Another tingling light switch thread.

[Keithmac]

Hello All, just looking for a bit of clarification of possible please.

I retiled our kitchen and fitted metal faced light switches at the same time. These are all correctly wired including the earth.

Wife noticed "tingling" a while back when touching the switch but I never experienced it so put it down to the brushed suface of the switches. Anyway she has mentioned it again so I had a quick look tonight.

There are two ceiling lights in the kitchen on a switch each (seperate walls). I got the multimeter out and checked between the two switch faces, it showed continuity. I then checked between these and the plug socket earths expecting the same but it was open circuit!.

After puting the multimeter in voltage mode it's showing 60v ac between the plug socket earth and the switch face which is not good at all!.

Reading up it seems I should have an earth coming out of the consumer unit, and on each ceiling rose there should be a total of 3 earths (earth in, earth out to switch and earth out to next ceiling rose) all connected together.

My plan is to drop all the light fittings and switches and try and find the problem.

The house has been rewired at some point and on twin and earth cable.

I noticed the lighting circuit is not protected by an rcd, just an mcb is this ok with metal faced switches?.

 

[JohnW2]

There are two ceiling lights in the kitchen on a switch each (seperate walls). I got the multimeter out and checked between the two switch faces, it showed continuity. I then checked between these and the plug socket earths expecting the same but it was open circuit!. ... After puting the multimeter in voltage mode it's showing 60v ac between the plug socket earth and the switch face which is not good at all!.

As you say, not good at all - and you don't yet even know whether it is the lighting circuit or the plug circuit (or both!) which is/are not properly earthed.

Do you have an accessible 'main earthing terminal' (a metal terminal box with green/yellow insulated wires connected to it) in the vicinity of your consumer unit?

Kind Regards, John

 

[Keithmac]

Hello John, thanks for the reply, the consumer unit is in the outhouse (concrete floor) I'll have to dig some stuff out and have a look.

I have checked before between the metal water inlet supply pipe and the socket ring main eath and it had good continuity. Have used a socket tester a few times which always shows the sockets are correctly wired.

We had a Shower circuit fitted approx 7 years ago and signed off, would they have checked the main earthing then would you think?.

 

[JohnW2]

I have checked before between the metal water inlet supply pipe and the socket ring main eath and it had good continuity. Have used a socket tester a few times which always shows the sockets are correctly wired.

OK. That certainly suggests that the sockets are probably properly earthed - which in turn suggests that the lighting circuit isn't.

We had a Shower circuit fitted approx 7 years ago and signed off, would they have checked the main earthing then would you think?.

... for the shower circuit, at least, but they would probably not have checked the earthing of lighting circuits.

Do you have more than one lighting circuit (e.g. another for 'upstairs')? If so, does the same problem exist if you measure between switches and sockets upstairs (assuming they are also metal)?

Kind Regards, John

 

[Keithmac]

Hello again John, I will check that tomorrow and get back with results (kids are in bed so can't make much noise).

We have 2 lighting circuits upstairs and down.

All upstairs are plastic or pull cord but the back boxes will be earthed (hopefully!).

If all the earths terminate at the consumer unit and I have relatively short runs (3 bed mid terrace house) whats a reasonable resistance reading between socket earth and lighting earth (less than 5 ohm?).

Bit worrying that at least the downstairs lighting has not been earthed correctly, just goes to show you can't assume anything!.

Thanks again, much appreciated!.

 

[JohnW2]

If all the earths terminate at the consumer unit and I have relatively short runs (3 bed mid terrace house) whats a reasonable resistance reading between socket earth and lighting earth (less than 5 ohm?).

Probably a lot less than that, although measurements with a multimeter can get a bit iffy at very low resistances. In practice, you would be most likely to see either 'very low' resistances (i.e. 'normal') or 'incredibly high' (i.e. literally or virtually no connection).

However, whilst you can look at voltages between things that should be earthed (as you have done), if you wanted to measure resistances you would have to completely switch off the electricity to the circuits concerned - not only for reasons of safety but also because resistance measurements could get totally screwed up by the presence of a voltage between the points being measured.

Kind Regards, John

 

[flameport]

(less than 5 ohm?).

Probably less than 1 ohm.

