The IT earthing system (unearthed neutral) in LV

The IT earthing system(unearthed neutral) in LVAlthough all Earthing Systems offer users the same degree of safety, theydo not all have the same operating characteristics.This is why, in certain countries, a specific earthing system is stipulated bylegislation or standards according to buildings. For example, in France theIT system is compulsory in hospital operating theatres, and the TN-C isforbidden in premises where there is a risk of explosion.These stipulations apart, dependability objectives (safety, availability,reliability, maintenability and proper operation of low currentcommunication systems) determine which earthing system should bechosen for which installation.The aim of this “Cahier Technique” is to describe the advantages andareas of application of the IT earthing system.After a brief introduction of the electrical hazard and the various earthingsystems, the first fault situations, followed by the double fault specific to theIT system, are studied, and the advantages and disadvantages of thisparticular earthing system are developed. This “Cahier Technique” alsooffers solutions for the surge limiter with the various types of possibleovervoltages.Finally, a choice table is provided for all earthing systems, based on criteriafor safety, availability, electromagnetic compatibility and operators’professional requirements.Contents1 Introduction 1.1 Protection of persons against electrical shocks p. 41.2 The various standardised earthing systems p. 41.3 Choosing an earthing system p. 71.4 Type of insulation p. 71.5 Equivalent system for an unearthed or impedance-earthedneutral network p. 82 The 1 st insulation fault with the 2.1 Calculating fault currents and contact voltage p. 9IT earthing systemon the first fault2.2 Permanent insulation monitors, history and principles p. 112.3 Tracking the 1 st insulation fault p. 133 The 2 nd insulation fault with the 3.1 Analysis of the double insulation fault p. 15IT earthing system 3.2 Elimination of the double insulation fault p. 164 Special features of the IT earthing system 4.1 Overvoltages in the IT system p. 184.2 Surge limiters p. 204.3 Why use an impedance? p. 215 Advantages and disadvantages 5.1 Increased availability p. 22of the IT earthing system in LV 5.2 Increased safety against risk of fire p. 225.3 Less downtime on control and monitoring circuits p. 235.4 Usage limits and precautions of the IT earthing system p. 236 Conclusion 6.1 Availability: an increasing need to be satisfied p. 266.2 The IT earthing system finds its true place p. 266.3 The added advantage of safety p. 266.4 In short p. 27Bibliography p. 28Cahier Technique Schneider Electric no. 178 / p.3

1 Introduction1.1 Protection of persons against electric shocksUse of Safety by Extra Low Voltage (< 25 V)-SELV- is the most drastic solution since iteliminates the electrical hazard. However it isapplicable only in low power distribution.Regarding everyday use of electricity, a number ofstudies have identified the causes of electric shocksand provided specific solutions for each one.Electric shocks have two causes, namely:c direct contact, i.e. a person or an animaltouching an exposed live conductor;c indirect contact, i.e. a person touching themetal frame of an electrical load on which aninsulation fault has occurred.Protection against direct contactTo provide protection against direct contact,insulation and/or distancing measures are taken.These measures can be reinforced in finaldistribution by additional protection in the form ofa high sensitivity Residual Current Device(RCD).Protection against indirect contactWith respect to protection against indirectcontact, between an accidentally energisedframe and the earth, the basic solution is to earthall the load frames via the protective conductors.However, this measure does not rule out theexistence of a contact voltage hazardous forpersons if it exceeds the conventional limit safetyvoltage U L defined by standard IEC 60479.This contact voltage depends on the earthingsystems defined in the international standardIEC 60364.1.2 The various standardised earthing systemsThe three earthing systems given official statusby international standards (IEC 60364) are alsostipulated by a large number of nationalstandards: in France by the LV installationstandard: NF C 15-100.A brief reminder of the protection principle ofthese systems will now be given beforedescribing the IT system in greater detail.The TN systemc Its principle:v the transformer neutral is earthed;v the electrical load frames are connected toneutral.This type of system has three possibilities:v the same conductor acts as a neutral and aprotective conductor: this is the TN-C system;v the neutral and the protective conductor areseparate: this is the TN-S system;v and the combination of these two systems,known as TN-C-S when the neutral and theprotective conductor are separated downstreamof part of the installation in the TN-C system.Note that the TN-S cannot be placed upstreamof the TN-C.c Its operation (see fig. 1 ):An insulation fault on a phase becomes a shortcircuitand the faulty part is disconnected by aShort-Circuit Protection Device (SCPD).The TT systemc Its principle:v the transformer neutral is earthed;v the electrical load frames are also earthed.c Its operation (see fig. 2 ):The current of an insulation fault is limited byearth connection impedance.Cahier Technique Schneider Electric no. 178 / p.4

a) b)I dI dNNPENSCPDPESCPDR BU dR dR BU dR dc)I d12N3NPENPEPESCPDR BU dR dFig. 1 : insulation fault on a network operated in TN-C [a], TN-S [b] and TN-C-S [c].NRCDI dProtection is provided by the Residual CurrentDevices (RCD): the faulty part is disconnectedas soon as the threshold I∆n, of the RCD placedupstream, is overshot by the fault current, so thatI∆n R B iU L .R B R AU dR dFig. 2 : insulation fault on a network operated in TT.The IT systemc Its principle:v the transformer neutral is not earthed, but istheoretically unearthed. In actual fact, it isnaturally earthed by the stray capacities of thenetwork cables and/or voluntarily by a highimpedance of around 1,500 Ω (impedanceearthedneutral);v the electrical load frames are earthed.Cahier Technique Schneider Electric no. 178 / p.5

c Its operation:v should an insulation fault occur, a low currentdevelops as a result of the network’s straycapacities (see fig. 3a ).The contact voltage developed in the frameearth connection (no more than a few volts) isnot dangerous;v if a second fault occurs on another phasebefore the first fault has been eliminated(see fig. 3b and 3c ), the frames of the loads inquestion are brought to the potential developedby the fault current in the protective conductor(PE) connecting them. The SCPDs (for theframes interconnected by the PE) or the RCDs(for the frames with separate earth connections)provide the necessary protection.This deliberately brief presentation of the variousearthing systems clearly cannot cover all thespecific installation possibilities. Readersrequiring more details can consult “CahiersTechniques” no. 114, 172 and 173.a)N321Permanentinsulationmonitor(PIM)Z N : optionalimpedanceZ NSurgelimiterI dI dI I dNI CR BU dC1 C2 C3PEI C1 I C2 I C3I Cb)Permanentinsulationmonitor(PIM)NSurgelimiterI dI dSCPDI dI dSCPD321NPEUd2R d2U d1R d1R Bc)Permanentinsulationmonitor(PIM)NSurgelimiterI d RCDPESCPDI dI dSCPD321NPER BI dUd2R d2R AU d1R d1I dFig. 3 : single [a] and double [b and c] insulation fault on a network operated in IT.Cahier Technique Schneider Electric no. 178 / p.6

