Downstairs lighting circuit MCB randomly tripping since electrician's visit last week.

[ban-all-sheds]

Maybe, but I don't think that adds anything useful to this discussion.

It ought to stop you saying that it is surely quite impossible that a correctly wired non-faulty RCD could fail to operate in the presence of a L-earth short on a circuit it was protecting.

ISTR that that principle is very important to you.

I have a (large) lawn to mow!

It's alright for people with gardens on chalk, isn't it.

 

[ban-all-sheds]

Perhaps you're not expressing yourself clearly enough, but that statement appears to make no sense.

It appears to indicate that he doesn't understand how RCDs work....

 

[JohnW2]

It's alright for people with gardens on chalk, isn't it.

Dunno, I've never had a garden on anything other than clay

In any event, poor light has definitely stopped play for today!

Kind Regards, John

 

[JohnW2]

I just knew you would say its not an RCD

... but, as I said, there's no guarantee that those who used the term used it correctly.

Is that was Residual current refers to, the imbalance, i never realised that

Yep, it's the 'residue' ('remainder') after one subtracts the N current from the L current (or the vector sum of all the L currents, in the case of a 3-phase one), or vice versa.

Kind Regards, John

 

[ceecjc1]

Thanks for everyone's further contributions! I think we're all in agreement that there is a L/N or L/E wiring fault of some description, which needs looked into before we can turn that circuit on again.

I should have mentioned in my last post but I didn't because a fault is a fault and I didn't want to confuse the thread until a solution was found, but we discovered this week that the RCD no longer works. In April when the lights were originally installed, the RCD (30mA covering whole house) was tested and functioned correctly. Last week when the electrician visited he checked the RCD as a tick box exercise at the end of his visit and could not get it to trip either by the usual testing method of tripping a socket or by physically pressing the test button on the RCD itself.

Given the fault condition this means there could realistically be a live/earth fault somewhere, all the more reason to keep the circuit switched off until properly tested. In the guy's defence, remember that when he left he thought he had fixed the original issue of a tripping MCB (and arguably he has), so he didn't recommend any action further than replacing the RCD. It seems obvious though that with an unexpected high current situation and a failed RCD, that he should have done further investigation.

I have a different electrician coming tomorrow, wish me luck. I am also getting quotes for the RCD to be fixed as well as potentially replacing the CU with a more up-to-date RCBO version, but I will need this fault fixed before I do the CU.

 

[ban-all-sheds]

Dunno, I've never had a garden on anything other than clay

I feel your pain.

Sorry - I thought the Cretaceous chalk geology extended further north than it does.

Mind you - your pain must be less than mine - I could grow rice in my back garden in the rainy season, and no way could I contemplate mowing my lawn. My neighbour did his the other day - I think it might have been to create a commemorative reconstruction of Passchendaele...

 

[JohnW2]

I feel your pain. Sorry - I thought the Cretaceous chalk geology extended further north than it does.

I'm very much in 'the north' - right in the 'top left-hand corner' of Bucks, a few hundred yards from Oxfordshire, and probably only about 6 or so miles from Northamptonshire.

Mind you - your pain must be less than mine - I could grow rice in my back garden in the rainy season, and no way could I contemplate mowing my lawn. My neighbour did his the other day - I think it might have been to create a commemorative reconstruction of Passchendaele...

Agreed, mine is not as bad as that, but one does not have to dig down very far to hit the water table, even during dry periods. My cellar is certainly usually well below the table, hence the need for an underfloor drainage system and a pump - which, again, stays fairly busy even during 'dry spells'!

Kind Regards, John

 

[ceecjc1]

RESOLVED (properly this time)

The electrician did a proper insulation check and discovered there was a live/earth fault for which my faulty RCD failed to trip. He disconnected everything from the circuit one by one and discovered in one light switch a nicked wire which was making contact with the metal back-box. This is the same light switch which the previous electrician had tightened up last week, neatly explaining the timing of this new problem. Once the wire was fixed the insulation test passed and the mysterious load has now disappeared! When all lights are off no power is drawn and with all lights on they draw roughly 450W which by adding up all the lights on the circuit is about right.

Very happy now, thanks to everyone for their suggestions and contributions.

Chris

 

[JohnW2]

RESOLVED (properly this time)

I'm pleased to hear that it is all sorted. I presume that you have added the name of your previous electrician to your 'little black book'!

Kind Regards, John

 

[ceecjc1]

I presume that you have added the name of your previous electrician to your 'little black book'!

Indeed! Need to take a look on check-a-trade for some more quotes for my RCD.

 

[John D v2.0]

RESOLVED (properly this time)

I hate to say it, but for a live earth fault dead short the MCB should trip within 0.4s if the earth loop impedance is within spec.
Did the new electrician confirm the loop impedance at the switch?
If not you still have a fault in your lighting circuit. If so the MCB should be suspected faulty too (maybe damage through overheating) and replaced.
While the fault was in occurrence, your terminal screws on the light switch will have been at a dangerous voltage.

 

[JohnW2]

I hate to say it, but for a live earth fault dead short the MCB should trip within 0.4s if the earth loop impedance is within spec. Did the new electrician confirm the loop impedance at the switch?

I'm not going to trawl back through the whole thread, but I don't think anything has been said that would preclude this being a TT installation. Indeed, what the OP said about the RCD ...

There is only a single RCD covering the entire house and it hasn't tripped once.

... could be consistent with TT. If that were the case then what the OP experienced (given that the RCD is apparently non-functioning) would all make reasonable sense, wouldn't it?

Kind Regards, John

 

[JohnW2]

Indeed! Need to take a look on check-a-trade for some more quotes for my RCD.

The electrician who diagnosed and cured the primary problem would seem to be a good starting point. One issue you may have is that "just" replacing an RCD is such a simple and quick job that 'overheads' (bothering to spend time travelling etc.) may dominate the price, which would be negligible if charged on the basis of 'hours' (probably minutes) of chargeable work - in other words, you could be in the territory of a 'minimum charge' (to make it worthwhile for the electrician).

Kind Regards, John

 

[EFLImpudence]

There's no way that such a fault could be over 2kW.

So, was the meter flashing rate a red herring - or some kind of warning?

 

[garyo]

Any sign of melting/blackening in the switch back box?