This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Application of BS7671 Table 54.7

Can anyone offer experience on the application of Table 54.7 for Main Earth size on non PME supplies for supplies that have parallel incoming cables.

For example, 2 x 185mm2 PCV/SWA installed in separate ducts.


  • The use of Table 54.7 for parallel cables is not applicable and the adiabatic must be used when faced with parallel cables.

    Humm. You certainly can't apply 54.7 where the line conductor has itself been sized using the adiabatic (i.e. if the line conductor is only just going to survive the fault current, it doesn't then make sense to use half the size for the c.p.c.). But if the overall c.s.a. of each of the line conductors doesn't require consideration of fault currents (e.g. all parallel paths together give Iz≥In) I would have thought you could apply table 54.7 to that. E.g. if each of your line conductors is 2x185mm² then plug in 370mm² which then gives you 185mm² (or 185mm² * k1/k2 if of differing metals) and and you should be safe. If you have parallel protective conductors, it's difficult to control how the fault current will divide (appendix 10 gives some background) - but use the selected/calculated value for each of the protective conductors and you should be safe regardless (but that's probably the same issue whether you select or calculate). As ever, the adiabatic often gives more economicaly attractive results.

        - Andy.

  • Well it depends if the twin 185mms are needed to meet voltage drop or thermal considerations - i.e. how long they are.

    I see the problem, because 'indicative design' really means you do not really have the access to all the numbers that would allow you to downsize the conductors and be sure that  it is safe to do so.

    It should be possible to put some safe upper bounds on it though, especially if you define the 'death or glory' fuse to be fitted.

    The numbers below are not right - and should be taken with a large spoon of salt, but they illustrate one line of thought.

    The worst case is probably when the 185s are short, and needed because the steady sate load current is about 450Amps per core, so 900A per phase total. That suggests a substation of 700kVA to 1MVA . So a PSSC of 25-35 kA is credible, and perhaps a phase fuse size of 1000-1200A.

    Let-through of a fuse of that rating is then maybe 10^7 Amps squared seconds, and use that to size the CPC...


  • I really don't see much point in this discussion. 2X185 phase conductors and a 185 Earth should be satisfactory on all counts, you need to realise that a short from phase to Earth will heat ALL the conductors involved, but the Earth is quite safe getting considerably hotter under worst-case conditions. Phase Earth faults at this point in the supply should be extremely unlikely and they will never be worse than a phase-phase fault at some point in one of the cables. Such faults may well cause cable damage, but the worst outcome is probably raising everything Earthed to half mains potential while the fault clears, a perfectly normal condition on the whole distribution system (DNO). For a short distance to cost saving of not using 2x185 Earth cables is not worthwhile. For a long distance, a single one will be good to go as the PSSC will be seriously reduced, and an Earth fault to that conductor cannot occur. A fault to the SWA armour is a different kettle of fish, it will almost certainly wreck the cable and probably (maybe) will not clear at the fuse.

    Your typing is going Mike you mean PSSC of 25-35 kA, probably about right.

  • sorry correcting....  part of the problem with this sort of query is that without more site specific info to provide an 'indicative' design has to be either quite conservative, or will for some situation still be inadequate. Equally the technically correct answer 'use the site specific figures and design it properly' is probably not helpful either.

    The chonky substation style fuse is helpful though as it puts an upper bound on the damage any fault downstream of it could  do, almost regardless of supply arrangement.

    Good point on self clearing faults to the SWA armour, it may well burn back from a nail or anything of that size without troubling any ADS . To me SWA seems an odd choice.


  • I really don't see much point in this discussion.

    I am not too sure I follow your sentiment.

    I am seeking advice in the application of Table 54.7 for Main Earth size on non PME supplies for supplies that have parallel incoming cables.

    I am seeking advice on whether this is a credible/application in the use of Table 54.7. It really is as simple as that. The cable size, type and number are all hypnotical and have no real bearing on the question.

    I am not seeking advice on whether this is the most economical method as we all know the adiabatic calculation will provide the most economic results.

  • if each of your line conductors is 2x185mm² then plug in 370mm² which then gives you 185mm²

    Thanks Andy, this is the method that I would adopt if I needed to use this method. However, there is a caveat that I would employ in the application.

    1.  I would size the incoming cable on the first 10m. Reason for this as after this the incoming cable size can balloon due to voltdrop... 10m as this is the typical length that most DNO would allow the cable to be run without protection.

    Absolutely agree, adiabatic equation is the best solution, I am really just quizzing the application of table 54.7 and whether it could be used. Should be is a different question.

  • I may be completely naïve here, but if you want to be lazy, 543.1.1 allows you to use table 54.7. In your example, given that the minimum cross-sectional area of the protective conductor = S/2, one of the cables needs to be accompanied by a protective conductor of the same size as the line conductor(s). Alternatively, each one could be accompanied by a protective conductor of S/2. Note that the separate protective conductor will not protect the cables between the supply and the load or downstream DB.

    That said, 543.1.1 goes on to say that if the size of the line conductor has been chosen by considering the PSSC, 543.1.3 must be applied rather than table 54.7.

    That's how to make an installation compliant, but perhaps it is better to start with the proper engineering principles, which are a bit above my pay grade.

  • Chris, it not really about being lazy ... its about understanding the applicable application of Table 54.7 and whether it can be used (and application of) for Main Earth size on non PME supplies for supplies that have parallel incoming cables.

    I do appreciate you comment with respect to the fact that calculation must be performed where line conductor has been chosen by considering the PSSC.
    Hence on these occasions Table 54.7 must not be used.

  • I may be missing the point of the question. When you say non-PME, is that TN-S or TT?

    543.1.3 may always be applied. As I understand it, 543.1.4 (and table 54.7) may be applied except as already mentioned. If you have a choice, it's up to you.

Reply Children
  • Many answers on this thread take the problem sideways and make the question more complicated than it is. I am really trying to keep it a simple question ....

    The Question is - "Can Table 54.7 for Main Earth size on non PME supplies for supplies that have parallel incoming cables?"

    The other details provided were hypothetical at best, to give readers a basic understanding of what I was thinking.

    If you consider that TN-S or TT gives different answers to the question perhaps you can comment on both?

  • Well clearly with TT, the core can be much smaller, as the fault current path includes the current limiting resistance of the ground, several ohms is likely, so a cable good for 100A might  do - 10mm perhaps ;-)    also you'd need an earth fault relay, as there is no chance of blowing a 1000A fuse, an LE fault in such a case without the Earth fault detection just leads to high bills, dry hot earth around the electrodes, and nasty step voltages as nothing disconnects.

    With TN-s similarly, the impedance of the earth path will limit the current, but depending on the physical construction how much metal is involved, it may be more or less than the PME case.

    Of course you have not said in  your PME supply where you expect the N-E bond to be - a lot of large sites with a private transformer actually do not do this at the transformer, but at first breaker or fuse panel - this also simplifies things when multiple transformers are being paralleled.


  • Mike, whilst I understand the difference in fault currents and the how that may affect the incoming supply cable, but with respect to the question where I am specifically trying to focus on the understanding of Table 54.7...

    Can Table 54.7 for Main Earth size on non PME supplies for supplies that have parallel incoming cables be applied for TT systems?

    If the system is a TT is the method of application different from a TN-S system?

  • 'no' or at least "you can, of course you can, but you should not, as it is not sensible. "