The IT earthing system (unearthed neutral) in LV

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