Ir Test

[Copperkev0]

Hi all, I've been working on a house and did an IR test on down stairs sockets. Did the test at the db with all the wires disconnected from the board. the results were L-N 196. M ohms N-E 201. M ohms but the L-E was 1.67 M ohms now I know this is just about a pass on the L-E as its over 1 M but needs investigating. I've disconnected everything from the sockets there nothing plugged in anywhere, the kitchen is completely ripped everything is exposed. My question is if i'm unable to find the fault and when I fit the new board i'm I likely to have issues with tripping? I've tested every other circuit and they fine. Thanks

 

[ProDave]

If it's all ripped out, you will be able to test each leg of the ring to find the one with the fault, and replace that leg.

Chances are a pinched cable somewhere or something like that.  I wouldn't want to let it go particularly while it's all exposed. You will never get a better opportunity to find and fix the fault.

 

[kerching]

If you do the calc for the L-E fault current what do you get?

Is it sufficient to trip an RCD?

Having said that....i would NOT pass it.....I have yet to see a fault that got better with age.

In fact there are not many things that improve with age, except

RR

emma Watson

Wine

Cheese

For example, look at Brucie. The clock aint done him any favours

Just saying

 

[Murdoch]

Do you not think that the cpc's should have been connected to the earth bar in the cu?

 

[pewter]

As above.

CPC should be connected

 

[Steve3948]

People keep saying its a fault, but above 1.0 meg ohm is clear and acceptable according to BS7671.

When I started above 0.5 meg ohm was clear of faults, if you turned the handle easy and the needle went half way all was good.

Cold damp day could cause this, splitting legs and it will most likely disappear.

 

[ProDave]

But as already pointed out, 2 Meg is NOT "normal"  good wiring in good condition will not show that.  It's an early indication of something wrong and it's not going to get better. You would have to be a fool in this situation not to investigate while it's all stripped bare and so easy to find and fix the fault.

 

[Steve3948]

No Dave 2 Meg is Not a problem, will it get worse, what proof do you have, you could spend hours trying to find something that's not there.

If instead the meter had OK for over 2 meg and NO for under would you waste time then.

In the bigger scheme of things how different is 2 to 900 in terms of of a fault reading, as far as the meter goes, is 2, 900 and infinity near to each other in terms of a fault reading.

 

[ProDave]

Well flame me for it if you want, but I have given a code 3 on an EICR last week for 2 M hms L-E on one ring final where everything else is way > 100M hms

The prospective buyer of the house who asked for the EICR is now aware there might be a problem and when he comes to ripping out the old kitchen I will be investigating further.

It's not the same house is it?

 

[Marvo]

It may be on the right side of the figure in the regs but the regs are the absolute minimum requirements, there's no reason the after considering all the factors you shouldn't hold an installation to a higher standard than the regs require if you see fit.

Given the circumstances outlined in the OP I'd track the fault down. If it was an electrician working in my house I'd expect him to do the same.

 

[Steve3948]

The reason why 1Meg is used and 2Meg preferred is because this a Clear reading, hence why its given.

If 100Meg was deemed Clear that's what the requirement would be.

I would not code 3 a reading above the required requirement, maybe comment but that's all.

I would not want to pay an Electrician for looking for something thats not wrong to start with.

2, 900, Infinity all the same in terms of a Clear reading, its not a fault reading they are Clear readings.

 

[apprentice87]

Anyone but an idiot would agree that a circuit with an IR of 2 meg had SERIOUSLY deteriorated since it was installed. Why would it be 150 times worse than other adjacent circuits/conductors?? There is obviously a reason for this.. You need to find the cause of this deterioration. To do otherwise would be negligent..

john...

 

[Sidewinder]

Is this a completely new install, or a partial, or an EICR?

Or, a mix of the above.

What is the "global" figure?

I would not be happy with this due to the differences between the readings on the same circuit.

That is indicative of an issue, whilst the "numbers" are better than the gutter level required by BS7671, I don't work in the gutter, so I would be advising in detail the issues, and would be strongly advising that the issue is further investigated and why when the opportunity is available.

Oh, and the cpc's must be connected in the CU.

Re-connect them and then re-test and see what you get then?...

 

[apprentice87]

A fright i suspect!!!

john...

 

[brman]

I find it scary that some people think that a <2M IR is ok with nothing connected to the ring. 

that said, you have two choices.

1) put the new board in and hope you don't get called back next time it rains

2) split the ring and find out wher the fault is.

Personally I would give the customer the option but point out that future call outs are not free and finding the fault isn't free unless I can do it in less than 1/2 hour. But then I always warn customers about the possibility of this sort of problem before I start a board change.

 

[NozSpark]

I had a board change the other day and found a low IR between N&E on the lighting circuit,,, spent an hour or so tracking it down to the outside lights and then isolated them..

Going back in a couple of weeks to sort it

 

[ProDave]

I tend to try and fix simple faults when I can even on on EICR where strictly speaking I am only inspecting.

Last weeks example, one ring final had an open circuit neutral. (I normally do testing at the board before any inspection of accessories)

So I powered up the ring with only one N connected, went round to find which sockets had N and which didn't. then took 2 socket plates off (part of the inspection anyway to sample some of them) and found the loose neutral.

Far better to spend 5 minutes and fix a fault rather than leaving it I feel.