Starting point for discontinuous earth fault on final ring.

[CDB]

Hi all,

I have a 32A Final Ring Circuit that I have swapped a single socket to a FSU for an outside light. I'm now testing it and finding it has a discontinuous earth - I'm also kicking myself for not checking from the off but we learn. Anyway from the attached image you should be able to see my results arranged in a rough layout of the property. If you were looking for a DE where would you start?

I didn't do anything on the day I was there because I didn't have a long lead kit. I'm now kicking myself because I feel I should have been able to suss it on the day.

Thanks in anticipation of your help.

 

[Midwest]

Not quite sure of your explanation (of the fault), but from your end to ends looks like there is a bad connection on r2 (>500mOmhs)?

Is the dodgy socket s2 in the dining room?

 

[telectrix]

disconnect the cpc's from the CU, then with a wander lead, go from one leg end round the sockets. when you run out of readings, then do same for other leg. the fault will be between the 2 sockets that have o/c readings.

 

[Idpost]

At the board connect live and cpc of one of the legs. Measure R1+R2 at each socket. At some point you will get 'open circuit'. Repeat for the other leg. It may help to to draw a diagram, as you go and note readings. You should now be able to narrow the fault down to a couple of sockets.

 

[SparkyChick]

I had a similar issue doing a CU change. I joined L+E on one leg, went round measured continuity between L+E at all sockets and drew a map recording the results. Then switched legs and did it again.

Where I got a sudden jump in value, checked that item and/or the one before it.

 

[Richard Burns]

Assuming the values given on the plan are Zs readings then having 1.7Ω in the dining room that close to the CU would indicate to me that you are measuring the resistance of the cpc over the length of the ring so the expected disconnection would be from CU to S1 (possibly S2) in the dining room.

Theoretically from the readings at a glance the ring would be laid out like the below diagram.
However just a guess.

 

[CDB]

At the board connect live and cpc of one of the legs. Measure R1+R2 at each socket. At some point you will get 'open circuit'. Repeat for the other leg. It may help to to draw a diagram, as you go and note readings. You should now be able to narrow the fault down to a couple of sockets.

I had to read this twice but what you are suggesting is like the fig8 tests which I've done but on 1 half of the ring only! Good call. Wish I had known that when at the property - I didn't think I could do anything because I didn't have a long lead with me.

 

[CDB]

Assuming the values given on the plan are Zs readings then having 1.7Ω in the dining room that close to the CU would indicate to me that you are measuring the resistance of the cpc over the length of the ring so the expected disconnection would be from CU to S1 (possibly S2) in the dining room.

Theoretically from the readings at a glance the ring would be laid out like the below diagram.
However just a guess.
View attachment 36157

Hi Richard,

That's really helpful. Thank you. They are r1+r2 fig8 results. So, if I understand correctly, the low values I see in the Kitchen are because they only have one short leg from the CU and the 'return' journey via the dining room is open circuit. The LIM at s2 is because its behind a large dresser.

 

[CDB]

disconnect the cpc's from the CU, then with a wander lead, go from one leg end round the sockets. when you run out of readings, then do same for other leg. the fault will be between the 2 sockets that have o/c readings.

Thanks. This is the method I knew but didn't have a long lead for.

 

[CDB]

Thank you guys (and gals) for all the replies. First post on the forum and super impressed with the helpful and supportive responses. Its hard being a lone worker as I'm sure many of us are when you cannot just knock on your colleagues office door for a second opinion. Its nice to know that help is only a post away.

Thanks again!

 

[Pete999]

Hi Richard,

That's really helpful. Thank you. They are r1+r2 fig8 results. So, if I understand correctly, the low values I see in the Kitchen are because they only have one short leg from the CU and the 'return' journey via the dining room is open circuit. The LIM at s2 is because its behind a large dresser.

R1+R2

 

[CDB]

R1+R2

Sorry, you are absolutely right. my mistake.

 

[Pete999]

Sorry, you are absolutely right. my mistake.

I can be a pedantic burger at times no offence meant

 

[CDB]

No. That's fine. My OCD just wishes I could edit it so it was correct. I'm getting use to a new keyboard which is a bit rubbery and smaller than my standard desk one. I'm all fingers and thumbs!

 

[davesparks]

Thanks. This is the method I knew but didn't have a long lead for.

You live conductor makes a perfectly good long lead, no need to go trailing a long wander lead all over the house

 

[telectrix]

You live conductor makes a perfectly good long lead, no need to go trailing a long wander lead all over the house

but then you wouldn't have the fun of tripping up the squeally rug rats.

 

[Octopus]

A wander lead is a must on some occasions..

I've done a Zs at all sockets to try and find the break.... It can work

 

[Dave OCD]

Hi Richard,

That's really helpful. Thank you. They are r1+r2 fig8 results. So, if I understand correctly, the low values I see in the Kitchen are because they only have one short leg from the CU and the 'return' journey via the dining room is open circuit. The LIM at s2 is because its behind a large dresser.

I'd tell her to hurry up getting dressed and shift out the way so you can also check that socket.

 

[tomspark93]

Reminds me of a job a couple of months ago on a house refurb. I narrowed it down to one room and it turned out they had built a stud wall directly in front of an existing wall and boarded straight over the socket where the fault was. Only managed to find it as I started lifting floor boards to trace back the wiring.