Rules and Guidelines
Want to participate in the discussions? Please read our Community Rules and Guidelines. Need some help? Visit Support and Answers.

  • State Not Answered
  • Locked Locked
  • Replies 119 replies
  • Subscribers 32 subscribers
  • Views 17529 views
  • Users 0 members are here
Options
You may also be interested in
This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

Type A rcd . EICR coding ? etc

aligarjon
Hi Guys.   Not been on for a long time, just had a bit of a search and couldn't really find anything so thought i would ask and see what you all thought.


1.  Are we or will we be coding type AC rcd's if there are LED's or induction hobs, lots of electronics  etc  present.

2. How much DC leakage does it actually take to saturate an rcd and cause  problem?

3. How much does a standard LED lamp or induction hob  leak ?

If we test an AC RCD with no load and it's fine then re-test it with all LED lights, induction hobs etc turned on and it operates correctly could we then say that it is ok with a note on EICR  OR EIC if installing any of the above.  


Obviously also on an EICR if the RCD then doesn't operate with it all on it becomes a C2 ?


Any thoughts



Gary
  • 0
    After this thread, I wouldn't install a CU with type AC RCDs: that's the learning point. So you might ask whether Schneider appreciate the difficulty and why they continue to make these boards for the UK market. ?


    The other issue is the plumbers! In fact, they are v. old school and have done excellent work; but when they specify this particular Worcester boiler, which may be the most reliable kit available, do they know about the electrical implications? No!
  • 0
    davezawadi (David Stone):

    Very well, Gentlemen, the boiler manufacturer cannot be bothered to correctly fuse his appliance, expects there to be faults in his electronics, safety coverings, etc. and so we must fit a type A RCD.


    Selectivity springs to mind here.


    The manufacturer's instructions apparently (according to earlier post) say that where an RCD is required for additional protection ... not protection.



    I guess the poor manufacturer can't win? Some people ask "How on earth can I tell which RCD to use?" - yet others berate them for helping out and clearly stating that?

     



    I think it would be reasonable for someone to describe what fault exactly could require a type A RCD.

     

    BS 7671 does exactly that - see Figure A53.1. Some of the faults would have to occur either downstream of electronics shown, or as a result of an N-E fault (simply because single-phase systems are "single-ended" transmission systems with respect to Earth) - but they still happen.

     


  • 0
    Hi Graham


    Figure A53.1 shows waveforms which may cause some RCDs difficulties. We discussed those at length above. However, that was not the question I asked which was (written more clearly perhaps) "How can any credible fault produce those waveforms, and if it can, will it operate various types of RCD?" which is very different. I can always say "We need to provide additional protection of the highest level for safety" but that is not the correct answer at all, although it smacks of the usual British Gold plating of any possible regulation or even idea. I am interested in Credible faults that could occur, and once one is established, how often could it happen, in other words, the normal risk assessment criterion. It is unacceptable to expect to make everything completely zero risk, although some like to try and do so, particularly when it costs them nothing or worse when it makes them a profit!


    In this boiler case, it is very difficult to imagine a fault which could require additional protection of the boiler or how a credible fault would need a type A RCD. Two unanswered questions, both of which need answers before I can decide on the additional protection measures needed. The instructions give no help at all. If I fit additional protection to protect the wiring, a type AC RCD is exactly what is needed, so why fit a type A, which seems to be mandated by these instructions? If we lose type ACs at some point (which seems likely), what should an EICR say, code C2 perhaps, just in case there could be some fault which causes a type AC to fail?


    I am sure that many of you have given instructions to others at work. It is usually found that some explanation is a good idea, as otherwise the intention of the instruction is lost, and the result may well be a disaster. The Army certainly has!


    Kind regards

    David
  • 0
    We can't go assuming the manufacturer has got it wrong just because it means the job is a little more difficult.

    Agreed. Also we can't really assume that something that was deemed acceptable last year should be assumed to be equally acceptable this year, especially if new evidence has come to light in the meantime. Grenfell should have taught us that by now.


    I've been pondering the mention of additional/supplementary protection - presumably against direct contact - so we're expecting the RCD to see a residual current that's just flowing though a person who I presume is a fair approximation to a simple resistor so we shouldn't expect any waveform distortion or d.c. components on that score - so in that respect AC type RCDs should be adequate. On the other hand we have a load that might draw distorted waveform current for its normal load - so the RCD will see distorted a.c. load current plus pure sinewave shock current flowing out in its L coil, and just the distorted N return (without the shock current) in its N coil - so the question is: can we be sure that all AC type RCDs can spot a pure sinewave residual current when both the L and N load currents equally are distorted? I.e. does the insensitivity to distorted sinewaves occur before or after summing in the toroid?


       - Andy.
  • 0
    davezawadi (David Stone):

    Hi Graham


    Figure A53.1 shows waveforms which may cause some RCDs difficulties. We discussed those at length above. However, that was not the question I asked which was (written more clearly perhaps) "How can any credible fault produce those waveforms, and if it can, will it operate various types of RCD?" which is very different. I can always say "We need to provide additional protection of the highest level for safety" but that is not the correct answer at all, although it smacks of the usual British Gold plating of any possible regulation or even idea. I am interested in Credible faults that could occur, and once one is established, how often could it happen, in other words, the normal risk assessment criterion. It is unacceptable to expect to make everything completely zero risk, although some like to try and do so, particularly when it costs them nothing or worse when it makes them a profit!



    Kind regards

    David


    No, David, Figure A53.1 shows faults that can and do occur downstream of the particular rectification or control arrangement shown - have a look at the diagram in Column 1, which shows the point of fault (effectively downstream of that point), and the fault waveform. The problem with half-wave rectified mains is it contains complex components, as I'm sure you're aware.


