A few random questions, if you will..

[cjard]

While trying to work out the logic of why rings were devised post war:
Surely, rings only saved copper if the distance between..the last socket and the CU (or fuse board as it would have been then) was dramatically shorter than the trip out to it?
Which got me wondering - would it not save more copper to split the ring in a sensible place (longest socket-socket bridge, farthest from the CU) and save that leg of copper instead - have 2 radials? Or is that the situation where those 2 radials would for sure have had to go in in 4mm2, and for the cost of joining the radials to form a ring, 2.5mm could be used, so yes, more copper is used than 2x 2.5mm radials, but less copper is used than 2x 4mm radials?

Next q:
My TF company are recommending a "service void" - I understand the reasoning but it seems like a large amount of room to lose for the convenience of easily running a few cables in some small percentage of the volume created by the void. Sure I could fill most the void up with insulation and then it have a practical benefit, but might it just be better to chase a route into the celotex that will line the house inner, put an oval duct in it, and run the cables in that? (the air tightness membrane would go on the outside(world) side of the duct rather than having the duct come through the AT membrane

Suppose I do go for a service void; the tf co are advising the battens to form it run vertically. This means for ease of reaching a socket the wires would circulate in the gf ceiling/ff floor void (there is no gf floor void unless I duct the gf screed) and drop down the walls to sockets on the ground floor? Isn't this incredibly wasteful on wire to have to loop down the wall and back up it for every socket? Might it be better to make the service void have horizontal runs of battens at socket height and just run the electrical wiring horizontally around the room to each socket? The SV can have vertical battens everywhere else no probs (and it helps pinch the air tightness membrane glue joints) but socket height and wall lamp height (SWMBO hates overhead lights) the sv would be better horizontal, yes?


I've long wondered why, when wires visit each socket i9n turn, are the conductors cut through? Physics lessons of old told me that resistance occurs at a surface, so would it be better in volt drop terms if the wire that came into the back of a socket was continuous with the wire that left - fold it, strip it and poke the folded bit into the terminals of the socket, but don't cut it. I'm guessing theyre cut because overall the contribution to voltage drop is minimal, but the inconvenience of trying to neatly strip insulation off a wire that hasn't been cut through is high and would slow installation considerably?

 

[plugwash]

While trying to work out the logic of why rings were devised post war:
Surely, rings only saved copper if the distance between..the last socket and the CU (or fuse board as it would have been then) was dramatically shorter than the trip out to it?
Which got me wondering - would it not save more copper to split the ring in a sensible place (longest socket-socket bridge, farthest from the CU) and save that leg of copper instead - have 2 radials? Or is that the situation where those 2 radials would for sure have had to go in in 4mm2, and for the cost of joining the radials to form a ring, 2.5mm could be used, so yes, more copper is used than 2x 2.5mm radials, but less copper is used than 2x 4mm radials?

The question is does a ring provide more or less utility than 2x20A radials.

And i'd say it provides more, with the ring you can have two 3KW heaters at once at fairly arbitary locations (the exception being you can't have both very near the same end) plus a smattering of small loads. With the radials you can also have two 3KW heaters plus a smattering of small loads but you are far more restricted in where you can put things.

 

[JohnD]

I've long wondered why, when wires visit each socket in turn, are the conductors cut through?

Many years ago, people were taught to strip the cores and bend them into a "U" then flatten so they went into the socket terminals. Yes, it does improve continuity, but it is quite fiddly and slow, and you need to get the amount of cable in each box quite carefully measured, and to cut the sheath and insulation with care.

AFAIK nobody does it like that any more.

 

[ericmark]

First do remember the ring was designed without having finger protectors on the live pins and with 7/0.029 cable not 2.5mm so not quite the same today as when designed.

But the idea is to provide sockets around the house where the permitted volt drop and the required maximum earth loop impedance will not be exceeded. One big point is although the fuse/MCB may be 30/32 amp the design current is taken as 26 amp. That's worked out at 20 amp in centre and 12 amp even spread. So when working out volt drop one is working on 26A.

However any radial will also be worked out at 20A at furthest point and remaining even spread so two 20A radials will be calculated at 20A each so 40A in total.

