Puzzled

[bernardgreen]

Interesting.

I have seen power sockets connected to next door's cables, but that cannot be the case here as turning OFF the main switch removes power from this socket.

It has to be some sneak wiring in the CU around the Live bus bar or something very strange..

Cannot be a UPS ( back up invertor ) as that would would still be active when the main switch was OFF.

Any solar PV panels and a feed in invertor. That invertor would shut down when the main switch was OFF as it would have no reference frequency to synch to.

 

[PrenticeBoyofDerry]

No secondary supply of any type.
It is a simple 6 way system with RCD as main iso

 

[AndyPRK]

Use a DVM between cpc and the outgoing MCB ways to see if one is stuck closed

 

[ban-all-sheds]

Use a DVM between cpc and the outgoing MCB ways to see if one is stuck closed

When all these MCBs are opened the one double socket remains powered, so I now doubt it can be a faulty breaker, as each device does seem to power down a circuit.

 

[PrenticeBoyofDerry]

Okay the puzzle continues I have taken some pictures

When the busbar is removed from RCD/ISO and MCBs, the socket still remains powered up when RCD/ISO is closed and powers down when RCD/ISO is opened.

 

[chivers67]

Can you remove the cable supplying the socket and tone it somehow to see what/where it connects to when powered down.

 

[mikeyd]

If all the MCB's function and there definitely isn't a live connected behind one of the bus bar connections which you haven't seen, then it must be something to do with the main switch / RCD isolating the neutral as well as live. (I'm thinking live connected upstream of consumer unit). But I can't think how this could allow the socket to to supply any power unless the RCD is faulty.

Edit: I'm a slow typer.

 

[PrenticeBoyofDerry]

Can you remove the cable supplying the socket and tone it somehow to see what/where it connects to when powered down.

I would need to investigate a little further to establish where the cable is routed and where it is terminated, the simple visual evidence seems to suggest it does not exist at the board! Although common sense says it must?

 

[Echo the husky]

Are you losing both Line and neutral at the socket when the RCD is off?
What happens when the neutral bar is disconnected from the bottom of the RCD?
Does the RCD trip when tested?

 

[deadshort]

Wire a lamp across L+E and plug it into the socket does it trip the RCD?

Regards,

DS

 

[sparkwright]

And your certain the polarity at the socket in question really is correct?

Lets think about this.

The neutral wouldn't get interupted when isolating the MCB.

But a neutral not connected somewhere would become live.

Just thinking aloud. I've no solution, but maybe something here may get you thinking.

Out of interest, are the wires at the socket in question old colours or new colours?

As said, when the main RCD is off, do you definitely lose L and N at the socket?

And when the main RCD is off, do you still have E at the socket?

 

[sparkwright]

Have you checked at the board when each breaker is turned off the outgoing side of each breaker is actually dead?

Any cross connected breakers, where two breakers isolate something?

Apologies for the random questions, but some very strange faults have been down to one or more strange issues.

 

[PrenticeBoyofDerry]

To try answer all the recently asked questions, I am sure polarity is correct at the socket, as my MFT would normally flag up a polarity issue and trip the RCD, when testing Zs.
I am sure I have done the auto RCD test at this socket (but sure with a question mark). As I was actually working on other parts of the installaion, prior to being asked to swap this socket face plate.
The core colours are the older version (red/black), if have not as yet tested continuity of any of the conductors back to the board.
If have not tested each individual breaker for power at the load side of the breaker, but each breaker does power down parts of the installation when open/off. Also the complete removal of the busbar from ISO load terminal and MCB feed terminals, does not solve the issue, nor is there any additional link between ISO load terminal or any other breaker that can be seen.
I shall return to the property today and do the test requested in the above posts.
Then come back later today with results and hopefully a conclusion/diagnosis!

 

[bernardgreen]

Have you tried using the suspect socket to supply a load greater than 15 watts ?

If the RCD does not trip when the suspect socket supplies a load more than 15 watts ( 60 mA at 230 volts ) then one of three possibilities :-

(1) both Live and Neutral to the socket are through the RCD ( equal and opposite so no trip )

(2) neither of the Live or Neutral to the socket are through the RCD ( if only one was through the RCD then the 60mA difference would trip the RCD )

(3) either Live or Neutral ( only one of them ) is through the RCD and the RCD is defective and not tripping on 60 mA

Removing the bus bar does mean the Live to the Socket cannot be going through the RCD hence (1) is not possible

Switching the RCD OFF appears to remove supply to the socket. This can only be by cutting the Neutral in the RCD as Live to the socket is not through the RCD hence (2) is not possible

This leaves (3) as the only logical possibility with the the socket's Neutral through the RCD and the socket's Live being taken from a junction somewhere between incoming supply and the CU ( or from next door's wiring that is not RCCD protected )

From the above it seems that the Neutral may be coming from the Neutral bar in the CU ( maybe connected somewhere in the house to a Neutral from the CU ) and this would explain the socket appearing to be dead when the RCD / Main switch is OFF. In reality the socket is not dead but has 230 volts on the Live but no Neutral. With the RCD / main switch OFF measure Live to Earth and Neutral to Earth at the socket. If possible use a water pipe that you know is earthed as the earth reference as the CPC at the socket may not be at earth potential. ( The CPC may not be connected at the far end of the cable feeding that socket ).

What is there between the incomer and the meter and between meter and the CU. Any Henley blocks or main isolators that may be the point at which the Live ( but not Neutral) in the cable from the socket is connected. Any evidence of an OFF PEAK supply to storage heating / immersions that have been removed. Any redundant but still connected Teleswitches or Timeswitches related to Off Peak.

or there could be a (4) that is totally illogical

EDIT the bit in red added

 

[Lectrician]

What are the voltages like with the trip off?