy Research, Vol. 25, pp. 375-387, 2001.38. Liu, M. et al., 1998, op. cit.39. IpMVp Committee, International Performance Measurement &Verification Protocol.- Concepts and Options for Determining Energyand Water Savings, Vol. 1, U.S. Dept. of Energy, DOE/GO-102001-1187, January, 2001, 86 pp40. Source: Liu, Claridge, and Turner, op. cit.41. Wei, G., Liu, M., and Claridge, D.E., “Signatures of Heating andCooling Energy Consumption for Typical AHUs,” The EleventhSymposium on Improving Building Systems in Hot and HumidClimates Proceedings, Fort Worth, Texas, pp. 387-402, June 1-2,1998.42. Liu, M. and Claridge, D.E., “Use of Calibrated HVAC SystemModels to Optimize System Operation,” ASME Journal of SolarEnergy Engineering, Vol. 120, pp. 131-138, 1998.43. Claridge, D.E., Bensouda, N., Lee, S.U., Wei, G., Heinemeier, K.,and Liu, M., Manual of Procedures for Calibrating Simulations ofBuilding Systems, submitted to the Lawrence Berkeley NationalLaboratory under Sponsored Contract #66503346 as part of theCalifornia Energy Commission Public Interest Energy ResearchProgram, 94 pp., October, 2003.44. Source: Liu, Claridge, and Turner, op. cit.45. Turner, W.D., Claridge, D.E., Deng, S., Cho, S., Liu, M., Hagge,T., Darnell, C., Jr., and Bruner, H., Jr., “Persistence of SavingsObtained from Continuous Conunissioning SM ,” Proc. of 9thNational Conference on Building Commissioning, Cherry Hill, NJ,pp. 20-1.1 - 20-1.13, May 9-11, 2001.46. Claridge, D.E., Turner, W.D., Liu, M., Deng, S., Wei, G., Culp, C.,Chen, H. and Cho, S.Y., “Is Commissioning Once Enough?,” Solutionsfor Energy Security & Facility Management Challenges: Proc.of the 25 h WEEC, Atlanta, GA, pp. 29-36, Oct. 9-11, 2002.47. Source: ibid.48. ibid.49. Source: ibid.50. Source: ibid.51. Lewis, T., H. Chen, and M. Abbas, “CC summary for the KlebergBuilding,” Internal ESL Report, July, 1999.52. Chen, H., Deng, S., Bruner, H., Claridge, D. and Turner, W.D.,“Continuous Commissioning SM Results Verification And Follow-UpFor An Institutional Building - A Case Study,” Proc.13th Symposium on Improving Building Systems in Hot and HumidClimates, Houston, TX, pp. 87-95, May 20-23, 2002.53. Source: Claridge, D.E. et al. 2002, op. cit.54. Source: ibid.55. PECI, “National Strategy for Building Commissioning,” PortlandEnergy Conservation, Inc., Portland, OR, 1999.56. Adapted from Haasl, T. and Sharp, T., A Practical Guide for CommissioningExisting Buildings, Portland Energy Conservation, Inc.and Oak Ridge National Laboratory for U.S. DOE, ORNL/TM-1999/34, 69 pp. + App., 1999.


CHAPTER 27MEASUREMENT AND VERIFICATION OF ENERGY SAVINGSJEFF S. HABERL, PH.D., P.E.CHARLES C. CULP, PH.D., P.E.Energy Systems LaboratoryTexas A&M University27.1 INTRODUCTION27.1.1 M&V Method SelectionMeasurement and verification (M&V) has a dualrole. First, M&V quantifies the savings being obtained.This applies to the initial savings and the long-term savings.Since the persistence of savings has been shownto decrease with time, 1 long term M&V provides datato make these savings sustainable. Second, M&V mustbe cost effective so that the cost of measurement andthe analysis does not consume the savings. 2,3 The 1997International Performance Measurement and VerificationProtocol (IPMVP) set the target costs for M&V to be in therange of 1% to 10%, depending upon the Option selected,of the construction cost for the life of the ECM. Mostapproaches fall in the recommended range of 3% to 10%of the construction cost. The IPVMP 2001 removed thisguidance on the recommended costs for M&V. Currently,a goal of about 5% of the savings per year has evolvedas a preferred criterion for costing M&V, since the costjustification directly results from the calculation.A general procedure for selecting an approach canbe summarized by the following five steps:a) In general one wants to try to... Perform MonthlyUtility Bill Before/After Analysis.b) And if this does not work, then… Perform Daily orHourly Before/After Analysis.c) And if this does not work, then… Perform ComponentIsolation Analysis.d) And if this does not work, then… Perform CalibratedSimulation Analysis.e) Then… Report savings and Finish Analysis.27.1.2 History of M&V27.1.2.1 History of Building Energy Measurement.The history of the measurement of building energyuse can be traced back to the 19th century for electricity,and earlier for fuels such as coal and wood, whichwere used to heat buildings. 4,5,6,7 By the 1890s, althoughelectricity was common in many new commercial buildings,its use was primarily for incandescent lighting and,to a lesser extent, for the electric motors associated withventilating buildings since most of the work in officebuildings was carried out during daylight hours. Themetering of electricity closely paralleled the spread ofelectricity into cities as its inventors needed to recoverthe cost of its production through the collection of paymentsfrom electric utility customers. 8,9 Commercial metersfor the measurement of flowing liquids in pipes canbe traced back to the same period, beginning with theinvention of the first commercial flow meter by ClemensHerschel in 1887, which used principles based on thepitot tube and venturi flow meter, invented in 1732 and1797 by their respective namesakes. 10 Commercial metersfor the measurement of natural gas can likewise betraced to the sale and distribution of natural gas, whichparalleled the development of the electric meters.27.1.2.2 History of M&V in the U.S.27.1.2.2.1 History of M&VThe history of the measurement and verification ofbuilding energy use parallels the development and useof computerized energy calculations in the 1960s, with amuch accelerated awareness in 1973, when the embargoon Mideast oil made energy a front page issue. 11,12 Duringthe 1950s and 1960s most engineering calculationswere performed using slide rules, engineering tablesand desktop calculators that could only add, subtract,multiply and divide. Since the public was led to believeenergy was cheap and abundant, 13 the measurementand verification of the energy use in a building was limitedfor the most part to simple, unadjusted comparisonsof monthly utility bills.In the 1960s several efforts were initiated to formulateand codify equations that could predict dynamicheating and cooling loads, including efforts at the NationalBureau of Standards to predict temperatures infallout shelters, 14 and the 1967 HCC program developedby a group of mechanical engineering consultants, 15which used the Total Equivalent Temperature Difference/TimeAveraging (TETD/TA) method. The popularityof this program prompted the American Society ofHeating, Refrigeration and Air-Conditioning Engineers(ASHRAE) to embark on a series of efforts that eventuallydelivered today’s modern, general purpose simulationprograms 16 (i.e., DOE-2, BLAST, EnergyPlus, etc.),707

708 ENERGY MANAGEMENT HANDBOOKwhich utilize thermal response factors, 17,18 as well asalgorithms for simulation of the quasi-steady-state performanceof primary and secondary equipment. 19 Oneof these efforts was to validate the hourly calculationswith field measurements at the Legal Aid Building onthe Ohio State University campus, 20 which is probablythe first application of a calibrated, building energysimulation program.Some of the earliest efforts to develop standardizedmethods for the M&V of building energy use beganwith efforts to normalize residential heating energy usein single-family and multi-family buildings, 21 whichinclude the Princeton Scorekeeping Method (PRISM), 22a forerunner to ASHRAE’s Variable-based Degree Day(VBDD) calculation method. In commercial buildings,numerous methods have been reported over theyears 23,24,25 varying from weather normalization usingmonthly utility billing data ,26,27,28 daily and hourlymethods, 29 and even dynamic inverse models usingresistance-capacitance (RC) networks. 30 Procedures andmethodologies to baseline energy use in commercialbuildings began to appear in several publications inthe 1980s 31,32,33 and the early 1990s. 34,35,36 Modelingtoolkits and software have been developed that are usefulin developing performance metrics for buildings, aswell HVAC system components. These efforts includethe Princeton Scorekeeping Software (PRISM), 37 whichis useful for developing variable-based degree daymodels of monthly or daily data, ASHRAE’s HVAC01software for modeling primary HVAC systems such asboilers, and chillers, 38 ASHRAE’s HVAC02 software formodeling secondary HVAC systems including air-handlers,blowers, cooling coils and terminal boxes, 39 andASHRAE’s Research Projects 827-RP for in-situ measurementof chillers, pumps, and blowers, 40 Research Project1004-RP for in-situ measurement of thermal storagesystems, 41 Research Project 1050-RP Toolkit for CalculatingLinear, Change-point Linear and Multiple-LinearInverse Building Energy Analysis Models, and ResearchProject 1093-RP toolkit for calculating diversity factorsfor energy and cooling loads. 43In 1989, a report by Oak Ridge National Laboratory44 classified the diverse commercial building analysismethods into five categories, including: annual totalenergy and energy intensity comparisons, linear regressionand component models, multiple linear regression,building simulation, and dynamic (inverse) thermal performancemodels. In 1997 a reorganized and expandedversion of this classification appeared in the ASHRAEHandbook of Fundamentals, and is shown in Tables 27.1and 27.2. In Table 27.1 different methods of analyzingbuilding energy are presented, which have been clas-Table 27.1 ASHRAE’s 1997 Classification of Methods for the Thermal Analysis of Buildings. 57

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 709Table 27.2 ASHRAE’s 1997 Decision Diagram for Selection of Model. 58sified according to model type, including: forward,inverse, and hybrid models. 45In the first method, forward modeling, a thermodynamicmodel is created of a building using fundamentalengineering principles to predict the hypothetical energyuse of a building for 8,760 hours of the year given thelocation, and weather conditions. This requires a completedescription of the building, system, or componentof interest, as well as the physical description of thebuilding geometry, geographical location, system type,wall insulation value, etc. Forward models are normallyused to design and size HVAC systems, and have begunto be used to model existing building, using a techniquereferred to as calibrated simulation.In the second method, inverse modeling, an empiricalanalysis is conducted on the behavior of thebuilding as it relates to one or more driving forces orparameters. This approach is referred to as a systemidentification, parameter identification or inverse modeling.To develop an inverse model, one must assumea physical configuration of the building or system, andthen identify the parameter of interest using statisticalanalysis. 46 Two primary types of inverse models havebeen reported in the literature, including: steady stateinverse models and dynamic inverse models. A thirdcategory, hybrid models, consists of models that havecharacteristics of both forward and inverse models. 47The simplest steady-state inverse model regressesmonthly utility consumption data against average billingperiod temperatures. More robust methods includemultiple linear regression, change-point linear regression,and Variable-Based Degree Day regressions asindicated in Table 27.1. The advantage of steady-stateinverse models is that their use can be automated andapplied to large datasets where monthly utility billingdata and average daily temperatures for the billing periodsare available. Steady-state inverse models can alsobe applied to daily data which allows one to compensatefor differences in weekday and weekend use. 48 Unfortunately,steady state inverse models are insensitive todynamic effects (i.e., thermal mass), and other variables(for example humidity and solar gain), and are difficultto apply to certain building types, for example buildingsthat have strong on/off schedule dependent loads, orbuildings that display multiple change-points.Dynamic inverse models include: equivalentthermal network analysis, 49 ARMA models, 50,51 Fourierseries models, 52,53 machine learning, 54 and artificialneural networks. 55,56 Unlike steady-state, inverse models,dynamic models are capable of capturing dynamiceffects such as thermal mass which traditionally haverequired the solution of a set of differential equations.The disadvantages of dynamic inverse models are thatthey are increasingly complex, and need more detailedmeasurements to “tune” the model.Hybrid models are models that contain forwardand inverse properties. For example, when a traditionalfixed-schematic simulation program such as DOE-2 or

710 ENERGY MANAGEMENT HANDBOOKBLAST (or even a component-based simplified model)is used to simulate the energy use of an existing buildingthen one has a forward analysis method that isbeing used in an inverse application, i.e., the forwardsimulation model is being calibrated or fit to the actualenergy consumption data from a building in much thesame way that one fits a linear regression of energy useto temperature.Table 27.2 presents information that is useful forselecting an inverse model where usage of the model(diagnostics - D, energy savings calculations - ES, designDe, and control - C), degree of difficulty in understandingand applying the model, time scale for the data usedby the model (hourly - H, daily - D, monthly - M, andsub-hourly - S), calculation time, and input variablesused by the models (temperature - T, humidity - H, solar- S, wind - W, time - t, thermal mass - tm), and accuracyare used to determine the choice of a particular model.27.1.2.2.2 History of M&V Protocols in the United StatesThe history of measurement and verification protocolsin the United States can be traced to independentM&V efforts in different regions of the country as shownin Table 27.3, with states such as New Jersey, California,and Texas developing protocols that contained varyingprocedures for measuring the energy and demand savingsfrom retrofits to existing buildings. These effortsculminated in the development of the USDOE’s 1996North American Energy Measurement and VerificationProtocol (NEMVP), 59 which was accompanied bythe USDOE’s 1996 FEMP Guidelines, 60 both relying onanalysis methods developed in the Texas LoanSTARprogram. 61 In 1997 the NEMVP was updated and republishedas the International Performance Measurementand Verification Protocols (IPMVP). The IPMVP wasthen expanded in 2001 into two volumes: Volume I coveringEnergy and Water Savings, 63 and Volume II coveringIndoor Environmental Quality. 64 In 2003 Volume IIIof the IPMVP was published, which covers protocolsfor new construction. 65 Finally, in 2002 American Societyof Heating Refrigeration Air-conditioning Engineers(ASHRAE) released Guideline 14-2002: Measurement ofEnergy and Demand Savings, 66 which is intended toserve as the technical document for the IPMVP.27.1.3 Performance ContractsIn order to reduce costs and improve the HVACand lighting systems in its buildings, the U.S. federalgovernment has turned to the private energy efficiencysector to develop methods to finance and deliverenergy efficiency to the government. One of these arrangements,the performance contract, often includes aguarantee of performance, which benefits from accurate,reliable measurement and verification. In such a contractTable 27.3: History of M&V Protocols————————————————————————————————————————2003 - IPMVP - 2003 Volume III (new construction)2002 - ASHRAE Guideline 14 - 20022001 - IPMVP - 2001 Volume I & II (revised and expanded IPMVP)1998 - Texas State Performance Contracting Guidelines1997 - IPMVP (revised NEMVP)1996 - FEMP Guidelines1996 - NEMVP1995 - ASHRAE Handbook - Ch. 37 “Building Energy Monitoring”1994 - PG&E Power Saving Partner “Blue Book”1993 - NAESCO M&V Protocols1993 - New England AEE M&V Protocols1992 - California CPUC M&V Protocols1989 - Texas LoanSTAR Program1988 - New Jersey M&V Protocols1985 - First Utility Sponsored Large Scale Programs to Include M&V1985 - ORNLs “Field Data Acq. For Bld & Eqp Energy Use Monitoring”1983 - International Energy Agency “Guiding Principles for Measurement”1980s - USDOE funds the End - Use Load and Consumer Assessment Program (ELCAP)1980s - First Utility Sponsored Large Scale Programs to Include M&V1970s - First Validation of Simulations1960s - First Building Energy Simulations on Mainframe Computers————————————————————————————————————————

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 711all costs of the project (i.e., administration, measurementand verification, overhead and profit) are paid for bythe energy saved from the energy or water conservationprojects. In principle, this is a very attractive option forthe government, since it avoids paying the initial costsof the retrofits, which would have to come from shrinkingtaxpayer revenues. Instead, the costs are paid overa series of years because the government agrees to paythe Energy Service Company (ESCO) an annual fee thatequals the annual normalized costs savings of the retrofit(plus other charges). This allows the government tofinance the retrofits by paying a pre-determined annualutility bill over a series of years, which equals the utilitybill during the base year had the retrofit not occurred. Inreality, because the building has received an energy conservationretrofit, the actual utility bill is reduced, whichallows funding the annual fee to the ESCO without realizingany increase in the total annual utility costs (i.e.,utility costs plus the ESCO fee). Once the performancecontract is paid off, the total annual utility bills for thegovernment are reduced and the government receivesthe full savings amount of the retrofit.27.1.3.1 Definitions, Roles and ParticipantsThere are many different types of performancecontracts, which vary according to risk, and financing,including Guaranteed Savings, and Shared Savings. 67In a Guaranteed Savings performance contract a fixedpayment is established that repays the ESCO’s debtfinancing of the energy conservation retrofit, and anyfees associated with the project. In return, the ESCOguarantees that the energy savings will cover the fixedpayment to the ESCO. Hence, in a Guaranteed Savingscontract the ESCO is responsible for the majority of theproject risks. In a Shared Savings performance contractpayments to the ESCO are based on an agreed-upon portionof the estimated savings generated by the retrofit.In such contracts the M&V methods selected determinethe level of risk, and the responsibilities of the ESCO,and building owner. In both types of contracts the measurementand verification of the energy savings plays acrucial role in determining payment amounts.27.2 OVERVIEW OF MEASUREMENTAND VERIFICATION METHODSNationally recognized protocols for measurementand verification have evolved since the publication ofthe 1996 NEMVP as shown in Table 27.4. This evolutionreflects the consensus process that the Departmentof Energy has chosen as a basis for the protocols. Thisprocess was chosen to produce methods that all partiesagree can be used by the industry to determine savingsfrom performance contracts, varying in accuracyand cost from partial stipulation to complete measurement.In 1996 three M&V methods were included in theNEMVP: Option A: measured capacity with stipulatedconsumption; Option B: end-use retrofits, which utilizedmeasured capacity and measured consumption; andOption C: whole-facility or main meter measurements,which utilize before after regression models.In 1997, Options A, B and C were modified andrelabeled, and Option D: calibrated simulation wasadded. Also included in the 1997 IPMVP was a chapteron measuring the performance of new construction,which primarily utilized calibrated simulation, and adiscussion of the measurement of savings due to waterconservation efforts. In 2001 the IPMVP was publishedin two volumes: Volume I, which covers Options A, B,and C, which were redefined and relabeled from the 1997IPMVP, and Volume II, which covers indoor environmentalquality (IEQ), and includes five M&V approaches forIEQ, including: no IEQ M&V, M&V based on modeling,short-term measurements, long-term measurements, anda method based on occupant perceptions of IEQ. In 2003the IPMVP released Volume III, which contains four M&Vmethods: Option A: partially measured Energy ConservationMeasure (ECM) isolation, Option B: ECM isolation,Option C: whole-building comparisons, and Option D:whole-building calibrated simulation.In 2002 ASHRAE released Guideline 14-2002:Measurement of Energy and Demand Savings, whichis intended to serve as the technical document for theIPMVP. As the name implies, Guideline 14 containsapproaches for measuring energy and demand savingsfrom energy conservation retrofits to buildings. Thisincludes three methods: a retrofit isolation approach,which parallels Option B of the IPMVP, a whole-buildingapproach, which parallels Option C of the IPMVP,and a whole-building calibrated simulation approach,which parallels Option D of the 1997 and 2001 IPMVP.ASHRAE’s Guideline 14 does not explicitly contain anapproach that parallels Option A in the IPMVP, althoughseveral of the retrofit isolation approaches use partialmeasurement procedures, as will be discussed in a followingsection.27.2.1 M&V Methods: Existing BuildingsIn general, a common theme between the NEMVP,IPMVP and ASHRAE’s Guideline 14-2002, is that M&Vmethods for measuring energy and demand savings inexisting building are best represented by the followingthree approaches: retrofit isolation approach, a whole-

712 ENERGY MANAGEMENT HANDBOOKbuilding approach, and a whole-building calibratedsimulation approach. Similarly, the measurement of theperformance of new construction, renewables, and wateruse utilize one or more of these same methods.27.2.1.1 Retrofit Isolation ApproachThe retrofit isolation approach is best used whenend use capacity, demand or power can be measuredduring the baseline period, and after the retrofit forshort-term period(s) or continuously over the life of theproject. This approach can use continuous measurementof energy use both before and after the retrofit. Likewise,periodic, short-term measurements can be used duringthe baseline and after the retrofit to determine the retrofitsavings. Often such short-term measurements areaccompanied by periodic inspections of the equipmentto assure that the equipment is operating as specified.In most cases energy use is calculated by developingrepresentative models of the isolated component load(i.e., the kW or Btu/hr) and energy end-use (i.e., thekWh or Btu).27.2.1.1.1 Classifications of RetrofitsAccording to ASHRAE’s Guideline 14-2002 retrofitisolation approach, components or end-uses can be classifiedaccording to the following definitions: 68• Constant Load, Constant use. Constant load, constantuse systems consist of systems where theenergy used by the system is constant (i.e., variesby less than 5%) and the use of the system is constant(i.e., varies by less than 5%) through both thebaseline and post-retrofit period.• Constant Load, Variable use. Constant load, variableuse systems consist of systems where the energyused by the system is constant (i.e., varies byless than 5%) but the use of the system is variable(i.e., varies by more than 5%) through either thebaseline or post-retrofit period.• Variable Load, Constant use. Variable Load, constantuse systems consist of systems where theenergy used by the system is variable (i.e., variesby more than 5%) but the use of the system is constant(i.e., varies by less than 5%) through eitherthe baseline or post-retrofit period.• Variable Load, Variable use. Variable Load, variableuse systems consist of systems where theenergy used by the system is variable (i.e., variesby more than 5%) and the use of the system is variable(i.e., varies by more than 5%) through eitherthe baseline or post-retrofit period.Use of these classifications then allows for asimplified decision table (Table 27.5) to be used indetermining which type of retrofit-isolation procedureto use. For example, in the first row (i.e., a CL/TS-preretrofitto CL/TS-post-retrofit), if a constant load witha known or timed schedule is replaced with a newdevice that has a reduced constant load, and a knownor constant schedule, then the pre-retrofit and postretrofitmetering can be performed with one-time loadmeasurement(s). Contrast this with the last row (i.e., aVL/VS-pre-retrofit to VL/VS-post-retrofit), if a variableload, with a timed or variable schedule is replaced withTable 27.4: Evolution of M&V Protocols in the United States.

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 713a new device that has a reduced variable load, and avariable schedule, then the pre-retrofit and post-retrofitmetering should use continuous or short-term measurementthat are sufficient in length to allow for thecharacterization of the performance of the componentto be accomplished with a model (e.g., regression, orengineering model).27.2.1.1.2 Detailed Retrofit IsolationMeasurement and Verification ProceduresAppendix E of ASHRAE’s Guideline 14-2002contains detailed retrofit isolation procedures for themeasurement and verification of savings, including:pumps, fans, chillers, boilers and furnaces, lighting,and large and unitary HVAC systems. In general, theprocedures were drawn from the previous literature,including ASHRAE’s Research Project 827-RP 70 (i.e.,pumps, fans, chillers), various published proceduresfor boilers and furnaces, 71,72,73,74,75,76,77 lightingprocedures, and calibrated HVAC calibration simulations.78,79 A review of these procedures, which varyfrom simple one-time measurements, to complex,calibrated air-side psychrometric models, is describedit the following sections.Table 27.5: Metering Requirements to Calculate Energy andDemand Savings from the ASHRAE Guideline 14-2002. 69

714 ENERGY MANAGEMENT HANDBOOKA. PumpsMost large HVAC systems utilize electric pumpsfor moving heating/cooling water from the building’sprimary systems (i.e., boiler or chiller) to the building’ssecondary systems (i.e., air-handling units, radiators,etc.) where it can condition the building’s interior. Suchpumping systems use different types of pumps, varyingcontrol strategies, and piping layouts. Therefore, thecharacterization of pumping electric power depends onthe system design and control method used. Pumpingsystems can be characterized by the three categoriesshown in Table 27.6. 80 Table 27.7 shows the six pumptesting methods, including the required measurements,applications and procedures steps.ASHRAE’s Research Project 827-RP 81 developedsix in-situ methods for measuring the performance ofpumps of varying types and controls. To select a methodthe user needs to determine the pump system typeand control, and the desired level of uncertainty, cost,and degree of intrusion. The user also needs to recordTable 27.6: Applicability of Test Methods toCommon Pumping Systems from the ASHRAE Guideline 14-2002. 82Table 27.7: Pump Testing Methods from ASHRAE Guideline 14-2002. 83

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 715Table 27.7 (Continued)the pump and motor data (i.e., manufacturer, modeland serial number), fluid characteristics and operatingconditions. The first two methods (i.e., single-point andsingle-point with a manufacturer’s curve) involve testingat a single operating point. The third and fourthprocedures involve testing at multiple operating pointsunder imposed system loading. The fifth method alsoinvolves multiple operating points, in this case obtainedthrough short term monitoring of the system withoutimposed loading. The sixth procedure operates thepump with the fluid flow path completely blocked.While the sixth procedure is not useful for generating apower versus load relationship, it can be used to confirmmanufacturer’s data or to identify pump impeller diameter.A summary of the methods is provided below. Additionaldetails can be found by consulting ASHRAE’sGuideline 14-2002.A-1. Constant Speed and Constant Volume PumpsConstant volume pumping systems use three wayvalves and bypass loops at the end-use or at the pump.As the load varies in the system, pump pressure andflow are held relatively constant, and the pump inputpower remains nearly constant. Because pump motorspeed is constant, constant volume pumping systemshave a single operating point. Therefore, measuring thepower use at the operating point (i.e., a single pointmeasurement) and the total operating hours are enoughto determine annual energy use.A-2. Constant Speed and Variable Volume PumpsVariable pumping systems with constant speedpumps use two-way control valves to modulate flow tothe end-use as required. In constant speed variable volumepumping systems, the flow varies along the pumpcurve as the system pressure drop changes in response tothe load. In some cases, a bypass valve may be modulatedif system differential pressure becomes too large. Suchsystems have a single possible operating point for anygiven flow, as determined by the pump curve at that flow

716 ENERGY MANAGEMENT HANDBOOKrate. In such systems the second and third testing methodscan be used to characterize the pumps energy use atvarying conditions. In the second procedure, measurementsof in-situ power use is performed at one flow rateand manufacturer’s data on the pump, motor, and drivesystem are used to create a part load power use curve. Inthe third testing method in-situ measurements are madeof the electricity use of the pump with varying loads imposedon the system using existing control, discharge, orbalancing valves. The fourth and fifth methods can alsobe used to characterize the pump electricity use. Usingone of these methods the part load power use curve anda representative flow load frequency distribution are usedto determine annual energy use.A-3. Variable Speed and Variable Volume PumpsLike the constant speed variable volume system,flow to the zone loads is typically modulated usingtwo-way control valves. However, in variable speedvariable volume pumping systems, a static pressurecontroller is used to adjust pump speed to match theflow load requirements. In such systems the operatingpoint cannot be determined solely from the pump curveand flow load because a given flow can be provided atvarious pressures or speeds. Furthermore, the systemdesign and control strategy place constraints on eitherthe pressure or flow. Such systems have a range of systemcurves which call for the same flow rate, dependingon the pumping load. 827-RP provides two options(i.e., multiple point with imposed loads and short termmonitoring) for accurately determining the in-situ partload power use. In both cases, the characteristics of thein-situ test include the pump and piping system (piping,valves, and controllers), therefore the control strategyis included within the data set. In the fourth method(i.e., multiple point with imposed load at the zone), thepump power use is measured at a range of imposedloads. These imposed loads are done at the zone levelto account for the in-situ control strategy and systemdesign. In the fifth method (i.e., multiple point throughshort term monitoring), the pump system is monitoredas the building experiences a range of thermal loads,with no artificial imposition of loads. If the monitoredloads reflect the full range of loads, then an accurate partload power curve can be developed that represents thefull range of annual load characteristics. For methods #4and #5, the measured part load power use curves andflow load frequency distribution are used to determineannual energy use.A-4. Calculation of Annual Energy UseOnce the pump performance has been measuredthe annual energy use can be calculated using the followingprocedures, depending upon whether the systemis a constant volume or variable volume pumping system.Savings are then calculated by comparing the annualenergy use of the baseline with the annual energyuse of the post-retrofit period.Constant Volume Constant Speed Pumping Systems.In a constant volume constant speed pumping systemsthe volume of the water moving through the pump isalmost constant, and therefore the power load of thepump is virtually constant. The annual energy calculationis therefore a constant times the frequency of theoperating hours of the pump.E annual = T * Pwhere:T = annual operating hoursP = equipment power inputVariable Volume Pumping Systems. For variablevolume pumping systems the volume of water movingthrough the pump varies over time, hence the powerdemand of the pump and motor varies. The annualenergy use then becomes a frequency distribution of theload times the power associated with each of the binsof operating hours. In-situ testing is used to determinethe power associated with the part load power use.where:iE annual =ΣiT i * P i= bin index, as defined by the load frequencydistributionT i = number of hours in bin iP i = equipment power use at load bin IB. FansMost large HVAC systems utilize fans or airhandlingunits to deliver heating and cooling to thebuilding’s interior. Such air-handling systems use differenttypes of fans, varying control strategies, andduct layouts. Therefore, the characterization of fan electricpower depends on the system design and controlmethod used. Fan systems can be characterized by thethree categories shown in Table 27.8. 84In a similar fashion as pumping systems, ASHRAE’sResearch Project 827-RP developed five in-situ methodsfor measuring the performance of fans of varying typesand controls. To select a method the user needs to determinethe system type and control, and the desired

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 717level of uncertainty, cost, and degree of intrusion. Theuser also needs to record the fan and motor data (i.e.,manufacturer, model and serial number), as well as theoperating conditions (i.e., temperature, pressure andhumidity of the air stream). The first two methods (i.e.,single-point and single-point with a manufacturer’scurve) involve testing at a single operating point. Thethird and fourth procedures involve testing at multipleoperating points under imposed system loading. Thefifth method also involves multiple operating points,in this case obtained through short term monitoring ofthe system without imposed loading. Additional detailsabout fan testing procedures can be found by consultingASHRAE’s Guideline 14-2002.B-1. Calculation of Annual Energy UseOnce the fan performance has been measured theannual energy use can be calculated using the followingprocedures, depending upon whether the system isa constant volume or variable volume system. Savingsare then calculated by comparing the annual energy useof the baseline with the annual energy use of the postretrofitperiod.Constant Volume Fan Systems. In a constant volumesystem the volume of the air moving across the fan isalmost constant, and therefore the power load of the fanis virtually constant. The annual energy calculation istherefore a constant times the frequency of the operatinghours of the fan.E annual = T * Pwhere:T = annual operating hoursP = equipment power inputVariable Volume Systems. For variable volume systemsthe volume of the air being moved by the fanvaries over time, hence the power demand of the fanand motor varies. The annual energy use then becomesa frequency distribution of the load times the power associatedwith each of the bins of operating hours. In-situtesting is used to determine the power associated withthe part load power use.Table 27.8: Applicability of Test Methods to Common Fan Systems from the ASHRAE Guideline 14-2002. 85Table 27.9: Fan Testing Methods from ASHRAE Guideline 14-2002. 86(Continued)

718 ENERGY MANAGEMENT HANDBOOKTable 27.9: (Continued)E annual =ΣiT i * P iwhere:i = bin index, as defined by the load frequencydistributionT i = number of hours in bin iP i = equipment power use at load bin IC. ChillersIn a similar fashions as pumps and fans, in-situchiller performance measurements have been alsobeen developed as part of ASHRAE Research Project827-RP. These models provide useful performancetesting methods to evaluate annual energy use andpeak demand characteristics for installed watercooledchillers and selected air-cooled chillers. Theseprocedures require short-term testing of the part loadperformance of an installed chiller system over arange of building thermal loads and coincident ambientconditions. The test methods determine chillerpower use at varying thermal loads using thermodynamicmodels or statistical models with inputs fromdirect measurements, or manufacturer’s data. Withthese models annual energy use can be determinedusing the resultant part load power use curve with aload frequency distribution. Such models are capableof calculating the chiller power use as a function ofthe building thermal load, evaporator and condenserflow rates, entering and leaving chilled water temperatures,entering condenser water temperatures,and internal chiller controls. ASHRAE’s Guideline14-2002 describes two models for calculating the

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 719power input of a chiller, including simple and temperaturedependent thermodynamic models. 87,88,89 Athird method, which uses a tri-quadratic regressionmodel such as those found in the DOE-2 simulationprogram, 90,91,92,93,94 also provides acceptable performancemodels, provided that measurements are madeover the full operating range.C-1. Simple Thermodynamic ModelBoth the simple thermodynamic model andthe temperature-dependent thermodynamic modelexpress chiller efficiency as 1/COP because it hasa linear relationship with 1/(evaporator load). Thesimpler version of the chiller model developed predictsa linear relationship between 1/COP and 1/Q evap , which is independent of the evaporator supplytemperature or condenser temperature returningto the chiller. The full form of simple thermodynamicmodel is shown in the equation below.1COP =–1+ T cwRT/T chwST+ 1Q evapq evap T cwRTT chwST– q cond + f HXwhere:T cwRT = Entering (return) condenser water temperature(Kelvin)T chwST = Leaving (supply) evaporator watertemperature (Kelvin)Q evap = Evaporator loadq evap = rate of internal losses in evaporatorq cond = rate of internal losses in condenserf HX = dimensionless term. 95This equation reduces to a simple form that allowsfor the determination of two coefficients usinglinear regression, which is shown in the followingequation.1COP =–C 11Q evap+C 0In this simplified form the coefficient c 1 characterizesthe internal chiller losses, while the coefficientc 0 combines the other terms of the simple model. TheCOP figure of merit can be converted into conventionalefficiency measures of COP or kW per ton usingthe following relationships:Coefficient of Performance (COP):COP =kW refrigeration effectkW inputEnergy Efficiency Ratio (EER):EER =Btu/hr refrigeration effectWatts inputPower per Ton (kW/ton):kW/ton == 3.412 COPkW inputtons refrigeration effect =12/EERThe simple thermodynamic model can be determinedwith relatively few measurements of the chillerload (evaporator flow rate, entering and leaving chilledwater temperatures) and coincident RMS power use.Unfortunately, variations in the chilled water supply(i.e., the temperature of the chilled water leaving theevaporator) and the condenser water return temperatureare not considered. Hence, this model is best usedwith chiller systems that maintain constant temperaturecontrol of evaporator and condenser temperatures. Insystems with varying temperatures a temperature-dependentthermodynamic model, or tri-quadratic modelyield a more accurate performance prediction.C-2. Temperature-dependent Thermodynamic ModelThe temperature dependent thermodynamic modelincludes the losses in the heat exchangers of the evaporatorand condenser, which are expressed as a functionof the chilled water supply and condenser water returntemperatures. The resulting expression uses three coefficients(A 0 , A 1 , A 2 ), which are found with linear regression,as shown in the equation that follows.1COP =–1+ T cwRT/T chwST+ –A 0 +A 1 T cwRT –A 2 T cwRT /T chwSTQ evapUse of this temperature dependent thermodynamicmodel requires the measurement of the chiller load(i.e., evaporator flow rate, entering and leaving chilledwater temperatures), coincident RMS power use, andcondenser water return temperature. Since this modelis sensitive to varying temperatures it is applicable to awider range of chiller systems.

720 ENERGY MANAGEMENT HANDBOOKTo use the temperature dependent model, measuredchiller thermal load, coincident RMS power use,chilled water supply temperature, and condenser waterreturn temperatures are used to calculate the three coefficients(A 0 , A 1 , A 2 ).To determine A 2 the following equation is plottedagainst T cwRT /T chwST (Kelvin), with value of A 2 beingdetermined from the regression lines, which shouldresemble a series of straight parallel lines, one for eachcondenser temperature setting.α = 1COP +1– T cwRT/T chwSTQ evapThe coefficients A 0 and A 1 are determined byplotting β from the next equation, using the alreadydetermined value of A 2 , versus the condenser waterreturn temperature T cwRT (Kelvin). This should resultin a group of data points forming a single straight line.The slope of the regression line determines the value ofcoefficient A 1 while the intercept determines the valueof coefficient A 0 .β = 1COP +1– T cwRT/T chwST+A 2 T cwRT /T chwSTQ evapAfter A 0 , A 1 , and A 2 have been determined usingα and β and from the equations above, the 1/COP canbe calculated and used to determine the chiller performanceover a wide range of measured input parametersof chiller load, chilled water supply temperature, andcondenser water return temperature.C-3. Quadratic Chiller ModelsChiller performance models can also be calculatedwith quadratic models, which can include models thatexpress the chiller power use as a function of the chillerload (quadratic), as a function of the chiller load andchilled water supply temperature (bi-quadratic), or as afunction of the chiller load, evaporator supply temperatureand condenser return temperature (tri-quadratic).Such models use the quadratic functional form used inthe DOE-2 energy simulation program to model partloadequipment and plant performance characteristics.Two examples of quadratic models are shown below,one for a monitoring project where chiller electricity use,chilled water production, chilled water supply temperature,and condenser water temperature returning to thechiller were available, which uses a tri-quadratic modelas follows:kW/ton = a + b x Tons + c x T cond + d x T evap + e x Tons 2+ f x T cond2+ g x T evap2 + h x Tons x T cond+ I x T evap x Tons+ j x T cond x T evap + k x Tons x T cond x T evapIn a second example, chiller electricity use ismodeled with a bi-quadratic model that includes onlythe chilled water production, and chilled water supplytemperature, which reduces to the following form. Eithermodel can easily be calculated from field data in aspreadsheet using multiple linearized regression.kW/ton = a + b x Tons + c x Tk evap + d x Tons 2+ e x T evap2 + f x T evap x TonsC-4. Example: Quadratic Chiller ModelsAn example of a quadratic chiller performanceanalysis model is provided from hourly measurementsthat were taken at to determine the baseline model ofa cooling plant at an Army base in Texas. Figure 27.1shows the time series data that were recorded duringJune and August of 2002. The upper trace is the chillerthermal load (tons), and the lower trace is the ambienttemperature during this period. Figure 27.2 showsa time series plot of the recorded temperatures of thecondenser water returning to the chiller (upper trace),and the chilled water supply temperatures (lower trace).In Figures 27.3 and 27.4 the performance of the chilleris shown as the chiller efficiency (i.e., kW/ton) versusthe chiller load (tons). In Figure 27.3 linear (R 2 = 34.3%)and quadratic (R 2 = 53.4%) models of the chiller are superimposedover the measured data from the chiller toillustrate how a quadratic model fits the chiller data. InFigure 27.4 a tri-quadratic model (R 2 = 83.7%) is shownsuperimposed over the measured data.A quick inspection of the R 2 goodness-of-fit indicatorsfor the linear, quadratic and tri-quadratic modelsbegins to shed some light on how well the models arefitting the data. However, one must also inspect howwell the model is predicting the chiller performance atthe intended operation points. For example, althougha linear model has an inferior R 2 when compared toa quadratic model, for this particular chiller, it givessimilar performance values for cooling loads rangingfrom 200 to 450 tons. Choosing the quadratic modelimproves the prediction of the chiller performance forvalues below 200 tons. However, it significantly underpredicts the kW/ton at 350 tons and over-predicts thekW/ton at values over 500 tons. Hence, both modelsshould be used with caution.

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 721The tri-quadratic model has an improved R 2 of83.7% and does not seem to contain any ranges wherethe model’s bias is significant from the measured data(excluding the few stray points which are caused bytransient data). Therefore, in the case of this chiller, theadditional effort to gather and analyze the chiller loadagainst the chilled water supply temperature and condenserwater return temperature is well justified.C-5. Calculation of Annual Energy UseOnce the chiller performance has been determinedthe annual energy use can be calculated using the simpleor temperature dependent models to determine thepower demand of the chiller at each bin of the coolingload distribution. For chillers with varying temperatures,a load frequency distribution, which contains thetwo water temperatures, provides the operating hoursof the chiller at each bin level. The energy use E i , andpower level P i are given by the equations below. Thetotal annual energy use is then the sum of the product ofthe number of hours in each bin times the chiller powerassociated with that bin. Savings are then calculated bycomparing the annual energy use of the baseline withthe annual energy use of the post-retrofit period.where:E i = T i * P iPi = (1/Eff i ) (Q evap,i )E annual =ΣiT i * P ii = bin index, as defined by loadfrequency distributionFigure 27.1: Example chiller analysis. Time series plotof chiller load (upper trace, tons) and ambient temperature(lower trace, degrees F).Figure 27.3: Example chiller analysis. Chiller performanceplot of chiller efficiency (kW/ton) versus thechiller cooling load. Comparisons of linear (R 2 =34.3%) and quadratic (R 2 = 53.4.3%) chiller models areshown.Figure 27.2: Example chiller analysis. Time series plotof condenser water return temperature (upper trace,degrees F) and chilled water supply temperature(lower trace, degrees F).Figure 27.4: Example chiller analysis. Chiller performanceplot of chiller efficiency (kW/ton) versus thechiller cooling load. In this figure a tri-quadratic chillermodel (R 2 = 83.7%) is shown.

722 ENERGY MANAGEMENT HANDBOOKT i = number of hours in bin iPi = equipment power use at load bin iEff i = chiller 1/COP in bin iQ evap,i = chiller load in bin iD. Boilers and FurnacesIn-situ boiler and furnace performance measurements,for non-reheat boilers and furnaces, are listed inAppendix E of ASHRAE Guideline 14-2002. These procedures,which were obtained from the previously notedpublished literature on performance measurements ofboilers and furnaces, 96,97,98,99,100,101,102 are groupedinto four methods (i.e., single-point, single-point withmanufacturer’s data, multiple point with imposed loads,and multiple point tests using short term monitoring)that use three measurement techniques (i.e., directmethod, direct heat loss method, and indirect combustionmethod), for a total of twelve methods.The choice of method depends on boiler type (i.e.,constant fire boiler, or variable-fire boilers), and availabilityof measurements (i.e., fuel meters, steam meters,etc.). For constant fire boilers, the boiler load is virtuallyconstant. Therefore, a single measurement or series ofmeasurements a full load will characterize the boiler orfurnace efficiency at a given set of ambient conditions.For variable fire boilers the fuel use and output of theboiler varies. Therefore, the efficiency of the boiler willvary depending upon the load of the boiler as describedby the manufacturer’s efficiency curve. Figure 27.5shows an example of the measured performance of avariable-fire, low pressure steam boiler installed at anarmy base in Texas. 103D-1. Boiler Efficiency MeasurementsThere are three principal methods for determiningboiler efficiency, the direct method (i.e., Input-Outputmethod), the direct heat loss method, also known as theindirect method, and the indirect combustion efficiencymethod. The first two are recognized by the AmericanSociety of Mechanical Engineers (ASME) and aremathematically equivalent. They give identical resultsif all the heat balance factors are considered and theboiler measurements performed without error. ASMEhas formed committees from members of the industryand developed the performance test codes 105 that detailprocedures of determining boiler efficiency by the firsttwo methods mentioned above. The accuracy of boilerperformance calculations is dependent on the quantitiesmeasured and the method used to determine the efficiency.In the direct efficiency method, these quantitiesare directly related to the overall efficiency. For example,if the measured boiler efficiency is 80%, then an errorof 1% in one of the quantities measured will result in a0.8% error in the efficiency. Conversely, in the direct heatloss method, the measured parameters are related to theboiler losses. Therefore, for the same boiler which hadan efficiency of 80%, a measurement error of 1% in anyquantity affects the overall efficiency by only 0.2% (i.e.,1% of the measured losses of 20%). As a result, the directheat loss method is inherently more accurate than thedirect method for boilers. However, the direct heat lossmethod requires more measurement and calculations. Ingeneral, boiler efficiencies range from 75% to 95% forutility boilers, and for industrial and commercial boilers,the average efficiency ranges from 76% to 83% on gas,78% to 89% on oil and 85% to 88% for coal. 106,107Figure 27.5: Example Boiler PerformanceCurve from Short TermMonitoring. 104

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 723where• Direct Method= 100 – L df – L fh – L am – L rad – L conv – L bdOn the right side of the equation the loss terms– L inc – L unacct are usually small for well insulated boilers. TheseThe direct method (i.e., the input-output method) Q loss = heat losses (Btu/hr),is the simplest method to determine boiler efficiency.L df = dry flue gas heat loss (%),In this method, the heat supplied to the L fh = fuel hydrogen heat loss (%),boiler and the heat absorbed by the water in the L am = combustion air moisture heat loss (%),boiler in a given time period are directly measured. L rad = radiation heat loss (%),Using the direct method, the efficiency of a nonreheatL conv = convection heat loss (%),boiler is given by: 108L inc = incombusted fuel loss (%),η b = Q a× 100Q iL bd = blowdown heat loss (%),L unacct = unaccounted for heat losses (%).Using this method the flue gas losses (sensiblewhereand latent heat) due to radiation and convection,Q a = heat absorbed (Btu/hr) = Σm o h o – Σm i h i incomplete combustion, and blowdown are accountedfor. In most boilers the flue gas loss ism o h o = mass flow-enthalpy products of workingfluid streams leaving boiler envelope,the largest loss, which can be determined by aincluding main steam, blowdown, sootflue gas analysis. Flue gas losses vary with flueblowing steam, etc.gas exit temperature, fuel composition and typem i h i = mass flow-enthalpy products of workingof firing. 110 Radiation and convection loss canfluid streams entering boiler envelope, includingfeedwater, desuperheating sprays,be obtained from the standard curves. 111 Unaccountedfor heat losses can also be obtained frometc.published industry sources, 112 which cite losses ofQ i = heat inputs (Btu/hr) = V fuel x HHV+Q c 1.5% for solid fuels, and 1% for gaseous or liquidV fuel = volumetric flow of fuel into boiler (SCF/fuel boilers. Losses from boiler blowdown shouldhr)also be measured. Typical values can be found inHHV = fuel higher heating value (Btu/SCF), andvarious sources. 113,114Q c = heat credits (Btu/hr). Heat credits are definedas the heat added to the envelope ofthe steam generating unit other than thechemical heat in the fuel “as fired.” Thesecredits include quantities such as sensibleheat in the fuel, the entering air and the atomizingsteam. Other credits include heat• Indirect Combustion MethodThe indirect combustion method can also be usedto measure boiler efficiency. The combustion efficiencyis the measure of the fraction of fuel-airenergy available during the combustion process,calculated from the folowing: 115,116from power conversion in the pulverizeror crusher, circulating pump, primary airfan and recirculating gas fan.η c = h p – h f + h aQ i× 100• Direct Heat Loss MethodIn the direct heat loss method the boiler efficiencywhereequals 100% minus the boiler losses. The directη c = combustion efficiency (%)heat loss method tends to be more accurate thanh p = enthalpy of products (Btu/lb)the direct method because the direct heat lossh f = enthalpy of fuel (Btu/lb)method focuses on determining the heat lost fromh a = enthalpy of combustion air, Btu/lbthe boiler, rather than on the heat absorbed by theQ i = heat inputs (Btu/hr) = Vfuel * HHV+Qcworking fluid. The direct heat loss method determinesefficiency using the following :109Indirect combustion efficiency can be relatedto direct efficiency or direct heat loss efficiencyη b = Q a× 100 = Q i – Q lossmeasurements using the following: 117,118× 100Q i ?η b = η c – L rad – L conv – L unacct

724 ENERGY MANAGEMENT HANDBOOKterms must be accounted for when boilers arepoorly insulated or operated poorly (i.e., excessiveblowdown control, etc.).Table 27-10 provides a summary of the performancemeasurement methods (i.e., single-point,single-point with manufacturer’s data, multiplepoint with imposed loads, and multiple pointtests using short term monitoring), which usethree efficiency measurement techniques (i.e., directmethod, direct heat loss method, and indirectcombustion method) that are listed in Appendix Eof ASHRAE Guideline 14-2002. For each methodthe pertinent measurements are listed along withthe steps that should be taken to calculate the efficiencyof the boiler or furnace being measured.D-2. Calculation of Annual Energy UseOnce the boiler performance has been measuredthe annual energy use can be calculated using the followingprocedures, depending upon whether the systemis a constant fire boiler or variable fire boiler. Savingsare then calculated by comparing the annual energy useof the baseline with the annual energy use of the postretrofitperiod.• Constant Fire BoilersIn constant fire boilers the method assumes the loadand fuel use are constant when the boiler is operating.Therefore, the annual fuel input is simply thefull-load operating hours of the boiler times the fuelinput. The total annual energy use is given by:E annual = T * Pwhere: T = annual operating hours under full loadP = equipment power use• Variable Fire BoilersFor variable fire boilers the output of the boilerand fuel input vary according to load. Hence afrequency distribution of the load is needed thatprovides the operating hours of the boiler at eachbin level. In-situ testing is then used to determinethe efficiency of the boiler or furnace for each bin.The total annual energy use for variable fire boilersis given by:E annual =ΣiT i * P iwhere:i = bin index, as defined by load variablefrequency distributionT i = number of hours in bin iP i = equipment fuel input (& efficiency) atload bin (i)E. LightingOne of the most common retrofits to commercialbuildings is to replace inefficient T-12 fluorescents andmagnetic ballasts with T-8 fluorescents and electronicballasts. This type of retrofit saves electricity associatedwith the use of the more efficient lighting, and dependingon system type, can reduce cooling energy usebecause of reduced internal loads from the removal ofthe inefficient lighting. In certain climates, depending onsystem type, this can also mean an increase in heatingloads, which are required to offset the heat from the inefficientlighting. Previously published studies show thecooling interaction can increase savings by 10 to 20%.The increased heating requirements can reduce savingsby 5 to 20%. 120 Therefore, where the costs can be justified,accurate measurement of total energy savings caninvolve before/after measurements of the lighting loads,cooling loads, and heating loads.E-1. Lighting MethodsASHRAE Guideline 14-2002 (see Table 27.11) providessix measurement methods to account for the electricityand thermal savings, varying from methods thatutilize sampled before/after measurements to methodsTable 27.10: Boiler and Furnace Performance Testing Methods from ASHRAE Guideline 14-2002. 119

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 725Table 27.10 (Continued)

726 ENERGY MANAGEMENT HANDBOOKTable 27.10 (Continued)

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 727Table 27.10 (Continued)that use sub-metered before-after lighting measurementwith measurements of increases or decreases tothe heating and cooling systems from the removal ofthe internal lighting load. In general, the calculationof savings from lighting retrofits involves ascertainingthe wattage or power reduction associated with thenew fixtures, which is then multiplied times the hoursper day (i.e., lighting usage profiles) that the lightsare used. The lighting usage profiles can be calculatedbased on appropriate estimates of use, measured at theelectrical distribution panel, or sampled with lightingloggers. Figure 27.6 shows an example of weekdayweekendprofiles calculated with ASHRAE’s DiversityFactor Toolkit. 121Some lighting retrofits involve the installation ofdaylighting sensors to dim fixtures near the perimeterof the building or below skylights when lighting levelscan be maintained with daylighting, thus reducing theelectricity used for supplemental lighting. Measuringthe savings from such daylighting retrofits usuallyinvolves before-after measurements of electrical powerand lighting usage profiles.Any lighting retrofit should include an assessmentof the existing lighting levels, which is measuredduring daytime and nighttime conditions. All lightingretrofits should achieve and maintain lighting levelsrecommended by the Illuminating Engineering Societyof North America (IESNA). 122 Any pre-retrofit lightinglevels not maintaining IESNA lighting levels shouldbe brought to the attention of the building owner oradministrator. In the following section the six methods,which are described in the ASHRAE Guideline 14-2002 are summarized. Table 27.11 contains the lightingperformance measurement methods from ASHRAE’sGuideline 14-2002.• Method #1: Baseline and post-retrofit measuredlighting power levels and stipulated diversity profiles.In Method #1 before-after lighting power levelsfor a representative sample of lighting fixturesare measured using a Wattmeter, yielding an averageWatt/fixture measurement for the pre-retrofitfixtures and post-retrofit fixtures. Lighting usageprofiles are estimated or stipulated using the bestavailable information, which represents the lightingusage profiles for the fixtures. This methodworks best for exterior lighting fixtures or lightingfixtures controlled by a timer or photocell. Lightingfixtures located in hallways, or any interiorlighting fixtures that is operated 24 hours per day,7 days per week or controlled by a timer is alsosuitable for this method. Savings benefits or penaltiesfrom thermal interactions are not included inthis method.Figure 27.6: Example Weekday-Weekend LightingProfiles.