 

[Keithmac]

I've got Thursday off, will being my decent Snap-on mutlimeter home from work (believe it's a rebadged Fluke) and aee what I can find.

Hopefully it'll just a fault in one of the ceiling roses (I live in hope!).

Thanks again.

 

[Keithmac]

Hello again, thought I'd update the tread with a conclusion!.

Had an interesting morning pulling 7 light fittings and painted ceiling roses down (a few taps with a nylon mallet works a treat!).

After going through the kitchen hall way and living room checking continuity I was a bit fed up!, one last fitting that I'd overlooked was in the outhouse.

Bingo!, two earth wires were not connected to the main earth out of the consumer unit.

All connected and gave it a good tug with the long nose pliers just to make sure!.

We have metal light fittings all through the downstairs and a few metal switches.

At a guess these have not been earthed correctly/ at all when we moved in some 13 years ago!.

Dread to think what could have happened if there was a live to metal casing fault as the whole downstairs fittings would have become live..

All down to one failed connection, surprised there isn't any kind of redundancy built in (like socket ring main).

I am genuinely a but upset, never dawned on me to check the light fittings to socket earth.

Lucky my wife mentioned it or it could have been far worse.

 

[JohnW2]

Hello again, thought I'd update the tread with a conclusion!. .... Had an interesting morning pulling 7 light fittings and painted ceiling roses down (a few taps with a nylon mallet works a treat!).

Thanks, but where's the 'conclusion'?

Kind Regards, John

 

[Keithmac]

Thanks, but where's the 'conclusion'?

Kind Regards, John

Sorry, phone was playing about and posted half a story!.

 

[flameport]

surprised there isn't any kind of redundancy built in

Redundancy isn't required - proper inspection and testing is, both when the circuit is installed and at various intervals after that.


This is what redundancy does:

 

[JohnW2]

After going through the kitchen hall way and living room checking continuity I was a bit fed up!, one last fitting that I'd overlooked was in the outhouse. ... Bingo!, two earth wires were not connected to the main earth out of the consumer unit. .... We have metal light fittings all through the downstairs and a few metal switches. At a guess these have not been earthed correctly/ at all when we moved in some 13 years ago!. .... I am genuinely a but upset, never dawned on me to check the light fittings to socket earth.

I'm glad that you found the problem. Do I take it that the continuity/voltage issues been metal faceplates have now gone away?

The worry is, of course, if your installation has had that one 'loose/missing connection' problem for years, one wonders what other issues might exist in the wiring of your installation - some would probably suggest that you should have the whole installation inspected.

All down to one failed connection, surprised there isn't any kind of redundancy built in (like socket ring main).

Redundancy of the 'earth' connections is, indeed, one of the (few) arguments in favour of ring circuits for sockets, but I have never heard it being suggested that a similar approach should (under normal circumstances) be taken in relation to anything other than (some) sockets circuits. Redundacy can be a two-edged sword, and there is an argument that it is better not to have it, but to have regular inspection/testing.

Kind Regards, John

 

[Keithmac]

I've replaced a few sockets over the years and all seems to be in order but as you say a full inspection would be a good idea.

To be honest looking at the offending light fitting (hand painted green!) I think the previous owners may have played a part in it.

Been upstairs and checked all the earth continuity on that lighting circuit and all good there.

This is my consumer unit (Wylex), would it be worth having a new one fitted do you think?.

 

[JohnW2]

I've replaced a few sockets over the years and all seems to be in order but as you say a full inspection would be a good idea. ... To be honest looking at the offending light fitting (hand painted green!) I think the previous owners may have played a part in it.

... and may have played a part in other (as yet undiscovered) things as well?

This is my consumer unit (Wylex), would it be worth having a new one fitted do you think?.

Not at all young, but nothing inherently wrong with it. However, it appears to have RCD protection of only three of the circuits, whereas current practice (with which you are not obliged to comply) would usually have two RCDs (between them covering most or all circuits) or separate RCBOs (essentially a combined RCD and MCB) for each circuit. If you did decide to have an inspection ('EICR') of your installation undertaken, this might be something you would want to discuss with the person undertaking the installation (who would probably 'note' that the CU was not up to 'current standards').

Kind Regards, John