Your feedback on regulation issues

Chris

I think that your description of trying to calculate diversity with real loads is interesting, and a serious part of proper installation design. However you will probably see at once that one 32A circuit for the clothes (cloths?) machines is adequate, and one for the kitchen (with dishwasher, kettle, microwave and small appliances) is also adequate. It is very unlikely that either of these will suffer nuisance tripping unless you really try, and even then the intermittent nature of them all will require considerable coordination of actions. If you have electric heating you would need to add another circuit. The potential loads for other socket circuits in the house is fairly trivial, but several final circuits may be installed for convenience. Ignoring cookers and showers for the moment, your maximum demand at any time on socket circuits is likely to be significantly less than 64A, and for any period of more than a few minutes, less than 32A. (that is 32A general sockets, 32A kitchen, 32A washing). However using the Hager "rules", you would need a 100A RCD for these circuits I have described, because all 3 circuits are the maximum rating so must be added at 100%. A reasonable designer would see that this is incorrect and would be quite safe with a 63A RCD, sustained operation at more than this is very unlikely indeed. If you add a few lighting etc. circuits at 6A each, a shower at 50A, a cooker at 40A, you may not use a 100A rated CU at all, you need at least 2! Even under the "rules" the outcome is dubious as 50 A and 40% of the rest comes out at over 100A, the main switch rating. The 100A DNO fuse will stand 140A for the best part of half an hour, but apparently the main switch in the CU will not. The 16mm internal cables would be quite hot by now, as would the tinny "busbars" and steel neutral strips. The 25mm tails you have used would be slightly warm to the touch.


The meaning of "protection against overload" here is being abused, in that overload is not an instantaneous current of a bit more than the maximum continuous current, but a fully diversified maximum current taking time and consumption into account. If one assumes that actual maximum without the time factor, almost all installation designs fail at once. Take a factory with electric motors. These take roughly six times the FLC at startup, and may take 30-60 seconds to reach FLC, so the whole installation must be designed to deal with a simultaneous start of all the motors, poor National Grid.


The point at issue here is that one can never be sure that the current will not be maximum for a long period, but it is extremely unlikely. This is the judgement, which is entirely missing from the Hager leaflet, although it is to some extent present in the BEAMA one. However, inspection is still a major problem, the MI make this pretty much impossible for most installations. One would have to give a C2 as potentially dangerous according to the MI.

My reading of that Mike, is that the RDCs require a protective device of B or C type MCB (Not a D) of not more than 50A, again emphasising that the RCD maximum current is not more than 63A (I assume they are 63A RCDs) at all. The fact that the leaflet was printed / written in 2012, is interesting and is probably not the current position, which is very poor. You now have to meet the BEAMA sheet which I would inspect your new installation against for compliance, and if you don't fit these MCBs you get an unsatisfactory C2! In fact as you are fitting RCBOs there is no problem (at the moment), but if that were a filled unit of older stock it may already fail their own paper unless you have a 60A main fuse.


In my view this is not either the position or intention of BS7671 itself, and it is the MI (or perhaps even the product) which is at fault.


On another note my playing with RCDs is still proving interesting, and I am not at all convinced that an RCD with a high fixed load still trips at 30mA or whatever, particularly if the waveform is not ideal. Perhaps there is more to this than meets the eye.

The only way to do this is very bureaucratic and would be a "Technical construction file" or TCF

I suspect we're stuck with manufacturer's instructions - there already needed for everything from bathroom appliances to being able to use MCBs with reduced c.p.c. (and likely ring circuits) with UK PFCs. We probably have a similar problem with published product standards (BS, BS EN etc) changing their technical content over time.


A better alternative to a TCF for every installation might be a national repository of manufacturers instructions and the like (perhaps hosted by the IET?) - ideally indexed by some unique product number that was printed somewhere on each device but also searchable on make/model/date-of-production together with an image to aid identification. It could be a requirement of the wiring regs that any item could only be installed if its instructions had been lodged in this repository by the manufacturer. (A bit like the Legal Deposit requirement that every book publisher has to give a copy of each book to the British Library). Future inspectors & designers of modifications would then have a reliable source for such information, without the inefficiency of making millions of copies across the country.


   -  Andy.

This is beginning to turn into my blog, I would welcome other comments given consideration of the above.


Andy, 536.4.2.4 & 536.4.3.2 actually give the game away completely, it says we need the manufacturers instructions (MI) or information several times, and says the OPD shall usually be UPSTREAM of the RCD or whatever. So adding up all the MCB ratings if they are downstream would be the same as 536 suggests, and this application of some "magic" diversity some kind of sop to complaints or perhaps reason. But even with that the instructions still say further protection is needed as in Mikes unit, so the situation is further confused.


Although I read the DPC for the 18th I don't remember this change, which is very significant. I think we have always assumed that the protection of RCDs was adequate, but this seems to be saying that someone has a significant problem with overcurrent, although I know of no cases. Perhaps someone does, in which case please reply.

The point is Andy that inspection is to the current regulations, and would also have to be the current MI, but of course this is allowing retrospective change to existing installation standards by manufacturers. Really standards should drive manufacturers, not the other way round. Clearly the best change would be to only sell 100A RCDs (Interesting thought, would they have to 130A ones to allow for the fusing factor?), probably a trivial cost increase over 40, 63 and 80A ones. In this case all existing domestic installations would be considered satisfactory, and new ones protected to the new rules. This is the proper way to handle something like this, and I am amazed at JPEL/64 allowing this mess to happen. It could be that manufacturers are having too much say somewhere, it would be interesting to find out.

You want incidents to pass on to the people who can make a difference. I'm just wondering what they would change it to, and why would they want to?


F

Keep going blog or no blog David, this is certainly very interesting. In Ireland we have some likely mitigation in that a main overcurrent device is required to be fitted. For domestic and similar installations these tend to be 63A type C MCBs although where the DSO fuse is 100A a max MCB rating of 80A is required. I am just trawling through IS10101 to identify any clauses that might be worthy of discussion.

The example that I gave of the utility and kitchen circuits is an example of the folly of this new approach. Most people would have the large appliances on the same circuit as the sockets above the work surfaces. Splitting the circuit into two has not, of course, made a jot of difference to the overall load, but it has potentially made the installation non-compliant under the new rules.


I would be surprised if this particular house ever draws 63 A total. It is well insulated, has gas CH, and coal fires (not that there will be any coal soon). If the boiler broke down, there might just be 1 x 3 kW immersion, 1 x 2 kW fan heater, 1 x 1.5 kW washing machine, 1 x 1.5 kW kettle, and a few other bits and bobs on simultaneously, but not for long.


The crucial thing is that there will be no electric showers, no EVCP (on this CU), and no heat pump for the pool (there isn't one).


My interpretation is that the installation is compliant.


Incidentally, the customer's "nuisance tripping" in his current home is due to inappropriate design or use (e.g. woodworking machinery on a type B MCB) and shoddy workmanship.