728 ENERGY MANAGEMENT HANDBOOK• Method #2: Baseline and post-retrofit measuredlighting power levels and sampled baseline andpost-retrofit diversity profiles.In Method #2 before-after lighting power levelsfor a representative sample of lighting fixtures aremeasured using a Wattmeter, yielding an averageWatt/fixture measurement for the pre-retrofitfixtures and post-retrofit fixtures. Lighting usageprofiles are measured with portable lighting loggers,or portable current meters attached to lighting circuitsto determine the lighting usage profiles for thefixtures. This method is appropriate for any interioror exterior lighting circuit that has predictable usageprofiles. Savings benefits or penalties from thermalinteractions are not included in this method.• Method #3: Baseline measured lighting powerlevels with baseline sampled diversity profilesand post-retrofit power levels with post-retrofitcontinuous diversity profile measurements.In Method #3 pre-retrofit lighting power levelsfor a representative sample of lighting fixtures aremeasured using a Wattmeter, yielding an averageWatt/fixture measurement for the pre-retrofitfixtures. Pre-retrofit lighting usage profiles aremeasured with portable lighting loggers, or portablecurrent meters attached to lighting circuitsto determine the lighting usage profiles for thefixtures. Post-retrofit lighting usage is measuredcontinuously using either sub-metered lightingelectricity measurements, or post-retrofit lightingpower levels for a representative sample of lightingfixtures times a continuously measured diversityprofile (i.e., using lighting loggers or currentmeasurements on lighting circuits). This methodis appropriate for any interior or exterior lightingcircuit that has predictable usage profiles. Savingsbenefits or penalties from thermal interactions arenot included in this method.• Method #4: Baseline measured lighting powerlevels with baseline sampled diversity profiles andpost-retrofit continuous sub-metered lighting.In Method #4 pre-retrofit lighting power levelsfor a representative sample of lighting fixtures aremeasured using a Wattmeter, yielding an averageWatt/fixture measurement for the pre-retrofitfixtures. Pre-retrofit lighting usage profiles aremeasured with portable lighting loggers, or portablecurrent meters attached to lighting circuitsto determine the lighting usage profiles for thefixtures. Post-retrofit lighting usage is measuredcontinuously using sub-metered lighting electricitymeasurements. This method is appropriate for anyinterior or exterior lighting circuit that has predictableusage profiles. Savings benefits or penaltiesfrom thermal interactions are not included in thismethod.• Method #5: Includes methods #1, #2, or #3 withmeasured thermal effect (heating & cooling).In Method #5 pre-retrofit and post-retrofitlighting electricity use is measured with Methods#1, #2, #3 or #4, and the thermal effect is measuredusing the component isolation method for the coolingor heating system.This method is appropriate for any interiorlighting circuit that has predictable usage profiles.Savings benefits or penalties from thermal interactionsare included in this method.• Method #6: Baseline and post-retrofit sub-meteredlighting measurements and thermal measurements.In Method #6 pre-retrofit and post-retrofitlighting electricity use is measured continuouslyusing sub-metering, and the thermal effect is measuredusing whole-building cooling and heatingsub-metered measurements. This method is appropriatefor any interior lighting circuit. Savingsbenefits or penalties from thermal interactions areincluded in this method.E-2. Calculation of Annual Energy UseThe calculation of annual energy use varies accordingto lighting calculation method as shown in TableTable 27.11: Lighting Performance Measurement Methods from ASHRAE Guideline 14-2002. 123


730 ENERGY MANAGEMENT HANDBOOK27.12. The savings are determined by comparing theannual lighting energy use during the baseline period tothe annual lighting energy use during the post-retrofitperiod. In Methods #5 and #6 the thermal energy effectcan either be calculated using the component efficiencymethods or it can be measured using whole-building,before-after cooling and heating measurements. Electricdemand savings can be calculated using Methods #5 and#6 using diversity factor profiles from the pre-retrofitperiod and continuous measurement in the post-retrofitperiod. Peak electric demand reductions attributable toreduced chiller loads can be calculated using the componentefficiency tests for the chillers. Savings are thencalculated by comparing the annual energy use of thebaseline with the annual energy use of the post-retrofitperiod.F. HVAC SystemsAs mentioned previously, during the 1950s and1960s most engineering calculations were performedusing slide rules, engineering tables and desktop calculatorsthat could only add, subtract, multiply anddivide. In the 1960s efforts were initiated to formulateand codify equations that could predict dynamic heatingand cooling loads, including efforts to simulate HVACsystems. In 1965 ASHRAE recognized that there was aneed to develop public-domain procedures for calculatingthe energy use of HVAC equipment and formed thePresidential Committee on Energy Consumption, whichbecame the Task Group on Energy Requirements (TGER)for Heating and Cooling in 1969. 125 TGER commissionedtwo reports that detailed the public domain proceduresfor calculating the dynamic heat transfer through thebuilding envelopes, 126 and procedures for simulatingthe performance and energy use of HVAC systems. 127These procedures became the basis for today’s publicdomainbuilding energy simulation programs such asBLAST, DOE-2, and EnergyPlus. 128,129In addition, ASHRAE has produced several additionalefforts to assist with the analysis of buildingenergy use, including a modified bin method, 130 theHVAC-01 131 and HVAC-02 132 toolkits, and HVACsimulation accuracy tests 133 which contain detailed algorithmsand computer source code for simulating secondaryand primary HVAC equipment. Studies have alsodemonstrated that properly calibrated simplified HVACsystem models can be used for measuring the performanceof commercial HVAC systems. 134,135,136,137Table 27.12: Lighting Calculations Methods from ASHRAE Guideline 14-2002. 124

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 731F-1. HVAC System TypesIn order to facilitate the description of measurementmethods that are applicable to a wide range of HVACsystems, it is necessary to categorize HVAC systems intogroups, such as single zone, steady state systems to themore complex systems such a multi-zone systems withsimultaneous heating and cooling. To accomplish thistwo layers of classification are proposed, in the first layer,systems are classified into two categories: systems thatprovide heating or cooling under separate thermostaticcontrol, and systems that provide heating and coolingunder a combined control. In the second classification,systems are grouped according to: systems that provideconstant heating rates, systems that provide varyingheating rates, systems that provide constant cooling rates,systems that provide varying cooling rates.• HVAC systems that provide heating or coolingat a constant rate include: single zone, 2-pipe fancoil units, ventilating and heating units, windowair conditioners, evaporative cooling. Systems thatprovide heating or cooling at a constant rate canbe measured using: single-point tests, multi-pointtests, short-term monitoring techniques, or in-situmeasurement combined with calibrated, simplifiedsimulation.• HVAC systems that provide heating or coolingat varying rates include: 2-pipe induction units,single zone with variable speed fan and/or compressors,variable speed ventilating and heatingunits, variable speed, and selected window airconditioners. Systems that provide heating orcooling at varying rates can be measured using:single-point tests, multi-point tests, short-termmonitoring techniques, or short-term monitoringcombined with calibrated, simplified simulation.• HVAC systems that provide simultaneous heatingand cooling include: multi-zone, dual ductconstant volume dual duct variable volume,single duct constant volume w/reheat, singleduct variable volume w/reheat, dual path systems(i.e., with main and preconditioning coils),4-pipe fan coil units, and 4-pipe induction units.Such systems can be measured using: in-situmeasurement combined with calibrated, simplifiedsimulation.F-2. HVAC System Testing MethodsIn this section four methods are described for thein-situ performance testing of HVAC systems as shownin Table 27.14, including: a single point method thatuses manufacturer’s performance data, a multiple pointmethod that includes manufacturer’s performance data,a multiple point that uses short-term data and manufacturer’sperformance data, and a short-term calibratedsimulation. Each of these methods is explained in thesections that follow.• Method #1: Single point with manufacturer’s performancedataIn this method the efficiency of the HVAC systemis measured with a single-point (or a series) offield measurements at steady operating conditions.On-site measurements include: the energy inputto system (e.g., electricity, natural gas, hot wateror steam), the thermal output of system, and thetemperature of surrounding environment. The efficiencyis calculated as the measured output/input.This method can be used in the following constantsystems: single zone systems, 2-pipe fan coil units,ventilating and heating units, single speed windowair conditioners, and evaporative coolers.Table 27.13: Relationship of HVAC Test Methods to Type of System.

732 ENERGY MANAGEMENT HANDBOOK• Method #2: Multiple point with manufacturer’sperformance dataIn this method the efficiency of the HVACsystem is measured with multiple points on themanufacturer’s performance curve. On-site measurementsinclude: the energy input to system(e.g., electricity, natural gas, hot water or steam),the thermal output of system, the system temperatures,and the temperature of surroundingenvironment. The efficiency is calculated as themeasured output/input, which varies accordingto the manufacturer’s performance curve. Thismethod can be used in the following systems:single zone (constant or varying), 2-pipe fan coilunits, ventilating and heating units (constant orvarying), window air conditioners (constant orvarying), evaporative cooling (constant or varying)2-pipe induction units (varying), single zone withvariable speed fan and/or compressors, variablespeed ventilating and heating units, and variablespeed window air conditioners.• Method #3: Multiple point using short-term dataand manufacturer’s performance dataIn this method the efficiency of the HVAC systemis measured continuously over a short-termperiod, with data covering the manufacturer’sperformance curve. On-site measurements include:the energy input to system (e.g., electricity, naturalgas, hot water or steam), the thermal output of system,the system temperatures, and the temperatureof surrounding environment. The efficiency is calculatedas the measured output/input, which variesaccording to the manufacturer’s performancecurve. This method can be used in the followingsystems: single zone (constant or varying), 2-pipefan coil units, ventilating and heating units (constantor varying), window air conditioners (constantor varying), evaporative cooling (constant orvarying) 2-pipe induction units (varying), singlezone with variable speed fan and/or compressors,variable speed ventilating and heating units, andvariable speed window air conditioners.• Method #4: Short-term monitoring and calibrated,simplified simulationIn this method the efficiency of the HVAC systemis measured continuously over a short-termperiod, with data covering the manufacturer’sperformance curve. On-site measurements include:the energy input to system (e.g., electricity, naturalgas, hot water or steam), the thermal output ofsystem, the system temperatures, and the temperatureof surrounding environment. The efficiency iscalculated using a calibrated air-side simulation ofthe system, which can include manufacturer’s performancecurves for various components. Similarmeasurements are repeated after the retrofit. Thismethod can be used in the following systems:single zone (constant or varying), 2-pipe fan coilunits, ventilating and heating units (constant orvarying), window air conditioners (constant orvarying), evaporative cooling (constant or varying),2-pipe induction units (varying), single zonewith variable speed fan and/or compressors, variablespeed ventilating and heating units, variablespeed window air conditioners, multi-zone, dualduct constant volume, dual duct variable volume,single duct constant volume w/reheat, single ductvariable volume w/reheat, dual path systems (i.e.,with main and preconditioning coils), 4-pipe fancoil units, 4-pipe induction unitsF-3. Calculation of Annual Energy UseThe calculation of annual energy use varies accordingto HVAC calculation method as shown in Table27.15. The savings are determined by comparing the annualHVAC energy use and demand during the baselineperiod to the annual HVAC energy use and demandduring the post-retrofit period.Whole-building or Main-meter ApproachOverviewThe whole-building approach, also called themain-meter approach, includes procedures that measurethe performance of retrofits for those projects wherewhole-building pre-retrofit and post-retrofit data areTable 27.14: HVAC System Testing Methods. 138,139


734 ENERGY MANAGEMENT HANDBOOKTable 27.14 (Continued)Table 27.15: HVAC PerformanceMeasurementMethods from ASHRAEGuideline 14-2002. 140

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 735available to determine the savings, and where the savingsare expected to be significant enough that the differencebetween pre-retrofit and post-retrofit usage canbe measured using a whole-building approach. Wholebuildingmethods can use monthly utility billing data(i.e., demand or usage), or continuous measurementsof the whole-building energy use after the retrofit ona more detailed measurement level (weekly, daily orhourly). Sub-metering measurements can also be usedto develop the whole-building models, providing thatthe measurements are available for the pre-retrofit andpost-retrofit period, and that meter(s) measures thatportion of the building where the retrofit was applied.Each sub-metered measurement then requires a separatemodel. Whole-building measurements can also be usedon stored energy sources, such as oil or coal inventories.In such cases, the energy used during a period needsto be calculated (i.e., any deliveries during the periodminus measured reductions in stored fuel).In most cases, the energy use and/or electricdemand are dependent on one or more independentvariables. The most common independent variable isoutdoor temperature, which affects the building’s heatingand cooling energy use. Other independent variablescan also affect a building’s energy use and peak electricdemand, including: the building’s occupancy (i.e., oftenexpressed as weekday or weekend models), parking orexterior lighting loads, special events (i.e., Friday nightfootball games), etc.Whole-building Energy Use ModelsWhole-building models usually involve the use ofa regression model that relates the energy use and peakdemand to one or more independent variables. The mostwidely accepted technique uses linear or change-pointlinear regression to correlate energy use or peak demandas the dependent variable with weather data and/orother independent variables. In most cases the wholebuildingmodel has the form:E = C + B 1 V 1 + B 2 V 2 + B 3 V 3 + …whereE = the energy use or demand estimated bythe equation,C = a constant term in energy units/dayor demand units/billing period,B n = the regression coefficient of anindependent variable V n ,V n = the independent driving variable.In general, when creating a whole-building modelfor a number of different regression models are triedfor a particular building and the results are comparedand the best model selected using R 2 and CV (RMSE).Table 27.16 and Figure 27.7 contain models listed inASHRAE’s Guideline 14-2002, which include steadystateconstant or mean models, models adjusted for thedays in the billing period, two-parameter models, threeparametermodels or variable-based degree-day models,four-parameter models, five-parameter models, andmultivariate models. All of these models can be calculatedwith ASHRAE Inverse Model Toolkit (IMT), whichwas developed from Research Project 1050-RP. 141The steady-state, linear, change-point linear, variable-baseddegree-day and multivariate inverse modelscontained in ASHRAE’s IMT have advantages overother types of models. First, since the models are simple,and their use with a given dataset requires no humanintervention, the application of the models can be on canbe automated and applied to large numbers of build-Table 27.16: Sample Models for the Whole-Building Approach from ASHRAE Guideline 14-2002. 152

736 ENERGY MANAGEMENT HANDBOOKings, such as those contained in utility databases. Sucha procedure can assist a utility, or an owner of a largenumber of buildings, identify which buildings haveabnormally high energy use. Second, several studieshave shown that linear and change-point linear modelcoefficients have physical significance to operation ofheating and cooling equipment that is controlled by athermostat. 142,143,144,145 Finally, numerous studies havereported the successful use of these models on a varietyof different buildings. 146,147,148,149,150,151Steady-state models have disadvantages, including:an insensitivity to dynamic effects (e.g., thermalmass), insensitivity to variables other than temperature(e.g., humidity and solar), and inappropriateness forcertain building types, for example building that havestrong on/off schedule dependent loads, or buildingsthat display multiple change-points. If whole-buildingmodels are required in such applications, alternativemodels will need to be developed.A. One-parameter or Constant ModelOne-parameter, or constant models are modelswhere the energy use is constant over a given period.Such models are appropriate for modeling buildingsthat consume electricity in a way that is independentof the outside weather conditions. For example, suchmodels are appropriate for modeling electricity use inbuildings which are on district heating and cooling systems,since the electricity use can be well represented bya constant weekday-weekend model. Constant modelsare often used to model sub-metered data on lightinguse that is controlled by a predictable schedule.B. Day-adjusted ModelDay-adjusted models are similar to one-parameterconstant models, with the exception that the final coefficientof the model is expressed as an energy use perday, which is then multiplied by the number of days inthe billing period to adjust for variations in the utilitybilling cycle. Such day-adjusted models are often usedwith one, two, three, four and five-parameter linear orchange-point linear monthly utility models, where theenergy use per period is divided by the days in thebilling period before the linear or change-point linearregression is performed.Figure 27.7: Sample Models for the Whole-buildingApproach. Included in this figure is: (a) mean or oneparametermodel, (b) two-parameter model, (c) threeparameterheating model (similar to a variable baseddegree-day model (VBDD) for heating), (d) three-parametercooling model (VBDD for cooling), (e) fourparameterheating model, (f) four-parameter coolingmodel, and (g) five-parameter model. 153C. Two-parameter ModelTwo-parameter models are appropriate for modelingbuilding heating or cooling energy use in extremeclimates where a building is exposed to heating orcooling year-around, and the building has an HVACsystem with constant controls that operates continuously.Examples include outside air pre-heating systemsin arctic conditions, or outside air pre-cooling systemsin near-tropical climates. Dual-duct, single-fan, constantvolumesystems, without economizers can also be modeledwith two-parameter regression models. Constantuse, domestic water heating loads can also be modeledwith two-parameter models, which are based on thewater supply temperature.D. Three-parameter ModelThree-parameter models, which include changepointlinear models or variable-based, degree day

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 737models, can be used on a wide range of building types,including residential heating and cooling loads, smallcommercial buildings, and models that describe the gasused by boiler thermal plants that serve one or morebuildings. In Table 27.16, three-parameter models haveseveral formats, depending upon whether or not themodel is a variable based degree-day model or threeparameter,change-point linear models for heating orcooling. The variable-based degree day model is definedas:E = C + B 1 (DD BT )whereC = the constant energy use below (or above)the change point, andB 1 = the coefficient or slope that describes thelinear dependency on degree-days,DD BT = the heating or cooling degree-days (ordegree hours), which are based on thebalance-point temperature.The three-parameter change-point linear model for heatingis described by 154whereE = C + B 1 (B 2 – T) +C = the constant energy use above thechange point,B 1 = the coefficient or slope that describes thelinear dependency on temperature,B 2 = the heating change point temperature,T = the ambient temperature for the periodcorresponding to the energy use,+ = positive values only inside theparenthesis.The three-parameter change-point linear model for coolingis described bywhereE = C + B 1 (T – B 2 ) +C = the constant energy use below the changepoint,B 1 = the coefficient or slope that describes thelinear dependency on temperature,B 2 = the cooling change point temperature,T = the ambient temperature for the periodcorresponding to the energy use,+ = positive values only for the parentheticalexpression.E. Four-parameter ModelThe four-parameter change-point linear heatingmodel is typically applicable to heating usage in buildingswith HVAC systems that have variable-air volume,or whose output varies with the ambient temperature.Four-parameter models have also been shown to beuseful for modeling the whole-building electricity useof grocery stores that have large refrigeration loads,and significant cooling loads during the cooling season.Two types of four-parameter models are listed in Table27.16, including a heating model and a cooling model.The four-parameter change-point linear heating modelis given bywhereE = C + B 1 (B 3 - T) + - B 2 (T - B 3 ) +C = the energy use at the change point,B 1 = the coefficient or slope that describes thelinear dependency on temperature belowthe change point,B 2 = the coefficient or slope that describes thelinear dependency on temperature abovethe change pointB 3 = the change-point temperature,T = the temperature for the period of interest,+ = positive values only for the parentheticalexpression.The four-parameter change-point linear cooling modelis given bywhereE = C - B 1 (B 3 - T) + + B 2 (T - B 3 ) +C = the energy use at the change point,B 1 = the coefficient or slope that describesthe linear dependency on temperaturebelow the change point,B 2 = the coefficient or slope that describesthe linear dependency on temperatureabove the change pointB 3 = the change-point temperature,T = the temperature for the period ofinterest,+ = positive values only for theparenthetical expression.F. Five-parameter ModelFive-parameter change-point linear models areuseful for modeling the whole-building energy usein buildings that contain air conditioning and electricheating. Such models are also useful for modeling the

738 ENERGY MANAGEMENT HANDBOOKweather dependent performance of the electricity consumptionof variable air volume air-handling units. Thebasic form for the weather dependency of either caseis shown in Figure 27.7f, where there is an increase inelectricity use below the change point associated withheating, an increase in the energy use above the changepoint associated with cooling, and constant energy usebetween the heating and cooling change points. Fiveparameterchange-point linear models can be describedusing variable-based degree day models, or a five-parametermodel. The equation for describing the energyuse with variable-based degree days isE = C - B 1 (DD TH ) + B 2 (DD TC )whereC = the constant energy use between theheating and cooling change points,B 1 = the coefficient or slope that describes thelinear dependency on heating degree-days,B 2 = the coefficient or slope that describes thelinear dependency on cooling degree-days,DD TH = the heating degree-days (or degree hours),which are based on the balance-pointtemperature.DD TC = the cooling degree-days (or degree hours),which are based on the balance-pointtemperature.The five-parameter change-point linear model that isbased on temperature isE = C + B 1 (B 3 - T) + + B 2 (T – B 4 ) +whereC = the energy use between the heating andcooling change points,B 1 = the coefficient or slope that describes thelinear dependency on temperature belowthe heating change point,B 2 = the coefficient or slope that describes thelinear dependency on temperature abovethe cooling change pointB 3 = the heating change-point temperature,B 4 = the cooling change-point temperature,T = the temperature for the period of interest,+ = positive values only for the parentheticalexpression.G. Whole-building Peak Demand ModelsWhole-building peak electric demand models differfrom whole-building energy use models in severalrespects. First, the models are not adjusted for the daysin the billing period since the model is meant to representthe peak electric demand. Second, the models areusually analyzed against the maximum ambient temperatureduring the billing period. Models for whole-buildingpeak electric demand can be classified according toweather-dependent and weather-independent models.G-1. Weather-dependentWhole-building Peak Demand ModelsWeather-dependent, whole-building peak demandmodels can be used to model the peak electricity use ofa facility. Such models can be calculated with linear andchange-point linear models regressed against maximumtemperatures for the billing period, or calculated with aninverse bin model. 155,156G-2. Weather-independentWhole-building Peak Demand ModelsWeather-independent, whole-building peak demandmodels are used to measure the peak electric usein buildings or sub-metered data that do not show significantweather dependencies. ASHRAE has developeda diversity factor toolkit for calculating weather-independentwhole-building peak demand models as partof Research Project 1093-RP. This toolkit calculates the24-hour diversity factors using a quartile analysis. Anexample of the application of this approach is given inthe following section.Example: Whole-building energy use modelsFigure 27.8 presents an example of the typical datarequirements for a whole-building analysis, includingone year of daily average ambient temperatures andtwelve months of utility billing data. In this exampleof a residence, the daily average ambient temperatureswere obtained from the National Weather Service (i.e.,the average of the published min/max data), and theutility bill readings represent the actual readings fromthe customer’s utility bill. To analyze these data severalcalculations need to be performed. First, the monthlyelectricity use (kWh/month) needs to be divided by thedays in the billing period to obtain the average dailyelectricity use for that month (kWh/day). Second, theaverage daily temperatures need to be calculated fromthe published NWS min/max data. From these averagedaily temperatures the average billing period temperatureneed to be calculated for each monthly utility bill.The data set containing average billing period temperaturesand average daily electricity use is then analyzedwith ASHRAE’s Inverse Model Toolkit (IMT) 157 todetermine a weather normalized consumption as shown

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 739in Figures 27.9 and 27.10. In Figure 27.9 the twelvemonthly utility bills (kWh/period) are shown plottedagainst the average billing period temperature alongwith a three-parameter change-point model calculatedwith the IMT. In Figure 27.10 the twelve monthly utilitybills, which were adjusted for days in the billing period(i.e., kWh/day) are shown plotted against the averagebilling period temperature along with a three-parameterchange-point model calculated with the IMT. Inthe analysis for this house, the use of an average dailymodel improved the accuracy of the unadjusted model(i.e., Figure 27.9) from an R 2 of 0.78 and CV (RMSE) of24.0% to an R 2 of 0.83 and a CV (RMSE) of 19.5% forthe adjusted model (i.e., Figure 27.10), which indicatesa significant improvement in the model.In another example the hourly steam use (Figure27.11) and hourly electricity use (Figure 27.13) for theU.S. DOE Forrestal Building is modeled with a dailyweekday-weekend three-parameter, change-point modelfor the steam use (Figure 27.12), and an hourly weekdayweekenddemand model for the electricity use (Figure27.14). To develop the weather-normalized model for thesteam use the hourly steam data and hourly weatherdata were first converted into average daily data, then athree-parameter, weekday-weekend model was calculatedusing the EModel software, 158 which contains similaralgorithms as ASHRAE’s IMT. The resultant model,which is shown in Figure 27.12 along with the dailysteam, is well described with an R 2 of 0.87 an RMSE of50,085.95 kBtu/day and a CV (RMSE) of 37.1%.In Figure 27.14 hourly weather-independent 24-hour weekday-weekend profiles have been created forFigure 27.8: Example Data for Monthly Whole-building Analysis (upper trace, daily average temperature, F,lower points, monthly electricity use, kWh/day).Figure 27.9 Example Unadjusted Monthly WholebuildingAnalysis (3P Model) for kWh/period (R 2 =0.78, CV (RMSE) = 24.0%).Figure 27.10. Example Adjusted Whole-building Analysis(3P Model) for kWh/day (R 2 = 0.83, CV (RMSE)= 19.5%).

740 ENERGY MANAGEMENT HANDBOOKthe whole-building electricity use using ASHRAE’s1093-RP Diversity Factor Toolkit. 159 These profiles canbe used to calculate the baseline whole-building electricityuse (i.e., using the mean hourly use) by multiplyingtimes the expected weekdays and weekends in the year.The profiles can also be used to calculate the peak electricityuse (i.e., using the 90th percentile).Calculation of Annual Energy UseOnce the appropriate whole-building model hasbeen chosen and applied to the baseline data, the annualenergy use for the baseline period and the post-retrofitperiod are then calculated. Savings are then calculatedby comparing the annual energy use of the baseline withthe annual energy use of the post-retrofit period.Whole-building Calibrated Simulation ApproachWhole-building calibrated simulation normallyrequires the hourly simulation of an entire building,including the thermal envelope, interior and occupantloads, secondary HVAC systems (i.e., air handlingunits), and the primary HVAC systems (i.e., chillers,boilers). This is usually accomplished with a generalpurpose simulation program such as BLAST, DOE-2or EnergyPlus, or similar proprietary programs. Suchprograms require an hourly weather input file forthe location in which the building is being simulated.Calibrating the simulation refers to the process wherebyselected outputs from the simulation are compared andeventually matched with measurements taken from anactual building. A number of papers in the literaturehave addressed techniques for accomplishing these calibrations,and include results from case study buildingswhere calibrated simulations have been developed forvarious purposes. 160, 161, 162, 163, 164, 165, 166, 167, 168, 169,170, 171,172,173,174,175Applications of Calibrated Whole-building Simulation.Calibrated whole-building simulation can be auseful approach for measuring the savings from energyconservation retrofits to buildings. However, it is generallymore expensive than other methods, and therefore itis best reserved for applications where other, less costlyapproaches cannot be used. For example, calibratedsimulation is useful in projects where either pre-retrofitor post-retrofit whole-building metered electrical dataare not available (i.e., new buildings or buildings withoutmeters such as many college campuses with centralfacilities). Calibrated simulation is desired in projectswhere there are significant interactions between retrofits,for example lighting retrofits combined with changesto HVAC systems, or chiller retrofits. In such cases thewhole-building simulation program can account forthe interactions, and in certain cases, actually isolateinteractions to allow for end-use energy allocations. Itis useful in projects where there are significant changesin the facility’s energy use during or after a retrofit hasbeen installed, where it may be necessary to accountfor additions to a building that add or subtract thermalloads from the HVAC system. In other cases, demandmay change over time, where the changes are not relatedto the energy conservation measures. Therefore,adjustments to account for these changes will be alsobe needed. Finally, in many newer buildings, as-builtdesign simulations are being delivered as a part of thebuilding’s final documents. In cases where such simulationsare properly documented they can be calibrated tothe baseline conditions and then used to calculate andmeasure retrofit savings.Unfortunately, calibrated, whole-building simulationis not useful in all buildings. For example, if abuilding cannot be readily simulated with available simulationprograms, significant costs may be incurred inFigure 27.11: Example Heating Data for Daily Whole-building Analysis.

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 741Figure 27.12: ExampleDaily Weekday-weekendWhole-buildingAnalysis (3P Model)for Steam Use (kBtu/day, R 2 = 0.87, RMSE =50,085.95, CV (RMSE)= 37.1%). Weekday use(x), weekend use ( ).Figure 27.13: Example Electricity Data for Hourly Whole-building Demand Analysis.Figure 27.14: Example Weekday-weekend Hourly Whole-building Demand Analysis (1093-RP Model) for ElectricityUse.modifying a program or developing a new program tosimulate only one building (e.g., atriums, undergroundbuildings, buildings with complex HVAC systemsthat are not included in a simulation program’s systemlibrary). Additional information about calibrated,whole-building simulation can be found in ASHRAE’sGuideline 14-2002.Figure 27.15 provides an example of the use ofcalibrated simulation to measure retrofit savings in aproject where pre-retrofit measurements were not avail-

742 ENERGY MANAGEMENT HANDBOOKable. In this figure both the before-after whole-buildingapproach and the calibrated simulation approach areillustrated. On the left side of the figure the traditionalwhole-building, before-after approach is shown for abuilding that had a dual-duct, constant volume system(DDCV) replaced with a variable air volume (VAV) system.In such a case where baseline data are available,the energy use for the building is regressed against thecoincident weather conditions to obtain the representativebaseline regression coefficients. After the retrofit isinstalled, the energy savings are calculated by comparingthe projected pre-retrofit energy use against themeasured post-retrofit energy use, where the projectedpre-retrofit energy use calculated with the regressionmodel (or empirical model), which was determined withthe facility’s baseline DDCV data.In cases where the baseline data are not available(i.e., the right side of the figure), a simulation of thebuilding can be developed and calibrated to the postretrofitconditions (i.e., the VAV system). Then, using thecalibrated simulation program, the pre-retrofit energyuse (i.e., DDCV system) can be calculated for conditionsin the post-retrofit period, and the savings calculated bycomparing the simulated pre-retrofit energy use againstthe measured post-retrofit energy use. In such a casethe calibrated post-retrofit simulation can also be usedto fill-in any missing post-retrofit energy use, which isa common occurrence in projects that measure hourlyenergy and environmental conditions. The accuracy ofthe post-retrofit model depends on numerous factors.Methodology for Calibrated Whole-building SimulationCalibrated simulation requires a systematic approachthat includes the development of the wholebuildingsimulation model, collection of data from thebuilding being retrofitted and the coincident weatherdata. The calibration process then involves the comparisonof selected simulation outputs against measureddata from the systems being simulated, and the adjustmentof the simulation model to improve the comparisonof the simulated output against the correspondingmeasurements. The choice of simulation program isa critical step in the process, which must balance themodel appropriateness, algorithmic complexity, userexpertise, and degree of accuracy against the resourcesavailable to perform the modeling.Data collection from the building includes the collectionof data from the baseline and post-retrofit periods,which can cover several years of time. Building datato be gathered includes such information as the buildinglocation, building geometry, materials characteristics,equipment nameplate data, operations schedules, temperaturesettings, and at a minimum whole-buildingutility billing data. If the budget allows, hourly whole-Figure 27.15: Flow Diagram for Calibrated Simulation Analysis of Air-Side HVAC System. 176

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 743building energy use and environmental data can begathered to improve the calibration process, which canbe done over short-term, or long-term period.Figure 27.16 provides an illustration of a calibrationprocess that used hourly graphical and statisticalcomparisons of the simulated versus measured energyuse and environmental conditions. In this example, thesite-specific information was gathered and used to developa simulation input file, including the use of measuredweather data, which was then used by the DOE-2program to simulate the case study building. Hourlydata from the simulation program was then extractedand used in a series of special-purpose graphical plotsto help guide the calibration process (i.e., time series, binand 3-D plots). After changes were made to the inputfile, DOE-2 was then run again, and the output comparedagainst the measured data for a specific period.This process was then repeated until the desired level ofcalibration was reached, at which point the simulationwas proclaimed to be “calibrated.” The calibrated modelwas then used to evaluate how the new building wasperforming compared to the design intent.A number of different calibration tools have beenreported by various investigators, ranging from simpleX-Y scatter plots to more elaborate statistical plots andindices. Figures 27.17, 27.18 and 27.19 provide examplesof several of these calibration tools. In Figure 27.17 anexample of an architectural rendering tool is shown thatassists the simulator with viewing the exact placementof surfaces in the building, as well as shading fromnearby buildings, and north-south orientation. In Figure27.18 temperature binned calibration plots are showncomparing the weather dependency of an hourly simulationagainst measured data. In this figure the upperplots show the data as scatter plots against temperature.The lower plots are statistical, temperature-binned boxwhisker-meanplots, which include the super positioningof measured mean line onto the simulated mean lineto facilitate a detailed evaluation. In Figure 27.19 comparativethree-dimensional plots are shown that showmeasured data (top plot), simulated data (second plotfrom the top), simulated minus measured data (secondplot from the bottom, and measured minus simulateddata (bottom plot). In these plots the day-of-the-year isthe scale across the page (y axis), the hour-of-the-day isthe scale projecting into the page (x axis), and the hourlyFigure 27.16: Calibration Flowchart. This figure showsthe sequence of processing routines that were used todevelop graphical calibration procedures. 178Figure 27.17: Example Architecture Rendering of theRobert E. Johnson Building, Austin, Texas. 179,180

744 ENERGY MANAGEMENT HANDBOOKelectricity use is the vertical scale of the surface abovethe x-y plane. These plots are useful for determininghow well the hourly schedules of the simulation matchthe schedules of the real building, and can be used toidentify other certain schedule-related features. For example,in the front of plot (b) the saw-toothed featureis indicating on/off cycling of the HVAC system, whichis not occurring in the actual building.Table 27.17 contains a summary of the proceduresused for developing a calibrated, whole-buildingsimulation program, as defined in ASHRAE’s Guideline14-2002. In general, to develop a calibrated simulation,detailed information is required for a building, includinginformation about the building’s thermal envelope(i.e., the walls, windows, roof, etc.), information aboutthe building’s operation, including temperature settings,HVAC systems, and heating-cooling equipment that existedboth during the baseline and post-retrofit period.This information is input into two simulation files, onefor the baseline and one for the post-retrofit conditions.Savings are then calculated by comparing the twosimulations of the same building, one that represents thebaseline building, and one that represents the building’soperations during the post-retrofit period.27.2.2 Role of M&VEach Energy Conservation Measure (ECM) presentsparticular requirements. These can be grouped infunctional sections as shown in Table 27.18. Unfortunately,in most projects, numerous variables exist sothe assessments can be easily disputed. In general, thelow risk (L)—reasonable payback ECMs exhibit steadyperformance characteristics that tend not to degradeor become easily noticed when savings degradationoccurs. These include lighting, constant speed motors,two-speed motors and IR radiant heating. The highrisk (H)—reasonable payback ECMs include EMCSs,variable speed drives and control retrofits. The savingsfrom these ECMs can be overridden by building operatorsand not be noticed until years later. Most otherECMs fall in the category of “it depends.” The attentionthat the operations and maintenance directs at thesedramatically impacts the sustainability of the operationand the savings. With an EMCS, operators can set uptrend reports to measure and track occupancy scheduleoverrides, the various reset schedule overrides, variablespeed drive controls and even monitor critical parameterswhich track mechanical systems performance. illustratesa “most likely” range of ratings for the variouscategories. 183Often, building envelope or mechanical systemsneed to be replaced. Building systems have finite life-Figure 27.18: Temperature Bin Calibration Plots. This figureshows the measured and simulated hourly weekdaydata as scatter plots against temperature in the upper plotsand as statistical binned box-whisker-mean plots in thelower plots. 181Figure 27.19: Comparative Three-dimensionalPlots. (a) Measured Data. (b) SimulatedData. (c) Simulated-Measured Data.(d) Measured-Simulated Data.

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 745Table 27.17: Calibrated, whole-building Simulation Procedures from ASHRAE Guideline 14-2002. 177times, ranging from two to five years for most light bulbsto 10 to 20+ years for chillers and boilers. Building envelopereplacements like insulation, siding, roof, windowsand doors can have lifetimes from 10 to 50 years. In theseinstances, life cycle costing should be done to comparethe total cost of upgrading to more efficient technology.Also, the cost of M&V should be considered when determininghow to sustain the savings and performance ofthe replacement. In many cases, the upgraded efficiencywill have a payback of less than 10 years when comparedto the current efficiency of the existing equipment. Currenttechnology high efficiency upgrades normally usecontrols to acquire the high efficiency. These controlsoften connect to standard interfaces so that they communicatewith today’s state of the art Energy Managementand Control Systems (EMCSs).27.2.3 Cost/Benefit AnalysisThe target for work for the USAF has been 5%of the savings. 184 The cost of the M&V can exceed 5%if the risk of losing savings exceeds predefined limits.The Variable Speed Drive ECM illustrates these opportunitiesand risks. VSD equipment exhibits highreliability. Equipment type of failures normally happenwhen connection breaks occur with the control input,the remote sensor. Operator induced failures occur thenthe operator sets the unit to 100% speed and does notre-enable the control. Setting the unit to 100% can occurfor legitimate reasons. These reasons include running atest, overriding a control program that does not provideadequate speed under specific, and typically unusual,circumstances, or requiring 100% operation for a limitedtime. The savings disappear if the VSD remains at 100%operating speed.For example, consider a VSD ECM with ten (10)motors with each motor on a different air handlingunit. Each motor has fifty (50) horsepower (HP). Thebase case measured these motors running 8760 hoursper year at full speed. Assume that the loads on themotors matched the nameplate 50 HP at peak loads.

746 ENERGY MANAGEMENT HANDBOOKTable 27.18: Overview of Risks and Costs for ECMs.Although the actual load on a AHU fan varies withthe state of the terminal boxes, assume that the loadaverage equates to 80% of the full load since the ductpressure will rise as the terminal boxes reduce flow atthe higher speed. Table 27.19 contains the remainingassumptions. To correctly determine the average powerload, the average power must either be integrated overthe period of consumption or the bin method must beused. For the purposes of this example, the 14.4% valuewill be used.The equation below shows the relationship betweenthe fan speed and the power consumed. The exponenthas been observed to vary between 2.8 (at highflow) and 2.7 (at reduced flow) for most duct systems.This includes the loss term from pressure increases ata given fan speed. Changing the exponent from 2.8 to2.7 reduces the savings by less than 5%.Pwr = Pwr 0 ×% SpeedFull SpeedDemand savings will not be considered in this example.Demand savings will likely be very low if the utilityhas a 12-month ratchet clause and the summer loadrequires some full speed operation during peak times.Assuming a $12.00/kW per month demand charge, demandsavings could be high for off-season months if thedemand billing resets monthly. Without a ratchet clause,rough estimates have yearly demand savings ranging upto $17,000 if the fan speed stays under 70% for 6 monthsper year. Yearly demand savings jump to over $20,000 ifthe fan speed stays under 60% for 6 months per year.2.8Table 27.19: VSD Example Assumptions.

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 747Figure 27.20 illustrates the savings expected fromthe VSD ECM by hours of use per year. The 5% and10% of Savings lines define the amount available forM&V expenditures at these levels. In this example, theECM savings exceeds $253,000 per year. Five percent(5%) of savings over a 20-year project life makes $253Kavailable for M&V and ten percent (10%) of the savingsmakes $506K available over the 20-year period. Ifthe motors run less frequently than continuous, savingsdecrease as shown in Figure 27.20. Setting up the M&Vprogram to monitor the VSDs on an hourly basis andreport savings on a monthly report requires monitoringthe VSD inverter with an EMCS to poll the data andcreate reports.To provide the impact of the potential losses fromlosing the savings, assume the savings degrades at aloss of 10% of the total yearly savings per year. Studieshave shown that control ECMs like the VSD examplecan expect to see 20% to 30% degradation in savings in2 to 3 years. Figure 27.21 illustrates what happens to theFigure 27.20: Example VSD EMC Yearly Savings/M&VCost.savings in 20 years with 10% of the savings spent onM&V. Note that the losses exceed the M&V cost duringthe first year, resulting in a net loss of almost $3,000,000over the 20-year period. Figure 27.22 shows the savingsper year with a 10% loss of savings. M&V costs remainat 10% of savings. At the end of the 20-year period, thesavings drop to almost $30,000 per year out of a potentialsavings of over $250,000 per year.This example shows the cumulative impact of losingsavings on a year by year basis. The actual savingsamounts will vary depending upon the specific factorsin an ECM and can be scaled to reflect a specific application.Increasing the M&V cost to reduce the loss of savingsoften makes sense and must be carefully thoughtthrough.27.2.4 Cost Reduction StrategiesM&V strategies can be cost reduced by loweringthe requirements for M&V or by statistical sampling.Reducing requirements involves performing trade-offswith the risks and benefits of having reliable numbers todetermine the savings and the costs for these measurements.27.2.4.1 Constant Load ECMsLighting ECMs can save 30% of the pre-ECMenergy and have a payback in the range of 3 to 6years. Assuming that the lighting ECM was designedand implemented per the specifications and the savingswere verified to be occurring, just verifying thatthe storeroom has the correct ballasts and lamps mayconstitute acceptable M&V on a yearly basis. This costsfar less than performing a yearly set of measurements,analyzing them and then creating reports. In this case,other safeguards should be implemented to assure thatthe bulb and ballast replacement occurs and meets theFigure 27.21: Yearly Impact of Ongoing Losses.Figure 27.22: Cumulative Impact of Savings Loss.

748 ENERGY MANAGEMENT HANDBOOKrequirements specified.High efficiency motor replacements provide anotherexample of constant load ECMs. The key shortterm risks with motor replacements involve installingthe right motor with all mechanical linkages and electricalcomponents installed correctly. Once verified, thelong term risks for maintaining savings occur when themotor fails. The replacement motor must be the correctmotor or savings can be lost. A sampled inspection reducesthis risk. Make sure to inspect all motors at leastonce every five (5) years.27.2.4.2 Major Mechanical SystemsBoilers, chillers, air handler units, cooling towerscomprise the category of manor mechanical equipmentin buildings. They need to be considered separately aseach carry their own set of short-term and long-termrisks. In general, measurements provide necessaryrisk reduction. The question becomes: What measurementsreduce the risk of savings loss by an acceptableamount?First a risk assessment needs to be performed. Theshort-term risks for boilers involve installing the wrongsize or installing the boiler improperly (not to specifications).Long-term savings sustainability risks tend tofocus on the water side and the fire side. Water deposits(K + , Ca ++ , Mg + ) will form on the inside of the tubes andadd a thermal barrier to the heat flow. The fire side canadd a layer of soot if the O 2 level drops too low. Eitherof these reduce the efficiency of the boiler over the longhaul. Generally this can take several years to impactthe efficiency if regular tune-ups and water treatmentoccurs.Boilers come in a wide variety of shapes andsizes. Boiler size can be used as a defining criterion formeasurements. Assume that natural gas or other boilerfuels cost about $5.00 per MMBtu. Although fuel priceconstantly changes, it provides a reference point for thisanalysis. Thus a boiler with 1MMBtu per hour output,an efficiency of 80% and operating at 50% load 3500hours per year, consumes about $11,000 per year. If thisboiler replaced a less efficient boiler, say at 65%, then thenet savings amounts to about $2,500 per year, assumingthe same load from the building. At 5% of the savings,$125 per year can be used for M&V. This does not allowmuch M&V. At 10% of the annual savings, $250 peryear can be used. At this level of cost, a combustion efficiencymeasurement could be performed, either yearlyor bi-yearly, depending on the local costs. In 2003 theASME’s Power Test Code 4.1 (PTC-4.1) 185 was replacedwith PTC4. Either of these codes allows two methodsto measuring boiler efficiency. The first method usesthe energy in equals energy out—using the first law ofthermodynamics. This requires measuring the Btu inputvia the gas flow and the Btu output via the steam (orwater) flow and temperatures. The second method measuresthe energy loss due to the content and temperatureof the exhausted gases, radiated energy from theshell and piping and other loss terms (like blowdown).The energy loss method can be performed in less thana couple of hours. The technician performing thesemeasurements must be skilled or significant errors willresult in the calculated efficiency. The equation belowshows the calculations required.Efficiency = 100% – Losses + CreditsThe losses term includes the temperature of theexhaust gas and a measure of the unburned hydrocarbonsby measuring CO 2 or O 2 levels, the loss due toexcess CO and a radiated term. Credits seldom occurbut could arise from sola r heating t he makeup wateror s imilar contributions. The Greek letter “η” us uallydenotes efficiency.As with boilers, a risk assessment needs to beperformed for chilers. The short term risks for chillersinvolve sizing or improper installation. Long term savingssustainability risks focus on the condenser watersystem, as circulation occurs in an open system. Waterdeposits (K + , Ca ++ , Mg + , organics) will form on theinside of the condenser tubes and add a barrier to thethermal flow. These reduce the efficiency of the chillerover the long haul. Generally this can take several yearsto impact the efficiency if proper water treatment occurs.Depending on the environmental conditions, the qualityof the makeup water and the water treatment, condensertube fouling should be checked every year or at leastevery other year.Chillers consume electricity in the case of mostcentrifugal, screw, scroll and reciprocating compressors.Direct-fired absorbers and engine driven compressorsuse a petroleum based fuel. As with boilers, chillersize and application sets the basic energy consumptionlevels. Assume, for the purpose of this example, thatelectricity provides the chiller energy. Older chillers withwater towers often operate at the 0.8 to 1.3 kW per tonlevel of efficiency. New chillers with water towers canoperate in the 0.55 to 0.7 range of efficiency. Note thatthe efficiency of any chiller depends upon the specificoperating conditions. Also assume the following: 500Tons centrifugal chiller with the specifications shown inTable 27.20. Under these conditions the chiller produces400 Tons of chilled water and requires an expenditureof $ 38,000 per year, considering both energy use and

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 749demand charges. Some utilities only charge demandcharges on the transmission and delivery (T&D) partsof the rate structure. In that case, the cost at $0.06/kWhwould be closer to $28,000. Using the 5% (10%) guidelinefor M&V costs as a percentage of savings leavesalmost $1,100 ($2,200) per year to spend on M&V. Thiscreates an allowable expenditure over a 20-year projectof $22,000 ($44,000) for M&V. If the utility has a ratchetclause in the rate structure, the amount for M&V increasesto $1,700 ($3,400) per year. At $1,100 per year,trade-offs will need to be made to stay within that“budget.” The risks need to be weighed and decisionsmade as to what level of M&V costs will be allowed.To determine the actual efficiency of a chiller requiresaccurate measurements of the chilled water flow,the difference between the chilled water supply andreturn temperatures and the electrical power providedto the chiller. Costs can be reduced using an EMCS ifonly temperature, flow and power sensors need to beinstalled.sustainability risks. When an operator overrides a strategyand forgets to re-enable it, the savings disappear. Acommon EMCS ECM requires the installation of equipmentand programs used to set back temperatures orturn off equipment. Short term risks involve setting upthe controls so that performance enhances, or at leastdoes not degrade, the comfort of the occupants. Whendiscomfort occurs, either occupants set up “portableelectric reheat units” or operators override the controlprogram. For example, when the night set-back controldoes not get the space to comfort by occupancy, operatorstypically override instead of adjusting the parametersin the program. These actions tend to occur duringpeak loading times and then not get re-enabled duringmilder times. Long term risks cover the same area asshort term risks. A new operator or a failure in remoteequipment that does not get fixed will likely cause theloss of savings. Estimating the savings cost for variousprojects can be done when the specifics are known.Table 27.20: Example of Savings with a 500 Ton Chiller.Cooling tower replacement requires knowledge ofthe risks and costs involved. As with boilers and chillers,the primary risks involve the water treatment. Controlscan be used to improve the efficiency of a chiller/towercombination by as much as 15% to 20%. As has beenpreviously stated, control ECMs often get overriddenand the savings disappear.Table 27.21: Sampling Requirements.27.2.4.3 Control SystemsControl ECMs encompass a wide spectrum of capabilitiesand costs. Upgrading a pneumatic control systemand installing EP (electronic to pneumatic) transducersinvolves the simple end. The complex side could spaninstalling a complete EMCS with sophisticated controls,with various reset, pressurization and control strategies.Generally, EMCSs function as basic controls and do notget widely used in sophisticated applications.Savings due to EMCS controls bear high

750 ENERGY MANAGEMENT HANDBOOKRisk abatement can be as simple as requiring atrend report weekly or at least monthly. M&V costs cangenerally be easily held under 5% when using an EMCSand creating trend reports.27.2.5 M&V Sampling StrategiesM&V can be made significantly lower cost by sampling.Sampling also reduces the timeliness of obtainingspecific data on specific equipment. The benefits ofsampling arise when the population of items increases.Table 27.21 (M&V Guidelines: Measurement and Verificationfor Federal Energy Projects, Version 2.2, AppendixD) illustrates how confidence and precision impact thenumber of samples required in a given population ofitems.Lighting ECMs may involve thousands of fixtures.For example, to obtain a savings estimate for 1,000 ormore fixtures, with a confidence of 80% and a precisionof 20%, 11 fixtures would need to be sampled. If therequirements increased to a confidence of 90% and a precisionof 10%, 68 fixtures would need to be sampled.The boiler ECM also represents opportunity forM&V cost reduction using sampling. Assume that theECM included replacing 50 boilers. If a confidence of80% and a precision of 20% satisfy the requirements,10 boilers would need to be sampled. The cost is thenreduced to 20% of the cost of measuring all boilers, asignificant savings. A random sampling to select thesample set can easily be implemented.References1. Claridge D.E., Turner, W.D., Liu, M., Deng, S., Wei, G., Culp, C.,Chen, H., and Cho, S. 2002. “Is Commissioning Once Enough?”Solutions for Energy Security and Facility Management Challenges:Proceedings of the 25th WEEC, Atlanta, GA, October19-11, 2002, pp. 29-36.2. Haberl, J., Lynn, B., Underwood, D., Reasoner, J., Rury, K. 2003.“Development an M&V Plan and Baseline for the Ft. HoodESPC Project,” ASHRAE Seminar Presentation, (June).3. C. Culp, K.Q. Hart, B. Turner, S. Berry-Lewis, 2003. “EnergyConsumption Baseline: Fairchild AFB’s Major Boiler Retrofit,”ASHRAE Seminar (January).4. Arnold, D., 1999. “The Evolution of Modem Office Buildingsand Air Conditioning,” ASHRAE Journal, American Society ofHeating Refrigeration Air-conditioning Engineers, Atlanta, GA,pp. 40-54, (June).5. Donaldson, B., Nagengast. 1994. Heat and Cold: Masteringthe Great Indoors. American Society of Heating RefrigerationAir-conditioning Engineers, Atlanta, GA.6. Cheney, M., Uth, R. 1999. Tesla: Master of Lightning. Barnesand Noble Books, New York, N,Y.7. Will, H. 1999. The First Century of Air Conditioning. AmericanSociety of Heating Refrigeration Air-conditioning Engineers,Atlanta, GA.8. Israel, P. 1998. Edison: A Life of Invention. John Wiley andSons, New York, N.Y.9. EEI 1981. Handbook for Electricity Metering, 8th Edition withAppendix, Edition Electric Institute, Washington D.C.10. Miller, R. 1989. Flow Measurement Engineering Handbook.McGraw Hill, New York, N.Y.11. American Institute of Physics, 1975. Efficient Use of Energy:The APS Studies on the Technical Aspects of the More EfficientUse of Energy, American Physical Society, New York, N.Y., (Areport on the 1973 summer study at Princeton University).12. National Geographic, February 1981. Special Report on Energy:Facing up to the Problem, Getting Down to Solutions, NationalGeographic Society, Washington, D.C.13. Scientific American 1971. Energy and Power. W.H. Freemanand Company, San Francisco, CA. (A reprint of eleven articlesthat appeared in the September 1971 Scientific American).14. Kusuda, T. 1999. “Early History and Future Prospects of BuildingSystem Simulation,” Proceedings of the Sixth InternationalBuilding Performance Simulation Association (IBPSA BS’ 99),Kyoto, Japan, (September).15. APEC 1967. HCC-heating/cooling load calculations program.Dayton, Ohio, Automated Procedures for Engineering Consultants.16. Ayres, M., Stamper, E. 1995. “Historical Development of BuildingEnergy Calculations,” ASHRAE Transactions, Vol. 101, Pt.1. American Society of Heating Refrigeration Air-conditioningEngineers, Atlanta, GA.17. Stephenson, D., and Mitalas, G. 1967. “Cooling Load Calculationsby Thermal Response Factor Method,” ASHRAE Transactions,Vol. 73, pt. 1.18. Mitalas, G. and Stephenson, D. 1967, “Room Thermal ResponseFactors,” ASHRAE Transactions, Vol. 73, pt. 2.19. Stoecker, W. 1971. Proposed Procedures for Simulating thePerformance of Components and Systems for Energy Calculations,2 “d Edition, American Society of Heating RefrigerationAir-conditioning Engineers, Atlanta, GA.20. Sepsy, C. 1969. “Energy Requirements for Heating, and CoolingBuildings (ASHRAE RP 66-OS), Ohio State University.21. Socolow, R. 1978. Saving Energy in the Home: Princeton’sExperiments at Twin Rivers, Ballinger Publishing Company,Cambridge, Massachusetts, (This book contains a collection ofpapers that were also published in Energy and Buildings, Vol.1, No. 3., (April)).22. Fels, M. 1986. Special Issue Devoted to Measuring Energy Savings:The Scorekeeping Approach, Energy and Buildings, Vol.9, Nos. 1 & 2, Elsevier Press, Lausanne, Switzerland, (February/May).23. DOE 1985. Proceedings of the DOE/ORNL Data AcquisitionWorkshop, Oak Ridge National Laboratory, Oak Ridge, TN,(October).24. Lyberg, M. 1987. Source Book for Energy Auditors: Vols. 1 &2, International Energy Agency, Stockholm, Sweden, (Reporton IEA Task XI).25. IEA 1990. Field Monitoring For a Purpose. International EnergyAgency Workshop, Chalmers University, Gothenburg, Sweden,(April).26. Oninicomp 1984. Faser Software, Omnicomp, Inc., State College,PA, (monthly accounting software with VBDD capability).27. Eto, J. 1988. “On Using Degree-days to Account for the Effectsof Weather on Annual Energy Use in Office Buildings,” Energyand Buildings, Vol. 12, No. 2, pp. 113-127.28. SRC Systems 1996. Metrix: Utility Accounting System, Berkeley,CA, (monthly accounting software with combined VBDD/multiple regression capabilities).29. Haberl, J. and Vajda. E. 1988. “Use of Metered Data Analysisto Improve Building Operation and Maintenance: Early ResultsFrom Two Federal Complexes,” Proceedings of the ACEEE1988 Summer Study on Energy Efficient Buildings, PacificGrove, CA, pp. 3.98 - 3.111, (August).30. Sonderegger, R. 1977. Dynamic Models of House HeatingBased on Equivalent Thermal Parameters, Ph.D. Thesis, Center

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 751for Energy and Environmental Studies, Report No. 57, PrincetonUniversity.31. DOE 1985. op.cit.32. Lyberg, M. 1987. op.cit.33. IEA 1990. op.cit.34. ASHRAE 199 1. Handbook of HVAC Applications, Chapter37: Building Energy Monitoring, American Society of HeatingRefrigeration Air-conditioning Engineers, Atlanta, GA.35. Haberl, J., and Lopez, R. 1992. “LoanSTAR Monitoring Workbook:Workbook and Software for Monitoring Energy inBuildings,” submitted to the Texas Governor’s Energy Office,Energy Systems Laboratory, Texas A&M University, (August).36. Claridge, D., Haberl, J., O’Neal, D., Heffington, W., Turner, D.,Tombari, C., Roberts, M., Jaeger, S. 1991. “Improving EnergyConservation Retrofits with Measured Savings.” ASHRAEJournal, Volume 33, Number 10, pp. 14-22, (October).37. Fels, M., Kissock, K., Marean, M., and Reynolds, C. 1995.PRISM, Advanced Version 1.0 User’s Guide, Center for Energyand Envirom-nental Studies, Princeton University, Princeton,N.J., (January).38. ASHRAE 1999, I-IVACO 1 Toolkit: A Toolkit for Primary HVACSystem Energy Calculation, ASHRAE Research Project -RP 665,Lebrun, J., Bourdouxhe, J-P, and Grodent, M., American Societyof Heating Refrigeration Air-conditioning Engineers, Atlanta,GA.39. ASHRAE 1993. HVAC02 Toolkit: Algorithms and Subroutinesfor Secondary HVAC System Energy Calculations, ASHRAEResearch Project - 827-RP, Authors: Brandemuehl, M., Gabel,S.,Andresen, American Society of Heating Refrigeration AirconditioningEngineers, Atlanta, GA.40. Brandemuehl, M., Krarti, M., Phelan, J. 1996, “827-RP FinalReport: Methodology Development to Measure In-Situ Chiller,Fan, and Pump Performance,” ASHRAE Research, ASHRAE,Atlanta, GA, (March).41. Haberl, J., Reddy, A., and Elleson, J. 2000a. “Determining Long-Term performance Of Cool Storage Systems From Short-TermTests, Final Report,” submitted to ASHRAE under ResearchProject 1004-RP, Energy Systems Laboratory Report ESL-TR-00/08-01, Texas A&M University, 163 pages, (August).42. Kissock, K., Haberl, J., and Claridge, D. 2001. “Developmentof a Toolkit for Calculating Linear, Changepoint Linear andMultiple-Linear Inverse Building Energy Analysis Models:Final Report,” submitted to ASHRAE under Research Project1050-RP, University of Dayton and Energy Systems Laboratory,(December).43. Abushakra, B., Haberl, J., Claridge, D., and Sreshthaputra, A.2001 “Compilation Of Diversity Factors And Schedules ForEnergy And Cooling Load Calculations; ASHRAE ResearchProject 1093: Final Report,” submitted to ASHRAE underResearch Project 1093-RP, Energy Systems Lab Report ESL-TR-00/06-01, Texas A&M University, 150 pages, (June).44. MacDonald, J. and Wasserman, D. 1989. Investigation ofMetered Data Analysis Methods for Commercial and RelatedBuildings, Oak Ridge National Laboratory Report No. ORNL/CON-279, (May).45. Rabl, A. 1988. “Parameter Estimation in Buildings: Methods forDynamic Analysis of Measured Energy Use,” Journal of SolarEnergy Engineering, Vol. 110, pp. 52-66.46. Rabl, A., Riahle, A. 1992. “Energy Signature Model for CommercialBuildings: Test With Measured Data and Interpretation,”Energy and Buildings, Vol. 19, pp. 143-154.47. Gordon, J.M. and Ng, K.C. 1994. “Thermodynamic Modelingof Reciprocating Chillers,” Journal of Applied Physics, Volume75, No. 6, March 15, 1994, pp. 2769-2774.48. Claridge, D.E., Haberl, J. S., Sparks, R., Lopez, R., Kissock,K. 1992. “Monitored Conu-nercial Building Energy Data: Reportingthe Results.” 1992 ASHRAE Transactions SymposiumPaper, Vol. 98, Part 1, pp. 636-652.49. Sonderegger, R. 1977, op.cit.50. Subbarao, K., Burch, J., Hancock, C.E, 1990. “How to accuratelymeasure the load coefficient of a residential building,” Journalof Solar Energy Engineering, in preparation.51. Reddy, A. 1989. “Application of Dynamic Building InverseModels to Three Occupied Residences Monitored Non-intrusively,”Proceedings of the Thermal Performance of ExteriorEnvelopes of Buildings IV, ASHRAE/DOE/BTECC/CIBSE.52. Shurcliff, W.A. 1984. “Frequency Method of Analyzing aBuilding’s Dynamic Thermal Performance, W.A. Shurcliff, 19Appleton St., Cambridge, MA.53. Dhar, A. 1995, “Development of Fourier Series and ArtificialNeural Networks Approaches to Model Hourly Energy Usein Commercial Buildings,” Ph.D. Dissertation, MechanicalEngineering Department, Texas A&M University, May.54. Miller, R., and Seem, J. 199 1. “Comparison of Artificial NeuralNetworks with Traditional Methods of Predicting Return Timefrom Night Setback,” ASHRAE Transactions, Vol. 97, Pt.2, pp.500-508.55. J.F. Kreider and X.A. Wang, (199 1). “Artificial Neural NetworksDemonstration for Automated Generation of EnergyUse Predictors for Commercial Buildings.” ASHRAE Transactions,Vol. 97, part 1.56. Kreider, J. and Haberl, J. 1994. “Predicting Hourly BuildingEnergy Usage: The Great Energy Predictor Shootout: Overviewand Discussion of Results,” ASHRAE Transactions-Research,Volume 100, Part 2, pp. 1104 - 1118, (June).57. ASHRAE 1997. Handbook of Fundamentals, Chapter 30:Energy Estimating and Modeling Methods, American Societyof Heating Refrigeration Air-conditioning Engineers, Atlanta,GA., p. 30.27 (Copied with perinission).58. ASHRAE 1997. op.cit., p. 30.28 (Copied with permission).59. USDOE 1996. North American Energy Measurement andVerification Protocol (NEMVP), United States Department ofEnergy DOE/EE-0081, (March).60. FEMP 1996. Standard Procedures and Guidelines for Verificationof Energy Savings Obtained Under Federal SavingsPerformance Contracting Programs, USDOE Federal EnergyManagement Program (FEMP).61. Haberl, J., Claridge, D., Turner, D., O’Neal, D., Heffington, W.,Verdict, M. 2002. “LoanSTAR After 11 Years: A Report on theSuccesses and Lessons Learned From the LoanSTAR Program,”Proceedings of the 2nd International Conference for EnhancedBuilding Operation,” Richardson, Texas, pp. 131-138, (October).62. USDOE 1997. International Performance Measurement andVerification Protocol (IPMVP), United States Department ofEnergy DOE/EE-0157, (December).63. USDOE 2001. International Performance Measurement andVerification Protocol (IPMVP): Volume 1: Concepts and Optionsfor Determining Energy and Water Savings, United StatesDepartment of Energy DOE/GO-102001-1187 (January).64. USDOE 2001. International Performance Measurement andVerification Protocol (IPMVP): Volume 11: Concepts andPractices for Improved Indoor Environrnental Quality, UnitedStates Department of Energy DOE/GO-102001-1188 (January).65. USDOE 2003. International Performance Measurement andVerification Protocol (IPMVP): Volume 11: Concepts andPractices for Improved Indoor Environmental Quality, UnitedStates Department of Energy DOE/GO-102001-1188 (January).66. ASHRAE 2002.Guideline 14: Measurement of Energy andDemand Savings, American Society of Heating RefrigerationAir-conditioning Engineers, Atlanta, GA (September).67. Hansen, S. 1993. Performance Contracting for Energy and EnvironmentalSystems, Fairmont Press, Lilbum, GA, pp. 99-100.68. ASHRAE 2002. op.cit., pp. 27-30.

