N - E Fault on TNS Supply?

[LeccyLad1990]

Hi All,

Right, I have a very strange fault occurring so I'll start from the beginning.

I have recently carried out a consumer unit change from an old rewireable to a new split load RCD protected consumer unit. So I had changed the board and put new meter tails in.

I have tested all the installation cabling and it would be classed as satisfactory (it's not perfect considering it's over 35 years old).

When it came to energising the board, I switched on the main switch and the two RCD's (63A 30mA) first then started energising the circuits. However when I energised the first circuit ( in this case a shower ) the RCD tripped. So knowing that there could be an issue with that particular circuit I tested another circuit (in this second instance it was a feed to a bell transformer with no earth as it was double insulated) and disconnected all other circuits including neutrals and earths and bonding leaving only the main earth. So now it was just a bell transformer on a 6A MCB connected to a 63A 30mA RCD connected to the 100A main switch and off to the cut-out. Once again the RCD tripped instantly.

I carried out a Ze which came back as 0.30 ohms which would likely be classed as a ‘pass’ from the DNO even though it's quite close to the limit.

I swapped the RCD’s around in case one was faulty, but it still tripped.

I checked the supply polarity and that too was correct.

I am genuinely stumped by the fault, all I can assume is that maybe there is a neutral - earth fault on the incoming supply?

If this is the case whats the solution? Would the DNO have to repair the cable?

I’d appreciate any help you all could give.

 

[Mikegh]

Is the board wiring correct

 

[Pretty Mouth]

I am genuinely stumped by the fault, all I can assume is that maybe there is a neutral - earth fault on the incoming supply?

Think about TN-C-S, and that will tell you why this wouldn't be a problem

 

[SJD]

Photo of the board, showing the internal wiring?

 

[LeccyLad1990]

All of the internal wiring tested out fine, I thought originally maybe a borrowed neutral somewhere but I disconnected all of the other outgoing circuits, then connected only one circuit at a time regardless which circuit I chose the rcd still tripped.

 

[LeccyLad1990]

Also Pretty Mouth I know where your going with them being connected at the star point of the transformer. Again it was just a stab in the dark.

Like I said I had disconnected all the outgoing circuits, I tried energising just a lighting circuit and the rcd tripped as soon as a load was called for. Same again with a socket circuit and same again with the doorbell transformer.

 

[timhoward]

Is the other side /RCD ok?
What I think you have is a N-E fault on the side you were working on. Any load on any circuit will find it's way back to the N bar, then to the faulty N, then to E and trip the RCD.
Best way forwards take all the N's out and do L+N to E IR test (250v will do) on each circuit. Identify the faulty circuit. Go from there.

 

[timhoward]

Also Pretty Mouth I know where your going with them being connected at the star point of the transformer. Again it was just a stab in the dark.

Like I said I had disconnected all the outgoing circuits, I tried energising just a lighting circuit and the rcd tripped as soon as a load was called for. Same again with a socket circuit and same again with the doorbell transformer.

Disconnected: do you mean you turned the breaker off? Or you also removed the Neutral.

 

[LeccyLad1990]

Disconnected: do you mean you turned the breaker off? Or you also removed the Neutral.

Completely removed L N and CPC

 

[timhoward]

Ok, prove the RCD. With all circuits completely out, ramp test the outgoing terminals.
Assuming this gives 15-30ma it's good.

 

[LastManOnline]

I am genuinely stumped by the fault, all I can assume is that maybe there is a neutral - earth fault on the incoming supply?

Then you would assume wrongly. The imbalance must be downstream of the rcd. An imbalance upstream would normally not affect rcd operation.
Your first step needs to be to check the wiring of each circuit individually and through the usual process of elimination you will establish where the fault lies

 

[timhoward]

If any load on it's own is causing this I'd check for a simple mistake, e.g. where is the N outgoing fly lead from the RCD going - definitely the right N bar?
(sorry to ask but this is unusual so I'm asking anything that comes into mind!)

 

[timhoward]

And also, how is the other side of the board behaving?

 

[LeccyLad1990]

Haven't ramp tested the rcds yet, All the fly leads go to the correct neutral bars (neutral link from the outgoing of the main switch to 1st neutral bar(non rcd protected citcuits), second link outgoing side of min switch though the rcd and outgoing to neutral bar 2, and a link from neutral incoming on rcd 1 to neutral incoming on rcd 2 then outgoing to neutral bar 3).

I even swapped the two rcds position wise, in case one was faulty but still the same result.

I have had to remove the rcds temporarily to get the power back on for the customer (while I figure out what the hell is going on). Would rigging them up to some leads and a 3 pin plug at the office suffice for ramp testing? I know it might trip the office leccy but hey its Christmas

 

[timhoward]

Would rigging them up to some leads and a 3 pin plug at the office suffice for ramp testing?

I've done this in the past. Finding a non-rcd protected circuit is obviously best.
I've abused a shaver socket (mini isolating transformer) is desperation once!

What brand board is it by the way?

 

[Pretty Mouth]

Also Pretty Mouth I know where your going with them being connected at the star point of the transformer. Again it was just a stab in the dark.

Or connected at the cut out perhaps...?

It shouldn't matter anyway, it is very rare that anything other than a downstream fault will cause an RCD to trip.

My best guess is you have the neutral conductors for the circuits protected by one RCD in the neutral bar for the other, and vise-versa. This is what I would be looking for, or some other similar wiring mishap.

It seems unlikely that all circuits have developed earth faults since testing prior to board change, and unlikely to have 2 RCDs that have the same defect (although not impossible, could be a dodgy batch)

 

[LastManOnline]

All of the internal wiring tested out fine, I thought originally maybe a borrowed neutral somewhere.

That would be my gut feeling. I would go right back to basics here. Starting with each individual neutral I would test them to every other individual neutral. Has someone possibly mucked up an old ring circuit?
A neutral to earth fault would normally require more than just a bell trafo to create a, significant enough volt drop to operate an rcd. It usually takes a heavy appliance or two.
These types of faults are typical of installations that never before had rcd, s

 

[LeccyLad1990]

Or connected at the cut out perhaps...?

It shouldn't matter anyway, it is very rare that anything other than a downstream fault will cause an RCD to trip.

My best guess is you have the neutral conductors for the circuits protected by one RCD in the neutral bar for the other, and vise-versa. This is what I would be looking for, or some other similar wiring mishap.

It seems unlikely that all circuits have developed earth faults since testing prior to board change, and unlikely to have 2 RCDs that have the same defect (although not impossible, could be a dodgy batch)

Sorry Pretty Mouth I think we have some crossed wires here it's a TNS supply not TNCS.

Also like I said to Tim, I had all of the circuits disconnected and had just 1 circuit connected at a time and the rcd was still tripping. There couldn't have been crossed neutrals because nothing else was connected.

 

[LeccyLad1990]

I've done this in the past. Finding a non-rcd protected circuit is obviously best.
I've abused a shaver socket (mini isolating transformer) is desperation once!

What brand board is it by the way?

Hi Tim, unfortunately they gave me a BG one :S

 

[timhoward]

Sorry Pretty Mouth I think we have some crossed wires here it's a TNS supply not TNCS.

Also like I said to Tim, I had all of the circuits disconnected and had just 1 circuit connected at a time and the rcd was still tripping. There couldn't have been crossed neutrals because nothing else was connected.

His point was that a TNCS links earth and neutral, so the earth to neutral 'supply fault' scenario you suggested exists all the time.