Guess what’s causing this fault

[RF Lighting]

Been to have a very quick look at an RCBO tripping fault today. So far haven’t had anything in bits or done any testing so I have no idea what is causing it.

The symptom is that the RCBO trips most times when the lights are switched off in one of the rooms. The lights switch on fine and stay on as long as they are left alone. As soon as you switch them off the RCBO trips.

The installation is in a bowling clubhouse. The installation is original fit from the mid 1970s and is T&E buried in the fabric of the building. The room which has the issue has 3no. 100W 8ft fluorescent tubes on the ceiling and is controlled by a single one way MK grid switch.
(This type)
There are lights in 4 other rooms on the same circuit which I think work fine.

The dist board is a 6 way single phase Schneider acti9 board supplying two circuits. The lights and sockets. This board is a few months old and installed to replace an old wylex BS3036 two way. It is fed in SWA from a panel board in the main building remote to the clubhouse. I believe this fault has existed since the board was installed (not by me)

The lights are fed from a B6A Schneider PoN type A RCBO.

I think you now know as much as me so will be interesting to see your guesses and what tests you think will highlight the problem!

 

[RandomGrinch]

Are you going to swap out the tubes for LED anyway?
The problem may just disappear!

 

[Adam_151]

Personanally I'd start with IR test to earth, if you find its good at 500v, and theres no electronic fittings, sensors, or MI cable on the cirucit, I'd even be tempted to test it at 1000v. Following that, check the polarity and just make sure the switch isn't breaking the neutral or anything (long shot, its just that its tripping when turned off is a little odd, I think I'd want to check that). If still no joy, as issue sounds pretty repeatable I think I'd disconnect the fititngs one at a time (both L+N) and see if if its any particular fitting doing it (I've seen chokes have faults which have failed to show up on IR tests), if still no joy, I'd swap the switch/grid module with one from an adjacent room thats not having issues and see what effect that has.

 

[ericmark]

To trip switching off, points to a back EMF which is possible from the ballast, but normally this would not trip an RCD, so I would think also running close to the imbalance needed to trip the RCD/RCBO and the back EMF is just that little extra to cause tripping.

My first test would be with the clamp on to see what the back ground leakage is, and how close to wind sailing, unusual with RCBO's for back ground leakage to be enough to worry about, but in a large building one does not know where the cables run and if close to earthed metal work is could be enough to tip the balance.

Second test would be with in insulation tester, if only the clamp on shows significant leakage then looking are where the cables run, and smoothing capacitors, but if both poor, then looking at dampness.

 

[333rocky333]

capacitors in the fittings possibly causing the issue

 

[SUNRAY]

Had same thing when changing from twin RCD to RCBO, wired in tube and singles in this basic arrangement, it was something like 5 rows of 4 70W discharge lamps at about 20ft and neutrals at that high level, dropping down to the DB in the adjacent space. the live wiring in trunking round the walls at about 8ft with other circuits. The live drops were one single in tube. All wiring being 7/0.029" the MCB being C25 and electrical tests were perfect.

never did suss out the actual mechanics of the problem but rewiring sorted the problem

It was one of those 'why on earth did they ever run all the way round the trunking when they were running into the tube fitted for the purpose' situations.
Not only that but due to the time and access equipment restrictions we even reused the existing wire with one joint.

 

[securespark]

There were two occasions where something remotely similar to this happened:

One was turning off a circuit with a fridge plugged in.

The RCD tripped on switch-off.

The fridge had N-E reversed in the plug.

This tripped every time the circuit was isolated while the fridge was plugged in, but never tripped when the fridge was disconnected. You say the RCBO trips "most times", not every time.

The other was a really sloppy, crackly switch.
I never really understood quite why an RCD would trip because of a poor connection, but swapping the socket (which IR'd OK) stopped the tripping.

So, on that basis, I suspect a dodgy fitting in that room. Or maybe a dodgy grid switch module.

Having said that, if it were a rubbish switch, you'd probably hear it on operation.

 

[SUNRAY]

There were two occasions where something remotely similar to this happened:

One was turning off a circuit with a fridge plugged in.