752 ENERGY MANAGEMENT HANDBOOK69. lbid, p. 30 (Copied with permission).70. Brandemuehl et al. 1996. op.cit.71. Wei, G. 1997. “A Methodology for In-situ Calibration of SteamBoiler Instrumentation,” MS Thesis, Mechanical EngineeringDepartment, Texas A&M University, August.72. Dukelow, S.G. 1991. The Control of Boilers. Research TrianglePark, NC: Instrument Society of America.73. Dyer, F.D. and Maples, G. 1981. Boiler Efficiency Improvement.Boiler Efficiency Institute. Auburn: AL.74. Garcia-Borras, T. 1983. Manual for Improving Boiler and FurnacePerformance. Houston, TX: Gulf Publishing Company.75. Aschner, F.S. 1977. Planning Fundamentals of Thermal PowerPlants. Jerusalem, Israel: Israel Universities Press.76. ASME 1974 Performance Test for Steam Units — PTC 4. 1 a1974.77. Babcock and Wilcox. 1992. “Steam: Its generation and Use,”Babcock and Wilcox, Barberton, Ohio, ISBN 0-9634570-0-4.78. Katipamula, S., and Claridge, D. 1992. “Monitored Air HandlerPerformance and Comparison with a Simplified System Model,”ASHRAE Transactions, Vol. 98, Pt 2., pp. 341-35 1.79. Liu, M., and Claridge, D. 1995 “Application of CalibratedHVAC Systems to Identify Component Malfunctions and toOptimize the Operation and Control Schedules,” ASME/JSMEInternational Solar Energy Conference, pp. 209-217.80. ASHRAE 2002. op.cit., p. 144.81. Brandemuehl, et al. 1996. op. cit.82. ASHRAE 2002, op cit., p. 144, (Copied with permission).83. ASHRAE 2002. op.cit., pp. 144-147, (Copied with permission).84. ASHRAE 2002. op.cit., p. 144.85. Ibid, p. 148, (Copied with permission).86. ASHRAE 2002. op.cit., pp. 147-149, (Copied with permission).87. Gordon, J.M. and Ng, K.C. 1994. op.cit.88. Gordon, J.M. and Ng, K.C., 1995. “Predictive and diagnosticaspects of a universal thermodynamic model for chillers,”International Journal of Heat Mass Transfer, 38(5), p.807.89. Gordon, J.M., Ng, K.C., and Chua, H.T., 1995. “Centrifugalchillers: thermodynamic modeling and a diagnostic casestudy,” International Journal of Refrigeration, 18(4), p.253.90. LBL. 1980. DOE-2 User Guide, Ver. 2. 1. Lawrence BerkeleyLaboratory and Los Alamos National Laboratory, Rpt No. LBL-8689 Rev. 2; DOE-2 User Coordination Office, LBL, Berkeley,CA.91. LBL. 198 1. DOE-2 Engineers Manual, Ver. 2. 1 A, LawrenceBerkeley Laboratory and Los Alamos National Laboratory,Rpt No. LBL-1 1353; DOE-2 User Coordination Office, LBL,Berkeley, CA.92. LBL. 1982. DOE-2.1 Reference Manual Rev. 2.1A. LawrenceBerkeley Laboratory and Los Alamos National Laboratory, RptNo. LBL-8706 Rev. 2; DOE-2 User Coordination Office, LBL,Berkeley, CA. LBL.93. LBL. 1989. DOE-2 Supplement, Ver 2. 1 D. Lawrence BerkeleyLaboratory, Rpt No. LBL-8706 Rev. 5 Supplement. DOE-2 UserCoordination Office, LBL, Berkeley, CA.94. Haberl, J. S., Reddy, T. A., Figueroa, I., Medina, M. 1997. “Overviewof LoanSTAR Chiller Monitoring and Analysis of In-SituChiller Diagnostics Using ASHRAE RP827 Test Method,” Proceedingsof the PG&E Cool Sense National Integrated ChillerRetrofit Forum (September).95. According to Gordon et al. 1995, fHx is a dimensionless termthat is norrrially negligible.96. Wei, G. 1997. op.cit.97. Dukelow, S.G. 1991. op.cit.9.8 Dyer, F.D. and Maples, G. 198 1. op.cit.99. Garcia-Borras, T. 1983. op.cit.100. Aschner, F.S. 1977. op.cit.101. ASME 1974. op.cit.102. Babcock and Wilcox. 1992. op.cit.103. Haberl, J., Lynn, B., Underwood, D., Reasoner, J., Rury, K. 2003.op.cit.104. Haberl, et al. 2003. ibid.105. ASME, 1974. Power Test Codes (PTC) 4. la, Steam GeneratingUnits. New York: ASME.106. Stallard, G.S. and Jonas, T.S. 1996. Power Plant Engineering:Combustion Processes. New York: Chapman & Hall.107. Payne, F.W. 1985. Efficient Boiler Operations Sourcebook. Atlanta,GA: The Fairmont Press.108. Wei 1997. op.cit.109. Wei 1997. ibid.110. Aschner 1977. op.cit.111. Babcock and Wilcox 1992. op.cit.112. Dukelow 1991. op.cit.113. Witte, L.C., Schmidt, P.S., and Brown, DR. 1988. IndustrialEnergy Management and Utilization. New York: HemispherePublishing Corporation.114. Aschner 1977. op.cit.115. Thumann, A. 1988. Guide to Improving Efficiency of CombustionSystems. Lilburn, GA: The Fairmont Press.116. Wei 1997. op.cit.117. Garcia-Borras, T. 1983. Manual for Improving Boiler and FurnacePerformance. Houston, TX: Gulf Publishing Company.118. Wei 1997. op.cit.119. ASHRAE 2002, op.cit., pp. 154-156, (Copied with permission).120. Bou Saada, T., Haberl, J., Vajda, J., and Harris, L. 1996. “TotalUtility Savings From the 37,000 Fixture Lighting Retrofit to theUSDOE Forrestal Building,” Proceedings of the 1996 ACEEESummery Study, (August).121. Abushakra, B., Sreshthaputra, A., Haberl, J., and Claridge, D.2001. “Compilation of Diversity Factors and Schedules for Energyand Cooling Load Calculations - Final Report,” submittedto ASHRAE under Research Project 1093-RP, Energy SystemsLab Report ESL-TR-01/04-01, Texas A&M University, (April).122. IESNA 2003. Lighting Handbook, 9h Edition, IlluminatingEngineering Society of North America, New York, N.Y.123. ASHRAE 2002, op.cit., pp. 156-159, (Copied with permission).124. ASHRAE 2002, op. cit., p. 160, (Copied with permission).125. Ayres, M., Stamper, E. 1995, op.cit.126. ASHRAE 1969. Procedures for Determining Heating and CoolingLoads for Computerized Energy Calculations: Algorithmsfor Building Heat Transfer Sub-routines. M. Lokmanhekim,Editor, American Society of Heating Refrigeration Air-conditioningEngineers, Atlanta, GA.127. ASHRAE 1971. Procedures for Simulating the Performance ofComponents and Systems for Energy Calculations. Stoecker,W.F. Stoecker, editor, 2 nd edition, American Society of HeatingRefrigeration Airconditioning Engineers, Atlanta, GA128. BLAST. 1993. BLAST Users Manual. BLAST Support Office,University of Illinois Urbana-Champaign.129. LBL 1980, 1981, 1982, 1989, op.cit.130. Kriebel, D.E., 1983. Simplified Energy Analysis Using theModified Bin Method, American Society of Heating, Refrigeratingand Air-Conditioning Engineers, Inc., Atlanta, Georgia.131. ASHRAE 1999. op.cit.132. ASHRAE 1993. op.cit.133. Yuill, G., K., Haberl, J.S. 2002. Development of Accuracy TestsFor Mechanical System Simulation. Final Report for ASHRAEResearch Project 865-RP, The University of Nebraska at Lincoln,(July).134. Katipamula, S. and Claridge, D.E., 1993. “ Use of SimplifiedSystems Models to Measure Retrofit Savings,” ASME Journalof Solar Energy Engineering, Vol. 115, pp. 57-68, May.135. Liu, M. and Claridge , D.E., 1995. “Application of CalibratedHVAC System Models to Identify Component Malfunctions

MEASUREMENT AND VERIFICATION OF ENERGY SAVINGS 753and to Optimize the Operation and Control Schedules,” SolarEngineering 1995, W.B. Stine, T. Tanaka and D.E. Claridge(Eds.), ASME/JSME/JSES International Solar Energy Conference,Maui, Hawaii, March.136. Liu, M. and Claridge, D.E., 1998. “Use of Calibrated HVACSystem Models to Optimize System Operation,” Journal ofSolar Energy Engineering, May 1998, Vol. 120.137. Liu, M., Wei, G., Claridge, D., E., 1998, “Calibrating AHUModels Using Whole Building Cooling and Heating EnergyConsumption Data,” Proceedings of 1998 ACEEE SurrurierStudy on Energy Efficiency in Buildings. Vol. 3.138. Haberl, J., Claridge, D., Turner. D. 2000b. “Workshop onEnergy Measurement, Verification and Analysis Technology,”Energy Conservation Task Force, Federal Reserve Bank, Dallas,Texas (April).139. This table contains material adapted from proposed HVACSystem Testing Methods for ASHRAE Guideline 14-2002,which were not included in the published ASHRAE Guideline14-2002.140. Haberl et al. 2000b, op. cit.141. Kissock et al. 2001. op.cit.142. Fels 1986. op.cit.143. Rabl 1988. op.cit.144. Rabl and Raihle 1992. op.cit.145. Claridge et al. 1992. op.cit.146. Reddy, T.A., Haberl, J.S., Saman, N.F., Turner, W.D., Claridge,D.E., Chalifoux, A.T. 1997. “Baselining Methodology for Facility-LevelMonthly Energy Use - Part 1: Theoretical Aspects,”ASHRAE Transactions-Research, Volume 103, Part 2, pp. 336-347, (June).147. Reddy, T.A., Haberl, J.S., Saman, N.F., Turner, W.D., Claridge,D.E., Chalifoux, A.T. 1997 “Baselining Methodology for Facility-LevelMonthly Energy Use - Part 2: Application to EightArmy Installations,” ASHRAE Transactions-Research, Volume103, Part 2, pp. 348-359, (June).148. Haberl, J., Thamilseran, S., Reddy, A., Claridge, D., O’Neal, D.,Turner, D. 1998. “Baseline Calculations for Measuring and Verificationof Energy and Demand Savings in a Revolving LoanProgram in Texas,” ASHRAE Transactions-Research, Volume104, Part 2, pp. 841-858, (June).149. Turner, D., Claridge, D., O’Neal, D., Haberl, J., Heffington,W., Taylor, D., Sifuentes, T. 2000. “Program Overview: TheTexas LoanSTAR Program: 1989 - 1999 A 10-year Experience,”Proceedings of the 2000 ACEEE Summery Study on EnergyEfficiency in Buildings, Volume 4, pp. 4.365-4.376, (August).150. Haberl, J., Sreshthatputra, A., Claridge, D., Turner, D. 2001.“Measured Energy Indices for 27 Office Buildings,” Proceedingsof the 1st International Conference for Enhanced BuildingOperation,” Austin, Texas, pp. 185-200, (July).151. Beasley, R., Haberl, J. 2002. “Development of a Methodologyfor Baselining The Energy Use of Large Multi-building CentralPlants,” ASHRAE Transactions-Research, Volume 108, Part 1,pp. 251-259, (January).152. ASHRAE 2002. op.cit. p. 25, (Copied with permission).153. Haberl et al. 2000b, op. cit.154. Temperatures below zero are calculated as positive increasesaway from the change point temperature.155. Thamilseran, S., Haberl, J. 1995. “A Bin Method for CalculatingEnergy Conservation Retrofits Savings in Commercial Buildings,”Proceedings of the 1995 ASME/JSME/JSES InternationalSolar Energy Conference, Lahaina, Maui, Hawaii, pp. 111- 124(March). .156. Thamilseran, S. 1999. “An Inverse Bin Methodology to Measurethe Savings from Energy Conservation Retrofits in CommercialBuildings,” Ph.D. Thesis, Mechanical Engineering Department,Texas A&M University, (May).157. Kissock et al. 2001. op.cit.158. Kissock, J.K, Xun,W., Sparks, R., Claridge, D., Mahoney, J. andHaberl, J., 1994. “EModel Version 1.4de,” Texas A&M University,Energy Systems Laboratory, Department of MechanicalEngineering, Texas A&M University, College Station, TX, December.159. Abushakra et al. 2001. op.cit.160. Haberl, J., Bou-Saada, T. 1998. “Procedures for CalibratingHourly Simulation Models to Measured Building Energy andEnvironmental Data,” ASME Journal of Solar Energy Engineering,Volume 120, pp. 193-204, (August).161. Clarke, J.A, Strachan, P.A. and Pemot, C.. 1993. An Approachto the Calibration of Building Energy Simulation Models.ASHRAE Transactions. 99(2): 917-927.162. Diamond, S.C. and Hunn, B.D.. 1981. Comparison of DOE-2Computer Program Simulations to Metered Data for SevenCommercial Buildings. ASHRAE Transactions. 87(l): 1222-1231.163. Haberl, J., Bronson, D., Hinchey, S. and O’Neal, D. 1993.“Graphical Tools to Help Calibrate the DOE-2 SimulationProgram to Non-weather Dependent Measured Loads,” 1993ASHRAE Journal, Vol. 35, No. 1, pp. 27-32, (January).164. Haberl, J., Bronson, D. and O’Neal, D. 1995. “An Evaluationof the Impact of Using Measured Weather Data Versus TMYWeather Data in a DOE-2 Simulation of an Existing Buildingin Central Texas.” ASHRAE Transactions Technical Paper no.3921, Vol. 101, Pt. 2, (June).165. Hinchey, S.B. 1991. Influence of Thermal Zone Assumptions onDOE-2 Energy Use Estimations of a Commercial Building. M.S.Thesis, Energy Systems Report No. ESL-TH-91/09-06, TexasA&M University, College Station, TX166. Hsieh, E.S. 1988, Calibrated Computer Models of CommercialBuildings and Their Role in Building Design and Operation.M.S. Thesis, PU/CEES Report No. 230, Princeton University,Princeton, NJ.167. Hunn, B.D., Banks, J.A. and Reddy, S.N. 1992. Energy Analysisof the Texas Capitol Restoration. The DOE-2 User News. 13 (4):2- 10.168. Kaplan, M.B., Jones, B. and Jansen, J. 1990a. DOE-2. I C ModelCalibration with Monitored End-use Data. Proceedings fromthe ACEEE 1990 Summer Study on Energy Efficiency in Buildings,Vol. 10, pp. 10. 11510.125.169. Kaplan, M.B., Caner, P. and Vincent, G.W. 1992. Guidelinesfor Energy Simulation of Commercial Buildings. Proceedingsfrom the ACEEE 1992 Summer Study on Energy Efficiency inBuildings, Vol. 1, pp. 1.137-1.147.170. Katipamula, S. and Claridge, D.E., 1993. “ Use of SimplifiedSystems Models to Measure Retrofit Savings,” ASME Journalof Solar Energy Engineering, Vol. 115, pp.57-68, May.171. Liu, M. and Claridge, D.E., 1995. “Application of CalibratedHVAC System Models to Identify Component Malfunctionsand to Optimize the Operation and Control Schedules,” SolarEngineering 1995, W.B. Stine, T. Tanaka and D.E. Claridge(Eds.), ASME/JSME/JSES International Solar Energy Conference,Maui, Hawaii, March.172. Liu, M. and Claridge, D. E., 1998. “Use of Calibrated HVACSystem Models to Optimize System Operation,” Journal ofSolar Energy Engineering, May 1998, Vol. 120.173. Liu, M., Wei, G., Claridge, D. E., 1998, “Calibrating AHUModels Using Whole Building Cooling and Heating EnergyConsumption Data,” Proceedings of 1998 ACEEE SummerStudy on Energy Efficiency in Buildings. Vol. 3.174. Manke, J., Hittle, D. and Hancock 1996. “Calibrating BuildingEnergy Analysis Models Using Short Term Test Data,” Proceedingsof the 1996 International ASME Solar Energy Conference,p. 369, San Antonio, TX175. McLain, H.A., Leigh, S.B., and MacDonald, J.M1993. Analysisof Savings Due to Multiple Energy Retrofits in a Large OfficeBuilding. Oak Ridge National Laboratory, ORNL Report No.ORNL/CON-363, Oak Ridge, TN.

754 ENERGY MANAGEMENT HANDBOOK176. Haberl et al. 2000b, op. cit.177. ASHRAE 2002. op.cit. p. 35-43, (Copied with permission).178. Bou-Saada, T. 1994. An Improved Procedure for Developing ACalibrated Hourly Simulation Model of an Electrically Heatedand Cooled Commecial Building, Master’s Thesis, MechanicalEngineering Department, Texas A&M University, (December),p. 54.179. Sylvester, K., Song, S., Haberl, J., and Turner, D. 2002. CaseStudy: Energy Savings Assessment for the Robert E. JohnsonState Office Building in Austin, Texas,” IBPSA Newsletter, Vol.12, Number 2, pp. 22-28, (Summer).180. Huang & Associates. 1993. DrawBDL user’s guide. 6720Potrero Ave., El Cerrito, California, 94530.181. Bou Saada, T. 1994. “An Improved Procedure for DevelopingA Calibrated Hourly Simulation Model of an ElectricallyHeated and Cooled Commercial Building,” Master’s Thesis,Mechanical Engineering Department, Texas A&M University,(December), p. 150.182. Bou-Saada 1994. op. cit. p. 144.183. C. Culp, K. Q. Hart, B. Turner, S. Berry-Lewis, 2003. “CostEffective Measurement and Verification at Fairchild AFB,International Conference on Enhance Building Operation,”Energy Systems Laboratory Report, Texas A&M University,(October).184. Culp et al. 2003, ibid.185. ASME 1974 op cit.

CHAPTER 28GROUND-SOURCE HEAT PUMPSAPPLIED TO COMMERCIAL BUILDINGSSTEVEN A. PARKER, P.E., C.E.M.DONALD L. HADLEYEnergy Science and Technology DirectoratePacific Northwest National Laboratory 1Richland, WA28.1 ABSTRACTGround-source heat pumps can provide an energy-efficient,cost-effective way to heat and cool commercialfacilities. While ground-source heat pumps are wellestablished in the residential sector, their applicationin larger, commercial-style, facilities is lagging, in partbecause of limited experience with the technology bythose in decision-making positions. Through the use ofa ground-coupling system, a conventional water-sourceheat pump design is transformed to a unique means ofutilizing thermodynamic properties of earth and groundwaterfor efficient operation throughout the year in mostclimates. In essence, the ground (or groundwater) servesas a heat source during winter operation and a heat sinkfor summer cooling. Many varieties in design are available,so the technology can be adapted to almost any site.Ground-source heat pump systems can be used widelyin commercial-building applications and, with properinstallation, offer great potential for the commercial sector,where increased efficiency and reduced heating andcooling costs are important. Ground-source heat pumpsystems require less refrigerant than conventional airsourceheat pumps or air-conditioning systems, with theexception of direct-expansion-type ground-source heatpump systems.Installation costs are relatively high but are offsetby low maintenance and operating expenses and efficientenergy use. The greatest barrier to effective use is improperdesign and installation; well-trained, experienced,and responsible designers and installers is of critical importance.1 Pacific Northwest National Laboratory is operated for the U.S.Department of Energy by Battelle Memorial Institute under contractDE-AC05-76RL01830.This chapter provides information and proceduresthat an energy manager can use to evaluate most groundsourceheat pump applications. Ground-source heatpump operation, system types, design variations, energysavings, and other benefits are explained. Guidelines areprovided for appropriate application and installation.Two case studies are presented to give the reader a senseof the actual costs and energy savings. A list of manufacturersand references for further reading are included forprospective users who have specific or highly technicalquestions not fully addressed in this chapter. Sample casespreadsheets are also provided.28.2 BACKGROUNDThis chapter is based on a Federal Technology Alertsponsored by the U.S. Department of Energy (DOE), FederalEnergy Management Program (FEMP). The originalFederal Technology Alert was published in 1995 andupdated in 2001. The material was updated in 2005 todevelop this chapter.28.2.1 The DOE Federal Energy Management ProgramThe federal government is the largest energy consumerin the nation. Annually, in its 500,000 buildingsand 8,000 locations worldwide, it uses nearly 1.4 quadrillionBtu (quads) of energy, costing approximately $8billion. This represents 1.5% of all primary energy consumptionin the United States 2 . The DOE Federal EnergyManagement Program was established in 1974 to providedirection, guidance, and assistance to federal agencies inplanning and implementing energy management programsthat will improve the energy efficiency and fuelflexibility of the federal infrastructure.Over the years several federal laws and ExecutiveOrders have shaped FEMP’s mission. These include theEnergy Policy and Conservation Act of 1975; the NationalEnergy Conservation and Policy Act of 1978; the Federal2 DOE. 2001. Annual Report to Congress on Federal Government EnergyManagement and Conservation programs, Fiscal Year 1999. DOE/EE-0252. U.S. Department of Energy. Washington, DC.755

756 ENERGY MANAGEMENT HANDBOOKEnergy Management Improvement Act of 1988; ExecutiveOrder 12759 in 1991; the National Energy Policy Actof 1992 (EPAct 1992); Executive Order 12902 in 1994; ExecutiveOrder 13123 in 1999; and the Energy Policy Act of2005 (EPAct 2005).The DOE Federal Energy Management Program iscurrently involved in a wide range of energy-assessmentactivities, including conducting new technology demonstrations,to hasten the penetration of energy-efficienttechnologies into the federal marketplace.28.2.2 The FEMP New TechnologyDemonstrations ActivityThe Energy Policy Act of 1992, and subsequent ExecutiveOrders, mandated that energy consumption infederal buildings be reduced by 35% from 1985 levels bythe year 2010. The Energy Policy Act of 2005 calls for evenmore energy reduction. To achieve this goal, the DOEFederal Energy Management Program sponsors a seriesof program activities to reduce energy consumption atfederal installations nationwide. One of these programactivities, new technology demonstrations, is tasked toaccelerate the introduction of energy-efficient and renewabletechnologies into the federal sector and to improvethe rate of technology transfer.In addition to technology demonstrations, FEMPsponsors a series of publications that are designed to disseminateinformation on new and emerging technologies.These publications include:Federal Technology Alerts—longer summary reportsthat provide details on energy-efficient, water-conserving,and renewable-energy technologies that havebeen selected for further study for possible implementationin the Federal sector. Additional information on FederalTechnology Alerts is provided below.Technology Installation Reviews— concise reportsdescribing a new technology and providing case studyresults, typically from another demonstration program orpilot project.Technology Focuses—brief information on new,energy-efficient, environmentally friendly technologiesof potential interest to the Federal sector.28.2.3 More on Federal Technology AlertsFederal Technology Alerts provide summary informationon candidate energy-saving technologiesdeveloped and manufactured in the United States. Thetechnologies featured in the Federal Technology Alertshave already entered the market and have some experiencebut are not in general use in the Federal sector.The goal of the Federal Technology Alerts is to improvethe rate of technology transfer of new energy-savingtechnologies within the Federal sector by providingthe right people in the field with accurate, up-to-dateinformation on the new technologies so that they canmake informed decisions on whether the technologiesare suitable for their sites. The information in the FederalTechnology Alerts typically includes a description of thecandidate technology; a description of its performance,applications and field experience to date; a list of manufacturers;and important sources for additional information.Appendixes provide supplemental information andexample worksheets for the technology.FEMP sponsors publication of the Federal TechnologyAlerts to facilitate information sharing between manufacturersand government staff. While the technologyfeatured promises potential Federal sector energy savings,the Federal Technology Alerts do not constitute FEMP’sendorsement of a particular product, because FEMP hasnot independently verified performance data provided bymanufacturers. Readers should note the publication dateand consider the Federal Technology Alerts as an accuratepicture of the technology and its performance at thetime of publication. Product innovations and the entranceof new manufacturers or suppliers should be anticipatedsince the date of publication. FEMP encourages interestedenergy and facility managers to contact the manufacturersand other sites directly, and to use the worksheets inthe Federal Technology Alerts to aid in their purchasingdecisions.28.3 INTRODUCTION TOGROUND-SOURCE HEAT PUMPSGround-source heat pumps are known by a varietyof names: geoexchange heat pumps, ground-coupledheat pumps, geothermal heat pumps, earth-coupled heatpumps, ground-source systems, groundwater source heatpumps, well water heat pumps, solar energy heat pumps,and a few other variations. Some names are used to describemore accurately the specific application; however,most are the result of marketing efforts and the need to associate(or disassociate) the heat pump systems from othersystems. This chapter refers to them as ground-source heatpumps except when it is necessary to distinguish a specificdesign or application of the technology. A typical groundsourceheat pump system design applied to a commercialfacility is illustrated in Figure 28.1.It is important to remember that the primary equipmentused for ground-source heat pumps are watersourceheat pumps. What makes a ground-source heatpump different (unique, efficient, and usually moreexpensive to install) is the ground-coupling system. In

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 757Figure 28.1. Typical ground-source heat pump system applied to a commercial facilityaddition, most manufacturers have developed extendedrangewater-source heat pumps for use as ground-sourceheat pumps. 3A conventionally designed water-source heat pumpsystem would incorporate a boiler as a heat source duringthe winter heating operation and a cooling tower to rejectheat (heat sink) during the summer cooling operation.This system type is also sometimes called a boiler/towerwater-loop heat pump system. The water loop circulatesto all the water-source heat pumps connected to the system.The boiler (for winter operation) and the coolingtower (for summer operation) provide a fairly constantwater-loop temperature, which allows the water-sourceheat pumps to operate at high efficiency.A conventional air-source heat pump uses theoutdoor ambient air as a heat source during the winterheating operation and as a heat sink during the summercooling operation. Air-source heat pumps are subject tohigher temperature fluctuations of the heat source andheat sink. They become much less effective—and lessefficient—at extreme ambient air temperatures. This isparticularly true at low temperatures. In addition, heattransfer using air as a transfer medium is not as effectiveas water systems because of air’s lower thermal mass.3 The extended-range designation is important. Conventional watersourceheat pumps are designed to operate with a water-loop as a heatsink that maintains a narrow temperature range. Ground-source heatpumps, however, are typically required to operate with a water-loopheat sink under a wider range of temperatures.A ground-source heat pump uses the ground (or insome cases groundwater) as the heat source during thewinter heating operation and as the heat sink during thesummer cooling operation. Ground-source heat pumpsmay be subject to higher temperature fluctuations thanconventional water-source heat pumps but not as highas air-source heat pumps. Consequently, most manufacturershave developed extended-range systems. Theextended-range systems operate more efficiently whilesubject to the extended-temperature range of the waterloop. Like water-source heat pumps, ground-source heatpumps use a water loop between the heat pumps and theheat source/heat sink (the earth). The primary exceptionis the direct-expansion ground-source heat pump, whichis described in more detail later in this chapter.Ground-source heat pumps take advantage of thethermodynamic properties of the earth and groundwater.Temperatures below the ground surface do not fluctuatesignificantly through the day or the year as do ambient airtemperatures. Ground temperatures a few feet below thesurface stay relatively constant throughout the year. Forthis reason, ground-source heat pumps remain extremelyefficient throughout the year in virtually any climate.28.4 ABOUT THE TECHNOLOGYIn 1999, an estimated 400,000 ground-source heatpumps were operating in residential and commercial ap-

758 ENERGY MANAGEMENT HANDBOOKplications, up from 100,000 in 1990. In 1985, it was estimatedthat only around 14,000 ground-source heat pumpsystems were installed in the United States. Annual salesof approximately 45,000 units were reported in 1997. Witha projected annual growth rate of 10%, 120,000 new unitswould be installed in 2010, for a total of 1.5 million unitsin 2010 (Lund and Boyd 2000). In Europe, the estimatedtotal number of installed ground-source heat pumps atthe end of 1998 was 100,000 to 120,000 (Rybach and Sanner2000). Nearly 10,000 ground-source heat pumps havebeen installed in U.S. Federal buildings, over 400 schoolsand thousands of low-income houses and apartments(ORNL/SERDP, no date).Although ground-source heat pumps are usedthroughout the United States, the majority of new groundsourceheat pump installations in the United States are inthe southern and mid-western states (from North Dakotato Florida). Oklahoma, Texas, and the East Coast havebeen particularly active with new ground-source heatpump installations. Environmental concerns, particularlyfrom the potential for groundwater contamination witha leaking ground loop, and a general lack of understandingof the technology by HVAC companies and installershave limited installations in the West (Lund and Boyd2000). Usually the technology does well in an area whereit has been actively promoted by a local utility or themanufacturer.Ground-source heat pumps are not a new idea.Patents on the technology date back to 1912 in Switzerland(Calm 1987). One of the oldest ground-source heatpump systems, in the United Illuminating headquartersbuilding in New Haven, Connecticut, has been operatingsince the 1930s (Pratsch 1990). Although ground-sourceheat pump systems are probably better established todayin rural and suburban residential areas because of theland area available for the ground loop, the market hasexpanded to urban and commercial applications.The vast majority of ground-source heat pumpinstallations utilize unitary equipment consisting ofmultiple water-source heat pumps connected to a commonground-coupled loop. Most individual units rangefrom 1 to 10 tons (3.5 to 35.2 kW), but some equipmentis available in sizes up to 50 tons (176 kW). Large-tonnagecommercial systems are achieved by using severalunitary water-source heat pumps, each responsible for anindividual control zone.One of the largest commercial ground-source heatpump systems is at Stockton College in Pomona, NewJersey, where 63 ground-source heat pumps totaling 1,655tons (5,825 kW) are connected to a ground-coupled loopconsisting of 400 wells, each 425 feet (129 m) deep (Gahran1993).Public schools are another good application forthe ground-source heat pump technology with over 400installations nationwide. In 1995, the Lincoln, Nebraska,Public School District built four new 70,000 square footelementary schools. Space conditioning loads are met by54 ground-source heat pumps ranging in size from 1.4to 15 tons, with a total cooling capacity of 204 tons. Gasfiredboilers provided hot water for pre-heating of theoutside air and for terminal re-heating. Compared withother similar new schools, these four ground-source heatpump conditioned facilities used approximately 26%less source energy per square foot of floor area (Shonderet al. 1999).Multiple unitary systems are not the only arrangementsuitable for large commercial applications. It is alsopossible to design large centralized heat-pump systemconsisting of reciprocating and centrifugal compressors(up to 19.5 million Btu/h) and to use these systems tosupport central-air-handling units, variable air-volumesystems, or distributive two-pipe fan coil units.28.4.1 How the Technology WorksHeat normally flows from a warmer medium to acolder one. This basic physical law can only be reversedwith the addition of energy. A heat pump is a device thatdoes so by essentially “pumping” heat up the temperaturescale, then transferring it from a cold material to awarmer one by adding energy, usually in the form of electricity.A heat pump functions by using a refrigerant cyclesimilar to the household refrigerator. In the heating mode,a heat pump removes the heat from a low temperaturesource, such as the ground or air, and supplies that heatto a higher temperature sink, such as the heated interiorof a building. In the cooling mode, the process is reversedand the heat is extracted from the cooler inside air andrejected to the warmer outdoor air or other heat sink. Forspace conditioning of buildings, heat pumps that removeheat from outdoor air in the heating mode and reject itto outdoor air in the cooling mode are common. Theseare normally called air-source or air-to-air heat pumps.Air-source heat pumps have the disadvantage that thegreatest requirement for building heating or cooling isnecessarily coincident with the times when the outdoorair is least effective as a heat source or sink. Below about37ºF (2.8°C), supplemental heating is required to meetthe heating load. For this reason, air-source heat pumpsare essentially unfeasible in cold climates with outdoortemperatures below 37ºF (2.8°C) for extended periods oftime.The efficiency of any heat pump is inversely proportionalto the temperature difference between the conditionedspace and the heat source (heating mode) or heat

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 759sink (cooling mode) as can be easily shown by a simplethermodynamic analysis (Reynolds and Perkins 1977).For this reason, air-source heat pumps are less efficientand have a lower heating capacity in the heating modeat low outdoor air temperatures. Conversely, air-sourceheat pumps are also less efficient and have a lower coolingcapacity in the cooling mode at high outdoor air temperatures.Ground-source heat pumps, however, are notimpacted directly by outdoor air temperatures. Groundsourceheat pumps use the ground, groundwater, orsurface water, which are more thermally stable and notsubjected to large annual swings of temperature, as a heatsource or sink.28.4.2 Other benefitsThe primary benefit of ground-source heat pumpsis the increase in operating efficiency, which translatesto a reduction in heating and cooling costs, but thereare additional advantages. One notable benefit is thatground-source heat pumps, although electrically driven,are classified as a renewable-energy technology. The justificationfor this classification is that the ground acts as aneffective collector of solar energy. The renewable-energyclassification can affect federal goals and potential federalfunding opportunities.An environmental benefit is that ground-source heatpumps typically use 25% less refrigerant than split-systemair-source heat pumps or air-conditioning systems.Ground-source heat pumps generally do not require tamperingwith the refrigerant during installation. Systems aregenerally sealed at the factory, reducing the potential forleaking refrigerant in the field during assembly.Ground-source heat pumps also require less spacethan conventional heating and cooling systems. Whilethe requirements for the indoor unit are about the sameas conventional systems, the exterior system (the groundcoil) is underground, and there are no space requirementsfor cooling towers or air-cooled condensers. In addition,the ground-coupling system does not necessarily limitfuture use of the land area over the ground loop, withthe exception of siting a building. Interior space requirementsare also reduced. There are no floor space requirementsfor boilers or furnaces, just the unitary systemsand circulation pumps. Furthermore, many distributedground-source heat pump systems are designed to fit inceiling plenums, reducing the floor space requirement ofcentral mechanical rooms.Compared with air-source heat pumps that use outdoorair coils, ground-source heat pumps do not requiredefrost cycles or crankcase heaters and there is virtuallyno concern for coil freezing. Cooling tower systems requireelectric resistance heaters to prevent freezing in thetower basin, also not necessary with ground-source heatpumps.It is generally accepted that maintenance requirementsare also reduced, although research continuesdirected toward verifying this claim. It is clear, however,that ground-source heat pumps eliminate the exterior fincoilcondensers of air-cooled refrigeration systems andeliminate the need for cooling towers and their associatedmaintenance and chemical requirements. This is a primarybenefit cited by facilities in highly corrosive areassuch as near the ocean, where salt spray can significantlyreduce outdoor equipment life.Ground-source heat pump technology offers furtherbenefits: less need for supplemental resistance heaters, noexterior coil freezing (requiring defrost cycles) such as thatassociated with air-source heat pumps, improved comfortduring the heating season (compared with air-source heatpumps, the supply air temperature does not drop whenrecovering from the defrost cycle), significantly reducedfire hazard over that associated with fossil fuel-fired systems,reduced space requirements and hazards by eliminatingfossil-fuel storage, and reduced local emissionsfrom those associated with other fossil fuel-fired heatingsystems.Another benefit is quieter operation, becauseground-source heat pumps have no outside air fans. Finally,ground-source heat pumps are reliable and longlived,because the heat pumps are generally installed inclimate-controlled environments and therefore are notsubject to the stresses of extreme temperatures. Becauseof the materials and joining techniques, the ground-couplingsystems are also typically reliable and long-lived.For these reasons, ground-source heat pumps are expectedto have a longer life and require less maintenance thanalternative (more conventional) technologies.28.4.3 Ground-Coupled System TypesThe ground-coupling systems used in groundsourceheat pumps fall under three main categories:closed-loop, open-loop and direct-expansion. These areillustrated in Figure 28.2 and discussed in the followingsections. The type of ground coupling employed will affectheat pump system performance (therefore the heatpump energy consumption), auxiliary pumping energyrequirements, and installation costs. Choice of the mostappropriate type of ground coupling for a site is usuallya function of specific geography, available land area, andlife-cycle cost economics.Closed-loop SystemsClosed-loop systems consist of an underground networkof sealed, high-strength plastic pipe 4 acting as a heat

760 ENERGY MANAGEMENT HANDBOOKexchanger. The loop is filled with a heat transferfluid, typically water or a water-antifreeze 5 solution,although other heat transfer fluids maybe used. 6 When cooling requirements cause theclosed-loop liquid temperature to rise, heat istransferred to the cooler earth. Conversely, whenheating requirements cause the closed-loop fluidtemperature to drop, heat is absorbed from thewarmer earth. Closed-loop systems use pumpsto circulate the heat transfer fluid between theheat pump and the ground loop. Because theloops are closed and sealed, the heat pump heatexchanger is not subject to mineral buildupand there is no direct interaction (mixing) withgroundwater.There are several varieties of closed-loopconfigurations including horizontal, spiral, vertical,and submerged.Horizontal LoopsHorizontal loops, illustrated in Figure 28.2a,are often considered when adequate land surfaceis available. The pipes are placed in trenches,typically at a depth of 4 to 10 feet (1.2 to 3.0 m).Depending on the specific design, from one to sixpipes may be installed in each trench. Althoughrequiring more linear feet of pipe, multiple-pipeconfigurations conserve land space, require lesstrenching, and therefore frequently cost less toinstall than single-pipe configurations. Trenchlengths can range from 100 to 400 feet per systemcooling ton (8.7 to 34.6 m/kW), depending on soil characteristicsand moisture content, and the number of pipes inthe trench. Trenches are usually spaced from 6 to 12 feet(1.8 to 3.7 m) apart.These systems are common in residential applicationsbut are not frequently applied to large-tonnage commercialapplications because of the significant land arearequired for adequate heat transfer. The horizontal-loopsystems can be buried beneath lawns, landscaping, andparking lots. Horizontal systems tend to be more popularwhere there is ample land area with a high water table.4 Acceptable piping includes high quality polyethylene or polybutylene.PVC is not acceptable in either heat transfer characteristics or strength.5 Common heat transfer fluids include water or water mixed with anantifreeze, such as: sodium chloride, calcium chloride, potassiumcarbonate, potassium acetate, ethylene glycol, propylene gycol, methylalcohol, or ethyl alcohol.6 Note that various heat transfer fluids have different densities andthermodynamic properties. Therefore, the heat transfer fluid selectedwill affect the required pumping power and the amount of heat transferpipe. Furthermore, some local regulations may limit the selection anduse of certain antifreeze solutions.Figure 28.2. Ground-coupling system types• Advantages: Trenching costs typically lower thanwell-drilling costs; flexible installation options.• Disadvantages: Large ground area required; groundtemperature subject to seasonal variance at shallowdepths; thermal properties of soil fluctuate withseason, rainfall, and burial depth; soil dryness mustbe properly accounted for in designing the requiredpipe length, especially in sandy soils and on hilltopsthat may dry out during the summer; pipe systemcould be damaged during backfill process; longerpipe lengths are required than for vertical wells;antifreeze solution viscosity increases pumpingenergy, decreases the heat transfer rate, and thus reducesoverall efficiency; lower system efficiencies.Spiral LoopsA variation on the multiple pipe horizontal-loopconfiguration is the spiral loop, commonly referred to asthe “slinky.” The spiral loop, illustrated in Figure 28.2b,consists of pipe unrolled in circular loops in trenches; thehorizontal configuration is shown.

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 761Another variation of the spiral-loop system involvesplacing the loops upright in narrow vertical trenches. Thespiral-loop configuration generally requires more piping,typically 500 to 1,000 feet per system cooling ton (43.3to 86.6 m/kW) but less total trenching than the multiplehorizontal-loop systems described above. For the horizontalspiral-loop layout, trenches are generally 3 to 6 feet(0.9 to 1.8 m) wide; multiple trenches are typically spacedabout 12 feet (3.7 m) apart. For the vertical spiral-looplayout, trenches are generally 6 inches (15.2 cm) wide;the pipe loops stand vertically in the narrow trenches. Incases where trenching is a large component of the overallinstallation costs, spiral-loop systems are a means ofreducing the installation cost. As noted with horizontalsystems, slinky systems are also generally associated withlower-tonnage systems where land area requirements arenot a limiting factor.• Advantages: Requires less ground area and lesstrenching than other horizontal loop designs; installationcosts sometimes less than other horizontalloop designs.• Disadvantages: Requires more total pipe lengththan other ground-coupled designs; relatively largeground area required; ground temperature subject toseasonal variance; larger pumping energy requirementsthan other horizontal loops defined above;backfilling the trench can be difficult with certainsoil types and the pipe system could be damagedduring backfill process.Vertical LoopsVertical loops, illustrated in Figure 28.2c, are generallyconsidered when land surface is limited. Wells arebored to depths that typically range from 75 to 300 feet(22.9 to 91.4 m) deep. The closed-loop pipes are insertedinto the vertical well. Typical piping requirements rangefrom 200 to 600 feet per system cooling ton (17.4 to 52.2m/kW), depending on soil and temperature conditions.Multiple wells are typically required with well spacingnot less than 15 feet (4.6 m) in the northern climatesand not less than 20 feet (6.1 m) in southern climates toachieve the total heat transfer requirements. A 300- to 500-ton capacity system can be installed on one acre of land,depending on soil conditions and ground temperature.There are three basic types of vertical-system heatexchangers: U-tube, divided-tube, and concentric-tube(pipe-in-pipe) system configurations.• Advantages: Requires less total pipe length thanmost closed-loop designs; requires the least pumpingenergy of closed-loop systems; requires leastamount of surface ground area; ground temperaturetypically not subject to seasonal variation.• Disadvantage: Requires drilling equipment; drillingcosts frequently higher than horizontal trenchingcosts; some potential for long-term heat buildupunderground with inadequately spaced bore holes.Submerged LoopsIf a moderately sized pond or lake is available, theclosed-loop piping system can be submerged, as illustratedin Figure 28.2d. Some companies have installedponds on facility grounds to act as ground-coupledsystems; ponds also serve to improve facility aesthetics.Submerged-loop applications require some special considerations,and it is best to discuss these directly withan engineer experienced in the design applications. Thistype of system requires adequate surface area and depthto function adequately in response to heating or coolingrequirements under local weather conditions. In general,the submerged piping system is installed in loops attachedto concrete anchors. Typical installations requirearound 300 feet of heat transfer piping per system coolington (26.0 m/kW) and around 3,000 square feet of pondsurface area per ton (79.2 m 2 /kW) with a recommendedminimum one-half acre total surface area. The concreteanchors act to secure the piping, restricting movement,but also hold the piping 9 to 18 inches (22.9 to 45.7 cm)above the pond floor, allowing for good convective flowof water around the heat transfer surface area. It is alsorecommended that the heat-transfer loops be at least 6to 8 feet (1.8 to 2.4 m) below the pond surface, preferablydeeper. This maintains adequate thermal masseven in times of extended drought or other low-waterconditions. Rivers are typically not used because theyare subject to drought and flooding, both of which maydamage the system.• Advantages: Can require the least total pipe length ofclosed-loop designs; can be less expensive than otherclosed-loop designs if body of water available.• Disadvantage: Requires a large body of water andmay restrict lake use (i.e., boat anchors).Open-Loop SystemsOpen-loop systems use local groundwater or surfacewater (i.e., lakes) as a direct heat transfer mediuminstead of the heat transfer fluid described for the closedloopsystems. These systems are sometimes referred tospecifically as “ground-water-source heat pumps” to

762 ENERGY MANAGEMENT HANDBOOKdistinguish them from other ground-source heat pumps.Open-loop systems consist primarily of extraction wells,extraction and reinjection wells, or surface water systems.These three types are illustrated in Figures 28.2e, 28.2f,and 28.2g, respectively.A variation on the extraction well system is thestanding column well. This system reinjects the majorityof the return water back into the source well, minimizingthe need for a reinjection well and the amount of surfacedischarge water.There are several special factors to consider inopen-loop systems. One major factor is water quality. Inopen-loop systems, the primary heat exchanger betweenthe refrigerant and the groundwater is subject to fouling,corrosion, and blockage. A second major factor is the adequacyof available water. The required flow rate throughthe primary heat exchanger between the refrigerant andthe groundwater is typically between 1.5 and 3.0 gallonsper minute per system cooling ton (0.027 and 0.054 L/skW).This can add up to a significant amount of water andcan be affected by local water resource regulations. A thirdmajor factor is what to do with the discharge stream. Thegroundwater must either be re-injected into the groundby separate wells or discharged to a surface system suchas a river or lake. Local codes and regulations may affectthe feasibility of open-loop systems.Depending on the well configuration, open-loopsystems can have the highest pumping load requirementsof any of the ground-coupled configurations. In idealconditions, however, an open-loop application can be themost economical type of ground-coupling system.• Advantages: Simple design; lower drilling requirementsthan closed-loop designs; subject to betterthermodynamic performance than closed-loopsystems because well(s) are used to deliver groundwaterat ground temperature rather than as a heatexchanger delivering heat transfer fluid at temperaturesother than ground temperature; typically lowestcost; can be combined with potable water supplywell; low operating cost if water already pumpedfor other purposes, such as irrigation.• Disadvantages: Subject to various local, state, andFederal clean water and surface water codes andregulations; large water flow requirements; wateravailability may be limited or not always available;heat pump heat exchanger subject to suspendedmatter, corrosive agents, scaling, and bacterial contents;typically subject to highest pumping powerrequirements; pumping energy may be excessiveif the pump is oversized or poorly controlled; mayrequire well permits or be restricted for extraction;water disposal can limit or preclude some installations;high cost if reinjection well required.Direct-Expansion SystemsEach of the ground-coupling systems describedabove uses an intermediate heat transfer fluid to transferheat between the earth and the refrigerant. Use ofan intermediate heat transfer fluid necessitates a highercompression ratio in the heat pump to achieve sufficienttemperature differences in the heat transfer chain (refrigerantto fluid to earth). Each also requires a pump tocirculate water between the heat pump and the groundcouple.Direct-expansion systems, illustrated in Figure28.2h, remove the need for an intermediate heat transferfluid, the fluid-refrigerant heat exchanger, and the circulationpump. Copper coils are installed underground fora direct exchange of heat between refrigerant and earth.The result is improved heat transfer characteristics andthermodynamic performance.The coils can be buried either in deep verticaltrenches or wide horizontal excavations. Vertical trenchestypically require from 100 to 150 square feet of land surfacearea per system cooling ton (2.6 to 4.0 m 2 /kW) andare typically 9 to 12 feet (2.7 to 3.7 m) deep. Horizontalinstallations typically require from 450 to 550 square feetof land area per system cooling ton (11.9 to 14.5 m 2 /kW)and are typically 5 to 10 feet (1.5 to 3.0 m) deep. Verticaltrenching is not recommended in sandy, clay or dry soilsbecause of the poor heat transfer.Because the ground coil is metal, it is subject tocorrosion (the pH level of the soil should be between5.5 and 10, although this is normally not a problem). Ifthe ground is subject to stray electric currents and/orgalvanic action, a cathodic protection system may berequired. Because the ground is subject to larger temperatureextremes from the direct-expansion system,there are additional design considerations. In winterheating operation, the lower ground coil temperaturemay cause the ground moisture to freeze. Expansion ofthe ice buildup may cause the ground to buckle. Also,because of the freezing potential, the ground coil shouldnot be located near water lines. In the summer coolingoperation, the higher coil temperatures may drive moisturefrom the soil. Low moisture content will change soilheat transfer characteristics.At the time this chapter was initially drafted (1995),only one U.S. manufacturer offered direct-expansionground-source heat pump systems. However, new companieshave released similar direct-expansion systems. InNovember 2005, the Geothermal Heat Pump Consortiumweb site identified four manufacturers of direct exchange

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 763systems. Systems were available from 16,000 to 83,000Btu/h (heating/cooling capacity) (4.7 to 24.3 kW). Largercommercial applications would require multiple unitswith individual ground coils.• Advantages: Higher system efficiency; no circulationpump required.• Disadvantages: Large trenching requirements foreffective heat transfer area; ground around the coilsubject to freezing (may cause surface ground tobuckle and can freeze nearby water pipes); coppercoil should not be buried near large trees where rootsystem may damage the coil; compressor oil returncan be complicated, particularly for vertical heatexchanger coils or when used for both heating andcooling; leaks can be catastrophic; higher skilledinstallation required; installation costs typicallyhigher; this system type requires more refrigerantthan most other systems; smaller infrastructure inthe industry.28.4.4 Variables Affecting Design and PerformanceAmong the variables that have a major impact onthe sizing and effectiveness of a ground-coupling system,the importance of underground soil temperatures andsoil type deserves special mention.Underground Soil TemperatureThe soil temperature is of major importance in thedesign and operation of a ground-source heat pump. Inan open-loop system, the temperature of groundwaterentering the heat pump has a direct impact on the efficiencyof the system. In a closed-loop system and in thedirect-expansion system, the underground temperaturewill affect the size of the required ground-coupling systemand the resulting operational effectiveness of theunderground heat exchanger. Therefore, it is importantto determine the underground soil temperature beforeselecting a system design.Annual air temperatures, moisture content, soil type,and ground cover all have an impact on undergroundsoil temperature. In addition, underground temperaturevaries annually as a function of the ambient surface airtemperature swing, soil type, depth, and time lag. Figure28.3 contains a map of the United States indicating meanannual underground soil temperatures and amplitudes ofannual surface ground temperature swings. Figure 28.4,though for a specific location, illustrates how the annualsoil temperature varies with depth, soil type, and season.For vertical ground-loop systems, the mean annual earthtemperature (Figure 28.3a) is an important factor in theground-loop design. With horizontal ground-loop systems,the ground surface annual temperature variation(Figure 28.3b and Figure 28.4b) becomes an importantdesign consideration.Soil and Rock ClassificationThe most important factor in the design and successfuloperation of a closed-loop ground-source heatpump system is the rate of heat transfer between theclosed-loop ground-coupling system and the surroundingsoil and rock. The thermal conductivity of the soiland rock is the critical value that determines the lengthof pipe required. The pipe length, in turn, affects the installationcost as well as the operational effectiveness,which in turn affects the operating cost. Because of localvariations in soil type and moisture conditions, economicdesigns may vary by location. Soil classificationsinclude coarse-grained sands and gravels, fine-grainedsilts and clays, and loam (equal mixtures of sand, silt,and clay). Rock classifications are broken down intonine different petrologic groups. Thermal conductivityvalues vary significantly within each of the nine groups.Each of these classifications plays a role in determiningthe thermal conductivity and thereby affects the designof the ground-coupling system. For more informationon the thermal properties of soils and rocks and how toidentify the different types of soils and rocks, see Soiland Rock Classification for the Design of Ground-CoupledHeat Pump Systems (STS Consultants 1989).Series versus Parallel FlowClosed-loop ground-coupled heat exchangers maybe designed in series, parallel, or a combination of both.In series systems, the heat transfer fluid can take onlyone path through the loop, whereas in parallel systemsthe fluid can take two or more paths through the circuit.The selection will affect performance, pumpingrequirements, and cost. Small-scale ground-couplingsystems can use either series or parallel-flow design, butmost large ground-coupling systems use parallel-flowsystems. The advantages and disadvantages of seriesand parallel systems are summarized below. In largesystems, pressure drop and pumping costs need to becarefully considered or they will be very high. Variablespeeddrives can be used to reduce pumping energy andcosts during part-load conditions. Total life-cycle costand design limitations should be used to design a specificsystem.• Series-System Advantages: Single path flow andpipe size; easier air removal from the system; slightlyhigher thermal performance per linear foot of

