Current "Leakage" between Circuits - Urgent!!

[Colonel Hathi]

Hi
I hope someone can help.
I am testing a 1920s semi (my own).
I have come across a radial circuit (supplying socket outlets) which, when supposedly isolated (mcb off and all cables free), is still showing 37 volts LE LN.
I have found that the voltage is reduced to zero by switching off one of the lighting circuits.

I am working on the basis that there is somehow a cross connection somewhere between the circuits (?) - but why/how only 37 volts - and how to further isolate/remedy the problem?

The installation is rcd protected and otherwise tests out ok.

Can anybody please advise on how I can address this?

I am not comfortable that somebody could think they had isolated a circuit that still carried 37 volts.

Thanks in advance

CH

 

[Farmelectrics]

possible induced voltage from other circut ie pressure

 

[Farmelectrics]

1920 good chance the property will have single cables possibly presure from other circuts which will give an induced voltage

 

[Silly Sausage]

Continuing on from 'farmelectrics'...
You are probably seeing a stray or ghost voltage, have a read of this...
Fluke Brand Portal | powered by SyncForce

I now see that Fluke have this http://assets.fluke.com/manuals/sv225___iseng0100.pdf which eliminates stray voltages.

 

[telectrix]

try measuring with an analogue meter.

 

[Colonel Hathi]

Thanks for the advice.

I will check this out this morning. I hadn't thought of that possibility and let you know how I get on

 

[Colonel Hathi]

Thanks.
I'll check out the link.

 

[Farmelectrics]

colonnel let us no how you get on

 

[Colonel Hathi]

Hi
Thanks for the advice.
The installation does have a lot of "singles" in metal conduit as you said.
I also found two circuits in which the singles from two different circuits were tightly wrapped together with insulation tape near the CU. I have separated these and am checking accessible areas to see where else i can separate conductors.

I'll keep checking.

I am a bit surprised that the Insulation Resistance readings are fine and yet still this apparent "leakage"?

 

[brman]

I regularly get my voltage indicator beeping to indicate voltage (but not showing any on the display) when testing dead circuits. Every time a quick check with my fluke MFT shows a few volts on the line. As others have said is it just pickup from adjacent live circuits where cables are bundled. It isn't just singles either, I've seen it on T&E. This pickup will (generally*) not supply any current so is safe and not to be worried about.
btw. this isn't breakdown of insulation (hence why your IR test is ok), it is capacitive/inductive coupling which will happen between any metal parts in close proximity where one is carrying an AC signal.

*I say "generally" because it is possible to create a transformer by coiling cables together so it is possible to induce a significant voltage/current in a dead circuit if you do this. It is wise to be a bit cautious!

 

[Farmelectrics]

with the metrel tester i use it wont let you test circut if theres any voltage on it went to do circut recently to Ir circut and a warning light comes up and dosnt allow you to test it was an induced voltage from another circut

 

[brman]

I have a feeling most testers do this but I heard somewhere that Metrels are more sensitive than most. Not if that is true though.
I also generally test IR with all circuits dead although I realise this can't always be the case.

I've got to admit the 37V quoted does sound pretty high, from memory I've seen no more than about 10 or 15V but that could just be my lack of experience- I haven't worked much on singles for example.

To the original poster - you say your IR measurements are ok. Have you tested continuity and/or IR between circuits just to make sure you really don't have leakage or a cross connection somewhere?

 

[snowhead]

Stick a 37volt lamp on it and it won't even glimmer.
Neither will a 12volt one.

 

[GB266]

If you have 1W GU10 LED lights, you will find that they will 'glow' dimly even when switched off due to the induced voltage!


You will also see similar symptoms when alarm cables run close to any mains (which is normally unavoidable). You will get around 20V ac at the alarem panel causing false alarms. The only way to get rid of this problem is to fit a spike suppressor!

Good luck.