Difficult fault to find?

[HappyHippyDad]

I have come across a fault that I am struggling to find. I cant remember being stumped like this before as I like fault finding and have always found the fault before.

6A MCB is tripping on lights. It sometimes trips the moment it is switched on and sometimes takes up to 30 mins. There is no RCD.

IR readings are poor (o.3Mohms L/N - E) but do not show a short circuit to earth. I have removed all lamps but IR readings for L-N still suggests something is on the circuit which I cannot find (0.05Mohms).

I have tried it on another 6A MCB, it trips, so not a faulty MCB.

It will not be an over current as there is little on the circuit (2 bathroom lights, 1 x kitchen, utility, garage, perhaps 2A absolute max).

My plan is now to break the circuit in half and leave. Customer can tell me if it trips or not. I can carry on like that but it means many trips and they are not close.

I could also try it on the RCD side to see if it trips the RCD (meaning a L-E fault, but would this knowledge help?)

Any suggestions?

 

[Deleted member 9648]

Break the circuit down and find the 0.05 meg. I would not want to find the cause of an OCPD tripping by trial and error, there are likely to be fault currents sufficient to cause a fire. If you can isolate the low reading energise the rest of the circuit and see what doesn't work. Sometimes if everything appears to work I leave the offending part disconnected until the owner notices what doesn't work
Not a criticism by the way, I know how frustrating some faults can be.

 

[SJD]

I once had something like you describe, that turned out to be an outside cable on the circuit with bare corroded ends that were touching. Mostly hardly any current due to the corrosion, but sometimes more current would flow and then rapidly increase and trip the MCB.

 

[KEV 1 N]

Could be water ingress some where......?

 

[buzzlightyear]

ask hammer jack the customer did they put a nail in the wall.

 

[HappyHippyDad]

Break the circuit down and find the 0.05 meg. I would not want to find the cause of an OCPD tripping by trial and error, there are likely to be fault currents sufficient to cause a fire. If you can isolate the low reading energise the rest of the circuit and see what doesn't work. Sometimes if everything appears to work I leave the offending part disconnected until the owner notices what doesn't work
Not a criticism by the way, I know how frustrating some faults can be.

How would finding the 0.05Meg help? I could find out which 2 points it's between but not the item itself. Also, the fault may be at this lost item but it's a bit of a long shot.
A bit embarrassing to say, but it may actually be quicker to remove all fittings (a total of 6 lights) and physically look at them. I cringe when I hear electricians doing this rather than testing, but what happens when the tests are not faulty and cannot lead you!

Could be water ingress some where......?

I thought that too. I thought I had found the fault with a poorly fitted outside light but it wasn't this.

 

[HappyHippyDad]

ask hammer jack the customer did they put a nail in the wall.

I have quizzed the customer quite a bit.
No new work.
No trades in recently.
No DIY work.
They did not initially switch something on for it to trip, it just tripped.

 

[SJD]

Might be interesting to monitor the current on the circuit (with all the lights turned off) when the MCB will hold. Is it zero then gradually creeps up; or increases very rapidly as the MCB trips.

 

[HappyHippyDad]

Might be interesting to monitor the current on the circuit (with all the lights turned off) when the MCB will hold. Is it zero then gradually creeps up; or increases very rapidly as the MCB trips.

Interesting Yes but how does it help?

 

[suffolkspark]

Just split the circuit in half and see which half that low reading is on and so on? Am I missing something

 

[SJD]

Interesting Yes but how does it help?

If there is a substantial current with everything turned off, you can find where is it going (before the MCB trips). If it just rapidly increases & trips, it is probably no help.

 

[FatAlan]

Live - neutral short somewhere with a loose varying amount of contact thus affecting the time it takes the MCB to trip?

 

[HappyHippyDad]

Just split the circuit in half and see which half that low reading is on and so on? Am I missing something

The low reading is 50000 ohms. I believe this is something 'plugged in' not the reason for the tripping MCB, therefore I would be tracking down this item rather than finding the fault.

 

[HappyHippyDad]

Live - neutral short somewhere with a loose varying amount of contact thus affecting the time it takes the MCB to trip?

Yep, that's what I think. Now tell me how to find it without any faulty readings whilst I sup my beer

 

[SparkyAndGeorge]

Have you checked the condition of all the fittings including any transformers and fans connected to that circuit?

 

[HappyHippyDad]

Have you checked the condition of all the fittings including any transformers and fans connected to that circuit?

I may have to do that, but I was hoping to track it down with fault finding/testing rather than undoing everything on the circuit.

 

[suffolkspark]

0.3 meg ohm l+n to e is still pish poor though? May be able to get you in the right direction

 

[HappyHippyDad]

0.3 meg ohm l+n to e is still pish poor though?