764 ENERGY MANAGEMENT HANDBOOKFigure 28.4. Soil temperaturevariation Source: OSU (1988)Figure 28.3. Mean annual soil temperatures. Source: OSU (1988)• Parallel-System Advantages:Smaller pipe diameter has lowerunit cost; lower volume requiresless antifreeze; smaller pressuredrop resulting in smaller pumpingload; lower installation laborcost.pipe because larger pipe size required in the seriessystem.• Series-System Disadvantages: Larger fluid volumeof larger pipe in series requires greater antifreezevolumes; higher pipe cost per unit of performance;increased installed labor cost; limited capacity(length) caused by fluid pressure drop characteristics;larger pressure drop resulting in larger pumpingload; requires larger purge system to remove airfrom the piping network.• Parallel-System Disadvantages: Special attentionrequired to ensure air removal and flow balancingbetween each parallel path to result in equal lengthloops.28.4.5 VariationsThe ground-coupling system is what makes theground-source heat pump unique among heating andair-conditioning systems and, as described above, thereare several types of ground-coupling systems. In addition,variations to ground-source heat pump design andinstallation can save additional energy or reduce installa-

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 765tion costs. Three notable variations are described below.Cooling-Tower-Supplemented SystemThe ground-coupling system is typically the largestcomponent of the total installation cost of a groundsourceheat pump. In southern climates or in thermallyheavy commercial applications where the cooling load isthe driving design factor, supplementing the system witha cooling tower or other supplemental heat rejection systemcan reduce the required size of a closed-loop groundcouplingsystem. The supplemental heat rejection systemis installed in the loop by means of a heat exchanger (typicallya plate and frame heat exchanger) between the facilityload and the ground couple. A cooling tower systemis illustrated in Figure 28.5. The cooling tower acts to precoolthe loop’s heat transfer fluid upstream of the groundcouple, which lowers the cooling-load requirement on theground-coupling system. By significantly reducing the requiredsize of the ground-coupling system, using a coolingtower can lower the overall installation cost. This typeof system is operating successfully in several commercialfacilities, including some mission-critical facilities at FortPolk in Louisiana.Solar-Assisted SystemIn northern climates where the heating load is thedriving design factor, supplementing the system withsolar heat can reduce the required size of a closed-loopground-coupling system. Solar panels, designed to heatwater, can be installed into the ground-coupled loop (bymeans of a heat exchanger or directly), as illustrated inFigure 28.6. The panels provide additional heat to theheat transfer fluid. This type of variation can reduce therequired size of the ground-coupled system and increaseheat pump efficiency by providing a higher temperatureheat transfer fluid.Hot Water Recovery/DesuperheatingThe use of heat pumps to provide hot water is becomingcommon. Because of their high efficiency, thispractice makes economic sense. Most manufacturers offeran option to include desuperheating heat exchangersto provide hot water from a heat pump. These dual-wallheat exchangers are installed in the refrigerant loop to recoverhigh temperature heat from the superheated refrigerantgas. Hot-water recovery systems can supplement,or sometimes replace, conventional facility water-heatingsystems. With the heat pump in cooling mode, hot-waterrecovery systems increase system operating efficiencywhile acting as a waste-heat-recovery device—and provideessentially free hot water. When the load is increasedduring the heating mode, the heat pump still providesheating and hot water more efficiently and less expensivelythan other systems.28.4.6 System Design and InstallationMore is becoming known about the design and installationof ground-source heat pumps. Design-day coolingand heating loads are determined through traditionaldesign practices such as those documented by the AmericanSociety of Heating, Refrigerating, and Air-ConditioningEngineers (ASHRAE). Systems are also zoned usingcommonly accepted design practices.The key issue that makes ground-source heat pumpsunique is the design of the ground-coupling system. Mostoperational problems with ground-source heat pumpsFigure 28.5. Cooling-tower-supplemented system forcooling-dominated loadsFigure 28.6. Solar-assisted system for heating-dominatedloads

766 ENERGY MANAGEMENT HANDBOOKstem from the performance of the ground-coupling system.Today, software tools are available to support thedesign of the ground-coupling systems that meet theneeds of designers and installers. These tools are availablefrom several sources, including the InternationalGround-Source Heat Pump Association (IGSHPA). Inaddition, several manufacturers have designed their ownproprietary tools more closely tuned to their particularsystem requirements.Ground loops can be placed just about anywhere—under landscaping, parking lots, or ponds. Selection of aparticular ground-coupling system (vertical, horizontal,spiral, etc.) should be based on life-cycle cost of the entiresystem, in addition to practical constraints. Horizontalclosed-loop ground-coupling systems can be installedusing a chain-type trenching machine, horizontal boringmachine, backhoe, bulldozer, or other earth-movingheavy equipment. Vertical applications (for both openand closed systems) require a drilling rig and qualifiedoperators. Most applications of ground-source heatpumps to large facilities use vertical closed-loop groundcouplingsystems primarily because of land constraints.Submerged-loop applications require some special considerationsand, as noted earlier, it is best to discuss thesedirectly with an experienced design engineer.It is important to assign overall responsibility forthe entire ground-source heat pump system to a singleindividual or contractor. Installation of the system, however,will involve several trades and contractors, many ofwhom may not have worked together in previous efforts.In addition to refrigeration/air-conditioning and sheetmetal contractors, installation involves plumbers, and(in the case of vertical systems) well drillers. Designatinga singular responsible party and coordinating activitieswill significantly reduce the potential for problems withinstallation, startup, and proper operation.In heating-dominated climates, a mixture of antifreezeand water must be used in the ground-couplingloops if loop temperatures are expected to fall belowabout 41ºF (5ºC). A study by Heinonen (1997) establishesthe important considerations for antifreeze solutions forground-source heat pump systems and provides guidanceon selection.One note of caution to the designer: some regulations,installation manuals, and/or local practices callfor partial or full grouting of the borehole. The thermalconductivity of materials normally used for groutingis very low compared with the thermal conductivity ofmost native soil formations. Thus, grouting tends to act asinsulation and hinders heat transfer to the ground. Someexperimental work by Spilker (1998) has confirmed thenegative impact of grout on borehole heat transfer. Underheat rejection loading, average water temperature wasnearly 11°F (6ºC) higher for a 6.5-in. (16.5-cm) diameterborehole backfilled with standard bentonite grout thanfor a 4.75-in. (12.1-cm) diameter borehole backfilled withthermally enhanced bentonite grout. Using fine sand asbackfill in a 6.5-in. (16.5-cm) diameter borehole loweredthe average water temperature over 14°F (8ºC) comparedwith the same-diameter bore backfilled with standardbentonite grout. For a typical system (Spilker 1998) witha 6.5-in. (16.5-cm) diameter borehole, the use of standardbentonite grout would increase the required bore lengthby 49% over fine sand backfill in the same borehole. Byusing thermally enhanced grout in a smaller 4.75-in.(12.1-cm) borehole, the bore length is increased by only10% over fine sand backfill in the larger 6.5-in. (16.5-cm)diameter borehole. Thus, the results of this study (Spilker1998) suggest three steps that may be taken to reducethe impact of grout on vertical borehole system performance:• Reduce the amount of grout used to the bare minimum.Sand or cuttings may be used where allowed,but take care to ensure that the entire interstitialspace between the piping and the borehole diameteris filled.• Use thermally enhanced grout wherever possible.For information on thermally enhanced grout consultASHRAE (1997) and Spilker (1998).• Reduce the borehole diameter as much as possible tomitigate the effects of the grout or backfill used. Theregulatory requirements for vertical boreholes usedfor ground-coupling heat exchangers vary widelyby state. Current state and federal regulations, aswell as related building codes, are summarized atthe Geothermal Heat Pump Consortium web site(www.geoexchange.org/publications/regs.htm).28.4.7 Summary of Ground-Loop Design SoftwareBecause of the diversity in loads in multi-zone buildings,the design of the ground-coupling heat exchanger(the ground loop) must be based on peak block loadrather than the installed capacity. This is of paramountimportance because ground coupling is usually a majorportion of the total ground-source heat pump systemcost, and over-sizing will render a project economicallyunattractive.In the residential sector, many systems have been designedusing rules-of-thumb and local experience, but forcommercial-scale systems such practices are ill advised.For all but the most northern climates, commercial-scale

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 767buildings will have significantly more heat rejection thanextraction. This imbalance in heat rejection/extractioncan cause heat buildup in the ground to the point whereheat pump performance is adversely affected and hencesystem efficiency and possibly occupant comfort suffer.(This is an important consideration in producintg accuratelife-cycle cost estimates of energy use.) Proper designfor commercial-scale systems almost always benefits fromthe use of design software. Software for commercial-scaleground-source heat pump system design should considerthe interaction of adjacent loops and predict the potentialfor long-term heat buildup in the soil. Some sourcesof PC-based design software packages that address thisneed are:• GchpCalc, Version 3.1, Energy Information Services,Tel: (205) 799-4591. This program includes built-intables for heat pump equipment from most manufacturers.Input is in the form of heat loss/gain duringa design day and the approximate equivalentfull-load heating hours and equivalent full-loadcooling hours. Primary output from the program isthe ground loop length required. This program willalso calculate the optimal size for a supplementalfluid cooler for hybrid systems, as discussed later.• GLHEPRO, International Ground Source HeatPump Association (IGSHPA), Tel: (800) 626-4747.Input required is monthly heating/cooling loadson heat pumps and monthly peak loads either entereddirectly by user or read from BLAST or TraneSystem Analyzer and Trane Trace output files.Output includes long-term soil temperature effectfrom rejection/extraction imbalance. The currentconfiguration of the program has some constraintson selection of borehole spacing, depth, and overalllayout that will be removed from a future versionnow being prepared.• GS2000, Version 2.0, Caneta Research Inc., Tel: (905)542-2890, email: caneta@compuserve.com. Heating/coolingloads are input as monthly totals onheat pumps or, alternatively, monthly loads on theground loop may be input. Equipment performanceis input at ARI/ISO rating conditions. For operatingconditions other than the rating conditions, theequipment performance is adjusted based on genericheat pump performance relationships.Each of these programs requires input about the soilthermal properties, borehole resistance, type of pipingand borehole arrangement, fluid to be used, and otherdesign parameters. Many of the required inputs willbe available from tables of default values. The designershould be careful to ensure that the values chosen arerepresentative of the actual conditions to be encounteredto ensure efficient and cost-effective designs. Test boringsand in situ thermal conductivity analysis to determinethe type of soil formations and aquifer locations will substantiallyimprove design accuracy and may help reducecosts. Even with the information from test borings, someuncertainty will remain with respect to the soil thermalproperties. These programs make it possible to vary designparameters easily within the range of anticipatedvalues and determine the sensitivity of the design to aparticular parameter (OTL 1999). In some instances, particularlyvery large projects, it may be advisable to obtainspecific information on ground-loop performance bythermal testing of a sample borehole (Shonder and Beck2000).28.5 APPLICATIONThis section addresses technical aspects of applyingground-source heat pumps. The range of applicationsand climates in which the technology has been installedare discussed. The advantages, limitations, and benefitsare enumerated. Design and integration considerationsfor ground-source heat pumps are highlighted, includingenergy savings estimates, equipment warranties, relevantcodes and standards, equipment and installation costs,and utility incentives.28.5.1 Application ScreeningA ground-source heat pump system is one of themost efficient technologies available for heating and cooling.It can be applied in virtually any climate or buildingcategory. Although local site conditions may dictate thetype of ground-coupling system employed, the high firstcost and its impact on the overall life-cycle cost are typicallythe constraining factors.The operating efficiency of ground-source heatpumps is very dependent on the entering water temperature,which, in turn, depends on ground temperature,system load, and size of ground loop. As with any HVACsystem, the system load is a function of the facility, internalactivities, and the local weather. Furthermore, withground-source heat pumps, the load on the ground-couplingsystem may impact the underground temperature.Therefore, energy consumption will be closely tied to therelationship between the annual load distribution and theannual ground loop-temperature distribution (e.g., theirjoint frequency distribution).

768 ENERGY MANAGEMENT HANDBOOKThere are several techniques to estimate the annualenergy consumption of ground-source heat pumpsystems. The most accurate methodologies use computersimulation, and several software systems now supportthe analysis of ground-source heat pumps. These methods,while more accurate than hand techniques, are alsodifficult and expensive to employ and are therefore moreappropriate when additional detail is required ratherthan as an initial screening tool.The bin method is another analytical tool for screeningtechnology applications. In general, a bin method is asimple computational procedure that is readily adaptableto a spreadsheet-type analysis and can be used to estimatethe energy consumption of a given application andclimate. Bin methods rely on load and ambient wet anddry bulb temperature distributions. This methodology isused in the case study presented later in this chapter.28.5.2 Where to Apply Ground-Source Heat PumpsGround-source heat pumps are generally applied toair-conditioning and heating systems, but may also be usedin any refrigeration application. The decision whether touse a ground-source heat pump system is driven primarilyby economics. Almost any HVAC system can be designedusing a ground-source heat pump. The primary technicallimitation is a suitable location for the ground-couplingsystem. The following list identifies some of the best applicationsof ground-source heat pumps.• Ground-source heat pumps are probably least costprohibitivein new construction; the technology isrelatively easy to incorporate.• Ground-source heat pumps can also be cost effectiveto replace an existing system at the end of its usefullife, or as a retrofit, particularly if existing ductworkcan be reused with minimal modification.• In climates with either cold winters or hot summers,ground-source heat pumps can operate much moreefficiently than air-source heat pumps or other airconditioningsystems. Ground-source heat pumpsare also considerably more efficient than other electricheating systems and, depending on the heatingfuel cost, may be less expensive to operate thanother heating systems.• In climates with by high daily temperature swings,ground-source heat pumps show superior efficiency.In addition, in climates characterized by large dailytemperature swings, the ground-coupling systemalso offers some thermal storage capability, whichmay benefit the operational coefficient of performance.• In areas where natural gas is not available or wherethe cost of natural gas or other fuel is high comparedwith electricity, ground-source heat pumps are economical.They operate with a heating coefficient ofperformance in the range of 3.0 to 4.5, comparedwith conventional heating efficiencies in the rangeof 80% to 97%. Therefore, when the cost of electricity(per Btu) is less than 3.5 times that of conventionalheating fuels (per Btu), ground-source heat pumpshave lower energy costs.• Areas of high natural gas (or fuel oil) costs will favorground-source heat pumps over conventionalgas (or fuel oil) heating systems. High electricitycosts will favor ground-source heat pumps over airsourceheat pumps.• In facilities where multiple temperature controlzones or individual load control is beneficial,ground-source heat pumps provide tremendouscapability for individual zone temperature controlbecause they are primarily designed using multipleunitary systems.• In areas where drilling costs are low, vertical-loopsystems may be especially attractive. The initial costof the ground-source heat pump system is one of theprime barriers to the economics. In locations with asignificant ground-source heat pump industry infrastructure(such as Oklahoma, Louisiana, Florida,Texas, and Indiana), installation costs may be lowerand the contractors more experienced. This, however,is changing as the market for ground-sourceheat pumps grows.28.5.3 What to AvoidThe following precautions should be followed whenthe application of ground-source heat pump technologyis considered:• Avoid threaded plastic pipe connections in theground loop. Specify thermal fusion welding. Unlikeconventional water-source heat pump systemswhere the water loop temperature ranges from 60°to 90°F (15.6° to 32.2°C), ground-source systemsare subject to wider temperature ranges (20° to110°F [-6.7° to 43.3°C]), and the resulting expansionand contraction may result in leaks at the threadedconnection. It is also generally recommended to

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 769specify piping and joining methods approved byInternational Ground-Source Heat Pump Association( IGSHPA).• Check local water and well regulations. Regulationsaffecting open-loop systems are common, andlocal regulations can vary significantly. Some localregulations may require reinjection wells ratherthan surface drainage. Some states require permitsto use even private ponds as a heat source/sink.• Have the ground-source heat pump system installedas a complete and balanced assemblage ofcomponents, each of which must be properly designed,sized, and installed (Giddings 1988). Also,have the system installed under the responsibilityof a single party. If the entire system is installedby three different professionals, none of whom understandsor appreciates the other two parts of thesystem, then the system may not perform satisfactorily.• One of the most frequent problems cited is impropersizing of the heat pump or the ground-couplingsystem. Approved calculation procedures shouldbe used in the sizing process—as is the case withany heating or air-conditioning system regardlessof technology. ASHRAE has established one of themost widely known and accepted standards for thedetermination of design heating and cooling loads.Sizing the ground-coupling system is just as critical.Because of the uncertainty of soil conditions, asite analysis to determine the thermal conductivityand other heat transfer properties of the local soilmay be required. This should be the responsibilityof the designing contractor because it can significantlyaffect the final design.• Avoid inexperienced designers and installers (seeabove). Check on the previous experience of potentialdesigners and installers. It is also generally recommendedto specify IGSHPA certified installers.28.5.4 Design and Equipment IntegrationThe purpose of this chapter is to familiarize the energymanager and facility engineer with the benefits andliabilities of ground-source heat pumps in their applicationto commercial buildings. It is beyond the scope ofthis chapter to fully explain the design requirements ofa ground-source heat pump system. It is, however, importantthat the reader know the basic steps in the designprocess.The design of a ground-source heat pump systemwill generally follow the following sequence:1. Determine local design conditions, climatic andsoil thermal characteristics.2. Determine local water, well, and grouting requirements.3. Determine building heating and cooling loads atdesign conditions.4. Select the alternative HVAC system components,including the indoor air-distribution system type;size the alternatives as required; and select equipmentthat will meet the demands calculated in Step2 (using the preliminary estimate of the enteringwater temperatures to determine the heat pump’sheating and cooling capacities and efficiencies).5. Determine the monthly and annual building heatingand cooling energy requirements.6. Make preliminary selection of a ground-couplingsystem type.7. Determine a preliminary design of the groundcouplingsystem.8. Determine the thermal resistance of the groundcouplingsystem.9. Determine the required length of the ground-couplingsystem; recalculate the entering and exitingwater temperatures on the basis of system loadsand the ground-coupling system design.10. Redesign the ground-coupling system, as required,to balance the requirements of the system load(heating and cooling) with the effectiveness of theground-coupling system. Note that designing andsizing the ground-coupling system for one season(such as cooling) will impact its effectiveness andability to meet system load requirements duringthe other season (such as heating).11. Perform life-cycle cost analysis on the system design(or system design alternatives).Although the design procedure for the groundcouplingsystem is an iterative and sometimes difficultprocess, several sources are available to simplify the task.First, an experienced designer should be assigned responsibilityfor the heat pump and ground-coupling system

770 ENERGY MANAGEMENT HANDBOOKdesigns. Several manufacturers of ground-source heatpump equipment have their own software tools to supportthe design of large, commercial-type systems. However,for those who typically design systems in-house,there are support tools available. Software programs areavailable to support the design of ground-source heatpump HVAC systems and the ground-coupling system.Several software tools are available through the IGSHPA,including an Earth-Coupled Analysis Program and aGround-Loop Heat Exchanger Design Program. In addition,several technical design manuals also are availablethrough IGSHPA, ASHRAE, and equipment manufacturers(refer to earlier section for an introduction to groundloopdesign software).There are several different approaches for incorporatingground-source heat pumps into the HVAC design.However, most applications in large facilities involvemultiple smaller heat pump units (

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 771comparing equipment efficiencies, it is important to makean appropriate comparison. The efficiency of any heatingor cooling equipment varies with application, load, andrelated heat-source and heat-sink temperatures.As partially illustrated in Table 28.1, ISO Standard13256-1 provides separate equipment rating conditionsfor water-source heat pumps, ground-water-source heatpumps, and closed-loop ground-source heat pumps. Thedifference among these rating conditions is in the application,not necessarily the equipment. Therefore, a given heatpump rated under closed-loop ground-source heat pumpconditions would have a different EER (cooling) and COP(heating) than if the same heat pump were rated under water-sourceheat pump rating conditions and different still ifthe equipment were rated under the ground-water-sourceheat pump rating conditions.Also, standard ratings are for specific temperaturesand operating conditions. Standard ratings do not necessarilyreflect the efficiency of systems under true seasonaloperating conditions. In determining the efficiency of heatingor cooling equipment for estimating energy consumptionor potential energy savings, the efficiency should becorrected for the appropriate operating conditions.28.5.7 Utility Incentives and SupportMany utilities are promoters of ground-source heatpumps, and many offer incentive programs and support.The Geothermal Heat Pump Consortium web site (www.geoexchange.org/incentives/incentives.htm) identifiesfederal and state incentives for ground-source heat pumpsystems for both residential and commercial applications.Incentives identified include rebates (by ton, byunit, and/or by kWh energy saved), low interest loans,sales tax exemptions, tax credits, and/or tax deductions.Rebates for ground-source heat pumps identified rangedfrom $150 to $600 per ton. For commercial facilities, someincentives ran into several thousand dollars. In additionto the technology-specific rebate programs, some utilitiesalso offer custom rebate programs or programs that arenot technology specific but are based on the energy savingsregardless of the technology employed. Readers areencouraged to contact their local utility to find out moreabout what programs, services, and rebates may be offeredby the utility to promote energy management andnew energy-efficient technologies.28.6 TECHNOLOGY PERFORMANCEIn 1999, an estimated 400,000 ground-source heatpumps were operating in residential and commercialapplications, up from 100,000 in 1990. With a projectedannual growth rate of 10%, 120,000 new units would beinstalled in 2010, for a total of 1.5 million units in 2010(Lund and Boyd 2000). The majority of new groundsourceheat pump installations in the United States are forresidential applications in the southern and mid-westernstates. Environmental concerns and a general lack of understandingof the technology by HVAC companies andinstallers have limited installations in the west.28.6.1 Field ExperienceObservations about field performance of groundsourceheat pumps obtained from federal and privatesectorusers are summarized in this section.The large numbers of reported installations testifyto the stability of this technology. Most sites contactedreport satisfaction with the overall performance (energyefficiency, maintenance, and comfort) of the technology.One of the world’s largest installations of ground-sourceheat pumps is at Fort Polk, Louisiana, where over 4000units were installed in family housing as part of a majorretrofit project (Hughes and Shonder 1998). They alsohave systems installed in several larger facilities, includingsome cooling-tower hybrid systems (discussed earlierin this chapter). Personnel at Fort Polk are obviouslypleased with the performance of their ground-source heatpumps, and the energy manager reports that the systemshave performed better than expected and they plan to usethe technology for all heating and cooling requirementson the base.The maintenance engineer at the hospital in LoveCounty, Oklahoma, reported that no problems have occurredsince nine units were installed in their facility inthe early 1990s. They perform routine maintenance on thesystem. They particularly like the technology because ittakes up less space and gives better heating and coolingcontrol than former systems.An interview with the energy manager primarilyresponsible for the ground-source heat pump installationat the Oklahoma State Capitol Building revealed no problemswith the heat pumps or the ground loop; however,the building was poorly zoned, which impeded effectivetemperature control. Approval from the local Water ResourceBoard for drilling the wells proved difficult butwas primarily a matter of educating the board aboutthe technology and the closed-loop system. The Capitolbuilding system, which is a cooling-tower-supplementeddesign, contains 855-cooling tons (3,009.6 kW) connectedto a common ground-loop system consisting of 372 250-foot (76.2 m) vertical wells.One facility manager did report problems with theground-source heat pump system installed in a facilityat the Dugway Proving Grounds, Utah. The system uses

772 ENERGY MANAGEMENT HANDBOOK22.5-ton (8.8-kW) units. Although the actual problem hadnot been identified, the system is not performing up toexpectations. The belief was that the ground loop wasundersized. Sizing of the ground loop is one of the mostimportant design factors and is a function of system load,ground temperature, and local soil conditions. As notedearlier in this chapter, the type of soil and rock in contactwith the pipe loop has serious implications for thelength of ground loop required for adequate heat transfer.Inexperienced designers or installers are more likely toundersize the systems, resulting in inadequate systemperformance, or to oversize the systems, resulting in increasedinstallation costs. The importance of good design,documentation, installation, and commissioning cannotbe overemphasized.In early 1995, at the National Training Center in FortIrwin, California, 220 new single-family houses were builtwith an innovative ground-source heat pump system. Thethermal sink/source for the heat pumps is geothermallyheated 75ºF (23.9°C) groundwater. The groundwater ispumped into a storage reservoir adjacent to the housingproject. The reservoir water is then circulated througha double-walled heat exchanger, transferring thermalenergy to/from a secondary closed circulation loop connectedto individual heat pumps in each residence. Operatingconditions and problems with maintaining correctwater flow through the central facility heat exchangercontributed to lower than expected energy savings duringthe first year of operation. There was no indicationthat the heat pumps themselves were not performing asdesigned. Limited anecdotal information from the occupantsindicates that overall satisfaction with the level ofthermal comfort from the system was high. 9Vertical-bore, ground-coupled heat pump systemswere installed in four new elementary schools in Lincoln,Nebraska, in 1995. Each school required 54 heat pumpsranging in size from 1.4 to 15 tons (204-tons total coolingcapacity). With the heat pump systems, utility costs forthese four schools are nearly half of that of other schools inthe district, and the systems are providing a comfortable,complaint-free environment (Shonder et al. 1999).Additional case studies are identified on the GeothermalHeat Pump Consortium web page (http://www.geoexchange.org/).9 D.L. Hadley and L. Lkevgard. August 1997. Family Housing EnergySavings Verification, National Training Center, Fort Irwin, CA.Preliminary Findings: Cooling Season Energy Savings. Letter Report,Pacific Northwest National Laboratory, Richland, Washington.28.6.2 Energy SavingsThe most important reason to consider the applicationof ground-source heat pumps to commercial buildingis the potential energy savings and its impact onoverall life-cycle cost of the heating and cooling system.Ground-source heat pumps save energy and money becausethe equipment operates more efficiently than conventionalsystems, the maintenance costs are lower (seenext section for maintenance benefits), and the equipmenthas a longer life expectancy than conventionalunitary equipment. In addition, a ground-source heatpump does not require a defrost cycle, or, in most situations,backup electric resistance heat, as do air-sourceheat pumps.By comparison, the average cooling efficiency atcommercial-type facilities is estimated to be an EER of8.0 (2.33 COP) for existing facilities and an EER of 10.0(2.93 COP) for new facilities. Ground-source heat pumpsystems have the potential to reduce consumption ofcooling energy by 30% to 50% and to reduce heating energyby 20% to 40% compared with typical air-source heatpumps.A review of manufacture’s literature on commerciallyavailable systems indicates that cooling efficiencies(EERs) of 13.4 to 20 Btu/W-h and heating efficiencies(COPs) of 3.1 to 4.3 are readily available. 10 A study preparedby DOE estimates energy-saving ranges of 17% to42% comparing ground-source heat pumps to air-sourceheat pumps, depending on region (Calm 1987). Figure28.7 illustrates the range of efficiencies typical of variousheating and cooling equipment.28.6.3 MaintenanceThe ground-source heat pump technology is matureand reliable. Systems have standard warranties rangingfrom 1 to 5 years. The heat pump units are self-contained,and maintenance requirements are relatively straightforward;no new maintenance skills are necessary. Becauseheat pump equipment is not exposed to outdoor elements,the units actually require less maintenance than typicalair-source heat pumps. One site reported a problem withcottonwood trees clogging outside condenser units ofthe previous air-conditioning systems. This maintenanceproblem was eliminated with the application of groundsourceheat pumps.In closed-loop systems, the ground loop is virtuallymaintenance free. The circulating pump(s) requires routinemaintenance, as with any pump and motor system,and the water loop (a closed system) should be routinelymonitored for temperature, pressure, flow, and antifreezeconcentration. Unless there is a leak, no action is generallyrequired.10 Rating based on ANSI/ARI/ASHRAE ISO Standard 13256-1-2005 forclosed-loop ground-source heat pumps.

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 773Figure 28.7. Typical heating and cooling equipment efficienciesIn open-loop systems, the well requires maintenancesimilar to any water well. The system should be routinelymonitored for temperature, pressure, and flow. Becausegroundwater is being supplied to the heat pump, the heatexchangers should be routinely inspected for potentialfouling and scale buildup.Maintenance costs for ground-source heat pumpsare about the same or less than conventional equipment.For example, maintenance costs for four schools in theLincoln, Nebraska, equipped with ground-source heatpump systems have been well documented (Martin et al.1999, 2000). The results of these studies are summarizedin Table 28.2.Table 28.2 Maintenance Costs for Public Schools in Lincoln,Nebraska—————————————————————————PreventiveMaintenance Repair TotalHVAC System Type (cents/yr;ft 2 ) (cents/hr-ft 2 ) (cents/hr-ft 2 )—————————————————————————GSHP System 7.1 2.1 9.2—————————————————————————Conventional system 5.9 to 12.6 2.9 to 6.1 8.8 to 18.7—————————————————————————Source: Martin et al. 1999, 2000Another analysis, this one by the Geothermal HeatPump Consortium, found that the average total preventiveand corrective maintenance costs for 25 ground-sourceheat-pump-equippedbuildings were approximately 11cents/ft 2 , compared to 30 to 40 cents/ft 2 for conventionalsystems (Cane 1998).28.6.4 Installation CostsApplication-specific parameters, such as equipmentcapacity, type, refrigerant, air-distribution system, controlsystem, plumbing configuration, and ground-couplingsystem type, significantly affect the total cost of the overallsystem. While equipment costs are competitive, installationcosts vary significantly.To illustrate the potential range of installation costs,the following examples have been reported.• Stockton State College in Pomona, New Jersey, retrofita system that totaled 1,655 tons (5,826 kW) at atotal cost of $5,246,000 (Gahran 1993). The facility receivedgrants and rebates reducing the capital outlayto $135,000. The system included 63 rooftop units, 500variable-air-volume (VAV) boxes, and a 3,500- pointenergy management control system (EMCS). Theunit cost was $3,170/ton in 1993 dollars.• WaterFurnace, a ground-source heat pump manufacturer,designed the technology using a submergedpond closed-loop system into its new officebuilding located in Fort Wayne, Indiana. The systemtotaled 134 tons (471.7 kW) at a cost of $239,800 (WaterFurnaceWF639). The unit cost was $1,790/ton in1991 dollars.• Salem Community College in Carney’s Point, NewJersey, retrofit a system that totaled 160 tons at a totalcost of $284,000 (Gahran 1994). The system included32 heat pumps. The unit cost was $1,775/ton in 1993dollars.

774 ENERGY MANAGEMENT HANDBOOK• Paint Lick Elementary School in Garrard County,Kentucky, designed the technology into the newschool building. The system totaled 123 tons (433 kW)at a total cost of $380,010 (WaterFurnace WF666). Theunit cost was $3,090/ton in 1992 dollars.• Maywood Elementary School in Hammond, Indiana,designed the technology into a new schoolbuilding. The unit totaled 250 tons (880 kW) at atotal cost of $1,277,190 (WaterFurnace WF925). Thesystem consisted of 74 heat pumps and a groundcouplingsystem consisting of 244 vertical wells. Theunit cost was $5,110/ton in 1994 dollars.• The Lincoln, Nebraska, school district installedground-source heat pump systems in four new elementaryschools. In each school, the system consistedof 54 heat pumps ranging in size from 1.4 tonsto 15 tons, with a total cooling capacity of 180 tons(630 kW). Four gas-fired boilers with a capacity of330,000 Btuh each provide hot water for preheat andterminal reheat. The total heat pump cost per schoolwas approximately $657,000 ($3,650/ton) in 1995dollars (Shonder et al. 1999).• Kavanaugh (1995) concluded that the average costof ground-source heat pumps (including unit, loop,duct, and installation) ranged from $2360/ton for a5-ton horizontal loop to $3000/ton for a 3-ton verticalloop system. Compared to a 3-ton conventionalsystem, the added cost was $1250 to $1550 per ton.Installation costs are expected to drop as theground-source heat pump industry infrastructure growsand designers and installers become more experienced.Reducing installation costs is one of the prime goals of theInternational Ground-Source Heat Pump Association andthe Geothermal Heat Pump Consortium.28.6.5 Other ImpactsThere are no significant negative environmental impactsassociated with ground-source heat pumps. Thereis, however, the potential for systems to be affected bysome local codes and regulations. The most likely sourceof conflict, if any, lies with the installation of the groundcouplingsystem. Working with an experienced installeris the best advice. However, local electric utilities andother local sites with existing ground-source heat pumpinstallations are other sources of information about localpermit and regulation issues.With the application of any electrotechnology, thereis a potential environmental benefit. Installing a groundsourceheat pump system in lieu of a fossil-fuel heatingsystem will reduce local emissions. Furthermore, installinga more efficient electrotechnology such as a groundsourceheat pump system for cooling will reduce sourceemissions at the utility power plant. Typical emissionreductions per MWh of energy conserved are 0.3 pounds(0.14 kg) of particulates, 3.3 pounds (1.5 kg) of sulfuroxides, 5.3 pounds (2.4 kg) of nitrogen oxides, and 1,720pounds (780 kg) of carbon dioxide. These numbers varywith time and region, depending on the power generationfuel mix (EPA 1994; Nemeth 1993).28.7 HYPOTHETICAL CASE STUDIESThe purpose of these hypothetical case studies is toassist the energy manager or facility engineer in estimatingthe energy consumption and costs associated with theconstruction and operation of ground-source heat pumpsystems and comparing them with those for conventionalHVAC technologies. The goal is to estimate energy consumptionand savings, not to design systems.There are several methods for estimating energyconsumption of HVAC technologies, from simplisticdegree-day calculations to sophisticated hour-by-hourenergy modeling and simulation systems supported bycomputer programs. The examples used in this chapterare based on an outdoor temperature bin method. Thismethod is described in more detail in Closed-Loop/Closed-Source Heat Pump Systems: Installation Guide(OSU 1988).Two case studies were developed for this chapter. Inboth, estimates of the potential energy consumption andlife-cycle costs of the ground-source heat pump technologyare compared with conventional HVAC technologies.In each study, the facility is a hypothetical administrativebuilding of typical single-story construction. The facilitiesin each example operate continuously (no night setbacktemperature control). In the first example, the building islocated in upstate New York; in the second example, thesame building is located in central Oklahoma.28.7.1 Example 1: Upstate New York FacilityThe hypothetical administrative office facility is inupstate New York. Mean local weather conditions are7,331 (base 65°F) heating degree-days (4,073 heating Celsiusdegree-days) and 472 (base 65°F) cooling degree-days(262 cooling Celsius degree-days). The 97.5% heating designtemperature is -5°F (-20.6°C) and the 2.5% coolingdesign temperature is 85°F (29.4°C). The mean annualearth temperature is 49°F (9.4°C).The electric rate schedule for this example consistsof a demand charge of $5.51/kW-month and an energy

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 775charge of $0.05/kWh. The gas rate schedule consists of anenergy charge of $0.54/therm.The heating load for this facility during the designday is estimated to be 3,309 kBtu/h (970.6 kW). The designcooling load is 1,273 kBtu/h (373.4 kW). The balancetemperature for the facility is 60°F (15.6°C). The balancetemperature is the temperature at which neither heatingnor cooling is required to meet the comfort requirementsof the facility.Technology DescriptionThree technologies are compared in this example.The conventional technology is a rooftop air-conditioningsystem with integral natural gas furnace heaters. Forthis example, the capacity and input power are based ona commercially available rooftop unit with a rated heatingcapacity of 270 kBtu/h (79.2 kW) and a rated coolingcapacity of 162 kBtu/h (47.5 kW) at ARI conditions. Theequipment selected has a cooling efficiency of 8.2 EERand 10.1 IPLV 11 at ARI conditions. The heating efficiencyis rated at 80%. The indoor air fan is rated at 2.50 kW inputand the outdoor air fans are rated at 2.67 kW input.To meet the heating requirements during the design day,13 rooftop units are required. This number also meetsthe cooling load requirements. The estimated equipmentlife of this alternative, according to ASHRAE, is 15years (ASHRAE HVAC Applications Handbook, 2005, pp36.3).The second alternative is a rooftop air-source heatpump system with electric resistance supplemental heaters.For this example, the capacity and input power arebased on a commercially available air-source heat pump.The rooftop units have a rated cooling capacity of 162kBtu/h (47.5 kW) and a rated heating capacity of 166kBtu/h (48.7 kW) (high temperature) and 102 kBtu/h(29.9 kW) (low temperature) at ARI Standard 340 testconditions. The equipment selected has a rated coolingefficiency of 8.5 EER and a rated heating efficiency of 3.0COP (high temperature) and 2.1 COP (low temperature).The indoor and outdoor air fan input power is included inthe power input loads. To meet the cooling requirementsduring the design day, seven rooftop units are required.To meet the added load during the heating design day,each rooftop unit is equipped with 125-kW supplementalheaters. The estimated equipment life of this alternative,according to ASHRAE, is 15 years (ASHRAE HVAC ApplicationsHandbook, 2005, pp 36.3).The third alternative is a ground-source heat pumpsystem using extended-range water-source heat pumpsand a vertical closed-loop ground-coupling system. For11 EER and IPLV have units of Btu/W-h in this example.this example, the capacity and input power are based ona commercially available ground-source heat pump. Theheat pumps have a rated cooling capacity of 57.5 kBtu/h(16.9 kW) and a rated heating capacity of 60.0 kBtu/h (17.6kW) at ARI 330 conditions. The equipment selected has arated cooling efficiency of 14.49 EER and a rated heatingefficiency of 4.0 COP. The indoor air fan input power isincluded in the power input loads. To meet the cooling requirementsduring the design day, 22 units are required.To meet the added load during the heating design day,each unit is equipped with a 33-kW supplemental heater.It is also assumed that the water-loop system employsa variable-speed drive for added energy savings. Theestimated equipment life of this alternative, accordingto ASHRAE, is 19 years (ASHRAE HVAC ApplicationsHandbook, 2005, pp 36.3).Savings PotentialEnergy consumption for the three alternatives isestimated by an outdoor temperature bin method. Thecalculations are performed in a spreadsheet using 5°F bindata because this is a form in which data are readily available(TM 5-785).The main assumption underlying this type of analysisis that the facility is “thermally light.” This implies thatthe heating and cooling loads on the building are proportionalto the outside air temperature. A “thermally heavy”facility would be a building in which the cooling load onthe building is not very proportional to the outside airtemperature because of significant internal heat loads ora thermally massive structure, which tends to “hold” theheat or cold. In the case of a “thermally heavy” facility,the building heating and cooling loads would have to becalculated using a methodology other than the one utilizedin the following analysis.The analysis for the ground-source heat pump optionshown in Table 28.3 begins with an estimate of the buildingload for each bin. Column 1 indicates the midpoint ofeach temperature bin, e.g., the bin from 95° to 99°F has amidpoint temperature of 97°F. Column 2 gives the numberof hours that occur in each bin over a “typical” yearfor that location. Column 3 estimates the correspondingbuilding load. This is calculated by linear interpolationbetween no heating or cooling load at the facility balancetemperature and the design heating or cooling load at thedesign heating or cooling temperature. The actual equationsused in this analysis are listed in Table 28.4The entering water temperature (EWT) is estimatedin Column 4 of Table 28.3. This is the temperature of thewater entering the heat pump from the ground-couplingsystem. For initial estimating, the entering water temperatureis a linear interpolation between two points for

776 ENERGY MANAGEMENT HANDBOOKTable 28.3. Upper New York Facility Ground-Source Heat Pump Energy Consumption Bin Method AnalysisExamplethe heating cycle and again for the cooling cycle. Duringzero building load conditions, which occurs at the facilitybalance temperature, the entering water temperature isassumed to be equal to the ground temperature. Duringthe peak load, the entering EWT is set to the maximumtemperature desired during the peak cooling season load;similarly, it is set to the minimum temperature desiredduring the peak heating season load. Publications by theInternational Ground-Source Heat Pump Association recommendfor initial calculations these temperatures be setto 100°F maximum (37.8°C) and 37°F (2.8°C) minimum,although in this example, the minimum temperature wasset to 35°F (1.7°C). Today, most commercial systems arecommonly designed at 90°F (32.2°C) maximum EWT. Theestimated entering water temperatures are refined later inthe design process, as the ground-coupling system designis finalized.Column 5 in Table 28.3 is the net capacity of theground-source heat pump equipment. This is taken fromequipment specifications based on the entering watertemperature. Columns 6, 7, and 8 are used to determinethe loss in efficiency caused by part-load operating characteristicsof equipment, which result in increased runtime on the units. The part-load factor is dependent onthe equipment “degradation factor.” The degradation factoris typically assumed to be 0.25 unless the equipmentmanufacturer has tested the unit and determined a lowervalue.Column 10 is the input power of the ground-sourceheat pump equipment. Like the net capacity, this is takenfrom equipment specifications based on the entering watertemperature. Column 9 is the efficiency of the groundsourceheat pump alone and is determined by dividingthe net capacity by the input power and correcting forthe appropriate units of measure. The cooling efficiencyis typically expressed as the energy-efficiency ratio ( EER)and has units of Btu/watt-h. Heating efficiency is typicallyexpressed as the coefficient of performance ( COP)and is a dimensionless measure.Column 11 is the average input power required bythe supplemental heaters to meet the estimated buildingheating load. It is estimated by taking the difference betweenthe building heating load and the capacity of theheat pumps. When the ground-source heat pumps canmeet the building load, there is no load on the supplementalheaters. The maximum load does not exceedthe installed capacity of the supplemental heaters. Notincluded in this simplified example is an air-to-air heat

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 777Table 28.4. Summary of Equations for Bin Method Energy Consumption Analysisrecovery system that would reduce the need for supplementalheating. These recovery systems are common incold region applications and can be specified to be part ofthe basic rooftop unit.The average pump flow rate in Column 12 is estimatedbased on the number of heat pumps, the actualrun time (Column 8), and the rated flow per unit. Mostground-source heat pumps are rated at around 3 gpmper ton of cooling (0.054 L/s-kW), although this varies bymanufacturer.The ground-source heat pump energy consumption,Column 13, is the result of the input power multipliedby the percent run time multiplied by the numberof hours in each bin (Column 10 x Column 8 x Column 2).The supplemental heater energy consumption is similarlyestimated as the result of the supplemental heater powermultiplied by the run time multiplied by the number ofhours in each bin (Column 11 x Column 8 x Column 2).The energy consumption of the ground-coupling loopis dependent on the configuration of the ground-couplingand control systems. For this example, it is assumed the netpressure rise of the circulation pump at full design flow is48 psi and, as a result of the control system, the variablespeeddrive reduces the energy requirement as a squarefunction of the flow rate (assumes automatic water shut-offvalves at each heat pump that close when the unit is off).The total energy consumption (Column 16) is the sum ofthe component energy consumptions (Column 13 + Column14 + Column 15). For the ground-source heat pumpsystem in this example, the annual energy consumption is

778 ENERGY MANAGEMENT HANDBOOKestimated to be 1,413,207 kWh/yr. The monthly-billed demandis estimated by examining the corresponding temperaturebin during the time the monthly peak demand istypically set for the facility. For this example, it is estimatedthat the demand is 4,355 kW-month/yr.Not included in the sample equipment design is anair-to-air heat recovery system that would reduce the needfor supplemental electric resistance heating. Heat pumpmanufacturers make these units available as part of therooftop unit for installations in colder northern climates.Similar spreadsheets were developed for each ofthe other technology alternatives using a standard binmethod. Table 28.5 shows the analysis for a conventionalair-conditioning with natural-gas furnace option. Theanalysis for an air-source heat pump option is shown inTable 28.6.Life-Cycle CostThe total installation cost, including material, labor,overhead, and profit, for the ground-source heat pumpoption is $329,350 compared with the conventional systemat $454,100 and the air-source heat pump systemat $213,000. The operations and maintenance cost forthe ground-source heat pump system is estimated tobe $3,700/yr compared with $8,775/yr for the conventionalsystem and $3,300/yr for the air-source heat pumpsystem. Using the Building Life-Cycle Cost 12 softwareavailable from the National Institute for Standards andTechnology (NIST), the total life-cycle costs for the threealternatives are estimated at $1,502,942 for the groundsourceheat pump system, $1,516,482 for the air-sourceheat pump system, and $1,639,262 for the conventionalsystem. A common life-cycle of 15 years was used in thisanalysis. A summary of the energy and cost factors andthe results of the life-cycle cost analysis for the three alternativesare shown in Table 28.7.————————12 BLCC is available at no charge from the DOE Federal Energy ManagementProgram web site at http://www.eere.energy.gov/femp/. Asof November 2005, BLCC version 5.3-05 was the most current versionavailable.Table 28.5. Upper New York Facility Conventional Air-Conditioning and Gas Furnace System Bin MethodAnalysis Example

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 779Table 28.6. Upper New York Facility Air-Source Heat Pump System Bin Method Analysis Example28.7.2 Example 2: Oklahoma City FacilityThis example places the same building in a newenvironment. The location of the facility is in OklahomaCity, Oklahoma. Typical local weather conditions are3,588°F-day/yr (base 65°F) heating degree-days (1,993heating Celsius degree-days) and 2,068 (base 65°F) coolingdegree-days (1,149 cooling Celsius degree-days). The97.5% heating design temperature is 13°F (-10.6°C) andthe 2.5% cooling design temperature is 96°F (35.6°C). Themean annual earth temperature is 62°F (16.7°C).The electric rate schedule for this example consistsof a summer demand charge of $12.35/kW-month, a winterdemand charge of $4.48/kW-month, and an energycharge of $0.0236/kWh. The gas rate schedule consists ofan energy charge of $3.40/mcf ($0.34/therm).The corresponding heating load for this facility duringthe design day is estimated to be 2,392 kBtu/h (701.7kW). The design cooling load is 1,832 kBtu/h (537.4 kW).The same balance temperature of 60°F (15.6°C) is estimatedfor the facility.Technology DescriptionThe same three technologies are compared in thisexample, but the number of units has changed becausethe building loads are now different. The conventionaltechnology is a rooftop air-conditioning system withintegral natural gas furnace heaters. To meet the coolingrequirements during the actual cooling design day, 10rooftop units are required. This number also meets theheating load requirements.The second alternative is a rooftop air-source heatpump system with electric resistance supplemental heaters.To meet the cooling requirements during the actualcooling design day, 10 rooftop units are required. To meetthe added load during the heating design day, each rooftopunit is equipped with a 50-kW supplemental heater.The third alternative is a ground-source heat pumpsystem using extended range water-source heat pumpsand a vertical closed-loop ground-coupling system. Tomeet the cooling requirements during the actual coolingdesign day, 35 units are required. To meet the added loadduring the heating design day, each unit is equipped witha 10-kW supplemental heater. The water-loop system employsa variable-speed drive for added energy savings.Savings PotentialEnergy consumption is estimated with the samespreadsheet temperature bin model as the previous case.

780 ENERGY MANAGEMENT HANDBOOKTable 28.7. Example 1: Summary of Results———————————————————————————————————————————Conventional Air-Source Ground-SourceUpstate New York Facility system Heat Pump Heat Pump———————————————————————————————————————————Number of units 13 7 22———————————————————————————————————————————Nominal capacity (tons)Each 13.5 13.5 4.8Total 175.5 94.5 105.6———————————————————————————————————————————Supplemental heaters (kW)Each n/a 125 33Total n/a 875 726———————————————————————————————————————————Equipment capacity (kBtuh)(at design conditions)Summer 2,535.0 1,360.1 1,270.1Winter 3,510.0 3,395.0 3,336.6———————————————————————————————————————————Energy consumption (/yr)Electricity (kWh) 252,908 1,656,555 1,413,207Demand* (kW-mo) 1,481 4,200 4,355Natural gas (therm) 110,380 0 0Total energy (10 6 Btu) 11,901 5,562 4,822———————————————————————————————————————————Energy Cost ($/yr)Electricity 12,645 82,828 70,660Demand 8,160 23,142 23,996Natural gas 59,605 0 0Total energy 80,411 105,970 94,656———————————————————————————————————————————O&M Costs ($/yr) 8,775 3,300 3,700———————————————————————————————————————————Installed Costs (S) 454,100 212,500 329,300———————————————————————————————————————————Equipment life (yr) 15 15 15———————————————————————————————————————————Total life-cycle cost ($) 1,639,262 1,516,482 1,502,942———————————————————————————————————————————*Assumptions are required to estimate the monthly billing demand. In this case, the authors matcheda temperature bin to when the peak demand window was likely to occur. A sensitivity analysis could beperformed by changing the assumptions.The results for the ground-source heat pump option areshown in Table 28.8. Table 28.9 shows the analysis for theconventional air-conditioning and natural-gas furnaceoption. Table 28.10 shows the analysis for the air-sourceheat pump option.Life-Cycle CostThe total installation cost for the ground-sourceheat pump option (including material, labor, overhead,and profit) is estimated to be $451,500, compared with theconventional system at $349,300 and the air-source heatpump system at $236,000. The operations and maintenancecost for the ground-source heat pump system is estimatedto be $5,880/yr, compared with $6,750/yr for theconventional system and $4,725/yr for the air-source heatpump system. Through BLCC, the total life-cycle costs forthe three alternatives were estimated to be $938,282 forthe ground-source heat pump system, $800,065 for theair-source heat pump system, and $974,924 for the conventionalsystem. A summary of the energy and cost factorsand the results of the life-cycle cost analysis for eachof the three options are shown in Table 28.11.The fact that the ground-source heat pump systemdid not have the lowest life-cycle cost in this evaluationshould not imply that ground-source heat pumps arenever cost effective in this region. On the contrary, thepurpose of this example is to illustrate that the cost effectivenessof ground-source heat pumps is a function ofsystem heating and cooling load requirements, local utilitycosts, and most importantly, local installation costs.

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 781Table 28.8. Oklahoma City Facility Ground-Source Heat Pump Energy Consumption Bin Method AnalysisExampleThe reader is encouraged to perform similar calculationsfor a specific site.The above examples assume that the facility willresponsible for all costs associated with the constructionand operation of the installed ground-source heat pumpsystem. However, if the equipment were installed underan energy-savings performance contract (ESPC) program,the installation cost could be zero and the economic justificationfor the project would be significantly different.28.8 THE TECHNOLOGY IN PERSPECTIVEThe future of ground-source heat pump technologylooks good because there are many potential commercialapplications. Although installation costs aretypically higher for ground-source heat pumps than forother technologies, the decision criteria should be basedon life-cycle costs rather than first costs; then, a groundsourceheat pump system can be the most cost-effectivealternative. According to the EPA study, Space Conditioning:The Next Frontier, ground-source heat pumps areconsistently the most energy-efficient, least pollutingof all space conditioning technologies throughout thecountry (EPA 1993).The limited industry volume has been one of the factorsholding back the development of a broader contractorbase. The technology has not enjoyed broad nationalpromotion. According to the results of one survey on thebarriers to ground-source heat pumps, HVAC, plumbing,and Architectural/Engineering contractors are conservative,and are therefore reluctant to commit to what theyregard as innovative and (to some) unproven technologyand equipment (Technical Marketing Associates 1988).Installation of ground-source heat pump systems requiresskills beyond those of most HVAC or plumbing contractors.Until recently, only a small number of contractorshave performed enough installations to develop an extensivebase of experience and expertise in ground-sourceheat pumps.28.8.1 The Technology’s DevelopmentThe ground-source heat pump technology has beenshown through laboratory testing, field testing, and theoreticalanalysis to be technically valid and economicallyattractive in many applications. Energy savings have beenverified in a large number of field tests over the past 30years. In several installations, reductions in maintenance

782 ENERGY MANAGEMENT HANDBOOKTable 28.9. Oklahoma City Facility Conventional Air-Conditioning and Gas Furnace System BinMethod Analysis Examplecosts have also been verified. The technology has gainedrapid acceptance from users.The remaining barriers to rapid implementation include(1) acceptance from new users and engineers unfamiliarwith the technology, (2) out-of-date cost estimatingguides for construction and maintenance, (3) general lackof engineers able to design ground-source heat pumpsystems, and (4) lack of contractors to install and serviceground-source heat pump systems. This chapter is intendedto address some of these concerns by reportingon the collective experience of ground-source heat pumpusers and evaluators and by providing application guidance.For actual design guidance, readers should refer toany of the many publications available, some of which arelisted at the end of this chapter.In addition to the U.S. Department of Energy andthe Environmental Protection Agency, two associationsare working to address the barriers to increased utilizationof ground-source heat pumps: the InternationalGround Source Heat Pump Association (www.igshpa.okstate.edu) and the Geothermal Heat Pump Consortium(www.geoexchange.org).The International Ground Source Heat Pump Association(IGSHPA) is a non-profit, member-driven organizationestablished in 1987 to advance ground-source heatpump technology on local, state, national and internationallevels. IGSHPA conducts ground-source heat pumpsystem installation training and geothermal research. Themission of the International Ground Source Heat PumpAssociation and its membership is to promote the use ofground-source heat pump technology worldwide througheducation and communication.The government-industry-utility consortium,Geothermal Heat Pump Consortium, Inc., was formedin 1994 to create a self-sustaining ground-source heatpump market. The Geothermal Heat Pump Consortiumis a nonprofit organization working to raise awarenessand increase the use of ground-source heat pump technologythroughout the United States. As a cooperativeventure, the Consortium counts among its partners electricutilities, equipment manufacturers, architects, designers,engineers, contractors, builders, drillers, energy

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 783Table 28.10. Oklahoma City Facility Air-Source Heat Pump System Bin Method Analysis Exampleservice companies, and other private sector companiesthat operate in the ground-source heat pump market aswell as national, state and local organizations and publicagencies.28.8.2 Technology OutlookThe outlook for ground-source heat pumps is bright.In 1999, an estimated 400,000 ground-source heat pumpswere operating in residential and commercial applications,up from 100,000 in 1990. In 1985, it was estimatedthat only around 14,000 ground-source heat pump systemswere installed in the United States. Annual sales ofapproximately 45,000 units were reported in 1997. With aprojected annual growth rate of 10%, 120,000 new unitswould be installed in 2010, for a total of 1.5 million unitsin 2010.In Europe, the estimated total number of installedground-source heat pumps at the end of 1998 was 100,000to 120,000. Nearly 10,000 ground-source heat pumps havebeen installed in U.S. federal buildings, over 400 schools,and thousands of low-income houses and apartments.28.9 MANUFACTURERSThere are a number of manufacturers of groundsourceheat pumps and water-source heat pump that canbe ground coupled. The following is list of U.S. manufacturersidentified on the Geothermal Heat Pump Consortiumweb site 13 .Advanced Geothermal TechnologyP.O. Box 6469Reading, PA 19610Tel: (610) 796-1450Fax: (610) 736-0571American Geothermal, Inc.1037 Old Salem RoadMurfreesboro, TN 37129Tel: (615) 890-6985; Tel: (800) 776-8039Fax: (615) 890-6926www.amgeo.com13 List obtained from the Geothermal Heat Pump Consortium, Inc. website, www.geoexchange.org/local/manufacturers.htm (last accessed 1November 2005).