    So, where could these faults occur? Some examples include:



    • Cable supplying equipment downstream from the particular source being discussed

    • Wiring in the equipment itself


    And as to "how often can it happen" ... well, we might as well all go home, because electrical faults in the grand scheme of things are far less frequent than road accidents, far less fatal, and cost far less, than RTAs
  • 0



    I've been pondering the mention of additional/supplementary protection - presumably against direct contact - so we're expecting the RCD to see a residual current that's just flowing though a person who I presume is a fair approximation to a simple resistor so we shouldn't expect any waveform distortion or d.c. components on that score - so in that respect AC type RCDs should be adequate. On the other hand we have a load that might draw distorted waveform current for its normal load - so the RCD will see distorted a.c. load current plus pure sinewave shock current flowing out in its L coil, and just the distorted N return (without the shock current) in its N coil - so the question is: can we be sure that all AC type RCDs can spot a pure sinewave residual current when both the L and N load currents equally are distorted? I.e. does the insensitivity to distorted sinewaves occur before or after summing in the toroid?


       - Andy.


    Yes, whilst 10 mA and 30 mA RCDs are used for additional protection (perhaps a fault through a person), they may already be subject to complex components in residual currents downstream, and effectively "blinded" by that.


    The situation with fault protection using RCDs is a little more complicated - because the fault may well be as described in A53.1.


    It's also worth considering the wording in, say, some of the Part 7 sections - is the RCD providing additional protection or fault protection? Some Part 7's say "shall be protected by an RCD" - without mention of either fault protection or (in the cases where 30 mA RCDs) additional protection, or even protection against fire! Some clue may be available from the numbering of the requirements (i.e. considering which of the General Requirements in Parts 3, 4, 5, 6, or (in the case of AMD 2 DPC) 8.


  • 0
    AJJewsbury:

    I've been pondering the mention of additional/supplementary protection - presumably against direct contact - so we're expecting the RCD to see a residual current that's just flowing though a person who I presume is a fair approximation to a simple resistor so we shouldn't expect any waveform distortion or d.c. components on that score - so in that respect AC type RCDs should be adequate. On the other hand we have a load that might draw distorted waveform current for its normal load - so the RCD will see distorted a.c. load current plus pure sinewave shock current flowing out in its L coil, and just the distorted N return (without the shock current) in its N coil - so the question is: can we be sure that all AC type RCDs can spot a pure sinewave residual current when both the L and N load currents equally are distorted? I.e. does the insensitivity to distorted sinewaves occur before or after summing in the toroid?


    That is a pretty fair summary. In other words, might a fault in the boiler "blind" a type AC RCD?


    The manufacturer's instructions don't require that this particular boiler be RCD protected so in a TN installation, ADS should provide shock protection for a fault inside it. Whether or not a distorted load current would trip a type AC RCD becomes irrelevant. So that leaves a fault in the supply cable (which need not require additional protection) or in the many cables which control the heating system, which are buried in walls < 50 mm from the surface, and which I assume are at mains voltage.


    So to my mind, it is all about blinding. I await a response from the manufacturer.


  • 0
    Chris Pearson:

    ADS should provide shock protection for a fault inside it.



    So to my mind, it is all about blinding. I await a response from the manufacturer.




    Well, yes ... selectivity? ADS may well not operate for faults of a few mA to Earth ... which in themselves are not always particularly dangerous of course, especially if they are "mapped" onto the Neutral because of an N-E fault from an ELV source with a voltage low enough not to drive that current through the human body.. but we know this might be sufficient to affect the operation of an upstream Type AC RCD ...


  • 0
    This is the reply from Worcester. I wonder whether it has been mangled by Google translate.


    "We do not specify a particular type of RCD in our manuals as you can use different types other than an A, all we advise is the correct type of RCD must be employed where additional protection is required that is suitable for a low energy DC modulating pump according to IET wiring regulations.


    We can confirm that it is a low energy DC modulating pump and a suitable RCD will need to be used in accordance with the electrical regulations."


    This is followed by an excerpt from a manual which is not quite the same as the manual which is available on their site.


  • 0
    I have spent a couple of hours yesterday experimenting with RCDs, rectifiers, and various degrees of leakage from the rectifier DC to Earth. The interesting effect is that the leakage seems to cause an increase in trip time, rather than a failure to trip when using my MFT to test. A different RCD tester shows much less effect although they both give similar results with a normal unloaded circuit and RCD. I am using a 1kW lamp load on the rectifier output, so the RCD transformer is actually having to do some work, rather than the unloaded condition. This is an attempt to demonstrate a real circuit rather than the unusual test condition of zero load.


    DC current produced by a 10k resistor to Earth has little or no effect. A 5k resistance increases trip time to a second or so, but does no prevent a trip. This is one make of RCD, BSEN 61008-1, I will try some others and other currents when I have got some more parts, more power resistors of various values. I have yet to add a smoothing capacitor to the DC, and I will have to move the lamp to be AC powered otherwise I will need a very big smoothing capacitor to get a low ripple supply.


    The fault I am simulating is undoubtedly unlikely in any real product, as 5k from a creepage distance of 3mm is hard to imagine, but that seems to be the point behind these diagrams in A53.1. Further testing to follow.

Join us to get the best from IET EngX.

Joining EngX lets you personalise your experience so you stay up to date on the topics that interest you, plus you’ll be able to make connections who are looking to collaborate, exchange ideas and more.

Join us Not now

Community Rules & Guidelines