Yes it is, but it's still not a cause of the tripping MCB so it will not aid in finding this particular fault. I've suggested finding this fault also to the lady but she is not interested. I don't mean any disrespect to the client, she's fairly old and believes 'if it aint broke don't fix it'.

 

[Lucien Nunes]

The 0.05MΩ is L > N, so a load, possibly a fan timer? Won't like insulation test volts up it.

But you say the L+N > E is 0.3MΩ which is abnormally low and quite possibly at the fault location. It might be variable, a dead short at times and a meg at others. A likely scenario is that a cable is damaged with the conductors almost touching, which actually make contact when something moves, e.g. a heating pipe expands or a floorboard rocks. The arc has left carbonised insulation which is the cause of your low IR.

This is one of the things about IR testing to find a fault. It doesn't necessarily reveal the low resistance that caused / causes the bang, but it can lead you to anything suspicious - e.g. a damp section of wiring or even the aftermath of the fault - which does aid location.

I'd be splitting the circuit up in search of any section below a few MΩ. There's no guarantee that it's related to the tripping but I would say much more likely than not.

 

[Deleted member 9648]

How would finding the 0.05Meg help? I could find out which 2 points it's between but not the item itself. Also, the fault may be at this lost item but it's a bit of a long shot.
.

Finding and eliminating any low reading is essential to isolate the cause. Whether the 0.05 megs is a load or a fault you need to end up with a circuit with an acceptable IR L-N and LN-E.

 

[HappyHippyDad]

The 0.05MΩ is L > N, so a load, possibly a fan timer? Won't like insulation test volts up it.

But you say the L+N > E is 0.3MΩ which is abnormally low and quite possibly at the fault location. It might be variable, a dead short at times and a meg at others. A likely scenario is that a cable is damaged with the conductors almost touching, which actually make contact when something moves, e.g. a heating pipe expands or a floorboard rocks. The arc has left carbonised insulation which is the cause of your low IR.

This is one of the things about IR testing to find a fault. It doesn't necessarily reveal the low resistance that caused / causes the bang, but it can lead you to anything suspicious - e.g. a damp section of wiring or even the aftermath of the fault - which does aid location.

I'd be splitting the circuit up in search of any section below a few MΩ.

Excellent response Mr LN
Apologies to SuffolkSpark who kind of suggested this as well and I poo poo'd it.
Do you really think tracking down the 0.3M ohms to earth will lead to the tripping MCB fault Lucien?
I was going to try terminating the line conductor in one of the MCB's on the RCD side of the board to see if it tripped the RCD. If it does trip the RCD then I know I am looking at a L-E fault and I will track down the 0.3Meg. However if it doesn't trip the RCD then would you agree that the 0.3Meg to earth is a red herring as the fault would then likely be L-N?

 

[HappyHippyDad]

Duplicate

 

[Lucien Nunes]

No. If there is rodent damage to a cable, for example, there might be all manner of simultaneous problems. In fact, the 0.05MΩ might be one of them. The tripping might be flashovers L-N hence won't take out the RCD, but the 0.3MΩ is where the bared length of L is resting against brickwork so won't go low enough to trip the RCD either.

No evidence is irrelevant until the case is closed.

 

[Deleted member 9648]

I sometimes whack 1000v across LN-E.....NOT L-N (!). It is often enough to break down an intermittent fault into a dead short, which is easier to locate.

 

[Squid]

Quote them for a full rewire......

 

[FatAlan]

Yep, that's what I think. Now tell me how to find it without any faulty readings whilst I sup my beer

Yep, supping a beer too. When I come up with something I’ll let you know

 

[Lucien Nunes]

whack 1000v across LN-E.....NOT L-N (!)

This is a good trick. Even if the fault doesn't avalanche, it can at least give a significantly different reading that indicates something unstable. Testing at 250V can give a similar indication by way of a much higher reading.

Here, I have a choice of insulation testers. 250V, 500V, 1kV, 2.5kV, 15kV and Cuthbert. In Cuthbert v. domestic wiring, if there are no faults, Cuthbert will create some for you. 60kV at enough current to carbonise you. He's really for testing HV lines.

 

[HappyHippyDad]

No. If there is rodent damage to a cable, for example, there might be all manner of simultaneous problems. In fact, the 0.05MΩ might be one of them. The tripping might be flashovers L-N hence won't take out the RCD, but the 0.3MΩ is where the bared length of L is resting against brickwork so won't go low enough to trip the RCD either.

No evidence is irrelevant until the case is closed.

I did forget to mention something!
A mouse has been recently seen in the property

 

[Lucien Nunes]

Eek!

 

[rolyberkin]

I have quizzed the customer quite a bit.
No new work.
No trades in recently.
No DIY work.
They did not initially switch something on for it to trip, it just tripped.

