Earth - neutral fault on ring final, any ideas on how to find with ease.

[ericmark]

I have split the neutrals and have half of the ring final back, I split on a USB socket, as thought the socket may have been faulty, and not have power back to 85% of sockets, that will do for tonight, now I need to find the fault, the RCBO type AC did not switch the neutral, so swapped for a type A which does switch the neutral so I can now easy switch off to continue fault-finding.

But many sockets had to get to, so considering any tests to reduce the amount of furniture I need to move. To access a socket I can cut a hole in the back of cupboards, but to remove the socket really need more room, so trying to think of a way to test without removing display cabinet.

 

[pete01]

Disconnect at a socket either side of the cabinet and test.

Its bound to be a problem at the socket you cant get to.

 

[Murdochcat]

Did you do any IR tests?

 

[ericmark]

Did not need IR test, was down with multi-meter test. When I got one section working, was surprised how much I had not unplugged. The socket which is still not powered is not where I expected to find it, I have been trying to work out likely run, but since earth neutral the fault may have been there for years, some sockets behind this and I am not looking forward to removing the left two units and going behind them, specially since not sure if that is where the fault is.

 

[JohnW2]

Did not need IR test, was down with multi-meter test.

Whta is the resistance of this N-E fault?

 

[ericmark]

The meter showed just an ohm or two at the socket, but then realised the RCBO was not switching the neutral, so not really a good reading, family say I should no work on it on my own, so I have a week until they come back from holiday, so only thinking about the best way to tattle.

Have one socket in living room, and possibly sockets in wife's craft room, but hard to get to both, so may even rope my son into helping me.

Out of interest, I had said a few times how I thought the surge protection device was why my LED bulbs lasted a long time, well the one in main CU the line leg was not connected, so only one SPD connected. Which was fitted Sept 2023 so did first 3 years without a SPD fitted, kind of gets rid of my argument that it stops LED bulbs failing.

 

[JohnW2]

The meter showed just an ohm or two at the socket, but then realised the RCBO was not switching the neutral, so not really a good reading ...

Is it a TN-C-S installation? If so, with the N still 'connected' one would obviously expect a very low N-E resistance, and even with TN-S not only would the resistance be pretty low but the (inevitable) small N-E pd might well totally muck up the measurement.

However, you obviously could have disconnected the N at the CU to make a measurement and, in any event, I gather that you have now replaced the SP RCBO with a DP one, so can now simply 'switch it off' for measurement. What is the N-E resistance of the circuit when you do that?

Given the recent discussion about such things, I presume that you are somehow 'certain' that the RCBO has been tripped by residuyal current and not overcurrent?

Out of interest, I had said a few times how I thought the surge protection device was why my LED bulbs lasted a long time, well the one in main CU the line leg was not connected, so only one SPD connected. Which was fitted Sept 2023 so did first 3 years without a SPD fitted, kind of gets rid of my argument that it stops LED bulbs failing.

Interesting. As you know, I was personally always extremely doubtful about that argument of yours

 

[robinbanks]

Are you doing this in halves? Disconnect at CU end, check for fail (should show on both), disconnect half way round, only one should fail (unless two faults), disconnect that one half way etc

If you can gauge the route of the ring that is..

If there isn’t an obvious route I suppose you could, disconnect at the CU, then any other socket and in the socket clip a 9v battery on one and a 1.5v on the other wire L/N, then visit the other sockets with a meter clipped to a plug, to quickly test which half (9 or 1.5) that socket’s on and form a map of the house, then split again at some other socket and move your batteries to that. Gradually you’ll work out the order of rooms

 

[ericmark]

The disconnection of neutral only to start with was because my screwdriver was not long enough to reach the line, so socket removed and a connector block fitted to keep wires safe and separated, fitted where I had the USB socket, as wondered if the socket was faulty, trips too fast to measure earth leakage, but total earth leakage for whole house is high at the moment showing around 100 mA, but when I tried around earth wires nowhere near that high, that was the differential around the meter tails. I am not sure about the SPD and if that was reason for some odd reading? The lead was not clamped but was touching, and it was because of the SPD I was not using the insulation tester, but given up for a few days, not safe standing on a step-up working on the CU on my own.

But why an earth - neutral fault should show now, when CU has been fitted for 5 years, has me scratching my head, mainly as working out possible cause, may lead me to find the fault faster.

My first thought was a mains filter gone faulty, hence unplugging all I could. The earth leakage was 8 mA actually took a picture so why now showing 100 mA don't know, 14 RCBO that's 12 type AC and 2 type A shed and faulty ring final on type A now, the 32 amp RCBO feeding back kitchen cooker had failed so had a new one in ready to change, so ring final one moved to cooker, and new one on the ring final, not as yet tested to see the ramp value, but since it was a handy ring final test my new RCD tester on, I know it was well within parameters and did not trip at ½ value, the problem is a set of steps between CU and the ring final sockets in question, so to reset have to walk outside and around the house to the flat under main house to reset, so really a job to do with two people.

I have a similar problem other side on the house, and last thing disconnected before it would reset was dishwasher, so though I would use insulation tester before buying new, and all OK, refitted all and tripping stopped, never did find the cause. The ramp test showed I had around 22 mA to spare, so not just on the edge of tripping.

So it does seem certain it is the neutral at fault, so what can change to cause a neutral earth fault? My thoughts were a multi-way extension with a filter gone faulty, but the two leads plugged in are not filtered.

 

[ericmark]

May have found it, laundry room on same ring, and I have a USB socket in there. Anyway that room needs testing before I move any furniture.