switched socket ring?

[ban-all-sheds]

Now price that up and compare it to the cost of just having more sockets arround.

Absolute fortune.

It would be utter madness to do it.

 

[SimonH2]

I could see this workable with a SPDT switch as a separate item next to each socket/group of sockets. You'd need a ring with two lives and one shared neutral (not a problem - read on). At the CU you have a contactor with takes a feed from the MCB/RCBO feeding the live feed down one live and selectively feeding it down the other live. You need a contactor with a pole for each circuit to be done this way.

Why a contactor, rather than just an appropriate switch?

Just an assumption that this would be a "going to bed/going out - switch everything off" type of thing, and CUs tend not to be located in handy places.
You're right, it could be done with just a suitably rated switch.

Also not sure how you'd fare with rings and the wiring regs once you deviate from the "standard" setup - but you could always use radial circuits.

I think the potential problem would arise if people with knowledge of the regs but a lack of basic understanding tried to apply the regs. In engineering terms, assuming the switches were adequately rated (and to BS 1363 - which could be a problem) and the cable CCC at least 20A, for any given set of settings of the switches, one would simply have two 'standard' rings, sharing a neutral, fed from the same MCB/RCBO - for which I can't see any prohibition in the regs.

I was thinking more about the issue of having what would become a spur from the switch onwards, with multiple 13A sockets on it. Ie, up to the switch you have a ring, then from the switch to sockets you have a spur. If you have 2xdouble sockets, would you want a separate switch for each double ?
Seems to be an argument for running both rings behind the sockets and moving the wiring as required - with the "not used for this socket" ring looped through with a choc block. That would also be cheaper !

 

[JohnW2]

Why a contactor, rather than just an appropriate switch?

Just an assumption that this would be a "going to bed/going out - switch everything off" type of thing, and CUs tend not to be located in handy places. You're right, it could be done with just a suitably rated switch.

Agreed that a contactor might be more convenient - and probably more practical given what I'm about to write....

I was thinking more about the issue of having what would become a spur from the switch onwards, with multiple 13A sockets on it. Ie, up to the switch you have a ring, then from the switch to sockets you have a spur. If you have 2xdouble sockets, would you want a separate switch for each double ?

I haven't had time to think about this properly, but if the contactor (or switch) switched L's of both ends of the rings, wouldn't the two circuits both remain as rings (albeit with a common N) throughout?

Seems to be an argument for running both rings behind the sockets and moving the wiring as required - with the "not used for this socket" ring looped through with a choc block. That would also be cheaper !

Indeed so.

Kind Regards, John

 

[SimonH2]

I was thinking more about the issue of having what would become a spur from the switch onwards, with multiple 13A sockets on it. Ie, up to the switch you have a ring, then from the switch to sockets you have a spur. If you have 2xdouble sockets, would you want a separate switch for each double ?

I haven't had time to think about this properly, but if the contactor (or switch) switched L's of both ends of the rings, wouldn't the two circuits both remain as rings (albeit with a common N) throughout?

Sorry, bit of confusion there !
Yes, the master switch/contactor would switch the common ends of the ring to the supply.

But at each socket, you've effectively a spur point at the individual switch - so the L (and probably N) for the group of sockets fed by that switch. If it's one switch, one socket (or one double) then no problem. What if it's 2 or 3 double sockets in a group - is that a problem ? Presumably you could use 4mm cable for the spur wiring, but you've still got a potentially large load attached at a point on the ring. Seems to me that unless you do calcs to ensure you don't do this too close either end of the ring, then you couldn't use a normal RFC. And if you do do those calcs and deal with it, what about someone coming along later and modifying the system ?

Still sounds a lot of hassle to me, when you can easily walk round on your way to bed and switch things off !

 

[JohnW2]

I haven't had time to think about this properly, but if the contactor (or switch) switched L's of both ends of the rings, wouldn't the two circuits both remain as rings (albeit with a common N) throughout?

Sorry, bit of confusion there ! Yes, the master switch/contactor would switch the common ends of the ring to the supply.

Right. As I said, I wrote before having had a chance to think too deeply! Firstly, at the CU end, one would obviously have to switch both ends of each of the rings (probably L only) - otherwise the entire circuits would be 'banjos', rather than rings.

