GUIDE TO CONE PENETRATION TESTING - SPGEO

GUIDE TOCONE PENETRATIONTESTINGwww.greggdrilling.com

Engineering UnitsMultiplesMicro () = 10 -6Milli (m) = 10 -3Kilo (k) = 10 +3Mega (M) = 10 +6Imperial UnitsSI UnitsLength feet (ft) meter (m)Area square feet (ft 2 ) square meter (m 2 )Force pounds (p) Newton (N)Pressure/Stress pounds/foot 2 (psf) Pascal (Pa) = (N/m 2 )Multiple UnitsLength inches (in) millimeter (mm)Area square feet (ft2) square millimeter (mm 2 )Force ton (t) kilonewton (kN)Pressure/Stress pounds/inch 2 (psi) kilonewton/meter 2 kPa)tons/foot 2 (tsf) meganewton/meter 2 (MPa)Conversion FactorsForce: 1 ton = 9.8 kN1 kg = 9.8 NPressure/Stress 1kg/cm 2 = 100 kPa = 100 kN/m 2 = 1 bar1 tsf = 96 kPa (~100 kPa = 0.1 MPa)1 t/m 2 ~ 10 kPa14.5 psi = 100 kPa2.31 foot of water = 1 psi 1 meter of water = 10 kPaDerived Values from CPTFriction ratio: R f = (f s /q t ) x 100%Corrected cone resistance: q t = q c + u 2 (1-a)Net cone resistance:q n = q t – voExcess pore pressure: u = u 2 – u 0Pore pressure ratio:Bq = u / q nNormalized excess pore pressure: U = (u t – u 0 ) / (u i – u 0 )where: u t is the pore pressure at time t in a dissipation test, andu i is the initial pore pressure at the start of the dissipation test

Guide toCone Penetration Testingandit’s Application toGeotechnical EngineeringByP. K. RobertsonandK.L. RobertsonGregg Drilling & Testing Inc.July 2006

Gregg Drilling & Testing, Inc.Corporate Headquaters2726 Walnut AvenueSignal Hill, California 90755Telephone: (562) 427-6899Fax: (562) 427-3314E-mail: info@greggdrilling.comWebsite: www.greggdrilling.comThe publisher and the author make no warranties or representations of any kind concerning the accuracy orsuitability of the information contained in this guide for any purpose and cannot accept any legalresponsibility for any errors or omissions that may have been made.Copyright © 2006 Gregg Drilling & Testing, Inc. All rights reserved.

TABLE OF CONTENTSGlossaryiIntroduction 1Risk Based Site Characterization 2Role of the CPT 3Cone Penetration Test (CPT) 5Introduction 5History 6Test Equipment and Procedures 9Additional Sensors/Modules 10Pushing Equipment 11Depth of Penetration 15Test Procedures 15CPT Interpretation 18Soil Profiling and Classification 19Equivalent SPT N 60 Profiles 23Undrained Shear Strength (s u ) 26Soil Sensitivity 27Undrained Shear Strength Ratio (s u /σ’ v ) 28Stress History - Overconsolidation Ratio (OCR) 29In-Situ Stress Ratio (Ko) 30Friction Angle 31Relative density (D r ) 33Stiffness and Modulus 35Modulus from Shear Wave Velocity 36Identification of Unusual Soils Using the SCPT 37Hydraulic Conductivity (k) 38Consolidation Characteristics 41Applications of CPT Results 44Shallow Foundation Design 45Deep Foundation Design 68Liquefaction 80Ground Improvement Compaction Control 101Design of Wick or Sand Drains 105Main References 107

CPT Guide – 2006GlossaryGlossaryThis glossary contains the most commonly used terms related to CPT and arepresented in alphabetical order.CPTCone penetration test.CPTUCone penetration test with pore pressure measurement – piezoconetest.ConeThe part of the cone penetrometer on which the cone resistance ismeasured.Cone penetrometerThe assembly containing the cone, friction sleeve, and any othersensors and measuring systems, as well as the connections to the pushrods.Cone resistance, q cThe force acting on the cone, Q c , divided by the projected area of thecone, A c .q c = Q c / A cCorrected cone resistance, q tThe cone resistance q c corrected for pore water effects.q t = q c + u 2 (1- a n )Data acquisition systemThe system used to record the measurements made by the conepenetrometer.Dissipation testA test when the decay of the pore pressure is monitored during a pausein penetration.Filter elementThe porous element inserted into the cone penetrometer to allowtransmission of pore water pressure to the pore pressure sensor, whilemaintaining the correct dimensions of the cone penetrometer.Friction ratio, R fThe ratio, expressed as a percentage, of the sleeve friction, f s , to thecone resistance, q t , both measured at the same depth.R f = (f s /q t ) x 100%i

CPT Guide - 2006GlossaryFriction reducerA local enlargement on the push rods placed a short distance above thecone penetrometer, to reduce the friction on the push rods.Friction sleeveThe section of the cone penetrometer upon which the sleeve friction ismeasured.Normalized cone resistance, Q tThe cone resistance expressed in a non-dimensional form and takingaccount of the in-situ vertical stresses.Q t = (q t – σ vo ) / σ' voNet cone resistance, q nThe corrected cone resistance minus the vertical total stress.q n = q t – σ voExcess pore pressure (or net pore pressure), ΔuThe measured pore pressure less the in-situ equilibrium pore pressure.Δu = u 2 – u 0Pore pressureThe pore pressure generated during cone penetration and measured bya pore pressure sensor:u 1 when measured on the coneu 2 when measured just behind the cone, and,u 3 when measured just behind the friction sleeve.Pore pressure ratio, B qThe net pore pressure normalized with respect to the net coneresistance.B q = Δu / q nPush rodsThick-walled tubes used to advance the cone penetrometerPush (thrust) machineThe equipment used to push the cone penetrometer and push rods intothe ground.Sleeve friction, f sThe frictional force acting on the friction sleeve, F s , divided by itssurface area, A s .f s = F s / A sii

CPT Guide – 2006IntroductionIntroductionThe purpose of this guide is to provide a concise resource for the applicationof the CPT to geotechnical engineering practice. This guide is a supplementto the book ‘CPT in Geotechnical Practice’ by Lunne, Robertson and Powell(1997). This guide is applicable primarily to data obtained using a standardelectronic cone with a 60-degree apex angle and either a diameter of 35.7mm or 43.7 mm (10 or 15 cm 2 cross-sectional area).Recommendations are provided on applications of CPT data for soilprofiling, material identification and evaluation of geotechnical parametersand design. The companion book provides details on the history of the CPT,equipment, specification and performance, as well as details ongeo-environmental applications. The book also provides extensivebackground on interpretation techniques. This guide provides only the basicrecommendations for the application of the CPT for geotechnical designA list of the main references is included at the end of this guide.1

CPT Guide - 2006Risk Based Site CharacterizationRisk Based Site CharacterizationRisk and uncertainty are characteristics of the ground and are never fullyeliminated. The extent and level of an investigation should be based on therisk of the project. Risk analysis answers three basic questions, namely:• What can go wrong?• How likely is it?• What are the consequences?Projects can be classified into low, moderate or high risk projects, dependingon the probability of the associated hazards occurring and the associatedconsequences. Low-risk projects could be projects with few hazards, lowprobability of occurrence, and limited consequences, whereas high riskprojects could be projects with many hazards, a high probability ofoccurrence, and severe consequences. Table 1 shows a generalized flowchart to illustrate the likely geotechnical ground investigation approachassociated with low risk, moderate risk and high risk projects.PROJECTPreliminary Site Evaluatione.g. geologic model, desk study,risk assessmentLOW RISKMODERATERISKHIGH RISKGround InvestigationIn-situ testing& Disturbed samplesGround InvestigationSame as for low riskprojects, plus the following:Preliminary groundinvestigationSame as for low riskprojects, plus the following:Detailedgroundinvestigation• In-situ testinge.g. SPT, CPT (SCPTu),DMT• Possibly specific testse.g. PMT, FVT• Index testinge.g. Atterberg limits, grainsize distribution, emin/e max, G s• Empirical correlationsdominateSitespecificcorrelationAdditional specificin-situ testsBasic laboratorytesting on selectedbulk sam ples(resp onse)In-situ testing• Identify criticalzonesSitespecificcorrelationAdditional in-situ tests&High qualityundisturbed samp lesHigh quality laboratorytesting (response)• Undisturbed samples• In-situ stresses• Appropriate stress path• Careful measurementsTable 1 Risk-based flowchart for site characterization.2

CPT Guide 2006Role of the CPTRole of the CPTThe objectives of any subsurface investigation are to determine thefollowing:• Nature and sequence of the subsurface strata (geologic regime)• Groundwater conditions (hydrologic regime)• Physical and mechanical properties of the subsurface strataFor geo-environmental site investigations where contaminants are possible,the above objectives have the additional requirement to determine:• Distribution and composition of contaminantsThe above requirements are a function of the proposed project and theassociated risks. An ideal investigation program should include a mix offield and laboratory tests depending on the risk of the project.Table 2 presents a partial list of the major in-situ tests and their perceivedapplicability for use in different ground conditions.Table 2 The applicability and usefulness of in-situ tests(Lunne, Robertson & Powell, 1997)3

CPT Guide – 2006Role of the CPTThe Cone Penetration Test (CPT) and its enhanced versions (i.e. piezocone-CPTU and seismic-SCPT) have extensive applications in a wide range ofsoils. Although the CPT is limited primarily to softer soils, with modernlarger pushing equipment and more robust cones, the CPT can be performedin stiff to very stiff soils, and in some cases soft rock.Advantages of CPT:• Fast and continuous profiling• Repeatable and reliable data (not operator-dependent)• Economical and productive• Strong theoretical basis for interpretationDisadvantage of CPT:• High capital investment• Requires skilled operators• No soil sample• Penetration can be restricted in gravel/cemented layersAlthough a disadvantage of the CPT is that no soil sample is obtained duringa CPT, it is possible to obtain soil samples using the CPT pushing equipment.The continuous nature of the CPT results provides a detailed stratigraphicprofile to guide in selective sampling appropriate for the project.4

CPT Guide - 2006Cone Penetration Test (CPT)IntroductionCone Penetration Test (CPT)In the Cone Penetration Test (CPT), a cone on the end of a series of rodsis pushed into the ground at a constant rate and continuous measurementsare made of the resistance to penetration of the cone and of a surfacesleeve. Figure 1 illustrates the main terminology regarding conepenetrometers.The total force acting on the cone, Q c , divided by the projected area of thecone, A c , produces the cone resistance, q c . The total force acting on thefriction sleeve, F s , divided by the surface area of the friction sleeve, A s ,produces the sleeve friction, f s . In a piezocone, pore pressure is alsomeasured, as shown in Figure 1.Figure 1 Terminology for cone penetrometers5

CPT Guide - 2006Cone Penetration Test (CPT)History1932The first cone penetrometer tests were made using a 35 mm outsidediameter gas pipe with a 15 mm steel inner push rod. A cone tip with a 10cm 2 projected area and a 60 o apex angle was attached to the steel innerpush rods, as shown in Figure 2.Figure 2 Early Dutch mechanical cone (After Sanglerat, 1972)1935Delf Soil Mechanics Laboratory designed the first manually operated 10ton cone penetration push machine, see Figure 3.Figure 3 Early Dutch mechanical cone (After Delft Geotechnics)6

CPT Guide - 2006Cone Penetration Test (CPT)1948The original Dutch mechanical cone was improved by adding a conicalpart just above the cone. The purpose of the geometry was to prevent soilfrom entering the gap between the casing and inner rods. The basic Dutchmechanical cones, shown in Figure 4, are still in use in some parts of theworld.Figure 4 Dutch mechanical cone penetrometer with conical mantle1953A friction sleeve (‘adhesion jacket’) was added behind the cone to includemeasurement of the local sleeve friction (Begemann, 1953), see Figure 5.Measurements were made every 8 inches (20 cm), and for the first time,friction ratio was used to classify soil type (see Figure 6).Figure 5 Begemann type cone with friction sleeve7

CPT Guide - 2006Cone Penetration Test (CPT)Figure 6 First soil classification for Begemann mechanical cone1965Fugro developed an electric cone, of which the shape and dimensionsformed the basis for the modern cones and the International ReferenceTest and ASTM procedure. The main improvements relative to themechanical cone penetrometers were:• Elimination of incorrect readings due to friction between inner rodsand outer rods and weight of inner rods.• Continuous testing with continuous rate of penetration without theneed for alternate movements of different parts of the penetrometerand no undesirable soil movements influencing the cone resistance.• Simpler and more reliable electrical measurement of coneresistance and sleeve friction.1974Cone penetrometers that could also measure pore pressure (piezocone)were introduced. Early design had various shapes and pore pressure filterlocations. Gradually the practice has become more standardized so thatthe recommended position of the filter element is close behind the cone atthe u 2 location. With the measurement of pore water pressure it becameapparent that it was necessary to correct the cone resistance for pore waterpressure effects (q t ), especially in soft clay.8

CPT Guide - 2006Cone Penetration Test (CPT)Test Equipment and ProceduresCone PenetrometersCone penetrometers come in a range of sizes with the 10 cm 2 and 15 cm 2probes the most common and specified in most standards. Figure 7 showsa range of cones from a mini-cone at 2 cm 2 to a large cone at 40 cm 2 . Themini cones are used for shallow investigations, whereas the large conescan be used in gravely soils.Figure 7 Range of CPT probes (from left: 2 cm 2 , 10 cm 2 , 15 cm 2 , 40 cm 2 )9

CPT Guide - 2006Cone Penetration Test (CPT)Additional Sensors/ModulesSince the introduction of the electric cone in the early 1960’s, manyadditional sensors have been added to the cone, such as;• Temperature• Geophones (seismic wave velocity)• Pressuremeter• Camera (visible light)• Radioisotope (gamma/neutron)• Electrical resistivity/conductivity• Dielectric• pH• Oxygen exchange (redox)• Laser/ultraviolet induced fluorescence (UVIF)• Membrane interface probe (MIP)The latter items are primarily for geo-environmental applications.One of the more common additional sensors is a geophone to allow themeasurement of seismic wave velocities. A schematic of the seismic CPT(SCPT) is shown in Figure 8.Figure 8 Schematic of Seismic CPT (SCPT)10

CPT Guide - 2006Cone Penetration Test (CPT)Pushing EquipmentPushing equipment consists of push rods, a thrust mechanism and areaction frame.On LandPushing equipment for land applications generally consist of speciallybuilt units that are either truck or track mounted. CPT’s can also becarried out using an anchored drill-rig. Figures 9 to 12 show a range of onland pushing equipment.Figure 9 Truck mounted 25 ton CPT unit11

CPT Guide - 2006Cone Penetration Test (CPT)Figure 10 Track mounted 20 ton CPT unitFigure 11 Small anchored drill-rig unit12

CPT Guide - 2006Cone Penetration Test (CPT)Figure 12 Ramset for CPT inside buildings or limited access13

CPT Guide - 2006Cone Penetration Test (CPT)Over WaterThere is a variety of pushing equipment for over water investigationsdepending on the depth of water. Floating or Jack-up barges are commonin shallow water (depth less than 80 feet), see Figures 13 and 14.Figure 13 Mid-size jack-up boatFigure 14 Quinn Delta ship with spuds14

