Thanks for that, but in practical terms I don't care if the cause was an earth leakage or a short circuit; I'm just relieved to know that the trip almost certainly happened because of a bulb failure and not something more serious.
Especially not on mine which are Edison screw. Yes you are right it has to be a plasma discharge dead short as the final puff of tungsten vapour shorts out the incoming leads. Spotlights seem to do it far more often than normal bulbs. Annoying thing is kitchen and loft lights are on the same circuit - and guess where the fuse box is.
When the lamp shorts, the fault current is very high, and it's limited by the supply impedance. This generates a very sudden step change in the voltage, and this step has lots of high frequency components. These high frequencies will capacitively leak quite easily, and most particularly through any mains interference suppressor in an earthed appliance which is plugged in. Ideally, an RCD will have designed in a short time delay so that spike leaks don't trip it, but this may well exceed it's allowance. After all, it's going to trip something, so it doesn't much matter what it trips.
Smaller lamps are less likely to have space for an internal fuse, which are always included in GLS lamps.
It would explain a RCD trip, but not an MCB trip generally.
In practical terms you probably would... if it caused earth leakage then in some installations an RCD would trip rather than the circuit MCB, and that may take out more than just the one lighting circuit ;-)
Indeed. The simplest plausible explanation normally wins ;-)
My worst offenders are golfball lights round a mirror.
If it's possible, it always wise to have the nearest lights to the CU on a separate circuit from the next nearest ones. Mine is at the top of the cellar stairs, so easy to have it on a different circuit from the hall lights.
If you actually understand the difference, might make fault finding easier when one next occurs.
It's nothing new. When I had re-wirable fuses, I've had one blow when a bulb failed.
Ah, true, but I meant *in this particular instance*.
Three hearty cheers for William of Ockham.
You have such appliances on your lighting circuit? ;-) It was, after all, just the lighting circuit which tripped.
It is annoying, though. Even with the knowledge a MCB does a better job of protecting things than a fuse.
According to the agent that manages the house we rent out, it was squirrels that did for the shower cable that runs through the loft. So 6, maybe even 10mm2 cable.
In article , Andrew Gabriel scribeth thus
Interesting theory, has anyone done any research into that?.
I know of a location where they had a lot of computers on the one ring main and the thing was tripping almost all the time what with the suppression capacitor leakage...
That's why you shouldn't use final ring circuits in a commercial environment. Should be radials. Rings are for the average domestic environment. Same sort of reason as you don't put permanent heating in a house on a ring.
that wont stop the RCD tripping mate.
So your statement is of course wrong.
neither will it stop temporarily overload on any MCB if 50 pluss SMPSU equipped computers are attached to it.
So your statement is doubly wrong.
Should be radials.
Doesn't help one iota.
Rings are for the average domestic
completely different argument.
You wouldn't have one RCD protecting the whole lot either.
You split them into a number of circuits.
It is the same argument. The correct wiring and protection for the load.
Except the protection isn't for the load. Its for the wiring.
No, a RFC is perfect for a suite of computers, they are a fixed load and need high leakage current protection, which the RFC CPC will give.
If there was a radial feeding the same equipment, you would need an extra CPC cable, have a lower cable capacity for the same cable size, and take just as long to install it. Testing would be slightly quicker, but that is done once every 5 years or so, so isnt really a consideration.
If the One RCD is in use no matter how many radials or rings are downstream of that its still the same as far as the Capactive leakage matters....
If more than the One RCD or if RCBO's are in use then the number of Capactive leakage devices that each RCD or RCBO device "supplies" will alter with the number of them (cap leakage devices) on the downstream circuits that each RCD or RCBO device supplies...
>
Make that the permissible load, then.
Of course you probably think it impossible to have a load of SMPS in the same building without everything tripping?
Not sure I follow the logic there...
Firstly the circuit topology has (practically) no effect on the total earth leakage - it will simply be a multiple of the number of filters in the appliances on the circuit.
Secondly for circuits of this type (i.e. those with high normal leakage currents) the wiring regs require the use of high integrity earthing. That is as a minimum ensuring that the earth connection *is* wired as a ring so as to lower the likelihood of a failing connection resulting in disconnection of the earth and a potentially dangerous direct contact risk that may result.
No, its unrelated.
Rings are ideal at flexibly supporting large diverse and dispersed loads. High point loads simply remove some of that flexibility. Hence if you know in advance that you have such loads (which you normally do) there makes little sense putting them on circuits optimised for other needs.
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.