Electrical fault finding

I have a problem with an underground mains cable that goes from my house to my garage, and was wondering if anyone could assist in working out the fault please. The setup is as follows:

My house has a standard consumer unit, with a 30mA RCD covering all circuits, and individual MCBs. The circuit to the garage is a 2.5mm twin & earth cable going outside under the garden and into the garage. The end coming into the garage is enclosed in a metal pipe, I'm not sure if this extends all the way along though.

Recently, the house RCD has started tripping, and I traced this to the garage circuit. Even with the wire disconnected completely at the garage end, the RCD trips. I tested the cable with it disconnected at both ends and found a 120K resistance between live & neutral, and a

0.4V voltage between earth and neutral and the same between earth & live. Does this sound normal?

I can't decide if the cable is faulty, or if it's just the RCD which has become too sensitive (it has been OK for the last 4 years). Would the problem be fixed if I disconnected the earth at the garage end and used a local earth? Or should I try usung a higher rated (100mA) RCD at the house end?

Any suggestions most welcome, many thanks in advance for any help!

Tim

Reply to
Tim Hayton
Loading thread data ...

In article , Tim Hayton writes

Never heard of an RCD increasing its tripping level!.

Disconnect all conductors in the cable, and measure from each one to the other to see if any leakage exists from any to any, measure from each conductor to earth and see if any leaks are their either. If none or greater than say 20 meg ohms, and really this should be totally open circuit, then I don't see your problem being the cable.

Course make sure that there are no earth-neutral shorts anywhere as these can cause some interesting nuisance tripping.

If your 120 K is true, then something is "very" leaky somewhere:(....

Reply to
tony sayer

In article , Tim Hayton writes

Never heard of an RCD decreasing its tripping level!.

Disconnect all conductors in the cable, and measure from each one to the other to see if any leakage exists from any to any, measure from each conductor to earth and see if any leaks are their either. If none or greater than say 20 meg ohms, and really this should be totally open circuit, then I don't see your problem being the cable.

Course make sure that there are no earth-neutral shorts anywhere as these can cause some interesting nuisance tripping.

If your 120 K is true, then something is "very" leaky somewhere:(....

Reply to
tony sayer

Nope! Sounds like that is the likely culprit. With the the cable disconnected at both ends you should have an open circuit between each of the conductors.

The 0.4V is in itself not conclusive (especially if you were measuring with a digital multimeter).

Chances are the cable is mechanically damaged somewhere along its length thus alowing the conductors to come into contact with the damp ground. Which will be enough to trip the RCD from time to time.

(I take it this cable was not a Steel Wire Armoured (SWA) one?)

The best solution sounds like replacement of the cable with SWA. Replacing the RCD with a higher trip threshold is a *bad* idea since all you are doing is masking a symptom and in the process reducing the protection that the RCD will give you should an accident occur.

Disconnecting the earth at the garage end and using a local earth rod _may_ be a good idea. It will depend on how far the garrage is from the house. If it is close (say less than 5m) then it is probably not worth doing. Note also that if you switch to a local earth for the garrage, the garrage will need its own RCD since an earth rod will not be able to offer the low earth fault loop resistance you get with the earth supplied by your electricity supplier. In these situations an RCD becomes essential since you will no longer be able to rely on the MCBs working for "live to earth" type faults.

Reply to
John Rumm

I'm not suggesting that this is the case here, or that it should ever be considered until all other possibilities have been eliminated but it has happened for me.

We had a MK 30 mA unit covering the whole system. This ran for many years without any trouble before starting to give nuisance trips which increased in frequency as more years went by.

The RCD was in it's own enclosure between the meter and the CU, I started to suspect a faulty RCD when it took several attempts to turn it back on even with the CU main switch turned off.

All circuits had well nigh infinite resistance with appliances disconnected. There was a small measurable earth current of about 10mA with all appliances plugged in and switched off, probably from built in RF suppressors (computers, dishwasher, washing m/c etc). When it got to the point that I couldn't run 2 computers at the same time I replaced the 20+ year old RCD with a new one and everything's been fine since then.