CPT Guide - 2006Cone Penetration Test (CPT)Depth of PenetrationCPT’s can be performed to depths exceeding 300 feet (100m) in soft soilsand with large capacity pushing equipment. To improve the depth ofpenetration, the friction along the push rods should be reduced. This isnormally done by placing an expanded coupling (friction reducer) a shortdistance (typically 3 feet) behind the cone. Penetration will be limited ifeither very hard soils, gravel layers or rock are encountered. It iscommon to use 15 cm 2 cones to increase penetration depth, since 15 cm 2cones are more robust and have a slightly larger diameter than the 10 cm 2push rods.Test ProceduresPre-drillingFor penetration in fills or hard soils it may be necessary to pre-drill inorder to avoid damaging the cone. Pre-drilling, in certain cases, may bereplaced by first pre-punching a hole through the upper problem materialwith a solid steel dummy probe with a diameter slightly larger than thecone.VerticalityThe thrust machine should be set up so as to obtain a thrust direction asnear as possible to vertical. The deviation of the initial thrust directionfrom vertical should not exceed 2 degrees and push rods should bechecked for straightness. Modern cones have simple slope sensorsincorporated to enable a measure of the non-verticality of the sounding.This is useful to avoid damage to equipment and breaking of push rods.For depths less than 50 feet, significant non-verticality is unusual,provided the initial thrust direction is vertical.Reference MeasurementsModern cones have the potential for a high degree of accuracy andrepeatability (0.1% of full-scale output). Tests have shown that the zeroload output of the sensors can be sensitive to changes, for example15

CPT Guide - 2006Cone Penetration Test (CPT)temperature. It is common practice to record zero load readings of allsensors to track these changes.Rate of PenetrationThe standard rate of penetration is 2 cm per second, which isapproximately 1 inch per second. Hence, a 60 foot sounding can becompleted (start to finish) in about 30 minutes. The cone results aregenerally not sensitive to slight variations in the rate of penetration.Interval of readingsElectric cones produce continuous analogue data. However, most systemsconvert the data to digital form at selected intervals. Most standardsrequire the interval to be no more than 8 inches (200mm). In general,most systems collect data at intervals of between 1 to 2 inches (25 -50mm), with 2 inches being the most common.Dissipation TestsDuring a pause in penetration, any excess pore pressure generated aroundthe cone will start to dissipate. The rate of dissipation depends upon thecoefficient of consolidation, which in turn, depends on the compressibilityand permeability of the soil. The rate of dissipation also depends on thediameter of the probe. A dissipation test can be performed at any requireddepth by stopping the penetration and measuring the decay of porepressure with time. If equilibrium pore pressures are required, thedissipation test should continue until no further dissipation is observed.This can occur rapidly in sands, but may take many hours in plastic clays.Dissipation is faster for smaller cones.Calibration and MaintenanceCalibrations should be carried out at regular intervals (approximatelyevery 3 months). For major projects, calibrations should be carried outbefore and after the field work, with functional checks during the work.Functional checks should include recording and evaluating the zero loadmeasurements.16

CPT Guide - 2006Cone Penetration Test (CPT)With careful design, calibration, and maintenance, strain gauge load cellsand pressure transducers can have an accuracy and repeatability of betterthan +/- 0.2% of full scale reading.Table 3 shows a summary of checks and recalibrations for the CPTMaintenanceStartofProjectStart ofTestEnd ofTestEnd ofDayOnce aMonthEvery 3monthsWear x x xO-ring seals x xPush-rods x xPorepressure-filterxxCalibrationxComputerConexxZero-load x xCables x xTable 3 Summary of checks and recalibrations for the CPTPore water effectsIn soft clays and silts and in over water work, the measured q c must becorrected for pore water pressures acting on the cone geometry, thusobtaining the corrected cone resistance, q t :q t = q c + u 2 (1 – a n )where a n is the net area ratio determined from laboratory calibration witha typical value of around 0.85.17

CPT Guide - 2006Cone Penetration Test (CPT)CPT InterpretationNumerous semi-empirical correlations have been developed to estimategeotechnical parameters from the CPT for a wide range of soils. Thesecorrelations vary in their reliability and applicability. Because the CPThas additional sensors (e.g. pore pressure-CPTU and seismic-SCPT), theapplicability to estimate soil parameters varies. Since CPT with porepressure measurements (CPTU) is commonly available, Table 4 shows anestimate of the perceived applicability of the CPTU to estimate soilparameters. If seismic is added, the ability to estimate soil stiffness (E, G& G o ) improves further.SoilTypeD r Ψ K o OCR S t s u φ E,G * M G 0*kc hSand 2-3 2-35 2-33-4 2-3 3 3-4Clay 2 1 2 1-24 3-4 4 3-4 2-32-3Table 4 Perceived applicability of CPTU for deriving soil parameters1=high, 2=high to moderate, 3=moderate, 4=moderate to low, 5=low reliability, Blank=no applicability,* improved with SCPTWhere:D r Relative density φ Friction angleΨ State Parameter K 0 In-situ stress ratioE, G Young’s and Shear moduli G 0 Small strain shear moduliOCR Over consolidation ratio M (or m v ) Compressibilitys u Undrained shear strength S t Sensitivityc H Coefficient of consolidation k Permeability18

CPT Guide - 2006Cone Penetration Test (CPT)Soil Profiling and ClassificationThe major application of the CPT is for soil profiling and classification.Typically, the cone resistance, (q t ) is high in sands and low in clays, andthe friction ratio (R f = f s /q t ) is low in sands and high in clays. CPTclassification charts cannot be expected to provide accurate predictions ofsoil type based on grain size distribution but provide a guide to themechanical characteristics of the soil, or the soil behavior type (SBT).CPT data provides a repeatable index of the aggregate behavior of the insitusoil in the immediate area of the probe. Hence, prediction of soil typebased on CPT is referred to as soil behavior type (SBT).Non-Normalized ChartsThe most commonly used CPT soil behavior type method is the chartsuggested by Robertson et al. (1986) and shown in Figure 15. This chartuses the basic CPT parameters of cone resistance, q t and friction ratio, R f .The chart is global in nature and can provide reasonable predictions ofsoil behavior type for CPT soundings up to about 60ft (20m) in depth.The chart identifies general trends in ground response, such as, increasingrelative density (D r ) for sandy soils, increasing stress history (OCR), soilsensitivity (S t ) and void ratio (e) for cohesive soils. Overlap in some zonesshould be expected and the zones should be adjusted somewhat based onlocal experience.Normalized ChartsSince both the penetration resistance and sleeve friction increase withdepth due to the increase in effective overburden stress, the CPT datarequires normalization for overburden stress for very shallow and/or verydeep soundings.A popular CPT soil classification chart based on normalized CPT data isthat proposed by Robertson (1990) and shown in Figure 16. A zone hasbeen identified in which the CPT results for most young, un-cemented,insensitive, normally consolidated soils will fall. Again the chart is globalin nature and provides only a guide to soil behavior type (SBT). Overlapin some zones should be expected and the zones should be adjustedsomewhat based on local experience.19

CPT Guide - 2006Cone Penetration Test (CPT)Zone123456789101112Soil Behavior TypeSensitive fine grainedOrganic materialClaySilty Clay to clayClayey silt to silty claySandy silt to clayey siltSilty sand to sandy siltSand to silty sandSandGravelly sand to sandVery stiff fine grained*Sand to clayey sand** Overconsolidated or cemented1 MPa = 10 tsfFigure 15 CPT Soil Behavior Type (SBT) chart(Robertson et al., 1986).20

CPT Guide - 2006Cone Penetration Test (CPT)Zone Soil Behavior Type I c1 Sensitive, fine grained N/A2 Organic soils – peats > 3.63 Clays – silty clay to clay 2.95 – 3.64 Silt mixtures – clayey silt to 2.60 – 2.95silty clay5 Sand mixtures – silty sand to 2.05 – 2.6sandy silt6 Sands – clean sand to silty 1.31 – 2.05sand7 Gravelly sand to dense sand < 1.318 Very stiff sand to clayey sand* N/A9 Very stiff, fine grained* N/A* Heavily overconsolidated or cementedFigure 16 Normalized CPT Soil Behavior Type (SBT N ) chart(Robertson, 1990).21

CPT Guide - 2006Cone Penetration Test (CPT)If no prior CPT experience exists in a given geologic environment it isadvisable to obtain samples from appropriate locations to verify theclassification and soil behavior type. If significant CPT experience isavailable and the charts have been modified based on this experiencesamples may not be required.Soil classification can be improved if pore pressure data is also collected.In soft clay the penetration pore pressures can be very large, whereas, instiff heavily over-consolidated clays or dense silts and silty sands thepenetration pore pressures can be small and sometimes negative relativeto the equilibrium pore pressures (u 0 ). Also the rate of pore pressuredissipation during a pause in penetration can guide in the soil type. Insandy soils any excess pore pressures will dissipate very quickly, whereas,in clays the rate of dissipation can be very slow.To simplify the application of the CPT SBT N chart shown in Figure 16,the normalized cone parameters Q t and F r can be combined into one SoilBehavior Type index, I c , where I c is the radius of the essentiallyconcentric circles that represent the boundaries between each SBT zone.I c can be defined as follows;I c = ((3.47 - log Q t ) 2 + (log F r + 1.22) 2 ) 0.5where:Q t = the normalized cone penetration resistance (dimensionless)= (q t – σ vo )/σ' voF r = the normalized friction ratio, in %= (f s /(q t – σ vo )) x 100%The boundaries of soil behavior types are then given in terms of the index,I c , as shown in Figure 16. The soil behavior type index does not apply tozones 1, 8 and 9. Profiles of I c provide a simple guide to the continuousvariation of soil behavior type in a given soil profile based on CPT results.Independent studies have shown that the normalized SBT N chart shown inFigure 16 typically has greater than 80% reliability when compared tosamples.22

CPT Guide - 2006Cone Penetration Test (CPT)Equivalent SPT N 60 ProfilesThe Standard Penetration Test (SPT) is one of the most commonly usedin-situ tests in many parts of the world, especially North America.Despite continued efforts to standardize the SPT procedure and equipmentthere are still problems associated with its repeatability and reliability.However, many geotechnical engineers have developed considerableexperience with design methods based on local SPT correlations. Whenthese engineers are first introduced to the CPT they initially prefer to seeCPT results in the form of equivalent SPT N-values. Hence, there is aneed for reliable CPT/SPT correlations so that CPT data can be used inexisting SPT-based design approaches.There are many factors affecting the SPT results, such as boreholepreparation and size, sampler details, rod length and energy efficiency ofthe hammer-anvil-operator system. One of the most significant factors isthe energy efficiency of the SPT system. This is normally expressed interms of the rod energy ratio (ERr). An energy ratio of about 60% hasgenerally been accepted as the reference value which represents theapproximate historical average SPT energy.A number of studies have been presented over the years to relate the SPTN value to the CPT cone penetration resistance, q c . Robertson et al.(1983) reviewed these correlations and presented the relationship shownin Figure 16 relating the ratio (q c /p a )/N 60 with mean grain size, D 50(varying between 0.001mm to 1mm). Values of q t are madedimensionless when dividing by the atmospheric pressure (p a ) in the sameunits as q c . It is observed that the ratio increases with increasing grainsize.The values of N used by Robertson et al. correspond to an average energyratio of about 60%. Hence, the ratio applies to N 60, as shown on Figure17. Other studies have linked the ratio between the CPT and SPT withfines content for sandy soils.23

CPT Guide - 2006Cone Penetration Test (CPT)Figure 17 CPT-SPT correlations with mean grain size(Robertson et al., 1983)The above correlations need the soil grain size information to determinethe mean grain size (or fines content). Grain characteristics can beestimated directly from CPT results using soil classification or soilbehavior type (SBT) charts. The CPT SBT charts show a clear trend ofincreasing friction ratio with increasing fines content and decreasing grainsize. Robertson et al. (1986) suggested (q c /p a )/N 60 ratios for each soilbehavior type zone using the non-normalized CPT chart. The suggestedratio for each soil behavior type is given in Table 5.These values provide a reasonable estimate of SPT N 60 values from CPTdata. For simplicity the above correlations are given in terms of q c . Forfine grained soft soils the correlations should be applied to total coneresistance, q t .One disadvantage of this simplified approach is the somewhatdiscontinuous nature of the conversion. Often a soil will have CPT datathat crosses different soil behavior type zones and hence producesdiscontinuous small changes in predicted SPT N 60 values.24

CPT Guide - 2006Cone Penetration Test (CPT)ZoneSoil Behavior Type( q / p )1 sensitive fine grained 22 organic soils - peats 13 clay 14 silty clay to clay 1.55 clayey silt to silty clay 26 sandy silt to clayey silt 2.57 silty sand to sandy silt 38 sand to silty sand 49 sand 510 dense sand to gravelly6sand11 very stiff fine grained 1cN60aTable 5 Suggested (q c /p a )/N 60 ratios.Jefferies and Davies (1993) suggested the application of a soil behaviortype index, I c to link with the CPT-SPT correlation. The soil behaviortype index, I c , can be combined with the CPT-SPT ratios to give thefollowing relationship:(qcN/p60a)⎞= 8.5 ⎜⎛ I − c1 ⎟⎝ 4.6 ⎠Jefferies and Davies (1993) suggested that the above approach canprovide a better estimate of the SPT N-values than the actual SPT test dueto the poor repeatability of the SPT.In very loose soils the weight of the rods and hammer can dominate theSPT penetration resistance and produce very low N-values, which canresult in high (q c /p a )/N 60 ratios due to the low SPT N-values measured.25

CPT Guide - 2006Cone Penetration Test (CPT)Undrained Shear Strength (s u )No single value of undrained shear strength exists, since the undrainedresponse of soil depends on the direction of loading, soil anisotropy, strainrate, and stress history. Typically the undrained strength in tri-axialcompression is larger than in simple shear which is larger than tri-axialextension (s uTC > s uSS > s uTE ). The value of s u to be used in analysistherefore depends on the design problem. In general, the simple sheardirection of loading often represents the average undrained strength.Since anisotropy and strain rate will inevitably influence the results of allin-situ tests, their interpretation will necessarily require some empiricalcontent to account for these factors, as well as possible effects of sampledisturbance.Recent theoretical solutions have provided some valuable insight into theform of the relationship between cone resistance and s u . All theoriesresult in a relationship between cone resistance and s u of the form:s u =qt− σNktvTypically N kt varies from 10 to 20, with 15 as an average. N kt tends toincrease with increasing plasticity and decrease with increasing soilsensitivity.For deposits where little experience is available, estimate s u using the totalcone resistance (q t ) and preliminary cone factor values (N kt ) from 15 - 20.For a more conservative estimate, select a value close to the upper limit.In very soft clays where there may be some uncertainty with the accuracyin q t , estimates of s u can be made from the excess pore pressure (Δu)measured behind the cone (u 2 ) using the following:s u =ΔuNΔu26

CPT Guide - 2006Cone Penetration Test (CPT)Where N Δu varies from 7 to 10. For a more conservative estimate, select avalue close to the upper limit.If previous experience is available in the same deposit, the valuessuggested above should be adjusted to reflect this experience.For larger, moderate to high risk projects, where high quality field andlaboratory data may be available, site specific correlations should bedeveloped based on appropriate and reliable values of s u .Soil SensitivityThe sensitivity (S t ) of clay is defined as the ratio of undisturbed undrainedshear strength to totally remolded undrained shear strength.S t =ssuu(Remolded)=qt− σNktv(1 / f s )The remolded undrained shear strength can be assumed equal to thesleeve friction stress, f s . Therefore, the sensitivity of a clay can beestimated by calculating the peak s u from either site specific or generalcorrelations with q t or Δu and s u(Remolded) from f s .For relatively sensitive clays (S t > 5), the value of f s can be very low withinherent difficulties in accuracy. Hence, the estimate of sensitivity shouldbe used as a guide only.27