1.3 Choosing an earthing systemAlthough all three earthing systems offer users thesame degree of safety against indirect contact,only the IT system guarantees risk-free continuityof supply in the presence of an insulation fault.This undeniable advantage also has certaindrawbacks: for example the need to locate thisfirst fault and the possibility of overvoltagesoccurring that may affect operation of sensitiveloads.However, choice of earthing system for aninstallation also depends on parameters otherthan safety of persons and continuity of supply,namely:c the environment (e.g. premises with a risk offire or sites frequently struck by lightning);c electromagnetic compatibility (EMC) (presencein the installation of harmonics and radiatingfields, and sensitivity of equipment to suchphenomena);c technicity of installation designers andoperators;c maintenance quality and cost;c network size;c etc.Although consideration of the above parametersguarantees choice of the earthing system mostsuited to the installation, it should be emphasisedthat the advantage offered by the IT system interms of availability (2 nd fault most unlikely)generates installation and operating costs thatshould be compared with the downtime costsgenerated by other earthing systems (operatinglosses and repair costs caused by the firstinsulation fault).1.4 Type of insulationCommon mode impedanceAll electrical networks have an impedance withrespect to earth known as the “common modeimpedance”, the origin of which is insulation ofnetwork cables and loads. This impedanceconsists of the leakage capacity and resistancebetween each live conductor and the earth.In LV, the leakage resistance of a new cable isaround 10 MΩ per kilometer and per phase,whereas its capacity evenly distributed withrespect to earth is approximately 0.25 µF, i.e.12.7 kΩ at 50 Hz.It should also be noted that in MV and HV thisleakage capacity is even greater and MUST betaken into account when drawing up a protectionplan (see “Cahier Technique” no. 62).Loads also have a natural leakage capacity,usually negligible.Effect of distributed capacity in the IT systemIn electrical installations, other capacities areadded to the network cable ones. This is thecase of certain electronic loads that aregenerators of HF harmonic currents, in particularwhen they use the chopper principle (e.g. pulsewith modulation converters). However,ElectroMagnetic Compatibility (EMC) standardsstate that these HF currents must be shunted toearth, resulting in the presence of filters and thuscapacitors between phases and frame.According to the number of loads, theircontribution to the network’s “leakage” capacitycan be significant or even important.Measurements taken on a variety of electricalpower networks show that capacity variesconsiderably from network to network andcovers a range of a few µF to a few dozen µF.Excessively high capacities may question theadvantage of the IT earthing system: if, on thefirst fault, the value of network impedance withrespect to earth means that contact voltageexceeds 50 V, safety of persons is notguaranteed. This is rare, however, as with a10 Ω earth connection, the network’s earthleakage capacity must exceed 70 µF (23 µF perphase if the neutral is not distributed).An IT network must therefore have a limitedcapacity with respect to earth, and the presenceof loads equipped with HF filters must be takeninto account in the network design stage.Cahier Technique Schneider Electric no. 178 / p.7

1.5 Equivalent system for a network with unearthed or impedance-earthed neutralA few definitions and assumptions are givenbelow in order to define the equivalent systemfor this network (see fig. 4 ):c the neutral point is unearthed or earthed by animpedance (Z N ) of high value (normally 1 kΩ to2kΩ) whose earth connection is equivalent to aresistance (R B );c the load frames are interconnected either fullyor by group. For EMC reasons (see“Cahier Technique” no. 187), it is advisable tointerconnect all the application frames of thesame installation and to connect them to thesame earth connection (resistance R A );c the earth connections (R A and R B ) areinterconnected (in most cases), or separate.NB: Two earth connections are considered to beseparate if they are more than 8 m apart;c each live conductor has, with respect to earth,an impedance made up of a resistance and acapacity.N321NPEZ NRN R1 R2 R3CN C1 C2 C3R BR AFig. 4 : equivalent system of a network with unearthed or impedance-earthed neutral.Cahier Technique Schneider Electric no. 178 / p.8

2 The 1 st insulation fault with the IT earthing systemIn normal operating conditions, safety of personsis guaranteed when contact voltage is less than50 V as per standard IEC 60364 (NF C 15-100).When this contact voltage is exceeded, thesestandards require automatic opening of thecircuit. The following section shows how use ofthe IT earthing system for network operationprevents tripping on the first insulation fault.2.1 Calculating fault currents and contact voltage on the first faultGeneral case (resistive fault)Should a fault with a resistive value Rd occurbetween phase 3 and the earth, a fault current I dflows in the neutral impedance and in thecapacities C1, C2 and C3 (see fig. 3a).Assuming that the phase-to-earth capacities arebalanced (C1 = C2 = C3 = C), the fault currenthas the following value:1+ 3j CωZI d =UN0Rd+ ZN + 3jCωZNRdThe capacitive current is written as:3j CωZI c =UN0Rd+ ZN + 3jCωZNRdand the current in the impedance Z N :UI N =0Rd+ ZN + 3jCωZNRdThe contact voltage U C (contact voltage betweenthe frame of a faulty device and another frame orthe earth) is calculated from the fault current I dflowing in the earth connection resistance R A ofthe application frames if they are notinterconnected, else R B (only network earthconnection):U C = R A I d .Case of the full faultThis paragraph calculates the configurationgenerating the highest contact voltage (U C ):thus for a fault occuring on a frame with earthconnection separate from that of Z N .By application of the above formulae, whereR d = 0, we obtain:UI d = 0ZN+ 3j CωUUc = R 0AZ N + 3j CωThe capacitive current is equal to:I C = +3j Cω U 0and the current in impedance Z N :UI N = 0ZNIn the various examples below, studied forZ N = ∞ (unearthed neutral) and Z N = 1 kΩ(impedance-earthed neutral), the calculationsare made for a network in the IT system,400 VAC (U 0 = 230 V), where:R A , earth connection resistance = 10 ΩR d , insulation fault value = 0 to 10 kΩ.c Case 1:Low capacity network (e.g. limited to anoperating theatre)C1 = C2 = C3 = C = 0.3 µF per phase.c Case 2:Power network, whereC1 = C2 = C3 = C = 1.6 µF per phase.c Case 3:Very long power network, whereC1 = C2 = C3 = C = 10 µF per phase, i.e.roughly 40 km of cables!The results of all these calculations, grouped inthe table in figure 5 , confirm the low faultvoltage (≈ 20 V in the most unfavourable cases),ensuring continuity of operation, without risk forpersons, of a network designed using the ITsystem. They prove that addition of animpedance between the neutral and the earthhas very little effect on contact voltage.R d (kΩ) 0 0.5 1 10Case 1 Z N = ∞ U C (V) 0.72 0.71 0.69 0.22C R =1µF I d (A) 0.07 0.07 0.07 0.02Z N = 1 kΩ U C (V) 2.41 1.6 1.19 0.21I d (A) 0.24 0.16 0.12 0.02Case 2 Z N = ∞ U C (V) 3.61 2.84 1.94 0.23C R =5µF I d (A) 0.36 0.28 0.19 0.02Z N = 1 kΩ U C (V) 4.28 2.53 1.68 0.22I d (A) 0.43 0.25 0.17 0.02Case 3 Z N = ∞ U C (V) 21.7 4.5 2.29 0.23C R =30µF I d (A) 2.17 0.45 0.23 0.02Z N = 1 kΩ U C (V) 21.8 4.41 2.26 0.23I d (A) 2.18 0.44 0.23 0.02Fig. 5 : comparison of fault currents and contactvoltages on a first fault.Cahier Technique Schneider Electric no. 178 / p.9