[Snip]

Yes, I'd be very worried about that, especially if it was spotted with a simple multimeter, it'd probably show up even worse with a high voltage tester.

Reply to
Mike Clarke

It would be a good idea to check this. Even if it is a steel pipe it could have rusted out or come apart at a join.

No. 240 volts into 120 kilohms gives you a leakage current of 2mA. However the resistance could vary depending on how damp the ground is and this might well trip the RCD. Have you checked the resistance between the cable and real earth?

As for the 0.4v is this with the cable connected? Is that an AC or DC voltage? You can have electrical fields in the ground due to leakage, especially if you happen to be near a DC railway line. It's not uncommon for a voltage to be detected between different earth points. The neutral wire is grounded at the substation.

I'd go along with JR and suspect that the cable has been physically damaged. I'd recommend getting rid of the conduit and using armoured cable. You could do a test by temporarily connecting another cable in place of the existing one, just laid on the surface. That should allow you to use the garage electrics without tripping the RCD. Changing the RCD will only mask the symptoms.

Reply to
Richard Porter

My guess on that one was electolyticaly generated emf from the point where the damp was getting to the cores. To the OP. As previously stated, if you can read 120K on a multimeter between cores there's damp getting into it. Time to get out the spade :-(

Jim.

Reply to
jim.

Testers are made for this possibility. Using mine I have found RCDs that trip at 1/2 current ie 30mA trips at 15mA but not often. Only way to test the cable is with a insulation tester 500v or more a 9v multimeter is not good enough unless the cable is dead shorted.

Reply to
Peter

Yes but that would trip the house supplier RCD anyway???....

Reply to
tony sayer

Yup, good point! I was forgetting he has whole house RCD rather than a split load CU...

Reply to
John Rumm

Don't even think about this. It is definitely contrary to wiring regs. Everything must be earthed to the suppliers earth.

luggsie

Reply to
luggsie

Did you by any chance have your fingers holding the meter probes on the wires? 120K is about right for slightly damp hands.

Reply to
Peter Parry

"luggsie" wrote | > Would the problem be fixed if I disconnected the earth at | > the garage end and used a local earth? | Don't even think about this. It is definitely contrary to | wiring regs.

It is not. In some cases, such as IIRC PME installations and caravans, it is prohibited to export the house earth and a local TT system is mandated. In those cases, the cable armour is earthed at the supply end but an insulated gland is used at the load end.

| Everything must be earthed to the suppliers earth.

Wrong (the suppliers are not bound to provide an earth.)

Owain

Reply to
Owain

Eh?

Reply to
Dave Plowman (News)

If so, he obviously wasn't using an insulation tester! (well, not for long...)

Reply to
Bob Eager

Gordon Bennett!, I'll have to rewire the whole caravan park now.

john

Reply to
luggsie

Completely untrue.

Christian.

Reply to
Christian McArdle

I got my digital insulation tester very cheap for this reason. The guy selling it didn't have much of a clue what it was. Wasn't sure if it was an electrical tester or a thermometer. In any case, after getting 500V through his fingers when fiddling, he was happy to pass it on!

Christian.

Reply to
Christian McArdle

Neighbour had a similar problem in my last house, the garage at end of garden RCD kept tripping. Anyway after a suggestion from me he managed to get a 10K or less reading from the cable armour to a metal fence post in the garden. After a bit of digging found the fence post had been driven through the armour of the cable and a condiuctor allowing moisture to get into the cable. This was in 1998 and the posts were present when I moved in 1990, so fault had been present for at least 8 years. Anyway cut cable, replaced section from cut to garage (section from cut to house was OK) connected together using an underground junction box (about £20 from who ever tlc-direct were called before they renamed) and worked fine.

Reply to
Ian Middleton

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.