The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement

This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

Earth resistance of Neutral Earthing

Good Evening,

Could anyone explain why a low earth resistance  of 20ohm is to be achieved for earthing the start point on the secondary side  of  a private distribution transformer with a TNS earthing system ? Rarely will any fault current pass through the ground to trip the protective device on the secondary. As I understand the purpose of earth is to provide a reference point for the neutral, in that case is 20 ohms required??
  • Where the earth is distributed from the transformer the earth is connected to the neutral at the star point of the transformer and to the earth electrode referencing the transformer to the general mass of earth.  This would be a TN-S earthing system.


    Where the the earth is not distributed by the supplier they may allow their neutral to be used as an earth return path. This would be a TN-C-S earthing system with the earth an neutral separated in the consumers installation.


    In both of the above installations earthing methods the transformer earth electrode is not in the earth fault path. The resistance between earth and neutral will be sub 1 ohm.


    If you want a comprehensive understanding of earthing and bonding I would highly recommend IET Guidance Note 8.

  • I think the OP's question was Why 20Ω for RB? As I think is stipulated in BS 7430 (the code of practice for Earthing). Why not 2Ω or even 1000Ω?


    I've asked that question before (especially in the context of small domestic embedded generation that might run during power cuts - where achieving 20Ω might be quite challenging). As far as I can tell there's no maths behind 20Ω at all - it's just a long standing convention that probably started out in the supply industry for public systems and has been written into various standards for general use - and as a result all sorts of assumptions have been based on it (e.g. when designing a TT system derived from a TN system it's usually taken that the max Ze (excluding RA of course) is 21Ω - 20Ω for RB and max 1Ω for the line conductor.


    Clearly it's good to have RB as low as practical - especially if there's any chance of earth leakage to true earth - as even quite modest currents returning from the electrode can raise the potential of the entire system quite significantly.


       - Andy.
  • Like many things the electrode resistance is a compromise - the electrode is needed to have a low enough resistance to satisfy two purposes.

    1) To complete the circuit for ADS in the form of RCDs and earth fault relays to operate correctly with faults to terra-firma, these need to trip on faults like live wires being caught on items that have no wired connection to the circuit earth (CPC) such as fences and so forth that are made of metal, and buried in the ground,  but not actually part of the circuit.  This is unlikely to blow any fuses however,except on very low current branches. (*)

    2) To not cause a large CPC to terra-firma earth offset during a fault on part of the line without an RCD.

    During fault the full phase voltage is divided between what is in effect two electrodes, of unequal impedance, so neither of the electrode to terra-firma voltages are zero. We might like the one at the faulty end to drop most of the volts, and the substation or genset earth and neutral point to be nearer to the voltage of true earth.

     If the fault to terra-firma earth has a lower "equivalent electrode" resistance than the official electrode at the substation  then everyone wired as TN-x on  the network sees live metalwork on their CPC, rather than just the metalworks at the fault, and this is a far more dangerous situation. (it is also an argument in favour of  TT in locations where it is hard to get a good low electrode resistance such as pole mounted transformers in the wilds feeding a few properties.)



    Ideally perhaps the electrode resistance should scale with the power in question - a higher current supply needing a lower resistance  earth electrode, but practical considerations of how much metal can be buried, the type of terrain  and how you test it set a limit that is rather higher than we might like.

    What we deicide instead is that the electriode resistance should be quite a bit lower than a typical fault to terra-firma, and accept that there will be corner cases, such as a cable fault to a building with steel pile foundations, that do not work as we would like.



    Mike.


    (*)20 ohms of earth fault loop will blow a 5A fuse (as seen in UK lighting circuits) in a respectable time and is just about the only thing that does not need an RCD.