Rewire questions

[JohnW2]

Go back to the 13th edition and the rules permitted an unlimited number of number of BS1363 sockets (in domestic premises) on a 30A ring, but a maximum of only six on a 30A radial (a double socket counting as two). Where was there any logic in that?

You tell me. What CSA cable for the radial were we talking about?

The 20A radial rules then were much more restrictive, specifying a maximum of either (a) two sockets located anywhere, or (b) three sockets all in one room not exceeding 200 sq. ft. and not a kitchen (again, a double socket counting as two).

If, per my comments, one takes the view that one cannot (should not) assume much about what might be plugged into a sockets circuit, I suppose that was not too unreasonable.

The 14th edition permitted a maximum of six sockets serving a single room not exceeding 300 sq. ft. and not a kitchen, otherwise limited to two as before. Then was it the 15th edition which changed it to an unlimited number of sockets within a specified area, or was that later?

The moment one gets beyond one or two double sockets, one is into the territory of crystal balls. It is rarely the case that installing more sockets will result in the load increasing, so it's all down to judgements/guesses as regards what the load "is likely to be". FWIW, I think the OSG now recommends a minimum number of sockets for each type/size of room, and contains other odd guidances, such as limiting the number of spurred sockets on a ring final to the number of sockets on the ring, and also limiting the length of those spurs (to one eighth of total ring length, or something like that??).

Kind Regards, John

 

[ban-all-sheds]

FWIW, I think the OSG now recommends ... limiting the length of those spurs (to one eighth of total ring length, or something like that??).

I've not been keeping up to date with what the OSG says, and I haven't even bought a green one, let alone yellow.

That would be a fun recommendation in my house - the ground floor is solid, with tiles and parquet, and every single downstairs socket apart from a radial in the kitchen is on a spur dropped down the wall from the first floor. The circuit, which is a ring, contains not one socket or FCU actually on the ring.

 

[JohnW2]

FWIW, I think the OSG now recommends ... limiting the length of those spurs (to one eighth of total ring length, or something like that??).

I've not been keeping up to date with what the OSG says, and I haven't even bought a green one, let alone yellow.

Same here, mine is also red.

That would be a fun recommendation in my house - the ground floor is solid, with tiles and parquet, and every single downstairs socket apart from a radial in the kitchen is on a spur dropped down the wall from the first floor. The circuit, which is a ring, contains not one socket or FCU actually on the ring.

Indeed. We've discussed that before and it is not an uncommon situation when there are solid floors. If my recollections of what the OSG says are correct, that means that the maximum number of sockets 'allowed' on such a ring would be zero!

Kind Regards, John

 

[bernardgreen]

Surely the solution is to bring the ring down to the socket and then back up to the first floor.

 

[EFLImpudence]

My question (although, unfortunately, I got it the wrong way round) was intended as humourous and sarcastic with the intent of showing the fundamental pointlessness of such a rule or restriction.

Subsequent posts list other similar peculiarities which have no bearing on the result.

 

[stillp]

I agree that it's very unlikely, but I gather (but have no proof) that the CPC redundancy was one of the things which convinced the then IEE to allow the introduction of ring finals.

I'm not sure I've ever seen anything to that effect. Where does that idea come from?

stillp, IIRC.

To me, it would seem a little odd to accept the earthing redundancy as being sufficient reason to override the drawbacks of rings, given that the IEE then and right through to the IET today obviously has never had a problem accepting single earthing conductors on radial circuits.

Yes, but CPC redundancy would obviously only have been a 'secondary' reason. I think it is generally accepted that the primary reason was the belief (right or wrong) that it would require less copper, at a time of a serious post-war copper shortage.

Kind Regards, John

I have mentioned it briefly in the past, so perhaps I should expand.