The RCD tripped on switch-off.

The fridge had N-E reversed in the plug.

This tripped every time the circuit was isolated while the fridge was plugged in, but never tripped when the fridge was disconnected. You say the RCBO trips "most times", not every time.

The other was a really sloppy, crackly switch.
I never really understood quite why an RCD would trip because of a poor connection, but swapping the socket (which IR'd OK) stopped the tripping.

So, on that basis, I suspect a dodgy fitting in that room. Or maybe a dodgy grid switch module.

Having said that, if it were a rubbish switch, you'd probably hear it on operation.

Yes an arching switch is a known problem for this.

 

[RF Lighting]

I did wonder about the switch but it feels ok and I can’t hear any arcing. Even when I jabbed the switch so it opened quickly it still tripped. If the weather stays poor I’m going to go and have a look tomorrow.

 

[hovertoast]

The other was a really sloppy, crackly switch.
I never really understood quite why an RCD would trip because of a poor connection, but swapping the socket (which IR'd OK) stopped the tripping.

So, on that basis, I suspect a dodgy fitting in that room. Or maybe a dodgy grid switch module.

Having said that, if it were a rubbish switch, you'd probably hear it on operation.

I once had a class II vacuum cleaner where the flex had frayed at the cord grip in the plug. Was using it one day and the RCD tripped after the vacuum stopped/started a few times when wriggling the flex around. Rewired the plug after chopping off the dodgy bit of flex and it never happened again.

Only other time that (single upfront) RCD tripped was when I accidentally touched neutral and earth together when changing a socket (with the MCB isolated). In hindsight I think I found the answer to that “issue” on this site many years ago.

Never could explain how an appliance with no earth connection could trip an RCD and have been thinking I should post about it here for some time but never got around to it.

My vote is sloppy switch too.

 

[Adam_151]

Never could explain how an appliance with no earth connection could trip an RCD and have been thinking I should post about it here for some time but never got around to it.

Having seen similar - Loose neutral on a henley block before the board that caused the RCD to trip only when the vacuum cleaner was in use - The kettle made it fizz and crackle but didn't trip the RCD. I suspect the answer is that the arcing of the connection, combined with the inductive load probably caused significant interference on the mains that either the RCD was directly affected and tripped, or the high frequency voltage spikes it caused resulted in significant capactive leakage across twin and earth cable elsewhere on the installation. Though I'm not sure if thats the case, how we don't get similar issues with a brushed motor in normal operation - much smaller and more regular arcs I suppose

 

[crystal ball]

I have said for years - RCDs are tricky chaps, seem to defy logic sometimes

 

[RF Lighting]

Having seen similar - Loose neutral on a henley block before the board that caused the RCD to trip only when the vacuum cleaner was in use - The kettle made it fizz and crackle but didn't trip the RCD. I suspect the answer is that the arcing of the connection, combined with the inductive load probably caused significant interference on the mains that either the RCD was directly affected and tripped, or the high frequency voltage spikes it caused resulted in significant capactive leakage across twin and earth cable elsewhere on the installation. Though I'm not sure if thats the case, how we don't get similar issues with a brushed motor in normal operation - much smaller and more regular arcs I suppose

I posted a similar one quite some time ago!!

Video of the month!!

Seeing as Securespark is trying to out do photo of the week, here is a new feature! This month we have a video of an intermittent RCD fault I was called to. I was the second bod from our firm to look for this fault, and it had been ongiong for a good few weeks. The rcd would trip every...

www.diynot.com

Unfortunately photobucket have broken the link to the video and want to charge me to use it - although it was pre smartphone so not exactly great quality

 

[securespark]

Me outdo photo of the week?

Never!

 

[Taylortwocities]

I have said for years - RCDs are tricky chaps, seem to defy logic sometimes

Yes. They sit there for years doing nothing. And then, all of a sudden, they go faulty and start tripping.
Obviously the solution is to go out and buy a new one, but there's some sort of RCD virus that infects the new one, and that new one is tripping too.
Yes, very tricky chaps.....