Odd RCD problem

[oldbutnotdead]

Doing some work on one of the lighting circuits, isolated via the RCD (rather than the MCB) by using the TEST button. All good, finished what I was doing, dropped all the other MCBs off, reset the RCD, then the individual MCBs and nothing working. Odd. Power to the top of the RCD, nothing at the bottom. With meter probes in the bottom of the RCD I found that by moving the switch very slowly I would get 240v at about half throw and then nothing at full throw (no feeling of innards clicking away or anything). Also when the RCD was (apparently) on the TEST button did nothing.
Eventually found that energising one of the MCBs and making sure there was a load on it caused the RCD to engage properly. Repeated the exercise (all MCBs off= RCD won't reset properly. no voltage and no TEST. Any MCB energised with a load on = RCD sets properly, TEST works properly). EDIT TEST also worked properly even if I switched off all the MCBs (so no load on the RCD)
Is this a feature of the model sir or a fault? CU is out of warranty alas but supplier (CED) still exists.

 

[JohnW2]

Doing some work on one of the lighting circuits, isolated via the RCD (rather than the MCB) by using the TEST button. All good, finished what I was doing, dropped all the other MCBs off, reset the RCD, then the individual MCBs and nothing working. Odd. Power to the top of the RCD, nothing at the bottom.

Are you sure that you're switching the RCD back on effectively? It is quite often case that they won't reset properly (after they have operated as the result of a fault or pressing the test button) unless you first push the lever all the way down (to the 'off' position).

Kind Regards, John

 

[oldbutnotdead]

Are you sure that you're switching the RCD back on effectively? It is quite often case that they won't reset properly (after they have operated as the result of a fault or pressing the test button) unless you first push the lever all the way down (to the 'off' position).

Kind Regards, John

Ahhh, hadn't thought of that- certainly looks as if the toggle is all the way down but.... will try that & see what happens. Ta

 

[JohnW2]

Ahhh, hadn't thought of that- certainly looks as if the toggle is all the way down but.... will try that & see what happens. Ta

You're welcome. Particularly in view of some of the other things you said, this may well not be the problem - but it is often the case that one has to push the lever firmly down as far as it will go before it will reset.

Kind Regards, John

 

[securespark]

Are you sure that you're switching the RCD back on effectively? It is quite often case that they won't reset properly (after they have operated as the result of a fault or pressing the test button) unless you first push the lever all the way down (to the 'off' position).

Once those "trip halfway" RCD's trip, the lever is disengaged from the switch mechanism until switched fully off then on.

You will not find a voltage appearing by manipulating the lever, nor will engaging an MCB cause the RCD switch mechanism to latch closed.


If it is working as the OP suggests, I strongly suspect a fault.

You could try ringing up CED to see if they will send a replacement. I cannot remember how long the warranty is on CED stuff, but I seem to remember it is quite long.

They're on 0208 5038500.

 

[flameport]

It's faulty - replace it.

 

[JohnW2]

If it is working as the OP suggests, I strongly suspect a fault.

That was also my first reaction, but it would be a very bizarre sort of fault....

.... if I understand the OP correctly, the RCD will reset provided there is some load on it, but not if there is no load. If there is no residual current associated with the load (i.e. no 'leak' in the load) then, at least as far as the sensing system is concerned, that is no different from 'no load'. I also presume that the latching into the 'on' position must be mechanical - or, at least, not dependent upon any current flowing through the RCD. I therefore somewhat struggle to understand how, even if the RCD is faulty, it could behave as has been described - but I guess that's just me!

Kind Regards, John

 

[mfarrow]

I fear dodgy mechanism coupled with coincidence. You do not know it is "reset", it's sounding like a high resistance connection when closed waiting to cause a fire. Replace it.

 

[JohnW2]

all MCBs off= RCD won't reset properly. no voltage and no TEST.
EDIT TEST also worked properly even if I switched off all the MCBs (so no load on the RCD)

I've just noticed - there appears to be a contradiction in the above. Not only that, but (in both cases) how could you 'test the TEST button" if the RCD could not be reset (when all MCBs were 'off')??

Kind Regards, John

 

[JohnW2]

I fear dodgy mechanism coupled with coincidence. You do not know it is "reset", it's sounding like a high resistance connection when closed waiting to cause a fire. Replace it.

There certainly seems to be something very odd going on, but I don't really understand what you suggest above. What makes you suspect a 'high resistance connection' and, in particular, how could a high resistance connection result in the RCD not being resettable when there was no current flowing through it??

Kind Regards, John

 

[securespark]

Suffice to say the RCD is fubar and needs changing pronto.

 

[JohnW2]

Suffice to say the RCD is fubar and needs changing pronto.

Probably - but I can't help feeling that we might be missing some important bit of information.

No-one has yet mentioned N-E faults which, at least theoretically, can cause an RCD to trip (or not be resettable) even when there are no loads on the RCD in question - although I admit that I would struggle to imagine how putting a load onto the RCD would then allow it to be reset!

Of course, we don't yet know whether the OP has yet tried pushing the lever right down before resetting.

Kind Regards, John

 

[securespark]

Pretty sure the CED RCD's only have a two-position mechanism.

 

[mfarrow]

There certainly seems to be something very odd going on, but I don't really understand what you suggest above. What makes you suspect a 'high resistance connection' and, in particular, how could a high resistance connection result in the RCD not being resettable when there was no current flowing through it??

"Loose" connection then, or whatever you want to call it; it's certainly not consistent. It may be low resistance when cold & at first but once a substantial current flows so will the smoke from the understairs cupboard.

 

[JohnW2]

"Loose" connection then, or whatever you want to call it; it's certainly not consistent. It may be low resistance when cold & at first but once a substantial current flows so will the smoke from the understairs cupboard.

I still don't understand - whether you talk of high resistance connections, loose connections or whatever.

It should be possible to 'reset' an RCD (push the lever up into the 'on' position) whether or not anything is connected to it (load and/or supply side) - so 'loose' etc. connections should not make any difference. If the device cannot be 'reset' when nothing is connected, one would normally automatically assume that the device was simply 'dead' (mechanically). However, the OP tells us that if a load is connected, then it magically becomes possible to reset it - which, to mind mind, is pretty bizarre!

Don't forget, as I said before, that the OP's problem arises specifically when there is (or should be) no current flowing though the RCD (at least, the L side) - so, again, loose / high resistance / whatever connections can hardly be blamed. I still can't get the idea of N-E faults out of my mind - although, as I said before, I don't see how/why loading the RCD could/should then 'mend' things.

Kind Regards, John