The curves in figure 6 representing these resultsshow the considerable effect of network capacityon the value of U C .In point of fact, regardless of the distributedcapacity of the sound network or network onwhich a first fault has occurred, users can becertain that this voltage will always be less thanthe conventional safety voltage and thus withoutrisk for persons. Also, the currents of a first fullfault are low and thus have minimum destructiveor disturbing (EMC) effect.Effect of distributed capacities, vector chartand neutral potentialc Effect of distributed capacities on a soundnetworkThe capacities of all 3 phases create an artificialneutral point. In the absence of an insulation fault,if network capacities are balanced, this neutralpoint is then at earth potential (see fig. 7 ).In the absence of a fault, the phase-to-earthpotential is thus equal to phase to neutralvoltage for each phase.U c (V) whereZ n = 1,000 Ω10050C R = 70 µFC R = 30 µF10C R = 5 µF1C R = 1 µF0.1 R d (Ω)1 10 100 500 1,000 10 4(recommended threshold)Fig. 6 : contact voltage on a first insulation fault is always less than safety voltage.11Artificial neutral32C C C 3C3 2Fig. 7 : the network’s distributed capacities create a connection between neutral and earth.Cahier Technique Schneider Electric no. 178 / p.10

c Vector chart in presence of a full faultIn event of a full fault on phase 1, the potential ofphase 1 is at earth potential (see fig. 8 ).The neutral-to-earth potential is thus equal tophase to neutral voltage V1, and that of phases2 and 3 with respect to earth is equal tophase-to-phase voltage. If the neutral isdistributed, the fault current is arithmeticallyincreased:I C = 4j Cω V1.However, detection, location and correction ofthis fault must be immediate in order to reducethe risk of a second simultaneous fault occurringwhich would result in opening of the faulty circuits.a) b)23V 1NvN2v 3V 3 V 2II C3 d1T I V 1-T I = I V 3-T I = I V 2-T I TI C2V 1-T = 0V 3-T = V 1 + V 3V 2-T = V 1 + V 2I d = I C = I C2 + I C3I C2 = j C ω v 2I C3 = j C ω v 3I C = 3j C ω V 1I I d I = 3 C ω I V 1 IFig. 8 : vector charts of a network in the IT system, without fault [a] and with an earth fault on phase 1 [b].2.2 Permanent insulation monitors, history and principlesThe first LV electrical distribution networks wereoperated using the IT earthing system.Operators rapidly sought to detect the presenceof the first insulation fault in order to prevent thehazards linked to a short-circuit current ofvarying impedance and the de-energisation of afaulty feeder (with the lowest rating protection) orof the two faulty feeders.The first PIMsThese devices used 3 lamps connected betweenthe phases and the earth (see fig. 9 ).On a sound network, the three lamps form abalanced three-phase load, all lit and with thesame brilliance. When an insulation fault occurs,one of the three lamps is short-circuited by thefault impedance. Voltage is reduced at theterminals of this lamp, and lamp brilliancedecreases. However, voltage at the terminals ofthe other two lamps increases until phase-tophasevoltage is reached. Their luminosity alsoincreases.This system is easy to install and use. However,given that its practical operating threshold is low,attempts were quickly made to try to detectimpedant faults in order to anticipate the full fault.For a DC network(supplied by batteries or by DC generator).The technique of the voltmeter balance(see fig. 10 ) was the first to be used, andindeed is still used today.(1)(2)(3)The OFF indicatorlight indicates thefaulty phase:in this case no. 3.(+) (-)RThe needle indicatesthe faulty polarity; in thiscase the (-) polarity.RFig. 9 : principle of the first PIM.Fig. 10 : principle of the PIM with voltmeter balance.Cahier Technique Schneider Electric no. 178 / p.11

The principle consists of measuring andcomparing voltages between the (+) polarity andthe earth, on the one hand, and the (-) polarityand the earth on the other. This principle makesauxiliary power sources unnecessary, since thenetwork supplies the PIM directly viameasurement sensors (resistances).This technique applies to two-phase AC and DCnetworks and does not allow live fault tracking.For AC networksThe most commonly used PIM are those withinsulation measurement by DC current injection.Permanent measurement of insulationresistance required use of active systems inplace of the previously used passive systems.This resistance can be measured accurately inDC (see fig. 11 ), which is why the first PIMs,placed between the network and the earth,injected a low DC current which flowed throughthe fault. This simple, reliable technique is stillextensively used today, but does not allow livefault tracking.Note that when these PIMs are used on mixednetworks (containing rectifiers without galvanicinsulation), they may be disturbed or evenVNI PIM321NPEI PIMR BFig. 11 : principle of PIM with current injection.“blinded” if a fault is present on the DC part ofthe network.These were followed by PIMs with AC currentinjection at low frequency (< 10 Hz), operatingon the same principle. Although these PIMsallow live fault tracking, they can be “misled” bycable capacities that are seen as insulationfaults and disturbed by frequency converters(variable speed controllers).For all AC and DC networksFinally, nowadays, given that networks arefrequently of the mixed AC/DC kind as well asvariable frequency, the new PIMs are able tomonitor insulation on all types of networks.c Some use squared wave pulses at very lowfrequency (≈ 1Hz). They allow PIM not to bedisturbed by earth leakage capacities, as theyare then immediately loaded then unloaded bythe next strobe pulse of opposite sign. They areuniversal in use and easily adapted to modernnetworks, in particular to those supplying powerelectronic devices which often deform the ACpulse. However, their response time, dependingon the network’s earth leakage capacity, may beas much as a few minutes and does not allowthe detection of intermittent faults.c In order to compensate the usage restrictionsof these PIMs for very long networks andnetworks with a large number of capacitivefilters, the low frequency AC current injectiontechnique has been improved by means of“synchronous demodulation” (see fig. 12 ): thistype of PIM applies a low frequency AC voltagebetween the network and the earth, measuresthe current flowing back via network insulationimpedance and calculates the voltage-currentshift.It is then possible to determine the resistive andcapacitive components of this current and thusrelate the threshold to the resistive componentonly. This upgrade, the result of digitaltechnology, combines the advantages of DCcurrent and low frequency AC current injectionwithout their disadvantages.I BFBF~mAVI R-BF= U BFU BFR NetworkI C-BFC NetworkI C-BFI BFZ NetworkI R-BFFig. 12 : the low frequency AC current injection technique has been improved by means of “synchronousdemodulation”, which enables the insulation drop (resistive leakage) to be distinguished from capacitive leakage.Cahier Technique Schneider Electric no. 178 / p.12

PIM standardsc The manufacturing standardsStandard IEC 61557-8 is in existence sinceFebruary 1997. It defines the specialspecifications governing insulation monitorsdesigned for permanent monitoring, irrespectiveof the measurement principle, of insulationresistance with respect to earth of unearthed ACand DC IT system networks, and of AC ITsystem networks containing rectifiers suppliedwithout galvanic separation (transformer withseparate windings).It places particular emphasis on three points.v Properly inform specifiers and contractors. Themanufacturer must provide the characteristics ofthe devices he produces and in particular thosethat are dependent on network capacity(response time and threshold values).v Ensure that these devices are properlyintegrated in their electrical environment. Thisrequires compliance with the specifications ofstandards IEC 61326-1 and 61326-10concerning ElectroMagnetic Compatibility(EMC).v Guarantee operating safety for users.The main stipulations are: device operatingtesting must be possible without inserting anadditional impedance between the monitorednetwork and the earth, settings must beprotected to prevent modification by error or byunauthorised users, and impossibility of devicedisconnection (the need to use a tool fordisassembly).c The operating standardsAs concerns PIM setting, standard IEC 60364provides an initial answer: “A PIM designedaccording to… is set at a value less than theminimum value of the insulation resistancedefined for the installation in question”, i.e.greater than or equal to 0.5 MΩ for a circuit witha nominal voltage greater than or equal to 500 V.Guide NF C 15-100 states: “…set at a valueroughly less than 20% of the resistance of theinstallation as a whole…”However, a clear distinction must be madebetween the insulation resistance of theinstallation, which only takes electricaldistribution into account, and the insulation levelwhich is set for overall network monitoring,including the various machines and switchgearconnected to it.In the previous chapter we saw that for faultsgreater than 500 Ω, contact voltage does notexceed 5 V with an earth connection of 10 Ω(see fig. 5). In practice, for a normal industrialinstallation, it is thus reasonable, without takingrisks, to set the lower alarm threshold at a valueof between 500 Ω and 1,000 Ω, ensuringeffective fault tracking (and thus location of thereported insulation fault). To organise preventivetracking, it is useful to have a first level thresholdaround 10 kΩ for example. This threshold mustbe adapted according to installationcharacteristics and operating requirements. Notethat short networks allow a higher preventionthreshold.2.3 Tracking the 1 st insulation faultWhen tracking a fault, although certain operatorsmerely identify the faulty feeder, accuratedetermination of the location of this fault isrecommended (e.g. damaged cable or insulationfault in a device) in order to put it right as quicklyas possible.Tracking by successive de-energisation offeedersThis means of fault tracking is quoted formemory only. It consists of opening the feedersone by one, beginning with the main feeders.When the faulty feeder is opened, the currentinjected by the PIM decreases markedly anddrops below the detection threshold. The audiblealarm normally controlled by the PIM then stops,enabling remote identification of the faultyfeeder.This procedure, which requires interruption ofoperation on each feeder, is contrary to theoperating philosophy of the IT earthing system,which stipulates continuity of supply. Althoughfrequently used in the past, it is graduallydisappearing with the development of the newfault tracking systems which allow live tracking(without power breaking).Live trackingc Detecting the fault currentAs seen above (see fig. 3a), a current I d flowsthrough the first insulation fault at the samefrequency as that of the network (50 Hz or60 Hz), returning to the source via the capacitiesof the other sound phases and via the neutralimpedance if any.An initial live tracking method (withoutinterrupting distribution) consisted of using aclamp-on probe to measure the earth “leakage”current on each feeder. The faulty feeder wasthe one on which the highest value wasmeasured.This method has two drawbacks, namely:v It is not reliable for networks with a largenumber of feeders some of which are highlycapacitive (how can the earth current of a shortfaulty feeder be distinguished from that of a longcapacitive feeder?).Cahier Technique Schneider Electric no. 178 / p.13

