Weird electrics - any ideas?

Just after buying the house we had new tableaus fitted, Linky etc replacing a lot of old gear. The sockets in one room however did not work (no idea if they did immediately before we got the house, or if they’d been dead for some time) and we had the electrician back to investigate, but he could find no reason for their non-functioning. Knowing that the wiring of these was likely dodgy anyway and we were putting in a new kitchen that would be wired direct from the tableau, we didn’t pursue the matter.

So today, I was going to take the old sockets off the wall ready for replastering in a few weeks, and just thought I’d check them for electricity before snipping wires. I get 122V across live & neutral (no earth :stuck_out_tongue: ). Plugging a fan heater in, the voltage drops to 0.2V and the fan heater doesn’t respond.

Any ideas why they might have a 122V potential? The house is on single phase, so it can’t be that. I could just snip off & bind up, but would rather do so knowing why and that the wires weren’t going to suddenly discover the joy of 230V.

The most likely is leakage current from somewhere, not a direct connection.


As said above.
Probably a shared neutral. Try testing with your multi meter by live to earth use an extension lead and then neutral to earth.


Wasnt much of an electrician?


Has your tableau got an RCD ? If you have and it didn’t trip with the fan heater it’s unlikely to be a neutral fault.

It sounds like a really bad connection - you get similar low volt readings if your MCB connections are loose - that’s if your lucky - but loose connectors somewhere in the line have the same effect.

1 Like

I think you will need to trace this out and/or remove sockets and all wiring to be certain.


I’m surely not the only one who flinches when the electrics are playing up… with the knowledge of what might (but only might) happen…
As has been said… this needs thorough investigation, right back to the board.

best of luck…

Thanks everyone. I will try the extension lead trick, but as chrisell points out, it’s a modern board with RCDs.

I’m not overly impressed, but as long as everything is safe then I don’t really care. He’s a local, qualified electrician, and while perhaps I could call him back again, I don’t really want to do that.

If the voltage drops away like you say then it’s probably a ghost voltage due to induced current.

Somewhere that old wiring runs close & parallel to some live wires for a long distance.

Finding ghost voltages is a flaw with modern digital meters. Old moving coil meters won’t see them.

1 Like

It’s because they have such high input impedances - even a whiff of current, mere nanoamps, will register.

That said it is an indication that the old and new wiring is in proximity and if the insulaton on the old is perished then there is the potential (sorry) for problems.

One thing which could be done is to short phase, neutral and earth on the old, supposedly dead, wiring - that would make certain no shock hazard could present itself.

1 Like

@Ancient_Mariner You mentioned ‘sockets’ in the plural. May I suggest removing all of them and then testing each piece of wire to see if there is anything indicated between it and a known good earth connection. In that way you should at least be able to reduce to a minimum the suspect wires.
Also, what type of sheathing is on the old wire ? Is it the old rubber type that just flakes off with age ?
One way to narrow down where the voltage is coming from is to switch off the other circuits one at a time until the suspect voltage goes away. That should help to give an indication of where to investigate further.

Thanks again gents, I will try the true-earth trick and also switching off circuits.

Sounds like poor neutral and maybe you are picking a voltage to earth but this depends on your earting arrangment. The readings indicate poor connection which break down under load ( high resistance) If you isolate the instalation and undertake testing for continuity of the neutral connection back to the board but will need to uplug anything on the circuit or split the live and neutral connections

Have you tried the light switch for the room? Some of the sockets in my house are worked by the main light switch for that room.


Ok crikey - hope you’re not using them for anything with a heavy load.

No - only lamps and fortunately it is only in two guest bedrooms that aren’t used very often

Thanks for the suggestion. The light switches in this room operate a relay in the tableau. It’s not connected to the socket.

As I/Badger said, it will be leakage or capacitively coupled voltage - modern test meters are very sensitive.

If the new/old conductors run in the same gaine for 10m they could have a core to core capacitance in the order of a few hundred picofarads to a nanofarad or so, even with no direct connection - just touching will be enough.

At 50Hz 300pF has an impedance of about 10MΩ - if you put that in series with a typical multimeter which also has an input impedance of 10MΩ and you will get a voltage divider which will show half the mains voltage - which is what @Ancient_Mariner saw.

Put it in series with a fan heater and it will drop down to almost nothing - assuming just the fan motor of about 100W or ~500Ω you’d get a few mV.

However given that there is evidence that the new and old wiring seem to be in proximity, and probably for some distance, I would have the question in my mind “just how well isolated is the old wiring”.

I’m probably being overcautious, I realise.

1 Like

Very much like these touch devices to check for live leads, if there is any doubt I go back with meter before touching it.

I’ll try to do the check with separate proper earth Tuesday. My expectation is that the wiring hasn’t been connected at the tableau and this is induction PD. As I see it, provided the wires are safely capped off to ensure that they can’t touch at the socket end, then unless they can physically move in some way to cause a short at another junction, they should be safe.