Bit of an odd one.

[studentspark]

Just been to a job to change a light.

Tried to switch off the circuit at the CU, but the lighting circuits mcbs did not switch off power to lights.
Tried all other mcbs and still lights stayed on.
No mcb would turn off the lights.

When I switch the lighting mcb off the lights dim a little.

When I switch the socket circuit of the lights dim a little

Only when I switch off the lighting mcb and the socket mcb do the lights go off.
So both mcb need to be off.

Some connecting link between the two circuits I can only presume.
Anybody come across an issue like this before.
Board was installed a few months ago!
Thanks

 

[ericmark]

I have had shared neutral faults do something like this, and they can be really hard to trace. I would start at consumer unit check neutrals there.

 

[studentspark]

Not had a look yet, will do so tomorrow,
Was thinking could be a fused spur connection to a light, or a ring final cable put into the lighting mcb rather than the sockets mcb.

 

[deadshort]

Board was installed a few months ago!

Scary stuff ! No paperwork , i bet…..

DS

 

[333rocky333]

Sounds like the lives of both circuits being linked through some sort of load hence the dimming, possible source something like a timer fan on the lights, with a permanent feed taken from a socket

 

[deadshort]

You have just reminded me of exactly that situation ! And no fusing down on the permanent live ,just straight off the back of a socket, the fan had been installed for years.

DS

 

[JohnW2]

Sounds like the lives of both circuits being linked through some sort of load hence the dimming, possible source something like a timer fan on the lights, with a permanent feed taken from a socket

It's difficult to imagine what could be going on. For the lights to dim when either of the MCBs were opened would presumably require that there were some sort of impedance in the feeds from both circuits - and it's not easy to think of how that could happen.

Furthermore, I don't think it could happen through a timer fan in the sort of way you are suggesting, since there is usually a very high impedance path between the permanent live and S/L of such a fan. The S/L is just a 'trigger' input, and is usually wired via a very value resistor (often to an IC which has an even higher impedance) - so one would expect the lights to go off completely if the (very slight, at most) path through the fan were to become the only path from L to the lights (when the lighting MCB was opened).

Were it not for the 'dimming' that has been reported, it could obviously just be a simple cross-connection between the circuits, and there are many ways in which that could arise 'by error'. I did once see one end of a ring final connected to a lighting MCB!

Kind Regards, John

 

[endecotp]

Imagine that you have a two-way-switched hall light wired in this "not recommended" way, and the switches are set as shown, so the light is off:

Now imagine that the left side is connected to the lighting circuit and the right side is connected to the socket circuit.

Now imagine that somewhere else on the lighting circuit, another light (with standard wiring) is switched on.

Turn off the lighting MCB but leave the sockets MCB on.

A current can flow through the two lamps in series, so they appear dim.

Similarly there can be a load on the sockets side that allows current to flow when the Lighting MCB is on and the sockets MCB is off.

Questions: is this a two-way-switched light? (Note that even if it isn't, you may be looking at the other light in the circuit.)
What happens if you remove the bulbs from all the hall two-way-switched lamps?
When both MCBs are ON, is this lamp on or off?


p.s. There are surely many possible explanations other than this one.
It's also more than possible that there are multiple problems with the wiring.
Efficiently diagnosing it will be a great test of your detective skills!

 

[JohnW2]

Imagine that you have a two-way-switched hall light wired in this "not recommended" way, and the switches are set as shown, so the light is off: .......... Now imagine that the left side is connected to the lighting circuit and the right side is connected to the socket circuit. .... Now imagine that somewhere else on the lighting circuit, another light (with standard wiring) is switched on. ....

Well, yes, that could do it ... but that would be a pretty extraordinary (and potentially dangerous) way of (deliberately) wiring two-way switching, even if the two switches were connected to the same circuit. Have you ever seen it done like that?

Kind Regards, John

 

[John D v2.0]

I was under the impression that way was quite common and the reason behind borrowed neutrals on the landing light. Borrowed live would be another alternative though!

 

[EFLImpudence]

No, that is this way where L is downstairs and the light and N upstairs.

 

[JohnW2]

I was under the impression that way was quite common and the reason behind borrowed neutrals on the landing light.

Sorry, I missed this one last night. In the meantime, EFLI has illustrated the 'old' method of implementing 2-way switching - as you say, still quite commonly found (and once-upon-a-time 'universal'). I find it hard to believe that the method suggested by endecotp will ever have been regarded as acceptable, not the least because, in one of the 'off' configurations, both sides of the lamp are at L potential (although the light obviously not 'on').

Borrowed live would be another alternative though!

I don't think that there is really any difference between 'borrowed live' and 'borrowed neutral' - both refer to a situation in which L comes from one circuit and N from a different one (and becomes noticed if those two circuits are protected by different RCDs). However, there is then only one connection to (an) L and one connection to (an) N, so that could not be an explanation for what the OP is describing.

To get the observed behaviour, the lights must be getting, directly or indirectly, an L feed from both the circuits concerned. As I've said, were it not for the 'dimming', it could just be a direct connection, which could have arisen in many ways (like the one I mentioned), but it taxes the brain more to think of ways in which the dimming can occur!

Kind Regards, John

 

[bernardgreen]

but that would be a pretty extraordinary (and potentially dangerous) way of (deliberately) wiring two-way switching, even if the two switches were connected to the same circuit. Have you ever seen it done like that?

I found that two way system in a couple of houses in the distant past. More recently I came across it when the house owner mentioned that the light switches on the stairs needed replacing frequently and often tripped the MCB. When the switch was operated the arc between the fixed contact and the moving contact continued until the moving contact landed on the other fixed contact. The arc was now crossing the 3 mm gap directly between Live and Neutral and maintained itself until the MCB tripped or the switch burnt out

In the old bakelite and ceramic two way switches there was a large gap between the two sets of fixed contacts. Often they were two separate make switches with a common toggle. The moving contact was far enough away from the fixed contact that the arc collapsed before the moving contact reached the other fixed contact

 

[JohnW2]

I found that two way system in a couple of houses in the distant past.

Interesting. As I said, I find it hard to believe that it was ever regarded as 'acceptable' or 'allowed'.

I can't even think why anyone would have wanted to do it. About the only 'advantage' I can think of is that it does not require any 3-core cable. However, on the 'downside', it requires cables from both switches to the light and also requires L&N feeds to both switches, not to mention the inherent 'dangers'. Am I missing some other 'advantage'?

KInd Regards, John

 

[bernardgreen]

Am I missing some other 'advantage'?

There probably isn't any advantage, these were in the days when houses were wired using singles so avoiding the need for 3 core wasn't a consideration.