Another question on spurs

[JohnW2]

I can see the SCR that fires and the contacts that open circuit the Live and Neutral but I cannot see the symbol for a coil or other means of operating the contacts.

Very good point - when I saw the SCR with its gate connected to the output of the IC, I subconsciously assumed that it was 'doing the obvious thing' and didn't look carefully enough at the diagram! In fact, not only is there no obvious connection to anything which operates the contacts, but it looks as if the only thing which will happen when the SCR fires is that the power supplied to the electronics will (if the polarity is correct) fall to the Vf of the SCR!

I think I need to ignore this 'Application note' and go back and look at the actual animal again. Unfortunately, some of the components of that RCD got largely destroyed by the 'incident', but I'll do my best!

Kind Regards, John

 

[JohnW2]

I can see the SCR that fires and the contacts that open circuit the Live and Neutral but I cannot see the symbol for a coil or other means of operating the contacts.

Very good point - when I saw the SCR with its gate connected to the output of the IC, I subconsciously assumed that it was 'doing the obvious thing' and didn't look carefully enough at the diagram! ... I think I need to ignore this 'Application note' and go back and look at the actual animal again. Unfortunately, some of the components of that RCD got largely destroyed by the 'incident', but I'll do my best!

OK, I think I've got it 'sussed'. The electronics of my dissected RCD are more-or-less identical to that diagram, with one crucial difference - there is a contact-operating solenoid between the neutral supply and the bridge rectifier. There is therefore normally very little current flowing through the solenoid (just the quiescent current of the IC) but when the SCR fires, it 'shorts' the output of the bridge rectifier and thereby produces a path for enough current to flow to operate the solenoid. ... does that sound reasonable?

Kind Regards, John

 

[bernardgreen]

Yes that is a way to achieve the function with minimum component count.

 

[JohnW2]

Yes that is a way to achieve the function with minimum component count.

Indeed. Having ascertained how at least one RCD works, we are obviously left with a number of questions.

Firstly, since this electronic operation of an RCD is contrary to what both you and I (and probably others) expected, I wonder how common it is. Given that the RCD I dissected is a 'cheap and possibly nasty' (Protek) one, one suspects that they wouldn't have used electronics unless this was a pretty common, or even ubiquitous, practice.

Given that at least some RCDs clear do rely on electronics, I wonder where that leaves eric's suggestions. There presumably has got to be an L-N voltage below which the electronics simply won't successfully trip the device. Although the capacitor across the IC's power may well keep the IC's voltage 'high enough for long enough', there has presumably got to be a level of L-N voltage pd below which the solenoid will not function, even if the SCR fires. Hence, it would seem that a very low impedance fault very close to the RCD might well not result in operation of the RCD. That's the opposite of what eric was suggesting, but it's still a situation in which there could be a problem. What do you think?

Kind Regards, John

 

[bernardgreen]

My initial opinion is using a few cheap electronic components and a non precision mechanism is a much easier way to get high yield production.

The non electronic method requires a mechanism assembled with care using precision manufactured mechanical components. The level of precision and care in assembly will mean high manfacturing costs and a higher percentage of items that fail on test.

An electronic method also means several different types of RCD ( trip current and time delay ) can be manufactured using the same mechanism and PCB with only a change of one or two components needed.

 

[JohnW2]

My initial opinion is using a few cheap electronic components and a non precision mechanism is a much easier way to get high yield production. The non electronic method requires a mechanism assembled with care using precision manufactured mechanical components. The level of precision and care in assembly will mean high manfacturing costs and a higher percentage of items that fail on test. An electronic method also means several different types of RCD ( trip current and time delay ) can be manufactured using the same mechanism and PCB with only a change of one or two components needed.

Yes, that all sounds possible. I guess the next step would be to look inside some expensive ones!

On the opposite side of the equation, I suppose that a non-electronic one is reliant on the mechanical properties and behaviour of the 'precision mechanism' remaining unchanged for potentially many years - a consideration that is much less relevant to with 'electronic' ones.

