Can anyone explain why a failing (tungsten) bulb should trip the main circuit switch, and not just the lighting circuit protection?
There is no way that the broken filament can arc internally to the bulb's bayonet/earth is there?
I presume you mean the RCD rather than switch?
Any other lights on the same circuit at the same time? If something with a capacitor etc it's possible that caused a current imbalance when the single pole MCB trips.
I probably do mean that. It looks like a 'switch'.
Just a chandalier type of light fitting with 4 of those candle-shaped bulbs. I can understand a spike might get generated when the filament ruptures, but why should that trip the entire house supply. It's not the first time this has occurred.
David J
Was the bulb pointing upwards or downwards? If it's pointing upwards, bits of broken filament could short out the two solid wires going to the base, and that could melt a wire outside the bulb and it might touch the earthed base. I'm sure I've had bulbs blow a fuse even when hanging downwards.
When a filament is simply broken, I have a party trick where I plug the bulb in and rotate it so the broken ends touch, and magically the bulb works again! It works for quite a few hours, good for an emergency when there are no new bulbs.
What do you exactly mean by "main circuit switch" and "lighting circuit protection"?
The arc in a expiring tungsten light bulb is very low impedance so there is quite a surge in current, this can easyly trip an MCB.
RCD trips are also possible depending on how high the existing leakage is. These days of everything having SMPSU's and filter components between all three mains wires the standing leakage through an RCD serveing a couple of rings etc can be quite high and thus the RCD only needing a little push to trip it.
When the filament fails it usually does so at one point only. As it fails the gap is initially small enough for an arc to form and be maintained for a very short time drawing a very high current. Incandescent bulbs have a built in fuse (a thinner section of wire) designed to burn out if this happens..
During the fractional time this takes to happen the current drawn by the bulb can often be over 100A for a few cycles and very unstable. This is not of sufficient duration to trip the 5A MCB but the very fast rise in current and rapid fluctuation in current causes a variety of normally insignificant capacitive and inductive effects in the wiring to generate a current imbalance in the supply wires for a fraction of a second.
As an RCD will trip in 2 cycles (40mS - 10 times or more faster than an MCB will react) the RCD will therefore often trip when a bulb blows. Some light fittings will usually be more susceptible to this than others depending upon the wiring layout in the house so it isn't unusual to find one lamp fitting that will nearly always trip an RCD when a bulb fails.
Sorry - I'm using technical terms very loosely here. The box appears to have individual circuit protection (just like the old fusewire days) plus one device that covers everything. I guess that is the RCD...
Thanks..
If it has a test button then its a RCD
Something like the DIN rail mounting one shown here:
It sounds as if you have a "sensitisation" issue.
See:
Ah, further to my other post, you also have the "whole house RCD" problem described in the article. These setups can be a PITA. Now the price of RCBOs have fallen so much, it might be worth investigating changing the RCD for a simple switch and swapping each of the MCBs for RCBOs (assuming you can get a type that fits your CU). That way you have full discrimination - only the circuit with the fault will trip.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.