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

High protective conductor currents - Deletion of 543.7.1.204

543.7.1.204 - the one about duplicate c.p.c.s needing to be 'terminated independently of each other' - has gone.

Does anyone understand the thinking behind this? It seems a bit odd to me.

Given that (in my experience) more problems occur at terminals rather than along cable runs, if we need the c.p.c.s to be duplicated, it seems like a bit of a flaw that one single loose connection can make them both useless simultaneously. I can see that in some instances (e.g. a terminal on a socket on a ring) that losing a single terminal carries a limited immediate risk (as the leakage current from a single socket should be small and all other sockets are still connected to a c.p.c.) but in other instances - say the connection of both c.p.c.s to the earth bar in a DB - that single fault could be very significant.

   - Andy.

Parents
  • So my lifes work is now in the trash can.

    Well, not quite yet Geoff. We still have accessories with not one but TWO terminals for CPCs, which i believe we wouldn't have if it wasn't for your suggestion. Thank you.

  • Agreed, Three Cheers for Geoff

  • Well, not quite yet Geoff. We still have accessories with not one but TWO terminals for CPCs, which i believe we wouldn't have if it wasn't for your suggestion. Thank you.

    If we are talking about socket-outlets, Regulation 411.3.3's development through 17th and 18th Editions has effectively removed the necessity for this. Socket-outlets up to 32 A are now required to have 30 mA RCD protection. 63 A socket-outlets are the next up in the range, meaning we would almost certainly have a cpc that meets the minimum 4 sq mm requirement of Regulation 543.7.1.203.

    I'm struggling to understand how you could meet the requirements of 411.3.3 and have a circuit with high protective conductor currents for socket-outlets of 32 A or less (i.e. requiring "dual cpc"), unless it's somewhere like a data centre where you have additional earthing in any case via another path to comply with BS EN 50310.

  • I am probably getting too old for this, so I don't want to get too involved in the discussion.

    However, my comments are -
    The idea behind 'duplicate earthing' as opposed to using more robust earthing - is to ensure that two simultaneous failures would have to happen before a shock risk could occur.

    Duplicate earthing consists of two earth paths and duplicate connections. A single failure could occur in an earth path or in a single connection - the result being that the other path or other connection would maintain protection.

    1) This is similar to class 1 protection you need:
    i) a fault to earth and;
    ii) a failure of the automatic disconnection or other measures to mitgate the shock risk.
    If i) and ii) occur a shock risk could exist.

    2) Class II - the equivalent of two layers of insulation must fail before the risk occurs.

    3) Duplicate earthing - two conducting paths for the cpc with duplicate connections at each termination - so two failures in the path of the circuit protective conductor must occur.

    4) The suggestion appears to be that we can remove part of the duplicate earthing measure (i.e. the duplicate connections) and rely on RCD protection to save the day. In this case in the event of one fault occurring in a single point of connection of the circuit protective conductor, an appliance or similar could rise to a dangerous potential with respect to earth. If nothing else happens this will be a stable situation so the the RCD will not respond.

    This potential has caused a shock risk to exist after just one failure - if a person touches the appliance current will divert though them causing an imbalance between line and neutral currents. This (we hope!) will trigger the RCD.

    So we are relying on the RCD to prevent injury after just one failure has occurred.

    If we take 1) above you could argue that the automatic disconnection could be achieved using an RCD so what is the difference between 1) and 4). Simply this - for 1) above the shock risk is only present after two failures and for 4), if it does not have duplicate connections, a failure in one connection could remove both paths of the duplicate cpc. So the shock risk could be present after just one failure.

    I am not entirely happy with that.

    Regards

    Geoff Blackwell

  • What are we looking to address with 2 terminals?

    If it's a case of the conductor itself breaking, 2 no. conductors in the same terminal would require 2 failures?

    If a terminal is subject to failure, surely that's the case regardless of the size of cpc connecting to it? So what's special regarding 4 sq mm over 2.5 sq mm or 1.5 sq mm connecting to what is potentially the same terminal? But BS 7671 never covered off that requirement - it only required 2 independent terminals if 2 conductors were used?

Reply
  • What are we looking to address with 2 terminals?

    If it's a case of the conductor itself breaking, 2 no. conductors in the same terminal would require 2 failures?

    If a terminal is subject to failure, surely that's the case regardless of the size of cpc connecting to it? So what's special regarding 4 sq mm over 2.5 sq mm or 1.5 sq mm connecting to what is potentially the same terminal? But BS 7671 never covered off that requirement - it only required 2 independent terminals if 2 conductors were used?

Children
  • Simply if the single terminal fails both conductors forming the duplicate cpc may become completely detached thus removing the path for functional earth currents.  With two terminals the failure of one involving one of the duplicate cpc conductors becoming disconnected leaves the other conductor intact in it's separate terminal - thus maintaining a path for functional earth currents.

    Regards

    Geoff Blackwell