784 ENERGY MANAGEMENT HANDBOOKTable 28.11. Example 2: Summary of Results———————————————————————————————————————————Conventional Air-Source Ground-SourceOklahoma City Facility system Heat Pump Heat Pump———————————————————————————————————————————Number of units 10 10 35———————————————————————————————————————————Nominal capacity (tons)———————————————————————————————————————————Each 13.5 13.5 4.8Total 135 135 168———————————————————————————————————————————Supplemental heaters (kW)Each n/a 50 10Total n/a 500 350———————————————————————————————————————————Equipment capacity (kBtuh)(at design conditions)Summer 1,848.0 1,822.7 1,864.6Winter 2,700.0 2,621.0 2,611.5———————————————————————————————————————————Energy consumption (/yr)Electricity (kWh) 467,708 850,469 687,333Demand* (kW-mo)- Winter season 455 2,748 1,931- Summer season 842 770 70Natural gas (therm) 54,081 0 0Total energy (10 6 Btu) 7,004 2,902 2.345———————————————————————————————————————————Energy Cost ($/yr) Electricity 11,038 20,071 16,221Demand 12,437 21,821 18,160Natural gas 18,388 0 0Total energy 41,863 41,892 34,381———————————————————————————————————————————O&M Costs ($/yr) 6,750 4,725 5,880———————————————————————————————————————————Installed Costs (S) 349,300 236,000 451,500———————————————————————————————————————————Equipment life (yr) 15 15 15———————————————————————————————————————————Total life-cycle cost ($) 974,924 800,065 938,282———————————————————————————————————————————*Assumptions are required to estimate the monthly billing demand. In this case, the authors matched atemperature bin to when the peak demand window was likely to occur. A sensitivity analysis could beperformed by changing the assumptions.ClimateMaster7300 SW 44th StreetP.O. Box 25788Oklahoma City, OK 73125Tel: (405) 745-6000Fax: (405) 745-3629www.climatemaster.comEarth To Air Systems, Inc.123 SE Parkway Court, Ste 160Franklin, TN 37064 - USAPh: (615) 595-2888Fax: (615) 595-1080www.earthtoair.comEconar Energy Systems Corp.19230 Evans StreetElk River, MN 55330Tel: (612) 241-3110Tel: (800) 4-ECONARFax: (612) 241-3111www.econar.comECR Technologies3536 DMG DriveLakeland, FL 33811Tel: (863) 701-0096Fax: (863) 701-7796www.ecrtech.com

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 785Enertran Technologies, Inc.517 McCormick Blvd.London, Ontario N5W 4C8CanadaTel: (519) 452-7049Fax: (519) 452-0106www.enertran.caFHP Manufacturing601 NW 65th CourtFt. Lauderdale, FL 33309Tel: (954) 776-5471Fax: (954) 776-5529www.fhp-mfg.comHydroDelta CorporationTel: (724) 728-4230Tel: (412) 373-58001000 Rico RoadMonroeville, PA 15146www.hydrodelta.comHydro-Temp CorporationP.O. Box 566Pocahontas, AR 72455Tel: (870) 892-8343Tel: (800) 382-3113Fax: (870) 892-8323http://www.hydro-temp.comMcQuay International4900 Technology Park Blvd.Auburn, NY 13021Tel: (315) 282-6243Fax: (315) 282-6417www.mcquay.comMillbrook Industries - Hydronic Division41659 256th StreetRR #3, Box 265Mitchell, SD 57301Tel: (605) 995-0241Fax: (605) 996-8837www.hydronmodule.comThe Trane Company182 Cotton Belt ParkwayMcGregor, TX 76657Tel: (254) 299-6329Fax: (817) 299-6671www.trane.comWaterFurnace International9000 Conservation WayFort Wayne, IN 46809Tel: (260) 478-5667 ext. 3221Tel: (800) 222-5667Fax: (260) 478-3029www.waterfurnace.com28.10 FOR FURTHER INFORMATIONTrade AssociationsAmerican Society of Heating, Refrigerating, Refrigeratingand Air Conditioning Engineers (ASHRAE)1791 Tullie Circle, NEAtlanta, GA 30329Tel: (404) 636-8400Fax: (404) 321-5478www.ashrae.orgCanadian Earth Energy Association130 Slater Street, Suite 1050Ottawa, ON Canada K1P6E2Tel: (613) 230-2332Fax: (613) 237-1480 Faxwww.earthenergy.caElectric Power Research Institute (EPRI)P.O. Box 504903412 Hillview Ave.Palo Alto, CA 94303www.epri.comGeothermal Heat Pump Consortium, Inc. (Geoexchange)1050 Connecticut Ave, NW, Suite 1000Washington, D.C.Tel: (202) 558-7175Fax: (202) 558-6759www.geoexchange.orgGeothermal Resource CouncilP.O. Box 1350Davis, CA 95617www.geothermal.orgInternational Ground-Source Heat Pump Association(IGSHPA)498 Cordell SouthOklahoma State UniversityStillwater, OK 74078Tel: (800) 626-4747Fax: (405) 744-5283www.igshpa.okstate.edu

786 ENERGY MANAGEMENT HANDBOOKNational Rural Electric Cooperative Assn. (NRECA)1800 Massachusetts Ave., NWWashington, D.C. 20036www.nreca.orgNational Ground Water Association601 Dempsey RoadWesterville, OH 43801Tel: (800) 551-7379www.ngwa.orgMichigan Geothermal Energy Association2859 West Jolly RoadOkemos, MI 48864Tel: (800) 417-5555www.earthcomfort.comNewsletters“Heat Pump News Exchange,” Electric Power ResearchInstitute (EPRI), publishedquarterly and distributed by EPRI, Palo Alto, California.“The Source,” International Ground Source Heat PumpAssociation, published quarterly and distributed byOklahoma State University Ground Source Heat PumpPublications, Stillwater, Oklahoma.“Geo-Heat Center Quarterly Bulletin,” Geo-Heat Center,published quarterly and distributed by Geo-Heat Center,Oregon Institute of Technology, KlamathFalls, OR.“Down to Earth,” published by the Michigan GeothermalEnergy Association, Okemos, MI.“Outside the Loop,” published quarterly by the Universityof Alabama, Tuscaloosa, Alabama (www.oit.edu/~geoheat/otl/index.htm).User and Third-Party Field and Lab Test Reports 14Collie, M.J. (editor) 1979. “Comparison of Water-Sourceand Air-Source Heat Pumps in Northern Environment,”a chapter in Heat Pump Technology for Saving Energy.Noyes Data Corporation, Park Ridge, New Jersey.*Hughes, P.J. and J.A. Shonder. 1998. The Evaluation of a4000-Home Geothermal Heat Pump Retrofit at Fort Polk,Louisiana: Final Report. ORNL/CON-460, Oak RidgeNational Laboratory, Oak Ridge, Tennessee.14 *Denotes literature cited in the technical body of this chapter.Phetteplace, G., H. Ueda, and D. Carbee. 1992. “Performanceof Ground-Coupled Heat Pumps in Military FamilyHousing Units.” Solar Engineering. G0656A-1992,American Society of Mechanical Engineers, New York,NY.Phetteplace, G. 1995. Ground-Coupled Heat Pumps forFamily Housing Units. FEAP-UG-CRREL-95/01, U.S.Army Center for Public Works, Alexandria, Virginia.Svec, O.J. 1987. “Potential of Ground Heat Source Systems.”International Journal of Energy Research, Vol. 11,pp. 573-581.Design and Installation Guides*ASHRAE. 1995. “Chapter 29: Geothermal Energy.”ASHRAE Handbook: Heating, Ventilating, and Air-ConditioningApplications. American Society of Heating,Refrigerating, and Air- Conditioning Engineers, Inc. Atlanta,Georgia.Bose, J.E., J.D. Parker, and F.C. McQuiston. 1985. Design/Data Manual for Closed-Loop Ground-Coupled HeatPump Systems. American Society of Heating, Refrigerating,and Air-Conditioning Engineers, Inc., Atlanta, Georgia.Caneta Research, Inc. 1995. Commercial/InstitutionalGround-Source Heat Pump Engineering Manual. AmericanSociety of Heating, Refrigerating, and Air-ConditioningEngineers, Inc., Atlanta, Georgia.*International Ground-Source Heat Pump Association(not dated). Ground Source Systems: Design and InstallationStandards. Oklahoma State University, Stillwater,Oklahoma.*Oklahoma State University. 1988. Closed-Loop/Ground-Source Heat Pump Systems: Installation Guide. InternationalGround-Source Heat Pump Association, Stillwater,Oklahoma.*STS Consultants. 1989. Soil and Rock Classification forthe Design of Ground-Coupled Heat Pump Systems: FieldManual. CU-6600, Electric Power Research Institute, PaloAlto, California.*WaterFurnace. October 1991. Polyethylene Pond/LakeLoop, Mat Design. WF390, WaterFurnace International,Fort Wayne, Indiana.

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 787Manufacturer’s Application NotesCommand-Aire. 1988. Earth Energy Heat Pumps: CommercialWater Source Heat Pump Systems; Application,Installation, Operation Manual. Bulletin C-A10-0191.Guarino, D.S. March 1993. “WaterFurnace: From theGround Up.” Reprint from Contracting Business.Utility, Information Service, orGovernment Agency Tech-Transfer LiteratureBas, E. February 13, 1995. “Geothermal Market ReceivesHefty Boost from Consortium.” The Air Conditioning,Heating, and Refrigeration News, p.1.*Calm, J.M. September 1987. Proceedings of the Workshopon Ground-Source Heat Pumps. From the workshopheld in Albany, New York, October 27 through November1, 1986. HPC-WR-2. International Energy Agency HeatPump Center.Goldfish, L.H., and R.A. Simonelli. 1988. Ground-Sourceand Hydronic Heat Pump Market Study. EM-6062, ElectricPower Research Institute, Palo Alto, California.Ground Source Systems: Educational and Marketing MaterialCatalog. International Ground Source Heat PumpAssociation, Stillwater, Oklahoma.GS-Systems: An Answer to U.S. Energy and EnvironmentalConcerns: A comprehensive report on ground-sourceheat pumps and their benefits (not dated). InternationalGround-Source Heat Pump Association, Oklahoma StateUniversity, Stillwater, Oklahoma.*Pratsch, L.W. 1990. “Geothermal Heat Pumps: A MajorOpportunity for the Utility Industry.” Presented at theGeothermal Energy Conference, Columbus, Ohio, November14, 1990.Scofield, M., and P. Joyner. September 1991. “Heat Pumpsfor Northern Climates.” EPRI Journal 16(6), pp. 28-33.Case StudiesDuffy, G. March 22, 1993. “120-ton geothermal system replaces30-year-old relics.” The Air Conditioning, Heating,and Refrigeration News.*Gahran, A. September 1993. “Grants, Util. Rebate Pay97% of Ground-Source Heat Pump Project Cost.” EnergyUser News 18(9) pp. 10, 51, 72.*Gahran, A. January 1994. “New Ground-Source HeatPumps Cut Energy, Maintenance Costs $68K.” EnergyUser News 19(1) p. 16.Greenleaf, J. July-September 1982. “College invests in anenergy-efficient future.” Brown and Caldwell Quarterly.KCPL. 1994. A Model of Efficiency from the Ground Up.Commercial case study 2M/3-135, Kansas City Powerand Light, Kansas City, Missouri.KCPL. 1994. Innovative Heating and Cooling System AllowsIncreased Efficiencies to Surface. Commercial casestudy 4-132, Kansas City Power and Light, Kansas City,Missouri.KCPL. 1994. Station’s Ground Source Heat Pump SystemFinds Efficiency in Unusual Places. Commercial casestudy 4-118, Kansas City Power and Light, Kansas City,Missouri.PSO. “Facility Type: Family Fitness Center, TechnologyApplication: Ground- Source Heat Pump.” Power Profile.Number 1, Public Service Company of Oklahoma, Tulsa,Oklahoma.Randazzo, M. November 1994. “Retrofit Cut Elec. Use by32MMkWh/Yr.” Energy User News 19(11) pp. 1,14.WaterFurnace. Dallas’ “Hope of the Sun” goes undergroundfor affordable, efficient, environmental housing.Case Study 5, WF915, WaterFurnace International, FortWayne, Indiana.*WaterFurnace. Elementary school teaches lesson in efficiency.Case Study 2, WF666, WaterFurnace International,Fort Wayne, Indiana.WaterFurnace. Geothermal comfort goes low-rise in Kisilano,British Columbia. Case Study 8, WF923, WaterFurnaceInternational, Fort Wayne, Indiana.WaterFurnace. John Deere dealership comforts its clientelefrom the floor up. Case Study 11, WF922, WaterFurnaceInternational, Fort Wayne, Indiana.WaterFurnace. Last century’s structure retrofitted for thenext. Case Study 7, WF924, WaterFurnace International,Fort Wayne, Indiana.WaterFurnace. Patients treated to geothermal comfort atDupont Medical Center. Case Study 12, WF942, Water-Furnace International, Fort Wayne, Indiana.

788 ENERGY MANAGEMENT HANDBOOKWaterFurnace. Salem Community college enhances environmentand education. Case Study 4, WF907, WaterFurnaceInternational, Fort Wayne, Indiana.*WaterFurnace. Three Indiana schools compare HVACnotes. Case Study 9, WF925, WaterFurnace International,Fort Wayne, Indiana.*WaterFurnace. WaterFurnace builds state-of-the-art facility.Case Study 1, WF639, WaterFurnace International,Fort Wayne, Indiana.WaterFurnace. WaterFurnace goes down under, down under.Case Study 10, WF921, WaterFurnace International,Fort Wayne, Indiana.WaterFurnace. Whitehawk Ranch looks earthward forheating and cooling. Case Study 6, WF916, WaterFurnaceInternational, Fort Wayne, Indiana.WaterFurnace. York County contributes to a healthierenvironment. Case Study 3, WaterFurnace International,Fort Wayne, Indiana.Codes and Standards*Energy-Efficient Design of New Buildings ExceptLow-Rise Residential Buildings. 2004. ASHRAE/IESNAStandard 90.1-2004, American Society of Heating, Refrigerating,and Air- Conditioning Engineers, Inc., Atlanta,Georgia.*Water-source Heat Pumps – Testing and Rating forPerformance – Part 1: Water-to-air and Brine-to-air HeatPumps. 2005. ANSI/ARI/ASHRAE ISO Standard 13256-1-2005, American Society of Heating, Refrigerating, andAir- Conditioning Engineers, Inc., Atlanta.*Ground Source Closed-Loop Heat Pumps. 1998. ANSI/ARI Standard 330-98, Air-Conditioning and RefrigerationInstitute, Inc., Arlington, Virginia.*Ground Water-Source Heat Pumps. 1998. ARI Standard325-98. Air-Conditioning and Refrigeration Institute, Inc.,Arlington, Virginia.*Water-Source Heat Pumps. 1998. ANSI/ARI Standard320-98. Air-Conditioning and Refrigeration Institute, Inc.,Arlington, Virginia.Other SourcesAmerican Society of Heating, Refrigerating, and Air-ConditioningEngineers (ASHRAE). 1997. Ground Source HeatPumps — Design of Geothermal Systems for Commercialand Institutional Buildings. ASHRAE, Atlanta, GA.Cane, D., A. Morrison, and C. Ireland. 1998. “Maintenanceand Service Costs of Commercial Building Ground-SourceHeat Pump Systems.” ASHRAE Transactions, Vol 104(2):699-706.Duffy, G. July/August 1989. “Commercial Earth-CoupledHeat Pumps Are Ready To Roll.” Engineered Systems,pp. 40-47.*Electric Power Research Institute (EPRI). May 1993. 1992Survey of Utility Demand-Side Management Programs.EPRI-TR 102193, Palo Alto, California.*EPA. 1993. Space Conditioning: The Next Frontier. Officeof Air and Radiation, 430-R-93-004 (4/93), EPA, Washington,D.C.*EPA. 1994. Voluntary Reporting Formats for GreenhouseGas Emissions and Carbon Sequestration. (Public ReviewDraft, May 1994). EPA, Washington, D.C.*Geothermal Heat Pumps.” February 1990. CustomBuilder. 5(2) pp. 32-35.*Giddings, T. February 1988. “Ground Water Heat Pumps:Why Doesn’t EverybodyWant One?” Well Water Journal 42(2) pp. 32-33.Gilmore, V.E. June 1988. “Neo-geo heat pump.” PopularScience 232(6) pp. 88-89,112.*Heinonen, E.W., R.E. Tapscott, M.W. Wildin, and A.N.Beall. February 1997. “Assessment of Antifreeze Solutionsfor Ground-Source Heat Pump Systems.” Report908RP, American Society of Heating, Refrigerating, andAir-Conditioning Engineers (ASHRAE), Atlanta.Hurlburt, S. February 1988. “Ground Water Heat Pumps- The Second Wave.” Well Water Journal 42(2) pp. 34-40.Kavanaugh, S., C Gilbreath, J. Kilpatrick. 1995. Cost Containmentfor Ground-Source Heat Pumps, Final Report.University of Alabama, Tuscaloosa.Kavanaugh, S. September 1992. “Ground-coupled heatpumps for commercial buildings.” ASHRAE Journal34(9), 30-37.*Lund, J.W. and T. L. Boyd. 2000. “Geothermal Direct-Usein the United States in 2000.” Geo-Heat Center QuarterlyBulletin, Vol. 21, No. 1 (March), Klamath Falls, Oregon,pp. 1-5.

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 789Martin, M.A., D.J. Durfee, and P.J. Hughes. 1999. “ComparingMaintenance Costs of Geothermal Heat PumpSystems with Other HVAC Systems in Lincoln PublicSchools: Repair, Service, and Corrective Action.” ASHRAETransactions, Vol. 105(2): 1208-1216.Martin, M.A., M.G. Madgett, and P.J. Hughes. 2000.“Comparing Maintenance Costs of Geothermal HeatPump Systems with Other HVAC Systems: PreventiveMaintenance Actions and Total Maintenance Costs.”ASHRAE Transactions, Vol. 106(1): 1-15.*Nemeth, R.J., D.E. Fornier, and L.A. Edgar. 1993.Renewables and Energy- Efficiency Planning. U.S. ArmyConstruction Engineering Research Laboratory, Champaign,Illinois.Oak Ridge National Laboratory (not dated). “Big Savingsfrom the World’s Largest Installation of GeothermalHeat Pumps at Fort Polk, Louisiana.” Oak Ridge NationalLaboratory, Oak Ridge, Tennessee.Outside the Loop. 1999. “Impact of Conductivity Error onDesign Results.” Outside the Loop - Volume 2, Number 3,University of Alabama, Tuscaloosa, Alabama.Phetteplace, G., H. Ueda, and D. Carbee. April 1992. “Performanceof Ground-Coupled Heat Pumps in MilitaryFamily Housing Units.” Proceedings of the 1992 ASMEInternational Solar Energy Conference. Maui, Hawaii, 4-8April, 1992, pp. 377-383.Reynolds, William C., and Herry C. Perkins. 1977. EngineeringThermodynamics, pp. 287-289. McGraw-HillBook Co., New York.Rybach, L., and B. Sanner. March 2000. “Ground-SourceHeat Pump Systems the European Experience.” Geo-HeatCenter Quarterly Bulletin 21(1):16-26, Klamath Falls, OR.*Shonder J.A., D. Durfee, M.A. Martin, P.J. Hughes, andT.R. Sharp. 1999. “Benchmark for Performance: GeothermalApplications in Lincoln Public Schools.” ASHRAETransactions 105(2): 1199-1207.Shonder, J. A., M. A. Martin, T. R. Sharp, D. J. Durfee, andP. J. Hughes. 1999. “Benchmark for Performance: GeothermalApplications in Lincoln Public Schools.” ASHRAESE-99-20-03, Atlanta.Shonder, J. A., V. Baxter, J. Thorton, and P. Hughes. June1999. “A New Comparison of Vertical Ground Heat ExchangerDesign Methods for Residential Applications.”ASHRAE Transactions.Shonder, J. A., V. Baxter, P. Hughes, and J. Thorton (draft- no date). “A Comparison of Vertical Ground Heat ExchangerDesign Software for Commercial Applications.”ASHRAE Transactions.Shonder, J.A. and J.V. Beck. 2000. “Field Test of a NewMethod for Determining Soil Formation Thermal Conductivityand Borehole Resistance from In Situ Measurements.”ASHRAE Transactions.*Spilker, E. H. January 1998. “Ground-Coupled HeatPump Loop Design Using Thermal Conductivity Testingand the Effect of Different Backfill Materials on VerticalBore Length.” Proceedings of the American Society ofHeating, Refrigerating, and Air-Conditioning Engineers(ASHRAE). January 1998 meeting, San Francisco, paperSF98-1-3.Swanson, G.L. August 1986. “Heating and Cooling withGWHPs.” Well Water Journal 40(8) pp. 44-46.*Technical Marketing Associates, Inc. 1988. Ground-Source and Hydronic Heat Pump Market Study. EM-6062,Electric Power Research Institute, Palo Alto, California.*TM 5-785. 1978. Engineering Weather Data. Departmentof the Army Technical Manual. TM 5-785, U.S. GovernmentPrinting Office, Washington, D.C.AppendixPERFORMANCE AND EFFICIENCY TERMINOLOGYThere are many terms used in the heating, airconditioning,and refrigeration industry that conveyperformance and efficiency. Many of these terms aresynonymous, others are not. When comparing varioussystems, it is important to understand what the assortedterms are, how they are determined, and their relationship.The following provides a brief description of many(but not all) of the terms used to convey efficiency. Use ofthe terms is also summarized in Table 28.A.1.Annual Fuel Utilization Efficiency (AFUE): For fuel-firedsystems such as boilers and furnaces, AFUE is defined asthe ratio of annual output energy to annual input energy.This term is generally applied to systems

790 ENERGY MANAGEMENT HANDBOOKlosses. AFUE is a unitless term generally expressed as apercentage.Coefficient of Performance (COP): The COP is the basicparameter used to report the efficiency of refrigerantbasedsystems. It is a unitless term. This term is universalin its use but not in its meaning. COP can be used to defineboth cooling efficiency or heating efficiency, such asfor a heat pump. For cooling, COP is defined as the ratioof the rate of heat removal to the rate of energy input tothe compressor, in consistent units. For heating, COP isdefined as the ratio of rate of heat delivered to the rate ofenergy input to the compressor, in consistent units. COPcan be used to define the efficiency at a single (standardor nonstandard) rated condition or a weighted average(seasonal) condition. Depending on its use, the term mayor may not include the energy consumption of auxiliarysystems such as indoor or outdoor fans, chilled waterpumps, or cooling tower systems. For purposes of comparison,the higher the COP the more efficient the system.For mathematical purposes, COP can be treated as an efficiency.(COP of 2.00 = 200% efficient) For unitary heatpumps, ratings at two standard (outdoor) temperatures(47°F and 17°F [8.3°C and -8.3°C]) may be reported basedon old ARI standards.Combustion Efficiency (n c or E c ): For fuel-fired systems,this efficiency term is defined as the ratio of the fuel energyinput minus the flue gas losses (dry flue gas, incompletecombustion and moisture formed by combustion ofhydrogen) to the fuel energy input. In the U.S., fuel-firedefficiencies are reported based on the higher heating valueof the fuel. Other countries report fuel-fired efficienciesbased on the lower heating value of the fuel. The combustionefficiency is calculated by determining the fuelgas losses as a percent of fuel burned. [E c = 1 – (flue gaslosses)]Energy Efficiency Ratio (EER): The EER is a term generallyused to define the cooling efficiency of unitary airconditioningand heat pump systems. The term impliesthat the efficiency is determined at a single rated conditionspecified by the appropriate equipment standard and isdefined as the ratio of net cooling capacity (heat removedin Btu/h) to the total input rate of electric energy required(watt). The units of EER are Btu/w-h. It is important tonote that this efficiency term typically includes the energyrequirement of auxiliary systems such as the indoor andoutdoor fans. For purposes of comparison, the higher theEER the more efficient the system. To convert to a COP,divide the EER by 3.412. [COP = EER/3.412]Heating Seasonal Performance Factor (HSPF): The termHSPF is similar to the term SEER, except it is used to signifythe seasonal heating efficiency of heat pumps. TheHSPF is a weighted average efficiency over a range ofoutside air conditions following a specific standard testmethod. The term is generally applied to heat pump systemsless than 60,000 Btu/h (rated cooling capacity.) Theunits of HSPF are Btu/w-h. It is important to note thatthis efficiency term typically includes the energy requirementof auxiliary systems such as the indoor and outdoorfans. For purposes of comparison, the higher the HSPFthe more efficient the system.Integrated Part-Load Value (IPLV): The term IPLV isused to signify the cooling efficiency related to a typical(hypothetical) season rather than a single rated condition.The IPLV is calculated by determining the weightedaverage efficiency at part-load capacities specified by anaccepted standard. It is also important to note that IPLVsare typically calculated using the same condensing temperaturefor each part-load condition and IPLVs do notinclude cycling or load/unload losses. The units of IPLVare not consistent in the literature; therefore, it is importantto confirm which units are implied when the termIPLV is used. ASHRAE Standard 90.1 (using ARI referencestandards) uses the term IPLV to report seasonalcooling efficiencies for both seasonal COPs (unitless) andseasonal EERs (Btu/w-h), depending on the equipmentcapacity category; and most chiller manufacturers reportseasonal efficiencies for large chillers as IPLV using unitsof kW/ton. Depending on how a cooling system loadsand unloads (or cycles), the IPLV can be between 5 and50% higher than the EER at the standard rated condition.kW/ton: The term kW/ton is generally used for largecommercial and industrial air-conditioning, heat pump,and refrigeration systems. The term is defined as the ratioof the rate of energy consumption (kW) to the rate of heatremoval (ton) at a rated condition. As the term suggests,the units are kW/ton. Because refrigeration systems ofthis capacity are typically custom designed, the reportedkW/ton generally implies only the compressor and doesnot include the auxiliaries. However, for specific references,auxiliaries can be added to report the overall systemefficiency using this term. It is important to note that thisterm is inverse to the other performance and efficiencyterminology. Therefore, for purposes of comparison,the lower the kW/ton the more efficient the system. Toconvert to a COP, divide 12,000 by the product of 3,412multiplied by the kW/ton. [COP = 12,000/{(3,412) * (kW/ton)}]

GROUND-SOURCE HEAT PUMPS APPIED TO COMMERCIAL BUILDINGS 791Seasonal Energy Efficiency Ratio (SEER): The term SEERis used to define the average annual cooling efficiency ofan air conditioning or heat pump system. The term SEERis similar to the term EER but is related to a typical (hypothetical)season rather than for a single rated condition.The SEER is a weighted average of EERs over a range ofrated outside air conditions following a specific standardtest method. The term is generally applied to systems lessthan 60,000 Btu/h. The units of SEER are Btu/W-h. It isimportant to note that this efficiency term typically includesthe energy requirements of auxiliary systems suchas the indoor and outdoor fans. For purposes of comparison,the higher the SEER the more efficient the system.Although SEERs and EERs cannot be directly compared,the SEERs usually range from 0.5 to 1.0 higher than correspondingEERs.Thermal Efficiency (n t or E t ): This efficiency term isgenerally defined as the ratio of the heat absorbed by thewater (or the water and steam) to the heat value of theenergy consumed. The combustion efficiency of a fuelfiredsystem will be higher than its thermal efficiency. SeeASME Power Test Code 4.1 for more details on determiningthe thermal efficiency of boilers and other fuel-firedsystems. In the U.S., fuel-fired efficiencies are typicallyreported based on the higher heating value of the fuel.Other countries typically report fuel-fired efficienciesbased on the fuel’s lower heating value. The differencebetween a fuel’s higher heating value and its lower heatingvalue is the latent energy contained in the water vapor(in the exhaust gas) which results when hydrogen (fromthe fuel) is burned. The efficiency of a system based on afuel’s lower heating value can be 10 to 15% higher than itsefficiency based on a fuel’s higher heating value.Table 28.A.1.Summary of Performance and Efficiency Terminology—————————————————————————Operating Design Rated Seasonal AverageMode Conditions Conditions—————————————————————————Cooling COP COPEERSEERkW/tonIPLV—————————————————————————Heating COP AFUEE cCOPE tHSPF—————————————————————————


CHAPTER 29SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGSLEED FOR NEW CONSTRUCTION AND EXISTING BUILDINGSNICK STECKYNJS Associates, LLC29.1 BEGINNINGSThe publication of Rachel Carson’s Silent Spring in 1962alerted the general public to the dangers of pesticides, inparticular the dangers to humans. This helped precipitatethe rise of an environmental movement, politics and lawsin the United States during the sixties and seventies. Newlaws were passed to protect the environment. These included:• The National Environmental Policy Act of 1969—this created the Federal Environmental ProtectionAgency.• The Clean Air Act was passed in 1970. This greatlyexpanded the protection of two previous laws, theAir Pollution Control Act of 1955 and the first CleanAir Act of 1963.• The Water Pollution Control Act of 1972.• The Endangered Species Act of 1973.• The formation of the Federal Department of Energyin the late seventies.But a problem with environmentalism was beginning tobrew. There came a tension, an apparent conflict betweenthe need to preserve the environment and the need togrow and expand the economy and jobs. Environmentalistsbegan to be seen as opponents of growth and industry.There appeared to be a contradiction between businessand protection of the environment. Environmentalismbegan to be seen as just another “special interest” groupwhich simply added cost to running a business with verylittle added value.In addition to legislation, other events were occurringduring the seventies, eighties, and nineties which encouragedthe development of sustainability:• Earth Day, launched April 22, 1970.• Nuclear power suffered major setbacks with the incidentsat Three Mile Island and Chernobyl.• Major oceanic oil spills, including the EXXON Valdez.• The OPEC oil crises of the mid-seventies and earlyeighties.• The mid-seventies natural gas shortages that causedmany plants to close during the winter to preservegas for home heating.• Discovery that ozone depleting compounds, such asCFC refrigerants, were destroying the ozone layerof the atmosphere.During the eighties, in reaction to the forces of highenergy costs, inadequate energy supplies, environmentalism,and pollution control, a new approach to designing,building and operating buildings began to develop.It was recognized that buildings consume significantpercentages of our resources, open space and energy.However, some of the new approaches were not withoutproblems. For example, architects experimentedwith half or even fully buried homes. But many timesthese homes had problems such as humidity, which ledto mold. As a result of the desire to save energy on ventilationair, ASHRAE modified Standard 62 and reducedventilation rates to five cubic feet per minute (CFM) foroffices. This contributed to what was later called sickbuilding syndrome, or SBS. SBS was a catchall phrasefor any building that provided an uncomfortable, irritating,and possibly unhealthy indoor environment forthe occupants. Employees in these buildings generallyhad higher rates of absenteeism, lower productivity, andhigher incidences of lawsuits against building ownersand employers.29.1.1 The Sustainability MovementWhile buildings and occupants were sufferingthrough difficulties as described above, the nineties gaverise to the sustainability movement. It was in 1999 that793

794 ENERGY MANAGEMENT HANDBOOKthe book Natural Capitalism by P. Hawken, A. Lovins, andL. H. Lovins was released. The authors present the connectionbetween economics and environmentalism, andargued that they are mutually supportive, not mutuallyexclusive. Natural Capitalism points out that as in its title,nature itself is capital. For example, what is the value ofa clean lake? It is drinking water that needs less expensivewater treatment before being potable. It is the recreationalvalue and income for some, through swimmersand boaters. If the lake were destroyed through pollution,water treatment costs would rise and lake use revenuesgo down. Seems simple enough, but it has mostly beenignored until now.Historically, the environmental movement consistedprimarily of regulation, legislation, and mandates. However,this regulatory approach has often been seen by thebusiness community as an obstacle to growth. But a newform of green revolution was emerging. This one maysucceed where traditional legislative environmentalismhas had limited success. This new form is a larger view,taking economic, community, and technological considerationsinto account, as well as environmental. This newsustainability recognizes the need to consider the costbenefitanalysis in evaluating and/or promoting variousprograms. And the themes presented in Natural Capitalismhelp us identify and quantify the complete costs andbenefits associated with sustainability.29.1.2 Sustainability DefinedSome of the many definitions of sustainability thatexist include:Design Ecology ProjectSustainability is a state or process that can be maintainedindefinitely. The principles of sustainability integratethree closely intertwined elements—the environment, theeconomy, and the social system—into a system that canbe maintained in a healthy state indefinitely.Brundtland Commission of the UNDevelopment is sustainable “if it meets the needs of thepresent without compromising the ability of future generationsto meet their own needs.”ASHRAE defines sustainability as“providing for the needs of the present without detractingfrom the ability to fulfill the needs of the future.”Note the commonality of theme. The concept isrooted in maintaining our current standards of livingwithout jeopardizing future generation’s standards ofliving.29.2 SUSTAINABILITY GIVES RISE TOTHE GREEN BUILDING MOVEMENTThe late eighties and early nineties were a crucialdevelopmental period for the green design movement.Leaders of green design included William McDonough,Paul Hawken, John Picard, Bill Browning, and DavidGottfried, who later went on to be one of the co-foundersof the USGBC. The movement acknowledged that buildingsrepresent a very significant usage of resources, land,and energy, and that improvements to the ways in whichwe design, construct, operate, and decommission buildingscould make significant contributions to improvementof the environment and overall sustainability.Today, owners, occupants, and communities are beginningto hold buildings to higher standards. Industryleaders are responding by creating physical assets thatsave energy and resources, and are more satisfying andproductive, economical as well as environmentally accountable.Building owners who want the greatest returnon investment can take a path that is green both economicallyand environmentally. How are they doing it? Theyare doing it through integrated solutions for the design,construction, maintenance, and operations, as well as theultimate disposal of a building.An integrated, or whole building approach, maymean that up-front costs may be no more than conventionalconstruction, but life cycle costs over the life of theasset will be lower. By designing, building, and operatingin an integrated way, owners can expect high performancebuildings that offer:• Increased efficiencies of systems and use of resourcesand energy.• Quality indoor environments that are healthy, secure,pleasing and productive for occupants andoperators.• Optimal economic and environmental performance.• Wise use of building sites, assets and materials.• Landscaping, material use and recycling efforts inspiredby the natural environment.• Lessened human impact upon the natural environment.29.2.1 Formation of the Unites StatesGreen Building Council, USGBCIn 1993, David Gottfried, a developer, Mike Italiano,an environmental attorney, and Rick Fedrizzi of CarrierCorporation got together to form the United States GreenBuilding Council, USGBC. They had become concernedabout the fragmentation of the building industry as it

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 795relates to sustainable design. At the time, there was noclear consensus among industry professionals as to whatconstituted a “green design.” The USGBC was initiallyformed to create an educational organization that wouldbring building professionals together to promote sustainabledesign. In 1997, the USGBC was awarded a $200,000grant from the Federal Department of Energy, and theUSGBC was off and running. The USGBC has made alarge impact on the design and building industry. Thereare more than 5,500 members consisting of individuals,large and leading corporations, governmental entities,universities, educational entities, consultants, productmanufacturers, trade associations, and more.29.2.2 Making It All Come Together: The USGBCThe U.S. Green Building Council, (USGBC.org), abalanced consensus coalition representing every sectorof the building industry, spent five years developing,testing, and refining the LEED Green Building RatingSystem. LEED stands for “Leadership in Energy & EnvironmentalDesign,” and when adopted from the start of aproject helps facilitate integration throughout the design,construction, and operation of buildings.MISSION STATEMENTThe U.S. Green Building Council is the nation’s foremost coalitionof leaders from across the building industry working topromote buildings that are environmentally responsible, profitable,and healthy places to live and work.The mission of this unprecedented coalition is toaccelerate the adoption of green building practices, technologies,policies, and standards. The USGBC is a committee-basedorganization endeavoring to move the greenbuilding industry forward with market-based solutions.Another vital function of the council is linking industryand government. The council has formed effective relationshipsand priority programs with key federal agencies,including the U.S. DOE, EPA, and GSA.The council’s membership is open and balanced.It is comprised of leading and visionary representationfrom all segments of the building industry includingproduct manufacturers, environmental groups, buildingowners, building professionals, utilities, city government,research institutions, professional societies and universities.This type of representation provides a unique, integratedplatform for carrying out important programs andactivities.LEED OverviewThe LEED Green Building Rating System is a priorityprogram of the US Green Building Council. It is avoluntary, consensus-based, market-driven building ratingsystem based on existing proven technology. It evaluatesenvironmental performance from an integrated or“whole building” perspective over a building’s life cycle,providing a definitive standard for what constitutes a“green building.”LEED is based on accepted energy and environmentalprinciples and strikes a balance between knowneffective practices and emerging concepts. Unlike otherrating systems currently in existence, the development ofLEED Green Building Rating System was initiated bythe US Green Council Membership, representing all segmentsof the building industry. It also has been open topublic scrutiny.LEED is an assessment system that incorporatesthird party verification and is designed for rating new andexisting commercial, institutional, and high-rise residentialbuildings. It is a feature-oriented system where credits,also called points, are earned for satisfying each criterion.Different levels of green building certification are awardedbased on the total credits earned. The system is designed tobe comprehensive in scope, yet simple in operation.29.2.3 General Introduction & DiscussionThere are several key points in this section:• LEED is becoming nationally accepted at local, stateand federal levels.• The engineering community such as ASHRAE, AEE,IESNA has the knowledge and skills that can addsignificant value to a LEED design team.• LEED has value to the engineering community, bothas owners/operators of facilities or as design teammembers.• It is a WIN-WIN proposition for us all.• Green buildings are a process, not a collection oftechnologies.• Encourage engineering participation/membershipin USGBC and its local chapters.• Encourage professional accreditation and attendanceat LEED workshops.• Encourage building owners/operators to registerand certify projects.There are many different terms for sustainable buildings,but basically they all convey the same message:• Sustainable Design• High Performance Buildings• High Efficiency Buildings• Integrated Building Design• Green Buildings

796 ENERGY MANAGEMENT HANDBOOKSustainable buildings may be our goal, but the mostcommon term used for these high-performance buildingsis green buildings. For some, the term green buildingsmay sound too environmentally focused, but the way itis used here, it represents the all-inclusive idea of sustainablebuildings.Characteristics of Sustainable Green Buildings• Optimal environmental and economic performance.• Increased efficiencies saving energy and resources.• Satisfying, productive, quality indoor spaces.• Whole building design, construction and operationover the entire life cycle.• A fully integrated approach—teams, processes, systems.29.2.4 Green BuildingsGreen buildings are designed and constructed in accordancewith practices that significantly reduce or eliminatethe negative impact of buildings on the environmentand its occupants. This includes design, construction, operations,AND, ultimately, demolition. Five fundamentalcategories constitute the USGBC green building designation.They are:• Sustainable Site Planning• Safeguarding Water and Water Efficiency• Energy Efficiency and Renewable Energy• Conservation of Materials and Resources• Indoor Environmental QualityAll relate back to the previous definitions and discussionsof sustainability.All of these are contained in the LEED standard,Leadership in Energy and Environmental Design. Thisis the trademark rating system developed by the UnitedStates Green Building Council, USGBC.Besides the LEED rating system to define and describesustainable buildings, there are others such as:• The British Research Establishment EnvironmentalAssessment Method (BREEAM) was launched in1990 and is increasing in its use.• Canada’s Building Environmental Performance AssessmentCriteria (BEPAC) began in 1994, but wasnever fully implemented due to its complexity.Canada has now licensed use of LEED from the US-GBC.• The Hong Kong Building Environmental AssessmentMethod (HK-BEAM) is currently in pilotform.• The USGBC LEED family of programs.• State and regional guides include high performancebuilding guidelines in NY and PA’s guidelines forcreating high performance buildings, as well as California’sprograms for school construction.• Green Globes is a web-based self-assessment programthat guides the integration of green principlesinto a building’s design.LEED strives to encompass a wide band ofsustainability that includes:• Society and Community—recognizes that buildingsexist to serve the needs of the community, but thatthey must also minimize their impacts.• Environment—again striving to minimize negativeimpacts on the environment• Economics—recognizes that the adoption of thesesustainability initiatives by business will requireeconomic benefits that can be delivered by greenbuildings.• Energy—recognizes that energy plays a key role inbuilding operating costs as well as a sustainable energyfuture.29.2.5 Benefits of LEED BuildingsThe EnvironmentAt the onset, LEED was created to standardize theconcept of building green to offer the building industry auniversal program that provided concrete guidelines forthe design and construction of sustainable buildings for alivable future. As such, it is firmly rooted in the conservationof our world’s resources. Each credit point awardedthrough the rating system reduces our demand and footprintupon the natural environment.EconomicsConventional wisdom says that the constructionof an environmentally friendly, energy efficient buildingbrings with it a substantial price tag and extended timetables.This need not be the case. Breakthroughs in buildingmaterials, operating systems and integrated technologieshave made building green not only a timely, cost effectivealternative, but a preferred method of constructionamong the nation’s leading professionals.From the USGBC“Smart business people recognize that high performancegreen buildings produce more than just a cleaner,healthier environment. They also positively impact thebottom line. Benefits include: better use of building materials,significant operational savings, and increased workplaceproductivity.”

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 797In many instances, green alternatives to conventionalbuilding methods are less expensive to purchase andinstall. An even larger number provide tremendous operationalsavings. The USGBC and LEED offer the buildingindustry a fiscally sound platform on which to build theircase for whole-building design and construction. Greenbuildings can show a positive return on investment forowners and builders.Economic benefits also can include:• Improved occupant performance—employee productivityrises, students’ grades improve. Californiaschools have analyses that show children in highperformance facilities have improved test scores.• Absenteeism is reduced.• Retail stores have observed measurable sales improvementsin stores with daylighting.The savings associated with productivity gains canbe the single largest category of savings in an office building.For example, salaries per square foot can be in therange of ten times or more than the cost of energy persquare foot in a typical office building. So while the energyengineering community agonizes over extractingeach cent in energy costs, there are greater savings potentialsby leveraging the increased employee productivitythrough high performance buildings. As described earlier,these buildings provide a superior indoor environmentwith regard to lighting, noise control, temperatureand humidity control, ventilation, and fresh air.Building green also enhances asset value. Accordingto organizations such as the International Facilities ManagementAssociation ( IFMA) and the Building Ownersand Managers Association ( BOMA), the asset value of aproperty rises at a rate of ten times the value of the operationalsavings. For example, if green building efficiencyreduces operating costs by $1/sq ft per year, the asset valueof that property rises by ten times that amount, or $10.For a 300,000 sq ft office complex, annual savings couldbe $300,000, and the asset value increase would be approximately$3,000,000. It pays to be efficient.Societal ImplicationsIn addition to the environmental and economic benefitsof building green, there are also significant societalbenefits. These include increased productivity, a healthywork environment, comfort, and satisfaction, to name afew. The good news about green buildings can then beleveraged with local and trade media through press releases,ceremonies, or events. By calling attention to thebuilding and its certification status, owners speak volumesabout themselves. The USGBC writes:“Like a strong prospectus, building green sends theright message about a company or organization: it’s wellrun, responsible, and committed to the future”This, too, has a direct effect on the bottom line.29.2.6 Benefits to the Architecturaland Engineering CommunityAll too often, market forces drive a building designteam to focus on minimum first cost regardless of whatthe overall life cycle costs of this minimum first cost designmay be. The end result is that all elements of society,from owner, to occupants, and to the community endup with a building which is less than what it could haveor should have been. It used too many resources in construction,its energy and operating costs are high, and itdoes not provide the optimum indoor environment foremployee productivity.However, installation of the LEED process of formingan integrated design team on conceptual design dayone promotes the formation of a creative solution to theparticular buildings needs being planned for. Within thiscreative roundtable of equals, the team is able to maximizeuse of the collective wisdom of the members anddevelop a design concept that can be both green and economic.For example, increasing the use of natural light todisplace some artificial light can result in a reduced loadon air conditioning systems. These AC systems can thenbe downsized, resulting in reduced equipment first costsand reduction in electricity use, both through the artificiallighting reduction and reduced cooling loads. Small savingsmultiply and reverberate throughout the design, ultimatelyhaving significant impacts on overall first costsand operating costs.When building “green,” the sum is larger than theindividual parts only when all components are integratedinto a single unified system. Integration draws upon everyaspect of the building to realize efficiencies, cost savings,and continuous returns on investment.The most significant benefit to the architectural/engineering (A/E) design community is that LEED promotesand rewards creative solutions. Sustainable andgreen designs are not yet commodity skills that all firmscan lay claim to. For those A/E firms seeking to providemore value to their clients, LEED is a way to achieve this.LEED can become a standard for design excellence, andprovides an A/E with a brand differentiator versus thecompetition. LEED can become an outstanding competitiveedge in the marketplace for firms seeking a leadershipposition of excellence.LEED AcceptanceThere’s a groundswell of acceptance taking place.

798 ENERGY MANAGEMENT HANDBOOKWhen LEED was first introduced in 2000, there was reluctanceto accept it. It represented a new way of lookingat the design, construction, operations, and disposal offacilities. It called for raising the bar, which many havebeen slow to accept. There were many questions abouthigher first costs, paybacks, overall benefits, ability of thedesign community to deliver, doubts about the technologiesinvolved, etc. The perception was that green buildingswere too futuristic, unobtainable, and required toomany tradeoffs to be able to deliver a practical, efficientfacility where people could live, work, and play.Some elements of these doubts remain, especially asto the financial benefits question. However, as we gainexperience with green buildings, we are developing theexperience and the data needed to resolve these doubts.The First Cost “Premium” of LEEDConventional wisdom says that green buildingscost more. However, as the industry becomes more experiencedwith the actual delivery of green buildings,green is becoming more cost-neutral. In an article titled“The Costs and Financial Benefits of High PerformanceBuildings,” Greg Kats of Capital E analyzed 40 CaliforniaLEED buildings for the “cost premium” of LEED. Thestudy consisted of 32 office buildings and eight schools:• The eight LEED-certified buildings (the basic level ofLEED certification) cost an average of 0.7% more.• The twenty-one Silver rated buildings cost an average1.9% more.• The nine Gold rated buildings cost average of 2.2%more.• The two Platinum rated buildings cost an average of6.8% more.29.2.7 LEED DescribedWhy was LEED created?• LEED Developed as a way to define and quantifywhat constitutes a sustainable green design.• Defines “green” by providing a standard for measurement.• Addresses the “greenwashing” issue, such as falseor exaggerated claims.• Facilitates positive results for the environment, occupanthealth, and financial return.• Use as a design guideline.• Recognizes leaders.• Stimulates green competition.• Establishes market value with recognizable national“brand.”• Raises consumer awareness.• Transforms the marketplace.• Promotes a whole-building, integrated design process.What is LEED• Consists of performance-based and prescriptivebasedcriteria.• Focuses on the whole building system instead of thecomponents.• Life cycle based, not first cost.• Promotes architectural and engineering innovation,i.e., innovation LEED credits.• Provides a third-party verification process to ensurequality and compliance.The growing family of LEED building rating systemsincludes:LEED NC for new construction,LEED EB for existing buildings,LEED CI for commercial interiors,LEED CS for commercial core & shell,LEED H for homes, currently in pilot.Other LEED programs are being developed.29.3 INTRODUCING THE LEED NC RATINGSYSTEM: A TECHNICAL REVIEWThe LEED format for rating a green building consistsof two categories:• Prerequisites—These are mandatory requirementsand all must be satisfied before a building can becertified.• Credits—Each credit is optional, with each contributingto the overall total of credits. This willdetermine the level a building will be rated at, i.e.Certified, Silver, Gold or Platinum.29.3.1 Sustainable Sites—14 Possible PointsPrerequisite:• Erosion & Sedimentation Control—Control erosionto reduce negative impacts on water and air qualityby complying with the EPA storm water managementrequirements for construction activities.Credits:• Site Selection—Avoid development of inappropriatesites and reduce the environmental impact fromthe location of a building on a site. One point.• Urban Redevelopment—Channel development tourban areas with existing infrastructures, protecting

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 799green fields and preserving habitat and natural resources.One point.• Brownfield Redevelopment—Rehabilitate damagedsites where development is complicated by real orperceived environmental contamination, thereby reducingpressure on undeveloped land. One point.• Alternative Transportation—Four points available:one each for public transportation access, bicyclestorage, alternative fuel vehicles, and parking capacity.• Reduced Site Disturbance—Conserve existing naturalareas and restore damaged areas to provide habitatand promote biodiversity. Two credits available:one each to protect or restore open space, and/orreduce the development footprint.• Storm Water Management—Limit disruption ofnatural water flows by minimizing storm waterrunoff, increasing on-site infiltration and reducingcontaminants. Two credits available: one each forno increase in the rate or quantity of runoff and/or treatment systems designed to remove total dissolvedsolids, phosphorous, and comply with EPAguidelines.• Heat Island Effect—Reduce heat islands, which arethermal gradient differences between developed andunderdeveloped areas. There is one point each forroof and non-roof applications, up to two points.• Light Pollution Reduction—one point. The intentis to eliminate light trespass from the building site,improve night sky access, and reduce developmentimpact on nocturnal environments.29.3.2 Water Efficiency—5 possible pointsNo prerequisites in this category.Credits:• Water Efficient Landscaping—Limit or eliminate theuse of potable water for landscape irrigation. Up totwo points available. One each for use of high efficiencyirrigation technology, and/or using capturedrain or recycled water for irrigation.• Innovative Wastewater Technologies—one point.Reduce the generation of wastewater and potablewater demand, while increasing the local aquiferrecharge.• Water Use Reduction—20% reduction is one point,30% reduction is two points. Maximize water efficiencywithin the building to reduce the burden onmunicipal water supply and wastewater systems.29.3.3 Energy & Atmosphere—17 possible pointsThree prerequisites:• Fundamental Building Systems Commissioning—Verifyand ensure that fundamental buildingelements and systems such as HVAC are designed,installed, and calibrated to operate as intended.• Minimum Energy Performance—Establish the minimumlevel of energy for the base building systemswhich is compliance with ASHRAE/IES 90.1-1999.• CFC Reduction in HVAC&R Equipment—This requireszero use of CFC based refrigerants in newbuildings such as R11 and R12.Credits:• Optimize Energy Efficiency—possible ten points.Achieve increasing levels of energy performanceabove the prerequisite standard, ASHRAE 90.1-1999, to reduce environmental impacts associatedwith excessive energy use. 20% better is two points,30% is four points, and on up to 60% better is worthten points.• Renewable Energy—three points possible, one pointeach for 5%, 10%, 20% of total energy. Encourage andrecognize increasing levels of self-supply throughrenewable technologies to reduce environmental impactsassociated with fossil fuel energy use.• Additional Commissioning—one point. Verify andensure that the entire building, including the buildingenvelope, is designed, constructed, and calibrated tooperate as intended. This as opposed to the prerequisitewhich called for only fundamental systems.• Elimination of HCFC s and HALONS—one point.Reduce ozone depletion and support early compliancewith the Montreal Protocol. This applies to refrigerantssuch as R22 and R123.• Measurement & Verification—one point. Providefor the ongoing accountability and optimizationof building energy and water consumption performanceover time.• Green Power—one point. Encourage the developmentand use of grid-source energy technologies ona net zero pollution basis by the purchase of greenpower that meets the Center for Resource SolutionsGreen-E products.29.3.4 Materials & Resources—13 possible pointsPrerequisite:• Storage & Collection of Recyclables—Facilitate thereduction of waste generated by building occupantsthat is hauled to and disposed on in landfills.Credits:• Building Reuse—two points possible at 75% and100% reuse of building shell and non-shell. A thirdpoint is available by maintaining 100% of an exist-

800 ENERGY MANAGEMENT HANDBOOKing building’s structure and 50% of the non-shellsuch as walls, floor coverings, and ceilings. Purposeis to extend the life cycle of existing building stock.• Construction Waste Management—two pointsavailable. This is to divert construction, demolition,and land clearing debris from landfill disposal. Redirectrecyclable material back into the manufacturingprocess. 50% recycled or salvaged materials byweight gets one point, 75% earns one more point.• Resource Reuse—two possible points at 5% or 10%of using salvaged or refurbished materials. The intentis to extend the life cycle of targeted buildingmaterials by reducing environmental impacts relatedto materials manufacturing and transport.• Recycled Content—two possible points. One pointspecifying a minimum of 25% building materialsthat contain post consumer recycled materials. Anadditional point is available if an additional 25% isrecycled content.• Local/Regional Materials—two possible points eithermanufactured or harvested locally. The purposeis to increase the demand for building products thatare manufactured locally, less than 500 miles, therebyreducing the environmental impacts resultingfrom long-distance transportation.• Rapidly Renewable Materials—one point. Reducethe use and depletion of finite raw and longcycle renewable materials by replacing them withrenewables.• Forest Stewardship Council, FSC, certified wood.One pointNote: regarding “certified products.” This does not mean“LEED certified” but certified by other entities such asthe Forest Stewardship Council FSC. Beware of manufacturesclaiming “LEED certified” products. The USGBCand LEED do not certify products. What they do is adoptindustry standards as applicable, such as the FSC certifiedwood.29.3.5 Indoor Environmental Quality (IEQ)—15 possible pointsTwo prerequisites:• Minimum IAQ Performance—Comply withASHRAE 62-2004 indoor air quality standard toprevent the development of air quality problems.• Environmental Tobacco Smoke (ETS) Control—Preventexposure of building occupants and systems toETS.Credits:• Carbon Dioxide Monitoring—one point. Provide anHVAC system which can monitor and ventilate abuilding based upon CO 2 levels.• Ventilation Effectiveness—one point. Provide forthe effective delivery of mixed and outdoor air tosupport health, safety, and comfort of occupants.Uses ASHRAE 129 methodology.• Construction IAQ Management Plan—two possiblepoints. Develop an IAQ management plan for duringconstruction and before occupancy, one point.An additional point is available by conducting atwo-week building flushout prior to occupancy using100% outside air.• Low Emitting Materials—four possible points. Onepoint for low VOC adhesives. One point for lowVOC paints. One point for carpet exceeding the Carpet& Rug Institute Green Label IAQ Program. Onepoint for composite wood and agrifiber productscontaining no additional urea formaldehyde resins.• Indoor Chemical & Pollutant Source Control—onepoint. This is to avoid exposing occupants to potentiallyhazardous chemicals that adversely affectIAQ.• Controllability of Systems—two possible points,one each for perimeter and non-perimeter. This is toprovide a high level of individual occupant controlof thermal, ventilation, and lighting systems.• Thermal Comfort—comply with ASHRAE Standard55—one point. This relates to providing a thermallycomfortable environment that supports healthy andproductive performance of occupants.• Permanent Monitoring System—one point. Theseare permanent systems to monitor temperature andhumidity which also allow occupants to have partialcontrol over these parameters.• Daylighting & Views—two points available. This isto provide a connection between indoor spaces andoutdoor environments through the introduction ofsunlight and views into the occupied areas of thebuilding. For one point, achieve a daylight factor of2% in 75% of all space occupied for critical visualtasks, but not including the likes of laundry rooms,copying rooms etc. For an additional point, achievethe 2% rating in 90% of spaces.29.3.6 Innovation & Design Process—5 possible points• Use of LEED-accredited professional—one point• Innovation in design—four possible points29.3.7 DiscussionNote: energy and engineering skills are applicable to asmany as 58% of total available points on water, energy,and IAQ. Energy related credits are the largest category

