Why was the RCD not tripping with a N-E short?

[HappyHippyDad]

I went to a property to carry out an EICR.

Sockets and cooker were RCD protected all the other circuits were not.

I realise there is an order of doing the tests but I thought I would check the RCD test button. It didn't trip. I then did an RCD test and the RCD didn't trip.

Took the cover off the CU, started some dead tests on the ring. N-E 0MΩ.

Upon re-energising the ring the RCD kept tripping instantly as you would expect with a N-E short. However all the sockets had been working fine before I arrived!

After an hour of fault finding (as I couldn't leave him without sockets, even though it was just an EICR) I found a neutral touching a back box as the insulation was broken and there was no grommet. Fixed it and the IR results went to 50MΩ and the RCD then worked as it should. I.e the button tripped it and it tested as 11mS at 5x rated current.

My question is why was the RCD test button not working AND why did it not trip when tested? It only became a working RCD once the fault was found which kind of defeats the object!

 

[telectrix]

had this several times. a dead short N-E can inhibit the RCD from tripping, even though you think it should. not sure of the science behind it though.

 

[HappyHippyDad]

had this several times. a dead short N-E can inhibit the RCD from tripping, even though you think it should. not sure of the science behind it though.

Not good enough! You need to find out quick smart and then let me know. I'm too busy making bread and am also about to have a cup of tea!

 

[telectrix]

did the RCD trip when you switched on a load?

 

[buzzlightyear]

My question is why was the RCD test button not working AND why did it not trip when tested? It only became a working RCD once the fault was found which kind of defeats the object!

electricity can be a wonderful thing ,some times it works and some time dont ,with all those atoms flying about see if you can catch one .

 

[buzzlightyear]

I'm too busy making bread and am also about to have a cup of tea!

probably got his head in the oven and drinking tea the same time .

 

[HappyHippyDad]

did the RCD trip when you switched on a load?

No, completely dead. The chap said it had never tripped.

 

[Gavin John Hyde]

It could possibly be that the switch mechanism had seized up, I have had them before where there is a obvious fault but didn't trip, had to put a bit of force on it to manually move it. then it wouldn't stop tripping. how many people actually bother to test the rcd and follow the little stickers?

 

[TheReturnOfHightower]

I'd say as @Gavin John Hyde - switch mechanism seized and by isolating you've freed it up a bit.

Was it a TN-C-S? Forgive me, it's late on a Friday so my brain may not be engaged, but with TN-C-S you've got a dead short at the service head and a dead short at the fault. Therefore, maybe the majority of fault current was still going down the neutral and only a bit, not enough to trip the device, going down the fault path?

 

[Shoei]

I think this is a known issue with RCD test buttons, if the neutral is linked to earth at source and a fault links N-E after the RCD, then there is a current path which bypasses the sense coil in the RCD. Little or no current flows in the coil so no trip.

 

[TheReturnOfHightower]

I think this is a known issue with RCD test buttons, if the neutral is linked to earth at source and a fault links N-E after the RCD, then there is a current path which bypasses the sense coil in the RCD. No current flows in the coil so no trip.
View attachment 47841

Hmmm, I've learned something. All the RCD wiring diagrams I've seen previously have the test button after the coil on the line to before the coil on the neutral.

 

[Shoei]

Hmmm, I've learned something. All the RCD wiring diagrams I've seen previously have the test button after the coil on the line to before the coil on the neutral.

Either test arrangement should work, the button only has to create an imbalance in the sense coils. I was just trying to think why an N-E fault would stop the RCD tripping and that's the only thing that makes sense.

 

[Midwest]

I’ve experienced this phernonmon’ before, when on a new install, due to my mistake I connected a circuit with neutral earth fault. Got to RCD tests with MFT, and RCD would not function. After a bit of head scratching, and testing from beginning, realised I’d connected faulty circuit. Disconnected faulty circuit and RCD functioned correctly. Posted before about and think someone like Lucien explained why.

 

[HappyHippyDad]

Interesting replies. It was a real head scratcher! Plus I was in a bit of a rush as I had to pick the boy up from School, couldn't believe it when all the sockets tripped and I was only doing an EICR...arghhhh!
I expect Lucien, Richard, Darkwood or Marvo will explain
E54 would have known too!

 

[Wilko]

Law according to Sod ...