Its possible that this is the only fault - that could leave a large section of the earth wiring in the circuit floating, and hence it probably would tend to acquire a mains potential due to proximity to other live wires. If this were the case you ought not be able to get a dangerous shock from it.
If you turn the light switch on (such that both wires in it are now live) and measure the voltage between one of them and the earth, what reading do you get then? Only if it is 0V (or close) would that suggest a physical low resistance fault between live and earth).
The other test to do would be to disconnect the circuit at the consumer unit, and measure the resistance between live and earth of the cable feeding the circuit. You should see an open circuit.
You haven't mis remembered and the nominal voltage in the UK is now 230v. But they played with the tolerances such that nothing had to actually change. IIRC it was 240v +/-6% (225 - 254v) it's now 230v +10% -6% (216 -
253v), There was talk of altering the tolerance to +/-10% but that seems to have been forgotten, just as well 207 to 253v is one helluva range for kit to deal with.
What no-one's yet mentioned is that if the earth fault isn't local to the switch only, reconnecting an earth wire could result in lots of other stuff in the house also becoming live. And if the earth fault is unfortunately situationed, you or your family may be frying tonight in the shower.
Really your last sentence is the best advice you can take. Despite your protestations of knowledge, some of the things you've said are plain scary, and some of the things proposed just dangerous.
Still here, alive and kicking! Nice to know that you are thinking of me!
After a bit more testing and much thought on my part, the aforementioned leccy turned up on Monday afternoon. He repeated many of the tests I had done with my AVO, but with more modern electronic/ digital test kit. After some more thought, the conclusion was that there is a probable live to earth fault in the lighting circuit (not sure where yet) and that there is a probable break in the earth line somewhere between the fault and the CU or even that it is not connected at all, which is why the fuse has not blown.
We then took the cover off the CU, and guess what ............ although we could count 8 live (red) wires and 8 neg (black) wires - cooker, 2 x power rings, 2 x lights and immersion heater - there were only 7 earth (yel/green) wires visible! . This probably goes back to the time when we had the kitchen re-fitted about 15 years ago, and the guy who did it had to move the CU about 1/2 - 1 mtr along the wall behind the cupboards. IIRC, he was a joiner by trade and did it himself - who said Part P was rubbish?
You will have guessed that the CU is fairly old (30 yrs), and has wired fuses. So, the first job is to replace the CU with a new mcb CU and at the same time see if the missing earth wire can be traced and connected properly. Estimate for this part of the job is =A3210 which sounds reasonable I suppose. After this we can set about tracing the fault by isolating each bit of the lighting circuit in turn. He can't give me a price for this obviously as we don't know how long it will take, could find it straight away or sod's law says it will be in the last bit that is tested.
He's coming round later this week to do it, but meanwhile he has replaced the stainless steel switch with a plastic one.
Just a couple of Part P type questions -
His business card says "JIB Approved Electrician". Is this OK in Part P terms?
After he has finished, what documentation if any should I expect? Test Cert perhaps?
It almost certainly indicates his competence to a good standard, but I don't think it implies directly that he's set up for Part P self notifications.
Ask if he's registered to self certify for Part P and with whom. Assuming you care!?
You should expect (and ask for in advance) a EIC (Electrical Installation Certificate) and some sort of evidence of Part P notification (never seen one, can't tell you what this looks like).
An earth that's has gone live - it's like discovering that someone you were trusting to protect you wants to kill you. No-one wants to go like Mrs Ghandi did.
Glad to hear you found out what the problems are the easy way, not the awful one.
Thanks for this advice Tim, and just to be on the safe side I have spoken to him again. He was a bit evasive, saying that because the new CU is a "replacement ", then Part P doesn't apply. After some pushing on my part he admitted that he is not qualified to self cert but that if I wanted a test cert he could get a "qualified tester" to do a "whole house" test. It would cost about =A3150 and he warned that it may show up some new faults which would need sorting out as well.
That is incorrect as he will be altering the circuit protection characteristics due to using differing overcurrent protection devices to the originals, not to mention the RCD requirements. This is why it should be written up on an EIC and not a MWC (Minor Works Certificate). I'm fairly sure (without checking) that like-for-like replacements explicity does not feature CUs in Part P.
The EIC would need to have every moved circuit tested for presence of earth (CPC) at all points, polarity at all points, worst case L-E resistance and a 500V insulation resistance test on all parts as practical, same as for a PIR which is the bit that he's saying would be 150. 150 is an expected fee for a full house PIR.
This is the requirement for doing it by the book.
Of course, most people would just put in a CU and not bother, with a comparatively small risk. But a good way to look at it is: doing it by the book gives you effectively a PIR, or a health check on your entire system, so it's not a total waste of money. This especially true given your original posting.
But if your mate changed the CU, assuming no dates are printed on the devices or case, it's doubtful anyone would notice, other than an obvious compliance with the 17th (but you could have had dual RCD or all RCBO boards ages ago).
That's correct. Sometimes things like bad L-E resistances can be worked around with a slightly smaller than normal breaker overcurrent rating, but most faults would need to be searched out and fixed, including any lack of earths on lighting, which was standard practise in the old days years ago. Those fixes are of course chargeable.
If you feel happy about the man's abilities, but accept that he can't self-certify (and doesn't own a tester by the sound of it) - then you could just make it a notifable job to the council. If you're lucky, you'll pay no more than 150 for a BNA and the council will send an inspector to assess it.
Then again, many councils are totally inept at handling Part P - but it would only take a phone call to your Building Control department at the Borough/District council to find out. Or start with their website - sometimes policy regarding Part P is there.
The wiring could date from before an ECC was required in lighting circuits. If so, it might well take more work to sort than just coupling up at the CU. It may not be connected anywhere in the circuit where the cable goes through any form of JB, etc. And that's assuming the cable itself is actually TW&E everywhere - two core was once the norm.
As was said earlier, an ECC not connected will often show full mains voltage due to capacitive coupling. The electrician should have checked if it was this or a real short - it's easy enough to do.
I would have run an earth wire and connected it by now.. if there is a short it will soon show. You may even see the flash so you know where the fault is. ;-)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.