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 801of available credits.May ‘04 Energy User News article by Peter D’Antonio,entitled “The LEEDing Way.”The article analyzes the activity in energy & atmosphere,E&A, and indoor environmental quality, IEQ, forthe first 53 LEED-certified buildings:• Regarding E&A, average points earned is only 5.3out of the possible 17!• This is the lowest % achieved in any of the five categories!So although E & A is the largest plum, fewappear to be taking advantage of it.• Renewable energy points are earned in fewer than10% of the certified buildings.• Regarding IEQ points, ventilation effectiveness andcontrollability points are achieved in less than onethird of buildings.This EUN article provides support for the propositionthat engineers, the AEE, ASHRAE, the IESNA, andothers are not maximizing the potential contributions toLEED buildings. Hence, there is a significant opportunityto take on a larger role in the design, construction, andoperations of green buildings.Other factors for the design team to consider are theforces that drive the LEED points on a project. Many times,for the design team of an LEED project, it boils down to,“How many points can we get?” This becomes especiallythe case for the mechanical, electrical, and plumbing(MEP) team members. They control or influence approximately75% the total LEED credits on a job. Commonlycalled “pointchasing,” it is an effort by the design team toachieve the maximum available points at the minimumcost and effort. And although it is a rather ugly approachto green building design, it has become a matter of factthat teams will focus on points. It is even possible that theacquisition of points is one of the elements in the designcontract. Possibly a bonus is linked to points achieved?But this can and should be “managed” by the owner.It gets back to the integrated design process and the settingof goals during the design charette. Does the ownerwant very high energy efficiency, or does he want to makean environmental statement with a green roof?29.4 LEED FOR EXISTING BUILDINGS RATINGSYSTEM (LEED-EB) ADOPTED IN 2004LEED for existing buildingsAddresses:• Operation and upgrades of existing buildings.• Pilot: 2002—2004.• Initial certification and ongoing re-certification.• Achievements:—More than 95 registered buildings.—4 certified buildings.• Range of Users:—Federal, state, and local government; schools, collegesand universities, commercial buildings.LEED-EB approval completed and released for use, Oct.2004.Why LEED-EB Is So ImportantDrawing on similarities to LEED NC, LEED EB hasa larger potential impact and resultant benefits to societysimply because there are many times more existing buildingsthan new construction. LEED EB focuses on wherethe greatest impact potential is.29.4.1 LEED for Existing Buildings Rating SystemLEED-EB Rating System Goals• Help building owners upgrade and operate theirbuildings in a sustainable way over the long term.Avoid the “saw tooth” approach (upgrade, decline,upgrade, decline).• Support high productivity by building occupants.• Operations—Helps building owners upgrade and operate theirbuildings in a sustainable way over the longterm.—Reduces building operating costs.—Solves building operating problems.—Improves indoor environment.—Supports higher productivity of building occupants.• Communications—Helps building managers, operators, and serviceproviders communicate the importance of effectiveongoing building operation and maintenanceto decision makers in their organization.—Helps building managers and operators makesustainability part of the culture of their organization.—Helps CEOs and CFOs make sustainability partof the culture of their organization.—Helps communicate organization’s sustainabilitycommitments and achievements to its customersand the community.Prerequisites and CreditsSame categories as for other LEED Rating Systems:• Sustainable Sites• Water Efficiency• Energy and Atmosphere

802 ENERGY MANAGEMENT HANDBOOK• Materials and Resources• Indoor Environmental Quality• Innovation and Accredited ProfessionalLEED-EB Rating SystemFour Levels of Certification:• LEED-EB Certified 32-39 points• Silver Level 40-47 points• Gold Level 48-63 points• Platinum Level 64-85 points29.4.2 LEED for EB Technical ReviewSimilar to LEED NC, all prerequisites must be satisfied,and the credits are optional depending upon thefinal points and certification level desired.29.4.2.1 Sustainable Sites—14 possible pointsTwo prerequisites:• Erosion and Sedimentation Control—Control erosionto reduce negative impacts on water and airquality.• Age of Building—two years old or more.Credits:• Plan for Green Site & Bldg Exterior Management—up to two points. Encourage grounds/site/buildingexterior management practices that have thelowest environmental impact possible and preserveecological integrity, enhance diversity, and protectwildlife while supporting building performance.• Hi Development Density Building & Area—onepoint. Channel development to urban areas withexisting infrastructure, protect greenfields, and preservehabitat and natural resources.• Environmentally Preferable Alt Transportation—upto four points available. One point each for: publictransportation access, bicycle storage and changingrooms, alternative fueled vehicles, car pooling, andtelecommuting.• Reduced Site Disturbance—up to two points. Onepoint for protecting or restoring 50% of site area. Anadditional point to protect or restore open space at75% of the site area.• Storm water Management—up to two points. Onepoint for measures that mitigate at least 25% of theannual storm water falling on the site. An additionalpoint for mitigation of at least 50% of storm water.• Reduce Heat Islands Effect (roof and non-roof)—upto two points. One point for reduction of heat islands.An additional point is available for an ENER-GYSTAR-compliant roof.• Light Pollution Reduction—one point. Eliminatelight trespass from the building and site, improvethe night sky, and reduce development impact onnocturnal environments.29.4.2.2 Water Use and Water Efficiency—5 possible pointsTwo Prerequisites:• Minimum Water Efficiency—maximize fixture waterefficiency within buildings to reduce the burdenon potable water supply and wastewater systems.• Discharge Water Compliance—protect natural habitat,waterways, and water supply from pollutantscarried by building discharge water.Credits:• Water Efficient Landscaping—up to two points. Requiresthe use of water efficient irrigation technologiesor captured rain and recycled water to reducepotable water consumption for irrigation. The firstpoint is based on a 50% reduction, and an additionalpoint is available for 95% reduction in potable wateruse.• Innovative Wastewater Technology—one point.Reduce the generation of wastewater and potablewater demand, while increasing the local aquiferrecharge.• Water Use Reduction—up to two points. Maximizefixture potable water efficiency to reduce burdenson potable and wastewater municipal systems. Thefirst point is for a 10% reduction and an additionalpoint is available for a 20% reduction.29.4.2.3 Energy and Atmosphere—23 possible pointsThree Prerequisites• Existing Building Commissioning—Verify that fundamentalbuildings systems are performing as intended.• Minimum Energy Performance—Satisfy the minimumlevel of energy efficiency using the ENER-GYSTAR portfolio manager. Needs a rating of 60 ormore.• Ozone Protection—Reduce ozone depletion potentialsby not using CFC refrigerants such as R11 andR12.Credits:• Optimize Energy Performance—up to ten pointsavailable. Achieve increasing levels of energy efficiencyabove the ENERGYSTAR prerequisitescore of 60. Thus a score 63 earns one point, 79earns five points, up to a maximum of ten points

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 803for a 99 rating.• On-site & Off-site Renewable Energy—up to fourpoints. The first point is for 5% on-site renewableOR 25% off-site renewables up to a maximum offour points for 30% on-site renewable energy OR100% off-site renewable energy.• Building Operations & Maintenance—up to threepoints. One point each for maintenance staff education,building systems maintenance, building systemsmonitoring.• Additional Ozone Protection—one point. Reduceozone depletion potential in compliance with MontrealProtocol. Thus HCFC refrigerants such as R22and R123 are not used.• Performance Measurement—up to four points.Have in place a continuous metering system for anumber of facilities functions as follows: lightingsystems, electric and gas metering, cooling load,chilled water system efficiency, irrigation water metering,boiler efficiencies, HVAC systems such aseconomizers, variable speed pumps and fans, airdistribution, and emissions monitoring. Note theseall can be incorporated into the building automationsystem (BAS).• Documenting Sustainable Building Cost Impacts—one point. Document overall building operatingcosts for previous five years and track changes inthe overall operating costs.29.4.2.4 Materials and Resources -16 possible pointsTwo prerequisites:• Source Reduction and Waste Management—establishminimum source reduction and recycling programelements.Toxic Material Source Reduction—reduced mercury inlamps.Credits:• Construction, Demolition and Renovation WasteManagement—up to two points. First point for diverting50% or more of construction, demolition andland clearing waste from landfills. An additionalpoint if 75% or more is diverted.• Optimize Use of Alternative Materials—up to fivepoints available. Maintain a sustainable purchasingprogram covering at least office paper, officeequipment, furniture, furnishings, and buildingmaterials. One point is awarded for each 10% of totalpurchases that achieve criteria such as 70% salvagedmaterials, 10% post consumer recycled, 50%rapidly renewables, FSC-certified wood, and materialsmanufactured within 500 miles of the site.• Optimize Use of IAQ Compliant Products—up totwo points. These relate to the purchase of low emittingmaterials such as for carpets, sealants, paintsand coatings, composite materials and agrifiberproducts with no added urea formaldehyde.• Sustainable Cleaning Products and Materials—up tothree points. Points accumulate based upon quantitiesof products that meet the Green Seal GS-37 orcomply with the California Code of Regulations forVOCs. Disposable janitorial paper products andtrash bags meet the requirements of the EPA comprehensiveprocurement guidelines.• Occupant Recycling—up to three points. Set updivert/recycle programs for occupants. 30% is onepoint, 20% another, and the third point if 30% of totalwaste stream is diverted or recycled.• Additional Toxic Material Source Reduction—onepoint. Establish a program to reduce the potentialamounts of mercury brought into the buildingthrough lamps.29.4.2.5 Indoor Environmental Quality—22 points availableFour prerequisites:• Outside Air Introduction and Exhaust Systems—Satisfy ASHRAE 62-2004 for IAQ.• Environmental Tobacco Smoke (ETS) Control—Preventor minimize occupant exposure to ETS.• Asbestos Removal or Encapsulation—Establishan asbestos remediation and control managementplan.• PCB Removal—Establish a PCB management planincluding a facility survey for PCBs.Credits:• Outside Air Delivery Monitoring—one point. Providepermanent monitoring systems on ventilationsystem performance measuring outdoor air andCO 2 .• Increased Ventilation—one point. Increase ventilationrates to exceed ASHRAE 62-2004 by 30%.• Construction IAQ Management Plan—one point.Prevent any IAQ problems from arising due to construction/renovationwork. Isolate occupied areasfrom dust, noise and other irritants.• Documenting Productivity Impacts—up to twopoints. Document the history of absenteeism, productivity,and health care costs and submit to theUSGBC.• Indoor Chemical & Pollution Source Control—upto two points. Reduce the exposure of occupants todusts and particulates by using filters of effective-

804 ENERGY MANAGEMENT HANDBOOKness of MERV 13 or greater. An additional point isearned by reducing occupants’ exposures to contaminantsthat may arise from operations such ascopying, faxing, etc.• Controllability of Systems—up to two points. Pointsavailable for occupant control of lighting systemsand another for HVAC and temperature control.• Thermal Comfort—ASHRAE Standard 55- up totwo points. First point for compliance with the standardand an additional point is available for a permanentmonitoring system to ensure compliance.• Daylighting and Views—up to four points. Providea connection between indoor spaces and the outdoorenvironment through the introduction of sunlightand views. Points available for 50% and 75% ofspaces that have a 2% daylight factor. And two morepoints available for 45% of spaces (1 point) and 90%of spaces (1 point) that have direct line of sight visionto the outdoors.• Contemporary IAQ Practice—one point. EnhanceIAQ performance by optimizing practices and developingprocedures to prevent the development ofIAQ problems.• Green Cleaning—up to six points. Points availablefor cleaning entryway systems, isolation of janitorialclosets, low environmental impact cleaning policy,low environmental impact pest management policy,and low environmental impact cleaning equipmentpolicy.29.4.2.6 Innovation and Accredited Professional5 possible points• LEED EB Innovation in Operation, Upgrades andMaintenance—up to four points.• LEED-accredited Professional—one point.29.5 SUMMARY DISCUSSION OF TWONEW LEED PROGRAMS:LEED-CI for Commercial Interiors andLEED-CS for Core and ShellLEED-CI addresses the specifics of tenant spacesprimarily in office, retail and industrial buildings. It wasformally adopted in the fall of 2004. A companion ratingis LEED for core & shell which is currently under developmentand in its pilot phase. Adoption is expected thefall of 2005. Together, LEED-CI and LEED-CS will establishgreen building criteria for commercial office real estatefor use by both developers and tenants.LEED-CI serves building owners and occupantsas well as the interior designers and architects who designbuilding interiors and the teams of professionalswho install them. It addresses performance areas includingwater efficiency, energy efficiency, HVAC systems &equipment, resource utilization, furnishings, and indoorenvironmental quality.29.5.1 LEED for Commercial Interiors (CI)• Addresses the design and construction of interiorsin existing buildings and tenant fit-outs in new coreand shell buildings.• Pilot: 2002—2004.• Achievements: More that 45 projects in pilot.• LEED CI Adopted in fall 2004.29.5.1.1 LEED-CI Point DistributionThe same five basic categories as the other LEEDrating systems are used.Possible PointsSustainable Sites 7Water Efficiency 2Energy & Atmosphere 12Materials & Resources 14Indoor Environmental Quality 17Innovation & Design Process 4LEED Accredited Professional 1Total Points Available 574 Levels of CertificationCertified 21-26Silver 27-31Gold 2-41Platinum 42-57Because of its nature, i.e., the interior parts of abuilding, the energy engineer has less opportunity to aidin earning points in this program. This is the only programto date wherein the E & A credits are not the largestcategory. However, engineers in general do have the potentialto aid points especially in the indoor environmentalquality category.29.5.2 LEED for Core & Shell, CSBased upon the LEED NC rating system for newconstruction and major renovation, LEED CS was developedin recognition of the unique nature of core and shelldevelopments. In particular, there is the lack of developercontrol over key aspects such as interior finishes, lighting,and HVAC distribution. Thus the scope of CS is limitedto those elements of the project under the direct control of

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 805the developer.With its standards for CS and CI, the USGBC addressesthe entire building—core, shell, and interiors. Theresponsibilities for particular sections, however are assignedto those parties having direct control over them.LEED CS is still in its pilot phase as of early 2006.29.5.2.1 LEED CS Credit CategoriesBelow is a summary of where the points will be forLEED CS. Note its similarities to LEED NC and that energy& atmosphere is the largest points category.Possible PointsSustainable Sites 15Water Efficiency 5Energy & Atmosphere 16Materials & Resources 11Indoor Environmental Quality 13Innovation and Design Process 4LEED Accredited Professional 1——Total Points Available 654 Levels of CertificationCertifiedSilverGoldPlatinum24-29 points30-35 points36-47 points48-64 pointsYou can download all four of the LEED rating systems.The rating systems download are free. However,other tools and workbooks such as the reference guidesdo have a fee associated with them, with discounts givento members.• Visit U.S. Green Building Council Web Site at www.usgbc.org/leed.• Choose rating system.• Click on rating system you would like to download.29.6 THE LEED PROCESS• Design Team Integration• Project Registration• Project Certification• DocumentationThe LEED Design Process:When does a Green Design Begin?“It Begins in the Beginning”Critical to success is the integration of the designTEAM on Day 1 of design. LEED is a marketplace transformer.It is a paradigm shift away from top down, minimumfirst cost emphasis. The hierarchical old-fashionedway was design, bid, build.But the LEED design process is one of integrated,holistic building design, construction, operations andmaintenance. All participate as equals on a constructionroundtable:• Owner Operations Personnel• Owner• Architect• Engineer• Construction Manager• Contractors & Subcontractors• Equipment Suppliers & Manufacturers• Commissioning Authority—Watchdog RoleDuring the very early stages of a green building’sdevelopment, a design charette should be held. This refersto meetings that are held over the course of a day ortwo, wherein the entire team, the construction roundtablegroup, gets together to develop the roadmap to successfulgreen building.• The ENTIRE team joins in—all stakeholders, includingthe owner, designers, commissioning authority, andoperations.• Gain buy-in and consensus.• Explore environmental issues.• Propose alternatives.• Identify modeling and resource allocation.• Use the LEED checklists as guide for level of greendesired.• Use an outside facilitator who specializes in integrateddesign.• Present examples of resources and ways to tracecosts and benefits of modeling.• Establish a task/responsible team to track and managecompliance with the process.• Determine an LEED Leader who will be the “watchdog”over points. The commissioning authority canbe a good choice for this.29.6.1 The Energy Engineer’s Goal: Get Invited!It is during the design charette which occurs at theearliest moments of a project, that the energy engineercan provide the maximum overall benefit to the project.It is during this time that key choices are made about thelighting, HVAC, and building envelope. The energy engineercan help guide the team to the most appropriateenergy efficient design strategies based upon the team’senergy goals and the available energy sources.

806 ENERGY MANAGEMENT HANDBOOKOn an LEED project, a major concern of the designteam is simply, “How do I get the points?” Commonlycalled “points chasing,” it is an effort by the design teamto achieve the maximum available points at the minimumcost and effort. However, this can be ameliorated or drivenby the owner. During the design charette, it is the owner’sresponsibility to clearly identify the goals and objectivesof the project. If it is simply, “get me the most points,”the team will point chase. If the goals are, for example, tohave the highest energy efficiency possible, or the mostdaylighting possible, then other choices, alternatives, andevaluations may be examined.29.6.2 Example and Discussion on ObtainingLEED Points on a ProjectThe following is an example of how in an integratedgreen building design, a “simple” decision such as tohave a computerized building automation system (BAS),can have substantial overall impact on acquiring points.Knowing that LEED buildings will generally requiresophisticated controls, the BAS can have a major role inobtaining and facilitating points. The following is a summaryof the impact a BAS may have on LEED points. Thedesign team can use this example to guide them in thesame process for other techniques and/or technologies.Again, the emphasis is on integrated design. We are designingan integrated building, not a collection of partsand systems.29.6.2.1 Some Examples of BAS Influence onLEED Credits:• Sustainable Sites—Light pollution reductionthrough use of controls.• Water Efficiency—use of metering to document waterconsumption—although not a credit itself, it canfacilitate credits in this category.Energy & Atmosphere—many credits and much influencehere:• Energy Prerequisite and Optimized Energy Performance—tencredits are in play here. The BAS asan integral part of the energy consuming systemsincluding lighting, HVAC, load management, etc.,and helps earn credits through performance improvementsthat will be quantified in the buildingenergy simulations required by ASHRAE 90.1• Commissioning—BAS aids the commissioning authorityin performing their duties, a time saver.• Measurement and Verification—one prescriptivecredit—provide for accountability & optimizationof energy and water consumption over time.• Optimize Energy Performance—in LEED CI, up tofour prescriptive credits for lighting and power controls.Other credits in energy performance, an additionalfour.• Energy Submetering—in LEED CI, measure for energyaccountability. Up to two prescriptive credits.• Building Operations & Maintenance—three prescriptivecredits, relating to staff education, buildingsystems maintenance, and building systemsmonitoring.• Performance Measurement—enhanced meteringand emission reduction reporting—up to four creditsprescriptive.Indoor Environmental Quality• CO 2 Monitoring—one prescriptive credit.• Increase Ventilation Effectiveness—BAS can aid inearning this credit.• Controllability of Systems—up to two prescriptivecredits.• Thermal Comfort—up to two prescriptive credits—comply with ASHRAE Standard 55 and permanentlymonitor temperature and humidity.• Outdoor Air Delivery Monitoring—one prescriptivecredit.The conclusion is that a BAS can directly add topoints accumulation, but indirectly has a great deal ofinfluence on other points. The intent of these examplesis to demonstrate how design decisions flow through theentire integrated design, having direct effects on somecredits and indirect effects on others.29.6.3 Marketing LEED and Sustainabilityto the Community, Owners, and DesignersLEED offers a great deal of value to various membersof the community. The owners benefit by having highperformance buildings that are cost efficient and providefor better employee productivity. The design communitybenefits by now having a way to craft a stronger valuemessage for superior architecture and design. LEED providesa way for designers to qualify and quantify theircompetitive advantage over other non-green designers.The community benefits by having a program such asLEED that promotes urban and brownfield development,reduces demands on infrastructure such as roads andwaste disposal, improves the environment, and providesfor a healthy living style.Elaborating as to why a design team should be pro-

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 807moting LEED design, let us examine the overall life cyclecost of a typical commercial office building.Ownership Cost Breakdown—40 Year Life Cycle CostsConstruction or First Cost is 11%Financing is 14%Alterations are 25%Operations are 50%It is interesting that the cost that most design teamsgrapple with, first costs, and keeping them low, is actuallythe least significant cost element in the overall lifecycle costs of a building. Thus, those decisions to keepfirst costs low by specifying cheaper designs and equipmentcan have a serious negative impact on the overalllife cycle performance of a facility.Additionally, considering that first cost is only 11%of the total life cycle cost and that A/E fees are only 6%to 8% of that 11%, or .88% of the total costs, could it bebeneficial to the owner to pay more for superior architectureand engineering? This is because the designs andselections made by the design team have a great deal ofleverage on the total costs of life cycle ownership costs.29.6.4 Credits that Engineers can help in acquiringMany times it is believed that the architectural professionhas the most potential to aid in acquiring LEEDcredits. However it is the engineering profession that infact influences the most points. For example:• Energy & atmosphere and indoor environments are40% of available points.• Minimum energy performance is a prerequisite!• Energy modeling is required.• Energy measurement and verification is a credit.• Commissioning is required.Other benefits engineers bring to the design team• Requires creativity vs. CAD commodity design.Creativity is desired and rewarded.• Promotes investment in A/E design $ to value engineerbefore, not after.• Promotes the collective wisdom of the integrateddesign team.• This is the interplay between professions that occursduring the design charette.• Catalyst for the design team to do the high performancejob that it is capable of.• Provides a value message of premium engineeringand design to the owner—this may be the path tohigher fees and/or more work?29.6.5 Impediments to Green AcceptanceTypically, the first and possibly most serious impedimentto the wide-scale adoption of LEED is the perceptionthat it costs more. The facts are that it may addcost, from 1% to 5%, depending upon the level of greenthe ownership team has identified in that design charette.But it does not necessarily cost more if the design team isclever about making design decisions and using all availableresources that may be at hand.Hint: If trying to promote LEED, look to identifymarket conditions in the project’s locale that supportsLEED. Many states have various programs to incentivizeenergy efficiency and other marketplace conditionswhich can affect the viability of a green project. The followinguses New Jersey as an example of “market conditions”that can drive LEED adoption:The Case for LEED in New Jersey:First cost is less of an issue because of high efficiencyequipment incentives. There is a program through the NJBoard of Public Utilities called NJ Smart Start Buildings,which provides rebates for high efficiency equipment suchas lighting, HVAC, boilers, and chillers, as well as commissioningand design team meetings. Essentially much of thecost differential between cheap inefficient equipment andhigh efficiency equipment is offset by the rebates.Renewable energy sources are promoted by statewideprograms such as the Clean Energy Program. Similarto Smart Start, renewables such as wind, solar PV, andbiomass projects are rebated up to 50% of the initial installedcost.ASHRAE Std 90.1-1999 is the State Energy Code aswell as the prerequisite for energy and atmosphere LEEDcredits. So there is no additional cost for NJ buildings tocomply with this. In other states that may not have thiscode requirement, compliance with ASHRAE 90.1 wouldadd cost.Many brownfields are available for developmentwith incentives from the NJ Economic Development Administration,NJEDA. This can be a simple prescriptivecredit.Mass transportation is generally adjacent to thebrownfields, which aids in acquiring more points. NJ, asthe most densely populated state in the nation, has manyformer industrial sites which are in inner city areas closeto mass transit and part of urban renewal. Thus a brownfieldsite can facilitate a number of other credits.Environmental—voluntary LEED adoption decreasesneed for additional regulation. NJ is one of themost regulated and legislated states, but if we have moreadoption of LEED, many of the goals of environmentallegislation can be achieved voluntarily.

808 ENERGY MANAGEMENT HANDBOOKHigh energy costs in NJ promote equipment andoperational cost efficiencies. Paybacks on high efficiencyequipment are quicker than in other states, which helpsdrive the recognized value of energy efficiency versuslow first cost equipment.Other states and regions may have similar incentivesand programs. New York and California are two thatcome to mind. In addition, the Energy Policy Act of 2005,promises to encourage energy efficiency through variousprograms.29.7 ASHRAE GUIDES DEVELOPEDTO SUPPORT LEEDIntroduction to ASHRAE’s GreenGuide and theAdvanced Energy Design GuideBesides the market conditions which may favorLEED adoption, organizations, such as ASHRAE, are developingtools to help the design teams accomplish theirgreen design goals.During the year 2000, the USGBC released its firstGreen Building Rating System. Called “Leadership in Energyand Environmental Design for New Construction,”LEED 2.0, later revised and reissued as LEED 2.1 NC.Although the LEED rating system incorporated manyASHRAE standards into the rating system, the architecturalcommunity, not the engineering community, was inthe vanguard of LEED. ASHRAE recognized the potentialvalue of LEED to the community and the building industryand the value that ASHRAE could bring to the LEEDprogram. Under then president-elect Bill Coad, Tech Committee1.10 Energy Resources was tasked with developinga handbook or guide for sustainable engineering design,specifically targeted for ASHRAE members. In addition,during this time, after dialog between both organizations,the USGBC and ASHRAE entered into a partnering agreement.The result of these efforts was the development ofthe ASHRAE GreenGuide, released in December 2003, toassist the USGBC in its efforts at promoting sustainabledesign. Additionally, in 2005, ASHRAE released the AdvancedEnergy Design Guide for Small Office Buildings (lessthan 20,000 square feet). This guide provides a prescriptivedescription and requirements for various buildingcomponents and systems that would be energy efficient.29.7.1 ASHRAE GreenGuideThe ASHRAE GreenGuide aids designers of sustainable,high performance green facilities. It offers various“green tips” to aid the integrated design team in developinga green building. In 2002, the American Society ofRefrigerating and Air-Conditioning Engineers, ASHRAE,and the United States Green Building Council, USGBC,entered into a partnering agreement to team together topromote green buildings. This GreenGuide was developedby ASHRAE to assist the USGBC in their efforts at promotingsustainable design.The guide was developed to provide guidance onhow to apply green design techniques. Its purpose is tohelp the designer of a “green design” with the question of“What do I do next?” It is organized to be relevant to theaudience, useful, and practical, and to encourage innovativeideas from the design team. A key component of theguide is the “green tips,” which will be covered in somedetail later in this section.However, the guide is not a consensus document,and one does not have to agree with all elements of theguide for it to be helpful. It was not developed to motivatethe use of green design.29.7.1.1 Green Design, Sustainability and Good Design“Green” has become one of those words that canhave too many possible meanings. One of the USGBC’sinitial goals was to provide a definition of green throughthe development and release of the LEED rating system.It was here that we had a measurable, quantifiable wayof determining how green a building was. It also addressesthe “greenwashing” issue wherein all types ofgreen technologies and techniques could be employed,some valid and others questionable, all in the effortto be able to label a building green. The conclusion isthat green buildings are LEED buildings. This messageis almost universally accepted, in the USA as wellas internationally. But do be aware that there are otherstrong green rating systems that have been developed inCanada and Europe. For example Canada has the GreenBuilding Challenge, GBC, and Britain has the BuildingResearch Establishment Environmental AssessmentMethod, BREEAM. According to the guide, the consensuson green buildings is that they achieve a high levelof performance over the full life cycle in the followingareas:• Minimal consumption of nonrenewable, depletableresources such as water, energy, land, and materials.• Minimal atmospheric emissions that have negativeenvironmental effects.• Minimal discharge of harmful liquid and solid materials,including demolition debris at the end of abuilding’s life.• Minimal negative impact on site ecosystems.• Maximum quality of indoor environment includinglighting, air quality, temperature, and humidity.

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 80929.7.1.2 “Good” DesignASHRAE asks whether good design is intrinsicallygreen design. The GreenGuide authors make the distinctionbetween green and good. Good design includes:• Meeting the purpose and needs of the building’sowners and occupants.• Meets the requirements of health and safety.• Achieves a good indoor environment consisting ofthermal comfort, indoor air quality, acoustical comfort,and is compatible with the surrounding buildings.• Creates the intended emotional impact on building’soccupants.But this is not green design in the sense that it doesnot address energy conservation, environmental impact,low impact emissions, and waste disposal. So integratinggood and green design, such as the LEED rating system,helps us achieve the optimum building for the owners’needs as well as the needs of society. Hence the authorsof GreenGuide “strongly advocate that buildings shouldstrive to achieve both.”The GreenGuide emphasizes the design process.This process is the first crucial element in producing agreen building. There needs to be an integrated designteam created in the beginning. This team should includethe owner, project manager, representative of theend user, architect, mechanical engineer, plumbing andfire protection engineer, electrical engineer, lighting designer,structural engineer, landscaping specialist, civilengineer, energy analyst, environmental consultant,commissioning authority, construction manager, cost estimator,building operator, and code enforcement representative.Each individual profession works together in ateam environment to establish the building’s goals andthe manner in which these goals will be achieved. Eachprofession must be able to recognize the impact of oneanother on others designs and process.For example, during conceptual design, the architectwill determine the size and number of floors of thebuilding. The building envelope will determine the sizeof HVAC equipment, as well as the types of systems beingevaluated. The energy analyst will advise the teamon the energy cost implications of their selections. Everyoneis interdependent upon the others. What shouldcome at the end of this iterative process is a single integrateddesign that functions a unit and not as a collectionof parts.Integration and interdependency of the design teamprofessionals are the keys to a successful green design.29.7.2 Conceptual Engineering DesignThe principle intent of the GreenGuide is to assistthe design-engineering professionals in integrating theirskills and bringing value into the green design. The guidediscusses a number of design responsibilities and suggestsa number “green tips” to the design team. You mayfind that these suggestions are not new to you, and manyof these concepts have been in use or at least in considerationfor years. However, with the advent of the desire tobuild green, which requires high performance systems,these techniques have a much better chance of being incorporatedinto the green building than in the past whenfirst cost was possibly the primary concern of on ownerand the design team. The tips are arranged so that theyare listed after the design section responsibility to whichit is most closely linked.29.7.2.1 Load DeterminationLoads are determined by summing up internal andexternal gains and losses. The HVAC engineer can assistthe architect in determining necessary characteristics ofthe building envelope. Working together with the energyanalyst and others, they will aid the architect in selectingbuilding orientation, insulation, fenestration, roofing,lighting, day lighting, systems sizes, and efficiencies, etc.A key element to strive for is the initial reduction of loadsresulting from efficient building envelope, lighting loads,power loads and A/C tonnages.Green Tip #1: Night PrecoolingThis involves the circulation of cool air during thenighttime hours during the cooling season with the intentof cooling the structure. There are two variations on thistheme. First is the use of fans only to bring in cool ventilationair; this is a somewhat passive technique. The otheris to actively run the HVAC plants to precool the facility,potentially below daytime occupied temperatures to takeadvantage of the building’s thermal mass. Parameters toconsider when evaluating this strategy are local diurnaltemperature variations, humidity levels, and thermal couplingof the circulated air to the building’s thermal mass.29.7.2.2 Space Thermal/Comfort Delivery SystemsOccupant comfort and health are important and thequality of the indoor environment promotes satisfied,productive workers within the building. Green buildingsprovide a more productive workplace environment.Thermal comfort is primarily concerned with satisfyingASHRAE Standard 55 for temperature and humidityrequirements.Indoor air quality is primarily concerned withASHRAE Standard 62 for fresh air and ventilation. How-

810 ENERGY MANAGEMENT HANDBOOKever, both of these can negatively impact energy consumptionand thus the following green tips are offered:Green Tip #2 Air-to-Air Heat Recovery—Heat Exchange Enthalpy WheelsThis is a rotary cylinder filled with an air-permeablemedium with a large internal surface area. Intake andexhaust air streams pass through opposite ends of thewheel in a reverse flow configuration. Latent and sensibleheat are then transferred from exhaust to inlet air, therebyrecovering some of the conditioning energy that was investedin the exhaust air.Green Tip #3 Air-to-Air Heat Recovery—Heat Pipe SystemsThese are passive devices, usually configured astubes with fins for maximum surface area. They contain athermal fluid that transports sensible heat only betweenexhausts and inlet air streams.Green Tip #4 Air-to-Air Heat Recovery—Run Around SystemsThis consists of energy recovery coils in the exhaustand inlet air streams and a circulating loop containing aheat transfer fluid. These systems do not transfer latentenergy. An option added to this system and the heat pipesystem is the use of an indirect evaporative water processthat can reduce cooling loads in addition to the heat recoveryprocess.Green Tip #5 Displacement VentilationThis technique supplies conditioned air at a temperatureslightly lower than the desired room temperature at lowvelocities horizontally at the floor level. Returns are locatedin the ceiling. This supply air rises by convection, picks upthe room load and exits through the ceiling returns.Green Tip #6 Dedicated Outdoor Air Systems ( DOAS)This uses a dedicated, separate air handler to conditionthe outdoor air before delivering it directly to theoccupants. The air delivered should be conditioned sufficientlyto not adversely affect the thermal comfort ofthe occupants. The only absolute with this system is thatthe ventilation air must be delivered directly to the spacefrom a separate system. Control strategy, energy recovery,and leaving air conditions are all variables that can befixed by the designer.Green Tip #7 Ventilation Demand Control Using CO 2CO 2 concentrations are measured in a space andventilation rates are automatically adjusted by the BASto maintain CO 2 concentrations within predeterminedlimits. This system is best used in buildings and spaceswith variable occupancies such as public spaces, theaters,meeting rooms etc.Green Tip #8 Hybrid VentilationThis allows the controlled introduction of outsideair ventilation into a building by both mechanical andpassive means. It is sometimes called “mixed mode ventilation.”It has built-in strategies to allow the mechanicaland passive portions to work with one another so as to notcause additional ventilation loads as would occur usingmechanical ventilation alone. This is a non- conventionaltechnique that has the promise of reducing operating expensesas well as providing a healthier stimulating environment.29.7.2.3 Energy Distribution SystemsThese provide the heating, cooling, lighting, andelectric power throughout the building. The most commonmedia to distribute energy are steam, hydronics (water),air and electricity. Steam supply, because of its pressurecharacteristics, does not need to be pumped although generallythe steam condensate is pumped back to the boilers.Water and air are the principle media that require mechanicalintervention for distribution, and hence can be majorconsumers of electric power for pumps and fans.Green Tip #9 Variable Flow/VariableSpeed Pumping SystemsPumps and fans can be significant users of electricpower. In a conventional application, the pumps and fansoperate at a fixed rate based on maximum design conditions,regardless of actual loads. Adding variability to thepumping and fans to allow modulation of flows basedupon actual systems needs as opposed to design conditionscan provide significant electrical savings.29.7.2.4 Energy Conservation SystemsThis section focuses on the equipment that generateselectricity, steam, hot water, and chilled water. Theseinclude distributed electrical generation, boilers, furnaces,electrically driven water chillers and thermally drivenabsorption chillers.Green Tip #10 Low-NO x BurnersThese are natural gas burners that improve energyefficiency and lower emissions of oxides of nitrogen, NO x .They can be purchased as an option for new equipmentor retrofitted to existing equipment.Green Tip #11 Combustion Air PreheatingThis tip preheats combustion air by using waste

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 811heat from the exhaust stack, increasing energy efficiencyof equipment such as boilers and furnaces. The principleis to introduce preheated hot air into the combustor insteadof cold air, thereby reducing energy consumption.Green Tip #12 Combustion Space/Water HeatersThese consist of a storage water heater, a heat deliverysystem such as fan coils or baseboards, and associatedpumps and controls. The single unit does dual duty,as a water heater for domestic hot water heater and as asource of hot water for the hydronic heating system.Green Tip #13 Ground Source Heat Pumps, GSHPThese extract heat stored in the ground when in theheating mode, and rejects heat removed from the buildinginto the ground in the cooling season. They consistof a loop of piping or a well in the ground and indoorunits consisting of evaporators and condensers connectedinto the ground water loop. GSHPs can reduce the energyrequired for space heating, cooling, and service water incommercial/institutional buildings by as much as 50%.Green Tip #14 Water Loop Heat Pump SystemsConsists of multiple water source heat pumps within abuilding and tied into a neutral temperature loop that servesas a heat source and a heat sink. The loop temperature inturn is maintained at this neutral point by supplementingwith heat from a boiler or cooling from a cooling tower.Green Tip #15 Thermal Energy Storage for CoolingThis is an active storage system that uses the building’scooling equipment to remove heat, usually duringthe night and off-peak periods, to take advantage oflower-cost electricity during those periods and make iceslurry or chilled water. This enables a number of controland operational strategies. For example, smaller chillerscan be purchased and the building peak loads are satisfiedwith ice made during the off- peak periods.Green Tip #16 Double Effect Absorption ChillersChilled water for facility cooling can be driven byelectricity or thermal energy such as steam. In absorptionchillers, thermal energy such as steam is used to drive aprocess using water and a salt solution such as lithiumbromide in a vacuum-sealed shell to produce chilledwater. There are no ozone depleting refrigerants used inthis process. Absorbers come in single and double effecttypes, the double effect having a COP of about 1.2 versusthe single effect COP of about .8.Green Tip #17 Gas Engine Driven ChillersChilled water systems that use energy sources besideselectricity can help offset high electricity costs.Generally, these are engines run on natural gas and arecoupled to a chiller compressor section. Essentially, it isan engine replacing the electric motor of an electricallydriven chiller. Gasoline and diesel fuel can also be useddepending upon engine types selected.Green Tip #18 Gas-Fired Chiller/HeaterGas-fired absorption chillers are a special type of absorptionchillers wherein the thermal energy source is adirect burner typically firing natural gas, although otherfuels could be used. This is as opposed to the conventionalabsorber operated with steam. The chiller/heater cando double duty both to make cold and hot water simultaneously.Green Tip #19 Desiccant Cooling andDehumidificationRotary desiccant dehumidifiers use solid desiccantssuch as silica gel to attract water from the air. Humid airis dehumidified in one part of the desiccant bed while adifferent part of the bed is dried for reuse by a second airstream.Green Tip #20 Evaporative CoolingThis technique can be used to reduce the amountof energy consumed by mechanical cooling equipment.There are two types of evaporation coolers, direct and indirect.Direct introduces water directly into the air stream,and the water evaporates, reducing the dry bulb temperatureof the air while raising the relative humidity. Indirectsystems spray water onto a coil and, through evaporationfrom the fins of the coils, reduces the dry bulb temperaturealso. However, no water is added to the air streamwith this method.29.7.2.5 Energy and Water SourcesThe designer may not have much option in the selectionsof energy sources for the building being designed.Typically, energy is provided by the area utilities in theform of gas and electric. However, the designer can consideroptions to supplement the conventional energy sourceswith renewables such as wind, solar photovoltaics, PV,and hydro as the site permits. PV is generally the mostapplicable renewable energy source for buildings. Theothers are too site specific.Green Tip #21 Solar Energy System—PhotovoltaicsSunlight shines on solid state crystals of silicon andgenerates low voltage direct current electricity. Applicationssimply require full access to the sun and sufficientspace, typically on the roof, to generate useful amounts of

812 ENERGY MANAGEMENT HANDBOOKelectricity. The low voltage DC can be inverted and voltageboosted to be supplied directly into the building’selectrical distribution system. Thus there is no need tolocate a specific low voltage DC load for the power produced.29.7.2.6 Lighting SystemsThe GreenGuide section here is designed to familiarizethe HVAC&R engineer with lighting systems andtheir potential impact on the equipment sizes and designs.However, it states that it is best to engage a lightingdesigner who will design according to IESNA standards(Illuminating Engineering Society-North America). However,the guide does make one suggestion for lighting.Green Tip #22 Light ConveyorA light conveyor is a large pipe or duct with reflectivesides that transmits artificial or natural light along itslength. There are occasions wherein designers have usedlight tracking to enhance the performance of the lightconveyors.29.7.2.7 Plumbing and Fire Protection SystemsAlthough not usually within the purview of the HVACdesigner’s expertise, both must be able to work togetherin developing a fully integrated design. There are severaltips that include aspects of both skill sets.Green Tip #23 Water Conserving Plumbing FixturesWater conserving strategies can save owners bothconsumption and demand charges. Options to be consideredfor water conservation are:• Infrared faucet sensors• Delayed action shut off valves• Low flow toilets• Faucets with flow restrictors• Metering faucets• Water efficient dishwashers• Waterless urinals• Closed cooling towers to eliminate drift and filtersfor cleaning tower waterGreen Tip #24 Graywater SystemsGraywater is defined as wastewater coming fromoperations such as showers, bathtubs, washing machinesand sinks. This is separate from blackwater, which iswastewater from toilets and sinks that contain organic ortoxic materials. Where allowed by code, graywater canbe filtered, treated, stored, and later used for nonpotableuses such as irrigation of landscaping and flushing oftoilets. Distribution would be through a piping systemclearly separated from all others.Green Tip #25 Point-of-Use Domestic Hot Water HeatersThese provide small quantities of hot water at thepoint of use, without a tie into a central hot water source.Generally electrically heated, savings are obtainedthrough the avoidance of large amounts of hot water storage,and thermal losses of hot water distribution piping.Green Tip #26 Direct Contact Water HeatersThis consists of a heat exchanger in which flue gasesare in direct contact with the water to be heated. Coldwater enters the top of the heat exchanger, natural gas isburned and flows up through the heat exchanger whereinthe water is cascading down, acquiring the heat of theburned gas. Although there is direct contact between exhaustgases and the water, there is very little contaminationof the water. These are suitable for dairy and foodprocessing uses, as well as many other processes.Green Tip #27 Rainwater HarvestingAlthough this is not a new concept, it has beenaround for thousands of years, and is a simple and effectivetechnology to apply. Rainwater is channeled into cisternsfor nonpotable uses as needed for irrigation, toiletflushing etc.29.7.2.8 ControlsThese can be thought of as the “nervous systems”of the building’s mechanical and electrical infrastructure.Controls can be used for temperature and humidity control,ventilation control, energy management and analysis,etc. (See also Chapter 22.)Green Tip #28 Mixed Air Temperature ResetThis refers to the mix of outside and return air thatexist on an operating system supply air handling unit priorto any “new” energy being applied to it. The conceptis to reset the mixed air temperature (MAT) to a temperaturethat just satisfies the lowest cold air demand. Resetcontrols involve raising the setpoint of the MAT controlsbased on input that indicates the demand of the zoneneeding the coldest air, limited by the minimum amountof outside air for IAQ purposes.Green Tip #29 Cold Deck Temperature Reset,CDT, with Humidity OverrideCDT is similar to MAT, but it applies to air leavinga cooling coil. The concept here is to allow the dischargetemperature from the cooling coil to go to higher temperatureswhen demand for cooling is low. However, thisstrategy can create high humidity conditions indoors onoccasion. Thus humidity sensors are located in the spaces

SUSTAINABILITY AND HIGH PERFORMANCE GREEN BUILDINGS 813to override the rising cold deck temperatures and drivetemperatures back down to extract more moisture basedupon demands for temperature and humidity control.Other elements of green, good design covered bythe GreenGuide but with no specific green tips suggestedare:• Expressing and Testing Concepts• Completing Design and Documentation for Construction• Post Design—Construction to Demolition• Builder/Contractor Selection• Construction• Commissioning• Operation/Maintenance/Performance Evaluation29.7.3 ASHRAE Advanced Energy Design GuideThis ASHRAE manual for small office buildings isintended for buildings up to 20,000 square feet in size.It is a prescriptive description of how designers andcontractors can achieve up to 30% reduction in energyconsumption in comparison to ASHRAE Standard 90.1-1999. This would be worth four credits in LEED NC. Themanual, although primarily of use for new construction(NC) design, can also be used as a planning tool for LEEDEB because the measures given are a form of best energypractices for new or existing buildings.The guide is divided into three basic chapters:(I) Integrated Process to Achieve Energy Savings1. Pre-Design Phase—Prioritize Goals2. Design Phase3. Construction4. Acceptance5. Occupancy6. Operation(II) The guide provides recommendations by climate. Itdivides the USA into 8 zones, with #1 the warmest zone(the southern tip of Florida) to #8 for Alaska.(III) For each climate zone there is a table providing prescriptiverecommendations for the building componentslisted below:1. Roofs2. Walls3. Floors4. Slabs5. Doors6. Vertical Glazing7. Skylighting8. Interior Lighting9. HVAC category includes air-conditioning, heating,air side economizers, ventilation and air handlingducts.10. Service Water HeatingReferences:ASHRAE GreenGuide, 2003—Editor David L. GrummanASHRAE Advanced Energy Design GuideASHRAE Website www.ASHRAE.ORGEnergy User News, May 2004 by Peter D’Antonio, entitled “The LEEDingWay.”The United States Green Building Council, www.USGBC.ORG


APPENDIX ITHERMAL SCIENCES REVIEWL.C. WITTEProfessor of Mechanical EngineeringUniversity of HoustonHouston, TexasPotential, kinetic, and internal energy forms generallyare nonchemical and nonnuclear in nature. Theyrelate to the position, velocity, and state of material ina thermodynamic system. More detailed representationsof these energy forms will be shown later.I.1. INTRODUCTIONMany technical aspects of energy management involvethe relationships that result from the thermal sciences:classical thermodynamics, heat transfer, and fluid mechanics.For the convenience of the user of this handbook,brief reviews of some applicable topics in thethermal sciences are presented. Derivation of equationsare omitted; for readers needing those details, referencesto readily available literature are given.I.2. THERMODYNAMICSClassical thermodynamics represents our understandingof the relationships of energy transport to propertiesand characteristics of various types of systems.This science allows us to describe the global behaviorof energy-sensitive devices. The relationships that can bedeveloped will find application in both fluid-flow andheat-transfer systems.A thermodynamic system is a region in space thatwe select for analysis. The boundary of the systemmust be defined. Usually, the system boundary willcoincide with the physical shell of a piece of hardware.A closed system is one where no mass may cross theboundary, whereas an open system, sometimes calleda control volume, will generally have mass flowingthrough it.We generally divide energy into two categories:stored and transient types of energy. The stored formsare potential, kinetic, internal, chemical, and nuclear.These terms are fairly self-descriptive in that they relateto ways in which the energy is stored. Chemicaland nuclear energy represent the energy tied up in thestructure of the molecular and atomic compounds themselves.These two types of stored energy presently formthe prime energy sources for most industrial and utilityapplications and thus are of great importance to us.815I.2.1 Properties and StatesThermodynamic systems are a practical necessityfor the calculation of energy transformations. But todo this, certain characteristics of the system must bedefined in a quantifiable way. These characteristics areusually called properties of the system. The propertiesform the basis of the state of the system. A state is theoverall nature of the system defined by a unique relationshipbetween properties.Properties are described in terms of four fundamentalquantities: length, mass, time, and temperature.Mass, length, and time are related to a force throughNewton’s second law.In addition to the fundamental quantities of asystem there are other properties of thermodynamicimportance. They are pressure, volume, internal energy,enthalpy, and entropy—P, V, U, H, and S, respectively.Equation of State. Returning to the concept of astate, we use an equation of state to relate the pertinentproperties of a system. Generally, we use P = P(m, V,T) as the functional equation of state. The most familiarform is the ideal gas equation of state written asorPV= mRTPV = nRT(I.1)(I.2)Equation I.1 is based on the mass in a system, whereasequation I.2 is molal-based. R is called the gas constantand is unique to a particular gas. R, on the otherhand, is called the universal gas constant and retainsthe same value regardless of the gas (i.e., R = 1545 ftlb ƒ /lb m• mol • °R = 1.9865 Btu/lb m• mol • °R. It canbe easily shown that R is simply R/M, where M is themolecular weight of the gas.The ideal gas equation is useful for superheatedbut not for saturated vapors. A vexing question thatoften arises is what is the range of application of theideal gas equation. Figure I.1 shows the limits of appli-

816 ENERGY MANAGEMENT HANDBOOKpressures. This uses the concept that the volume of asystem is occupied by all the components. Using a twocomponentsystem as an example, we writeP 1 V 1 = n 1 RT 1and P 2 V 2 = n 2 RT 2Also P 1 V 1 = n 1 RT 2Since V, R, and T are the same for all three equationsabove, we see thatFig. I.1 Applicability of ideal gas equation of state.cability. The shaded areas demonstrate where the idealgas equation applies to within 10% accuracy.If high pressures or vapors near saturation areinvolved, other means of representing the equation ofstate are available. The compressibility factor Z for exampleZ = PV/RT, is a means of accounting for nonidealgas conditions. Several other techniques are available;see Ref. 1 for example.Changes of state for materials that do not behavein an ideal way can be calculated by use of generalizedcharts for property changes. For example, changesin enthalpy and entropy can be presented in terms ofreduced pressure and temperature, P r and T r . The reducedproperties are the ratio of the actual to the criticalproperties. These charts (see Appendix II) can be usedto calculate the property changes for any change in statefor those substances whose thermophysical propertiesare well documented.Ideal Gas Mixtures. Where several gases aremixed, a way to conveniently represent the properties ofthe mixture is needed. The simplest way is to treat thesystem as an ideal gas mixture. In combustion systemswhere fuel vapor and air are mixed, and in the atmospherewhere oxygen nitrogen and water vapor formthe essential elements of air, the concept of the ideal gasmixture is very useful.There are two ways to represent gas mixtures. Oneis to base properties on mass, called the gravimetricapproach. The second is based on the number of molesin a system, called a molal analysis. This leads to thedefinition of a mass fraction, x i = m i /m t , where m i isthe mass of the ith component in the mixture and m t isthe total mass of the system, and the mole fraction, y i= n i /n t , where n represents the number of moles.Commonly, the equation of state for an idealmixture involves the use of Dalton’s law of additiveandP t = P 1 + P 2P 1 P 2 V = n 1 n 1 RTfor a mixture. Each gas in the mixture of ideal gases thenbehaves in all respects as though it exists alone at thevolume and temperature of the total system.I.2.2 Thermodynamic ProcessesA transformation of a system from one state toanother is called a process. A cycle is a set of processesby which a system is returned to its initial state. Thermodynamically,it is required that a process be quantifiableby relations between properties if an analysis is to bepossible.A process is said to be reversible if a system canbe returned to its initial state along a reversed processline with no change in the surroundings of the system.In actual practice a reversible process is not possible.All processes contain effects that render them irreversible.For example, friction, nonelastic deformation,turbulence, mixing, and heat transfer are all effects thatcause a process to be irreversible. The reversible process,although impossible, is valuable, because it serves as areference value. That is, we know that the ideal processis a theoretical limit toward which we can strive byminimizing the irreversible effects listed above.Many processes can be described by a phrasewhich indicates that one of its properties or characteristicsremains constant during the process. Table I.1shows the more common of these processes togetherwith expressions for work, heat transfer, and entropychange for ideal gases.I.2.3 Thermodynamic LawsThermodynamic laws are relationships betweenmass and energy quantities for both open and closedsystems. In classical form they are based on the conservationof mass for a system with no relativistic effects.Table I.2 shows the conservation-of-mass relations for

THERMAL SCIENCES REVIEW 817Table I.1 Ideal Gas Processes aProcess Describing Equations 1 W 2 1 Q 2 S 2 – S 1Isometric or 0V = c,constantυ = c,t υ = cU 2 – U 1 = m (u 2 – u 1 )volume = mc ν (T 2 – T 1 ) (2)ms 2 0 – s 1 0 – R ln ρ 2ρ 1(3)= mc υ ln T 2T 1Isobaric, isopiestic,P = c, V/T = c, υ t = c p (V 2 – V 1 ) H 2 – H 1 = m (h 2 – h 1 ) m s 0 2 – s01or constantpressure = mR(T 2 – T 1 ) = mc p (T 2 – T 1 ) = mc p ln T 2T 1Isothermal or T = c, pV = pv = c p 1 V 1 ln r (4) = (p 2 – V 2 ) ln r 1 W 2constant= mRT ln rtemperaturemR ln rIsentropic or s = c U 1 – U 2 = m (u 1 – u 2 )reversible 0 0adiabaticPolytropicpV k 5 = c, TV k –1= c, p k –1 T k –1 = cpV k 6 = c, TV n –1P 1 V 1 – P 2 V 2k –1P 1 V 1 – P 2 V 2n –1= mR T 1 – T 2k –1= mR T 1T 2n –1Use first lawm s 2 0 – s 1 0 – R ln ρ 2ρ 1= k – nk –1 ln r= c, p n –1 T n –1 = c———————————————————————————————————————————————————a (1) c stands for an unspecified constant; (2) the second line of each entry applies when cp and c v are independent oftemperature; (3) s 2 – s 1 = s 0 (T 2 ) – s 0 (T 1 ) – R ln (p 2 /p 1 ); (4) r = volume ratio or compression ratio = V 2 /V 1 ; (5) k = c p /c v> 1; (6) n = polytropic exponent.various types of systems. In energy conservation, thefirst and second laws of thermodynamics form the basisof most technical analysis.For open systems two approaches to analysis canbe taken, depending upon the nature of the process. Forsteady systems, the steady-state steady-flow assumptionTable I.2 Law of Conservation of Mass—————————————————————————Closed system, any process: m = constantOpen system, SSSF:m = Σ minΣ outOpen system, USUF:m 2 – m 2 c.v. = Σ m = min outΣ tOpen system, general case:ddt m c.v. = Σ m = minΣ or d ρ dV = ρ dVin dtrn dAAv—————————————————————————(SSSF) is adequate. This approach assumes that the stateof the material is constant at any point in the system.For transient processes, the uniform flow uniform state(UFUS) assumption fits most situations. This involvesthe assumption that at points where mass crosses thesystem boundary, its state is constant with time. Also,the state of the mass in the system may vary with timebut is uniform at any time.Tables I.3 and I.4 give listings of the conservationof-energy(first-law) and the second-law relations forvarious systems.The first law simply gives a balance of energyduring a process. The second law, however, extends theutility of thermodynamics to the prediction of both thepossibility of a proposed process or the direction of asystem change following a perturbation of the system.Although the first law is perhaps more directly valuablein energy conservation, the implications of the secondlaw can be equally illuminating.