v It is not applicable on networks with fewcapacitive leakages (the fault current is virtuallyundetectable).In order to improve detection of the fault currentpath (at power frequency) using a clamp-onprobe, two “tricks” were used.The first consisted of increasing this fault currentby temporarily placing a low impedance inparallel on the PIM.The second consisted of distinguishing thecapacitive leakage currents from the fault currentby the periodic use of the above-quotedimpedance by means of a beating relay(approx. 2 Hz).c Detecting an injected currentThis method uses a low frequency sinusoidalpulse (i 10 Hz) injected by a generator or a PIM.Choice of low frequency for fault trackingensures no disturbance by network leakagecapacities, but this frequency cannot be lessthan 2.5 Hz, as detection with a magnetic sensorbecomes difficult. This method uses devicessensitive to the injected pulse only, that caneither be fixed with detection toroids placed onall feeders, or portable with a clamp-on probetuned to signal frequency in order to locate theexact position of the fault (see fig. 13 ).When the devices (generator, sensors and load)are fixed, live fault tracking can be automatic ondetection of a fault, with transmission of an orderby the PIM.c Measuring insulation of each feederOperators, with their ever-increasing need forcontinuity of supply, are no longer prepared evento wait for the first fault, but want to be able toprogramme maintenance work and thusanticipate the next feeder likely to be affected byan insulation fault.It is thus necessary to “monitor” the changes ininsulation of each feeder and to carefully identifythe resistive and capacitive insulation components.The synchronous demodulation principle canalso be used by measuring, first, the injectioncurrent flowing in the feeders (by the toroidsensors) and, second, the injection voltage.Development of this tracking method isencouraged by application of digital techniquesto the management of electrical power distribution(see “Cahier Technique” no. 186): the user cannow remotely and continually monitor insulationchanges of the various feeders. Use of digitalbuses enables data to be centralised on asupervisor, displayed and logged, thus allowingintelligent, predictive maintenance.Fixed load with manualor automatic scanningLF generator(G)GPIM ( )N12 3 N/ / / /PER BManual loadFig. 13 : tracking can take place with fixed or portable devices sensitive to the injected pulse.PECahier Technique Schneider Electric no. 178 / p.14

3 The 2 nd insulation fault with the IT earthing systemAs we have already seen in the previouschapter, the advantage of using the IT system innetwork operation lies in the possibility ofcontinuity of supply even though an insulationfault has occurred on a circuit.This message has been received loud and clearby standard drafters who, in order to maintain ahigh level of availability, stipulate in installationstandards indication and tracking of this first faultso as not to fear a second fault. Protectiondevices are also provided for this second fault inorder to guarantee the same level of safety ofpersons as for the TN and TT earthing systems.The two sections below study the fault currentsand contact voltage which depend on how theframes are earthed. There are two possibilities,namely:c The load frames are all interconnected by aPE protective conductor: this is the generalcase.c The frames are not interconnected and areconnected to separate earth connections(configuration to be avoided due to EMC:see “Cahier Technique” no. 187).3.1 Analysis of the double insulation faultIn this section, fault currents and contact voltageare calculated by considering two full insulationfaults on two different live conductors (on onephase and the neutral if the neutral is distributed,or on two different phase conductors if theneutral is not distributed) of two circuits ofidentical cross-section and length.This assumption, which results in a minimumfault current, is the one normally chosen tocalculate the maximum lengths protected by theshort-circuit protection devices.Contact voltage and double fault currentwhen the frames are interconnectedWhen a fault current occurs between two faultyframes, a current flows in the phase conductorsand the PE protective conductor ensuringinterconnection of frames (see fig. 3b).This current is only limited by the impedance ofthe fault loop equal to the sum of theimpedances of the live conductors concernedand the circuit of the equipotential links (PE).There are a number of methods for calculatingfault currents for an electrical installation (see“Cahier Technique” no. 158).In this case, the conventional method has beenchosen, as it enables calculation of fault currentand contact voltage values without making toomany assumptions on installation characteristics.It will thus be used from now on in this “CahierTechnique” to give an idea of the value of thecurrents and voltages involved on a double faultin the IT system.It is based on the simplified assumption thatconsiders that, during the duration of the fault,the voltage at the origin of the feeder consideredis equal to 80% of installation nominal voltage.This assumes that the impedance of the feederin question accounts for 80% of total impedanceof the faulty loop, and that upstream impedanceaccounts for 20%.For the following calculations:U’ = phase to neutral voltage, (= U 0 if one of thetwo faults is on the distributed neutral),orU’ = phase-to-phase voltage, (= e U 0 if theneutral is not distributed).R a = ρ L = resistance of the live conductorS a(phase or neutral) of the circuit on which the faultoccurred.R pe = ρ L = resistance of the circuit protectiveS peconductor.S a = cross-section of the live conductor.S pe = cross-section of the protective conductor.L = length of the faulty circuits.m = Sa= ratio of live conductor cross-sectionSpeover protective conductor cross-section(normally i 1).c If we consider that the live and PE conductorsof the two faulty feeders have the same crosssectionand length and if we ignore their reactance:v if one of the faults is on the neutral0,8 UI0Sd =, i.e. Ia2R ( a + R pe )d =0,8U 02ρ ( 1+mL ),v if the double fault concerns two phaseSconductors Iad =0,8eU0 .2 ρ 1 + mL( )Cahier Technique Schneider Electric no. 178 / p.15

c The corresponding contact voltage isU C =R pe I d , i.e.:v if one of the faults is on the neutralmU =0,8U ,c 0 or21 ( + m)v if the double fault concerns two phasemconductors U c=0,8eU021 ( + m ).NB: this method is not applicable for installationssupplied by generator set, as, due to highgenerator impedance compared with theimpedance of the supplied network, voltage atthe origin of the network in question is low whena fault occurs (> 400 0.02 0.08Fig. 16 : maximum breaking time specified for the ITearthing system by installation standards (* for singlephasenetworks).c Protection of the neutral conductor when it isdistributed.Figure 3b shows that when a double fault occurs,the two SCPDs detect the fault current but eachone on a single phase or on the neutral.This situation calls for particular monitoring ofSCPD characteristics: this is because if thecables of the two feeders have similar crosssections,the two SCPDs play an equal role inbreaking, but if the cross-sections are different,there is a risk of only one SCPD, the one withCahier Technique Schneider Electric no. 178 / p.16

