RCD tripped when MCB was already off

[skotl]

Last year I had an electrician run SWA to the garage and put a small CU in the garage. This was part of work we were having in the kitchen so there's a new small CU in the house, in addition to the original CU, which has the garage feed plus the new feeds for the various kitchen kit.

As you would expect, the new CU in the house is RCD protected. The feed to the garage is via a C32 (cos of the welder!).

The garage CU has three MCBs; lights, 13A sockets and 32A socket for the welder.

Today, I replaced the two rather feeble lights in the garage with fluorescent tubes. I checked for live at the light fittings, flipped the lighting MCB off, and checked for dead - all good.


However, when pushing the T+E through each fitting, the cores inevitably touched. Each time, the RCD in the house tripped.


Apologies for the long lead-up to the punchline; just wanted to make sure all the info was there. Why would an (apparently) non-energised circuit trip the RCD? Capacitance from the still-live socket circuits?

 

[Mattenstein]

Just to beat the experts to it

An RCD detects ANY current leaking to Earth and even if you turn off the MCB of the circuit you're working on, thus disconnecting the live feed, the neutral carries a small charge from the backfeed of other active circuits. This will trip the RCD should the neutral touch the CPC.

When working on an RCD protected circuit you either need to isolate both poles or take extra care not to touch cores together.

I came across the issue with a toaster once, I needed to poke a fork in there to remove some toast so I switched off the socket at the wall. However as the socket wasn't double pole and I was effectively touching the element of the toaster to the casing, I was touching neutral to earth and thus tripped the RCD.

 

[bernardgreen]

Here is the reason

Neutral_Earth_trip

• bernardgreen
• 16 Jan 2011
• 48

 

[ericmark]

Line & Neutral are both considered as Live. Where the supply earth system is TN one can often get away with only switching the line. Where the supply is TT (uses an earth rod) then full isolation is required that means turning off at the isolator not just the MCB.

The longer the distance between RCD and the load the more likely there is of a neutral earth fault causing the RCD to trip.

 

[bernardgreen]

Line & Neutral are both considered as Live. Where the supply earth system is TN one can often get away with only switching the line.

You can only "get away with it" if there is no load on any other circuit to create a current along the neutral bar.

 

[skotl]

Thanks for the responses.

So is the conclusion that that this is expected behaviour, and there is probably no fault?

The reason I ask is that I am unsure whether Eric is suggesting there may be a fault;

Line & Neutral are both considered as Live. Where the supply earth system is TN one can often get away with only switching the line. Where the supply is TT (uses an earth rod) then full isolation is required that means turning off at the isolator not just the MCB.

The longer the distance between RCD and the load the more likely there is of a neutral earth fault causing the RCD to trip.

 

[JohnW2]

So is the conclusion that that this is expected behaviour, and there is probably no fault? ... The reason I ask is that I am unsure whether Eric is suggesting there may be a fault;

I can understand your being a bit confused. The answer to your question is that if there are any loads on any any of the other circuits protected by the same MCB (and,in some cases, even just loads on circuits protected by other RCDs), then touching the neutral wire to earth WILL almost inevitably cause the RCD to trip, even if the MCB (single pole, switching only line/live) is switched off. So, yes, 'expected behaviour', and no indication of any 'fault' (in the normal sense).

To put eric's comment into context, in your case the 'fault' to which he was referring would be your touching neutral and earth together!

Kind Regards, John

 

[skotl]

Ah - roger that - thanks John, and everyone else.

Good to hear that I'm the fault, rather than the wiring

 

[JohnW2]

Line & Neutral are both considered as Live. Where the supply earth system is TN one can often get away with only switching the line.

As bernard has implied, it would be extremely unusual that one would 'get away with it' (in terms of being able to touch N to earth without an RCD tripping) with only L isolated.

Where the supply is TT (uses an earth rod) then full isolation is required that means turning off at the isolator not just the MCB.

That's a regulation about isolation, rather than anything to do with the OP's question/issue. I have to say that the regs seem to be ultra-cautious in this respect. I suspect that one is probably more likely to be 'struck by lightning' than to be working on a TT installation at a point in time when there was a dangerous potential between N and E! T'other way around, given their ultra-caution, I'm a little surprised that they appear totally confident that there could never be a dangerous potential between neutral and a TN-S earth - the only really 'foolproof' situation is with TN-C-S, where N and E are joined actually withinthe property!

Kind Regards, John