Strange goings on with a wall lamp.

[JohnW2]

On TN installations working with only single pole isolation seems to be widely accepted practice even if not strictly ideal.

Indeed this is correct. Only on TT installations is it imperative to isolate both poles.

I've never fully understood the logic (even though, unusually, it's actually stated!) of BS7671 about this. In the absence of faults, there obviously ought never be a dangerous voltage between N and E within an installation. However, if one is going to consider the possibility of faults, then I would have thought the the only (almost) 'totally safe' situation for SP isolation would be with TN-C-S, since the N and E are actually joined within the installation. However, BS7671 seems to think that TN-S is also OK for SP isolation (which it will be, unless there are supply-side faults).

Kind Regards, John

 

[JohnW2]

Touching neutral and earth on a single pole isolated circuit is almost certain to trip the RCD.

That's not really true - in fact, it's unlikely to trip an RCD. If one considers an impedance through a person of, say, 1kΩ, then to get a 30mA fault current would require a 30V N-E potential difference at the point of contact, which is exceedingly unlikely. Given that, in the absence of faults, there is no current though (hence no VD along) either the N or CPC conductors of the isolated circuit back to the CU, there would have to be a 30V N-E difference at the CU for that situation to arise.

Kind Regards, John

 

[plugwash]

Sorry I meant touching them together, not touching them with your hands.

 

[JohnW2]

Sorry I meant touching them together, not touching them with your hands.

Oh sorry - I misunderstood! Yes (as I'm sure that virtually all of us know from bitter experience!) actually 'connecting' N & E together on an SP-isolated circuit will, if there are any loads on the installation at all, virtually always cause an RCD to trip.

I have to say that I'm a little surprised that this is always the case with TN-C-S (maybe it's not - most of my experience in recent years has been with TT!), since the only scope for any N-E potential difference at the CU (and there has to be some N-E pd, even if only tiny, for thereto be any 'fault current' to trip the RCD) is due to the voltage drop in the tails between CU and cutout.

Kind Regards, John

 

[88v8]

IMO it is pretty crazy that RCDs were thrown into our electrical systems without providing a way to isolate circuits in a RCD compatible manner making the installation trickier to work on and meaning that a single fault in a final circuit can render the whole installation unusable. SPSN MCBs exist but are almost unheard of in the UK.

Exactly.
A so-called safety device that plunges the whole house into darkness and crashes the computer because I touch two wires on a 'dead' circuit, it just plain daft. I've even had this happen with a blown bulb.

If ever I have nothing else to do - unlikely - I might just install a system that works. It would involve, errrm, porcelain fuse holders in a porcelain carrier, on a polished oak plinth with a mica cover.
Rather like the setup that worked fine for 31 years in our previous house.
There'll be one on eBay eventually. :}

 

[ban-all-sheds]

A so-called safety device that plunges the whole house into darkness and crashes the computer because I touch two wires on a 'dead' circuit, it just plain daft. I've even had this happen with a blown bulb.

It is not daft.

But your continued protestations about it are doing nothing but make you appear more and more that.

If ever I have nothing else to do - unlikely - I might just install a system that works. It would involve, errrm, porcelain fuse holders in a porcelain carrier, on a polished oak plinth with a mica cover.

FGS.

How many times do you have to be told that fuses vs MCBs makes no difference whatsoever before it sinks in?

 

[stillp]

porcelain fuse holders in a porcelain carrier, on a polished oak plinth with a mica cover.

Don't forget the asbestos supports for the fuse wire...

 

[JohnW2]

IMO it is pretty crazy that RCDs were thrown into our electrical systems without providing a way to isolate circuits in a RCD compatible manner making the installation trickier to work on and meaning that a single fault in a final circuit can render the whole installation unusable. SPSN MCBs exist but are almost unheard of in the UK.

Firstly, for my education, is their some subtle difference between SPSN, SPN and DP - or are they all synonymous?

There was an interesting thread recently in which an OP was expressing concerns (concerns I'd never heard voiced before) that, similarly to DP fusing, having DP switching/isolation carried the risk that the switch could fail with L contacts closed (e.g. welded) but N open with the switch in the 'off' position - hence giving the impression of isolation, but with L still connected to the load. As I said, I've never heard such concerns voiced before, and I imagine that the chances of it happening are greatly outweighed by the advantages of DP switching.

