Light tripping RCD

[RMS]

Hi all,

Many years ago I went to a job where there was a fluorescent light tripping the RCD when switched off.
The light fitting was wired to the non RCD side of the CU. No mixed up neutrals and tests were fine. I changed the light fitting and all was fine.

I had another job the other day. There was a 16watt 2d enclosed light fitting in a bathroom that tripped the RCD when switched off. The light was on the non RCD side of the CU. Again tests were fine and no mixed up neutrals. I changed the fitting and all was fine.

Can anyone explain the problem? I assume its to do with the ballast in the fittings.

Any help appreciated.

 

[ColJack]

back emf from a colapsing magnetic field of the balast.. it has to go somewhere and with the live disconnected it goes up the neutral causing an inballance between live and neutral at the RCD ( it's got current returning but none going out )...

if the RCD is loaded close to it's operating current anyway due to other normal leakage, then it can be enough to tip it over the edge, it may also happen if the RCD is faulty and operating lower than it's rating.

 

[Lectrician]

This usually happens if there is a N to E fault of a lowish impedance on the NON rcd side of the consumer unit.

 

[pushrod]

As coljack says there is a back emf from the inductance in the ballast there is also a capacitor ( the one for power factor correction) across the inductance and tube. It could be that this failing affects the ability of the circuit to deal with it. If this capacitor fails i believe the light can still work it is just drawing a bigger than necessary current.
Just a suggestion. (Not an expert on this by any means)

 

[RMS]

Thanks for all replies.

I suppose if the capacitor fails then the inductance effect produced by the ballast should be greater as there is no longer a capacitive effect to cancel the inductive reactance.

Would this produce a greater back emf than normal? Maybe enough to trip the RCD.

Also how does the RCD pick this up? The line and neutral conductors that supply the light do not pass through the RCD coils. therefore should not cause an in-balance within the coils.

 

[pushrod]

I am no expert as i said , but i suspect the emf would be higher and in any case much higher than 230. All the neutrals are connected. Rcds are normally earth fault devices as they detect an imbalance in live and neutral - i suspect this could trigger it as the neutral would have a bigger current.

I take it that this fault occurred after everything had been working normally for sometime previously and not from installation.