But at each socket, you've effectively a spur point at the individual switch - so the L (and probably N) for the group of sockets fed by that switch. If it's one switch, one socket (or one double) then no problem.

Right. I now see what you mean,and agree that there is an issue - I hadn't thought of this before. If one wanted to maintain a true ring structure (without spurs), then one would have to have two cables going to each socket and, again, switch (the Ls of) both of them (to one circuit or the other) - so the switches would then be doing exactly the same as if one had true rings and physically re-wired. However, I think that the switching would also have to 'complete' the ring that wasn't being used for the socket (probably by connecting another adjacent socket to the other ring) - which would require 4PDT switches, just for switching the L's! I really don't see the need to switch the N's, since there is only one N conductor for the two rings.

What if it's 2 or 3 double sockets in a group - is that a problem ? Presumably you could use 4mm cable for the spur wiring, but you've still got a potentially large load attached at a point on the ring.

Indeed. We've discussed that one before, in relation to 'normal' rings. I have pointed out that a 4mm² unfused spur supplying multiple sockets (on a 32A/2.5mm² RingFC) would be OK, CCC-wise. As you say, the only real issues with that are (a) point loading of ring and (b) that since such an arrangement is not explicitly mentioned in BS 7671 (not even in the 'informative' Appendix 15) some people would be uncertain about its compliance without doing calculations from first principles.

Seems to me that unless you do calcs to ensure you don't do this too close either end of the ring, then you couldn't use a normal RFC. And if you do do those calcs and deal with it, what about someone coming along later and modifying the system ?

Even if one does do the calculations, they are only as good as the assumptions one makes about loads - so, in the absence of any official/recognised figures to use for those assumptions, there probably would not be any guarantee that the results of one's calculations would be acceptable to others.

The 'simple' answer (for a switched arrangement) would, as you say, be to switch each socket separately (as a 'spur'), rather than to have any switching of 'groups of sockets';. In fact,I think that's what was being suggested.

Kind Regards, John

 

[ljarrald]

i've not read all of the posts so forgive me if its already been suggested...

could he use 2.5MM triple and earth? using one core as permanent live, one as switched live, and one as neutral?
then just wire it as a ring through all the backboxes, connect the permanent live to the sockets that he wants to be live 24/7, and the same with the switched live in the switched sockets.

its not too difficult to change the config in the future then.

 

[JohnW2]

i've not read all of the posts so forgive me if its already been suggested... could he use 2.5MM triple and earth? using one core as permanent live, one as switched live, and one as neutral?
then just wire it as a ring through all the backboxes, connect the permanent live to the sockets that he wants to be live 24/7, and the same with the switched live in the switched sockets. ... its not too difficult to change the config in the future then.

Yes, it's already been suggested, and discussed. One issue is that 2.5mm² 3C+E is not available as standard T&E, so one would have to use some other type of cable. The same obviously could be done with two separate 2.5mm² T&E cables (and then without the need for a common neutral) - and that would probably be one of the simplest approaches.

Kind Regards, John

 

[ljarrald]

i've not read all of the posts so forgive me if its already been suggested... could he use 2.5MM triple and earth? using one core as permanent live, one as switched live, and one as neutral?
then just wire it as a ring through all the backboxes, connect the permanent live to the sockets that he wants to be live 24/7, and the same with the switched live in the switched sockets. ... its not too difficult to change the config in the future then.

Yes, it's already been suggested, and discussed. One issue is that 2.5mm² 3C+E is not available as standard T&E, so one would have to use some other type of cable. The same obviously could be done with two separate 2.5mm² T&E cables (and then without the need for a common neutral) - and that would probably be one of the simplest approaches.

Kind Regards, John

ah right, so triple and earth only comes in 1.5mm? that's a shame.

i did think about two rings in and out of each socket, but you'd have to have very deep backboxes for that.

 

[JohnW2]

ah right, so triple and earth only comes in 1.5mm? that's a shame.

Yep - well, 1.5mm² and 1.0mm². There's not really any common 'normal' use for anything larger.

i did think about two rings in and out of each socket, but you'd have to have very deep backboxes for that.

Probably, particular given that (at least, after any change) the ends of the 'other' cables would have to be terminated in something. However, deep back boxes do exist.

Kind Regards, John