"17 edition"consumer units still being sold.

What is your objection to type AC RCDs? Even the objection to using them for car charging is rather more tenuous than just "it is an electronic load". That is not the basis of any reasonable objection, unless you can explain exactly why. A webinar is not an adequate reason unless it is technically correct, which I rather doubt! Give us the URL and we will watch and comment.

By 17th Ed ones do you mean plastic cased ones? These can still legitimately be used in non-domestic situations (e.g. small offices etc) and possibly detached domestic outbuildings.


Likewise AC type RCDs might still be OK for some situations - you pays your money and you takes your choice.


As for why, I guess some of the supply chain moves more slowly than others - and those with the stock aren't going to want to bin it at a loss if it can still be sold for a profit.


    - Andy.

​​​​​​Here's a video showing problems with type ACs and DC leakage: https://youtu.be/rW4FmssJvHs


Also, I asked a CU manufacturer a year ago why they still supply type ACs in CUs, given Hager had stopped manufacturing them, and was told it's because that's what suppliers ask for! Which is presumably, in part, because installers still want them because they're cheaper than boards with type As. Perhaps an injury and prosecution or two may change installers' perspective.


Around that time, I also asked an niceic installer whether he now installs type As in domestics and he replied, 'No. The regs are just getting too complicated'.


F

This is a severely misleading video. Whilst what he says is correct, DC leakage is not a normal phenomenon in installations. We have discussed it in car charging at length because there it is possible due to the car communications system with the charger and a single fault, but it is probably not possible with other electronic loads which never leak DC because the design runs with AC! Asymetric AC and other things do not result in DC (and it needs to be pure and smooth) leakage. If someone can design a load which results in actual DC leakage I would be most interested to simulate it, and in normal PSU designs multiple faults would be needed to even start to get to this problem. It can happen with cars because the car CPC is used as a DC signal return, but it still needs a short from the signal wire to neutral to cause a problem. A short to live would destroy the signalling system, and probably be very expensive to fix, and unless there is no neutral to Earth voltage a N-signal fault would also probably be damaging.

I suppose that in say a TN domestic installation where there is good earthing and boning, M.C.B.s will trip off if an exposed-conductive-part becomes live due to a fault. No R.C.D. needed. But there are other situations where perhaps water leaks onto electrical parts, or a pet chews through some flex insulation, or a person touches a live damaged garden appliance flex then an effective swift disconnection is required by an R.C.D.


Z.

Plus there is not (and has never been) such an item as a 17th Edition board , or an 18th either. Any board can be made to comply or not comply with quite a number of Editions of BS7671. All these terms such as a 17th Edition or High Integrity Board are just a silly name used by some suppliers/manufacturers and others

Perhaps I have to get an 18th edition (book) sometime ... what's the change about RCDs? Anything firm, or just a bit more stuff about making wise choices, being thoughtful, reading appliance manufacturers' demands on RCDs (what does the word "externality" mean in economics?), etc.   Perhaps there's no point in my little battle against the AC type!  A year and a half ago I was surprised at how hard it was to find a simple type A in the UK. Since then I've been noting how the ability to use type-AC "where appropriate" leads to their general use where not really appropriate, and how the price of A tends to be more in the UK than in countries where AC is not permitted.  I was working on writing something up about this, but perhaps I'm unaware of recent changes in the rules that would make this superfluous - although it doesn't look as if AC has gone down in popularity in terms of what the retailers offer.

Whilst what he says is correct, DC leakage is not a normal phenomenon in installations. We have discussed it in car charging at length because there it is possible due to the car communications system with the charger and a single fault, but it is probably not possible with other electronic loads which never leak DC because the design runs with AC! Asymetric AC and other things do not result in DC (and it needs to be pure and smooth) leakage.

I suspect problem is a bit wider than just AC type RCDs being blinded by pure d.c. Have a look at Figure A53.1 in the regs (page 192) - and then imagine a fault in an appliance from L after some of those diodes to the metallic case - especially on a TT system were we can't rely on the earth fault current being sufficient to trigger overcurrent protective devices or even blow the rectifier to smithereens. Then look at the right most column and note how quickly AC types disappear from the list of acceptable types.


    - Andy.

I never refered to ev charging or metal enclosures,only induction ovens dishwashers etc,which may have an electronic

speed control.The Napit webinar suggested the dc leakage fron the above can "blind" the type AC rcd from tripping

on a normal earth fault.I referred to the cu as 18th edition cos to my knowledge,the 17th made no mention of overload protection

for the rcds in a duel cu or the use of type A rcds in same.

                                                                Regards,Hz

Andy are you suggesting that a short to case after a rectifier makes a DC fault? First the fault current available is on average half the PSSC, and secondly if there is a large capacitor there as well the fault current will be the whole PSSC. This will not trip the CPD I suppose! Also it is not a DC fault, hence my comment above about smooth and continuous DC, because a single rectifier will give an AC fault current which WILL operate the RCD. Getting a DC fault on a mains powered piece of equipment is much more difficult than you are imagining, and is very unlikely without multiple faults being present, most of which would stop the equipment operating at all. Imagine the PSSC at a socket is only 100A, how long do you think even a 10A rectifier diode will last, across the mains? It might last 1 or 2 cycles and by then the CPD will have tripped. Take an induction hob. Is there any smooth DC there at all? Probably not in a form you would recognise, as the induction circuit probably works off straight rectified DC, with the usual full wave rectified waveform in order to provide unity power factor! The idea that electronics can provide sources of the DC current as used in the video, or my RCD experiments, by accident due to faults is living in a fantasy where broken electronics still work perfectly (which is never!). This rectified waveform will trip an RCD just fine, but just how many faults do you need to get it to a person who gets a shock from the earthed case?


I am not particularly suggesting that type AC RCDs are ideal or adequate in all cases, but as they are almost exactly the same as a type A, perhaps with a better current transformer costing very little extra, perhaps the manufacturers would like to sell them at the same price? There is no reason why they should cost more here than across Europe, we are supposed to have a level playing field (Barnier!).


Hertzal123, there is a lot more to this than the words DC might indicate, and the complexities need sorting by JPEL/64 if anywhere. Just stick to BS7671 and there will not be any problems for you. Anyone with any doubts should probably study the problems themselves by experiment, you will find the learning to be fun, and very worthwhile. Transformers of all sizes respond to changing currents through the windings. DC does not stop them completely even if saturated, a larger AC current in the opposite direction still gives secondary current some of the time, and so this is only a potential difficulty where the RCD is used for additional (personal) protection. Those used for ADS on TT will not stop working with Earth faults even with some DC. In the video you will notice that the DC curents used are fairly small, nothing like that which Andy is considering above with direct short faults. With amperes of DC no type will work at 30mA and will need amperes of fault to trip at all. It is a subtle matter of degree.