the lowest rating, performing breaking. It is thusnecessary to verify that its breaking capacity onone phase, thus under e U 0 , is greater than I d .For this reason, circuit-breaker manufacturersspecify the single-phase breaking capacities oftheir devices according to each nominal voltage,and standard IEC 60947-2 specifies a testsequence for circuit-breakers designed forprotection of IT networks. Devices failing tosatisfy the requirements of these tests must bemarked: ITProtection must also be confirmed for the neutralconductor when its cross-section is less thanthat of the phase conductors. Note that four-polecircuit-breakers (the fourth pole has a half rating)can be used to protect cables with neutral crosssectionhalf of phase cross-section.It should be stressed that four-pole SCPDs arebecoming increasingly necessary, irrespective ofthe installation earthing system used (TN, TT orIT), due to the proliferation of harmonics innetworks, and thus that the neutral can beoverloaded by harmonic currents of rank 3 andmultiples.c Fuse protectionThe fuse blowing zone is located between twoenvelope curves.Using the expression of current I d , defined in theprevious chapter, and the condition I fu < I d , it ispossible to determine the maximum length of theprotected circuit.v If the neutral conductor is distributed:L = 0,8 U 0 S 1max.2ρ ( 1+ m) Ifuv If the neutral conductor is not distributed:L = 0,8 e U 0 S phmax.1+ m2ρ ( ) I fuI fu corresponds to the fuse blowing current withina maximum time stipulated by the standards.It should be checked that this time is compatiblewith protection of persons in event of a doublefault.Note that use of fuses in the IT earthing systemoften clashes with the need for multi-polebreaking, including that of the neutral conductorwhen distributed.c Circuit-breaker protectionProtection of persons is guaranteed when thefault current is greater than the circuit-breaker’sshort time delay protection setting.Just as with fuses, the maximum length of theprotected circuit can be determined according tothe expression of current I d , defined in theprevious chapter and the condition I m < I d .The maximum length of the circuit protected by acircuit-breaker is:v with the distributed neutral conductor:SL =0,8U amax 02ρ ( 1+m) I.mv with the non-distributed neutral conductor:SL max =0,8eUa02 ρ ( 1+m ) I.mNote that whether protection is provided by fuse orcircuit-breaker, the fact of distributing the neutralin IT divides by e the maximum length protected.c Improvement of tripping conditions.When tripping conditions are not satisfied(lengths greater than maximum lengthsprotected), the following measures can be taken:v reduce the value of I m of the circuit-breakers:however current discrimination between circuitbreakersmay be reduced as a result;v increase PE conductor cross-section. Theimpedance of the return circuit of the double faultcurrent is thus reduced and enables an increasein maximum length for protection of persons.However, although contact voltage will bereduced, the electrodynamic stresses on thecables will increase.v Increase live conductor cross-section. This isthe most expensive solution and also results inan increase in three-phase short-circuit currents.v Finally, there is a simple solution that requiresno calculation: use of low sensitivity RCDs onvery long feeders. This solution is also possiblein IT, as the PE conductor is separate from theneutral conductor which is not the case in TN-C.Case of application frames with separateearth connectionsWhen an installation supplies a number ofseparate buildings at a distance from oneanother, their application frames are oftenconnected to separate earth connections. Theimpedance of the path of fault current I d is thenincreased by the resistance of the two earthconnections in question, and the conditionnecessary for protection of persons (respect ofmaximum breaking times) can no longer beguaranteed by the short-circuit protection devices.The simplest study and installation solution is touse RCDs. Their settings follow the same rulesas in TT.To derive maximum benefit from the continuity ofsupply offered by the IT system, the RCDs mustbe prevented from tripping on the first fault bynot setting their threshold IDn at too low a level,particularly for circuits with a high leakagecapacity, while at the same time respecting theinequation: I∆n

4 Special features of the IT earthing system4.1 Overvoltages in the IT systemElectrical networks can be subject toovervoltages of varying origins.Some overvoltages, such as differential modeones (between live conductors) affect all theearthing systems. Readers requiring moreinformation on this subject should consult“Cahier Technique” no. 179.This section is particularly concerned withcommon mode overvoltages, which mainly affectthe IT system as the network is then“unearthed”:c overvoltages due to insulation faults,c overvoltages due to internal disruptivebreakdown in the MV/LV transformer,c overvoltages due to lightning striking theupstream LV network,c overvoltages due to lightning striking thebuilding in which the installation is housed.These overvoltages are particularly taken intoaccount by standard NF C 15-100 whichstipulates installation of a surge limiterdownstream of an MV/LV transformer and whenthere is a risk of lightning (overhead lines).Overvoltages due to insulation faultsc When the first insulation fault occurs, thephase-to-earth voltage of the sound phases ispermanently brought to the phase-to-phasevoltage of the network.LV equipment must thus be designed towithstand a phase-to-earth voltage of U 0 e andnot the phase to neutral voltage U 0 for the timerequired to track and eliminate the fault. Thisparticularly applies to:v “Y” capacitive filters fitted on many electronicdevices;v installation PIM when installed between phaseand earth because the neutral is not accessible.When choosing a PIM, it is thus important toverify the voltage of the network to be monitoreddeclared by its manufacturer.These recommendations are specified inparticular in standard IEC 60950.c On occurrence of the first fault, a transientovervoltage appears with a possible peak of2.7 x r U 0 (U 0 = phase to neutral voltage of theLV network). On a 230 V/400 V network, thisvalue is 880 V, an overvoltage level that is notdangerous for equipment with an insulation of1,800 V (voltage constraint at power frequencyon the LV side as per IEC 60364-4-442).Note that these overvoltages do not causepermanent short-circuiting of the surge limiter.Overvoltages due to intermittent insulationfaultsIntermittent faults (according to internationalelectrotechnical vocabulary, or “restricting” or“arcing” faults in Anglo-saxon literature) behavelike a series of transient faults.Experience and theoretical studies show thatintermittent faults can generate overvoltages andthus result in equipment destruction. Suchovervoltages are particularly observed on MVnetworks operated with an earthed connectionby a tuned limiting reactance (Petersen coil).These overvoltages are caused by incompletedischarging of zero sequence capacity onre-arcing. The zero sequence voltage thereforeincreases each time the arc is ignited. Assumingthat the arc is ignited at the highest value of thephase-to-earth voltage of the faulty phase, andzero sequence voltage increases each time,overvoltages of 5 to 6 times phase to neutralvoltage may be generated.Yet again, in the IT system, protection isprovided by the surge limiter, and presence of animpedance between neutral and earthencourages rapid discharging of the zerosequence capacity.Overvoltages due to internal disruptivebreakdown of the MV/LV transformerVoltage withstand at power frequency of LVequipment is defined in standardIEC 60364-4-442 which specifies their valuesand durations (see fig. 17 ).c Internal disruptive breakdown between theMV/LV windings. This kind of overvoltage is atnetwork frequency.Acceptable AC voltageson LV equipmentU 0 + 250 V (i.e. 650 V in IT) > 5Breakingtime (s)U 0 + 1,200 V (i.e. 1600 V in IT) * i 5(*) For an IT network, the voltage U 0 must be replacedby the voltage e U 0 .Fig. 17 : acceptable AC voltage constraints on LVinstallation equipment in the IT system for a 230/400 Vnetwork.Cahier Technique Schneider Electric no. 178 / p.18

