R C Ds

Hitting the 'test' button on the downstream one may in some cases trip the upstream one too. It will always be at least as safe, and in the event of one of them failing, will still provide protection.

Wot U doin with the old RCD's?..

Assuming both are 30mA non-delayed devices you lose discrimination as to which will trip in the event of a fault, ideally you shouldn't cascade RCDs for this reason. Chances are the one nearest the supply will trip first as it will also be seeing the leakage on all the other circuits fed from it.

In my experience, they normally both trip. Unlike fuses and MCB's, the amount by which you exceed the trip current doesn't seem to have much effect on the trip time.

As long as I'm not compromising safety issues.

The ones in question feed outside sockets which are only used occasionally, and if they tripped the main one as well, I'd be no worse off, as it were.

They're flush mounted FCU types, so if I removed them I'd have to fit a blank plate etc.

A supplementary question. 'Additional' RCDs come with a warning about disconnecting before performing earth testing. Does this apply to a CU one, and if so what is the method of testing such circuits?

Either bypass it or use a tester designed for the purpose.

The reason to bypass them is that the standard earth loop imdedence test passes 20A from live to earth, which given that most RCDs are rated at 30mA might just cause them to trip!

Christian.

"Dave Plowman (News)" wrote

Depends on what else is fed by the same ring, and that you realised promptly that it had tripped (and did something about it).

It's a split load unit, and I've fed everything I can off the RCD. Only things not are the oven, immersion, alarm, and utility room ring.

Right. I doubt I could justify a new one, though.

I realise that, but wondered about the warning that comes with RCD sockets. It suggests (to me) they may be damaged by testing?

And the lights, I hope?

Christian.

All RCDs I've seen come with a warning to be removed when insulation testing (presumably they either don't like the high voltage, or would affect the result). This is usually in the form of a sticky label that you're supposed to put by the consumer unit so that the sparky knows they should be looking for one to disconnect. I can't see any reason to disconnect an RCD socket for an earth loop impedence test unless you are intending to use that very socket for the test.

Christian.

I'd take the lighting off the rcd, RCDing that creates more risk than it solves.

Regards, NT

It would be easy enough to do, but I'm going to see what happens.

The old 8-Way CU with rewirable fuses has only once blown a fuse - and that was the immersion heater.

Ho ho, or not as the case may be. The point about taking the lighting off the RCD is so that the whole household is not plunged into darkness when someone does something stupid up a ladder or with a drill or simply a faulty flex.

RCDs are not overload devices as I'm sure you are aware and fuses take a considerable overload to blow anything like quickly. 30mA leakage is not a great deal, especially if you have much computer(*) kit running through it. Surges and other supply quirks can trip an RCD as well.

(*) Stuff with switched mode PSUs, with lots of suppression stuff on the mains side.

Correct but I suspect that both going is under "fault" conditions, ie the leakage seen by both RCDs is >30mA. If one RCD is already seeing a total of 20mA of leakage and the other only 5mA if you add between

In Daves case with exterior sockets I wonder what happens if a bit of damp gets in and increases the leakage a bit...

Thats why I'm fitting radials for fridge/freezers as I refurbish this place. Eventually (ha!) these will brought back to "maintained" CU which will have it's feed switchable between the incoming supply and a gene, though as a incoming is very reliable this is very low priority.

should be pretty easy to figure out :)

rcd trips are way more common of course. Hope you never have a fire, you could bitterly regret your lights being on the rcd.

Regards, NT

Well, at the moment the whole household consists of me. ;-) And if you're fiddling with a live circuit where this could happen, isn't the whole point of the RCD to protect you from injury? Like say changing a bulb?

I realise what you're saying, but remember this is London where it's never pitch black.

I was just making the point that I don't often have circuit failures of any sort.

The computer stuff is already on a radial with its own RCD - as is the Hi-Fi. And they've never tripped either. ;-)

The more I think about it, the more it would make sense to have one RCD per circuit.