Fault finding - perplexed - help!!!

[Octopus]

Last night one of my neighbours asked me to look at a fault which had developed after new skirting boards were fitted!

Today I went round, isolated the circuit and the IR test from the board confimed a L-E very low IR, not too much of a surprise as the RCD wouldn't hold if anything was switched on on the ring.

So I split the ring in the "offending" room and confirmed IR >30 on the leg going back to the board, and IR of >30 on each "link" after that. Odd I thought.

So I refitted the first socket - IR down to about 0.80 Mohms - changed the socket for a new socket, IR value hardly budged - very odd.

Took the socket off the wall and re-tested each leg either side of the socket (and from both ends) , IR>30, scratched my head, put socket back on - IR back down to sub 1 Mohm!!, changed socket again, still below 1Mohm!!

Put if all back together, as I've got previous booked job this afternoon and obviously ring is out of action.

Going back for another look, Friday am probably

Thoughts/suggestions/ideas ???????????


Ta ta for now.

 

[Sparky Ninja]

Are you measuring back at the board when fixing back the sockets or at the socket itself?

 

[bozobro0]

If cables to the sockets come from below,and they have had new skirting fitted,i would be looking for screws/nails in the skirting under the sockets.

 

[truckster]

Does seem coincidental that probs start after skirting is fixed but from what you are saying is it possible that the insulatio is damaged on the cores inside the socket box such that when refiited its touching the box?

 

[Octopus]

Widdlet

I'm testing the "part" of the reconnected circuit at the end of the next leg (socket also off the wall) i.e. so I'm combining the leg from the CU, the socket I've just refitted and the next leg (as previously stated both legs IR test OK).

bozobro0

There are 3 sockets along this wall and the new skirting goes under each but none of the nails seem to be inline with the cables, but that's based on the cables running vertically!

Truckster

Hum, thought of that, Did look at the cables in the box at all 3 points as I took them off the wall.


Thanks for your comments.

 

[Paul.M]

Murdoch is it possible there maybe a Jb or a spur coming off this part of the ring and it may have been disturbed when old skirting taken off?

 

[Octopus]

Murdoch is it possible there maybe a Jb or a spur coming off this part of the ring and it may have been disturbed when old skirting taken off?

Thats what I've been wondering about. Looking around the house has undergone many changes over recent years. I'll be quizzing the owner when I go back on Friday am.

That still wouldn't explain why parts of the circuit test OK until a socket is fitted though.

 

[bozobro0]

I had something like this a few years ago,The main switch in the board was a 30mA RCD.The RCD kept going out when a light or two were switched on.I tested the lighting circuits and found no faults.The landlord was in the process of fitting a new kitchen,so i asked him if he had been fixing anything to the walls,and if he had where.Just the kitchen cupboards had been fixed,and no accessories around fixings.I then looked at a gap in the cupboards where the washer was going,running on the wall was a baton to hold the worktop,which had been fixed with screws to the wall,there was a single socket under there,directly in line with a screw.The screw had nicked the insulation of the neutral and gone through the earth.Fault fixed.
I was convinced it was the lighting!!!!!!!!

 

[electromonkey]

I'd put down the tester at this point and go looking for the fault. I'd defo go down the route of taking the legs out of all the sockets where the skirting boards had been fitted and connect legs one by one and re energise the circuit as you go and see what happens.

The IR may have been low in the first place and a new issue may have arisen to trip the rcd. Have you carried out a continuity test?

 

[Octopus]

Continuity test - yes - first thing I did at the board. CPC is a bit high, have also done continuity on the parts of the circuit I've tested so far too.

Have to say I enjoy this type of fault, not sure the client will like the bill - but they have a problem and need it fixed!

 

[electromonkey]

Hmmmmm. It does sound like a fun one. I'd remove all the sockets and add them one by one and energise as I go and see if the RCD stays on. When it pops, at least you know where the fault is. I find this process can be quicker than testing and seeing as you've already done a load of testing..........

 

[Octopus]

The saga continued...

So I went back - first thing I did was to see if the MCB and/or the RCD would hold and they did.

IR for circuit is very low on L-E, so I split the ring and concentrated on the "bad side", after a short while I was back to the same section as before, IR good on 2 cables, reintroduce the the socket and IR drops.

I took the box out of the wall and reconnected the socket, IR again - this time OK, what is different? I hadn't connected the earth lead to the box, reconnected the earth lead - reading dropped. I took the socket off again and had a good look at the ring cables and there was no damage. Strange!

I should note that when the owner lifted the floor boards I could see that the ring had been extended/amended in a seemingly illogical way!

Can't imagine that breeze blocks are very good conductors!

 

[Sparky Ninja]

Can't imagine that breeze blocks are very good conductors!

I have experienced similar before. I had a shower rail which was 230v to earth. It was eventually found to be a repositioned switch line on the other side of the wall and a foot to the right had been crushed 15 or so years ago making the wall at live potential. It was only discovered when the insulated hand rail was replaced for a metal one.

 

[Octopus]

I have experienced similar before. I had a shower rail which was 230v to earth. It was eventually found to be a repositioned switch line on the other side of the wall and a foot to the right had been crushed 15 or so years ago making the wall at live potential. It was only discovered when the insulated hand rail was replaced for a metal one.

So this suggests that a nail has nicked the live, and the box being in contact is allowing the "path" through.....

An interesting find if that is the case

 

[telectrix]

main problem with this type of fault is when you've found it after 6 hours, don't you just think " should have found it sooner" and having to justify the 6 hours labour time to the customer. i had one where the whole house was in 1 ring, split the ring in what i estimated was the approx. mid point, and then went from socket to socket from there along the faulty leg, at all sockets, split again until, finally after 6 sockets up and down, got back to the CU. after wriggling all the circuit tails out of the way, found that a few years before, some kind spark had fitted a new CU and the leg of the ring, being too short, was joined with a choc block virtually in the wall. all corroded and damp. ( old house ).

 

[Electro-tech]

Breeze blocks conduct if damp,do a continuity test around the wall. Test the wall for voltage.

 

[Daveyb100]

Thinking out of the box here...also try nulling the test leads and then try a few readings there..if youre doing lots of measurements on pendants the leads start coming apart and produce all sorts of funny readings as the dogleg between the lead and the plug starts to wear....you never know!

 

[Octopus]

The saga continues..... here are my findings from a brief visit today!

Here is a photo of a reading off my metrel immediately I've done an IR (at 1000v) which shows that there is a "retained" voltage on the cores of the offending "half" of the ring and the IR reading between L & E of 0.96.



It takes about 30 seconds for the voltage to drop back to almost zero - suggesting that something is acting as a "capacitor" on said circuit. The other "half" of the ring drops back to zero immediately after the test.

Both the MCB and RCD will hold!

I've tried to trace the circuit round the house and all I can say is that the ring has been extended in a rather unusual fashion!

Any comments, questions, or thoughts would be useful

 

[electromonkey]

I'm stumped. A re wire of the offending part of the circuit may be more cost effective for the customer?

 

[Octopus]

Anyone else got any bright ideas?