Do RCDs actually work?

If that's an accurate guess, I'm happy, if a little surprised.

I've never looked inside an RCD but it seems likely that either they use a very sensitive relay mechanism, sensitive enough to detect small current imbalances between two coils, or they use some electronics to amplify those imbalances. Either method would seem to be prone to failures.

OTOH how many ordinary MCBs (which almost never get tested) would turn out to be faulty? Moving parts are less reliable than fuses. Until someone fits the wrong fuse, of course.

Now here is the wierdest thing. On Monday I took the RCD apart and had a look inside. The thing has been sat on my compurer desk since then, After reading your post I pressed the test button - the RCD tripped and will now not reset. How odd.

There are some electronics in the RCD but I will put this one down to mechanical failure judging by the dust that has now come out of it.

OTOH how many ordinary MCBs (which almost never get tested) would turn

How many MCBs get tested?

I am surpised at the lack of suggestions (by certain people) that if an RCD trips then everyone immediately falls down the stairs breaking their necks because the lights have gone out. One of the problems of RCDs is that we have no idea how many lives or injuries they have saved because the RCD has tripped and the householder is not in "the slightest bit killed or injured".

I can happily go down stairs at night without putting the light on. Most live in towns where there is rarely total darkness.

Total darkness in a strange place would be a different matter.

I believe that rats behave the same way.

Robert

Well if you work back from the 20 or so killed a year - including those from "misuse of appliances". That does not leave much scope for deaths where there is a working RCD. Think of all the cases that have been reported over the years (MPs daughter in the kitchen with a live saucepan rack, kid playing with discarded plug and flex etc. I have not heard of any that would not have been protected by a RCD as yet.

Experienced a faulty RCD recently (1980's era MK). No indication it was faulty until the test button was prodded one day immediately prior to carrying out some minor wiring changes. Managed to get a secondhand direct replacement which was a damn sight easier to fit than modding the busbars to fit something more modern or replacing the entire consumer unit.

They probably should be more reliable than an RCD.

Resistance (or conductivity) of some things like your body, can vary with the applied voltage.

Which is why 500v Meggars are used when carrying out insulation resistance tests on a wiring installation.

The highest resistance of the body, is that of the skin, particularly hard dead skin. Wet skin obviously is more conductive, due to either sweating or having just taken a bath or shower. Which is why the regulations are much more demanding where there is likely to be moisture about.

Rod Speed used his keyboard to write :

Wrong! The steady DC voltage is 50v, the ringing voltage can be much higher, in part due to the inductance.

I worked in the exchanges and we worked live. No problems at all until the line rang :-)

ISDN is even higher (90V??).

Windmill submitted this idea :

I once read some research which suggested that 10mA was enough for some people, but I suppose it would depend upon your state of health. It stated that 30mA was a bit of a compromise between safety and nuisance tripping. Obviously if they trip too often, no one would fit them.

It happens that Robin formulated :

MK made one.

Proper way to eat leftover pizza, isn't it?

In an ordinary household, I'd guess none. Something like a hospital is hopefully another matter.

It could also be because 500+ volts might cause a spark to jump between two not-quite-touching conductors. Especially in a damp environment.

In my limited experience with RCD trips, heating devices are the worst offenders (one panel heater, one combination microwave's convection function, one toasty maker). As one of the experts said a while back.

Personally, I'd be happier if the limit was 5 mA, even if that did cause a few nuisance trips. But first you'd need to train people so that they knew what to expect.

Clue - what is the peak voltage of a 230V supply