RCBO trips when light turned on (shared line/neutral?)

[RickDastardly]

Here's the situation: I have (had) upstairs and downstairs lighting on separate circuits with 6A/30mA RCBOs. When I turned on the landing light (which is wired into the downstairs circuit) from ether hallway or landing switch, the downstairs RCBO trips. It also trips if I turn on the kitchen light (which is not dual switched), which (surely not by coincidence) runs through the same triple switch wall switch (hall light, landing light and kitchen light... don't ask - I just moved here, I didn't wire it).

OK... shared neutrals and all that, so I connected both circuits to the same RCBO. It still trips on the kitchen and landing lights if bulbs are fitted.

The RCBO does not trip if I take the bulbs out of the fitting, but does if I replace the LED bulbs for a filament, so it's not LED driver inrush (not that I thought it would be).

I tested the circuit's insulation and got 68 Mohm between earth and line/neutral shorted, and over 600 Mohm line to neutral. That was with the switches on but no bulbs in. So nothing to see there. I've checked the wiring in the switch and light fittings, all seems as it should. I haven't pulled up the floor boards yet.

Any hints what to look for?

 

[ericmark]

I know my 4 gang light switch at front door it was simply the wrong line used. With non RCD protected circuits borrowed neutrals could have been there since the house was built, but in the main introduced in the 90's when people thought quartz halogen down lights were cool, and so were overloading lighting circuits, since the ceiling rose was rated at 5 or 6 amp, fitting a larger MCB was out, so they split up and down stairs, so giving a total of 12 amps, without the inspection and testing which should have been done. So where twin and earth used between up and down stairs and the line borrowed for an adjacent lamp, we got a borrower neutral.

To fit a new consumer unit so putting lights on different RCBO's the person doing it should inspect and test and also notify the LABC. So putting is simple, should not be your job to sort out, however if you do need to sort it, you will need test equipment. The idea of knit one pearl one does not work with electrics, you need to test, I have two testers both cost around £35, one is an insulation tester (I would use the 250 volt range) the other is a clamp on ammeter, with the latter you may need to fit the MCB's back in while you test.

It is hard to give step by step instructions. You have already worked out wrong neutral, I used a tungsten bulb while testing so I could read the 40/230 = 170 mA on the meter with ease, which was well over any current due to induction or capacitive linking. I know my whole system (14 RCBO's) has around 24 mA leakage, which is around 6 watt, so hard to tell which is which with small bulbs.

But it does depend where one can gain safe access, I am personally not keen on standing on a jump up stool to access my consumer unit leaning over an earthed metal sink, so tried to measure else where. So you seem to understand what has happened, so have some common sense, so it is down to where can one get safe access to measure, and clearly ensure your not alone, and there is some one to call for help.

I made that mistake, so simple, slipped on the steps, in the cold, and then stuck until my daughter found me, not an electrical accident simply the step broke, so make sure there is some one able to help before you start.

 

[RickDastardly]

Thank you for the reply. My current thinking is they've 'borrowed' something from another circuit entirely, so yes, testing is needed to find out where. The previous owners stuffed everything up here so much it's unbelievable; electrics, plumbing, plaster, structure... you name it, they broke it and wallpapered over it. I know they rebuilt (Edit: that's too kind a way of putting it; 'destroyed' would be more apt) the kitchen so no wonder there's (another!) problem with the wiring in it.

I have an older three-piece Robin kit. Not in calibration but a few tests with known readings and against my HP benchtop multimeter say it's all OK. I've done Ze, PSCC, PEFC, checked Zs around the place etc.; everything is within acceptable range. At least it is now I've replaced the crumbling VIR cable I found junctioned into and hidden under the bathroom (and under cement rendering in the outbuildings).

I've just got a low current clamp meter for hunting for earth leakage too. I guess that's going to see a lot more use than I expected now.

I also have a copy of the Brown Book and know how to use it,

 

[ericmark]

Nice to hear you have the test gear. I had problems with dad's old house, he would not let me re-wire it, it was built 1954, I could not locate the earth leakage, but clamp on only measured 0.01 amp, and insulation tester showed all OK, when he died finally got whole house rewired, I am guessing either some VIR cable not found, or some overloaded thermal plastic cable. And the plasticiser had leached out.

Since no earths on lighting, it needed a rewire, and speed was important as paying for mothers care home while being done, so cheaper not to DIY as care homes are expensive. But dad seemed to think a spur of spur was no problem, either no FCU or two FCU's in series with one hidden, so would take me ages to find it.

But can't see how I can help you, seems you know what you are doing, wish I had got the clamp on I use now with dad's house, as measures down to 0.001 amp, interested to see what you find in the end, as said I gave up and had a rewire, after mother put an extension lead with a neon on it into a bucket of water as she thought it was on fire. Poor eye sight more than the alzheimer's. But it made me realised the rewire would not wait.

 

[sparkwright]

Here's the situation: I have (had) upstairs and downstairs lighting on separate circuits with 6A/30mA RCBOs. When I turned on the landing light (which is wired into the downstairs circuit) from ether hallway or landing switch, the downstairs RCBO trips. It also trips if I turn on the kitchen light (which is not dual switched), which (surely not by coincidence) runs through the same triple switch wall switch (hall light, landing light and kitchen light... don't ask - I just moved here, I didn't wire it).

OK... shared neutrals and all that, so I connected both circuits to the same RCBO. It still trips on the kitchen and landing lights if bulbs are fitted.

The RCBO does not trip if I take the bulbs out of the fitting, but does if I replace the LED bulbs for a filament, so it's not LED driver inrush (not that I thought it would be).

I tested the circuit's insulation and got 68 Mohm between earth and line/neutral shorted, and over 600 Mohm line to neutral. That was with the switches on but no bulbs in. So nothing to see there. I've checked the wiring in the switch and light fittings, all seems as it should. I haven't pulled up the floor boards yet.

Any hints what to look for?

Photos of inside a few light switches may help here.

 

[plugwash]

Thank you for the reply. My current thinking is they've 'borrowed' something from another circuit entirely,

Another possibility is that they have abused the earth as a neutral.

 

[RickDastardly]

Photos of inside a few light switches may help here.

I'll get onto that tomorrow or Saturday. I need to have a closer look myself anyway.

Another possibility is that they have abused the earth as a neutral.

Hmm... I need to put that clamp meter to work!

 

[flameport]

I tested the circuit's insulation and got 68 Mohm between earth and line/neutral shorted, and over 600 Mohm line to neutral. That was with the switches on but no bulbs in.

Install at least 1 filament lamp, light switch on, and with the whole consumer unit switched off, test insulation resistance between the L&N of the lighting circuit and the N of each other circuit.
If there is a connection between circuits it will show as a very low resistance.

 

[Murdochcat]

I also have a copy of the Brown Book and know how to use it,

But won't help you in many situations like this ................

 

[RickDastardly]

Thank you @flameport I'll do that this weekend when I get chance to look at it again.

 

[Murdochcat]

Given your have another post about 1960's wiring who installed the new CU and what were the test results on the EIC?