CPT Guide - 2006Cone Penetration Test (CPT)Undrained Shear Strength Ratio (s u /σ’ v )It is often useful to estimate the undrained shear strength ratio from theCPT, since this relates directly to overconsolidation ratio (OCR). CriticalState Soil Mechanics presents a relationship between the undrained shearstrength ratio for normally consolidated clays under difference directionsof loading and the effective stress friction angle, φ’. Hence, a betterestimate of undrained shear strength can be obtained with knowledge ofthe friction angle. For normally consolidated clays;From the CPT:(S u /σ v ’) NC = 0.22 in direct simple shear (φ’ = 30 o )⎛ qt− σvo(S u /σ v ’) = ⎟ ⎞⎜⎝ σ'vo ⎠(1/N kt ) = Q t / N ktSince N kt ~ 15(S u /σ v ’) ~ 0.067 Q tFor a normally consolidated clay where (S u /σ' v ’) NC = 0.22;(Q t ) NC ~ 3.3Based on the assumption that the sleeve friction measures the remoldedshear strength, S uRemolded = f sTherefore:S u /σ v ’ = f s /σ v ’ = (F . Q) / 100Hence, it is possible to represent (S u /σ v ’) contours on the normalizedSBT N chart (Figure 16). These contours represent OCR for insensitiveclays with high values of (S u /σ v ’) and Sensitivity for low values of (S u/σ v ’).28

CPT Guide - 2006Cone Penetration Test (CPT)Stress History - Overconsolidation Ratio (OCR)Overconsolidation ratio (OCR) is defined as the ratio of the maximumpast effective consolidation stress and the present effective overburdenstress:OCR =σ'σ'pvoFor mechanically overconsolidated soils where the only change has beenthe removal of overburden stress, this definition is appropriate. However,for cemented and/or aged soils the OCR may represent the ratio of theyield stress and the present effective overburden stress. The yield stresswill depend on the direction and type of loading.For overconsolidated clays:(S u /σ v ’) OC = (S u /σ v ’) NC (OCR) 0.8(S u /σ v ’) DSS = 0.22 (OCR) 0.8 (φ’ = 30 o )The easiest method to estimate OCR in cohesive soils is:OCR = k⎛ qt− σ⎜⎝ σ'vovo⎟ ⎞⎠= k Q tAn average value of k = 0.3 can be assumed, with an expected range of0.2 to 0.5. Higher values of k are recommended in aged, heavilyoverconsolidated clays. If previous experience is available in the samedeposit, the values of k should be adjusted to reflect this experience and toprovide a more reliable profile of OCR.For larger, moderate to high-risk projects, where additional high qualityfield and laboratory data may be available, site-specific correlationsshould be developed based on consistent and relevant values of OCR.29

CPT Guide - 2006Cone Penetration Test (CPT)The estimated OCR is influenced by soil sensitivity, pre-consolidationmechanism, soil type and local heterogeneity.In-Situ Stress Ratio (Ko)There is no reliable method to determine K o from CPT. However, anestimate can be made based on an estimate of OCR, as shown in Figure18.Kulhawy and Mayne (1990) suggested a similar approach, using:K o = 0.1⎛ qt− σ⎜⎝ σ'vovo⎟ ⎞⎠These approaches are generally limited to mechanically overconsolidatedsoils. Considerable scatter exists in the database used for thesecorrelations and therefore they must be considered only as a guide.Figure 18 OCR and K o from s u /σ’ vo and I p (after Andresen et al., 1979)30

CPT Guide - 2006Cone Penetration Test (CPT)Friction AngleThe shear strength of uncemented, cohesionless soil is usually expressedin terms of a peak secant friction angle, φ'.Numerous studies have been published for assessing φ' from the CPT inclean sands and basically the methods fall into one of three categories:• Bearing capacity theory• Cavity expansion theory• Empirical, based on calibration chamber testsSignificant advances have been made in the development of theories tomodel the cone penetration process in sands (Yu and Mitchell, 1998).Cavity expansion models show the most promise since they are relativelysimple and can incorporate many of the important features of soilresponse. However, empirical correlations based on calibration chambertest results are still the most commonly used.A review of calibration chamber test results was made by Robertson andCampanella (1983) to compare cone resistance to measured peak secantfriction angle. The peak secant friction angle was measured in drainedtriaxial compression tests performed at the confining stress approximatelyequal to the horizontal stresses in the calibration chamber before the CPT.The recommended correlation for uncemented, unaged, moderatelycompressible, predominately quartz sands proposed by Robertson andCampanella (1983) is shown in Figure 19. For sands of highcompressibility (i.e. carbonate sands or sands with high mica content), thechart will tend to predict low friction angles.31

CPT Guide - 2006Cone Penetration Test (CPT)Note: 0.1MPa = 100 kPa = 1 bar ≈ 1 tsf ≈ 1 kg/cm 2tan φ' =12.68⎡ ⎛ q⎢log⎜⎣ ⎝ σ'cvo⎞ ⎤⎟ + 0.29⎥ ⎠ ⎦Figure 19 Friction angle, φ', from CPT in uncemented silica sand( Robertson and Campanella, 1983)32

CPT Guide - 2006Cone Penetration Test (CPT)Relative density (D r )For cohesionless soils, the density, or more commonly, the relativedensity or density index, is often used as an intermediate soil parameter.Relative density, D r , or density index, I D , is defined as:where:I D = D r =eemaxmax− e− emine max and e min are the maximum and minimum void ratios and e isthe in-situ void ratio.The problems associated with the determination of e max and e min are wellknown. Also, research has shown that the stress strain and strengthbehavior of cohesionless soils is too complicated to be represented byonly the relative density of the soil. However, for many years relativedensity has been used by engineers as a parameter to describe sanddeposits.Research using large calibration chambers has provided numerouscorrelations between CPT penetration resistance and relative density forclean, predominantly quartz sands. The calibration chamber studies haveshown that the CPT resistance is controlled by sand density, in-situvertical and horizontal effective stress and sand compressibility. Sandcompressibility is controlled by grain characteristics, such as grain size,shape and mineralogy. Angular sands tend to be more compressible thanrounded sands as do sands with high mica and/or carbonate comparedwith clean quartz sands. More compressible sands give a lowerpenetration resistance for a given relative density then less compressiblesands.Based on extensive calibration chamber testing on Ticino sand, Baldi etal. (1986) recommended a formula to estimate relative density from q c . Amodified version of this formula, to obtain D r from q c1 is as follows:D r =⎛⎜⎝1C⎞ ⎛ q⎟ln⎜⎠ ⎝c12C0⎟ ⎞⎠33

CPT Guide - 2006Cone Penetration Test (CPT)where:C 0 and C 2 are soil constantsσ' vo = effective vertical stressq c1 = (q c / p a ) / (σ' vo /p a ) 0.5= normalized CPT resistance, corrected for overburdenpressurep a = reference pressure of 1 tsf, in same units as q c and σ' voq c = cone penetration resistanceFor moderately compressible, normally consolidated, unaged anduncemented, predominantly quartz sands the constants are: C o = 15.7 andC 2 = 2.41.Kulhawy and Mayne (1990) suggested a simpler formula for estimatingrelative density:D r 2 =305QCqc1QOCRQAwhere:q c1 and p a are as defined aboveQ C = Compressibility factor ranges from 0.91 (low compress.) to1.09 (high compress.)Q OCR = Overconsolidation factor = OCR 0.18Q A = Aging factor = 1.2 + 0.05log(t/100)A constant of 350 is more reasonable for medium, clean, uncemented,unaged quartz sands that are about 1,000 years old. The constant is closerto 300 for fine sands and closer to 400 for coarse sands.The equation can then be simplified to:D r 2 = q c1 / 35034

CPT Guide - 2006Cone Penetration Test (CPT)Stiffness and ModulusCPT data can be used to estimate modulus for subsequent use in elastic orsemi-empirical settlement prediction methods. However, correlationsbetween q c and moduli (E) are sensitive to stress and strain history, agingand sand mineralogy.A useful guide for estimating Young's moduli for uncementedpredominantly silica sands is given in Figure 20. The modulus has beendefined as that mobilized at 0.1% strain. For more heavily loadedconditions (i.e. larger strain) the modulus would decrease. The results forNC sands are applicable for young recent fills with an age less than 10years and the results for Aged NC sands are applicable for natural sandswith an age greater than 1000 years.Figure 20 Evaluation of drained Young's modulus from CPTfor silica sands35

CPT Guide - 2006Cone Penetration Test (CPT)Modulus from Shear Wave VelocityA major advantage of the seismic CPT is the additional measurement ofthe shear wave velocity, V s . The shear wave velocity is measured using adownhole technique during pauses in the CPT resulting in a continuousprofile of V s . Elastic theory states that the small strain shear modulus, G ocan be determined from:G o = ρ V s2where ρ is the mass density of the soil (ρ = γ/g).Hence, the addition of shear wave velocity during the CPT provides adirect measure of soil stiffness.The small strain shear modulus represents the elastic stiffness of the soilsat shear strains (γ) less than 10 -4 percent. Elastic theory also states that thesmall strain Young’s modulus, E o is linked to G o , as follows;E o = 2(1 + υ)G oWhere υ is Poisson’s ratio, which ranges from 0.1 to 0.3 for most soils.Application to engineering problems requires that the small strainmodulus be softened to the appropriate strain level. For most welldesigned structures the degree of softening is often close to a factor of 3.Hence, for many applications the equivalent Young’s modulus (E s ’ ) canbe estimated from;E s ’ ~ G o = ρ V s2The shear wave velocity can also be used directly for the evaluation ofliquefaction potential. Hence, the seismic CPT provides two independentmethods to evaluate liquefaction potential.36

CPT Guide - 2006Cone Penetration Test (CPT)Identification of Unusual Soils Using the SCPTAlmost all available empirical correlations to interpret in-situ tests assumethat the soil is well behaved, i.e. similar to the soils in which thecorrelation was based. Many of the existing correlations apply to soilssuch as, unaged, uncemented, silica sands. Application of the existingempirical correlations in sands other than unaged and uncemented canproduce incorrect interpretations. Hence, it is important to be able toidentify if the soils are ‘well behaved’. The combined measurement ofshear wave velocity and cone resistance provides an opportunity toidentify these ‘unusual’ soils. The cone resistance (q c ) is a good measureof soil strength, since the cone is inducing very large strains and the soiladjacent to the probe is at failure. The shear wave velocity (V s ) is a directmeasure of the small strain soil stiffness (G o ), since the measurement ismade at very small strains. Recent research has shown that unaged anduncemented sands have data that falls within a narrow range of combinedq c and G o , as shown in Figure 21 and the following equations.Upper bound, unaged & cemented G o = 280 (q c σ' vo p a ) 0.3Lower bound, unaged & cemented G o = 110 (q c σ' vo p a ) 0.3Figure 21 Characterization of uncemented, unaged sands(after Eslaamizaad and Robertson, 1997)37

CPT Guide - 2006Cone Penetration Test (CPT)Hydraulic Conductivity (k)An approximate estimate of soil hydraulic conductivity or coefficient ofpermeability, k, can be made from an estimate of soil behavior type usingthe CPT SBT charts. Table 6 provides estimates based on the nonnormalizedchart shown in Figure 15, while Table 7 provides estimatesbased on the normalized chart shown in Figure 16. These estimates areapproximate at best, but can provide a guide to variations of possiblepermeability.Zone Soil Behavior Type (SBT) Range of permeabilityk (m/s)1 Sensitive fine grained 3x10 -9 to 3x10 -82 Organic soils 1x10 -8 to 1x10 -63 Clay 1x10 -10 to 1x10 -94 Silty clay to clay 1x10 -9 to 1x10 -85 Clayey silt to silty clay 1x10 -8 to 1x10 -76 Sandy silt to clayey silt 1x10 -7 to 1x10 -67 Silty sand to sandy silt 1x10 -5 to 1x10 -68 Sand to silty sand 1x10 -5 to 1x10 -49 Sand 1x10 -4 to 1x10 -310 Gravelly sand to dense sand 1x10 -3 to 111 Very stiff fine-grained soil 1x10 -8 to 1x10 -612 Very stiff sand to clayey sand 3x10 -7 to 3x10 -4Table 6 Estimation of soil permeability (k) from the non-normalizedCPT SBT chart by Robertson et al. (1986) shown in Figure 15Baligh and Levadoux (1980) recommended that the horizontal coefficientof permeability can be estimated from the expression:k h =⎛ γw⎜⎝ 2.3 σ'vo⎟ ⎞⎠RR c hwhere RR is the re-compression ratio in the overconsolidated range. Itrepresents the strain per log cycle of effective stress during recompressionand can be determined from laboratory consolidation tests. Baligh andLevadoux recommended that RR should range from 0.5x10 -2 to 2x10 -2 .38

CPT Guide - 2006Cone Penetration Test (CPT)ZoneSoil Behavior Type(SBT N )Range of permeabilityk (m/s)1 Sensitive fine grained 3x10 -9 to 3x10 -82 Organic soils 1x10 -8 to 1x10 -63 Clay 1x10 -10 to 1x10 -94 Silt mixtures 3x10 -9 to 1x10 -75 Sand mixtures 1x10 -7 to 1x10 -56 Sands 1x10 -5 to 1x10 -37 Gravelly sands to dense sands 1x10 -3 to 18 Very stiff sand to clayey sand 1x10 -8 to 1x10 -69 Very stiff fine-grained soil 1x10 -8 to 1x10 -6Table 7 Estimation of soil permeability (k) from the normalized CPTSBT N chart by Robertson (1990) shown in Figure 16Robertson et al. (1992) presented a summary of available data to estimatethe horizontal coefficient of permeability from dissipation tests, and isshown in Figure 22. Since the relationship is also a function of therecompression ratio (RR) there is a wide variation of + or – one order ofmagnitude. Jamiolkowski et al. (1985) suggested a range of possiblevalues of k h /k v for soft clays as shown in Table 8.39

CPT Guide - 2006Cone Penetration Test (CPT)Figure 22 Summary of data for estimating horizontal coefficient ofpermeability from dissipation tests (after Robertson et al., 1992).Nature of clayNo macrofabric, or only slightly developedmacrofabric, essentially homogeneous depositsFrom fairly well- to well-developedmacrofabric, e.g. sedimentary clays withdiscontinuous lenses and layers of morepermeable materialVarved clays and other deposits containingembedded and more or less continuouspermeable layersk h /k v1 to 1.52 to 43 to 15Table 8 Range of possible field values of k h /k v for soft clays(after Jamiolkowski et al., 1985)40

CPT Guide - 2006Cone Penetration Test (CPT)Consolidation CharacteristicsFlow and consolidation characteristics of a soil are normally expressed interms of the coefficient of consolidation, c, and hydraulic conductivity, k.They are inter-linked through the formula:c =k Mγwwhere M is the constrained modulus relevant to the problem (i.e.unloading, reloading, virgin loading).The parameters c and k vary over many orders of magnitude and are someof the most difficult parameters to measure in geotechnical engineering.It is often considered that accuracy within one order of magnitude isacceptable. Due to soil anisotropy, both c and k have different values inthe horizontal (c h , k h ) and vertical (c v , k v ) direction. The relevant designvalues depend on drainage and loading direction.Details on how to estimate k from CPT soil classification charts are givenin another section.The coefficient of consolidation can be estimated by measuring thedissipation or rate of decay of pore pressure with time after a stop in CPTpenetration. Many theoretical solutions have been developed for derivingthe coefficient of consolidation from CPT pore pressure dissipation data.The coefficient of consolidation should be interpreted at 50% dissipation,using the following formula:where:c =⎛ T⎜⎝ t5050⎟ ⎞⎠r o2T 50 = theoretical time factort 50 = measured time for 50% dissipationr o = penetrometer radius41

CPT Guide - 2006Cone Penetration Test (CPT)It is clear from this formula that the dissipation time is inverselyproportional to the radius of the probe. Hence, in soils of very lowpermeability, the time for dissipation can be decreased by using smallerprobes.Robertson et al. (1992) reviewed dissipation data from around the worldand compared the results with the leading theoretical solution by Teh andHoulsby (1991), as shown in Figure 23.Figure 23 Average laboratory c h values and CPTU results(Robertson et al., 1992)The review showed that the theoretical solution provided reasonableestimates of c h . The solution shown in Figure 23 applies to pore pressuresensors located just behind the cone tip (i.e. u 2 ).42

