Very odd problem what do you think ?

[Dave OCD]

I might be overlooking something that's right under my nose but opinions would be much appreciated. I visited a customer today with a detached garage supplied with power from a 13a RCD fused spur in the conservatory [spurred off RFC].
The RCD is 'tripping' randomly so I went to investigate.
The unit in question must be around 15 years old or more and immediately I found that the test button did nothing, a test at a garage socket also failed at x1 and x5 and the RCD refused to operate and I then found the same with the load disconnected testing at the spur itself. [SFCU in case of pedantry].
So after isolating and doing an insulation resistance test at 250v with everything switched on in the garage [freezer and lights] which showed around 40 Megohms L-E & N-E I fitted a brand new BG Nexus RCD SFCU, upon powering up the test button worked perfectly but in the garage both my tester and Kewtech plug in socket tester are showing L-E reversed !
With the supply and load cables connected together everything is fine [Highest Zs is 0.79 at the furthest garage socket].
So I have another replacement unit to fit tomorrow, it has to be that, right ?
I can't think of anything else which could cause this but opinions are most welcome even if I'm being dumb.
Sorry about the long winded tale !

 

[Wilko]

Hi - I was just going to suggest to check through the installation earthing. Perhaps a wander lead from met to that outlet so it's 100% that earth is where it's supposed to be

 

[Octopus]

I was fault finding a 15 year old house recently .... the garage CU had reversed polarity and no earth connection .......................

 

[Dave OCD]

Hi - I was just going to suggest to check through the installation earthing. Perhaps a wander lead from met to that outlet so it's 100% that earth is where it's supposed to be

Thanks Wilko, yes I'm certain about that, with the new spur in place the garage lights wouldn't come on either. I'm 99% certain it's a dodgy unit but doubts creep in as it's brand new and I can't see how the internal circuitry could create this. But if I'm right it's a very dangerous RCD.

 

[telectrix]

with the RCD out of circuit, then test for polarity. if it's right then then it is a duff RCD. dangerous and should be brought to manufacturers attention in case they have made a faulty batch.

 

[Octopus]

Does it need to be an RCD?

Does the socket circuit have rcd protection?

 

[ipf]

with the RCD out of circuit, then test for polarity. if it's right then then it is a duff RCD. dangerous and should be brought to manufacturers attention in case they have made a faulty batch.

That's it. By pass the RCD for testing purposes.

 

[Dave OCD]

with the RCD out of circuit, then test for polarity. if it's right then then it is a duff RCD. dangerous and should be brought to manufacturers attention in case they have made a faulty batch.

That's exactly what I did Tel, I know it has to be the RCD but it just seems so odd. I have another to fit in the morning so will update tomorrow, thanks.

 

[Dave OCD]

Does it need to be an RCD?

Does the socket circuit have rcd protection?

Unfortunately it's needed Murdoch as the socket circuit isn't RCD protected, I'll be doing a new CU later in the year but the Customer is a pensioner and can't afford all that just yet.

 

[Flanders]

Has the rcd unit also got surge protection as I have seen surge protection exstention leads showing crossed polarity when tested ,just a thought as I have notice some rcd sockets have surge protection fitted now

 

[Dave OCD]

I don't think so, also this one appears to be reversing L & E, so if you want to commit murder by RCD, this unit is just the job.

 

[SparkyChick]

Did you do a Zs at a socket in the garage? If the L-E really are reversed I would have expected your MFT to complain.

 

[Dave OCD]

Yes SC but only after I'd removed the RCD and temporarily joined the supply and load together, with the RCD connected the MFT complained loudly.

 

[SparkyChick]

Sorry Dave, I came back to the thread and forgot what I'd read in your OP

Bizarre situation. Definitely want to know the outcome

 

[Dave OCD]

You can see how it had me puzzled eh ?! I'll post an update tomorrow.

 

[SparkyChick]

Another question... did you prove the voltages across the 3 conductors using say an AVI?

Brain fart time I think... I'm just musing out loud

 

[Dave OCD]

No but that's a very good point, I'd convinced myself it had to be the RCD spur though as all was well with it removed and connected straight through from Conservatory to garage.

 

[loplug]

I was thinking check for voltage as well and you could do a R1+R2 and R1+Rn from the supply side of the rcd spur to the garage outlets to verify polarity that way

 

[Richard Burns]

When a tester says line and earth reversed the comparative voltages are being measured and when you have a situation where this reversal is unlikely, as in your case, it is useful to note that the same comparative voltages would be measured if the outgoing neutral from the RCD were disconnected and there was a load connected at the end of circuit.
At the garage Line-earth would be 230V, Line-Neutral would be both the same at 230V (as the load would not drop voltage) and so the pd would be 0V, Neutral-Earth would read 230V because the Neutral is at 230V. These are the potential differences you would measure if Line and earth were reversed.

So in your case I would expect the original RCD had failed and that the neutral connection on the outgoing side of the new RCD is faulty to cause the symptoms you describe.

 

[Dave OCD]

I'm wondering if it might be the RCD simply not putting the neutral through for example, then the internal tripping circuitry is fooling the MFT and socket tester into saying L-E reversal, I can't see how it could physically reverse L & E but convinced it's dodgy, I'll know for sure tomorrow.