These overvoltages are rare and their “sudden”appearance means that the surge limiter, whosecertain arcing voltage is set at least at 2.5 timestype voltage (NF C 63-150), i.e. for example750 V for a limiter placed on the neutral of a230/400 V network, immediately earths the LVnetwork, preventing it from rising to MV potential.c MV/frame internal disruptive breakdown alsoknown as “return disruptive breakdown”When the transformer frame and the LV networkare connected to the same earth connection(see fig. 18 ) there is a risk of LV equipmentdisruptive breakdown if the voltage R p Ih MTexceeds equipment dielectric withstand, with R p(earth connection resistance) and Ih MT (zerosequence current due to MV disruptivebreakdown).One solution is to connect the LV installationframes to an earth connection that is electricallyseparate from that of the substation frames.However, in practice, this separation is difficultdue to frame meshing in MV/LV substations.Consequently standard IEC 60364-4-442 statesthat the LV installation frames can be connectedto the earth connection of the transformersubstation frames if the voltage R p Ih MT iseliminated within the stipulated times.Overvoltages due to lightning striking theupstream MV networkWhen lightning strikes the MV network, a waveis transmitted to the live conductors on the LVside as a result of capacitive coupling betweenthe transformer windings.If the installation is in IT, the surge limiterabsorbs the overvoltage occurring on the liveconductor to which it is connected (neutral orphase) and is short-circuited if this overvoltage isvery high: the network can then be compared toa network in TN-S. Experience andmeasurements have resulted in the followingobservations:c Overvoltages of around 2 kV occur at the endof short cables (10 m) irrespective of load andearthing system.c Higher overvoltages occur at the end of cableswith open end or which supply loads likely togenerate resonance. Even with a resistive load,overvoltages exist (see fig. 19 ), caused by wavepropagation and reflection phenomena and bycapacitive coupling between conductors.In view of the waveform of these overvoltages,the surge limiter is effective on the conductor towhich it is connected. Consequently, regardlessof the earthing system, we strongly recommendthat surge arresters be installed at the origin ofthe LV network, between all live conductors andHVMVNIh MTMVLVR T (R p )Fig. 18 : when the substation frames (MV) and the PEearth connection (LV) are connected to the same earthconnection, the LV load frames are brought to thepotential Ih MT R p.û (kV) Ph/Ph Ph/PE Ph/N N/PE PE/deepearthSystem :c IT 0.38 4.35 0.20 4.30 1.62c TN-S 0.36 4.82 0.20 4.72 1.62Fig. 19 : overvoltages, caused by a lightning shockwave, measured at the end of a 50m cable supplying aresistive load.the earth, if there is a risk of the upstream MVnetwork being directly struck by lightning (caseof overhead lines) and especially if the LVnetwork is also at risk. The surge limitercontinues to perform its function for MV/LVdisruptive breakdown.Overvoltages due to lightning striking thebuilding housing the installationThese overvoltages are caused by lightningcurrent flowing through the building’s earthconnection, particularly when lightning strikes abuilding equipped with a lightning rod.The entire earth network then markedly rises inpotential with respect to the deep earth. The LVnetwork, immediately earthed by the surgelimiter, changes from the IT to the TN-S system ifall the application frames are interconnected.The lightning energy thus flown off can beconsiderable and require replacement of thelimiter.In order to minimise these overvoltages onelectrical installations, the building’s horizontaland vertical equipotentiality must be the bestpossible in low and high frequency. A singleearth circuit (PE network) is naturallyrecommended, and use of metal cable trays withproper electrical connections (braids) is highlyadvisable for distribution.Cahier Technique Schneider Electric no. 178 / p.19

4.2 Surge limitersNominal voltage Arcing U (V) Example : limiter to beof a limiter -Un- at power on 1.2/50 chosen for a 230/400 V(V) (NF C 63-150) frequency impulse wave network…250 400 < U < 750 < 1,750 … if connected betweenthe earth and the neutral.440 700 < U < 1,100 < 2,500 … if connected betweenthe earth and the phase.660 1100 < U < 1,600 < 3,500Fig. 20 : the nominal voltage of a surge limiter must be adapted to network voltage.The previous section clearly explains why thesurge limiter is an “essential accessory” of theIT earthing system and thus stipulated bystandards. It also protects the PIM againstovervoltages.Its clipping thresholds for overvoltages at powerfrequencies and for common mode impulseovervoltages are defined by standardNF C 63-150 (see fig. 20 ). These thresholds arelower than the specified withstand of equipmentused on LV networks (230/400 V).It must be connected as close as possible to theMV/LV transformer between neutral and earth,or between a phase and earth if thetransformer’s secondary connection is of thedelta or non-distributed neutral kind.NB:c limiters are not necessary on networksdownstream of a LV/LV transformer,c standard IEC 60364 does not specify use ofsurge limiters, as it considers that occurrence ofan MV/LV fault is rare. However, when this faultdoes occur, its consequences are frequentlyserious.OperationA surge limiter consists of two conductivecomponents separated by an insulating film(see fig. 21 ).Impulse overvoltages generate arcing betweenthe two conductive components, but do notshort-circuit the limiter.Energetic overvoltages melt the insulating film,thus allowing the run-off of a high current toearth. The cartridge must then be replaced: itsshort-circuiting is reported by the PIM just as aninsulation fault. Moreover, it is useful, for livefault tracking, to consider its earth connection inthe same way as a feeder, particularly if thisconnection is normally inaccessible (such aswhen, for example, the limiter is placed in thetransformer cubicle).Important characteristicWhen all the application frames are properlyinterconnected, the double fault concerning boththe arced surge limiter and an insulation fault on“Insulating film”disappears duringhigh powerovervoltagesArcing zone duringlow power overvoltagesInsulating caseConnection padFig. 21 : surge limiter principle (Merlin Gerin Cardewtype).a phase, becomes a short-circuit. Limiterwithstand must then be sufficient for the timerequired to eliminate the fault current (Forexample, 40 kA must be withstood for 0.2 sec forMerlin Gerin Cardew limiters).In the rare case of the second insulation faultoccurring upstream of the incoming circuitbreaker,the double fault is eliminated by theMV protection devices (just as for an upstreamshort-circuit on the main LV switchboard).For this reason, the time delay setting of thetransformer’s MV protection must take the thermalwithstand [f (I 2 t)] of the surge limiter into account.The cross-section of the connection conductorupstream and downstream of the surge limitermust also have the same thermal withstand. Itscross-section is calculated in standardNF C 15-100.Cahier Technique Schneider Electric no. 178 / p.20

4.3 Why use an impedance?An impedance can be connected between thenetwork and the earth, normally between thetransformer neutral and the earth. Its value isapproximately 1,700 Ω at 50 Hz.Its purpose is to reduce variations in potentialbetween network and earth, caused by MVdisturbances or fluctuations in potential of thelocal earth. It is therefore particularlyrecommended for short networks supplyingmeasurement instruments sensitive to thispotential and for networks placed next tocommunication networks (Bus).A reading of the table in figure 5 shows thatwhen the network is very slightly capacitive(case 1), the neutral impedance Z N causes thefault current to increase, which neverthelessremains very low (≈ 250 mA in figure 5). Thiseffect is even slighter when the network is highlycapacitive (cases 2 and 3). In practice, thisimpedance effects only very slightly the contactvoltage U C which remains less than U L in soundnetworks.Finally, presence of a resistance in the impedanceenables a reduction of the ferromagneticresonance hazard.Cahier Technique Schneider Electric no. 178 / p.21