CPT Guide - 2006Cone Penetration Test (CPT)The ability to estimate c h from CPT dissipation results is controlled bysoil stress history, sensitivity, anisotropy, rigidity index (relativestiffness), fabric and structure. In overconsolidated soils, the porepressure behind the cone tip can be low or negative, resulting indissipation data that can initially rise before a decay to the equilibriumvalue. In these cases, the pore pressure sensor can be moved to the faceof the cone or the t 50 time can be estimated using the maximum porepressure as the initial value. Care is required to ensure that the dissipationis continued to the correct equilibrium and not stopped prematurely afterthe initial rise.Based on available experience, the CPT dissipation method shouldprovide estimates of c h to within + or – half an order of magnitude.However, the technique is repeatable and provides an accurate measure ofchanges in consolidation characteristics within a given soil profile.An approximate estimate of the coefficient of consolidation in the verticaldirection can be obtained using the ratios of permeability in the horizontaland vertical direction given in the section on hydraulic conductivity,since:c v = c h⎛ k⎜⎝ kvh⎞⎟⎠Table 8 can be used to provide an estimate of the ratio of hydraulicconductivities.43

CPT Guide - 2006Cone Penetration Test (CPT)Applications of CPT ResultsThe previous sections have described how CPT results can be used toestimate geotechnical parameters which can be used as input in analyses.An alternate approach is to apply the in-situ test results directly to anengineering problem. A typical example of this approach is theevaluation of pile capacity directly from CPT results without the need forsoil parameters.As a guide, Table 9 shows a summary of the applicability of the CPT fordirect design applications. The ratings shown in the table have beenassigned based on current experience and represent a qualitativeevaluation of the confidence level assessed to each design problem andgeneral soil type. Details of ground conditions and project requirementscan influence these ratings.In the following sections a number of direct applications of CPT/CPTUresults are described. These sections are not intended to provide fulldetails of geotechnical design, since this is beyond the scope of this guide.However, they do provide some guidelines on how the CPT can beapplied to many geotechnical engineering applications. A good referencefor foundation design is the Canadian Foundation Engineering Manual(CFEM, 1993).Type of soilPiledesignBearingcapacitySettlementCompactioncontrolLiquefactionSand 1 – 2 1 – 2 2 – 3 1 – 2 1 – 2Clay 1 – 2 1 – 2 3 – 4 3 – 4 1 – 2Intermediate soils 1 – 2 2 – 3 3 – 4 2 – 3 1– 2Reliability rating: 1=High; 2=High to moderate; 3=Moderate; 4=Moderate to low; 5=lowTable 9 Perceived applicability of the CPT/CPTU for various directdesign problems44

CPT Guide - 2006Cone Penetration Test (CPT)Shallow Foundation DesignGeneral Design PrinciplesTypical Design Sequence:1. Select minimum depth to protect against:• external agents: e.g. frost, erosion, trees• poor soil: fill, organic soils, etc.2. Define minimum area necessary to protect against soil failure:• perform bearing capacity analyses2. Compute settlement and check if acceptable3. Modify selected foundation if required.Typical Shallow Foundation ProblemsStudy of 1200 cases of foundation problems in Europe showed that theproblems could be attributed to the following causes:• 25% footings on recent fill (mainly poor engineering judgment)• 20% differential settlement (50% could have been avoided with gooddesign)• 20% effect of groundwater• 10% failure in weak layer• 10% nearby work(excavations, tunnels, etc.)• 15% miscellaneous causes(earthquake, blasting, etc.)In design, settlement is generally the critical issue. Hence, bearingcapacity is generally not of great importance.45

CPT Guide - 2006Cone Penetration Test (CPT)ConstructionConstruction details can alter significantly the conditions assumed in thedesign.Examples are provided in the following list:• During Excavation• bottom heave• slaking, swelling, and softening of expansive clays or rock• piping in sands and silts• remolding of silts and sensitive clays• disturbance of granular soils• Adjacent construction activity• groundwater lowering• excavation• pile driving• blasting• Other effects during or following construction• reversal of bottom heave• scour, erosion and flooding• frost action46

CPT Guide - 2006Cone Penetration Test (CPT)Shallow Foundation - Bearing CapacityGeneral PrinciplesLoad-settlement relationships for typical footings (Vesic, 1972):1. Approximate elastic response2. Progressive development of local shear failure3. General shear failureIn dense granular soils failure typically occurs along a well defined failuresurface. In loose granular soils, volumetric compression dominates andpunching failures common. Increased depth of overburden can change adense sand to behave more like loose sand. In (homogeneous) cohesivesoils, failure occurs along an approximately circular surface.Significant parameters are:• nature of soils• density and resistance of soils• width and shape of footing• depth of footing• position of load.A given soil does not have a unique bearing capacity; the bearing capacityis a function of the footing shape, depth and width as well as loadeccentricity.General Bearing Capacity TheoryInitially developed by Terzaghi (1936); now over 30 theories with thesame general form, as follows:Ultimate bearing capacity, (q f ):where:q f = 0.5 γ B N γ s γ i γ + c N c s c i c + γ D N q s q i qN γ N c N q = Bearing capacity coefficients (function of φ')s γ s c s q = Shape factors (function of B/L)47

CPT Guide - 2006Cone Penetration Test (CPT)i γ i c i qBDL= Load inclination factors= width of footing= depth of footing= length of footingComplete rigorous solutions are impossible since stress fields areunknown. All theories differ in simplifying assumptions made to writethe equations of equilibrium. No single solution is correct for all cases.Shape FactorsShape factors are applied to account for 3-D effects. Based on limitedtheoretical ideas and some model tests, recommended factors are asfollows:s c = s q = 1 +s γ = 1 - 0.4⎛ B ⎞⎛N⎜ ⎟⎜⎝ L ⎠⎝N⎛ B ⎞⎜ ⎟⎝ L ⎠qc⎟ ⎞⎠Load Inclination FactorsWhen load is inclined (δ), shape of failure surface changes and reducedarea of failure. Hence, reduced resistance. At the limit of inclination, δ =φ, q f = 0, since slippage can occur along footing-soil interface.In general:i c = i q =i g =⎛ δ ⎞⎜1 −ο⎟⎝ 90 ⎠⎛ δ ⎞⎜1 − ⎟⎝ φ ⎠22For an eccentric load, Terzaghi proposed a simplified concept of anequivalent footing width, B'.B' = B - 2 e48

CPT Guide - 2006Cone Penetration Test (CPT)where e is the eccentricity. For combined inclined and eccentric load, useB' and relevant values of shape factors. For footings near a slope, usemodified bearing capacity factors (see Bowles, 1982). They will be smallfor clay but large for granular soils.Effect of GroundwaterThe bearing capacity is based on effective stress analyses, hence, positionof the groundwater table affects the value of the soil unit weight.• If depth to the water table, d w = 0, use γ' in both terms• If d w = D (depth of footing), use γ' in width term and γ in depth term.• In general, install drainage to keep d w > D.Indirect Methods Based on Soil ParametersGranular SoilsBearing capacity generally not calculated, since settlements control,except for very narrow footings.Cohesive SoilsShort term stability generally controls; hence application of undrainedshear strength, s u .where:q f = N c s u + γ DN c = function of footing width and shape; for strip footings at the groundsurface, N c = (π + 2).s u = apply Bjerrum’s correction, based on past experience, to field vaneshear strength or from CPT.Allowable bearing capacity:q all= q f - γ D49

CPT Guide - 2006Cone Penetration Test (CPT)Hence, q all =N cs uFSWhere: FS is Factor of Safety, typically = 3.0.High FS to account for limitations in theory, underestimation of loads,overestimation of soil strength, avoid local yield in soil and keepsettlements small.Direct Approach to Estimate Bearing Capacity (In-Situ Tests)Based on in-situ tests, theory, model tests and past foundationperformance.SPT• Empirical direct methods• Limited to granular soils, however, sometimes applied to very stiff clays• Often linked to allowable settlement of 25mm (Terzaghi & Peck)• SPT of poor reliability, hence, empirical methods tend to be veryconservativeCPTEmpirical direct methods.Granular soils:q f = K φ q c (av)where:q c (av) = average CPT penetration resistance below depth of footing, z =BEslaamizaad & Robertson (1996) suggested K φ = 0.16 to 0.30 dependingon B/D and shape. In general, assume K φ = 0.16 (see Figure 24).Meyerhof (1956) suggested K φ = 0.30. However, generally settlementswill control.50

CPT Guide - 2006Cone Penetration Test (CPT)Figure 24 Correlation between bearing capacity of footing oncohesionless soils and average cone resistance(Eslaamizaad and Robertson, 1996)Cohesive soils:q f = K su q c (av) + γ DK su = 0.30 to 0.60 depending on footing B/D and shape and soil OCR andsensitivity. In general, assume K su = 0.30 in clayShallow Foundation Design – SettlementGeneral Design PrinciplesRequires:• magnitude of settlement• rate of settlement• compatibility with acceptable behavior of building51

CPT Guide - 2006Cone Penetration Test (CPT)For well designed foundations, the magnitude of strains in the ground isgenerally very small (ε 4), approximately 50%of settlement can be due to immediate distortion settlement and secondarysettlement is generally small.Methods for Granular SoilsDue to difficulty in sampling, most methods are based on in-situ tests,either direct or via estimate of equivalent elastic modulus (E' s ).For most tests, the link between test result and modulus is empirical, sinceit depends on many variables; e.g. mineralogy, stress history, stress state,age, cementation, etc.SPTQuick estimates can be made using the simplified chart by Burland et al.(1977), as shown in Figure 25. The SPT has poor reliability, hence,empirical methods tend to be very conservative.52

CPT Guide - 2006Cone Penetration Test (CPT)CPTFigure 25 Approximate range of settlement for footings in sand(Burland et al., 1977)Meyerhof (1974) suggested that the total settlement, s, could be calculatedusing the following formula:s =Δp B2q c(av)where:Δp = net footing pressureB = footing widthq c (av) = average CPT penetration resistance below depth offooting,z = B53

CPT Guide - 2006Cone Penetration Test (CPT)Schmertmann (1970) recommended using the following equation:zs = C 1 C 2 Δp ∑ ⎜C ⎟ 3E's ⎠⎛⎝I⎞Δzwhere:C 1 = correction for depth of footing= 1 – 0.5(σ' 1 /Δp)C 2 = correction for creep and cyclic loading= 1 + 0.2 log (10 t yr )C 3 = correction for shape of footing= 1.0 for circular footings= 1.2 for square footings= 1.75 for strip footingsσ' 1 = effective overburden pressure at footing depth (see Figure 26)Δp = net footing pressuret yr = time in years since load applicationI z = strain influence factor (see Figure 3)Δz = thickness of sublayerE' s = Equivalent Young's modulus = α q cα = function of degree of loading, soil density, stress history,cementation, age, grain shape and mineralogy (see Figure 27)= 2 to 4 for recent, normally consolidated sands;= 4 to 6 for aged (>1,000years), normally consolidated sands;= 6 to 20 for overconsolidated sandsq c= average CPT resistance for sublayer54

CPT Guide - 2006Cone Penetration Test (CPT)Figure 26 Strain influence method for footings on sand(Schmertmann, 1970)55

CPT Guide - 2006Cone Penetration Test (CPT)Figure 27 Evaluation of drained Young's modulus from CPT for silicasands (Bellotti et al., 1989)In this method, (see Figure 26) the sand is divided into a number of layers,n, of thickness, Δz, down to a depth below the base of the footing equal to2B for a square footing and 4B for a strip footing (length of footing, L >10B). A value of q c is assigned to each layer, as illustrated in Figure 28.56

CPT Guide - 2006Cone Penetration Test (CPT)Figure 28 Application of Schmertmann (1970) method for settlement offootings on sand57

CPT Guide - 2006Cone Penetration Test (CPT)Seismic Shear Wave VelocityEslaamizaad and Robertson (1996) suggested using shear wave velocity(V s ) to measure small strain stiffness (G o ) directly, as follows:G o = gγ (Vs ) 2Then, the equivalent Young’s modulus can be estimated as follows:where:E' s = 2(1 + υ)ψ G o ≈ 2.6ψ G oΨ = a function of the degree of loading and stress history (see Figure29).Since settlement is a function of degree of loading, it is possible tocalculate the load settlement curve, as follows:where:s =⎛ ΔpB⎟ ⎞⎜⎝ E' s ⎠i ci c = influence coefficientIn general, for most well designed shallow foundations, Ψ ~ 0.3, hence,E' s ≈ G o58

CPT Guide - 2006Cone Penetration Test (CPT)Figure 29 Factor Ψ versus q/q ult for sands with various densities andstress historiesShear wave velocity has the advantage of providing a direct measure ofsoil stiffness without an empirical correlation. The only empiricism is toadjust small strain modulus for effects of stress level and strain levelbelow the footing. The shear wave velocity approach can also be appliedto estimate settlements in very stiff clays where consolidation settlementsare very small.Methods for Cohesive SoilsThe key parameter is the preconsolidation pressure, σ' p . This can only bemeasured accurately in the laboratory on high quality samples. However,OCR and σ' p profiles can be estimated from the CPT. It is useful to linkresults from high quality laboratory tests with continuous profiles of theCPT.In general, to keep settlements small, the applied stress must be < σ' p . Insoft ground this may require some form of ground improvement.59

CPT Guide - 2006Cone Penetration Test (CPT)Components of settlement are:s i = immediate (distortion) settlements c = consolidation settlements s = secondary time dependent (creep) settlementImmediate SettlementsBased on elastic theory. Janbu (1963) proposed:s i =⎛ ΔpB⎟ ⎞⎜⎝ E u ⎠μ o μ 1where:B = footing widthΔp = net pressureE u = soil modulus (undrained)μ o ,μ 1 = influence factors for depth of footing and thickness ofcompressible layerUndrained modulus can be estimated from undrained shear strength (s u )from either field vane tests and/or the CPT, but requires a knowledge ofsoil plasticity.E u = n. s uwhere n varies from 40 to 1000, depending on degree of loading andplasticity of soil (see Figure 30).60

CPT Guide - 2006Cone Penetration Test (CPT)Figure 30 Selection of soil stiffness ratio for clays(after Ladd et al., 1977)Consolidation SettlementsTerzaghi’s 1-D theory of consolidation applies; 2-D and 3-D effects aregenerally small. The key parameters are the preconsolidation pressure, σ' pand compression index, C c . These can only be measured accurately in thelaboratory on high quality samples. However, the CPT can be used toestimate profiles of parameters and to link with lab results.61

CPT Guide - 2006Cone Penetration Test (CPT)Rate of settlement is important, hence, need for coefficient ofconsolidation, c v . Experience shows that c v can be highly variable due tonon-linearity of stress-strain relationship as well as change in permeabilityas soils compress. Values of c v can be best estimated either:1. Separately from coefficient of volume compressibility (m v ) fromoedometer tests on high quality samples and permeability (k) fromin-situ tests, using:c v =m vkγwor2. Directly from CPTU dissipation tests.c v values vary by orders of magnitude; hence, accuracy of the calculationis generally very poor. Drainage conditions play a major role, yet aredifficult to identify. The CPTU can provide an excellent picture of thedrainage conditions. Avoid a design that depends on the time-settlementrelationship. For settlement sensitive structures, try to minimizedifferential settlements (e.g. Osaka Airport - mechanical adjustments dueto very large long term settlements).Secondary SettlementsTime dependent settlements depend on soil mineralogy and degree ofloading. Organic soils can have high secondary settlement. In general,avoid soils with high secondary settlements. Mesri & Godlewski, (1977)simplified approach links coefficient of secondary consolidation (C α ) andcompression index, C c , as follows:C α = 0.04⎛ Cc⎜⎝1e+ o⎟ ⎞⎠Provided that the applied stress is less than 80% of σ' p , secondaryconsolidation is generally small.62

