not aborrowed earth - but a fault to earth!!!!

[DirtyDog368]

Hello All,

After posting about a CU change and the one light circuit that kept on tripping the RCD on the new CU i had a lot of great advice. Armed with my Megger 1553 i approached the CU and condcuted some IR and continuity tests.

At first a borrowed earth was suspected but from the following test results i think maybe not. I think there is a small IR fault and defiantely a fault to earth on the circuit 2. Seen by the resutls below. Can some one confirm that it is possible to have a small enough earth fault that will not trip a normal over current protective device but big enough to trip the more sensitive RCD protective devices.

If indeed i do have an earth fault, how can i go about finding it without ripping up half the house to find it?

cheers
DD368

upstair lights = 1
downstair lights = 2
IR test	P/N	P/E	N/E
1	99.9	99.9	99.9	(MΩ)
2	0	0	0	(MΩ)
nb switches off, 250v used with loads & Dimmers in
Continuity
	P1	N1	CPC1	P2	N2	CPC2
P1	_	open	open	open	open	open
N1	open	_	open	open	open	open
CPC1	open	open	_	open	open	open
P2	open	open	open	_	69.3Ω	69.3Ω
N2	open	open	open	68.4	_	0.27Ω
CPC2	open	open	open	68.4	0.27Ω	_
	continuity is evident between live conductors on circuit 2, possibly due to switches in on position
	however there should not be any continuity between live conductors and CPC.

 

[ezzzekiel]

loads and dimmers in circuit any reason why - join l/n to earth for quickness,
a borrowed earth shouldnt cause any problems
circuit 2 showing either a fault across all conductors or loads connected, have you checked borrowed neutral on landing.

otherwise identify circuit with fault and all points on circuit then split in half and ir test untill fault clears identifying the part with fault.

should you of changed the cu without being able to locate and repair faults?

 

[telectrix]

first remove all lamps/loads. then you can determine if it's a L/E or a N/E fault. then, as ezzzy says, split the circuit at approx. mid point and test each half of the circuit, then split the faulty section again. eventually, you will have narrowed it down to a fault either in a fitting, or on a single section of cable. however, a quick look for a nail through a switch cable might shorten the job.

 

[ezzzekiel]

also its not the job of the mcb to interrupt earth faults thats why we use an rcd i assume you are either in training or newly qualifed and should really have someone with you to advise correctly.

 

[oldtimer]

A borrowed earth does not do anything but as telectrix said a borrowed neutral best thing to do is switch all the lights off then reset the both RCDs now then switch on each light one at a time and check to see what light trips it. Youi may find that it is a light that sits between the 2 circuits. ie hall if a flat or stair landing if a house for a quick fix move both light circuits to either the same RCD row or CB.

Also after 36 years never came across a borrowed neutral until 3 months ago funny old world

 

[ezzzekiel]

fed up to back teeth of them at moment, every house on estate im working on.

remove hall switch and see if there is a link between commons of the two gang switch this is usually a good indication of borrowed neutral.

 

[WDMDL]

Not a borrowed neutral, but a resistance 0f 0.27 Ohm between N2 and CPC2 points to a direct short between N and cpc at some point in the circuit.
With a reading so low, for 1.0T&E I'd look at the first ceiling rose / luminaire for starters.

You could then check for N-E continuity at that and other accessible points until you found the culprit.
Most likely a Neutral trapped in a metal fitting.

Simon.

 

[telectrix]

or even a light connected L/E instead of L/N. that would only cause tripping when the light was switched on.

 

[DirtyDog368]

ezzzekiel, fancy popping round tomorrow then for a cuppa and a chin wag ;-) north london area N11?

 

[ezzzekiel]

unfortunaely im busy all day, but im sure youll find it eventually