A rather daffy friend of mine was so annoyed/concerned about the issue you mention above that he installed a CU enclosure above his CU, fully populated it with DP 'main switches' and ran each of the final circuits through one of those switches. For radial circuits, this was no real issue, other than the introduction of a couple of 'unnecessarily' connections. Ring final circuits technically became 'lollipops', since there was a single cable (I imagine 4mm²) from MCB to the external 'main switch' (where the two legs of the ring originated). Given that DP MCBs are not only "almost unheard of" in the UK, but also expensive if one finds them, I imagine that this was a much cheaper option than using DP MCBs.

I suppose the 'idealists' would like to see a CU full of DP RCBOs - but, at least at present, that would be a very expensive option.

Kind Regards, John

 

[plugwash]

IMO it is pretty crazy that RCDs were thrown into our electrical systems without providing a way to isolate circuits in a RCD compatible manner making the installation trickier to work on and meaning that a single fault in a final circuit can render the whole installation unusable. SPSN MCBs exist but are almost unheard of in the UK.

Firstly, for my education, is their some subtle difference between SPSN, SPN and DP - or are they all synonymous?

AIUI SPSN means a device which switches both line and neutral but only has overcurrent detection in the line conductor.

 

[ban-all-sheds]

SPSN vs DP - my understanding is that (for MCBs), the former monitors the L current, but switches both poles, the latter monitors and switches both poles.

And SPN & SPSN are synonymous, unless you postulate an MCB with N passing through it but not being switched.

With RCD/RCBO, there can't be any distinction between SPSN & DP for the RCD function, but there could be for overcurrent detection.

One could build SPN RCBOs and RCDs, but I've never heard of the latter.


As an aside - I saw a photo recently of an RCBO which had two levers, so you could tell why it had tripped.

 

[JohnW2]

AIUI SPSN means a device which switches both line and neutral but only has overcurrent detection in the line conductor.

Ah - that makes sense. Thanks. I'd never really thought of this before - are you saying that a true DP MCB (some manufactures seem to talk of both DP and SPSN ones) senses overcurrent in both conductors? I have to say that one would have to postulate a pretty bizarre faults to result in overcurrent in N but not L.

Kind Regards, John

 

[JohnW2]

SPSN vs DP - my understanding is that (for MCBs), the former monitors the L current, but switches both poles, the latter monitors and switches both poles. And SPN & SPSN are synonymous, unless you postulate an MCB with N passing through it but not being switched.
With RCD/RCBO, there can't be any distinction between SPSN & DP for the RCD function, but there could be for overcurrent detection. One could build SPN RCBOs and RCDs, but I've never heard of the latter.

That all makes sense. Thanks.

As an aside - I saw a photo recently of an RCBO which had two levers, so you could tell why it had tripped.

Interesting. I've often thought that would be a good idea. It can be a bit of a pain not knowing what has tripped an RCBO.

Kind Regards, John

 

[ban-all-sheds]

 

[securespark]

porcelain fuse holders in a porcelain carrier, on a polished oak plinth with a mica cover.

Don't forget the asbestos supports for the fuse wire...

Extinguishing pads, no less.

 

[88v8]

A so-called safety device that plunges the whole house into darkness and crashes the computer because I touch two wires on a 'dead' circuit, it just plain daft. I've even had this happen with a blown bulb.

It is not daft.

But your continued protestations about it are doing nothing but make you appear more and more that.

If ever I have nothing else to do - unlikely - I might just install a system that works. It would involve, errrm, porcelain fuse holders in a porcelain carrier, on a polished oak plinth with a mica cover.

FGS.

How many times do you have to be told that fuses vs MCBs makes no difference whatsoever before it sinks in?

You can tell me the earth is flat as well while you're at it :}

But I can tell you that working on a circuit where you've pulled the fuse, you can rattle the conductors together like castanets, and it does not trip any other circuit, nor, oddly enough, does it trip the 100A main fuse.

Unlike the supposedly better installation here, where switching off the relevant circuit MCB and then touching the wires together is likely to result in the main RCD tripping and switching everything off. Especially good when you're in the attic.

I've even had bulbs blow, and the main RCD trips instead of the circuit MCB.

Daft is the kindest I can call it.