CPT Guide - 2006Cone Penetration Test (CPT)Allowable SettlementsLoads considered in settlement analyses depend on nature of soil andtime-dependence of settlement. Differential settlements generally control.Sands• Load: maximum possible load due to immediate settlement• Differential settlement: can be up to 100% of maximum settlement dueto natural variability of sand. Typically less than or equal to 25mmClays• Load: dead load plus % of live load (LL) depending on duration oflive load• 50% of LL for buildings• 30% of LL for bridges• 75% of LL for reservoirs• Settlements: are more uniform and can be larger than 25mmTypical Design Sequence1. Check for possible isolated footing design2. Check for possible raft foundation3. Ground improvement4. Deep foundationsRaft FoundationsConsider a raft when:• Area of footing > 50% of building area• Need to provide underground space in location of high groundwater• Need to reduce magnitude of total settlements (i.e. floating foundation)• Need to reduce differential settlements63

CPT Guide - 2006Cone Penetration Test (CPT)A raft is an inverted slab, designed to distribute structural loads fromcolumns and walls, while keeping deformations within acceptable limits.Structural characteristics of rafts are optimized by accounting for theinteraction between the raft and the supporting ground. Structuralengineers usually perform an elastic analysis using elastic (Winkler)springs. Hence, they would like the spring constant, k s .k s = coefficient of subgrade reaction (kN/m 3 )k s = spwhere:p = net applied stresss = settlement resulting from applied stress, pThe process is governed by the relative stiffness of structure and theground. The coefficient of subgrade reaction is not a soil parameter, sinceit depends on the size of the footing and degree of loading. Oftenestimates are made from global tables (Terzaghi; see Table 10).64

CPT Guide - 2006Cone Penetration Test (CPT)Soil typeSubgrade reaction(kN/m 3 )Loose sand 5,000 – 16,000Medium dense sand 10,000 – 80,000Dense sand 60,000 – 125,000Clayey sand 30,000 – 80,000Silty sand 20,000 – 50,000Clayey soil:s u < 50 kPa 10,000 – 20,00050kPa < s u < 100kPa 20,000 – 50,000100 kPa < s u >50,000Table 10 Recommended coeffiecient of subgrade reaction (k s ) fordifferent soil types (Terzaghi)It is best to obtain estimates based on in-situ testing.Plate Load Tests (PLT)Provide a direct measure of relationship between p and s, but size effectscan dominate results. Terzaghi (1955) suggested a link between a 1 footsquare plate (k s1 ) and width of footing B, as follows:k s = k s12⎛ B + 1⎞⎜ ⎟⎝ 2B ⎠However, there is very large scatter in the results, due to variability inground stiffness with depth.65

CPT Guide - 2006Cone Penetration Test (CPT)Shear Wave Velocity (Vs)Based on work by Vesic (1961) and elastic theory, the modulus ofsubgrade reaction is:where:k'EB⎛ Es⎜⎝1-ν4s= 12s0.652EfIf⎞⎟⎠E s = modulus of elasticity of soil E sE f = modulus of elasticity of footingB = footing widthI f = moment of inertiaν = Poisson's ratio for soilk' s = modulus of subgrade reaction:k' s = k s BFor most values of E s and E f , the expression simplifies to:k' s ≈⎛ E⎜⎝1-sν 2⎟ ⎞⎠Bowles (1974) suggested:k s = 120 q allwhere q all is in kPa and k s is in kN/m 3 .It is possible to estimate E s from shear wave velocity, V s . The small strainshear modulus is given by the following:G o = gγ (Vs ) 266

CPT Guide - 2006Cone Penetration Test (CPT)In addition:G eq = Ψ G oandE s = 2(1 + υ) G eqSince ν ≈ 0.2 to 0.3,k' s = k s B ≈ 2.9 Ψ G oHence:where:k s ≈ 2.9 Ψγg( V ) 2BsΨ = a function of the degree of loading and stress history (see Figure29).67

CPT Guide - 2006Cone Penetration Test (CPT)Deep Foundation DesignPilesPiles can be used to:• Transfer high surface loads, through soft layers down to strongerlayers• Transfer loads by friction over significant length of soil• Resist lateral loads• Protect against scour, etc.• Protect against swelling soils, etcPiles are generally much more expensive than shallow footings.Types of PilesGenerally classified based on installation method (Weltman & Little,1977):• Displacement• Preformed• Driven Cast-in-place• High pressure grouted• Non(low)-displacement• Mud bored• Cased bored• Cast-in-place screwed (auger)Contractors are developing new pile types and installation techniquesconstantly to get better capacity and improved cost effectiveness fordifferent ground conditions. Hence, it is difficult to predict capacity andload-settlement response for all piles using simple analytical techniques,since the capacity and load response characteristics can be dominated bythe method of installation.68

CPT Guide - 2006Cone Penetration Test (CPT)Selection of Pile Type1. Assess foundation loads2. Assess ground conditions3. Are piles necessary?4. Technical considerations:• Ground conditions• Loading conditions• Environmental considerations• Site and equipment constraints• Safety5. List all technically feasible pile types and rank in order of suitabilitybased on technical considerations6. Assess cost of each suitable pile type and rank based on costconsiderations7. Assess construction program for each suitable pile type and rank8. Make overall ranking based on technical, cost and programconsiderationsGeneral Design PrinciplesAxial CapacityThe total ultimate pile axial capacity, Q ult , consists of two components:end bearing load (or point resistance), Q b , and side friction load(sometimes referred to as the shaft or skin friction), Q s , as follows:Q ult = Q s + Q bIn sands, the end bearing, Q b , tends to dominate, whereas in soft clays, theside friction, Q s , tends to dominate. The end bearing, Q b , is calculated asthe product between the pile end area, A p , and the unit end bearing, q p .The friction load, Q s , is the product between the outer pile shaft area, A s ,by the unit side friction, f p .Q ult = f p A s + q p A p69

CPT Guide - 2006Cone Penetration Test (CPT)Obviously, different f p values are mobilized along different parts of thepile, so that, in practice, the calculation is performed as a summation ofsmall components. For open ended piles, some consideration should bemade regarding whether the pile is plugged or unplugged (de Ruiter andBeringen, 1979), but the procedure is essentially as outlined above. Theallowable or design pile load, Q all will be then given by the total ultimateaxial capacity divided by a factor of safety. Sometimes separate factors ofsafety are applied to Q b and Q s .Basic approaches are:• Static Methods• Pile Dynamics• Pile Load TestsStatic MethodsPseudo-theoretical ApproachPseudo-theoretical methods are based on shear strength parameters.Similar to bearing capacity calculations for shallow foundations - thereare over 20 different bearing capacity theories. No single solution isapplicable to all piles and most can not account for installation technique.Hence, there has been extensive application of in-situ test techniquesapplied via empirical direct design methods.The most notable is the application of the CPT, since the CPT is a closemodel of the pile process. Detailed analysis is generally limited to highriskpile design, such as large off-shore piles.Effective Stress Approach (β)The effective stress (β) approach (Burland, 1973), has been very useful inproviding insight of pile performance.Unit side friction, f p = β σ v ’70

CPT Guide - 2006Cone Penetration Test (CPT)Unit end bearing, q p = N t σ b ’Soil Type Cast-in-placePilesDrivenPilesSilt 0.2 - 0.3 0.3 - 0.5Loose sand 0.2 - 0.4 0.3 - 0.8Medium sand 0.3 - 0.5 0.6 - 1.0Dense sand 0.4 - 0.6 0.8 - 1.2Gravel 0.4 - 0.7 0.8 - 1.5Table 11 Range of β coefficients: cohesionless soilsSoil Type Cast-in-placePilesDrivenPilesSilt 10 - 30 20 - 40Loose sand 20 - 30 30 - 80Medium sand 30 - 60 50 - 120Dense sand 50 - 100 100 - 120Gravel 80 - 150 150 - 300Table 12 Range of N t factors: cohesionless soilsThe above coefficients are approximate since they depend on groundcharacteristics and pile installation details. In the absence of pile loadtests assume FS = 3.Randolph and Wroth(1982) related β to the overconsolidation ratio (OCR)for cohesive soils and produced tentative design charts. In general, forcohesive soils:β = 0.25 - 0.32, and N t = 3 - 10Effective stress concepts may not radically change empirical based designrules, but can increase confidence in these rules and allow extrapolation tonew situations.71

CPT Guide - 2006Cone Penetration Test (CPT)Total Stress Approach (α)It has been common to design piles in cohesive soils based on total stressand undrained shear strength, s u .Unit side friction, f p = α s uUnit end bearing, q p = N t s uWhere α varies from 0.5 - 1.0 depending on OCR and N t varies from 6 to9 depending on depth of embedment and pile size.Empirical ApproachCPT MethodResearch has shown (Robertson et al., 1988; Briaud and Tucker, 1988;Tand and Funegard, 1989; Sharp et al., 1988) that CPT methods generallygive superior predictions of axial pile capacity compared to mostconventional methods. The main reason for this is that the CPT provides acontinuous profile of soil response. Almost all CPT methods usereduction factors to measured CPT values. The need for such reductionfactors is due to a combination of the following influences: scale effect,rate of loading effects, difference of insertion technique, position of theCPT friction sleeve and differences in horizontal soil displacements. Theearly work by DeBeer (1963) identified the importance of scale effects.Despite these differences, the CPT is still the test that gives the closestsimulation to a pile. Superiority of CPT methods over non CPT methodshas been confirmed in other studies (e.g. O'Neill, 1986).The main CPT method by Bustamante and Gianeselli (1982 - LCPCMethod) is outlined below. The LCPC CPT method is recommendedsince it provides simple guidance to account for different pile installationmethods and provides good estimates of axial capacity of single piles.72

CPT Guide - 2006Cone Penetration Test (CPT)LCPC CPT method (Bustamante and Gianeselli, 1982)The method by Bustamante and Gianeselli was based on the analysis of197 pile load (and extraction) tests with a wide range of foundation andsoil types, which may partly explain the good results obtained with themethod. The method, also known as the LCPC method, is summarized inTable 13 and Table 14.Table 13 Bearing capacity factors, k c(Bustamante and Gianeselli, 1982)73

CPT Guide - 2006Cone Penetration Test (CPT)The pile unit end bearing, q p , is calculated from the calculated equivalentaverage cone resistance, q ca , multiplied by an end bearing coefficient, k c(Table 13). The pile unit side friction, f p , is calculated from measured q cvalues divided by a friction coefficient, α LCPC (Table 14).q p = k c q caf p =αq cLCPCMaximum f p values are also recommended based on pile and soil type.Only the measured CPT q c is used for the calculation of both side frictionand pile end bearing resistance. This is considered an advantage by manydue to the difficulties associated in interpreting sleeve friction (f s ) in CPTdata.Table 14 Friction coefficient, α(Bustamante and Gianeselli, 1982)74

CPT Guide - 2006Cone Penetration Test (CPT)The equivalent average cone resistance, q ca , at the base of the pile used tocompute the pile unit end bearing, q p , is the mean q c value measured alongtwo fixed distances, a, (a = 1.5D, where D is the pile diameter) above (-a)and below (+a) the pile tip. The authors suggest that q ca be calculated inthree steps, as shown in Figure 31. The first step is to calculate q' ca , themean q c between -a and +a. The second step is to eliminate values higherthan 1.3q' ca along the length -a to +a, and the values lower than 0.7q' caalong the length -a, which generates the thick curve shown in Figure 31.The third step is to calculate q ca , the mean value of the thick curve.Figure 31 Calculation of equivalent average cone resistance(Bustamante and Gianeselli, 1982).Other Design ConsiderationsFactor of SafetyIn order to obtain the design load, factors of safety are applied to theultimate load and a deterministic approach is usually adopted to definethese values. The selection of an appropriate factor of safety depends on75

CPT Guide - 2006Cone Penetration Test (CPT)many factors, such as reliability and sufficiency of the site investigationdata, confidence in the method of calculation, previous experience withsimilar piles in similar soils and whether or not pile load test results areavailable.Factors of safety are generally of the order of 2, although real values aresometimes greater, as partial factors of safety are sometimes appliedduring calculations (particularly to soil strengths) before arriving to theultimate pile capacity.Recommended factors of safety for calculating the axial capacity of pilesfrom the CPT are given in Table 15.MethodBustamante andGianeselli (1982)de Ruiter and Beringen(1979)Factor of safety (FS)2.0 (Qs)3.0 (Qb)2.0 (static loads)1.5 (static + storm loads)Table 15 Recommended factors of safety for axial capacity of piles fromCPTThe design of high capacity large diameter bored piles in stiff clay ordense sand can be difficult due to the fact that settlement criteria usuallycontrol rather than capacity. Hence, very high factors of safety are oftenapplied to limit settlement.Pile DynamicsThe objective of methods that rely on pile dynamics is to relate thedynamic pile behavior to the ultimate static pile resistance. Hence, piledynamics can work well in drained soils (sands, gravels, etc.) but can bedifficult in undrained soils (silts, clays, etc.).The early approach was to use simple pile driving equations (Hiley,Engineering News, etc.) based on equating the available energy of the76

CPT Guide - 2006Cone Penetration Test (CPT)hammer to the work performed by the pile. However, these were basedon a rigid pile concept, which is fundamentally incorrect.Current approaches are based on 1-D wave-equation analyses (Goble etal., 1970) This method takes into account the characteristics of the;hammer, driving cap, pile and soil. The method is commonly appliedusing commercial software (i.e. WEAP). This method is good to assist inselection of hammers and prediction of driving stresses and the choice ofdriving criteria. It is useful for dynamic monitoring during construction.Pile Load TestsSince there is much uncertainty in the prediction of pile capacity andresponse, it is common to perform pile load tests on major projects.For major projects, it is common to apply static methods (i.e. LCPC CPTmethod) to obtain a first estimate of capacity, apply pile dynamics ifdriven piles selected (aid in hammer selection, driving stresses, drivingcriteria) and perform a small number of pile load tests to evaluate pileresponse and to calibrate the static method. Results from the pile loadtests can be used to modify the static prediction (i.e. CPT prediction) ofpile capacity and the modified method applied across the site. For lowriskprojects, pile load tests may not be warranted and a slightlyconservative prediction should be applied using the static (CPT) method.Group CapacityThe capacity of a group of piles is influenced by the spacing, pileinstallation and ground conditions. The group efficiency is defined as theratio of the group capacity to the sum of the individual pile capacities.Driven piles in cohesionless soils develop larger individual capacitieswhen installed as a group since lateral earth pressures and soil densityincrease due to pile driving. Hence, it is conservative to use the sum ofthe individual pile capacities.For bored pile groups, the individual capacity can reduce due to reducedlateral stresses. Meyerhof (1976) suggested a reduction factor of 0.67.77

CPT Guide - 2006Cone Penetration Test (CPT)For piles in cohesive soils the capacity of the pile group should beestimated based on the ‘block’ of piles since the soil between the pilesmay move with the pile group.Design of Piles in RockPiles can be placed on or socketed into rock to carry high loads. Theexact area of contact with rock, depth of penetration into rock and qualityof rock are largely unknown, hence, there is much uncertainty. Thecapacity is often confirmed based on driving or installation details, localexperience and test loading. End bearing capacity can be based onpressuremeter test results or strength from rock cores. Shaft resistanceshould be estimated with caution, due to possible poor contact betweenrock and pile, possible stress concentration and resulting progressivefailure.Pile SettlementAlthough the installation of piles changes the deformation andcompressibility characteristics of the soil mass governing the behavior ofsingle piles under load, this influence usually extends only a few pilediameters below the pile base. Meyerhof (1976) suggested that the totalsettlement of a group of piles at working load can generally be estimatedassuming an equivalent foundation. For a group of predominately frictionpiles (i.e. Q s > Q b ), the equivalent foundation is assumed to act on the soilat an effective depth of 2/3 of the pile embedment. For a group of pileswhich are predominately end bearing (i.e. Q b > Q s ), the equivalentfoundation is taken at or close to the base of the piles. The resultingsettlement is calculated in a manner similar to that of shallow foundations.Sometimes large capacity piles are installed and used as single piles andthe load settlement response of a single pile is required. The loadsettlement response of a single pile is controlled by the combinedbehavior of the side resistance (Q s ) and base resistance (Q b ). The sideresistance is usually developed at a small settlement of about 0.5 percentof the shaft diameter and generally between 5 to 10 mm. In contrast to theside resistance, the base resistance requires much larger movements todevelop fully, usually about 10 to 20 percent of the base diameter. Hence,an estimate of the load settlement response of a single pile can be made by78