Back in the mists of time, when I was starting to get involved with standards sometime in the 1980s, I had a long dinner with an elderly gent who'd been part of the original group that developed the ring final concept. He was adamant that the brief for the study group was to develop the best domestic wiring system in the world. Among the problems they tried to overcome were: the inflexibility of the then common system of radials to each socket-outlet (usually one per room), the conductor being sized for the single socket; the voltage drop on sockets furthest from the fuseboard in the meter cupboard; the unavailability that could result from a single conductor breakage; the dependence of the ADS on a single cpc; the number of fuses required for a system using a separate fuse for each socket; and no doubt others that I've forgotten. The solution to these problems was the ring final, that had two sources of current at each accessory, as well as a duplicated cpc. A little later it was realised that the conductors being effectively in parallel, their CSA could be reduced somewhat. The rfc system was not used straight away, since as a nation we were a little busy disagreeing with Germany at the time. It was only later, when reconstruction began in earnest, that the Ministry of Works requested the installations committee of the IEE to look at reducing the amount of copper required by a further CSA reduction of the conductors, to about 2/3 of the CSA that would seem at first sight to be necessary. The risk of an open-circuit allowing the CCC of the conductors to be exceeded was considered to be acceptable, given the relatively small loads that were common at the time, and the considerable safety margin built in to the CCC tables for conductors.

 

[EFLImpudence]

That's very interesting but would I be wrong in thinking it really doesn't make much sense? - the basic principle, that is; not your post.

 

[JohnW2]

I have mentioned it briefly in the past, so perhaps I should expand.

Many thanks - that's very interesting and informative.

Kind Regards, John

 

[stillp]

That's very interesting but would I be wrong in thinking it really doesn't make much sense? - the basic principle, that is; not your post.

Which part?

 

[PBC_1966]

You tell me. What CSA cable for the radial were we talking about?

Minimum 7/.036 for the 30A radial, changed to 4 sq. mm with the metric 14th edition in 1970.

I think the OSG now recommends a minimum number of sockets for each type/size of room, and contains other odd guidances, such as limiting the number of spurred sockets on a ring final to the number of sockets on the ring, and also limiting the length of those spurs (to one eighth of total ring length, or something like that??).

Didn't you also note a while ago that the former 50 & 75 sq. m limits (for 20 & 32A radials respectively) which had been in place recently have now also been relegated to just guidance?

The rule about the number of spurs not exceeding the number of sockets on the ring itself goes back many years, although this also now appears to be just a recommendation.

 

[PBC_1966]

the unavailability that could result from a single conductor breakage

I'm really struggling to see how adopting a system in which a conductor breakage leaving the circuit such that the remaining cable carrying the full demand of the circuit could then easily be overloaded goes with a brief to develop "the best domestic wiring system in the world."

 

[JohnW2]

Didn't you also note a while ago that the former 50 & 75 sq. m limits (for 20 & 32A radials respectively) which had been in place recently have now also been relegated to just guidance?

I did, but I can't remember what the actually figures are.

The rule about the number of spurs not exceeding the number of sockets on the ring itself goes back many years, although this also now appears to be just a recommendation.

Well, it's not even in an ('informative') Appendix of BS7671 (only in the OSG), so I don't know whether it really counts as a 'recommendation'.

Kind Regards, John

 

[PBC_1966]

I did, but I can't remember what the actually figures are.

The figures I quoted are the ones which were definitely in the regulations themselves at one point: Max. 50 sq. m on 20A or 75 sq. m on 30/32A.

 

[stillp]

the unavailability that could result from a single conductor breakage

I'm really struggling to see how adopting a system in which a conductor breakage leaving the circuit such that the remaining cable carrying the full demand of the circuit could then easily be overloaded goes with a brief to develop "the best domestic wiring system in the world."

Try reading my post again.

 

[JohnW2]

The figures I quoted are the ones which were definitely in the regulations themselves at one point: Max. 50 sq. m on 20A or 75 sq. m on 30/32A.

Just checked. As you know, the regs themselves say nothing. Even Appendix 15 gives no 'recommendations'. It merely notes that, "historically", there were limits of 50 m² for 2.5mm² (20A) radials, 75m² for 4mm² (30/32A) radials and 100 m² for 2.5mm² (30/32A) ring finals.

Kind Regards, John