Can a dodgy bulb cause a lighting circuit to fail?

[JohnW2]

The reason that an RCD may trip when a lamp blows is most likley to be as follows.

Maybe "most likely" but, IMO still extremely unlikely.

The high fault current flowing through the impedances along the Neutral causes the Neutral in the lighting ciruit to rise to a significant voltage above ground and the CPC. Capacitive coupling between Neutral and the CPC allows for some of the fault current on the Neutral to flow into the CPC. The portion of the fault current that flows along the CPC bypasses the RCD sensor and hence Live current and Neutral current through the RCD sensor are not equal.

The difficulty I have with that hypothesis is that I would have expected that such capacitive coupling effects would affect L and L more-or-less equally. L-CPC capacitive coupling in the circuit should be essentially the same as N-CPC capacitive coupling; the very high fault current should be essentially the same in both L & N; the change of voltage (and, more importantly, rate of change of voltage) due to the transient very high fault current should be essentially the same in both L and N. Accordingly, I would have expected the leakage to CPCh (through capacitive coupling) due to the transient high fault current to be essentially equal in L and N, thereby not resulting in any current imbalance in the RCD - but, of course, my 'expectations' may well be incorrect!

Kind Regards, John

 

[stillp]

my 'expectations' may well be incorrect!

They are! Bernard is correct. It can be very instructive to play about with a high-speed recording oscilloscope, and look at things like the inrush surge into anything with a transformer or with significant capacitance. There can be a surprising amount of asymmetry for a few milliseconds.

 

[JohnW2]

my 'expectations' may well be incorrect!

They are! Bernard is correct. It can be very instructive to play about with a high-speed recording oscilloscope, and look at things like the inrush surge into anything with a transformer or with significant capacitance. There can be a surprising amount of asymmetry for a few milliseconds.

... but that is surely only if the 'load' is very reactive (which arcing in a dying lamp/bulb isn't)?

Kind Regards, John

 

[lizzy]

To Answer the earlier question, it isn't a repo. It was a late 80's self build by an old couple that looks like it hasn't been touched since. Well built mind. 48 hours later and no further problems yet!

 

[stillp]

surely only if the 'load' is very reactive (which arcing in a dying lamp/bulb isn't)?

Not really. At the high frequencies generated in the current transient, it only needs a little inductance or capacitance.

 

[JohnW2]

surely only if the 'load' is very reactive (which arcing in a dying lamp/bulb isn't)?

Not really. At the high frequencies generated in the current transient, it only needs a little inductance or capacitance.

Fair enough - I probably should not have included the word "very". However, my point remains that you need a reactive load to get transient L/N current asymmetry, don't you?

Kind Regards, John

 

[stillp]

Yes, but at high enough frequencies almost anything has enough reactance, even a straight length of wire.

 

[JohnW2]

Yes, but at high enough frequencies almost anything has enough reactance, even a straight length of wire.

As a general statement, that's obviously true, but I'm not sure that an arc would represent. Furthermore, I haven't really got any idea as to what sort of frequencies we might be talking about - have you?

Whatever, all of this is really purely academic speculation, which may actually have nothing to do with the phenomenon we are discussing. The fact is that some people do seem to have witnessed RCDs operating in response to the death of a bulb/lamp - and we can but speculate as to the mechanism.

Kind Regards, John

 

[stillp]

and we can but speculate as to the mechanism

Or, as Bernard has done, we can make a logical deduction.

 

[JohnW2]

and we can but speculate as to the mechanism

Or, as Bernard has done, we can make a logical deduction.

aka speculation! In prior discussions about this, others have speculated about other possible mechanisms.

Kind Regards, John