Sparkies - Ring fault?

[A.Lee]

I've done a bit of testing this afternoon, and found a fault on a ring main. Continuity results: E-E 0.40 ohms N-N 0.19 ohms L-L 37 ohms - clearly a fault here.

I checked the 4 sockets on the ring for loose connections, and found no apparent fault. So next I joined up opposite ends of L+E, and measured R1 + R2 at the sockets: 0.42 ohms, 0.42, 0.48, and 0.51ohms.

Clearly the resistance goes up as the sockets get further away.

Muddled thinking got me nowhere. After a while, I was thinking that the cable must be damaged between the final socket and the return cable. But, a resistance of 37 ohms? That doesnt seem to be enough to be a damaged cable. (Insulation resistance was checked, and all fine at >1k M ohms)

Any thoughts? Ta Alan.

[Skipweasel]

How would a socket-switch affect the ring continuity?

[Harry Bloomfield]

A.Lee explained on 14/01/2011 :

Connect to one end of the ring live with the ring parted, then work your way along with a long wire with a 13amp plug fitted temporarily on the end - check each socket in turn, until you find a sudden variation in impedance. That should make localising it easier.

[John Rumm]

You may have made it harder to spot by joining both ends at the same time...

If you disconnect both ends at the CU, and then join L to E on just

*one* end only, then measure resistance between L & E at each socket. Then move the join between L & E to the other end of the ring, and then repeat the measurements at the sockets. You should be able to see where the problem is.

Obviously on a working ring you would expect a set of matching rising and falling resistances as you go round the sockets depending on which end has the join. In your case you would expect normal results with the normal results up to a point, and then a big jump in resistance as you pass the fault. Once you have passed the fault in both directions, you should know exactly which cable has the problem.

[ARWadsworth]

You have actually answered your own question.

Split the ring at all 4 sockets and use a fly lead of known resistance to check both ends of r1 from the CU before adding sockets as you work to confirm the topology of the circuit.

It is a damaged cable or an unknown juction box or socket that is at fault.

[Richard Russell]

Are you sure you disconnected *every* appliance on the ring? I once had a very similar fault, and like you was puzzled at the measured loop resistance of a few tens of ohms, but it went up to the expected 'infinity' when everything was properly disconnected. A wire had come astray behind a socket outlet.

Richard.

[ARWadsworth]

The clue is in the words "ring continuity":-)

You don't test that from the front of the socket!

[Skipweasel]

I'm glad someone's awake.

[A.Lee]

Tried replying by email, but your address is dead:

Final-Recipient: rfc822; [e-mail] Action: failed Status: 5.0.0 (permanent failure) Remote-MTA: dns; [209.85.227.27] Diagnostic-Code: smtp; 5.1.0 - Unknown address error 550-'5.2.1 The email account that you tried to reach is disabled. 34si4611472wba.45' (delivery attempts: 0)

Yes, I was trying to say it was like a radial where the resistance got higher as I went along, but it did not go down as I progressed back toward the CU, so the theory of the break between the last socket and the CU holds some weight.

2 of the sockets have faulty outlets, so I'll change them on Tuesday when I go back, and I'll do some further testing to pin down where the fault is. Thanks Alan.

[ARWadsworth]

It certainly does, either that or an odd topology,

Well it will obviously be the last cable you test!