5 Advantages and disadvantages ofthe IT earthing system in LVThe main advantage of using the IT earthingsystem for network operation is without doubtthe continuity of supply it offers, as there is noneed to trip on the first fault (as described in thesection below). Another of this system’s strongpoints is guaranteed safety against the firehazard and for control and monitoring circuits ofmachine tools.However, to benefit fully from such advantages,the restrictions of this system must also beconsidered.5.1 Increased availabilityThis advantage can be confirmed by a simpleprobability calculation.Let us assume that the occurrence of aninsulation fault in an electrical installation is onefault every three months (90 days),1i.e. λ =90 jand the time needed to track and put right thefaulty part is one day,i.e. µ =1j.The Markof graph technique gives therepresentation shown in figure 22 and enablesus to calculate that the average time betweentwo double faults is 8,190 days!This corresponds to an average electrical poweravailability that is 91 times better in IT than in TNor TT.Consequently, preference is frequently given tothe IT earthing system for use in:c hospitals,c airport take-off runways,Nofaultλ = 1 90 jµ = 1j1faultλ = 1 90 jµ = 1j2faultsFig. 22 : a Markof graph shows that average electricalpower availability is 91 times better in IT than in TN orTT.c vessels,c plants with continuous manufacturingprocesses,c laboratories,c cold storage units,c electrical power plants.5.2 Increased safety against the fire hazardElectricity is often the cause of fire. Standardsset the threshold for this risk at 500mA on aninsulation fault (NF C 15-100, part 482.2.10). Thisvalue can be considerably exceeded, particularlywith stray currents that flow through buildingstructures when faults occur in the TN system.Also worthy of note is that the IT is the onlyearthing system that monitors insulation of theneutral conductor, compared with the TNS whichcan insidiously turn into a TNC on a neutral-PEfault with an increase in the fire hazard.It is because the current of the first fault isparticularly low that the IT earthing system hasbeen chosen for use in certain establishments atrisk from fire and explosion (see chapter 1).Furthermore the first PIMs were used infiredamp mines.Cahier Technique Schneider Electric no. 178 / p.22

5.3 Less downtime on control and monitoring circuitsThe relay diagram illustrated in figure 23 withthe TN earthing system shows three possibleinsulation faults which, when full, result inimmediate downtime, whose material andeconomic consequences are rarely negligible.These faults have the same consequences withthe TT system.In particular, faults c and d cause tripping of themaster protection device, and prevent allsubsequent operations, such as for example theorder to change direction on a transporter bridge!These same first faults which can causeoperating malfunctions or even accidents withthe TN and TT systems, have no effect with theIT system, except if they occur as the secondfault (extremely unlikely, see section 5.1).These examples show that even if safety ofpersons with respect to the electrical hazard isguaranteed by the various earthing systems, orby use of Safety by Extra Low Voltage (SELV),safety of persons with respect to mechanicalhazards may not be guaranteed in certain cases.More care must therefore be taken when wiringsuch circuits in the TT and TN than in the ITsystem, as the latter warns the operator of theincident (first insulation fault), thereby guardingagainst electrical and mechanical hazards. PIMsare increasingly used for just this purpose, tomonitor automation networks.An additional solution is often advisable,particularly with relays using electronic devicessensitive to electromagnetic disturbances. Theaim is to supply all the control and monitoringcircuits separately by means of a LV/LVtransformer with separate windings.Despite this, as stated in chapter 2, use of the ITearthing system has its limits which aredescribed in the section below.32N1NPEI dAMR a b c dBFault a cannot be detected.Fault b prevents the off function.Faults c and d cause ashort-circuit.Ma b c dFault a cannot be detected.Faults b, c and d cause ashort-circuit.AFig. 23 : monitoring circuit may be concerned by several types of insulation faults always resulting in downtimewith TT and TN system.5.4 Restrictions and precautions for using the IT earthing systemThe restrictions for using the IT system arelinked to loads and networks.Limits linked to loadsc With a high earth capacitive coupling(presence of filters).A number of devices fitted with capacitive filters(see fig. 24 ) offer the same disadvantage, dueto their number, as very long networks when theIT system is used.These capacitive leakages have a particularity,with respect to distributed capacity mainly due tonetwork cables, i.e. they can be unbalanced.DeviceMicro-computerUPSVariable speed controlersFluorescent tubes(in ramps of 10)Network/earth capacity20 nF to 40 nF40 nF70 nF20 nFFig. 24 : guideline capacitive values for HF filters builtinto various devices.Cahier Technique Schneider Electric no. 178 / p.23

D a(A)123ND b(B)RCD30 mAC FC FFig. 25 : in the IT system, capacitive current flow can cause nuisance tripping of the RCDs “by sympathy”.In this case, in presence of a fault on feeder B, circuit-breaker D a placed on a highly capacitive feeder (presence ofseveral filters) may open instead of D b .Office computer equipment: micro-computers,monitors and printers, concentrated on the samesingle-phase feeder, is an example of this. Itshould be borne in mind that interferencesuppression filters (compulsory according to theEuropean Directive for EMC) placed on thesedevices, generate in single-phase permanentleakage currents at 50 Hz that can reach 3.5 mAper device (see IEC 95); these leakage currentsadd up if the devices are connected on the samephase.To prevent nuisance tripping (see fig. 25 ),especially when the RCDs installed have lowthresholds, the permanent leakage current mustnot exceed 0.17 I∆n in IT. In practice, the supplyby a 30 mA RCD of three micro-computerstations is the maximum recommended.This problem also exists with the TT and TNsystems.For memory:v to guarantee safety of persons (U C i U L ), thelimit not to be exceeded is 3C i 70 µF.v for insulation monitoring, PIMs with DC currentinjection are not affected by these capacities.Note that if the devices are connected on allthree phases, these capacitive currents canceleach other out when they are balanced (vectorsum).c With a low insulation resistanceThis particularly applies to induction furnacesand arc welding machines, as well as very oldcables.A low insulation resistance is equivalent to apermanent insulation fault: the IT system is“transformed” into a TN or TT system, with a PIMon permanent alert.PIMARPIMCPIMBFig. 26 : insulation monitoring of the various parts of anetwork with a replacement source.Limits due to the physical characteristics ofnetworksHigh “capacitive leakages” disturb insulationmonitoring using PIMs with AC current injectionand tracking of the first fault using a very LFgenerator (see chapter 2).When an insulation fault occurs, they can alsocause flow of residual currents likely to generatenuisance tripping “by sympathy” of the RCDsplaced on very long or highly capacitive feeders(see “Cahier Technique” no. 114).Use of the IT system is thus advised against forvery long networks, containing long feeders, forexample for electrical power distribution in anumber of buildings at a distance from one another.Case of networks with replacement powersupplyThe fact that a network can be supplied byseveral sources makes it necessary to detect thefirst fault and to trip on the second fault,irrespective of the voltage source in operation.Cahier Technique Schneider Electric no. 178 / p.24