CPT Guide - 2006Cone Penetration Test (CPT)combining the two components of resistance according to the aboveguidelines. In this way, a friction pile (i.e. Q s >> Q b ), will show a clearplunging failure at a small settlement of about 0.5% of the pile diameter.On the other hand, an end bearing pile (i.e. Q b >> Q s ), will not show aclear plunging failure until very large settlements have taken place andusually settlement criteria control before failure can occur. In both cases,the side friction is almost fully mobilized at working loads. Hence, it isoften important to correctly define the proportions of resistance (Q b /Q s ).Methods have been developed to estimate the load-transfer (t-z) curves(Verbrugge, 1988). However, these methods are approximate at best andare strongly influenced by pile installation and soil type. Therecommended method for estimating load settlement response for singlepiles is to follow the general guidelines above regarding the developmentof each component of resistance.Negative Shaft Friction and Down Drag on PilesWhen the ground around a pile settles, the resulting downward movementcan induce downward forces on the pile.The magnitude of the settlement can be very small to develop thesedownward forces. For end bearing piles, the negative shaft friction plusthe dead load can result in structural failure of the pile. For friction piles,the negative shaft friction can result in greater settlements. No pilesubject to down drag will settle more than the surrounding ground.Lateral Response of PilesVertical piles can resist lateral loads by deflecting and mobilizingresistance in the surrounding ground. The response depends on therelative stiffness of the pile and the ground. In general, the response iscontrolled by the stiffness of the ground near the surface, since most pilesare relatively flexible.A common approach is to simulate the ground by a series of horizontalsprings. The spring stiffness can be estimated based on a simple subgrademodulus approach (assumes the ground to be linear and homogeneous) oras non-linear springs (p-y curves) (Matlock, 1970). The p-y curves can be79

CPT Guide - 2006Cone Penetration Test (CPT)estimated using empirical relationships based on lab results or in-situ tests(e.g. pressuremeter, DMT, SCPT) (Baguelin et al., 1978; Robertson et al.,1986).Another approach is to simulate the ground as an elastic continuum.Poulos and Davis, (1980) and Randolph, (1981) suggested design chartsthat require estimates of equivalent ground modulus for uniformhomogeneous ground profiles.The above approaches apply to single piles. When piles are installed ingroups, interaction occurs and lateral deformations can increase. Thesecan be estimated using simplified theoretical solutions (Poulos and Davis,1980, Randolph, 1981). The direction of the applied load relative to thegroup is important for laterally loaded pile groups.Liquefaction(see Robertson & Wride, 1998 and Zhang et al., 2002 & 2004 for more updateddetails)Soil liquefaction is a major concern for structures constructed with or onsand or sandy soils. The major earthquakes of Niigata in 1964 and Kobein 1995 have illustrated the significance and extent of damage caused bysoil liquefaction. Soil liquefaction is also a major design problem forlarge sand structures such as mine tailings impoundment and earth dams.To evaluate the potential for soil liquefaction, it is important to determinethe soil stratigraphy and in-situ state of the deposits. The CPT is an idealin-situ test to evaluate the potential for soil liquefaction because of itsrepeatability, reliability, continuous data and cost effectiveness. Thissection presents a summary of the application of the CPT to evaluate soilliquefaction. Full details are contained in a report prepared for NCEER(1997) (Youd et al 2001) as a result of a workshop on liquefaction held inJanuary 1996 and in a paper by Robertson and Wride (1998).Liquefaction DefinitionsSeveral phenomena are described as soil liquefaction, hence, a set ofdefinitions are provided to aid in the understanding of the phenomena.80

CPT Guide - 2006Cone Penetration Test (CPT)Flow Liquefaction• Applies to strain softening soils only (i.e. susceptible to strength loss)• Requires a strain softening response in undrained loading resulting inapproximately constant shear stress and effective stress• Requires in-situ shear stresses to be greater than the residual orminimum undrained shear strength• Either monotonic or cyclic loading can trigger flow liquefaction• For failure of a soil structure to occur, such as a slope, a sufficientvolume of material must strain soften. The resulting failure can be aslide or a flow depending on the material characteristics and groundgeometry. The resulting movements are due to internal causes and canoccur after the trigger mechanism occurs• Can occur in any metastable saturated soil, such as very loose finecohesionless deposits, very sensitive clays, and loess (silt) depositsCyclic (softening) Liquefaction• Requires undrained cyclic loading during which shear stress reversaloccurs or zero shear stress can develop• Requires sufficient undrained cyclic loading to allow effective stressesto reach essentially zero• Deformations during cyclic loading can accumulate to large values,but generally stabilize shortly after cyclic loading stops. The resultingmovements are due to external causes and occur mainly during thecyclic loading• Can occur in almost all saturated sands provided that the cyclic loadingis sufficiently large in magnitude and duration• Clayey soils can experience cyclic liquefaction but deformations aregenerally small due to the cohesive strength at zero effective stress.Rate effects (creep) often control deformations in cohesive soils.Note that strain softening soils can also experience cyclic liquefactiondepending on ground geometry. Figure 32 presents a flow chart(Robertson and Wride, 1998) to clarify the phenomena of soilliquefaction.81

CPT Guide - 2006Cone Penetration Test (CPT)Figure 32Flow chart to evaluate liquefaction of soils.If a soil is strain softening (i.e. can experience strength loss), flowliquefaction is possible if the soil can be triggered to collapse and if thegravitational shear stresses are larger than the minimum undrained shearstrength. The trigger can be either monotonic or cyclic. Whether a slopeor soil structure will fail and slide will depend on the amount of strainsoftening soil relative to strain hardening soil within the structure, thebrittleness of the strain softening soil and the geometry of the ground.The resulting deformations of a soil structure with both strain softeningand strain hardening soils will depend on many factors, such asdistribution of soils, ground geometry, amount and type of triggermechanism, brittleness of the strain softening soil and drainageconditions. Examples of flow liquefaction failures are the Aberfan flowslide (Bishop, 1973), Zealand submarine flow slides (Koppejan et al.,1948) and the Stava tailings dam failure. In general, flow liquefactionfailures are not common, however, when they occur, they typically takeplace rapidly with little warning and are usually catastrophic. Hence, thedesign against flow liquefaction should be carried out cautiously.If a soil is strain hardening, flow liquefaction will not occur. However,cyclic liquefaction can occur due to cyclic undrained loading. The82

CPT Guide - 2006Cone Penetration Test (CPT)amount and extent of deformations during cyclic loading will depend onthe density of the soils, the magnitude and duration of the cyclic loadingand the extent to which shear stress reversal occurs. If extensive shearstress reversal occurs and the magnitude and duration of the cyclicloading are sufficiently large, it is possible for the effective stresses toessentially reach zero possibly resulting in large deformations. Examplesof cyclic liquefaction were common in the major earthquakes in Niigata in1964 and Kobe in 1995 and manifest in the form of sand boils, damagedlifelines (pipelines, etc.) lateral spreads, slumping of embankments,ground settlements, and ground surface cracks.If cyclic liquefaction occurs and drainage paths are restricted due tooverlying less permeable layers, the sand immediately beneath the lesspermeable soil can loosen due to pore water redistribution, resulting inpossible subsequent flow liquefaction, given the right geometry. In caseswhere drainage is restricted, caution is required to allow for possible voidredistribution.CPT for Cyclic Liquefaction – Level Ground Sites(see Robertson & Wride, 1998 and Zhang et al., 2002 & 2004 for more updateddetails)Most of the existing work on cyclic liquefaction has been primarily forearthquakes. The late Prof. H.B. Seed and his co-workers developed acomprehensive methodology to estimate the potential for cyclicliquefaction for level ground sites due to earthquake loading. Themethodology requires an estimate of the cyclic stress ratio (CSR) profilecaused by the design earthquake and the cyclic resistance ratio (CRR) ofthe ground. If the CSR is greater than the CRR cyclic liquefaction canoccur. The CSR is usually estimated based on a probability of occurrencefor a given earthquake. A site specific seismicity analysis can be carriedout to determine the design CSR profile with depth. A simplified methodto estimate CSR was also developed by Seed and Idriss (1971) based onthe maximum ground surface acceleration (a max ) at the site. Thesimplified approach can be summarized as follows:CSR =τσ'avvo⎡a= 0.65 ⎢⎣ gmax⎤ ⎛ σ⎥⎜⎦ ⎝ σ'vovo⎟ ⎞⎠rd83

CPT Guide - 2006Cone Penetration Test (CPT)where τ av is the average cyclic shear stress; a max is the maximumhorizontal acceleration at the ground surface; g is the acceleration due togravity; σ vo and σ' vo are the total and effective vertical overburdenstresses, respectively; and r d is a stress reduction factor which isdependent on depth. The factor r d can be estimating using the followingtri-linear function, which provides a good fit to the average of thesuggested range in r d originally proposed by Seed and Idriss (1971):r d = 1.0 – 0.00765zif z < 9.15 m= 1.174 – 0.0267zif z = 9.15 to 23 m= 0.744 – 0.008zif z = 23 to 30 m= 0.5if z > 30 mwhere z is the depth in meters. These formulae are approximate at bestand represent only average values since r d shows considerable variationwith depth.The sequence to evaluate cyclic liquefaction for level ground sites is:1. Evaluate susceptibility to cyclic liquefaction2. Evaluate triggering of cyclic liquefaction3. Evaluate post-earthquake deformations.1. Evaluate Susceptibility to Cyclic LiquefactionThe response of soil to seismic loading varies with soil type and state(void ratio, effective confining stress, stress history, etc.). Boulanger andIdriss (2004) correctly distinguished between sand-like and clay-likebehavior. The following criteria can be used to identify soil behavior:Sand-like BehaviorSand-like soils are susceptible to cyclic liquefaction when their behavioris characterized by Plasticity Index (PI) < 12 and Liquid Limit (LL) < 3784

CPT Guide - 2006Cone Penetration Test (CPT)and natural water content (w c ) > 0.8 (LL). More emphasis should beplaced on PI, since both LL and w c tend to be less reliable.• Low risk project: Assume soils are susceptible to cyclicliquefaction unless previous local experience shows otherwise.• High risk project: Either assume soils are susceptible to cyclicliquefaction or obtain high quality samples and evaluatesusceptibility based on appropriate laboratory testing, unlessprevious local experience exists.Clay-like BehaviorClay-like soils are generally not susceptible to cyclic liquefaction whentheir behavior is characterized by PI > 12 but they can experience cyclicfailure.• Low risk project: Assume soils are not susceptible to cyclicliquefaction unless previous local experience shows otherwise.Check for cyclic failure.• High risk project: Obtain high quality samples and evaluatesusceptibility to either cyclic liquefaction and/or cyclic failurebased on appropriate laboratory testing, unless previous localexperience exists.These criteria are generally conservative. Boulanger and Idriss (2004)suggested that sand-like behavior is limited to PI < 7. Use the abovecriteria, unless local experience in the same geology unit shows that alower PI is more appropriate.2. Evaluate Triggering of Cyclic LiquefactionSand-like MaterialsSeed et al., (1985) developed a method to estimate the cyclic resistanceratio (CRR) for clean sand with level ground conditions based on theStandard Penetration Test (SPT). Recently the CPT has become morepopular to estimate CRR, due to the continuous, reliable and repeatablenature of the data (NCEER, 1997; Youd et al., 2001).Apply the simplified (NCEER) approach as described by Youd et al(2001) using generally conservative assumptions. The simplifiedapproach should be used for low to medium risk projects and forpreliminary screening for high risk projects. For low risk projects, wherethe simplified approach is the only method applied, conservative criteria85

CPT Guide - 2006Cone Penetration Test (CPT)should be used. The recommended CPT correlation for clean sand isshown in Figure 33 and can be estimated using the following simplifiedequations:CRR 7.5 =⎡93⎢⎣( q )c1N1000cs3⎤⎥⎦+ 0.08if 50 ≤ (q c1N ) cs ≤ 160CRR 7.5 =⎡0.833⎢⎣( q )c1N1000cs3⎤⎥⎦+ 0.05if (q c1N ) cs < 50The field observations used to compile the curve in Figure 33 are basedprimarily on the following conditions:• Holocene age, clean sand deposits• Level or gently sloping ground• Magnitude M = 7.5 earthquakes• Depth range from 1 to 15 m ( 3 to 45 feet) (85% is for depths < 10 m(30 ft)• Representative average CPT values for the layer considered to haveexperienced cyclic liquefaction.86

CPT Guide - 2006Cone Penetration Test (CPT)Figure 33Cyclic resistance ratio (CRR) from CPT for clean sands(After Robertson and Wride, 1998)Caution should be exercised when extrapolating the CPT correlation toconditions outside the above range. An important feature to recognize isthat the correlation is based primarily on average values for the inferredliquefied layers. However, the correlation is often applied to all measuredCPT values, which include low values below the average. Therefore, thecorrelation can be conservative in variable deposits where a small part ofthe CPT data can indicate possible liquefaction.It has been recognized for some time that the correlation to estimateCRR 7.5 for silty sands is different than that for clean sands. Typically acorrection is made to determine an equivalent clean sand penetration87

CPT Guide - 2006Cone Penetration Test (CPT)resistance based on grain characteristics, such as fines content, althoughthe corrections are due to more than just fines content.One reason for the continued use of the SPT has been the need to obtain asoil sample to determine the fines content of the soil. However, this hasbeen offset by the generally poor repeatability of the SPT data. It is nowpossible to estimate grain characteristics directly from the CPT.Robertson and Wride (1998) suggest estimating an equivalent clean sandcone penetration resistance, (q c1N ) cs using the following:(q c1N ) cs = K c (q c1N )where K c is a correction factor that is a function of grain characteristics(fines content and plasticity)of the soil.Robertson and Wride (1998) suggest estimating the grain characteristicsusing the soil behavior chart by Robertson (1990) and the soil behaviortype index, I c , where:2[ ] 0. 52I c = ( 3 .47 − log Q) + ( log F + 1.22)and Q =⎛ q⎜⎝c− σPa2vo⎞ ⎛ P⎟⎜⎠ ⎝ σ'avo⎞⎟⎠nis the normalized CPT penetration resistance (dimensionless); n = stressexponent; F = f s /[(q c - σ vo )] x 100% is the normalized friction ratio (inpercent); f s is the CPT sleeve friction stress; σ vo and σ' vo are the totaleffective overburden stresses respectively; P a is a reference pressure in thesame units as σ' vo (i.e. P a = 100 kPa if σ' vo is in kPa) and P a2 is a referencepressure in the same units as q c and σ vo (i.e. P a2 = 0.1 MPa if q c and σ voare in MPa).The soil behavior type chart by Robertson (1990) (see Figure 16) uses anormalized cone penetration resistance (Q) based on a simple linear stressexponent of n = 1.0 (see above), whereas the chart recommended here forestimating CRR is essentially based on a normalized cone penetrationresistance (q c1N ) based on a stress exponent n = 0.5. Olsen and Malone(1988) correctly suggested a normalization where the stress exponent (n)88

