Fault Finding Conundrum

[EFLImpudence]

Yes, I was talking about the diagram in the link where that is not the case.

I don't see how your shower current can 'back-feed' through the other RCD.

 

[JohnW2]

I'm still thinking about it and getting confused, but I don't think so.

I'm got to go out shortly, but if I get a chance when I come back, I'll try to draw a diagram.

Yes, but you are saying that an appliance elsewhere - hence anywhere - would cause it to trip.

If any circuit, protected by any RCD or RCBO, in any CU, has an N-E fault, then some of the current from any/every circuit in any CU will, instead of flowing into the supply neutral, instead flow 'backwards' (to earth, through the fault) through the RCD/RCBO protecting the circuit with a fault, won't it?

Before you ask, I'm not even sure that that is restricted to load currents in the installation in question - some of the neighbour's neutral current might well 'flow backwards' though an RCD/RCBO protecting a circuit with a N-E fault (thereby potentially resulting in a trip), mightn't it?

Kind Regards, John

 

[EFLImpudence]

I'm got to go out shortly, but if I get a chance when I come back, I'll try to draw a diagram.
If any circuit, protected by any RCD or RCBO, in any CU, has an N-E fault, then some of the current from any/every circuit in any CU will, instead of flowing into the supply neutral, instead flow 'backwards' (to earth, through the fault) through the RCD/RCBO protecting the circuit with a fault, won't it?

Yes, I suppose I must be wrong otherwise RCDs wouldn't trip when you touch Neutral to Earth.

Before you ask, I'm not even sure that that is restricted to load currents in the installation in question - some of the neighbour's neutral current might well 'flow backwards' though an RCD/RCBO protecting a circuit with a N-E fault (thereby potentially resulting in a trip), mightn't it?

Yes, that is what is confusing me with it being a different CU.

Sorry.

 

[EFLImpudence]

Yes, my mistake.

 

[crystal ball]

I was called to a similar fault, various appliances tripping the RCD, I found a melted cable in the shower, stripped it back and reconnected, problem solved

 

[funkydiver]

Thankfully it's not the shower, with that isolated the fault still occurs.

I'm going to have to get everything unplugged tomorrow and use a bit of time testing.

 

[JohnW2]

Yes, I suppose I must be wrong otherwise RCDs wouldn't trip when you touch Neutral to Earth.

Exactly. If the 'touching of N to E' is on a loaded circuit, then it's obvious that the RCD will trip. However, as I've said, even if the circuit has no load, neutral currents from other circuits (even from other CUs, and maybe even other installations) can still 'go backwards' through the RCD (whence via the fault to earth), thereby causing a trip.

Yes, that is what is confusing me with it being a different CU. Sorry.

As above, I think that it might even be a 'different installation'. If some of the current (from wherever) which is meant to flow into the supplier's neutral in fact 'goes backwards' through someone's (anyone's) RCD which is protecting a circuit with an N-E fault, then that RCD is likely to trip.

As is apparent, I'm back now, so will see what I can do about a diagram!

Kind Regards, John

 

[EFLImpudence]

My diagram above is alright for two CUs on the same installation?

However, where this occurrence is from two installations (next door properties) surely the fault would only result in tripping on A through stray current from B if it is TN-S and not TN-C-S.
Which (because of the two CUs) is how I mistakenly drew it first.

Unless of course the fault was to true earth and not the CPC.
I suppose it could go to true earth anyway through the bonding.

 

[JohnW2]

My diagram above is alright for two CUs on the same installation?

I have to say that I'm having some difficulty in understanding your 'above diagram'.

However, where this occurrence is from two installations (next door properties) surely the fault would only result in tripping on A through stray current from B if it is TN-S and not TN-C-S.

Yes ... or TT ... or, as you go on to say, if it is a fault from neutral to true earth, rather than to CPC (which, with TN-C-S is the same as the neutral once one is upstream of the RCD).

Even with TN-S is probably fairly unlikley that a load 'next door' would produce this issue since the impedance of the between-houses-neutral would probably be sufficiently high that the vast majority of the neighbour's neutral current would flow into the supplier's neutral, with only a tiny fraction (quite probably not enough to trip the RCD) flowing through the other house's RCD and fault to earth.

Kind Regards, John

 

[funkydiver]

So the upshot is... The fault is most likely on CU1, but a load being applied on CU2 would be enough to trigger the RCD?

Presumably I would be able to test sockets with a simple socket tester from Screwfix once I've discounted appliances?

 

[JohnW2]

So the upshot is... The fault is most likely on CU1, but a load being applied on CU2 would be enough to trigger the RCD?

If by "on CU1", you mean on a circuitr supplied by CU1 then, yes, that would be my suspicion.

Presumably I would be able to test sockets with a simple socket tester from Screwfix once I've discounted appliances?

That I rather doubt. It's also far more likely that some a fault in some load (maybe one you've overlooked and therefore not yet disconnected) is responsible than the installations's wiring/accessories.

Kind Regards, John

 

[funkydiver]

Grrrr.... Talk about frustrating. Checked every appliance yesterday. Plugged everything in, turned on the shower, 3 vacuum cleaners and the oven.

RCD didn't activate. So I'm no further forward.

The only event that I can think of, we were looking after my brother in law's dog last week, and he peed on the bare floor board landing, above the light fitting below. Ergo, dropping through the floor boards (Old dog, on his way out bless him)

Without ripping the floor boards up to check, is there a chance that this could have been causing the issues I've been having, and now it's had chance to dry out the problem has effectively resolved itself?

 

[JohnW2]

.... The only event that I can think of, we were looking after my brother in law's dog last week, and he peed on the bare floor board landing, above the light fitting below. Ergo, dropping through the floor boards (Old dog, on his way out bless him) .... Without ripping the floor boards up to check, is there a chance that this could have been causing the issues I've been having, and now it's had chance to dry out the problem has effectively resolved itself?

Water/fluid ingress is certainly one of the most common reasons for RCD trips.

However, that would normally result primarily in a L-E (rather than N-E) fault, so probably could not explain RCD trips occurring as a result of loads on the other CU.

I think all you can do is wait and see.

Kind Regards, John