818 ENERGY MANAGEMENT HANDBOOKClosed system, cyclic process:Table I.3 First Law of ThermodynamicsdQ =dWClosed system, state 1 to state 2:1 Q 2 = E 2 – E 1 + 1 W 2E = internal energy + kinetic energy + potential energy = m(u + V 2 /2 + gz)Open system:Q c.v. = d dtρ u + V 2 ⁄ 2+gz dV +υwhere enthalpy per unit mass h = u + pv;wherealternative form:ρ h + V 2 ⁄ 2+gz V rn dA +W c.v. ,AQ c.v. + Σ m h + V 2 ⁄ 2+gz = W c.v. + minΣ h + V 2 ⁄ 2+gz + E c.v. ,outE c.v. = d dtvρ u + V 2 /2 + gz dVOpen system, steady state steady flow (SSSF):ΣinQ c.v. + mh+ V 2 /2 + gz = W c.v. + m h+V 2 /2 + gzΣ outOpen system, uniform state uniform flow (USUF):1 Q c.v. + mh+ V 2 /2 + gz = 1 W 2c.v. + Σ mh+ V 2 /2 + gz + mu+ V 2 /2 + gz 2c.v.out1c.v.ΣinThere are two statements of the second law. Thetwo, although appearing to be different, actually can beshown to be equivalent. Therefore, we state only one ofthem, the Kelvin-Planck version:It is impossible for any device to operate in a cycle andproduce work while exchanging heat only with bodiesat a single fixed temperature.The other statement is called the Clausius statement.The implications of the second law are many. Forexample, it allows us to (1) determine the maximumpossible efficiency of a heat engine, (2) determine themaximum coefficient of performance for a refrigerator,(3) determine the feasibility of a proposed process,(4) predict the direction of a chemical or other type ofprocess, and (5) correlate physical properties. So we seethat the second law is quite valuable.I.2.4 EfficiencyEfficiency is a concept used to describe the effectivenessof energy conversion devices that operate incycles as well as in individual system components thatoperate in processes. Thermodynamic efficiency, η, andcoefficient of performance, COP, are used for devicesthat operate in cycles. The following definitions apply:η = W netQ HCOP = Q LWwhere Q L and Q H represent heat transferred from coldand hot regions, respectively, W net is useful work pro-

THERMAL SCIENCES REVIEW 819Table I.4 Second Law of ThermodynamicsClosed system, cyclic process:dQT ≤ 0Closed system, state 1 to state 2:12Open system:ordQT ≤ S 2 – S 1 = m S 2 – S 1dρ sdV+ Σ ms ≥ ms +dt V outΣinAQT dAshould not be confused with efficiencies applied to devicesthat operate along a process line. This efficiencyis defined asactual energy transferη device = ———————————ideal energy transferfor a work-producing device andideal energy transferη device = ———————————actual energy transferfor a work-consuming device. Note these definitionsare such that η < 1. These efficiencies are conveniencefactors in that the actual performance can be calculatedfrom an ideal process line and the efficiency, whichgenerally must be experimentally determined. Table I.5shows the most commonly encountered versions of efficiencies.ddtVρ sdVOpen system, SSSF:Σ ms + msoutΣinOpen system, USUF:+ ρ sV rn dA ≥VAQT dAQT dAAI.2.5 Power and Refrigeration CyclesMany cycles have been devised to convert heatinto work, and vice versa. Several of these take advantageof the phase change of the working fluid: forexample, the Rankine, the vapor compression, and theabsorption cycles. Others involve approximations ofthermodynamic processes to mechanical processes andare called air-standard cycles.duced in a heat engine, and W is the work required todrive the refrigerator. A heat engine produces usefulwork, while a refrigerator uses work to transfer heatfrom a cold to a hot region. There is an ideal cycle, calledthe Carnot cycle, which yields the maximum efficiencyfor heat engines and refrigerators. It is composed of fourideal reversible processes; the efficiency of this cycle isand the COP isΣ outm 2 s 2 – m 1 s 1 c.v. + ms ≥ msη c =1– T LT HCOP c =TL/TH1–TL/THThese represent the best possible performance of cyclicenergy conversion devices operating between temperatureextremes, T H and T L . The thermodynamic efficiencyΣinAQ c.v.T—————————————————————————dATable I.5 Thermodynamic Efficiency—————————————————————————Heat engines and refrigerators:Engine efficiency η ≡ W/Q H ≤ η Carnot = (T H – T L )/ T H < 1Heat pump c.o.p. β' ≡ Q H /W ≤ β' Carnot = T H /(T H – T L ) > 1Refrigerator c.o.p. β ≡ Q L /W ≤ β Carnot = T L /(T H – T L ),0 < β < ∞, (Q H /Q L ) Carnot = T H – T LProcess efficienciesη ad, turbine = w actual, adiabatic /w isentropicη ad, compressor = w isentropic /w actual, adiabaticη ad, nozzle = K.E. actual, adiabatic /K.E. isentropicη nozzle = V a 2 /2gcV s 2 /2gcη cooled nozzle = w isentropic rev. /w actual—————————————————————————

820 ENERGY MANAGEMENT HANDBOOKThe Rankine cycle is probably the most frequentlyencountered cycle in thermodynamics. It is used in almostall large-scale electric generation plants, regardlessof the energy source (gas, coal, oil, or nuclear). Manymodern steam-electric power plants operate at supercriticalpressures and temperatures during the boilerheat addition process. This leads to the necessity ofreheating between high- and lower-pressure turbines toprevent excess moisture in the latter stages of turbineexpansion (prevents blade erosion). Feedwater heatingis also extensively used to increase the efficiency of thebasic Rankine cycle (see Ref. 1 for details).The vapor compression cycle is almost a reversedRankine cycle. The major difference is that a simpleexpansion valve is used to reduce the pressure betweenthe condensor and the evaporator rather than being awork-producing device. The reliability of operation ofthe expansion valve is a valuable trade-off compared tothe small amount of work that could be reclaimed. Thevapor compression cycle can be used for refrigeration orheating (heat pump).In the energy conservation area, applications of theheat pump are taking on added emphasis. The device isuseful for heating from an electrical source (compressor)in situations where direct combustion is not available.Additionally, the device can be used to upgrade thetemperature level of waste heat recovered at a lowertemperature.Air-standard cycles, useful both for power generationand heating/cooling applications, are the thermodynamicapproximations to the processes occurring inthe actual devices. In the actual cases, a thermodynamiccycle is not completed, necessitating the approximations.Air-standard cycles are analyzed by using the followingapproximations:1. Air is the working fluid and behaves as an idealgas.2. Combustion and exhaust processes are replaced byheat exchangers.Other devices must be analyzed component bycomponent using property data for the working fluids(see Appendix II). Figure I.2 gives a listing of variouspower systems with their corresponding thermodynamiccycle and other pertinent information.I.2.6 Combustion ProcessesThe combustion process continues to be the mostprevalent means of energy conversion. Natural andmanufactured gases, coal, liquid fuel/air mixtures, evenwood and peat are examples of energy sources requiringcombustion.There are two overriding principles of importancein analyzing combustion processes. They are the combustionequation and the first law for the combustionchamber. The combustion equation is simply a massbalance between reactants and products of the chemicalreaction combustion process. The first law is the energybalance for the same process using the results of thecombustion equation as input.In practice, we can restrict our discussion to hydrocarbonfuels, meaning that the combustion equation(chemical balance) is written asTable I.6 Characteristics of Some of the Hydrocarbon FamiliesFamily Formula Structure SaturatedParaffin C n H 2n + 2 Chain YesOlefin C n H 2n Chain NoDiolefin C n H 2n – 2 Chain NoNaphthene C n H 2n Ring YesAromatic Benzene C n H 2n – 6 Ring NoAromatic Naphthalene C n H 2n – 12 Ring NoH HMolecular structure of some hydrocarbon fuels:H C HH H H HH HC CH—C —C—C—C—H H—C—C—C—C—HH C HH H H HH H H HH HChain structure, Chain structure, Ring structure,saturated unsaturated saturated

THERMAL SCIENCES REVIEW 821This equation neglects the minor components of air;that is, air is assumed to be 1 mol of O 2 mixed with3.76 mol of N 2 . The balance is based on 1 mol of fuelC x H y . The unknowns are determined for each particularapplication. Table I.6 givesthe characteristics of some of thehydrocarbons. Table I.7 showsthe volumetric analyses of severalgaseous fuels.Once a combustion processis decided upon (i.e., the fuel to beused and the heat transfer/combustionchamber are selected),the relative amount of fuel andair become of prime importance.This is because the air/fuel ratio(AF) controls the temperatureof the combustion zone and theenergy available to be transferredto a working fluid or convertedto work. Stoichiometric air is thatquantity of air required such thatno oxygen would appear in theproducts. Excess air occurs whenmore than enough air is providedto the combustion process. Idealcombustion implies perfect mixingand complete reactions. In thiscase theoretical air (TA) wouldyield no free oxygen in the products.Excess air then is actual airless theoretical air.Most industrial combustionprocesses conform closely toa steady-state, steady-flow case.The first law for an open controlvolume surrounding the combustionzone can then be written. Ifwe assume that Q and W are zeroand that ∆K.E. and ∆P.E. are negligible,then the following equationresults:Fig. I.2 Air standard cycles.ΣproductsΣreactantsH e – H ref =H i – H ref+ H comb(I.3)C x H y + α(O 2 + 3.76N 2 )→ b CO 2 + c CO 2+ e H 2 O + d O 2 + 3.76a N 2Subscripts i and e refer to inlet and exit conditions,respectively. H ref is the enthalpy of each componentat some reference temperature. ∆Hcomb represents theheat of combustion for the fuel and, in general, carriesa negative value, meaning that heat would have tobe transferred out of the system to maintain inlet andexit temperatures at the same level.The adiabatic flame temperature occurs when the

822 ENERGY MANAGEMENT HANDBOOKTable I.7 Volumetric Analyses of Some Typical Gaseous FuelsVarious Natural Gases ProducerGas fromBituminous Carbureted CokeConstituent A B C D Coal Water Gas Oven GasMethane 93.9 60.1 67.4 54.3 3.0 10.2 32.1Ethane 3.6 14.8 16.8 16.Propane 1.2 13.4 15.8 16.2Butanes plus a 1.3 4.2 7.4Ethene 6.1 3.5Benzene 2.8 0.5Hydrogen 14.0 40.5 46.5Nitrogen 7.5 5.8 50.9 2.9 8.1Oxygen 0.6 0.5 0.8Carbon monoxide 27.0 34.0 6.3Carbon dioxide 4.5 3.0 2.2a This includes butane and all heavier hydrocarbons.combustion zone is perfectly insulated. The solution ofequation I.3 would give the adiabatic flame temperaturefor any particular case. The maximum adiabatic flametemperature would occur when complete combustionoccurs with a minimum of excess O 2 appearing in theproducts.Appendix II gives tabulated values for the importantthermophysical properties of substances importantin combustion.Gas Analysis. During combustion in heaters andboilers, the information required for control of theburner settings is the amount of excess air in the fuelgas. This percentage can be a direct reflection of the efficiencyof combustion.The most accurate technique for determining thevolumetric makeup of combustion by-products is theOrsat analyzer. The Orsat analysis depends upon thefact that for hydrocarbon combustion the products maycontain CO 2 , O 2 , CO, N 2 , and water vapor. If enoughexcess air is used to obtain complete combustion, no COwill be present. Further, if the water vapor is removed,only CO 2 , O 2 , and N 2 remain.The Orsat analyzer operates on the followingprinciples. A sample of fuel gas is first passed over adesiccant to remove the moisture. (The amount of watervapor can be found later from the combustion equation.)Then the sample is exposed in turn to materials thatabsorb first the CO 2 , then the O 2 , and finally the CO (ifpresent). After each absorption the volumetric change iscarefully measured in a graduated pipette system. Theremaining gas is assumed to be N 2 . Of course, it couldcontain some trace of gases and pollutants.I.2.7 PsychrometryPsychrometry is the science of air/water vapormixtures. Knowledge of the behavior of such systems isimportant, both in meteorology and industrial processes,especially heating and air conditioning. The concepts canbe applied to other ideal gas/water vapor mixtures.Air and water vapor mixed together at a totalpressure of 1 atm is called atmospheric air. Usually, theamount of water vapor in atmospheric air is so minutethat the vapor and air can be treated as an ideal gas. Theair existing in the mixture is often called dry air, indicatingthat it is separate from the water vapor coexistingwith it.Two terms frequently encountered in psychrometryare relative humidity and humidity ratio. Relativehumidity, φ , is defined as the ratio of the water vaporpressure to the saturated vapor pressure at the temperatureof the mixture. Figure I.3 shows the relationbetween points on the T–s diagram that yield the relativehumidity. Relative humidity cannot be greater thanunity or 100%, as is normally stated.The humidity ratio, ω, on the other hand, is definedas the ratio of the mass of water vapor to themass of dry air in atmospheric air, ω = m v /m a . This canbe shown to be ω = v a /v v , and a relationship between ωand φ exists, ω = (va/v g )φ, where v g refers to the specificvolume of saturated water vapor at the temperature ofthe mixture.

THERMAL SCIENCES REVIEW 823TsFig. I.3 Behavior of water in air: φ = P 1 /P 3 ; T 2 = dewpoint.A convenient way of describing the condition ofatmospheric air is to define four temperatures: dry-bulb,wet-bulb, dew-point, and adiabatic saturation temperatures.The dry-bulb temperature is simply that temperaturewhich would be measured by any of several typesof ordinary thermometers placed in atmospheric air.The dew-point temperature (point 2 on FigureI.3) is the saturation temperature of the water vapor atits existing partial pressure. In physical terms it is themixture temperature where water vapor would begin tocondense if cooled at constant pressure. If the relativehumidity is 100% the dew-point and dry-bulb temperaturesare identical.In atmospheric air with relative humidity lessthan 100%, the water vapor exists at a pressure lowerthan saturation pressure. Therefore, if the air is placedin contact with liquid water, some of the water wouldbe evaporated into the mixture and the vapor pressurewould be increased. If this evaporation were done in aninsulated container, the air temperature would decrease,since part of the energy to vaporize the water must comefrom the sensible energy in the air. If the air is broughtto the saturated condition, it is at the adiabatic saturationtemperature.A psychrometric chart is a plot of the properties ofatmospheric air at a fixed total pressure, usually 14.7 psia.The chart can be used to quickly determine the propertiesof atmospheric air in terms of two independent properties,for example, dry-bulb temperature and relative humidity.Also, certain types of processes can be describedon the chart. Appendix II contains a psychrometric chartfor 14.7-psia atmospheric air. Psychrometric charts canalso be constructed for pressures other than 14.7 psia.I.3 HEAT TRANSFERHeat transfer is the branch of engineering sciencethat deals with the prediction of energy transport caused32P 3 P 11by temperature differences. Generally, the field is brokendown into three basic categories: conduction, convection,and radiation heat transfer.Conduction is characterized by energy transfer byinternal microscopic motion such as lattice vibration andelectron movement. Conduction will occur in any regionwhere mass is contained and across which a temperaturedifference exists.Convection is characterized by motion of a fluidregion. In general, the effect of the convective motion isto augment the conductive effect caused by the existingtemperature difference.Radiation is an electromagnetic wave transportphenomenon and requires no medium for transport. Infact, radiative transport is generally more effective in avacuum, since there is attenuation in a medium.I.3.1 Conduction Heat TransferThe basic tenet of conduction is called Fourier’slaw,Q =–kA dTdxThe heat flux is dependent upon the area across whichenergy flows and the temperature gradient at that plane.The coefficient of proportionality is a material property,called thermal conductivity k. This relationship alwaysapplies, both for steady and transient cases. If the gradientcan be found at any point and time, the heat fluxdensity, Q/A, can be calculated.Conduction Equation. The control volume approachfrom thermodynamics can be applied to give anenergy balance which we call the conduction equation.For brevity we omit the details of this development; seeRefs. 2 and 3 for these derivations. The result isG + K∇ 2 T =–ρC ∂T∂τ(I.4)This equation gives the temperature distribution inspace and time, G is a heat-generation term, causedby chemical, electrical, or nuclear effects in the controlvolume. Equation I.4 can be written∇ 2 T + G K = ρC k∂T∂τThe ratio k/ρC is also a material property called thermaldiffusivity u. Appendix II gives thermophysical propertiesof many common engineering materials.For steady, one-dimensional conduction with noheat generation,

824 ENERGY MANAGEMENT HANDBOOKD 2 Tdx 2 =0This will give T = ax + b, a simple linear relationshipbetween temperature and distance. Then the applicationof Fourier’s law givesQ = kA T xa simple expression for heat transfer across the ∆x distance.If we apply this concept to insulation for example,we get the concept of the R value. R is just the resistanceto conduction heat transfer per inch of insulation thickness(i.e., R = 1/k).Multilayered, One-Dimensional Systems. Inpractical applications, there are many systems that canbe treated as one-dimensional, but they are composedof layers of materials with different conductivities. Forexample, building walls and pipes with outer insulationfit this category. This leads to the concept of overallheat-transfer coefficient, U. This concept is based on thedefinition of a convective heat-transfer coefficient,Q = hA TThis is a simplified way of handling convection at aboundary between solid and fluid regions. The heattransfercoefficient h represents the influence of flowconditions, geometry, and thermophysical properties onthe heat transfer at a solid-fluid boundary. Further discussionof the concept of the h factor will be presentedlater.Figure I.4 represents a typical one-dimensional,multilayered application. We define an overall heattransfercoefficient U asQ = UA (T i – T o )We find that the expression for U must beU = 11 + x 1+ x 2+ x 3+ 1 h 1 k 1 k 2 k 3 h 0This expression results from the application of theconduction equation across the wall components andthe convection equation at the wall boundaries. Then,by noting that in steady state each expression for heatmust be equal, we can write the expression for U, whichcontains both convection and conduction effects. The Ufactor is extremely useful to engineers and architects ina wide variety of applications.Fig. I.4 Multilayered wall with convection at the innerand outer surfaces.The U factor for a multilayered tube with convectionat the inside and outside surfaces can be developedin the same manner as for the plane wall. The result isU = 1r1 0 ln r j +1/r j+ Σ + 1r 0h 0 j k j h i r iwhere r i and r o are inside and outside radii.Caution: The value of U depends upon which radius youchoose (i.e., the inner or outer surface).If the inner surface were chosen, we would getU = 11r ir i ln r j +1/r j+ Σh 0 r 0 j k j+ 1 h iHowever, there is no difference in heat-transfer rate;that is,so it is apparent thatfor cylindrical systems.Q o = U i A i T overall = U o A o T overallU i A i = U o A oFinned Surfaces. Many heat-exchange surfacesexperience inadequate heat transfer because of lowheat-transfer coefficients between the surface and theadjacent fluid. A remedy for this is to add material tothe surface. The added material in some cases resemblesa fish “fin,” thereby giving rise to the expression “afinned surface.” The performance of fins and arrays offins is an important item in the analysis of many heat-exchangedevices. Figure I.5 shows some possible shapesfor fins.

THERMAL SCIENCES REVIEW 825The analysis of fins is based on a simple energybalance between one-dimensional conduction downthe length of the fin and the heat convected from theexposed surface to the surrounding fluid. The basicequation that applies to most fins isd 2 θ 1dA dθ h 1 dS—— + ———— – ——— θ = 0 (I.5)dx 2 A dx dx k A dxwhen θ is (T – T ∞ ), the temperature difference betweenfin and fluid at any point; A is the cross-sectional areaof the fin; S is the exposed area; and x is the distancealong the fin. Chapman 2 gives an excellent discussionof the development of this equation.The application of equation I.5 to the myriad ofpossible fin shapes could consume a volume in itself.Several shapes are relatively easy to analyze; for example,fins of uniform cross section and annular fins canbe treated so that the temperature distribution in the finand the heat rate from the fin can be written. Of moreutility, especially for fin arrays, are the concepts of finefficiency and fin surface effectiveness (see Holman 3 ).Fin efficiency η ƒ is defined as the ratio of actualheat loss from the fin to the ideal heat loss that wouldoccur if the fin were isothermal at the base temperature.Using this concept, we could writeQ fin = AhfinT b – T Ü η fη ƒ is the factor that is required for each case. Figure I.6shows the fin efficiency for several cases.Surface effectiveness K is defined as the actual heattransfer from a finned surface to that which would occurif the surface were isothermal at the base temperature.Taking advantage of fin efficiency, we can write(A – A f )h θ 0 + η f A f θ 0K = ——————————A h θ 0Equation I.6 reduces toA fK = 1 —— (1 – η f )A(I.6)which is a function only of geometry and single fin efficiency.To get the heat rate from a fin array, we writeQ array = Kh (T b – T ∞ ) Awhere A is the total area exposed.Transient Conduction. Heating and cooling problemsinvolve the solution of the time-dependent conductionequation. Most problems of industrial significanceoccur when a body at a known initial temperature issuddenly exposed to a fluid at a different temperature.The temperature behavior for such unsteady problemscan be characterized by two dimensionless quantities,the Biot number, Bi = hL/k, and the Fourier modulus,Fo = ατ/L 2 . The Biot number is a measure of the effectivenessof conduction within the body. The Fouriermodulus is simply a dimensionless time.If Bi is a small, say Bi ≤ 0.1, the body undergoingthe temperature change can be assumed to be at a uniformtemperature at any time. For this case,T – T fT i – T f= exp –hAρCV τFig. I.5 Fins of various shapes. (a) Rectangular, (b) Trapezoidal,(c) Arbitrary profile, (d ) Circumferential.where T ƒ and T i are the fluid temperature and initialbody temperature, respectively. The term (ρCV/hA) takeson the characteristics of a time constant.If Bi ≥ 0.1, the conduction equation must be solvedin terms of position and time. Heisler 4 solved the equationfor infinite slabs, infinite cylinders, and spheres. Forconvenience he plotted the results so that the temperatureat any point within the body and the amount ofheat transferred can be quickly found in terms of Bi andFo. Figures I.7 to I.10 show the Heisler charts for slabsand cylinders. These can be used if h and the propertiesof the material are constant.

826 ENERGY MANAGEMENT HANDBOOKI.3.2 Convection Heat TransferConvective heat transfer is considerably more complicatedthan conduction because motion of the mediumis involved. In contrast to conduction, where many geometricalconfigurations can be solved analytically, thereare only limited cases where theory alone will giveconvective heat-transfer relationships. Consequently,convection is largely what we call a semi-empirical science.That is, actual equations for heat transfer are basedstrongly on the results of experimentation.Convection Modes. Convection can be split intoseveral subcategories. For example, forced convectionrefers to the case where the velocity of the fluid is completelyindependent of the temperature of the fluid. Onthe other hand, natural (or free) convection occurs whenthe temperature field actually causes the fluid motionthrough buoyancy effects.We can further separate convection bygeometry into external and internal flows. Internalrefers to channel, duct, and pipe flow andexternal refers to unbounded fluid flow cases.There are other specialized forms of convection,for example the change-of-phase phenomena:boiling, condensation, melting, freezing, and soon. Change-of-phase heat transfer is difficult topredict analytically. Tongs 5 gives many of thecorrelations for boiling and two-phase flow.CμPrandtl number: Pr = ——kratio of momentum transport to heat-transport characteristicsfor a fluid: it is important in all convective cases,and is a material propertyg β(T – T ∞ )L 3Grashof number: Gr = ——————υ 2serves in natural convection the same role as Re inforced convection: that is, it controls the character ofthe flowhStanton number: St = ———ρ uC pDimensional Heat-Transfer Parameters.Because experimentation has been required todevelop appropriate correlations for convectiveheat transfer, the use of generalized dimensionlessquantities in these correlations is preferred.In this way, the applicability of experimentaldata covers a wider range of conditions and fluids.Some of these parameters, which we generallycall “numbers,” are given below:hLNusselt number: Nu = ——kwhere k is the fluid conductivity and L is measuredalong the appropriate boundary betweenliquid and solid; the Nu is a nondimensionalheat-transfer coefficient.LuReynolds number: Re = ——υdefined in Section I.4: it controls the characterof the flowFig. I.6 (a) Efficiencies of rectangular and triangular fins, (b) Efficienciesof circumferential fins of rectangular profile.

THERMAL SCIENCES REVIEW 827Fig. I.7 Midplane temperature for an infinite plate of thickness 2L. (From Ref. 4.)Fig. I.8 Axis temperature for an infinite cylinder of radius r o . (From Ref. 4.)also a nondimensional heat-transfer coefficient: it is veryuseful in pipe flow heat transfer.In general, we attempt to correlate data by usingrelationships between dimensionless numbers: for example,in many convection cases, we could write Nu =Nu(Re, Pr) as a functional relationship. Then it is possibleeither from analysis, experimentation, or both, towrite an equation that can be used for design calculations.These are generally called working formulas.Forced Convection Past Plane Surfaces. The averageheat-transfer coefficient for a plate of length L maybe calculated fromNu L = 0.664 (Re L ) 1/2 (Pr) 1/3if the flow is laminar (i.e., if Re L ≤ 4,000). For this casethe fluid properties should be evaluated at the meanfilm temperature T m , which is simply the arithmetic

828 ENERGY MANAGEMENT HANDBOOKFig. I.9 Temperature as a function of center temperaturein an infinite plate of thickness 2L. (From Ref. 4.)average of the fluid and the surface temperature.For turbulent flow, there are several acceptable correlations.Perhaps the most useful includes both laminarleading edge effects and turbulent effects. It isNu = 0.0036 (Pr) 1/3 [(Re L ) 0.8 – 18.700]where the transition Re is 4,000.Forced Convection Inside Cylindrical Pipes orTubes. This particular type of convective heat transferis of special engineering significance. Fluid flowsthrough pipes, tubes, and ducts are very prevalent, bothin laminar and turbulent flow situations. For example,most heat exchangers involve the cooling or heating offluids in tubes. Single pipes and/or tubes are also usedto transport hot or cold liquids in industrial processes.Most of the formulas listed here are for the 0.5 ≤ Pr ≤100 range.Laminar Flow. For the case where Re D < 2300,Nusselt showed that Nu D = 3.66 for long tubes at aconstant tube-wall temperature. For forced convectioncases (laminar and turbulent) the fluid properties areevaluated at the bulk temperature T b . This temperature,also called the mixing-cup temperature, is defined byFig. I.10 Temperature as a function of axis temperature inan infinite cylinder of radius r o . (From Ref. 4.)if the properties of the flow are constant.Sieder and Tate developed the following moreconvenient empirical formula for short tubes:Nu D =1.86 Re D 1/3 Pr 1/3 D L1/3 ÇÇs0.14The fluid properties are to be evaluated at T b except forthe quantity μ s , which is the dynamic viscosity evaluatedat the temperature of the wall.Turbulent Flow. McAdams suggests the empiricalrelationNu D = 0.023 (Pr D ) 0.8 (Pr) n(I.7)where n = 0.4 for heating and n = 0.3 for cooling. EquationI.7 applies as long as the difference between thepipe surface temperature and the bulk fluid temperatureis not greater than 10°F for liquids or 100°F for gases.For temperature differences greater then the limitsspecified for equation I.7 or for fluids more viscous thanwater, the following expression from Sieder and Tatewill give better results:NU D = 0.027 Pr D 0.8 Pr 1/3 ÇÇs0.14T b =R0R0uTr drur drNote that the McAdams equation requires only a knowledgeof the bulk temperature, whereas the Sieder-Tateexpression also requires the wall temperature. Manypeople prefer equation I.7 for that reason.

THERMAL SCIENCES REVIEW 829Nusselt found that short tubes could be representedby the expressionNu D = 0.036 Pe D 0.8 Pr 1/3 ÇÇs 0.14 DL1/18For noncircular ducts, the concept of equivalent diametercan be employed, so that all the correlations forcircular systems can be used.Forced Convection in Flow Normal to SingleTubes and Banks. This circumstance is encounteredfrequently, for example air flow over a tube or pipecarrying hot or cold fluid. Correlations of this phenomenonare called semi-empirical and take the form Nu D= C(Re D ) m . Hilpert, for example, recommends the valuesgiven in Table I.8. These values have been in use formany years and are considered accurate.Flows across arrays of tubes (tube banks) may beeven more prevalent than single tubes. Care must beexercised in selecting the appropriate expression for thetube bank. For example, a staggered array and an in-linearray could have considerably different heat-transfercharacteristics. Kays and London 6 have documentedmany of these cases for heat-exchanger applications. Fora general estimate of order-of-magnitude heat-transfercoefficients, Colburn’s equationcomplicated type of heat transfer. This is caused primarilyby the electromagnetic wave nature of thermalradiation. However, in certain applications, primarilyhigh-temperature, radiation is the dominant mode ofheat transfer. So it is imperative that a basic understandingof radiative heat transport be available. Heat transferin boiler and fired-heater enclosures is highly dependentupon the radiative characteristics of the surface and thehot combustion gases. It is known that for a body radiatingto its surroundings, the heat rate isQ = εσA T 4 – T s4where ε is the emissivity of the surface, σ is the Stefan-Boltzmann constant, σ = 0.1713 × 10 – 8 Btu/hr ft 2 • R 4 .Temperature must be in absolute units, R or K. If ε = 1for a surface, it is called a “blackbody,” a perfect emitterof thermal energy. Radiative properties of varioussurfaces are given in Appendix II. In many cases, theheat exchange between bodies when all the radiationemitted by one does not strike the other is of interest.In this case we employ a shape factor F ij to modify thebasic transport equation. For two blackbodies we wouldwriteQ 12 = F 12 σAT 1 4 – T 24Nu D = 0.33 (Re D ) 0.6 (Pr) 1/3is acceptable.Free Convection Around Plates and Cylinders.In free convection phenomena, the basic relationshipstake on the functional form Nu = ƒ(Gr, Pr). The Grashofnumber replaces the Reynolds number as the drivingfunction for flow.In all free convection correlations it is customary toevaluate the fluid properties at the mean film temperatureT m , except for the coefficient of volume expansionβ, which is normally evaluated at the temperature of theundisturbed fluid far removed from the surface—namely,T ƒ . Unless otherwise noted, this convention should beused in the application of all relations quoted here.Table I.9 gives the recommended constants and exponentsfor correlations of natural convection for verticalplates and horizontal cylinders of the form Nu = C • Ra m .The product Gr • Pr is called the Rayleigh number (Ra)and is clearly a dimensionless quantity associated withany specific free convective situation.I.3.3 Radiation Heat TransferRadiation heat transfer is the most mathematicallyTable I.8 Values of C and m for Hilpert’s EquationRange of N ReD C m1-4 0.891 0.3304-40 0.821 0.38540-4000 0.615 0.4664000-40,000 0.175 0.61840,000-250,000 0.0239 0.805Table I.9 Constants and Exponentsfor Natural Convection CorrelationsVertical Plate aHorizontal Cylinders bRa c m c m10 4 < Ra < 10 9 0.59 1/4 0.525 1/410 9 < Ra < 10 12 0.129 1/3 0.129 1/3a Nu and Ra based on vertical height L.bNu and Ra based on diameter D.

830 ENERGY MANAGEMENT HANDBOOKfor the heat transport from body 1 to body 2. FiguresI.11 to I.14 show the shape factors for some commonlyencountered cases. Note that the shape factor is a functionof geometry only.Gaseous radiation that occurs in luminous combustionzones is difficult to treat theoretically. It is toocomplex to be treated here and the interested reader isreferred to Siegel and Howell 7 for a detailed discussion.In words, this is simply a balance between mass enteringand leaving a control volume and the rate of massstorage. The ρ(υ•n) terms are integrated over the controlsurface, whereas the ρ dV term is dependent upon anintegration over the control volume.For a steady flow in a constant-area duct, the continuityequation simplifies tom = ρ f Α c u = constantI.4 FLUID MECHANICSIn industrial processes we deal with materials thatcan be made to flow in a conduit of some sort. The lawsthat govern the flow of materials form the science thatis called fluid mechanics. The behavior of the flowingfluid controls pressure drop (pumping power), mixingefficiency, and in some cases the efficiency of heat transfer.So it is an integral portion of an energy conservationprogram.I.4.1 Fluid DynamicsWhen a fluid is caused to flow, certain governinglaws must be used. For example, mass flows in and outof control volumes must always be balanced. In otherwords, conservation of mass must be satisfied.In its most basic form the continuity equation(conservation of mass) isThat is, the mass flow rate m is constant and is equal tothe product of the fluid density ρ ƒ , the duct cross sectionA c , and the average fluid velocity u.If the fluid is compressible and the flow is steady,one getsmρ f= constant = uΑ c uΑ c 2where 1 and 2 refer to different points in a variablearea duct.I.4.2 First Law—Fluid DynamicsThe first law of thermodynamics can be directlyapplied to fluid dynamical systems, such as duct flows.If there is no heat transfer or chemical reaction and if theinternal energy of the fluid stream remains unchanged,the first law isÌÌc.s.ρυ •n dA + ∂ ÌÌÌ [ ρ dV =0∂ tc.v.V i 2 _ V e22g c+ z i – z eg g + p i – p ec ρ + w p – w f =0(I.8)Fig. I.11 Radiation shape factor for perpendicular rectangles with a common edge.

THERMAL SCIENCES REVIEW 831Fig. I.12 Radiation shape factor for parallel, concentricdisks.Fig. I.14 Radiation shape factor for parallel, directlyopposed rectangles.Fig. I.15 The first law applied to adiabatic flow system.In the English system, horsepower ishp = m lb msec w p = ft • lb flb m×1 hp – sec500 ft – lb = mw p550Fig. I.13 Radiation shape factor for concentric cylindersof finite length.where the subscripts i and e refer to inlet and exit conditionsand w p and w ƒ are pump work and work requiredto overcome friction in the duct. Figure I.15 shows schematicallya system illustrating this equation.Any term in equation I.8 can be converted to a rateexpression by simply multiplying by , the mass flowrate. Take, for example, the pump horsepower,W energytime= mw pmasstimeenergymassReferring back to equation I.8, the most difficult term todetermine is usually the frictional work term w ƒ . This isa term that depends upon the fluid viscosity, the flowconditions, and the duct geometry. For simplicity, w ƒ isgenerally represented asp fw f = ——ρwhen ∆p ƒ is the frictional pressure drop in the duct.Further, we say thatp fρ = 2 fu2 Lg D cin a duct of length L and diameter D. The friction factorƒ is a convenient way to represent the differing influenceof laminar and turbulent flows on the friction pressuredrop.

832 ENERGY MANAGEMENT HANDBOOKThe character of the flow is determinedthrough the Reynolds number,Re = ρuD/μ, where μ is the viscosity ofthe fluid. This nondimensional groupingrepresents the ratio of dynamic toviscous forces acting on the fluid.Experiments have shown that if Re≤ 2300, the flow is laminar. For larger Rethe flow is turbulent. Figure I.16 showshow the friction factor depends uponthe Re of the flow. Note that for laminarflow the ƒ vs. Re curve is single-valuedand is simply equal to 16/Re. In theturbulent regime, the wall roughness ecan affect the friction factor because ofits effect on the velocity profile near theduct surface.If a duct is not circular, the equivalentdiameter D e can be used so that allthe relationships developed for circularsystems can still be used. D e is defined as4A cD e = ——PP is the “wetted” perimeter, that part of the flow crosssection that touches the duct surfaces. For a circularsystem D e = 4(πD 2 /4πD) = D, as it should. For an annularduct, we getD e = ÉD o 2 ⁄ 4–ÉD i 2 ⁄ 44ÉD o + ÉD i= É D o + D i D o + D iÉD o + ÉD i= D o + D iPressure Drop in Ducts. In practical applications,the essential need is to predict pressure drops in pipingand duct networks. The friction factor approach is adequatefor straight runs of constant area ducts. But valvesnozzles, elbows, and many other types of fittings are necessarilyincluded in a network. This can be accounted forby defining an equivalent length L e for the fitting. TableI.10 shows L e/ D values for many different fittings.Pressure Drop across Tube Banks. Another commonlyencountered application of fluid dynamics is thepressure drop caused by transverse flow across arraysof heat-transfer tubes. One technique to calculate thiseffect is to find the velocity head loss through the tubebank:N v = ƒNF dFig. I.16 Friction factors for straight pipes.where ƒ is the friction factor for the tubes (a functionof the Re), N the number of tube rows crossed by theflow, and F d is the “depth factor.” Figures I.17 and I.18show the ƒ factor and F d relationship that can be used inpressure-drop calculations. If the fluid is air, the pressuredrop can be calculated by the equationp = N 30 BT G21.73 × 10 5 10 3where B is the atmospheric pressure (in. Hg), T is temperature(°R), and G is the mass velocity (lbm/ft 2 hr).Bernoulli’s Equation. There are some cases where theequationp u 2— + — + gz = constantρ 2which is called Bernoulli’s equation, is useful. Strictlyspeaking, this equation applies for inviscid, incompressible,steady flow along a streamline. However, even inpipe flow where the flow is viscous, the equation canbe applied because of the confined nature of the flow.That is, the flow is forced to behave in a streamlinedmanner. Note that the first law equation (I.8) yieldsBernoulli’s equation if the friction drop exactly equalsthe pump work.I.4.3 Fluid-Handling EquipmentFor industrial processes, another prime applicationof fluid dynamics lies in fluid-handling equipment.

THERMAL SCIENCES REVIEW 833Table I.10 L e /D for Screwed Fittings, TurbulentFlow Only a—————————————————————————FittingL e/ D—————————————————————————45° elbow 1590° elbow, standard radius 3190° elbow, medium radius 2690° elbow, long sweep 2090° square elbow 65180° close return bend 75Swing check valve, open 77Tee (as el, entering run) 65Tee (as el, entering branch) 90Couplings, unionsNegligibleGate valve, open 7Gate valve, 1/4 closed 40Gate valve, 1/2 closed 190Gate valve, 3/4 closed 840Globe valve, open 340Angle valve, open 170—————————————————————————a Calculated from Crane Co. Tech. Paper 409, May 1942.Fig. I.17 Depth factor for number of tube rows crossedin convection banks.Pumps, compressors, fans, and blowers are extensivelyused to move gases and liquids through the processnetwork and over heat-exchanger surfaces. The generalconstraint in equipment selection is a matching of fluidhandler capacity to pressure drop in the circuit connectedto the fluid handler.Pumps are used to transport liquids, whereascompressors, fans, and blowers apply to gases. Thereare features of performance common to all of them. Forpurposes of illustration, a centrifugal pump will be usedto discuss performance characteristics.Centrifugal Machines. Centrifugal machines operateon the principle of centrifugal acceleration of afluid element in a rotating impeller/housing system toachieve a pressure gain and circulation.The characteristics that are important are flow rate(capacity), head, efficiency, and durability. Q ƒ (capacity),h p (head), and η p (efficiency) are related quantities,dependent basically on the fluid behavior in the pumpand the flow circuit. Durability is related to the wear,corrosion, and other factors that bear on a pump’s reliabilityand lifetime.Figure I.19 shows the relation between flow rateand related characteristics for a centrifugal pump at constantspeed. Graphs of this type are called performancecurves; fhp and bhp are fluid and brake horsepower, respectively.The primary design constraint is a matchingFig. I.18 Friction factor ƒ as affected by Reynolds numberfor various in-line tube patterns, crossflow gas or air, d o ,tube diameter; l ⊥ , gap distance perpendicular to the flow;l || , gap distance parallel to the flow.

834 ENERGY MANAGEMENT HANDBOOKof flow rate to head. Note that as the flow-rate requirementis increased, the allowable head must be reducedif other pump parameters are unchanged.Analysis and experience has shown that there arescaling laws for centrifugal pump performance that givethe trends for a change in certain performance parameters.Basically, they are:Efficiency:η p =ƒ 1Q fD 3 nDimensionless head:h p gD 2 n 2 = f 2Q fD 3 nFig. I.19 Performance curve for a centrifugal pump.Dimensionless brake horsepower:bhp • gγD 2 n 3 = f 3Q fD 3 nwhere D is the impeller diameter, n is the rotary impellerspeed, g is gravity, and γ is the specific weight offluid.The basic relationships yield specific proportionalitiessuch as Q ƒ ∝ n (rpm), h p ∝ n 2 , fhp ∝ n 3 , Q f ∝ 1 D 2,Q f ρgh pfhp = ————550g cQ f ρgh p 550g c fhpη p p = —————— = ——bhp bhpsystem efficiency η s = η p × η m(motor efficiency)h p ∝ 1 D 4, and fh p ∝ 1 D 4.For pumps, density variations are generally negligiblesince liquids are incompressible. But for gas-handlingequipment, density changes are very important.The scaling laws will give the following rules for changingdensity:h p ∝ ρfh p ∝ ρnfhp ∝ρ–1/2!Q fnQ fh p∝ 1 ρ1fhp ∝ —— ρ 2(Q f , n constant)(h p constant(m constant)For centrifugal pumps, the following equations hold:It is important to select the motor and pump so thatat nominal operating conditions, the pump and motoroperate at near their maximum efficiency.For systems where two or more pumps are present,the following rules are helpful. To analyze pumpsin parallel, add capacities at the same head. For pumpsin series, simply add heads at the same capacity.There is one notable difference between blowersand pump performance. This is shown in Figure I.20.Note that the bhp continues to increase as permissiblehead goes to zero, in contrast to the pump curve whenbhp approaches zero. This is because the kinetic energyimparted to the fluid at high flow rates is quite significantfor blowers.Manufactures of fluid-handling equipment provideexcellent performance data for all types of equipment.Anyone considering replacement or a new installationshould take full advantage of these data.Fluid-handling equipment that operates on a principleother than centrifugal does not follow the centrifugalscaling laws. Evans 8 gives a thorough treatment ofmost types of equipment that would be encountered inindustrial application.

THERMAL SCIENCES REVIEW 835k specific heat ratio: C p /C v ƒg difference in property for saturated vaporK.E. kinetic energylb f pound forcelb m pound masslb mol pound molem massm mass rate of flowM molecular weightn number of molesn polytropic exponentP pressureP i partial pressure of component i in a mixtureP.E. potential energyFig. I.20 Variation of head and bhp with flow rate for a P r relative pressure as used in gas tablestypical blower at constant speed.q, Q heat transfer per unit mass and total heattransferQ rate of heat transferReferencesQH, QL heat transfer from high- and low-temperaturebodiesR gas constant1. G.J. Van Wylen and R.E. Sonntag, Fundamentals of Classical R universal gas constantThermodynamics, 2nd ed., Wiley, New York, 1973.s, S specific entropy and total entropy2. A.S. Chapman, Heat Transfer, 3rd ed., Macmillan, New York, t time1974.T temperature3. J.P. Holman, Heat Transfer, 4th ed., McGraw-Hill, New York,u, U specific internal energy and total internal1976.energy4. M.P. Heisler, Trans. ASME, Vol. 69 (1947), p. 227.v, V specific volume and total volume5. L.S. Tong, Boiling Heat Transfer and Two-Phase Flow, Wiley,New York, 1965.V velocity6. W.M. Kays and A.L. London, Compact Heat Exchangers, 2nd V r relative velocityed., McGraw-Hill, New York, 1963.w, W work per unit mass and total work7. R. Siegel and J.R. Howell, Thermal Radiation Heat Transfer, W rate of work, or powerMcGraw-Hill, New York, 1972.8. FRANK L. Evans, JR., Equipment Design Handbook for Refineriesw rev reversible work between two states assumingheat transfer with surroundingsand Chemical Plants, Vols. 1 and 2, Gulf Publishing, x mass fractionHouston, Tex., 1974.Z elevationZ compressibility factorSYMBOLSGreek Lettersβ coefficient of performance for a refrigeratorThermodynamicsβ' coefficient of performance for a heat pumpAF air/fuel ratioη efficiencyC p constant-pressure specific heatρ densityC v constant-volume specific heatφ relative humidityCp 0 zero-pressure constant-pressure specific heat ω humidity ratio or specific humidityC v0 zero-pressure constant-volume specufic heate, E specific energy and total energySubscriptsg acceleration due to gravityc property at the critical pointg, G specific Gibbs function and total Gibbs functionc.v. control volumee state of a substance leaving a control vol-g e a constant that relates force, mass, length, andumetimeƒ formationh, H specific enthalpy and total enthalpyƒ property of saturated liquid

836 ENERGY MANAGEMENT HANDBOOKgrsand saturated liquidproperty of saturated vaporreduced propertyisentropic processSuperscripts- bar over symbol denotes property on a molalbasis (over V, H, S, U, A, G, the bar denotespartial molal property)° property at standard-state condition* ideal gasL liquid phaseS solid phaseV vapor phaseHeat Transfer—Fluid FlowA surface areaA m profile area for a finBi Biot number, (hL/k)c p specific heat at constant pressurec specific heatD diameterD e hydraulic diameterF i-j shape factor of area i with respect to area jƒ friction factorGr Grashof number, g βÄTL 3 c/υ 2g acceleration due to gravityg c gravitational constanth convective heat-transfer coefficientk thermal conductivitym massm mass rate of flowN number of rowsNu Nusselt number, hL/kPr Prandtl number, μC p /kp pressureQ volumetric flow rateQ rate of heat flowRa Rayleigh number, g βÄTL 3 c/υ∝Re Reynolds number, ρu av L c /μr radiusSt Stanton number, h/Cp ρu ∝T temperatureU overall heat-transfer coefficientu velocityux free-stream velocityV volumeV velocityW rate of work doneGreek Symbolsα thermal diffusivityβ coefficient of thermal expansion∆ difference, changeε surface emissivityη f fin effectivenessμ viscosityv kinematic viscosityρ densityσ Stefan-Boltzmann constantτ timeSubscriptsb bulk conditionscr critical conditionc convectioncond conductionconv convectione entrance, effectiveƒ fin, fluidi inlet conditionso exterior condition0 centerline conditions in a tube at r = 0o outlet conditionp pipe, pumps surface condition∝ free-stream condition

APPENDIX IICONVERSION FACTORS AND PROPERTY TABLESCompiled byL.C. WITTEProfessor of Mechanical EngineeringUniversity of HoustonHouston, TexasTable II.1 Conversion FactorsTo Obtain: Multiply: By:Acres Sq miles 640.0Atmospheres Cm of Hg @ 0 deg C 0.013158Atmospheres Ft of H 2O @ 39.2 F. 0.029499Atmospheres Grams/sq cm 0.00096784Atmospheres In. Hg @ 32 F 0.033421Atmospheres In. H 2O @ 39.2 F 0.0024583Atmospheres Pounds/sq ft 0.00047254Atmospheres Pounds/sq in. 0.068046Btu Ft-lb 0.0012854Btu Hp-hr 2545.1Btu Kg-cal. 3.9685Btu kW-hr 3413Btu Watt-hr 3.4130Btu/(cu ft) (hr) kW/liter 96,650.6Btu/hr Mech. hp 2545.1Btu/hr kW 3413Btu/hr Tons of refrigeration 12,000Btu/hr Watts 3.4127Btu/kW hr Kg cal/kW hr 3.9685Btu/(hr) (ft) (deg F) Cal/(sec) (cm) (deg C) 241.90Btu/(hr) (ft) (deg F) Joules/(sec) (cm) (deg C) 57.803Btu/(hr) (ft) (deg F) Watts/(cm) (deg C) 57.803Btu/(hr) (sq ft) Cal/(sec) (sq cm) 13,273.0Btu/min Ft-lb/min 0.0012854Btu/min Mech. hp 42.418Btu/min kW 56.896Btu/lb Cal/gram 1.8Btu/lb Kg cal/kg 1.8Btu/(lb) (deg F) Cal/(gram) (deg C) 1.0Btu/(lb) (deg F) Joules/(gram) (deg C) 0.23889Btu/sec Mech. hp 0.70696Btu/sec Mech. hp (metric) 0.6971Btu/sec Kg-cal/hr 0.0011024Btu/sec kW 0.94827Btu/sq ft Kg-cal/sq meter 0.36867837

838 ENERGY MANAGEMENT HANDBOOKTable II.1 ContinuedTo Obtain: Multiply: By:Calories Ft-lb 0.32389Calories Joules 0.23889Calories Watt-hr 860.01Cal/(cu cm) (sec) kW/liter 0.23888Cal/gram Btu/lb 0.55556Cal/(gram) (deg C) Btu/(lb) (deg F) 1.0Cal/(sec) (cm) (deg C) Btu/(hr) (ft) (deg F) 0.0041336Cal/(sec) (sq cm) Btu/(hr) (sq ft) 0.000075341Cal/(sec) (sq cm) (deg C) Btu/(hr) (sq ft) (deg F) 0.0001355Centimeters Inches 2.540Centimeters Microns 0.0001Centimeters Mils 0.002540Cm of Hg @ 0 deg C Atmospheres 76.0Cm of Hg @ 0 deg C Ft of H 2O @ 39.2 F 2.242Cm of Hg @ 0 deg C Grams/sq cm 0.07356Cm of Hg @ 0 deg C In. of H 2O @ 4 C 0.1868Cm of Hg @ 0 deg C Lb/sq in. 5.1715Cm of Hg @ 0 deg C Lb/sq ft 0.035913Cm/deg C In./deg F 4.5720Cm/sec Ft/min 0.508Cm/sec Ft/sec 30.48Cm/(sec) (sec) Gravity 980.665Cm of H 2O @39.2 F Atmospheres 1033.24Cm of H 2O @39.2 F Lb/sq in. 70.31Centipoises Centistokes DensityCentistokes Centipoises l/densityCu cm Cu ft 28,317Cu cm Cu in. 16.387Cu cm Gal. (USA, liq.) 3785.43Cu cm Liters 1000 03Cu cm Ounces (USA, liq.) 29.573730Cu cm Quarts (USA, liq.) 946.358Cu cm/sec Cu ft/min 472.0Cu ft Cords (wood) 128.0Cu ft Cu meters 35.314Cu ft Cu yards 27.0Cu ft Gal. (USA, liq.) 0.13368Cu ft Liters 0.03532Cu ft/min Cu meters/sec 2118.9Cu ft/min Gal. (USA, liq./sec) 8.0192Cu ft/lb Cu meters/kg 16.02Cu ft/lb Liters/kg 0.01602Cu ft/sec Cu meters/min 0.5886Cu ft/sec Gal. (USA, liq.)/min 0.0022280Cu ft/sec Liters/min 0.0005886Cu in. Cu centimeters 0.061023Cu in. Gal. (USA, liq.) 231.0Cu in. Liters 61.03Cu in. Ounces (USA. liq.) 1.805

CONVERSION FACTORS AND PROPERTY TABLES 839Table II.1 ContinuedTo Obtain: Multiply: By:Cu meters Cu ft 0.028317Cu meters Cu yards 0.7646Cu meters Gal. (USA. liq.) 0.0037854Cu meters Liters 0.001000028Cu meters/hr Gal./min 0.22712Cu meters/kg Cu ft/lb 0.062428Cu meters/min Cu ft/min 0.02832Cu meters/min Gal./sec 0.22712Cu meters/sec Gal./min 0.000063088Cu yards Cu meters 1.3079Dynes Grams 980.66Dynes Pounds (avoir.) 444820.0Dyne-centimeters Ft-lb 13 ,558,000Dynes/sq cm Lb/sq in. 68947Ergs Joules 10,000,000Feet Meters 3.281Ft of H 2O 39.2 F@Atmospheres 33.899Ft of H 2O @ 39.2 F Cm of Hg @ 0 deg C 0.44604Ft of H 2O @ 39.2 F In. of Hg @ 32 deg F 1.1330Ft of H 2O @ 39.2 F Lb/sq ft 0.016018Ft of H 2O @ 39.2 F Lb/sq in. 2.3066Ft/min Cm/sec 1.9685Ft/min Miles (USA. statute)/hr 88.0Ft/sec Knots 1.6889Ft/sec Meters/sec 3.2808Ft/sec Miles (USA, statute)/hr 1.4667Ft/(sec) (sec) Gravity (sea level) 32.174Ft/(sec) (sec) Meters/(sec) (sec) 3.2808Ft-lb Btu 778.0Ft-lb Joules 0.73756Ft-lb Kg-calories 3087.4Ft-lb kW-hr 2,655,200Ft-lb Mech. hp-hr 1,980,000Ft-lb/min Btu/min 778.0Ft-lb/min Kg cal/min 3087.4Ft-lb/min kW 44,254.0Ft-lb/min Mech. hp 33,000Ft-lb/sec Btu/min 12.96Ft-lb/sec kW 737.56Ft-lb/sec Mech. hp 550.0Gal. (Imperial, liq.) Gal. (USA. Liq.) 0.83268Gal. (USA, liq.) Barrels (petroleum, USA) 42Gal. (USA. liq.) Cu ft 7.4805Gal. (USA. liq.) Cu meters 264.173Gal. (USA, liq.) Cu yards 202.2Gal. (USA. liq.) Gal. (Imperial, liq.) 1.2010Gal. (USA. liq.) Liters 0.2642Gal. (USA. liq.)/min Cu ft/sec 448.83Gal. (USA, liq.)/min Cu meters/hr 4.4029

840 ENERGY MANAGEMENT HANDBOOKTable II.1 ContinuedTo Obtain: Multiply: By:Gal. (USA. liq.)/sec Cu ft/min 0.12468Gal. (USA. liq.)/sec Liters/min 0.0044028Grains Grams 15.432Grains Ounces (avoir.) 437.5Grains Pounds (avoir.) 7000Grains/gal. (USA. liq.) Parts/million 0.0584Grams Grains 0.0648Grams Ounces (avoir.) 28.350Grams Pounds (avoir.) 453.5924Grams/cm Pounds/in. 178.579Grams/(cm) (sec) Centipoises 0.01Grams/cu cm Lb/cu ft 0 .016018Grams/cu cm Lb/cu in. 27.680Grams/cu cm Lb/gal. 0.119826Gravity (at sea level) Ft/(sec) (sec) 0.03108Inches Centimeters 0.3937Inches Microns 0.00003937Inches of Hg @ 32 F Atmospheres 29.921Inches of Hg @ 32 F Ft of H 2O @ 39.2 F 0.88265Inches of Hg @ 32 F Lb/sq in. 2.0360Inches of Hg @ 32 F In. of H 2O @ 4 C 0.07355Inches of H 2O@ 4 C In. of Hg @ 32 F 13.60Inches of H2O @ 39.2 F Lb/sq in. 27.673Inches/deg F Cm/deg C 0.21872Joules Btu 1054.8Joules Calories 4.186Joules Ft-lb 1.35582Joules Kg-meters 9.807Joules kW-hr 3,600,000Joules Mech. hp-hr 2,684,500Kg Pounds (avoir.) 0.45359Kg-cal Btu 0.2520Kg-cal Ft-lb 0.00032389Kg-cal Joules 0.0002389Kg-cal kW-hr 860.01Kg-cal Mech. hp-hr 641.3Kg-cal/kg Btu/lb 0.5556Kg-cal/kW hr Btu/kW hr 0.2520Kg-cal/min Ft-lb/min 0.0003239Kg-cal/min kW 14,33Kg-cal/min Mech. hp 10.70Kg-cal/sq meter Btu/sq ft 2.712Kg/cu meter Lb/cu ft 16.018Kg/(hr) (meter) Centipoises 3.60Kg/liter Lb/gal. (USA, liq.) 0.11983Kg/meter Lbm 1.488Kg/sq cm Atmospheres 1.0332Kg sq cm Lb/sq in . 0.0703Kg/sq meter Lb/sq ft 4.8824

CONVERSION FACTORS AND PROPERTY TABLES 841Table II.1 ContinuedTo Obtain: Multiply: By:Kg/sq meter Lb/sq in. 703.07Km Miles (USA, statute) 1.6093kW Btu/min 0.01758kW Ft-lb/min 0.00002259kW Ft-lb/sec 0.00135582kW Kg-cal/hr 0.0011628kW Kg-cal/min 0.069767kW Mech. hp 0.7457kW-hr Btu 0.000293kW-hr Ft-lb 0.0000003766kW-hr Kg-cal 0.0011628kW-hr Mech. hp-hr 0.7457Knots Ft/sec 0.5921Knots Miles/hr 0.8684Liters Cu ft 28 . 316Liters Cu in. 0.01639Liters Cu centimeters 999.973Liters Gal. (Imperial. liq.) 4.546Liters Gal. (USA, liq.) 3.78533Liters/kg Cu ft/lb 62.42621Liters/min Cu ft/sec 1699.3Liters/min Gal. (USA. liq.)/min 3.785Liters/sec Cu ft/min 0.47193Liters/sec Gal./min 0.063088Mech. hp Btu/hr 0.0003929Mech. hp Btu/min 0.023575Mech. hp Ft-lb/sec 0.0018182Mech. hp Kg-cal/min 0.093557Mech. hp kW 1.3410Mech. hp-hr Btu 0.00039292Mech. hp-hr Ft-lb 0.00000050505Mech. hp-hr Kg-calories 0.0015593Mech. hp-hr kW-hr 1.3410Meters Feet 0.3048Meters Inches 0.0254Meters Miles (Int., nautical) 1852.0Meters Miles (USA, statute) 1609.344Meters/min Ft/min 0.3048Meters/min Miles (USA. statute)/hr 26.82Meters/sec Ft/sec 0.3048Meters/sec Km/hr 0.2778Meters/sec Knots 0.5148Meters/sec Miles (USA, statute)/hr 0.44704Meters/(sec) (sec) Ft/(sec) (sec) 0.3048Microns Inches 25,400Microns Mils 25.4Miles (Int., nautical) Km 0.54Miles (Int., nautical) Miles (USA, statute) 0.8690Miles (Int., nautical)/hr Knots 1.0

842 ENERGY MANAGEMENT HANDBOOKTable II.1 ContinuedTo Obtain: Multiply: By:Miles (USA, statute) Km 0.6214Miles (USA, statute) Meters 0.0006214Miles (USA, statute) Miles (Int., nautical) 1.151Miles (USA, statute)/hr Knots 1.151Miles (USA, statute)/hr Ft/min 0.011364Miles (USA, statute)/hr Ft/sec 0.68182Miles (USA, statute)/hr Meters/min 0.03728Miles (USA, statute)/hr Meters/sec 2.2369Milliliters/gram Cu ft/lb 62.42621Millimeters Microns 0.001Mils Centimeters 393.7Mils Inches 1000Mils Microns 0.03937Minutes Radians 3437.75Ounces (avoir. ) Grains (avoir. ) 0.0022857Ounces (avoir.) Grams 0.035274Ounces (USA, liq.) Gal. (USA, liq.) 128.0Parts/million Gr/gal. (USA, liq.) 17.118Percent grade Ft/100 ft 1.0Pounds (avoir.) Grains 0.0001429Pounds (avoir.) Grams 0.0022046Pounds (avoir.) Kg 2.2046Pounds (avoir.) Tons, long 2240Pounds (avoir.) Tons, metric 2204.6Pounds (avoir.) Tons, short 2000Pounds/cu ft Grams/cu cm 62.428Pounds/cu ft Kg/cu meter 0.062428Pounds/cu ft Pounds/gal. 7.48Pounds/cu in . Grams/cu cm 0.036127Pounds/ft Kg/meter 0.67197Pounds/hr Kg/min 132.28Pounds/(hr) (ft) Centipoises 2.42Pounds/inch Grams/cm 0.0056Pounds/(sec) (ft) Centipoises 0.000672Pounds/sq inch Atmospheres 14.696Pounds/sq inch Cm of Hg @ 0 deg C 0.19337Pounds/sq inch Ft of H 2O @ 39.2 F 0.43352Pounds/sq inch In. Hg @ l 32 F 0.491Pounds/sq inch In. H 2O @ 39.2 F 0.0361Pounds/sq inch Kg/sq cm 14 . 223Pounds/sq inch Kg/sq meter 0.0014223Pounds/gal. (USA, liq.) Kg/liter 8.3452Pounds/gal. (USA, liq.) Pounds/cu ft 0.1337Pounds/gal. (USA, liq.) Pounds/cu inch 231Quarts (USA, liq.) Cu cm 0.0010567Quarts (USA, liq.) Cu in. 0.01732Quarts (USA, liq.) Liters 1.057Sq centimeters Sq ft 929.0Sq centimeters Sq inches 6.4516

CONVERSION FACTORS AND PROPERTY TABLES 843Table II.1 ContinuedTo Obtain: Multiply: By:Sq ft Acres 43,560Sq ft Sq meters 10.764Sq inches Sq centimeters 0.155Sq meters Acres 4046.9Sq meters Sq ft 0.0929Sq mlles (USA. statute) Acres 0.001562Sq mils Sq cm 155.000Sq mils Sq inches 1,000.000Tons (metric ) Tons (short) 0.9072Tons (short) Tons (metric) 1.1023Watts Btu/sec 1054.8Yards Meters 1.0936


CONVERSION FACTORS AND PROPERTY TABLES 845



848 ENERGY MANAGEMENT HANDBOOKTable 11.2-1 Continued————————————————————————————————————————————————————Abs. Specific Volume Enthalpy EntropyPress. Temp. Sat. Sat. Sat. Sat. Sat. Temp.p t Sat. Liquid Evap Vapor Liquid Evap Vapor Liquid Evap Vapor t(psi) (°F) v fv fxv xh fh fxh xs fs fxs f(°F)————————————————————————————————————————————————————









CONVERSION FACTORS AND PROPERTY TABLES 857Mollier Diagram for SteamSource: Modified and greatly reduced from J.H. Keenan and F.G. Keyes, Thermodynamic Properties of Steam, JohnWiley & Sons Inc., New York, 1936; reproduced by permission of the publishers.

