Kind Regards, John

 

[terryb630]

OK gentlemen many thanks, that certainly got some responses!
As I understand it I need a 13A double pole passive rcd protected fcu, the switch will be inside off an rcd protected ring main, the cable will go through the wall conduited 18 inches under two paths so unlikely to get rotavated, it will surface next to the summer house and go through its wall. Once inside it will feed two double sockets in parallel, since the fcu will be fused at 13A and the lights will not draw much as they are UV-B heat lamps (eco style ) the only problem will be ensuring that the heater doesn't exceed 2KW, since it's only there to provide background warmth that shouldn't be a problem.
Have I got that right?

 

[JohnW2]

As I understand it ... conduited 18 inches under two paths so unlikely to get rotavated ...

That bit will probably generate some discussion. Do I take it that you are talking about ordinary 'win and earth' cable? What sort of conduit?

Kind Regards, John

 

[terryb630]

Yes 2.5 T&E with steel conduit, would that be ok do you think?

 

[ban-all-sheds]

As I understand it I need a 13A double pole passive rcd protected fcu, the switch will be inside off an rcd protected ring main,

you will need an RCD FCU if the house circuit is not already RCD protected.

 

[ban-all-sheds]

Yes 2.5 T&E with steel conduit, would that be ok do you think?

How will you earth the conduit?

Forget conduit - use armoured cable.

 

[JohnW2]

...Having ascertained how at least one RCD works, we are obviously left with a number of questions. ... Firstly, since this electronic operation of an RCD is contrary to what both you and I (and probably others) expected, I wonder how common it is. Given that the RCD I dissected is a 'cheap and possibly nasty' (Protek) one, one suspects that they wouldn't have used electronics unless this was a pretty common, or even ubiquitous, practice.

I wonder if anyone has any fairly recent experience of dissecting 'reputable brand' RCDs? The only ones I have (MK & Wylex) which are not in service are brand new and unused, so I don't want to take them apart, but I'd be very interested to know whether the current/recent 'upmarket' ones are using electronics to operate the trip mechanism.

Kind Regards, John

 

[terryb630]

To earth the conduit I shall fit a clip around it and run a lead to the earth point on the back box which of course in its turn is earthed through the normal wiring system.
I did wire in a shed years ago using armoured cable, not the easiest medium I thought, the cable had to run the length of the garden some 12 yards, because the people who had built the house hadn't buried the drains deep enough it was not possible to dig it in the required 18 inches and ended up around 10 to 12 inches deep, because of this I conduited it as well, belt and braces you understand, I felt cost was less an imperative to safety, the other problem with armoured cable is I haven't got any, and the run is only 6 foot, and I think a steel conduit coming out of a wall going straight down into the ground is more pleasing to the eye than armoured cable, and it's easier to seal to the wall, but mainly because I haven't got any.
Ok why a rcd/fcu? because then any fault within the summerhouse will not trip the whole house, and it's belt and braces again, I guess I'm that sort of bloke, cautious!

 

[JohnW2]

To earth the conduit I shall fit a clip around it and run a lead to the earth point on the back box which of course in its turn is earthed through the normal wiring system. ... the other problem with armoured cable is I haven't got any, and the run is only 6 foot, and I think a steel conduit coming out of a wall going straight down into the ground is more pleasing to the eye than armoured cable, and it's easier to seal to the wall, but mainly because I haven't got any.

You'll probably continue to see some argument about that, with people trying to persuade you to use armoured cable. The latter is certainly the ideal, but I think it would probably be difficult to find a particularly compelling argument against what you propose - at least, in terms of electrical common sense.

Ok why a rcd/fcu? because then any fault within the summerhouse will not trip the whole house, and it's belt and braces again, I guess I'm that sort of bloke, cautious!

You may have misunderstood or misread ...

you will need an RCD FCU if the house circuit is not already RCD protected.

If the circuit is already RCD protected in the house, then you do not need an RCD FCU. Indeed, if you addeed one, there is absolutely no gurantee that a fault in the summerhouse would not trip 'the whole house' (at least, everything protected by the RCD in question) as well as (or instead of) the one in the summerhouse.

Kind Regards, John

 

[terryb630]

Ok I think I understand, on the previous job I used a rcd and a fuse, but I see your point because on the earlier one there was no rcd protection on the mains only a cb, hence on that I used as stated a rcd, armoured cable to a fuse box in the shed, the belt and braces approach again! so on this one only a fcu is required, fair enough.
Many thanks to all for the wealth of knowledge you have so generously shared with me and the time you have taken.

Gratefully Terry B