So to remove the centre link and make a ring final into two 20A radials increase the current from 26A to 40A and so we start to have problems with volt drop.

To comply with volt drop a ring final using 2.5mm cables has a limit of 106 meters but a radial using 2.5mm and 20A MCB is limited to 26 meters OK drop to 16A and 34 meters but not 53 meters as one might expect before doing the calculations.

Even when not taking the ring final to the limits it still have less volt drop and a better earth loop impedance than you would get with a radial.

There are other advantages with out system which have nothing to do with being ring or radial which is one fuse/MCB can protect 50 sockets but with the 16A system used in Europe although they will not have a 16A MCB to every socket they will likely have four times the number of MCB's to what we use.

The building regulations have reduced the gains in using a ring final with the 1/3 rule when drilling beams which has increased the amount of cable used but unless tampered with and as a result the ring is lost it is still the best system.

 

[JohnW2]

The question is does a ring provide more or less utility than 2x20A radials.

That's one question but, if one makes that comparison, the 'increased utility' of a ring really relates to the fact that one is comparing 20A and 32A circuits. As you say, a 20A circuit is quite restrictive, because it won't even 'take' 2x13A loads.

However, the other (and maybe a more fair/relevant) question would relate to the comparison between a 32A ring and a 32A (4mm²) radial. 'Utility'-wise thy are then equivalent, although the radial will usually use more copper. [not inevitably - if the sockets are arranged in roughly a straight line radiating from the CU, a 4mm² radial could actually use less copper].

Kind Regards, John

 

[JohnW2]

One big point is although the fuse/MCB may be 30/32 amp the design current is taken as 26 amp. That's worked out at 20 amp in centre and 12 amp even spread. So when working out volt drop one is working on 26A. ... However any radial will also be worked out at 20A at furthest point and remaining even spread so two 20A radials will be calculated at 20A each so 40A in total.

Although you often mention it, I'm not sure that we actually know where (other than people's mouths) this design current of 26A for a 32A ring final (or 32A radial) actually comes from. However, as I've recently written, the difference you are talking about only exists because you are comparing a 20A circuit with a 32A one, not because one is radial and the other a ring - i.e. it would be equally true if you were comparing a 20A radial with a 32A radial.

Kind Regards, John

 

[ericmark]

There are variations on the humble ring. The lollipop where a heaver cables takes supply to where required and then the ring starts and also where the ring supplies an array of junction boxes and each box has a spur so getting rid of the need for twin cables down the walls. I have done this in a warehouse with 10 meter high walls but not really required with standard domestic.

As to routes it's more to do with fire regulations than electrical wiring. Using fire walls within a duct is far easier than individual fire barriers. Insulation can cause cables to over heat and putting cables together in a bunch can cause both higher and lower temperatures in the cables according to loading. The problems of crossing an escape route are also helped when cables are run together.

We use to consider an escape route as just escape but now we are told we must consider fire fighters returning later to fight the fire and the dangers when cables fall and get entangled in the fire fighters BA gear it seems there have been some deaths as a result of cables falling.

Clearly a house, and a high rise set of flats are very different also I have never worked out why 450 mm in a domestic and 400 mm in non domestic do people grow when they get home then shrink again when they go out?

The chimney effect is very strong. I did heat treatment and was amazed at the pressure with just 30 odd foot that would blow off my bungs and once started it was near impossible to stop. OK quite hot 600 to 750°C but it is the same problem with open stair cases any vertical shaft has to be carefully designed or it will have a gale blowing through it.

We of course only really see the problems when it goes wrong. 5 psi seems a lot of pressure for walls to stand but it was because of it all going wrong that flats with gas supplies need to withstand that pressure. So I leave it to the designer and just do as I am told.

 

[ericmark]

One big point is although the fuse/MCB may be 30/32 amp the design current is taken as 26 amp. That's worked out at 20 amp in centre and 12 amp even spread. So when working out volt drop one is working on 26A. ... However any radial will also be worked out at 20A at furthest point and remaining even spread so two 20A radials will be calculated at 20A each so 40A in total.

