Random RCD trip & interconnected circuits. Suggestions appreciated.

[Marti]

Hi There,
I was called out to a house RCD trip which would not reset. An old style 10 way split board with a single 30 ma RCD (half without RCD protection).
RCD checks out fine but I found a faulty chandelier & replaced it with a single bulb for now. Also a faulty 12v transformer flickering; now replaced. The RCD now resets with occasional trips. Various sign of poor work, driveway lights plugged into a garage socket,Circuits mis-labelled, lack of sleeving etc but nothing that screams disaster. Switching off everything and adding circuits gave no conclusive indication of the issue (still trips). I suspect borrowed or interconnected circuits but I'm not happy I've proven anything definitively. So, my query is, what would a good test routine include to indicate the issue(s)? Can testing really find the cause or is it likely to be buried in the wall's, floors and history of the house and need drastic remedial action? Appreciate your thoughts.

 

[PEG]

Hi,what test routine has been employed already?

If it is a decent size installation,a prescribed,deliberate and thorough testing plan,from the beginning,would be vital.

Random,sporadic and hopeful testing,can yield results,but they will not allow joined up thinking,should further faults occur,and indeed,can make things harder.

Whenever i get called in,to delve into one of these conundrums,it is not unusual to be told of all the places and items which do NOT need looking at or replacing,because they have been attended to already....these all go to the start of my list.

It does not show disrespect or indifference,to a previous tradesdude,to begin and end,in the correct manner. Any person who chirps in with "no need to check that,pal,it's new..." gets put on the pay-no-heed list.

Right,so you are driving across a foreign country,no sat-nav,no language skills,and you only get paid,on arrival....

You get directions in 5 mile stages. Get to the end of a stage,and you are given directions to the next set,and so on...

You would not try to start that journey from where you think is halfway,nor would leaving a stage out,be helpful...and some local,speaking a foreign language and telling you "you don't want to start from here,mate..." is not bringing payday nearer...

That's just how i see it...

 

[Strima]

What testing have you done? IR results?

 

[Marti]

Random,sporadic and hopeful testing,can yield results,but they will not allow joined up thinking,should further faults occur,and indeed,can make things harder.

Exactly the error I think I've made if I'm honest.
RCD tested with the Megger. Introducing each circuit in sequence then 250v IR between each live at the board then each neutral (not 100% I had found every plugged in device). A sweep of sockets and switches. All this gave me the chandelier and the LED but still occasional tripping.

 

[DPG]

Was IR testing done from neutrals to earth? And was CPC connected to earth bar during tests?

 

[Deleted member 9648]

It appears your testing in terms of fault finding is a bit random. Adopt a logical approach. RCD's normally trip because of a low resistance between live conductors and earth.
The first thing to do is carry out a global IR test @250v L-E and N-E. There is no point in testing between L-N because short circuits will not be the cause of an RCD other than an RCBO tripping. Do not disconnect or unplug anything, you want all possible sources of trouble in place. Bear in mind that N-E faults on any part of the system, even non RCD protected circuits can cause an RCD to trip because a N-E fault creates a parallel path for normal neutral current to divert down.
This test will almost certainly produce a low reading, so you now know for sure what you are dealing with, from which you can then start breaking down to individual circuits to isolate the low reading culprit.
I fault find every week using this approach and it's extremely rare that further testers such as earth leakage clamps and ramp tests etc are needed, but it often seems that IR testing is a last resort for many.

 

[Wilko]

If you're going back, perhaps write a test plan and record results next to each step. My top "to be ignored guess" is the driveway lights. Good luck !

 

[Bob Geldoff1234]

Rcd's tripping are usually caused by damp or slugs!
Outdoor sockets and lights being the obvious,ponds and ovens being the next.Damp in underfloor joint boxes is quite common.And finally damp behind sockets.
But as above you won't find anything without IR testing first.

 

[Marti]

Solid gold advice; thank you. I'm going to make up a simple chart to give method and structure to testing for the next time. I'll let you know the result.

 

[Lucien Nunes]

N-E faults on any part of the system, even non RCD protected circuits can cause an RCD to trip because a N-E fault creates a parallel path for normal neutral current to divert down.

I'm not sure I'm with you here. An N-E fault on an RCD-protected circuit can cause the RCD to trip when a load is applied to a circuit that's not on the RCD, or on another RCD. The load current (wherever it is drawn from) divides between the supply N and supply E via the fault, which passes through the RCD as an unbalanced current. But if there's no fault on the RCD-protected circuit, that can't happen. What situation are you thinking of?

 

[Deleted member 9648]

I have had instances on split '16th' boards where a N-E fault on the non RCD side has been the cause of nuisance tripping of the RCD side. I have put it down to the fact that neutral load currents can divert, however if that is not the case I stand corrected. Regardless though I can recall a few cases where correcting a fault on the non RCD side has stopped regular nuisance tripping.