Bathroom notifiability

[JohnW2]

Indeed. Any passive RCD (such as the one I tested) will obviously 'hold in' (stay 'on') with any supply voltage (including zero) - but, yes, with a 6V supply it tripped immediately with a ~30mA imbalance. For no real reason other than curiosity, I'll be interested to see how low a voltage will still enable tripping, but I'm probably going to have to do some improvisation to get a 50Hz supply lower than 6V. Watch this space.

A small update. My 'improvisation' has so far enabled me to get down to a supply voltage of 1.7 V, and at that supply voltage the RCD still trips reliably with a ~30mA imbalance.

Being persistent, I'll probably try to go down further, but I'm already so close to zero voltage that it is apparent that, at least for this particular RCD, there is effectively no lower limit to the supply voltage at which it will function satisfactorily.

Kind Regards, John

 

[ban-all-sheds]

How long before the internal impedances will stop you getting to 30mA?

 

[JohnW2]

How long before the internal impedances will stop you getting to 30mA?

I can't see that happening. I wouldn't use a source that couldn't supply well over 30mA, and the 'internal impedance' of the RCD (essentially just 2 or 3 inches of substantial wire) is going to be negligible - don't forget that it has to be able to 'safely' carry 80A.

I can't see that I can be all that far away from the minimum. There has to be rectification to produce a supply for the electronics, and I doubt that they would use anything other than a silicon diode for that, and I would think that the circuitry would need at least 1V or thereabouts - so that would require an absolute minimum of around 1.6V (peak). At 1.7V RMS, I'm already down to about 2.4V peak.

Kind Regards, John

 

[JohnW2]

I can't see that I can be all that far away from the minimum. There has to be rectification to produce a supply for the electronics, and I doubt that they would use anything other than a silicon diode for that, and I would think that the circuitry would need at least 1V or thereabouts - so that would require an absolute minimum of around 1.6V (peak). At 1.7V RMS, I'm already down to about 2.4V peak.

It seems that I was probably wrong ...

.... I'm now getting down to the limits of resolution of my measuring equipment, but I'm now down to only a 0.7V RMS (about 1V peak) supply voltage. With a 33Ω 'fault' resistor (about 21.2 mA) it doesn't trip, but with a 22Ω one (about 31.8 mA) is still does trip!

Assuming the measurements are roughly correct, that means that unless they are using something 'fancy' like a Shottky diode to rectify, the electronics must somehow be managing with appreciably less than 0.5V.

I'm really not convinced that it's worth trying to get any lower!

Kind Regards, John

 

[John D v2.0]

Can you do a tear down? Would be interesting to see if there are any recognisable components in there.

 

[JohnW2]

Can you do a tear down? Would be interesting to see if there are any recognisable components in there.

I'd prefer not to do so with this one, unless you'd like to pay for it - it's brand new, unused, and sitting on my shelf . However, I'll see if I can find one that I'm more comfortable about opening up.

Kind Regards, John

 

[rsgaz]

Can you do a tear down?

This might interest you, skip to 12:55...

 

[JohnW2]

This might interest you, skip to 12:55...

Many thanks.

If my Wylex one is similar, that goes some way to explaining how my Wylex one may work - it's not what I was expecting at all. I was expecting some electronics (probably including an amplifier) powered by a supply-derived 'power supply' of some sort - not just a thyristor triggered directly from the sense coil (the trigger voltage from being derived from the sensed imbalance current, regardless of the supply voltage - indeed, seemingly the same even in the absence of any supply voltage).

It's not the right time of day to be trying to think, but it's not immediately obvious to me why they've had to include the bridge rectifier, rather than just using a thyristor (which would give half-wave rectification) or, perhaps better, a triac directly in the path of the release coil.

As that design stands, there appears to still be the forward voltage drop of two (at any point in time) of the IN4007 bridge rectifier diodes (around 1.2V total) plus the forward voltage drop of the MCR 100-8 thyristor (seemingly about 0.6V at room temp) in the path between the supply and the release coil. That would seemingly be far too much total forward voltage drop for it to work with a supply of 0.7V RMS, which is what I seem to be seeing with my Wylex - so, if I'm thinking straight, mine cannot be quite like that ... or maybe I'm thinking all wrong or totally missing something at this unearthly time of day!

I've got to disappear now, but will give more thought to this when I'm back later in the day.

Kind Regards, John

 

[JohnW2]

Can you do a tear down? Would be interesting to see if there are any recognisable components in there.

As I said, I'm hesitant to dissect the brand new Wylex RCD I've been experimenting with. However, some of you may remember, from four or so years ago, my daughter's RCD which blew up when she touched the 'test' button ....

The electronics of that (Protek) RCD looked like this ....

I've just dusted off that PCB and had a quick look at it. It is more like what I expected than the one in the video, in that central to it is a VG54123 IC, which contains an amplifier, together with a voltage regulator, reference voltage source and a latching circuit - and, as in the video, has a thyristor/SCR (BT169D) triggered by the output of the IC.

The IC seems to be powered by a bridge rectifier using four discrete silicon diodes (again, similar to the video). I haven't yet attempted to trace the circuit, but it looks as if, as in the video, there are two silicon diodes and the thyristor/SCR in the supply path - which, if true, again means that there would be too much forward voltage drop across those devices for it to work with a 0.7V RMS supply (as my Wylex seems to do).

More after I've looked at the PCB in more detail.

Kind Regards, John