Although you often mention it, I'm not sure that we actually know where (other than people's mouths) this design current of 26A for a 32A ring final (or 32A radial) actually comes from. However, as I've recently written, the difference you are talking about only exists because you are comparing a 20A circuit with a 32A one, not because one is radial and the other a ring - i.e. it would be equally true if you were comparing a 20A radial with a 32A radial.

Kind Regards, John

The information was given at one of the many IET lectures I have attended I questioned where the 106 meters came from. I have not seen it written in any official paper and I agree it's easy for rummer control to take over.

But it does make sense to expect the maximum from any single point to be 20A and this is only for volt drop not the ELI.

To be frank back in the 1980's I was told one reel of cable only for a ring but never really worked out why. At a 0.35 Ω incomer and a 1.44 Ω maximum for ELI you need somewhere around 195 meters before there is a problem with ELI. Even with 0.8 Ω incomer still 115 meters so although we may not have realised why we have worked on the 85 meters maximum for a ring final and to again to be frank until the volt drop went from 4% to 5% so the maximum length went from 85 meters to 106 meters I did not even think about it.

 

[JohnW2]

Although you often mention it, I'm not sure that we actually know where (other than people's mouths) this design current of 26A for a 32A ring final (or 32A radial) actually comes from.

The information was given at one of the many IET lectures I have attended I questioned where the 106 meters came from. I have not seen it written in any official paper and I agree it's easy for rummer control to take over.

That's what worries me a bit. I agree that (albeit arbitrary) it makes some sense, and that it is only of relevance in relation to calculation of VD, but I've never seen it written anywhere.

Kind Regards, John

 

[flameport]

Surely, rings only saved copper if the distance between..the last socket and the CU (or fuse board as it would have been then) was dramatically shorter than the trip out to it?

Compared to 20A radials, there is little or no saving in cable.

However when rings were designed in the 1940s, BS1363 outlets did not exist and instead there were many different types of socket outlets in use, where the rating of the outlet also determined it's physical size.

In order for one room to have outlets suitable for different appliances, two or more circuits would be required, with two or more different types of sockets, for example 15A and 5A. This arrangement would use far more cable than a ring. It would also require larger fuseboxes due to the number of circuits.

 

[stillp]

I was told some time ago, by someone who was involved with the first discussions about the ring final, that the original intention of the committee was to achieve greater safety by using redundant CPCs and having a lower fault loop impedance. The original design used a larger CSA cable. it was only later that someone suggested that the cable size could be reduced

 

[JohnW2]

I was told some time ago, by someone who was involved with the first discussions about the ring final, that the original intention of the committee was to achieve greater safety by using redundant CPCs and having a lower fault loop impedance.

Yes, I've heard that suggested before. The worry with that seems to be that should the CPC ring become broken, no-one would be aware of it but the EFLI would increase considerably. [similarly, if either L or N rings became broken, it would again not be 'apparent', but the cable could become overloaded].

Kind Regards, John

 

[stillp]

Isn't the same true of a radial?

 

[JohnW2]

Isn't the same true of a radial?

I suppose so, in the case of the CPC. A break in it would obviously result of total loss of a path to earth downstream of the break, which I suppose is even worse than the corresponding situation with a ring. In the case of the L & N, there clearly is no such problem - a break in one of those will stop any current being drawn from downstream and, in any event, the cable would (should) be sized to take the full circuit current anyway, so theoretically cannot be 'overloaded', no matter what happens.

Kind Regards, John

 

[ericmark]

The whole concept of "What if some one does not obey the rules" bugs me. Be it we need to reduce to speed limit because some one doing 65 MPH in a 40 MPH limit caused an accident or some one worked on a ring final without testing and inspecting missed it was not a ring, the problem is the same, you can't make rules for some one not obeying the rules.

I look at Europe with reversible line and neutral, Using cables rated at 6 amp on 16 amp supplies and thank my lucky stars I live in the UK.

OK they would use 5 core cables not 4 core with a earth wire strapped to the outside and they adopted RCD protection well before us. But all in all I agree with Flounders and Swan well maybe not English I would say British but to quote "The English the English the English are best the English are best and to hell with the rest" Well OK not quite as blatant but yes we have worked out a system which protects our inhabitants as long as we follow the rules.

There are of course exceptions be it the ring or the ceiling rose there are times when the other system excels but on average ours works out the best.