CPT Guide - 2006Cone Penetration Test (CPT)varies from around 0.5 in sands to 1.0 in clays. However, thisnormalization for soil type is somewhat complex and iterative.The Robertson (1990) procedure using n = 1.0 is recommended for soilclassification in clay type soils when I c > 2.6. However, in sandy soilswhen I c ≤ 2.6, it is recommended that data being plotted on the Robertson(1990) chart be modified by using n = 0.5. Hence, the recommendedprocedure is to first use n = 1.0 to calculate Q and, therefore, an initialvalue of I c for CPT data. If I c > 2.6, the data should be plotted directly onthe Robertson (1990) chart (and assume q c1N = Q). However, if I c ≤ 2.6,the exponent to calculate Q should be changed to n = 0.5 and I c should berecalculated based on q c1N and F. If the recalculated I c remains less than2.6, the data should be plotted on the Robertson (1990) chart using q c1Nbased on n = 0.5. If, however, I c iterates above and below a value of 2.6,depending which value of n is used, a value of n = 0.75 should be selectedto calculate q c1N and plot data on the Robertson (1990) chart. Note that ifthe in-situ effective overburden stresses are in the order of 50 kPa to150 kPa, the choice of normalization has little effect on the calculatednormalized penetration resistance. Zhang, Robertson and Brachman(2002 and 2004) updated the procedure to estimate the stress exponent ‘n’as a function of the CPT soil behavior type, I c (see Figure 34).The recommended relationship between I c and the correction factor K c isgiven by the following:K c = 1.0 if I c ≤ 1.64K c = - 0.403 I c 4 + 5.581 I c 3 – 21.63 I c 2 + 33.75 I c – 17.88 if I c > 1.64The correction factor, K c , is approximate since the CPT responds to manyfactors such as soil plasticity, fines content, mineralogy, soil sensitivity,age, and stress history. However, in general, these same factors influencethe CRR 7.5 in a similar manner. Caution should be used when applyingthe relationship to sands that plot in the region defined by 1.64 < I c < 2.36and F < 0.5% so as not to confuse very loose clean sands with sandscontaining fines. In this zone, it is suggested to set K c = 1.0. Soils with I c> 2.6 fall into the clayey silt, silty clay and clay regions of the CPT soilbehavior chart and, in general, are not susceptible to liquefaction.However, samples should be obtained and liquefaction potential evaluatedusing other criteria based primarily on plasticity, e.g. soils with plasticity89

CPT Guide - 2006Cone Penetration Test (CPT)index greater than 12 are likely not susceptible to liquefaction. Soils thatfall in the lower left region of the CPT soil behavior chart defined by I c >2.6 but F < 1.0% can be sensitive and hence, possibly susceptible to bothcyclic and flow liquefaction. The full methodology to estimate CRR 7.5from the CPT is summarized in Figure 34.For low risk projects and for preliminary screening in high risk projects,Robertson and Wride (1998) suggested that when I c > 2.6 the soil wouldhave clay-like behavior and would likely not be susceptible toliquefaction. Youd et al (2001) recommends that soils be sampled whereI c > 2.4 to verify the behavior type. When I c > 2.4 selected (disturbed)soil samples (for grain size distribution, Atterberg limits and watercontent) should be obtained and tested to confirm susceptibility to cyclicliquefaction using the criteria in the previous section. Selective soilsampling based on I c should be carried out adjacent to some CPTsoundings. Disturbed samples can be obtained using either direct pushsamplers (using CPT equipment) or conventional drilling/samplingtechniques close to the CPT sounding.The factor of safely against cyclic liquefaction is defined as:Factor of Safety =CRR 7.5 MSFCSRWhere:MSF is the Magnitude Scaling Factor to convert the CRR 7.5 for M =7.5 to the equivalent CRR for the design earthquake.The recommended MSF is given by:174MSF =2. 56MThe above recommendations are based on the NCEER Workshop in 1996(NCEER, 1997 and Youd et al., 2001)90

CPT Guide - 2006Cone Penetration Test (CPT)q c : tip resistance, f s : sleeve frictionσ vo , σ vo’ : in-situ vertical total and effective stressunits: all in kPainitial stress exponent : n = 1.0 and calcualte Q, F, and I cif I c = 3.30, n = 1.0iterate until the change in n, Δ n < 0.01if σ vo’ > 300 kPa, let n = 1.0 for all soilsC n⎛= ⎜⎝ σ100'vo⎞⎟⎠n( q c− σvo )Q = ⋅ C n100f s, F = ⋅ 100( q − σ )[( 3 . 47 − log Q ) ]2 + ( 1 . 22 log F )2I c=+cvoif Ic = 2.60, evaluate using other criteria; likely non- liquefiable if F > 1%BUT , if 1.64 < I c < 2.36 and F < 0.5%, set K c = 1.0( q c1N)cs= Kc⋅ Q3⎛ ( q c1N) cs ⎞CRR 7 . 5 = 93 ⋅ ⎜ ⎟ + 0 . 08 , if 50 = 2.60, evaluate using other criteria; likely non- liquefiable if F > 1%Figure 34 Flow chart to evaluate cyclic resistance ratio from CPT(Updated from Robertson and Wride, 1998)91

CPT Guide - 2006Cone Penetration Test (CPT)An example of the CPT method to evaluate cyclic liquefaction is shownon Figure 35 for the Moss Landing site that suffered cyclic liquefactionduring the 1989 Loma Prieta earthquake in California (Boulanger et al.,1995).Figure 35Example of CPT to evaluate cyclic liquefaction at MossLanding Site(After Robertson and Wride, 1998)92

CPT Guide - 2006Cone Penetration Test (CPT)Clay-like MaterialsBecause of the cohesive nature of clay-like materials, they develop muchsmaller pore pressures under undrained cyclic loading than sand-likematerials. Hence, clay-like materials do not reach zero effective stresswith resulting large deformations under cyclic loading. Hence, clay-likematerials are not susceptible to cyclic liquefaction. However, when thecyclic stress ratio (CSR) is large relative to the undrained shear strengthratio of clay-like materials, deformations can develop. Boulanger andIdriss (2004) used the term ‘cyclic failure’ to define this build-up ofdeformations under cyclic loading in clay-like soils. Boulanger and Idriss(2004) showed that the CRR for cyclic failure in clay-like materials iscontrolled by the undrained shear strength ratio, which in term iscontrolled by the stress history (OCR). Boulanger and Idriss (2004)recommended the following expressions for CRR M=7.5 (for a magnitude7.5 earthquake) in natural deposits of clay-like soils:andCRR M=7.5 = 0.8 (S u /σ’ vc ) K αCRR M=7.5 = 0.18 (OCR) 0.8 K αWhere:S u /σ’ vc is the undrained shear strength ratio for the appropriate direction ofloading.K α is a correction factor to account to static shear stress. For welldesigned structures where the factor of safety for static loading is large,K α is generally close to 0.9.For seismic loading where CSR < 0.6, cyclic failure is possible only innormally to lightly overconsolidated (OCR < 4) clay-like soils.Boulanger and Idriss (2004) recommended three approaches to determineCRR for clay-like materials, which are essentially:1. Estimate using empirical methods based on stress history2. Measure S u using in-situ tests3. Measure CRR on high quality samples using appropriate cycliclaboratory testing.93

CPT Guide - 2006Cone Penetration Test (CPT)The CPT can be used to estimate both undrained shear strength ratio(S u /σ’ vc ) and stress history (OCR). The CPT has the advantage that theresults are repeatable and provide a detailed continuous profile of OCRand hence CRR. Contours of CRR M=7.5 (for K α = 1.0) in the clay-likeregion of the CPT SBT chart are shown in Figure 36. Figure 36 alsoshows the contours of CRR M=7.5 in the sand-like region using the R&Wmethod. If the R&W method (based on clean sand equivalent values) isused in the clay-like region (where I c > 2.6) the estimated values ofCRR M=7.5 are generally conservative (i.e. smaller than those derived fromOCR for clay-like soils).For low risk projects, the CRR M=7.5 for cyclic failure in clay-like soils canbe estimated using conservative correlations from the CPT (see Figure36). For medium risk projects, field vane tests (FVT) can also be used toprovide site specific correlations with the CPT. For high risk projectshigh quality undisturbed samples should be obtained and appropriatecyclic laboratory testing performed. Since sampling and laboratorytesting can be slow and expensive, sample locations should be based onpreliminary screening using the CPT.2. Evaluation of Post-earthquake DeformationsVertical settlementsFor low to medium risk projects and for preliminary estimates for highrisk projects, post earthquake settlements can be estimated using variousempirical methods to estimate post-earthquake volumetric strains (Wu,2002 and Zhang et al., 2002). The method by Zhang et al (2002) has theadvantage that it is based on CPT results and can provide a detailedvertical profile of volumetric strains at each CPT location. The CPTbasedapproach is generally conservative since it is applied to all CPTdata often using either commercially available software or in-housespreadsheet software. The CPT-based approach captures low (minimum)cone values in soil layers and in transition zones at soil boundaries. Theselow cone values in transition zones often result in accumulated volumetricstrains that tend to increase the estimated settlement. Engineeringjudgment can be used to remove excessive conservatism in highly interbeddeddeposits where there are frequent transition zones at soilboundaries.94

CPT Guide - 2006Cone Penetration Test (CPT)Engineering judgment is required to evaluate the consequences of thecalculated vertical settlements taking into account soil variability, depth ofthe liquefied layers and project details (see Zhang et al., 2002). For highrisk projects, selected high quality sampling and appropriate laboratorytesting may be necessary in critical zones identified by the simplifiedapproach.Lateral DeformationsFor low to medium risk projects and for preliminary evaluation for highrisk projects, post earthquake lateral deformation (spreading) can beestimated using various empirical methods (Youd et al, 2002 and Zhang etal, 2004). The method by Zhang et al (2004) has the advantage that it isbased on CPT results and can provide a detailed vertical profile of strainsat each CPT location. The CPT-based approach is generally conservativesince it should be applied to all CPT data and captures low (minimum)cone values in soil layers and in transition zones at soil boundaries. Theselow cone values in transition zones often result in accumulated shearstrains that tend to increase the estimated lateral deformations.Engineering judgment can be used to remove excessive conservatism inhighly inter-bedded deposits where there are frequent transition zones atsoil boundaries.Engineering judgment is required to evaluate the consequences of thecalculated lateral displacements taking into account, soil variability, sitegeometry, depth of the liquefied layers and project details. In general,assume that any liquefied layer located at a depth more than twice thedepth of the free-face will have little influence on the lateral deformations.For high risk projects, selected high quality sampling and appropriatelaboratory testing may be necessary in critical zones identified by thesimplified approach.When the calculated lateral deformations using the above empiricalmethods are very large (i.e. shear strains of more than 30%) the soilshould also be evaluated for susceptibility for strength loss (see nextsection on sloping ground) and the overall stability against a flow slideevaluated.95

CPT Guide - 2006Cone Penetration Test (CPT)Figure 36 Cyclic Resistance Ratio (CRR) M = 7.5 using CPT(based on Robertson & Wride, 1998 and Boulanger and Idriss, 2004)Zone A 1Zone A 2Zone BZone CCyclic liquefaction likely depending on size and duration ofcyclic loadingCyclic liquefaction and strength loss likely depending onloading and ground geometryLiquefaction and post-earthquake strength loss unlikely,check cyclic failureCyclic liquefaction and strength loss possible depending onsoil plasticity, brittleness/sensitivity, strain to peak undrainedstrength and ground geometry96

CPT Guide - 2006Cone Penetration Test (CPT)CPT for Flow Liquefaction – Steeply Sloping SitesA steeply sloping ground site is defined as:1. Steeply sloping ground (slope angle > 5 degrees)2. Earth embankments (e.g. dams, tailings structures)Sequence to evaluate flow liquefaction (i.e. strength loss)1. Evaluate susceptibility for strength loss2. Evaluate stability using post-earthquake shear strengths3. Evaluate if earthquake will trigger strength loss4. Evaluate deformations.1. Evaluate Susceptibility For Strength LossUse CPT, plus disturbed samples (for grain size distribution,Atterberg limit and water content) to identify materials that aresusceptible to strength loss due to earthquake shaking, i.e. loose,sand-like and sensitive clay-like materials. Use conservativeevaluation techniques, since experience has shown that whenstrength loss occurs, instability can be rapid with little warning anddeformations can be very large.a. Loose, sand-like materials (susceptible to strength loss)i. Either fines content < 20% or fines content > 35% andPlasticity Index (PI) < 12 and water content (w c ) > 0.80Liquid Limit (LL)ii. CPT (q c1N ) < 70 and SPT (N 1 ) 60 < 14. These values willbe conservative for soils with high fines content.b. Sensitive Clay-like materials (test for susceptibility, function ofsensitivity and strain to failure)i. Fines content > 35%, and water content (w c ) > 0.85 LLii. CPT Zone C (see CPT chart, Figure 36) and/or FVT(Field Vane Test), where sensitivity, S t > 3iii. Strain to failure less than 1%c. Clay-like materials (not susceptible to strength loss)i. Fines content > 20% and PI > 12, and water content (w c )< 0.80 Liquid Limit (LL)97

CPT Guide - 2006Cone Penetration Test (CPT)ii. CPT Zone B (see CPT chart, Figure 36)If layers/zones of low permeability exist that could inhibit pore waterredistribution after seismic loading and promote void redistribution,increase conservatism when evaluating susceptibility for strength loss.2. Evaluate Stability Using Post-earthquake Shear Strengthsa. Initial stability analysis assuming strength loss is triggered andusing conservative estimates of minimum (liquefied/residual/steady state) undrained shear strength, S ur , based oneither empirical correlations with in-situ tests or field measuredvalues:i. S ur /σ’ v or S ur versus CPT q c1N and/or SPT (N 1 ) 60 in loosesand-like materials (assume S ur /σ’ v ~ 0.1, if no dataavailable)ii. Use remolded undrained shear strength, S ur , for sensitiveclay-like materials measured from either CPT or FVTiii. 80% of peak undrained shear strength, S up , measuredusing either CPT or FVT in clay-like materialsiv. In zones where strength loss does not occur, use peakundrained shear strength, S up (or drained strength,whichever is lower)If Factor of Safety (FS) > 1.2, assume stability is acceptableand check deformationsIf FS < 1.2, evaluate material behavior and triggering in moredetail• For low risk structures, redesign or modify• For moderate and high risk structures, carry out moredetailed investigationb. If project risk is moderate to high risk and FS < 1.2, evaluatepost-earthquake shear strength in more detail:i. Additional in-situ testing, e.g. SCPT (seismic cone), FVT,geophysical logging.ii. High quality undisturbed samples and appropriatelaboratory testing.98

CPT Guide - 2006Cone Penetration Test (CPT)• If FS > 1.2, assume stability is acceptable and checkdeformations• If FS < 1.2, check triggeringIf layers/zones of low permeability exist that could inhibit porewater redistribution after seismic loading and promote voidredistribution, increase conservatism when evaluating postearthquake shear strengths. For high risk projects, the potential forvoid redistribution can be evaluated using complex effective stressnumerical models.3. Evaluate If Earthquake Will Trigger Strength LossWhen FS < 1.0, using best estimates of post-earthquake shearstrength parameters, assume that strength loss will be triggered.When 1.0 < FS < 1.2 using best estimates of post-earthquake shearstrength parameters, check if earthquake will trigger strength lossby applying either of the following approaches:a. Pore-pressure approach, using CRR (Youd et al. 2001)b. Strain-based approach (Castro, 1999)c. Yield-strength approach (Sladen , 1985, Olsen and Stark, 2003)All approaches should be based on improved knowledge ofmaterials based on combined results from in-situ tests andappropriate laboratory testing on high quality samples. When soilsare susceptible to strength loss and slopes are steep, a triggermechanism should always be assumed to be present (Silvis and deGroot, 1995). Hence, for high risk structures caution andconservatism should be exercised.If one or more zones are not expected to trigger strength loss, reevaluatestability using higher shear strength in these zones.• If FS > 1.2, assume stability is acceptable and checkdeformations.• If FS < 1.2 assume unsafe and redesign or modify.99

