Eratic tripping of RCD

[Biggles..]

Evening folks,

Have been asked to look into a job today where the RCD trips for no apperent reason and im stuck as to why.

Installation is a seperate 80A 30mA RCD supplying an old wylex board with 4 circuits protected by 3871 cbs.

Circuits are downstairs ring 30A, Upstairs ring 30A, Down stairs lights 5A and uP stairs lights 5amp

Sytem is PME with correct main bonding etc.(ze 0.12)

On testing the circuits found the down ring to be O/C end to end on the neutral so in the process of tracking that down(still trips erratically with the circuit isolated) other than that end to end and R1+R2 come back with values of around 0.3 for the rings and 0.6 for the lights.
All IR checks are full scale on my robin so everything good there.
Polarities are correct and zs values are well within the required limits.

Rcd works on manual test and with the tester.

Everthing seems ok all switches etc function correctly but cant work out why the RCD trips so eratically,

Any suggestions(have swapped the RCD for one in van and that one does it as well)

Lee.

 

[securespark]

For some strange reason (maybe because your spelling has only one "t"), I thought it said "erotic"...

 

[Lectrician]

Double and tripple check your meggaring! There must be something somewhere to cause it!

Forget L to N, test L to E and N to E of each circuit.

 

[Biggles..]

Have done all of those tests with what seem like good results, no cross connections no shorts no spurs on spurs and deffinately no ring on ring.

Have used 2 seperate meggerss both are calibrated closest i could get to a reading on anything was 197M between L-N on downstairs ring all the others were full scale.

really got me confused because the install chects out ok few mister meaners no labels, identifcation of conductors and the usuall's but nothing i could see to cause the rcd to be such a pain.

Thought about changing board and fiiting RCBO's to the rings and leaving the lights off to see what effect that has.

 

[jj4091]

Faulty RCD? sorry did not notice you had swapped it, will think again. How often is it doing it? Can you isolate it to lights or ring main?Had a similar problem a few years ago where it only happened when the customer slammed a particular door & the switch next to the door had a poor connection which arced & caused the trip.

 

[Biggles..]

Thanks so far this really has got me stumped, did that much testing yesterday when i got all the results together, i'd completed a PIR by the end of it but no reason as to the RCD tripping.

 

[bernardgreen]

Double and tripple check your meggaring! There must be something somewhere to cause it!

Forget L to N, test L to E and N to E of each circuit.

I assume the tests do not include a measurement of the capacitive effect between live and earth. This capacity will create a live to earth current the magnitude of which will depend on the amount of cable ( live conductor to earth conductor) and method of installation ( live conductor to ajacent metal work ) such as twin and earth in earth metal conduit or ajacent to damp brick work.

This capacitance will not be seen with a DC measurement other than the speed at which the needle moves to the open circuit reading.

Bernard
Sharnbrook

 

[securespark]

My old chesnut of a story involves a board change where I discovered the RCD tripping upon commissioning.

Trouble was, the circuit causing the RCD to fail was on the incomer...

Eventually traced it to an N/E fault on a ring, but how the other circuit got involved is a mystery.

 

[Lectrician]

Had an MCB pop everytime the circuit was turned on, but meggaring it found nothing at all - clean.

Turned out to be a light fitting with a dusk till dawn photocell inside! The fault was actually inside the light fitting, and only appeared when the photocell operated - which it did everytime the MCB was thrown to do it's warm up!

 

[plugwash]

Eventually traced it to an N/E fault on a ring, but how the other circuit got involved is a mystery.

placing a heavy load on the non-rcd side will increase the potential difference between neutral and earth on the RCD side and thus drive more current down your N-E fault.

 

[Biggles..]

During the initial inspecction found a few things whch i corrected

Front room light:brass chandeleer, flex shorted between L&E which i replaced with pendant)
Front room brass light switch short between L&E which i replaced with white
Landing socket which gave good R1+R2 values came up when doing zs as 120v and wird incorrect on the 3 lights on the robin, new front fitted all ok.

Thought these were where the faults lay but still the RCD tripped even having completed the best part of a PIR including removing all sockets for inspection and lights rather than taking a sample with no further faults showing(functionally and wiring wise)

Asked the customer if it happens when she does anything in particular but one time it could be light being turned on the next time the extractor fan coming on so no real help there.

Do you guys have any suggestions thus far other than me fitting a 5 way board with 100A DP and then RCBo's for the sockets and MCB's for the lights.(no cooker supply all gas)

 

[mst]

im not expert,but did you do a ramp test?

if yes what rating was the tripping current?

 

[Biggles..]

Didn't ramp test it as my robins only meaure the tripping time.

Last time i saw a ramp test done was on a TT where the ze was megger high and the rcd failed even on x5 0&180 and the ramp was done to see at what rating it would go.

This RCD does trip(too much and with times of 12 and 9ms resectively).

Correct me if wrong but ramping it would only prove what current it trips at and as have fitted 2 RCD's that both do it wouldn't give any further help compared to one that wont trip.Or am i now too lost in the job to be seeing sense.

Lee

 

[mst]

no,

im sure ya right lee, as i say im learning mate.

 

[bernardgreen]

Maybe a silly question but have you removed all the fuses and tested the live to earth resistance on the feed from the RCD to the Wylex. Damp or other cause of leakage on the RCD side of the fuses or the cable RCD to Wylex.

I found a door bell transformer neutral wired to the wrong neutral bar that caused occasional false trips when something else in the house switched.

Bernard

Sharnbrook