More Odd Faults: Stumped and not ashamed to admit it!!

[securespark]

Got a faulty 2.5 RF circuit today.

Tripping 32A CB, sometimes instantly, sometimes after 10 mins or so. No RCD protection. Suspect water ingress.

Disconnect all appliances from all known sockets.

Meg the circuit. 420 Ohm LE, 0 Ohms NE.

Suspect something still in circuit.

Find a cable still connected via fused unswitched spur which reads down NE. May be faulty, may just be something on the end of it. Cannot locate other end of cable.... Disconnect.

Still down!

Test ring continuity. Live - 0.86. Neutral 1.54

Split the ring. Meg circuit. One faulty leg, one healthy.

Reconnect healthy leg: All sockets now working.

Not sure if I've struck lucky and just happened to open up the socket with the first (or last) leg going to it OR there are more outlets I can't locate & one has an appliance plugged into it, this being the "faulty" leg.

Here are the other things I am stumped on:

1. Would an NE short on PME without RCD protection trip a CB? Having found faults in the past where N & E are shorted together and the circuit works without tripping, I can't quite understand this.

2. What explains the huge difference in the L & N ring continuity readings? AFAICS, all cabling is 2.5.

 

[Spark123]

1 - a N-E short won't operate an overcurrent protection device as that is in the phase only.

2 - bad connection or faulty wiring.

 

[securespark]

That is what I can't understand!!!

The circuit is tripping the MCB, but the only fault I can find is on one leg of the ring across NE...

Yeah, I've come across lots of bad connections before, but never have they raised the resistance that much.. Still, you never know.

 

[Spark123]

Did you find an outdoor socket laying in a puddle in the garden?

 

[RF Lighting]

JB under the floor?

Water ingress sounds favorite to me too.

 

[securespark]

No, I wish it was that straightforward!

This is work in progress, I have to go back tomorrow, but I'm hoping you can give me some ideas!! [Cheeky Git!]

Just had a brainwave, though.. have not Meg'd LN as I'm not entirely convinced that everything is out of circuit. Customer is sure that I have found all outlets on that circuit & I can't find any more.

Just have to go over the house with a fine-toothed comb.

 

[Spark123]

Do you have any cable locating equipment, such as a CAT and a signal generator? (we use that sort of stuff for big underground cables but don't see why it wouldn't work to trace cables).

 

[securespark]

JB under the floor?

Water ingress sounds favorite to me too.

Solid floor. But you never know, there may be a leak into a JB somewhere in the ceiling void.

If there really is nothing else in circuit, I suppose water ingress could be dragging down the LE reading to 400 ish??

But I would've thought if water ingress was causing a dead NE short, the chances are it will do the same with LE.

I'm thinking more & more that there is something else in circuit that is showing as a 000 reading NE and a lowish reading of 400M LE.

Do you have any cable locating equipment, such as a CAT and a signal generator? (we use that sort of stuff for big underground cables but don't see why it wouldn't work to trace cables).

Unfortunately, no...

 

[RF Lighting]

What about tracking across a socket or something?

 

[securespark]

As I said, all sockets (that I know of) came back on once I'd disconnected the leg that was showing zero meg on NE, but there may well be other sockets I don't know about on that leg.

 

[Spark123]

As you had 0 ohms N-E, could the 420 ohms you were seeing from L-E in reality be 420 ohms L-N?

 

[RF Lighting]

Have you left that led disconnected for the time being to see if the householder notices anything which has suddenly stopped working (Like the heating connected to that hidden spur )

 

[Spark123]

Does the electric shower still work?

 

[securespark]

Numerical Spark, Yes, I suppose it could well be 400 ish L-N. Hold that thought for a second...

Let's suppose that there is something still in circuit somewhere, and that is why NE is down on that leg. ie, it is not a fault. Perhaps it may be a fault, for example, a floodlight letting in water. Let's also suppose that again, because of that something in circuit (or the floodlight letting in water again), the LE reading is not completely clear.

Having said that, it is still an acceptable reading, as is the LN reading 123 is suggesting.

So, having already established an NE fault will not trip a CB, the fault must be across LE or LN.

That IR reading is 400 Meg. So, maybe the fault is intermittent, and not present at the time of testing?

About the shower: yes, it still works!!

Rob: No, I thought of that, but did not have a suitable breaker to feed that single leg (old MEM breaker QEB?), so the circuit is off at the moment.

 

[Spark123]

That IR reading is 400 Meg. So, maybe the fault is intermittent, and not present at the time of testing?

Could be, or could be something hidden behind a relay etc for example a photocell floodlight. Lights etc though should be on a fcu if on a socket circuit - or has there been a bit of dubious diy work gone on?