CPT Guide - 2006Cone Penetration Test (CPT)4. Evaluate Seismic DeformationsIf embankment is considered stable, evaluate seismic deformationsbased on size and duration of earthquake shaking.a. Preliminary screeningi. If no initial liquefaction is identified and earthquake issmall, assume deformations are small.b. Pseudo-static analysisi. If earthquake is small, M < 7, and,ii. If no significant zones indicate initial liquefaction, and,iii. Small deformations (less than 3 feet) are not significant tothe performance of the embankmentUse limit equilibrium stability analyses using pseudo-staticseismic coefficient of 0.5 PGA and 80% of peak undrainedstrength for clay-like and sand-like materials (but not toexceed 80% of drained strength).• If 1.0 < FS < 1.2, deformations are likely to be less than1m (3 feet).c. Newmark-type analyses (no seismic liquefaction/cyclicmobility)Perform a Newmark-type analysis if no zones of materialindicate initial liquefaction or cyclic mobility.d. Numerical modeling (seismic liquefaction/cyclic mobility)Perform appropriate numerical analyses (finiteelement/finite difference) that incorporate specialprovisions for pore pressure build-up.100

CPT Guide - 2006Cone Penetration Test (CPT)Ground Improvement Compaction ControlGround Improvement can be in many forms depending on soil type andproject requirements. For non-cohesive soil such as sands, silty sands andso on, deep compaction is a common ground improvement technique.Deep compaction can comprise: vibrocompaction, vibroreplacement,dynamic compaction, compaction piles, and deep blasting.The CPT has been found to be one of the best methods to monitor anddocument the effect of deep compaction due to the continuous, reliableand repeatable nature of the data. Most deep compaction techniquesinvolve cyclic shear stresses in the form of vibration to induce an increasein soil density. Vibratory compaction is generally more effective in soildeposits with a friction ratio less than 1%. When the friction ratioexceeds about 1.5% vibratory compaction is usually not effective. Theserecommendations apply to average values in a soil deposit. Local seamsor thin layers with higher friction ratio values are often of little practicalimportance for the overall performance of a project and their effect shouldbe carefully evaluated when compaction specifications are prepared. Thezone of soil behavior where vibratory compaction is most applicable isshown on the CPT soil behavior charts in Figure 37. Soils with an initialcone resistance below about 3 MPa (30 tsf) can be highly compressible orcontain organic matter, silt or clay and will not respond well to vibratorycompaction. Soils with a high initial cone resistance will not showsignificant compaction and generally do not need compaction. It is alsoimportant to establish the level and variation of the groundwater tablebefore compaction since some compaction methods are less effective indry or partially saturated soils. The CPTU provides the requiredinformation on groundwater conditions.Often the aim of deep compaction is for one or more of the following:• increase bearing capacity (i.e. increase shear strength)• reduce settlements (i.e. increase stiffness)• increase resistance to liquefaction (i.e. increase density).The need for deep compaction and geotechnical conditions will be projectspecific and it is important that design specifications take account of thesesite specific requirements. Cone resistance in cohesionless soils isgoverned by many factors including soil density, in-situ stresses, stress101

CPT Guide - 2006Cone Penetration Test (CPT)history, and soil compressibility. Changes in shear strength, stiffness anddensity can be documented with changes in measured cone resistance.A common problem in many deep compaction projects is to specify aminimum value of q c for compaction over a large depth range. Thisresults in a variation of relative density with depth, with the requireddegree of compaction near the surface being much higher than at depth.For certain projects, a high degree of compaction close to the groundsurface may beFigure 37 Guidelines for soils suitable for vibrocompaction techniques.justified. However, this can be very difficult to obtain with certain deepcompaction techniques and this decision should be based on engineeringjudgment related to the geotechnical project requirements. It is generallypreferred to specify a minimum normalized value of cone resistance102

CPT Guide - 2006Cone Penetration Test (CPT)corrected for overburden stress, q c1 . Since, grain characteristics can varyrapidly in many sandy deposits, it is preferred to specify an acceptancecriteria based on normalized clean sand equivalent values of coneresistance,(q c1N ) cs , using the methodology shown in Figure 34, especiallywhen compaction is performed to reduce the potential for liquefaction.Specification using (q c1N ) cs can reduce problems in silty zones, wheretraditional approaches have often resulted in excessive groundimprovement in an effort to reach unrealistic criteria.An important aspect of deep compaction which is not yet fullyunderstood, is the increase in cone resistance with time after compaction.This time effect has been observed in different ground conditions andwith different compaction methods. Often no measurable change in porepressure has been observed and the increase takes place without visibleground settlements. Charlie et al. (1992) studied a number of cases wherecone resistance was measured with time after compaction. A range ofcompaction techniques were used and the results are shown in Figure 38.The cases were representative of a wide range of climates and geologicconditions with average temperatures varying from -10 o C (Beaufort Sea)to +27 o C (Nigeria). Charlie et al. (1992) suggested that the time effectcould be linked to the average air temperature. The possibility of timeeffects should be evaluated for each project. For very large projects, itmay be necessary to perform field trials.103

CPT Guide - 2006Cone Penetration Test (CPT)Figure 38 Influence of time after disturbance on CPT results(After Charlie et al., 1992)For projects where deep compaction is recommended to either increaseresistance to liquefaction or decrease future settlements for shallowfoundations, the seismic CPT should be considered, since it provides bothpenetration resistance and shear wave velocity. The combined values canimprove interpretation, especially in silty sands.Ground improvement can also include many other techniques, such asgrouting, soil mixing and stone columns as well as pre-loading. The CPTcan also be used to evaluate the effectiveness of these other techniquesalthough this will depend on soil conditions and the ground improvementmethod. The CPT has also found some limited use in monitoring surfacecompaction. Since surface compaction is often carried out in thin layerswith frequent quality control, the CPT has not found extensive applicationin this area.104

CPT Guide - 2006Cone Penetration Test (CPT)Design of Wick or Sand DrainsPre-loading is a common form of ground improvement in fine grainedsoils where the rate of consolidation is important. Installation of sanddrains or wick drains can significantly decrease the time for consolidationsettlements. Prior to 1975, vertical sand drains were common to aidconsolidation with temporary pre-load. Since 1975, geosynthetics in theform of wick drains have dominated the market. Wick drains are usuallyfluted or corrugated plastic or cardboard cores within geotextile sheathsthat completely encircle those cores. They are usually 100 mm wide by 2to 6 mm thick. The wick drain is usually pushed or driven into the groundto the desired depth using a lance or mandrel. The drain then remains inplace when the lance or mandrel is removed. Installation can be in therange of 1 to 5 minutes depending on ground conditions, pushingequipment and depth of installation. The design of wick drains is notstandardized but most attempt to equate the diameter of the particular typeof drain to an equivalent sand drain diameter.The method developed by Hansbo (1970) is commonly used, and therelevant design equations are as follows:Dt =8c2h[ ln(D / d) − 0.75]11n1 − UWhere: t = consolidationc h = coefficient of consolidation for horizontal flowd = equivalent diameter of the wick drain( −~ circumference/π)D = sphere of influence of the wick drain (for a triangular patternuse 1.05 times the spacing, for a square pattern use 1.13 times thespacing).U = average degree of consolidationThe key input parameter for the soil is the coefficient of consolidation forhorizontal flow, c h . This parameter can be estimated from dissipationtests using the CPTU. The value derived from the CPTU is particularlyuseful since, the cone represents a very similar model to the installation105

CPT Guide - 2006Cone Penetration Test (CPT)and drainage process around the wick drain. Although there is somepossible smearing and disturbance to the soil around the CPT, similarsmearing and disturbance often exists around the wick, and hence, thecalculated value of c h from the CPTU is usually representative of the soilfor wick drain design.Details on estimation of c h from dissipation tests were given in the sectionon (geotechnical parameters) consolidation characteristics. To provide areasonable estimate of c h a sufficient number of dissipation tests should becarried out through the zone of interest. The dissipation tests should becarried out to at least 50% dissipation. Several dissipation tests should becarried out to full dissipation to provide an estimate of the equilibriumgroundwater conditions prior to pre-loading.106

CPT Guide 2006Main ReferencesMain ReferencesLunne, T., Robertson, P.K., and Powell, J.J.M 1997. Cone penetrationtesting in geotechnical practice, E & FN Spon Routledge, 352 p, ISBN0-7514-0393-8.Jefferies, M.G. and Davies, M.P. 1993. Estimation of SPT N values fromthe CPT, ASTM.Robertson, P.K. 1990. Soil classification using the cone penetration test.Canadian Geotechnical Journal, 27 (1), 151-8.Robertson, P.K., Campanella, R.G. and Wightman, A. 1983. SPT-CPTCorrelations, ASCE J. of Geotechnical Engineering 109(11): 1449-59.Baligh, M.M. and Levadoux, J.N. 1986. Consolidation after undrainedpiezocone penetration, II: Interpretation, J. of Geotech. Engineering,ASCE, 112(7): 727-745.Jamiolkowski, M., Ladd, C.C., Germaine, J.T. and Lancellotta, R. 1985.New developments in field and laboratory testing of soils, State-ofthe-artreport, Proceedings of the 11 th International Conference onSoil Mechanics and Foundation Engineering, San Francisco, 1, 57-153, Balkema.Robertson, P.K., Sully, J.P., Woeller, D.J., Lunne, T., Powell, J.J.M. andGillespie, D.G. 1992. Estimating coefficient of consolidation frompiezocone tests, Canadian Geotechnical J., 29(4): 551-557.Teh, C.I. and Houlsby, G.T. 1991. An analytical study of the conepenetration test in clay, Geotechnique, 41(1): 17-34.Robertson, P.K. and Wride, C.E., 1998. Cyclic Liquefaction and itsEvaluation based on the CPT Canadian Geotechnical Journal, 1998,Vol. 35, August.107

CPT Guide - 2006Main ReferencesYoud, T.L., Idriss, I.M., Andrus, R.D., Arango, I., Castro, G.,Christian, J.T., Dobry, R., Finn, W.D.L., Harder, L.F., Hynes, M.E.,Ishihara, K., Koester, J., Liao, S., Marcuson III, W.F., Martin, G.R.,Mitchell, J.K., Moriwaki, Y., Power, M.S., Robertson, P.K., Seed, R.,and Stokoe, K.H., Liquefaction Resistance of Soils: Summary Reportfrom the 1996 NCEER and 1998 NCEER/NSF Workshop onEvaluation of Liquefaction Resistance of Soils, ASCE, Journal ofGeotechnical & Geoenvironmental Engineering, Vol. 127, October, pp817-833Zhang, G., Robertson. P.K., Brachman, R., 2002, EstimatingLiquefaction Induced Ground Settlements from the CPT, CanadianGeotechnical Journal, 39: pp 1168-1180Zhang, G., Robertson. P.K., Brachman, R., 2004, EstimatingLiquefaction Induced Lateral Displacements using the SPT and CPT,ASCE, Journal of Geotechnical & Geoenvironmental Engineering,Vol. 130, No. 8, 861-871108

GEOTECHNICAL SERVICESGregg Drilling & Testing, Inc. (Gregg) offers a wide range of services for environmental,geotechnical, and marine site investigation. The Gregg professionals have the training and experienceto fully understand the needs of each client, identify the unique job requirements, and provide thehighest quality services.Geotechnical Capabilities:▪ Cone Penetration Testing (CPT)▪ Seismic Cone Penetration Testing▪ Mud Rotary and Hollow StemAuger Drilling▪ Soil and Rock Coring▪ SPT Energy Calibration▪ Diamond Coring▪ Soil Coring▪ Vane Shear Test▪ Pressuremeter Testing▪ Dilatometer Testing▪ Vibra Core SamplingQ uality isn’t something we test for after the job is done. It’s a way of doing business,an integral part of our daily activities that we instill in our personnel, something we build into ourtechnologies, a guarantee of value that we deliver with all our services.Q U A L I T Y • S A F E T Y • V A L U E

ENVIRONMENTAL SERVICESEnvironmental Capabilities:▪ Well Installation, Development & Abandonment▪ Groundwater Extraction & Monitoring Wells▪ Vapor Extraction, Sparge & Bio Vent Wells▪ Nest Wells, Cluster Wells,Multipoint Sampling Systems▪ Cone Penetration Testingwith UVIF & Resistivity▪ Hollow Stem Auger& Mud Rotary Drilling▪ Direct Push Soil, Groundwater& Vapor Sampling▪ Air Vacuum Excavation▪ Well Rehabilitation(submersible & turbine)▪ Pump Installation & Repair▪ Video Logging▪ Hydraulic FracturingProtecting our workforce, our clients, and the general public from injuries and hazardousincidents that may be harmful and disruptive is a core element of our corporate culture. Becauseof this we have dedicated ourselves to a “safety first” philosophy which applies to all areas ofoperation and is put into practice by all Gregg personnel on a daily basis. This philosophy hasallowed us to effectively manage the risks inherent in our industry and consistently provide a safe,sensible, and productive work environment.Q U A L I T Y • S A F E T Y • V A L U E

MARINE SERVICESMarine Capabilities:▪ Mud Rotary Drilling▪ Cone Penetration Testing (CPT)▪ Mini-Cone Penetration Testing▪ Seismic CPT▪ Downhole Vane Shear Testing (VST)▪ Micropile Installation▪ Vibra Core Sampling▪ Piston Sampling▪ Gravity CoringPlatforms Offered:▪ Quin Delta Drill Ship ▪ Tugboats▪ Jack-up Boats & Barges ▪ Support Vessels▪ Sectional BargesTesting Equipment:▪▪▪▪▪Mobile B-80/22 and B80/14 Mud Rotary Drill RigMiniature CPT System (2 cm 2 cone) - depths of up to 5000 feetCPT Portable System (10 cm 2 & 15 cm 2 cone, drill rig deployed)Vibracore Sampling System - up to 20 feet of coreGravity Core Sampling System - up to 5 feet of coreQ U A L I T Y • S A F E T Y • V A L U E

SPECIALIZED SERVICESGregg acquires geotechnical and environmental data using a variety of unique in-houseequipment, tools, sensors, and techniques ranging from cone penetration testing and geophysical methodsto conventional drilling procedures and borehole tests. In addition we offer an array of highly specializedservices including:▪ CPT with UVIF & Resistivity▪ Seismic CPT▪ SPT Energy Calibration▪ Vane Shear Testing▪ Pressuremeter Testing▪ Dilatometer Testing - Flat & Rock▪ Settlement Monitoring▪ Goodman Jack Testing▪ CPT Data Cross SectionsOur advanced data collection techniques, equipment, and monitoring systems combined withthe expertise of our personnel enables us to provide exceptional performance and value time after time.Q U A L I T Y • S A F E T Y • V A L U E

6 Feet10 FeetCOMMUNICATIONCATV

GREGG DRILLING & TESTING, INC.CORPORATE HEADQUARTERS2726 WALNUT AVENUE; SIGNAL HILL, CALIFORNIA 90755Tel: (562) 427-6899 Fax: (562) 427-3314www.greggdrilling.com• NORTHERN CALIFORNIA• SOUTHERN CALIFORNIASOUTHEAST•SOUTHWEST•SOUTHERN CALIFORNIASignal Hill, California(Corporate Headquarters)(562) 427-6899NORTHERN CALIFORNIAMartinez, California(925) 313-5800SOUTHWESTLivingston, Texas(936) 327-4004SOUTHEASTMonroe, Georgia(770) 207-9722QUALITY • SAFETY • VALUE