c Permanent monitoring of network insulation,regardless of the supply source, makes itnecessary to choose PIM position carefully. Insome positions insulation monitoring may bepartial (see fig. 26 ).Permanent connection of two PIMs at positionsA and B is not acceptable as they would eachobstruct the other on coupling.On the other hand, position C could beacceptable, as access to supply sources isreserved for authorised users, but there is therisk on source switching of finding that a faultalready exists on the new source.It is thus preferable to provide a PIM on eachsource [A and B] with a relay [R] preventingsimultaneous operation of both devices on thesame circuit.There are also new insulation monitoringsystems which exchange digital data via bus andautomatically adapt to network configuration.These PIMs make use of special, oftencomplicated, relays unnecessary (see fig. 27 ).c Tripping on a double fault, whatever thevoltage source, calls for (just like the TN system)verification of SCPD compatibility with thepresumed fault currents, particularly when thereplacement source is a generator set. This isbecause the short-circuit current it delivers is farlower than that supplied by an MV/LVtransformer supplied by the public distributionnetworks: the SCPD threshold must be setaccordingly.The first solution is to lower the threshold ofthese protection devices, but the problem is thatthis also reduces current discrimination ability.A second, simpler, solution, is to provide lowsensitivity RCDs.c With an Uninterruptible Power Supply (UPS)The problems experienced are the same as witha generator set. Only insulation monitoring ismore complex as it is linked to the variousoperating configurations that a UPS canassume.In practiceIn all these restriction cases, the mostappropriate solution is to reserve use of IT forthe network only supplying the devices requiringa high degree of electrical power availability.c For existing installations, in order to restorecontinuity of supply, it is necessary to identifylow insulation feeders and supply themseparately with, for example, a TN system, andsave the network in IT for the more demandingapplications. This solution requires use of a newtransformer, either LV/LV, or directly MV/LV,according to power requirements. A similarapproach can be applied in order to supplymachines experiencing operating problems in IT.c For new installations, electrical powerdistribution in IT must be provided as soon asthe need for continuity of supply is apparent. It ispreferable, in order to reduce the incidence ofnetwork capacity with respect to earth, to limitthe size of this network to a building, forexample.Finally, if phase to neutral voltage is required,distribution of the neutral conductor should notbe a cause for concern as:v neutral insulation is monitored,v use of circuit-breakers with B or G curve and oflow sensitivity RCDs simplifies the protectionstudy,…and avoids the installation (and thus theadditional cost) of a specific transformer or aspecial line.CommunicationinterfaceXTUInformation exchange busXM300C XM300C XM300CPIMPIMPIMFig. 27 : insulation monitoring system of the various parts of a network with several sources (Merlin Gerin’sVigilohm System).Cahier Technique Schneider Electric no. 178 / p.25

6 ConclusionEvolution of the various earthing systems shouldmirror the changing needs of electrical powerusers.6.1 Availability: an increasing need to be satisfiedThe increasing number of computer, automationand control/monitoring equipment has resulted inall major economic entities (industrial,commercial, etc.) calling for greater availability ofelectrical power.Today, electrical power is considered to be asimple product with which a number of qualitycriteria, particularly availability, are associated.To ensure that users benefit from this increasedavailability, this demand, already acknowledgedby electricity utilities, must also be incorporatedin the design of the new internal and privatedistribution installations… precisely there wherethe IT earthing system assumes its fullimportance by indicating the very first fault (notdangerous) and preventing tripping.However, for the network to benefit from all theadvantages of the IT system, designers mustcarefully consider the future operation of thenetwork and have excellent knowledge of thedevices to be supplied.6.2 The IT earthing system finds its true placeUsable in a very large number of electricalinstallationsThe IT earthing system can be used in a verylarge number of electrical installations inindustrialised countries, with the exception ofapplications (e.g. arc furnace, old lighting circuit)and situations (e.g. damp environment, very longnetwork) normally or frequently exhibiting a lowinsulation level. These countries possess skilledelectricians, sufficiently reactive to offer rapidinstallation servicing (the same day). Moreover,their infrastructures allow use of remotesupervision.For adapted distribution circuitsChanges in continuity of service requirementsand implementation of new machines withspecific characteristics, particularly in the field ofelectromagnetic compatibility (EMC) mean thatthe electrical power supply sometimes requiresspecially adapted distribution circuits. Thisaccounts for the emergence of privatedistribution networks comprising a variety of subnetworkswith an appropriate earthing system.In these conditions, the IT system easilyguarantees the necessary continuity of supply.6.3 The added advantage of safetyInstallation designers must also identify fire andexplosion hazards and satisfy EMC requirements(disturbance of measurements andcommunications).The IT earthing system offers the mostadvantages and best meets operators’requirements with such specific features as:c better EMC (interconnection of frames and intheory a single earth connection),c minimum fire and explosion hazards (low firstfault currents).Moreover, its use is encouraged by theupgrading of equipment (PIM, tracking device,supervisor, etc.) allowing:c anticipation of maintenance (prediction),c quicker tracking of the first insulation faults(automation), or even remote tracking (remotesupervision via digital connections),c preparation of troubleshooting (remotediagnosis).Cahier Technique Schneider Electric no. 178 / p.26

6.4 In shortOur readers now understand the importance ofproperly listing the requirements relating toequipment used, the environment and the studyconditions of the installation and subsequentmodifications, before choosing the earthingsystem for an electrical distribution network.A brief reminder of the advantages anddisadvantages inherent in each earthing systemis essential at this point: this is the purpose offigure 28 .NB: The installation cost is not included in thistable as the possible additional cost of an ITsystem (PIM, fault tracking system) must becompared with the financial loss generated byunexpected downtime on the first fault… thismust be evaluated for each activity.TT TN-C TN-S ITSafety of persons (perfect installation) c c c c c c c c c c c cSafety of equipmentc against the fire hazard c c c v v v c c cc for machine protection on an insulation fault c c c v v c c cAvailability of electrical power v v v v v v c c c cElectromagnetic compatibility v v v v v v vFor installation and maintenancec skill c c c c c c c c c c c c cc availability v v v v v c c cc c c c excellentc c c goodv v averagev poorFig. 28 : summary of the advantages and disadvantages of the various earthing systems.Cahier Technique Schneider Electric no. 178 / p.27

BibliographyStandards and decreesc IEC 60364: Electrical installations of buildings.c IEC 60479-1: Effects of current on humanbeings and livestock.c IEC 60947-2: Low voltage switchgear andcontrolgear - Part 2: Circuit-breakers.c IEC 60950: Safety of information technologyequipment.c IEC 61000: Electromagnetic compatibility.c IEC 61557, NF EN 61557: Electrical safety inlow voltage distribution systems up to 1,000 V ACand 1,500 V DC - Equipment for testing, measuringor monitoring of protective measures -.Part 8: Insulation monitoring devices forIT systems.c NF C 15-100: Installations électriques à bassetension.Schneider Electric’s Cahiers Techniquesc Neutral earthing in an industrial HV network.F. SAUTRIAU, Cahier Technique no. 62c Residual current devices.R. CALVAS, Cahier Technique no. 114c EMC: Electromagnetic compatibility.F. VAILLANT, Cahier Technique no. 149c Harmonics in industrial networks.N. QUILLON, P. ROCCIA,Cahier Technique no. 152c Calculation of short-circuit currents.B. De METZ-NOBLAT, G. THOMASSET,R. CALVAS and A. DUCLUZAUX,Cahier Technique no. 158c Earthing systems in LV.R. CALVAS, B. LACROIX,Cahier Technique no. 172c Earthing systems worldwide and evolutions.R. CALVAS, B. LACROIX,Cahier Technique no. 173c Perturbations des systèmes électroniques etschémas des liaisons à la terre.R. CALVAS, Cahier Technique no. 177c LV surges and surge arresters - LV insulationco-ordination -.Ch. SERAUDIE, Cahier Technique no. 179c Intelligent LV switchboards.A. JAMMES , Cahier Technique no. 186c Cohabitation of high and low currents.R. CALVAS, J. DELABALLE,Cahier Technique no. 187Cahier Technique Schneider Electric no. 178 / p.28

Schneider ElectricDirection Scientifique et Technique,Service Communication TechniqueF-38050 Grenoble cedex 9Fax: (33) 04 76 57 98 60DTP: AXESS - Saint-Péray (07)Edition: Schneider ElectricPrinting: Imprimerie du Pont de Claix - Claix - France - 1,000- 100 FF -© 1999 Schneider Electric2828906-99