874 ENERGY MANAGEMENT HANDBOOKTable II.5-1, Continued——————————————————————————————————————————————————————Properties at 68°Fk(Btu/hr-ft-.F)——————————————————————————————————————————————————————ρ C Pk α –148°F 32°F 212°F 392°F 572°F 752°F 1112°F 1472°F 1832°F 2192°FMetal (lb m/ft 3 ) (Btu/lb m•°F) (Btu/hr • ft • °F) (ft 2 /hr) –100°C 0°C 100°C 200°C 300°C 400°C 600°C 800°C 1000°C 1200°C——————————————————————————————————————————————————————CopperPure 559 0.0915 223 4.353 235 223 219 216 — 210 204Aluminum bronze: 95 541 0.098 48 0.903Cu, 5 AlBronze: 75 Cu. 25 Sn 541 0.082 15 0.333Red brass: 85 Cu. 9 Sn. 6 544 0.092 35 0.699 — 34 41ZnBrass: 70 Cu. 30 Zn 532 0.092 64 1.322 51 — 74 83 85 85German silver 62 Cu, 15 538 0.094 14.4 0.290 11.1 — 18 23 26 28Ni. 22 ZnConstantan: 60 Cu, 40 Ni 557 0.098 13.1 0.237 12 — 12.8 15MagnesiumPure 109 0.242 99 3.762 103 99 97 94 91Mg-Al (electrolytic) 6-8% 113 0.24 38 1.397 — 30 36 43 48Al, 1-2% ZnMg-Mn: 2% Mn 111 0.24 66 2.473 54 64 72 75Molybdenum 638 0.060 79 2.074 80 79 79NickelPure (99.9%) 556 0.1065 52 0.882 60 54 48 42 37 34Impure (99.2%) 556 0.106 40 0.677 — 40 37 34 32 30 32 36 39 40Ni-Cr: 90 Ni. 10 Cr 541 0.106 10 0.172 — 9.9 10.9 12.1 13.2 14.280 Ni, 20 Cr 519 0.106 7.3 0.129 — 7.1 8.0 9.0 9.9 10.9 13.0SilverPurest 657 0.0559 242 6.601 242 241 240 238Pure (99.9%) 657 0.0559 235 6.418 242 237 240 216 209 208Tungsten 1208 0.0321 94 2.430 — 96 87 82 77 73. 65 44Zinc. Pure 446 0.0918 64.8 1.591 66 65 63 61 58 54Tin. pure 456 0.0541 37 1.505 43 38.1 34 33——————————————————————————————————————————————————————Source: From E.R.G. Eckert and R.M. Drake, Heat and Mass Transfer-, copyright 1959 McGraw-Hill; used with the permission of McGraw-Hill Book Company.Table II.5-2 Thermal Properties of Some Nonmetals——————————————————————————————————————————————————————C Pρ t k αSubstance (Btu/lb m• °F) lb m/°F 3 (°F) (Btu/hr • ft 2 • °F) (ft 2 /hr)——————————————————————————————————————————————————————StructuralAsphalt 68 0.43 aBakelite 0.38 b 79.5 b 68 0.134 b 0.0044BricksCommon 0.20 d 100 d 68 0.40 a 0.02Face 128 d 68 0.76 aCarborundum brick 1110 10.7 a2550 6.4 a392 1.34 a 0.036Chrome brick 0.20 d 188 d 1022 1.43 a 0.0381652 1.15 a 0.031Diatomaceous earth 400 0.14 a(fired) 1600 0.18 a932 0.60 a 0.020Fire clay brick (burnt 0.23 d 128 d 1472 0.62 a 0.0212426 F) 2012 0.63 a 0.021Fire clay brick (burnt 912 0.74 a 0.0222642 F) 0.23 d 145 d 1472 0.79 a 0.0242012 0.81 a 0.024392 0.58 a 0.015Fire clay brick (Missouri) 0.23 d 165 f 1112 0.85 a 0.0222552 1.02 a 0.027400 2.2 aMagnesite 0.27 d 1200 1.6 a2200 1.1 aCement, portland 94 0.17 aCement, mortar 75 0.67 aConcrete 0.21 b 119-144 b 68 0.47-0.81 b 0.019-0.027Concrete, cinder 75 0.44 a——————————————————————————————————————————————————————













APPENDIX IIIREVIEW OF ELECTRICAL SCIENCERUSSELL L. HEISERMAN, Ed.D.School of TechnologyOklahoma State UniversityStillwater, OklahomaIII. 1 INTRODUCTIONThis brief review of electrical science is intended forthose readers who may use electrical principles only onoccasion and is intended to be supportive of the materialfound in those chapters of the handbook based onelectrical science. The review consists of selected topicsin basic ac circuit theory presented at a nominal analyticallevel. Much of the material deals with power in accircuits and principles of power-factor improvement.III.2 REVIEW OF VECTOR ALGEBRAVector algebra is the mathematics most appropriatefor ac circuit problems. Most often electric quantities,voltage and current, are not in phase in ac circuits, sophase relationships as well as magnitude have to beconsidered. This brief review will cover the basic idea ofa vector quantity and then refresh the process of adding,subtracting, multiplying, and dividing vectors.Fig. III.1 The generalized vector Ae jθ shown in the complexplane. If j is positive, it is referenced to the positivereal axis with a counterclockwise displacement. If j isnegative, it is referenced to the positive real axis with aclockwise displacement.III.2.1 ReviewA vector is a quantity having both direction andmagnitude. Familiar vector quantities are velocity andforce. Other familiar quantities, such as speed, volume,area, and mass, have magnitude only.A vector quantity is expressed as having bothmagnitude and direction, such asAe ± j θwhere A is the magnitude and e ± j θ expresses the directionin the complex plane (Figure III.1).The important feature of this vector notation is tonote that the angle of displacement is in fact an exponent.This feature is significant, since it will allow theuse of the law of exponents when multiplying, dividing,or raising to a power.887Fig. III.2 The vector A ∠θ shown in the polar coordinatesystem. A vector expressed as A ∠θ is said to be in polarform.Common practice has created a shorthand for expressingvectors. This method is quicker to write andfor many, more clearly expresses the idea of a vector:A Â θ

888 ENERGY MANAGEMENT HANDBOOKThis shorthand is read as a vector magnitude A operatingor pointing in the direction θ. It is termed the polarrepresentation of a vector as shown in Figure III.2.Now the function e j θ may be expressed or resolvedinto its horizontal and vertical components inthe complex plane:e j θ = cos θ + j sin θThe vector has been resolved and expressed in rectangularform. Using the shorthand notationA ∠θ = A cos θ + jA sin θwhere A cos θ is the vector projection on the real axisand jA sin θ is the vector projection on the imaginaryaxis, as shown in Figure III.3.Both rectangular and polar expressions of a vectorquantity are useful when performing mathematicaloperations.20∠30° = 20 cos 30° + j20 sin 30°= 17.3 + j1025∠–45° = 25 cos 45° – j25 sin 45°= 17.7 – j17.7A calculator is a handy tool for resolving vectors. Manycalculators have automatic programs for converting vectorsfrom one form to another.Now adding we obtain17.3 + j10(+) 17.7 – j17.735.0 – j7.7By inspection, this vector is seen to be slightlygreater in magnitude than 35.0 and at a small anglebelow the positive real axis. Again using a calculator toexpress the vector in polar form: 35.8∠–12.4°, an answerin agreement with what was anticipated. Figure III.4shows roughly the same result using a graphical technique.Subtraction is accomplished in much the sameway. Suppose that the vector 25∠–45° is to be subtractedfrom the vector 20∠30°.20∠30° = 17.3 + j1025∠–45° = 17.7 - j17.7To subtract17.3 + j1017.7 – j17.7first change sign of the subtrahend and then add:Fig. III.3 The vector A ∠θ shown together with itsrectangular components.III.2.2 Addition and Subtraction of VectorsWhen adding or subtracting vectors, it is most convenientto use the rectangular form. This is best demonstratedthrough an example. Suppose that we have twovectors, 20∠30° and 25∠–45°, and these vectors are to beadded. The quickest way to accomplish this is to resolveeach vector into its rectangular components, add the realcomponents, then add the imaginary components, and,if needed, express the results in polar form:17.3 + j10– 17.7 + j17.7– 0.4 + j27.7The effect of changing the sign of the subtrahendis to push the vector back through the origin. as shownin Figure III.5.The resulting vector appears to be about 28 unitslong and barely in the second quadrant. The calculatorgives 27.7∠90.8°.III.2.3 Multiplication and Division of VectorsVectors are expressed in polar form for multiplicationand division. The magnitudes are multipliedor divided and the angles follow the rules governing

REVIEW OF ELECTRICAL SCIENCE 889or consider(20∠30) 3 = 8000∠90°(20∠30°) l/2 = 4.47/15°III.2.4 SummaryVector manipulation is straightforward and easyto do. This presentation is intended to refresh thosetechniques most commonly used by those working ata practical level with ac electrical circuits. It has beenthe author’s intent to exclude material on dot and crossproducts in favor of techniques that tend to allow theuser more of a feeling for what is going on.Fig. III.4 Use of the graphical parallelogram method foradding two vectors. The result or sum is the diagonaloriginating at the origin of the coordinate system.Fig. III.5 Graphical solution to subtraction of vectors.exponents, added when multiplying, subtracted whendividing. Consider20∠30° × 25∠–45° = 500∠–15°The magnitudes are multiplied and the angles areadded. Consider20 ∠30°——— = 0.8 ∠75°25∠–45°The magnitudes are divided and the angle of the divisoris subtracted from the angle of dividend.Raising to powers is a special case of multiplication.The magnitude is raised to the power and the angleis multiplied by the power. ConsiderIII.3 RESISTANCE, INDUCTANCE,AND CAPACITANCEThe three types of electric circuit elements havingdistinct characteristics are resistance, inductance, andcapacitance. This brief review will focus on the characteristicsof these circuit elements in ac circuits to supportlater discussions on circuit impedance and power-factorimprovementprinciples.III.3. 1 ResistanceResistance R in an ac circuit is the name given tocircuit elements that consume real power in the formof heat, light. mechanical work, and so on. Resistanceis a physical property of the wire used in a distributionsystem that results in power loss commonly called I 2 Rloss. Resistance can be thought of as a name given thatportion of a circuit load that performs real work, thatis, the portion of the power fed to a motor that resultsin measurable mechanical work being accomplished.If resistance is the only circuit element in an accircuit, the physical properties of that circuit are easilysummarized, as shown in Figure III.6. The importantproperty is that the voltage and current are in phase.Since the current and voltage are in phase and theac source is a sine wave, the power used by the resistoris easily computed from root-mean-square (rms) (effective)voltage and current readings taken with a typicalmultimeter. The power is computed by taking the productof the measured voltage in volts or kilovolts and themeasured current in amperes:P(watts) = V(volts) × I (amperes)where V is the voltage measured in volts and I is thecurrent measured in amps. Both quantities are measuredwith an rms reading meter.In many industrial settings the voltage may be

890 ENERGY MANAGEMENT HANDBOOKFig. III.6 Circuit showing an ac source with radian frequencyω. The current through the resistor is in phasewith the voltage across the resistor.measured in kilovolts and current in amperes. Thepower is computed as the product of current and voltageand expressed as kilowatts:P (kilowatts) = V (kilovolts) × I (amps)If it is unhandy to measure both voltage and current,one can compute power using only voltage or currentif the resistance R is known:orP (watts) = I 2 (amps) × R (ohms)V 2 (volts)P(watts) = ————R(ohms)III.3.2 InductanceInductance L in an ac circuit is usually formed ascoils of wire, such as those found in motor windings,solenoids, or inductors. In a real circuit it is impossibleto have only pure inductance, but for purposes of establishingbackground we will take the theoretical caseof a pure inductance so that its circuit properties can beisolated and presented.An inductor is a circuit element that uses no realpower; it simply stores energy in the form of a magneticfield and will give up this stored energy, alternately storingenergy and giving it up every half-cycle. The resultof this storing and giving up energy when an inductoris driven by a sine-wave source is to put the measuredmagnetizing current (i c .) 90° out of phase with thedriving voltage. The magnetizing current lags behindthe driving voltage by 90°. If pure inductance were theload of a sine-wave generator, we could summarize itscharacteristics as in Figure III.7.An inductor limits the current flowing throughit by reacting with the voltage change across it. Thisproperty is called inductive reactance X L . The inductiveFig. III.7 (a) Ac circuit with pure inductance. (b) Plotof the voltage across the inductor v L , and the current i Lthrough it. The plot shows a 90° displacement betweenthe current and the voltage.reactance of a coil whose inductance is known in henrys(H) may be computed using the expressionX L = 2πƒLwhere ƒ is the frequency in hertz and L is the coil’sinductance in henrys.III.3.3 CapacitanceCapacitance C, like inductance, only stores andgives up energy. However, the voltage and current phasingis exactly opposite that of a inductor in an ac circuit.The current in an ac circuit containing only capacitanceleads the voltage by ∠90°. Figure III.8 summarizes thecharacteristics of an ac circuit with a pure capacityload. A capacitor also reacts to changes. This propertyis called capacitive reactance X c . The capacity reactancemay be computed by using the expression1X c = ———2πfC

REVIEW OF ELECTRICAL SCIENCE 891In the preceding section it was mentioned that pureinductance does not occur in a real-world circuit. This isbecause the wire that is used to form the most carefullymade coil still has resistance. This section considers circuitscontaining resistance and inductive reactance andcircuits containing resistance and capacity reactance. Attentionwill be given to the notation used to describe suchcircuits since vector algebra must be used exclusively.Fig. III.8 (a) Ac circuit with pure capacitance. (b) Plotof the voltage across the capacitor v c , and the current i cthrough it. The plot shows a 90° displacement betweenthe current and the voltage.III.4.1 Circuits with Resistance and Inductive ReactanceFigure III.9 shows a circuit that has both resistiveand inductive elements. Such a circuit might representa real inductor with the resistance representing the wireresistance, or such a circuit might be a simple model of amotor, with the inductance reflecting the inductive characteristicsof the motor’s windings and the resistancerepresenting both the wire resistance and the real powerconsumed and converted to mechanical work performedby the motor.In Figure III.9 the current is common to both circuitelements. Recall that the voltage across the resistor isin phase with this current while the voltage across theinductor leads the current. This idea is shown by plottingthese quantities in the complex plane. Since i is thereference, it is plotted on the positive real axis as shownin Figure III.10.The voltage across the resistor is in phase with thecurrent, so it is also on the positive real axis. whereasthe voltage across the inductor is on the positive j axiswhere ƒ is the frequency in hertz and C is the capacityin farads.III.3.4 SummaryCircuit elements are resistance that consumes realpower and two reactive elements that only store andgive up energy. These two reactive elements, capacitorsand inductors, have opposite effects on the phase displacementbetween the current and voltage in ac circuits.These opposite effects are the key to adding capacitorsin an otherwise inductive circuit for purposes of reducingthe current-voltage phase displacement. Reducingthe phase displacement improves the power factor of thecircuit. (Power factor is defined and discussed later.)Fig. III.9 Circuit with both resistance and inductance. Thecircuit current i is common to both elements.III.4 IMPEDANCEFig. III.10 Circuit voltages and current plotted in thecomplex plane. Both i and V R are on the positive real axissince they are in phase, V L is on the positive j axis sinceit leads the current by 90°.

892 ENERGY MANAGEMENT HANDBOOKsince it leads the current by 90°. However, the sum ofthe voltages must be the source voltage e. Figure III.10shows that the two voltages must be added as vectors:e = v r + jv Le = i R + jiX LIf we call the ratio of voltage to current the circuit impedance,theneZ = — = R + jX LiZ, the circuit impedance, is a complex quantity and maybe expressed in either polar or rectangular form:Fig. III.12 Summary of the relationship among R, X c ,and Z shown in the complex plane.orZ = R + jX LZ = |Z | θIn circuits with resistance and inductance the compleximpedance will have a positive phase angle and if R andX L are plotted in the complex plane, X L is plotted on thepositive j axis, as shown in Figure III.11.III.4.2 Circuits with Resistance and Capacity ReactanceCircuits containing resistance and capacitance areapproached about the same way. Going through a similaranalysis and looking at the relationship among R, X c ,and Z would show that X c is plotted on the negative jaxis, as shown in Figure III.12.III.4.3 SummaryFig. III.11 Plot in the complex plane showing the complexrelationship of R, X L , and Z.In circuits containing both resistive and reactive elements,the resistance is plotted on the positive real axiswhile the reactances are plotted on the imaginary axis.The fact that inductive and capacitive reactance causesopposite phase displacements (has opposite effects inac circuits) is further emphasized by plotting their reactanceeffects in opposite directions on the imaginaryaxis of the complex plane. The case is building for whycapacitors might be used in an ac circuit with inductiveloading to improve the circuit’s power factor.III.5 POWER IN AC CIRCUITSThis section considers three aspects of power in accircuits. First, the case of a circuit containing resistanceand inductance is discussed, followed by the introductionof the power triangle for circuits containing resistanceand inductance. Finally, power-factor improvementby the use of capacitors is presented.III.5.1 Power in a Circuit ContainingBoth Resistance and InductanceFigure III.13 reviews this situation through a circuitdrawing and the voltages and currents shown in thecomplex plane. Meters are in place that read the effectiveor rms voltage V across the complex load and theeffective or rms line current I.Power is usually thought of as the product of voltageand the current in a circuit. The question is: Thecurrent I times which voltage will yield the correct ortrue power? This is an important question, since FigureIII.13b shows three voltages in the complex plane.Each of the three products may be taken, and each

REVIEW OF ELECTRICAL SCIENCE 893This is the power that is alternately stored and given upby the inductor to maintain its magnetic field. None ofthis reactive power is actually used.If the voltages in the foregoing examples weremeasured in kilovolts, the three values computed wouldbe the more familiar:P(apparent) = kVAP(real) = kWP(imaginary) = kVARThis discussion, together with Figure III.13b, leads tothe power triangle.Fig. III.13 (a) Circuit having resistance and inductance;meters are in place to measure the line current I andthe voltage V. (b) Relationship between the variousvoltages and the line current for this circuit.has a name and a meaning. Taking the ammeter reading Itimes the voltmeter reading yields the apparent power. Theapparent power is the load current-load voltage productwithout regard to the phase relationship of the currentand voltage. This figure by itself is meaningless:P(apparent) = IVIf the voltmeter could be connected across the resistoronly, to measure v R , then the line current-voltage productwould yield the true power, since the current andvoltage are in phase.P(true) = Iv RUsually, this connection cannot be made, so the truepower of a load is measured with a special meter calleda wattmeter that automatically performs the followingcalculation.III.5.2 The Power TriangleThe power triangle consists of three values, kVA,kW, and kVAR, arranged in a right triangle. The anglebetween the line current and voltage, Θ, becomes animportant factor in this triangle. Figure III.14 shows thepower triangle.To emphasize the relationship between these threequantities, an example may be helpful. Suppose that wehave a circuit with inductive characteristics and using avoltmeter, ammeter, and wattmeter the following valuesare measured:watts = 1.5 kWline current = 10 Aline voltage = 240 VFrom this information we should be able to determinethe kVA, Θ, and the kVAR.The kVA can be computed directly from the voltmeterand ammeter readings:kVA = (10 A)(0.24) kV = 2.4 kVAP(true) = IV cos θNote that in Figure III.13b, the circuit voltage V and theresistance voltage V R are related through the cosine of θ.The third product that could be taken is called imaginarypower or VAR, the voltampere reactive product.P(imaginary) = Iv LFig. III.14 Power triangle for an inductive load. The angleθ is the angle of displacement between the line voltageand the line current.

894 ENERGY MANAGEMENT HANDBOOKLooking at the triangle in Figure III.14 and recallingsome basic trigonometry, we haveand θ is the angle whose cosine equals 0.625. This canbe looked up in a table or calculated using a hand calculatorthat computes trig functions:θ = cos –1 0.625 = 51.3°Again referring to the power triangle and a little trig,we see thatkVAR = kVA sin θ= 2.4 kVA sin 51.3°= 1.87 kVARFigure III.15 puts all these measured and calculated datatogether in a power triangle.Of particular interest is the ratio kW/kVA. Thisratio is called the power factor (PF) of the circuit. Sothe power factor is the ratio of true power to apparentpower in a circuit. This is also the cosine of the angleθ, the angle of displacement between the line voltageand the line current. To improve the power factor, theangle θ must be reduced. This could be accomplishedby reducing the kVAR side of the triangle.Fig. III.15 Organizationof the measured andcomputed data of theexample into a powertriangle.III.5.3 Power-Factor ImprovementRecall that inductive reactance and capacity reactanceare plotted in opposite directions on the imaginaryaxis, j. Thus it should be no surprise to consider thatkVAR produced by a capacitive load behave in an oppositeway to kVAR produced by inductive loads. Thisis the case and is the reason capacitors are commonlyadded to circuits having inductive loads to improvepower factor (reduce the angle θ).Suppose in the example being considered thatenough capacity is added across the load to offset theeffects of 90% of the inductive load. That is, we will tryFig. III.16 (a) Inductive circuit with capacity added tocorrect power factor. (b) Power vectors showing therelationship among kW, kVAR inductive, and kVARcapacitive.to improve the power factor by better than 90%. FigureIII.16 shows the circuit arrangement with the kW andkVAR vectors drawn to show their relationship.Following the example through, consider FigureIII.17, where 90% of the kVAR inductive load has beenneutralized by adding the capacitor.Working with the modified triangle in Figure III.17,we can compute the new θ, call it θ 2 .0.19θ 2 = tan–1 ——1.5= 7.2Again, a calculator comes in handy.Since the new power factor is the cosine of θ 2 , wecomputePF new = cosine 7.2 = 0.99certainly an improvement.Recall that the power factor can be expressed as aratio of kW to kVA. From this idea we can compute anew kVA value:1.5 kWPF = 0.99 = —————kVA newor

REVIEW OF ELECTRICAL SCIENCE 895Fig. III.17 Resulting net power triangle when the capacitoris added. A new kVA can be calculated as wellas a new θ.1.5 kWkVA new = ———0.99= 1.52 kVAThe line voltage did not change, so the line current mustbe lower.1.52 kVA new = 0.24 kV × I new1.52 kVAI new = ————— = 6.3 A0.24Comparing the original circuit to the circuit after addingcapacity, we have:Inductive Circuit Improved Circuit—————————————————————————Line voltage 240 V 240 VLine current 10 A 6.3 APF 62.5% 99%kVA 2.4 kVA 1.52 kVAkW 1.5 kW 1.5 kWkVAR 1.87 kVAR 0.19 kVARThe big improvement noted is the reduction of linecurrent by 37% with no decrease in real power, kW,used by the load. Also note the big change in kVA; lessgenerating capacity is used to meet the same real powerdemand (generator input power is determined by KVAoutput).III.5.4 SummaryThrough an example it has been demonstratedhow the addition of a capacitor across an inductiveload can improve power factor, reduce line current, andreduce the amount of generating capacity required tosupply the load. The way this comes about is by havingthe capacitor supply the inductive magnetizing currentlocally. Since inductive and capacitive elements storeand release power at different times in each cycle, thisreactive current simply flows back and forth betweenthe capacitor and inductor of the load. This idea isFig. III.18 (a) Pictorial showing the inductive load ofthe example in this section. (b) The load with a capacitoradded. With the exchange of the kVAR currentbetween the capacitor and inductive load, very littlekVAR current is supplied by the generator.reinforced by Figure III.18. Adding capacitors to inductiveloads can free generating capacity, reduce line loss,improve power factor, and in general be cost effectivein controlling energy bills.III.6 THREE-PHASE POWERThree-phase power is the form of power most oftendistributed to industrial users. This form of transmissionhas three advantages over single-phase systems: (1)less copper is required to supply a given power at givenvoltage; (2) if the load of each phase of the three-phasesource is identical, the instantaneous output of the alternatoris constant; and (3) a three-phase system producesa magnetic field of constant density that rotates at theline frequency—this greatly reduces the complexity ofmotor construction.The author realizes that both delta systems andwye systems exist, but will concentrate on four-wire wyesystems as being representative of internal distributionsystems. This type of internal distribution system allowsthe customer both single-phase and three-phase service.Our focus will be on measuring power and determiningpower factor in four-wire three-phase wye-connectedsystems.III.6.1 The Four-Wire Wye-Connected SystemFigure III.19 shows a generalized four-wire wyeconnectedsystem. The coils represent the secondarywindings of the transformers at the site substation whilethe generalized loads represent phase loads that are the

896 ENERGY MANAGEMENT HANDBOOKTable III.1 How to Select Capacitor Ratings for Induction Motors/Source: 1.Fig. III.19 Generalized four-wire wyeconnectedsystem. The coils A, B, and Crepresent the three transformer secondariesat the site substation; while Z A , ZB.and Z C are the generalized loads seenby each phase.

REVIEW OF ELECTRICAL SCIENCE 897sum loads on each phase. These loads may be compositesof single-phase services and three-phase motorsbeing fed by the distribution system. N is the neutralor return.To determine the power and power factor of anyphase A, B, or C, consider that phase as if it were asingle-phase system. Measure the real power, kW, deliveredby the phase by use of a wattmeter and measureand compute the volt-ampere product, apparent power,kVA, using a voltmeter and ammeter.The power factor of the phase can then be determinedand corrected as needed. Each phase can betreated independently in turn. The only caution to noteis to make the measurements during nominal load periods,this will allow power-factor correction for the mostcommon loading.If heavy motors are subject to intermittent duty,additional power and power-factor information can begathered while they are operating. Capacitors used tocorrect power factor for these intermittent loads shouldbe connected to relays so that they are across the motorsand on phase only when the motor is on; otherwise,overcorrection can occur.

898 ENERGY MANAGEMENT HANDBOOKTable III.2 How to Select Capacitor Ratings for Induction Motors/Source: 2.In the special case of a four-wire wye-connectedsystem with balanced loading, two wattmeters may beused to monitor the power consumed on the service andalso allow computation of the power factor from the twowattmeter readings.III.6.1.1 Balanced Four-Wire Wye-Connected SystemFigure III.20 shows a balanced system containingtwo wattmeters. The sum of these two wattmeter readingsare the total real power being used by the service:P T = P 1 + P 2Further, the angle of displacement between each linecurrent and voltage can be computed from P 1 and P 2 :θ = tan –1 3 P 2 –P 1P 2 +P 1and the power factor PF = cos θ.

REVIEW OF ELECTRICAL SCIENCE 899This quick method for monitoring power and powerfactor is useful in determining both fixed capacitors tobe tied across each phase for the nominal load, and thecapacitors that are switched in only when intermittentloads come on-line.The two-wattmeter method is useful for determiningreal power consumed in either wye- or delta-connectedsystems with or without balanced loads:P T = P 1 + P 2However, the use of these readings for determiningphase power factor as well is restricted to the case ofbalanced loads.III.6.2 SummaryThis brief coverage of power and power-factordetermination in three-phase systems covers only thevery basic ideas in this important area. It is the aim ofthis brief coverage to recall or refresh ideas once learnedbut seldom used.Tables III.1 and III.2 were supplied by GeneralElectric. who gave permission for the reproduction oftheir materials in this handbook.

900 ENERGY MANAGEMENT HANDBOOKFig. III.20 Four-wire wye-connected system with wattmeter connectionsdetailed. Solid circle voltage connections to wattmeter; open circle, currentconnections to wattmeter.

INDEXSymbols1995 Model Energy Code (MEC)5412000 International Energy ConservationCode (IECC) 541Aabsolute pressure 126absorber plate 475absorptance 476absorption chillers 272abuse 445acceptable indoor air quality 499accidents 623accuracy 588active power 279actuator 588adaptive control 318adjustable speed drive 291administrative sequence logic 333AEE (Association of Energy Engineers)xi, 1, 501after-tax cost of capital 43AFUE 789, 791AGA 172aggregation 643AHU 339systems 684air-handling units 716air-source heat pump 757airflow measurement devices 25air change method 238air collectors 477air compressors 406maintenance for 407air conditioning optimization 345air handler 748systems 684air velocity and airflow 260alarms 346monitoring and reporting 321all-air systems 250all-water systems 255alternative energy source 471ambient temperature 299, 449American Boiler Manufacturers Association(ABMA) 415American Council for an EnergyEfficient Economy 379American Physical Society 7American Society of Heating, Refrigeration,and Air Conditioning(ASHRAE) 397, 499, 540American Society of Testing Materials437ammetersclamp on 24amperage 284ampere 397amps 275, 298anaerobic digestion processes 492analog 616input 342, 349output (AO) 342, 349to digital A/D converter 349annual cost 468annual energy review 193annual energy use 716annual expenses 42annual worth 55anode 493ANSI 397anti-wind-up 583application requirements 443approach 604arc tube 397ASHRAE 397, 499, 540ASHRAE “Zone Method” 227ASHRAE 90.1-1999 540ASHRAE 90.2-1993 540ASHRAE Equipment Handbook407ASHRAE Guidelines 3-1990Reducing Emission ofFully Halogenated Chlorofluorocarbon(CFC) 544ASHRAE Handbook of Fundamentals237, 246, 460, 473ASHRAE Standard 501ASHRAE Standard 62-1999 543ASHRAE Standard 90-80 540ASHRAE Standard 90.1 226, 229ASHRAE ventilation standard, 62-1989 543901Association of Energy Engineers(AEE) xi, 1, 501ASTM 450as built drawings 348auditor’s toolbox 24auditscommercial 36industrial 34planning 14residential 37available forms 440average rated life 397avoided costs 177Bbaffle 397ballast 360, 397cycling 397efficiency factor 397factor 360, 397bare-surface heat loss 461base 222bath-tub curve 623bearings 407before tax cash flows (BTCF) 44Betz coefficient 485bin 775, 777method 246, 768blower door attachment 25boiler 338, 402, 722, 748economizer 216efficiency 128, 414maintenance 413measurements 722optimization 320boiling point 126BOMA 797bonds 655borehole 766bottoming cycle 157Bourdon gauge 428British thermal unit (Btu) 126building automation system 616building balance temperature 245building envelope 221, 402building load coefficient (BLC) 239,242

902 ENERGY MANAGEMENT HANDBOOKBuilding Officials and Code AdministratorsInternational(BOCA) 541building related illness 499causes 499by-pass 291Ccalcium silicate 441calibration 615California Title 24 540candela 397distribution 397candle power 397capacitive switching HID fixtures361capacitor 279capitalize 666capital investments 41, 42cost categories 42capital or financial lease 666capital rationing 42cash flow diagrams 42cathode 493CELCAP 174cellular glass 441cell structure 440central chiller plants 693ceramic metal halide lamps (CMH)376ceramic recuperator 213Certified Energy Manager 1CFC 793, 802CFL 393chilled water reset 320chilled water storage 523chiller 271, 339, 531, 693, 718, 720,748consumption profile 522demand limiting 321optimization 321system capacity 526chlorofluorocarbons (CFCs) 543circulators 338Clean Air Act Amendment 544Climatic Change Action Plan 547closed loop 589, 616heat pumps 257closed heat exchangers 200code 299coefficient of heat transmission 439coefficient of performance (COP)198, 719, 776, 790, 791coefficient of utilization (CU) 362,397cogeneration 545facility 176COGENMASTER 175collectively exhaustive 60collector 478efficiency 475color 356color rendering index (CRI) 356,397color temperature 397combined cycle 157combustion analyzers 25combustion or gas turbines 169commissioning 329, 799a new building 704for energy management 671measures 680compact fluorescent 397lamps (CFLs) 358compound interest 47, 48comprehensive, explicit, and relevanttraining and support(CERTs) 331compression of insulation 224compressive strength 440computer programs 468concentrating collectors 478condensate systems 125, 147condensation 125control 454conduction 437constant-worth dollars 66constant fire boilers 724constant volume dual duct 254constrained deterministic analysis59contaminants 38, 39contaminant amplification 499contingency 59planning 632contingent 60continuous commissioning 672, 687contrast 356, 397controls 503control drawings 347control loop 349, 616control relays 343control systems 264, 404, 481control valves 337, 339convection 437convective recuperator 212converter 300conveyor systems 423cooling systems 271cooling towers 272, 339, 693, 748cool white lamps 359coordinated color temperature(CCT) 356COP 198, 719, 776, 790, 791correlation coefficient, R 243corrosion 492control 209costs 41cost effectiveness 41cost factors 446cost of capital 44Council of American Building Officials(CABO) 541countermeasures 630, 633criteria for selection 208cross-contamination 214cubic feet per minute (CFM) 290current transformers 343cut-in 578, 616cut-out 578, 616cut set 630Ddaily scheduling 319daisy chain 349dampers 300, 337, 338daylighting 383, 804daylight harvesting 364DDC 579, 580, 581DDC EMCS 317, 319, 324dead band 345, 349, 595, 616deal time 349debt financing 43dedicated outdoor air systems 810deductive analysis 624deductive method 627deferred maintenance 634Degree Days 240demand controlled ventilation(DCV) 598, 612demand limiting or load shedding320demand management 261demand shedding 345demand side management (DSM) 1depreciation 44, 666

INDEX 903accelerated cost recovery system45clining balance 45straight line 45sum-of-the-years digits 45Dept. of Energy 539deregulation 634, 640, 641derivative control 349desiccant 811design 298life 447of the waste-heat-recoverysystem 206desuperheating 765deterministic analysisconstrained 59economic 64deterministic dew-point determination454dew-point temperature 449DG economics 168diaphragm gauge 428differential temperature controller481diffuser 397digital communication bus 349digital inputs 342digital outputs (DO) 342digital to analog D/A converter 349dilution 501, 543dimmable CFLs 375direct digital control 315, 579direct glare 397discounted cash flow (DCF) 466discounted payback 465disk traps 421distributed control 349distributed generation 167distribution energy 270distribution systems 690DOAS 810domestic hot water (DHW) 272, 339doors 237double failure matrix (DFM) 625,626, 627downlight 397downsizing 634drift 584, 588, 589drive system efficiency 273dry contact 617dual-technology sensors (DT) 366dual duct systems 254dual duct VAV 254dumping waste heat 199duplex steels 210duty 299cycling 320, 345logs 321dynamic system 525EE.O. 13123 542economics 481economic analysis 41, 58economic calculations 464economic thickness (ETI) 466, 468economizers 263eddy-current clutch 300educational planning 15EER 776, 790, 791efficacy 354, 398efficiency 276index 299elastomeric cellular plastic 442electric/pneumatic relays 317, 341electrical measurements 427electrical system 404electric energy management 273electric industry deregulation 638electric lift trucks 422electric motor efficiency 275electric power industryhistorical perspective 637Electric Power Research Institute(EPRI) 35, 368electronic ballast 398electronic ballasts 361EMCS application 322EMCS installation 339EMCS Retrofit 324EMCS software specifications 333emissivity 235emittance 451, 476end-of-year cash flow 42energy-saving ballast 398energy-saving lamp 398energy analysis and diagnosticcenters 4energy audit 23, 32, 247energy audit format 32energy broker 640energy conservation 634analyzing 138Energy Conservation and Produc-tion Act in 1977 541energy conservation opportunities(ECOs) 24energy consumption 222energy efficiency provisions 539energy flux 202Energy Information Administration539energy maintenance 403energy management 9, 315, 633control system 315functions 344in materials handling 430program 7, 247energy management systems315, 617energy manager 10energy marketer 640energy monitoring and controlsystem 617energy policy 13, 18Energy Policy Act of 1992 237, 539,544, 546Energy Policy and ConservationAct 541Energy Productivity Center of theCarnegie-Mellon Institute ofResearch 5energy savings 707calculations 463measurement of 696energy security 621energy servicesin-house vs. outsourced 643energy service company (ESCO)666energy systems maintenance 401energy team 11enthalpy 128envelope analysis 223, 242Environmental Protection Agency416equal percentage 586equipment efficiency 267equipment failure 622equipment list 326equity financing 43equivalence 52equivalent thickness 450ESCOs (energy service companies) 2eutectic salts 525, 531eutectic storage 531

904 ENERGY MANAGEMENT HANDBOOKevaporative-condensing cycle 216evaporative cooling 597, 811event tree analysis 625excess CO 415excess O 2 413executive orders 542exempt wholesale generators(EWGs) 546exhaust fans 339exit signs 363expenses 42extraction 501Ffacility appraisal 326facility layout 28facility specific instructions 331failure mode and effect analysis(FMEA) 624failure mode effect and criticalityanalysis (FMECA) 624fan-coils 256fan law 290fans 716faradaic or current efficiency 494fault hazard analysis (FHA) 624fault tree 627, 630analysis 627FCU 349Federal Energy Administration 467Federal Energy Management ImplementationAct (FEMIA) 542Federal Energy Management Program(FEMP) 2Federal Energy Regulatory Commission(FERC) 546, 638Federal Power Act 545feedback 617fenestration 234FERC 177FERC Order No. 436 547FERC Order No. 636 547FERC Order No. 636A 547FHA 625fiber optics 378fill factor 483filtration 498, 502, 503, 543finance terminology 652financial analysis 464financing 43, 649finned tube 215, 218fireproofing 444fire hazard classification 440first cost 41fixture 399dirt depreciation (LDD) 369efficiency 399flash steam 128, 149, 420flash tanks 148flat-plate collector 473, 477floating control 579, 582, 617float and thermostatic traps 420floors 233below grade 234on grade 234flow hoods 429flow meters 343fluorescent lamps 358, 398follow-up 412foot-candles 353, 398footlambert 398form required 446fouling 415, 492frame 298size 276freeze protection 346, 459free cooling 321, 336frequency 276fuel cells 170, 493molten carbonate (MCFC) 495full-load speed 274full storage systems 521, 526, 528functions 319furnaces 722future cash flows 66constant-worth dollars 66then-current dollars 66Ggain 581, 617gas or Liquid-to-Liquid Regenerators216Gas Research Institute (GRI) 35gauge pressure 126general test form 282geographic location 28Gibbs free energy 493glare 356, 398glass fiber 441glazed flat-plate collectors 476global data exchange 342gradient series 52graphics 579greenhouse effect 475green buildings 795, 796green power 799ground-source heat pumps 755,756, 757, 783group relamping 370grout 766guidevanes 300Hharmonic 281, 380, 398distortion 361health effect consequences 499heat-balance diagram 203heat/power ratio 157heating, ventilating, air conditioning(HVAC) 315systems 402heat exchangersconcentric-tube 217shell-and tube 217heat flow 449, 450heat flux 449heat gain 221heat loss 221from a floor to a crawl space234heat pipe 216array 215heat plant 447heat pumps 198heat recovery systems 263heat transfer 438heat wheel 208, 213HID 398high-bay 398high-intensity discharge (HID) 359high-temperature controller 481higher heating value 203high inertia load 276high output (HO) 398high pressure sodium 359high pressure sodium lamp 398holiday programming 319host 666hot deck/cold deck temperaturereset 321hot water distribution 405hot water reset 320HP 298, 349HSPF 790human dimension 317, 331hunting 583

INDEX 905hurdle rate 58HVAC 398HVAC systems 247, 448, 497, 730types 249HZ 298IEQ (indoor environmental quality)500ice storage 525, 531ideal heat pump 198IES (Illuminating Engineering Society)353, 405, 428IFMA 797IGSHPA 769illuminance 398Illuminating Engineering Society(IES) 353, 405, 428improper drainage 419incandescent 358income 43indifference 53indirect glare 398indoor air quality (IAQ) 38, 260,497, 498, 501, 542manager 501problems, solutions and preventionof 500induction lighting 377induction motors 274, 279, 285, 286induction systems 257inductive methods 624industrial assessment centers 4industrial light meter 428infiltration 237air flow 237commercial buildings 238residential buildings 237inflation 41, 46, 66infrared cameras 24infrared equipment 429initial costs 41, 42input/output point 326input/output units 342inputs 342insolation 473installation 340instant start 398insulationclass 276, 299cost considerations 461economics 461equivalent thickness 449flexible 441formed-in-place 441materials 439nonrigid 445penetrations 224properties 439removable-reusable 441selection 443thickness determination 448integral control 350integrated solid waste management489intelligent building 350interest 46interest rateeffective 65nominal 65period 65interface temperature 450interlock 583, 596internal rate of return 56International Code Council (ICC)541International Conference of BuildingOfficials (ICBO) 541inverted bucket traps 420inverter 300, 301investment 41investment analysis 42IPLV 790, 791isotherms 225JJustification of EMCSs 321KKyoto protocol 544LLAG 350laminar wheel 214lamp lumen depreciation (LLD) 369factor 398latent heat 128of fusion 525, 530transfer 224lay-in troffer 398leasing 659least cost 497LED 393, 398LEED 795, 796, 798lender 667lens 398lessee 667lessor 667leveraged lease 667liability 39life-cycle cost 503, 778, 780analysis 41, 42lift trucks, non-electric 422lighting 724fundamentals 353quality 355quantity 353lightmeter 24light color 355light generated current 483light loss factor (LLF) 398light measurements 428light meter 353limitations of metal recuperators212line of credit 667liquidity 667litigation, risk of 500load 282diversity factor 304factor 431limiting 596profile 303, 637types 277loans 654local distribution companies (LDCs)546location 444loose-fill insulation 440louver 398low-pressure sodium 359, 399low-temperature controller 481lumen 353, 399depreciation compensation 365luminaires 361Mmagnetic ballasts 360maintainability 211maintenance 42, 424, 503actions 416for air compressors 407manual 347of the lighting 405procedures 413program 401

906 ENERGY MANAGEMENT HANDBOOKschedule 411management control systems 7management decisions 423manifold 478manometer 428manual override 321, 346Markey Bill 294MARR (minimum attractive rate ofreturn) 668mass flow rate 202, 449mass resistances 438materials and construction 209materials handling energy savings432materials handling maintenance 421maximum theoretical COP 198MCS (multiple-chemical-sensitivity)497, 499mean light output 399mean temperature 440, 449, 450mean time between failure 631measurement and verification(M&V) 707, 711measuring instruments 426mechanical equipment 748mercury vapor 359lamp 399metallic radiation recuperator 211metal building roofs 232metal building walls 228metal elements envelope 225metal halide 359, 399microturbines 170mineral Fiber/Rock Wool 441miniature data loggers 25minimum annual cost analysis 465minimum attractive rate of return(MARR) 43, 54minimum on-minimum off times320model energy code 541Model Energy Codes 540modified accelerated cost recoverysystem (MACRS) 45modulation 579, 618molten carbonate fuel cell (MCFC)170monitoring 346, 431Montreal protocol 543most-open valve 618motor-generator sets 300motor 407code letters 275efficiency 284, 294manager 294operating loads 281, 282performance managementprocess 294record form 295, 296rpm 290speed control 321multimeters 427multizone systems 253mutually exclusive 59, 60set 60NNational Electrical Code 279National Electrical ManufacturersAssociation (NEM) 273, 285,379National Fenestration Rating Council237natural convection 256natural disasters 623natural gas 550environmental advantages 565purchase of 567consumption 554interstate pipeline specifications561marketers 567markets 566pricing 570purchasing 549supply 555transportation 558Natural Gas Policy Act (NGPA) 1,545of 1978 546near term results 41NEBs 322negative-sequence voltage 274NERTs 322net present value (NPV) 668new construction EMCS 326night set-back 262, 320non-annual compounding of interest41, 65non-energy benefit 322non-energy related tasks (NERTs)316North American Insulation ManufacturersAssociation 467Ooccupancy sensor 365, 366, 393, 399Occupational Safety and HealthAdministration (OSHA) 543off-balance sheet financing 668Omnibus Reconciliation Act of 199344opaque envelope components 223open-circuit voltage 483open bucket traps 419open loop 589, 618open protocol 580open waste-heat exchangers 199operating and maintenance costs 41operating conditions 460operating hours 28operating practices 425operating temperature 443operator’s terminal 350optimal start/stop 595optimization 578, 618optimum start/stop 320, 337, 345organic binders 444orifice plates 430Orsat apparatus 428outlay 43outputs 342ovensmaintenance 407over-the-Purlin 230ownership 17Pp-n junction diodes 482package boilers 416, 418parabolic 478partial load storage 526partial load system 529partial storage systems 521, 529par value or face value 668passive air preheaters 215passive infrared sensors (PIR) 365,366payback period 58, 468PCB ballasts 383peer-to-peer network 318performance contracting 661, 710performance contractors 2periodic replacement 42personnel protection 450phase 277, 299phase-change materials 480, 525

INDEX 907phosphoric acid (PAFC) 494phosphoric acid fuel cell (PAFC) 170photocells 364, 399photographic light meter 428photovoltaics 170PID control 350pipelines 558Pitot tubes 429PI control 350planning 13horizon 42, 63plastic foams 442plate-type 215pneumatic 579, 581pocket thermometers 429poles 277poll/response system 318polyimide foams 442potential 193transformers 343power coefficient 486power factor 277, 279, 281, 399, 405controller 291meter 25, 427power meter 282power survey 281, 282preferred stock 668present-value cost analysis 466present dollars 41present worth 54factor 50savings investment ratio 54pressure measurements 428pressure sensors 343pressure switches 343proactive monitoring 501process control 456process wastes 489process work 448programmable logic controllers 315project financing 668project measures of worth 54annual worth 54internal rate of return 54payback period 54proton exchange membrane (PEM)495properties of thermal storage materials197proportional-integral (PI) control316, 583proportional control 350, 583proportional output 618protective coatings and jackets 442protective countermeasures 631proton exchange membrane (PEM)170public awareness 498Public Utility Holding CompanyAct (PUHCA) 539, 546of 1935 545Public Utility Regulatory PoliciesAct (PURPA) 177, 545pulse accumulators 342pulse width modulation 301, 350pumps 714purge section 214purlin 230PURPA 177, 545pyrolysis 492Qqualifying facilities (QFs) 545quantifying 194Rrachet clause 26radiant heating 256radiation 438radon gas 39rapid recovery 631rapid start (RS) 399rate structure 26reactive power 279, 399recessed 399reciprocating engines 169recirculation 321recommended light levels 354, 355recommissioning 672recording ammeter 427recuperators 211reducing heat loss 228, 232redundant systems 631reed relays 343reflectors 362refractories 442refuse-derived fuel (RDF) 489, 490refuse combustion 492refuse preparation 490regenerators 207reheat 610systems 252relative humidity 449repeatability 589reporting 16reset 599, 603, 606, 607, 609, 812resistance 445Resource Conservation and RecoveryAct of 1976 (RCRA) 545retained earnings 44retrocommissioning 672retrofit 399revenues 41, 42risk analysis 67, 624risk management 646rock-bed storage system 480roll-runner 232room surface dirt depreciation(RSDD) 369root-mean-square (rms) 399routine maintenance 408rpm (revolutions per minute) 290,298runaround systems 200Ssabotage 623safety checklist 29safety equipment 25, 29salvage value 42saturated steam 128savings 533, 707saving investment ratio 57secured loan 668Securities and Exchange Commission(SEC) 545SEER 791selectivity 476selling stock 657sensible heat 128, 133sensitivity analysis 41, 67sequencing 584serial use 200series cash flows 50service factor 277, 298setpoint 350shared savings providers 2short-circuit current 483sick building syndrome 498, 499,542silicon controlled rectifier 301simple interest 47simulation 69single-duct VAV systems 251single duct systems 250single factor sensitivity analysis 67

908 ENERGY MANAGEMENT HANDBOOKsingle point failures 625single sum 50cash flows 49sizing 481slip 278, 282smart windows 237smoke detectors 429smoke generator 25smoke sources 33software 318routines 319specifications 327soil temperature 763, 764solar arrays 484solar cells 482, 483solar collecting systems 472, 473solar constant 472solar energy 471, 472solar thermal energy 471solid fuel pellets 492solid oxide fuel cell (SOFC) 170solid state relays (SSRs) 343source control 501, 503, 543Southern Building Codes CongressInternational (SBCCI) 541spacers 236space criterion 399specific heat 449specified heat loss 457specular 399speed ratio correction factor 290spot market 1, 546stack-gas analysis 428stack-gas stream 206stack-gas temperature 415stagnation 475state codes 540static system 525steady-state heat exchangers 208steam systems 125, 133steam traps 139, 145, 402, 418, 421failure 418stethoscope 430stochastic techniques 59storage 196mediums 523systems 521capacity 528stored heat 196stranded costs 637, 639strategic planning 16stroboscope 430summary 635superheated steam 128super insulations 443surface air film coefficient 449surface film coefficient 449surface resistance 438, 449, 450, 451surface temperature 449surge protection 343sustainability 793, 794, 806, 808synchronous speed 278system capacitance 526, 581system controllers 317system manual 346system modifications 270system performance method 540system programming 347systems checkout 348systems configuration 327systems integration 326, 329TT12 lamp 399T2 lamps 371T5 lamps 372T8 lamps 373tandem wiring 399task lighting 368tax benefits 663tax considerations 44tax effects 466TCLP compliant fluorescent lamps373temperature 278control 345difference 438drop 458inputs 342measurement 429rise 299use range 440terminal unit 350TES systems 533then-current dollars 66thermal-break 235thermally heavy building 240, 244thermally light buildings 240thermal break 229thermal comfort 248, 260thermal conductivity 196, 437, 440,449thermal energy storage systems 633thermal equilibrium 449thermal insulation 437Thermal Insulation ManufacturersAssociation 231thermal mass 221thermal performance 41of roof 230thermal pollution 193thermal resistance 223, 438, 449thermal spacers 233thermal storage 272systems 479, 526thermal stratification 523thermal weight 240thermocouples 429thermodynamic model 719temperature dependent 719thermal girts 228thermometers 24thermostatic traps 421thermowells 342The National Energy ConservationPolicy Act of 1978 541throttling 291tight building syndrome 497, 498time clocks 364, 408time rating 278time standards 408time value of money 42, 46, 465calculations 41factors 52principles 64tires 492topping cycles 157torque 278torque 279total harmonic distortion (THD)399total quality management (TQM) 3trading of electricity 640training 15, 346transducers 350, 585, 620transformers, potential 343transmittance 476transmitters 585, 620tri-phosphor lamps 359truck operation and maintenance423true lease 668trusses 230tube-and-shell heat exchanger 211twisted pair 350two-position 350, 620

INDEX 909control 581type 299typical Applications 447UU-factor 223ultrasonic (US) 365sensors 365unbalanced voltages 274unbundled rates 637unconstrained deterministic analysis59uniformity 356uniform series cash flows 50United Nations Environment Program543unit heaters 339USGBC 794, 795, 796utility billings 241utility deregulation 637utility network 622utility outages 632utility systems 622Vvariable-speed drive 214variable air volume (VAV) 250variable consumption 222variable fire boilers 724variable frequency drives 300, 302,304variable inlet vanes (VIV) 304variable pitch pulleys drive 300variable speed drive 291variable speed loads 302VAV (Variable Air Volume) 350, 498VCP 399velocity and flow-rate measurement429ventilation 498, 543rate procedure 501vertical-axis wind turbine (VAWT)486very high output (VHO) 400vibration 444analysis equipment 26measurement 430viscosity 760visual comfort probability 356VOC 800volatile organic compounds 497voltage 279, 281voltmeter 24volts 298volumetric flow rate 202WWACC (weighted average cost ofcapital) 668walk-through audit 431waste heat 193and load diagrams 205boilers 218exchangers 207quality 194source 194survey 201waterwall steam generator 491water storage 531tank 479water treatment 272watt-hour meters 343transducers 343wattmeter 25, 428watt (W) 400WECS 489wet-side economizer 272WHAT IF motor comparison form297whole-building approach 732wind characteristics 486wind devices 485wind energy 484aerodynamic efficiency 485availability 484conversion system 488loadings and acoustics 489power coefficient Cp. 485power density 484systems 471wind speed 488wind Systems 486wind up 583workplane 400Xxpanded perlite 441Yyearly scheduling 319

energy management handbook, sixth edition - serviciilocale.md

Transform your PDFs into Flipbooks and boost your revenue!

energy management handbook, sixth edition - serviciilocale.md

Transform your PDFs into Flipbooks and boost your revenue!

Delete template?

Save as template ?