Insulation resistance fault on two way switch.

[Cornubian]

Evening all. Thought I would finally seek advice of those who know better! This is my first post so be gentle with me!

Just wired a new 3 bed house. All great until I test for ins. res. on final lighting circuit. Result came in at less than 1Mohm- [ L-E = 0.39, LN-E = same ] did all the usual checks and soon got it down to the 3 and earth for the two way switches on stairs.

Now, if I isolate the cable in question, rest of the light circuit passes ins. test as expected.

If I link both switches with new cable [ just dangled on the floor ] test passes so switches are fine.

Fine, just to prove it is the cable in question I ins. test just the installed cable and it passes! I am now left wondering why this is? Am I missing something totally obvious here???

It also passed cont test as well.

Major frustration- as with all fault finding when on quote work although if I find it's a plasterboard screw then its happy days!

Thanks in advance for any help!

 

[derek]

Have you IR tested between all the cores of the cable to the cpc and extraneous conductive parts and each other? If it's a screw then it probably will show up on IR test

 

[Deleted member 9648]

Are you certain you havent somehow linked the 2 wayed light to another circuit?....a mistake in the switch resulting in two separate circuits linked?....sounds possibly like you are picking up a reading from a connected neutral through lamps etc.

 

[Octopus]

Given that you said that the 3 core cable is the faulty one and the IR for the cable is OK, and that you've checked continuity of the 4 cores is OK, when its not connected to ANYTHING, I must admit I'm puzzled!

 

[Cornubian]

thanks for the speedy replies lads!

Derek- yes i tested all possible combinations with the cores. Although my testing to ecp's ended at the metal back boxes in the two switches.

Wirepuller- Good idea- I hadn't thought of that although that would have given me the same reading on the other lighting circuit that passed?

Murdoch- yea thats what i mean- I have exhausted my knowledge and logic and am just stumped! Like Derek said- if it was as simple as a screw it would have shown up in the tests. Never had such difficulty before!

 

[Octopus]

thanks for the speedy replies lads!

Derek- yes i tested all possible combinations with the cores. Although my testing to ecp's ended at the metal back boxes in the two switches.

Wirepuller- Good idea- I hadn't thought of that although that would have given me the same reading on the other lighting circuit that passed?

Murdoch- yea thats what i mean- I have exhausted my knowledge and logic and am just stumped! Like Derek said- if it was as simple as a screw it would have shown up in the tests. Never had such difficulty before!

I'd retest with the CPC's disconnected from the back boxes. If there is damp in the building this could explain the "leakage". Seen it once or twice. Got to be worth a try.

 

[itselectric]

Have you tried having all cables connected but not screwing the faces back.
It could have been a cable squashed a bit behind face plate.

 

[Sintra]

Out of interest what were the L-N & N-E IR readings?

 

[Cornubian]

I'd retest with the CPC's disconnected from the back boxes. If there is damp in the building this could explain the "leakage". Seen it once or twice. Got to be worth a try.

Cant honestly remember if i tried this- i'll have another look in the morning! It's a timber frame house with a mansions worth of insulation although I do know the plaster in the walls is still damp from the insulation res of the other circuits!


May be of use but just to clarify the circuit is set up as follows- upstairs lights feeds two bedrooms and one cupboard, and stairway lighting. Stairs consist of two 240v down lights [feed cable is fine in tests] two way switch at top of stairs in metal back box screwed to timber. Two way cable clipped behind walls as usual down to 2 gang two way switch- plastic dry line box, one gang for stairs, one gang for hallway on separate circuit.

Cheers Murdoch

 

[Cornubian]

Out of interest what were the L-N & L-E IR readings?

Haha you are going to shoot me now- for some insane reason I didn't take that reading! So fixated on a fault L-E and isolating where the fault lay I missed the wood for the trees!

 

[Sintra]

:41: Done

 

[Cornubian]

L-n passed test- checked all the above with no luck. Am reluctant to rip the ceiling down if the cable itself shows no fault when isolated??

 

[Octopus]

L-n passed test- checked all the above with no luck. Am reluctant to rip the ceiling down if the cable itself shows no fault when isolated??

If the 3 core tests OK you need to add the switches back in and test again. Could be a faulty switch but only you can tell that.

 

[Cornubian]

If the 3 core tests OK you need to add the switches back in and test again. Could be a faulty switch but only you can tell that.

Thanks Murdoch- thats what i assumed so just connected the two switches with new 3 core running down the stairs and the test results were fine so i dont think its the switches. Im going to replace the cable over the weekend and see.

 

[PEG]

Hi fella and welcome. Did you say you have tested the disconnected cable conductors to the MET? Its just that i had a very similar situation with an extension wired by a spark mate and he had the four core tested good for continuity and IR,but an IR between one of the live cores and the MET showed up a fail. Cable had been fed through Rota-broach hole in UB steel and picked up a nick. He had his head in his hands for that one but like my other pals dad used to say,two heads are better than one,even if they're sheeps heads...

 

[Cornubian]

Hi all just thought I'd clear this thread up- after all that fuss and trying all the options above- I really didn't want to rip the wall down again and so just changed the switches [customer had supplied his house with Click mode modules]. Tested again and the tests were perfect! Sometimes it's easy to overlook the most simple solution!

Thanks for your help lads.