You have simply forgotten golden rules 1 & 2

Golden Rule No 1 = Customers Lie
Golden Rule No 2 = Refer to rule number 1

See my recent post in The Arms, everything now is rodent damage!

 

[HappyHippyDad]

I sometimes whack 1000v across LN-E.....NOT L-N (!). It is often enough to break down an intermittent fault into a dead short, which is easier to locate.

This is a good trick. Even if the fault doesn't avalanche, it can at least give a significantly different reading that indicates something unstable. Testing at 250V can give a similar indication by way of a much higher reading.

Here, I have a choice of insulation testers. 250V, 500V, 1kV, 2.5kV, 15kV and Cuthbert. In Cuthbert v. domestic wiring, if there are no faults, Cuthbert will create some for you. 60kV at enough current to carbonise you. He's really for testing HV lines.

Thanks for this.
Would you say that testing at varying voltage gives varying results ...
always, usually, sometimes, rarely? I mean this with regards a fault, not when testing a healthy circuit (as the results will be similar).
It's just useful to know how much credence to give to the results hence the options above.

 

[Toneyz]

I have come across a fault that I am struggling to find. I cant remember being stumped like this before as I like fault finding and have always found the fault before.

6A MCB is tripping on lights. It sometimes trips the moment it is switched on and sometimes takes up to 30 mins. There is no RCD.

IR readings are poor (o.3Mohms L/N - E) but do not show a short circuit to earth. I have removed all lamps but IR readings for L-N still suggests something is on the circuit which I cannot find (0.05Mohms).

I have tried it on another 6A MCB, it trips, so not a faulty MCB.

It will not be an over current as there is little on the circuit (2 bathroom lights, 1 x kitchen, utility, garage, perhaps 2A absolute max).

My plan is now to break the circuit in half and leave. Customer can tell me if it trips or not. I can carry on like that but it means many trips and they are not close.

I could also try it on the RCD side to see if it trips the RCD (meaning a L-E fault, but would this knowledge help?)

Any suggestions?

Find the fault that is giving you 0.05Mohms. halving the circuit and asking the customer to let you know if it trips does not come over very well apologise for being blunt but there it is.

 

[KEV 1 N]

I did forget to mention something!
A mouse has been recently seen in the property

Is the resistance of a mouse around 0.05Mohms by any chance!

 

[HappyHippyDad]

I did forget to mention something!
A mouse has been recently seen in the property

Perhaps I should take my fairly newly acquired air rifle instead of my MFT!
I bought an air rifle to shoot a squirrel in the loft that was giving me sleepless nights.
I had moved my bed downstairs it was so bad.
It would go against my beliefs to just kill it without making use of it so I had read up on how to skin it and cook it.
I practised with the rifle for a couple of weeks so that I would be a good shot and not merely injure the animal.
I heard it one night.
I went up with my head torch and rifle.... into the loft.
I saw it right in the corner!
I took aim......
I continued to take aim.....
I remained continuing to take aim.....
for around 2 minutes....
It ran away!

It left soon after and I moved back into my bedroom .
I realise this story is a bit odd, but it is true.

 

[Andy78]

Perhaps I should take my fairly newly acquired air rifle instead of my MFT!
I bought an air rifle to shoot a squirrel in the loft that was giving me sleepless nights.
I had moved my bed downstairs it was so bad.
It would go against my beliefs to just kill it without making use of it so I had read up on how to skin it and cook it.
I practised with the rifle for a couple of weeks so that I would be a good shot and not merely injure the animal.
I heard it one night.
I went up with my head torch and rifle.... into the loft.
I saw it right in the corner!
I took aim......
I continued to take aim.....
I remained continuing to take aim.....
for around 2 minutes....
It ran away!

It left soon after and I moved back into my bedroom .
I realise this story is a bit odd, but it is true.

You bought an air rifle to shoot an animal even though this was against your beliefs ? Surely a humane trap would have been a smarter buy ?

Good to know that threatening it worked well enough anyway

 

[James]

How about changing to a 50A breaker,
Turn it on and wait to see where the smoke comes from?
P.s. I get smarter on a Sunday after a booze lunch!!

 

[HappyHippyDad]

Find the fault that is giving you 0.05Mohms. halving the circuit and asking the customer to let you know if it trips does not come over very well apologise for being blunt but there it is.

I agree Anthony.
I'm going to go down there tomorrow and find it!
Thanks for reminding me of my moral duty and I shall update you all.

 

[rolyberkin]

Silly question but are you testing this at 250v or 500v?

 

[Gary latto]

Interesting to know the outcome

 

[Toneyz]

A fresh on Monday you will be in a better mindset treat it as building blocks start at the c.u. and work through one part of the circuit at a time.