Ring final with 4mm legs

[kingeri]

So I have been working on a four storey house, in which the wiring is pretty sound, the sparks who did the last rewire (approx 2004) seems to have loved 4mm T+E, all the sockets are on 4mm radials, with either 16, 20 or 32 amp breakers. The exception is the very top floor sockets, which he did as a ring final, with 4mm 'legs' and 2.5mm looped between sockets. I can only think he did this to combat voltage drop, as as far as I can see installation methods would have allowed 2.5 to be used throughout the circuit. Anyway, I know this is fine, just wondered if its a method anybody else has used over the years?

 

[ruston]

Yes , probably his calculations for the fourth floor have indicated that he needs a larger CSA for that floor.

 

[Octopus]

Or simple lack of commutation between the people pulling the cables!

 

[telectrix]

as here:

https://www.youtube.com/watch?v=yBBWUZfgRiw

 

[7029 dave]

Trying to use up his knock off 4mm t&e.

 

[Marvo]

Yeah, probably due to the cable he had on had rather than by conscious design. A ring final circuit will run with a far higher load on one leg if it's made up of different CSA cables. Also a bit dodgy for someone upping the MCB in future if it's not clearly labeled in the CU. Not something I'd personally install.

 

[kingeri]

Yeah, probably due to the cable he had on had rather than by conscious design. A ring final circuit will run with a far higher load on one leg if it's made up of different CSA cables. Also a bit dodgy for someone upping the MCB in future if it's not clearly labeled in the CU. Not something I'd personally install.

Both 4mm legs are the same length and follow the same route, they are approx 40m (by calculation). The loop between them in 2.5mm is around 15m, serving two bedrooms and ten double sockets. I can't see there being a problem with the ring being unbalanced here, as such. I wouldn't install it either (why not just do it all in 4mm?) though!

 

[Leesparkykent]

Assuming a load of 20A at the furthest point and the balance to the rating of the protective device evenly distributed so for a 32A protective device this equates to a load of 26A at the furthest point. mV/A/m x ib x L /1000 /4 so in your case 18x26x95/1000/4=11.115V. Would of been close to the recommended maximum value of 5% (11.5V) using 2.5mm. But TBH I could never see a load of 26A being used on the top floor of a house unless they have a cannabis factory up there :goofy:. I would of probably done it on a 2.5 RFC as would of complied for volt drop and its what I would of had on the van....Im not the biggest fan of 32A 4mm radials but accept there are times its warranted to use such a circuit. For one the maximum length would be around 32-33 meters if you took the design current to be 32A.

 

[R G]

The use of 4mm2 T&E along with 2,5mm2 T&E is nothing new , it was commonly used for spurs from the ring circuit , probably someone using the old ideas here..

 

[EastAnglian]

In new designs (i.e. new builds & rewires) I will usually run radials rather than RFCs. 4mm cables for 32A circuits and 2.5mm cables for 16/20A circuits. I believe that a larger number of short radial circuits is a better design than a smaller number of RFCs - partly because of the improved load distribution between circuits but mainly because it avoids the hidden danger of a ring with a single discontinuity in a conductor. I have discovered sufficient discontinuous rings on 32A & 30A MCBs/fuses in my time to regard these as unsafe practice.

 

[Leesparkykent]

In a new build I struggle to see how you would get 4mm T&E to comply on a 32A circuit breaker unless its supplying a fixed load that pulls no where near 32A.

 

[EastAnglian]

In a new build I struggle to see how you would get 4mm T&E to comply on a 32A circuit breaker unless its supplying a fixed load that pulls no where near 32A.

Comply with which parameter specifically?

 

[Leesparkykent]

Comply with which parameter specifically?

current carrying capacity.

 

[EastAnglian]

Yes - fair point in a new build other than 'clipped direct' reference method ...... I must learn not to generalise when posting

 

[kingeri]

How about an unfused spur wired in 4mm but serving 2 double sockets, off a ring final wired in 4mm? Found this today. I don't think its actually dangerous, but does it comply? Ring final is about 20m total length and spur about 3m.

 

[EastAnglian]

It certainly doesn't comply with Figure 15B of the Informative Annex 15.

 

[kingeri]

True enough but could it be regarded as a 4mm radial, after all you could remove one leg from the DB and that's what it would be.

 

[HandySparks]

It certainly doesn't comply with Figure 15B of the Informative Annex 15.

Maybe not, but it's as safe as a radial wired in 4mm² in the same location.

 

[kingeri]

We must have been thinking along the same lines at the same time there!

 

[EastAnglian]

Ah yes, I just re-read the original and see its a 4mm RFC - doh! In which case I would agree with the above!

 

[Risteard]

How about an unfused spur wired in 4mm but serving 2 double sockets, off a ring final wired in 4mm? Found this today. I don't think its actually dangerous, but does it comply? Ring final is about 20m total length and spur about 3m.

I have come across this arrangement a number of times recently (although with much more than two double sockets on it).

I would not in any way be happy with this arrangement as it has the potential to overload one leg of the ring unless coming right from the centre of it.

 

[Adam W]

I appreciate the reasoning behind it, but I wouldn't describe the situation as "fine". "Acceptable" maybe, but mixing different cable sizes in the same circuit seems to be inviting confusion. A reduced size for one spurred socket maybe, but say for example a chippy drilled into a cable, the customer didn't want their laminate flooring lifted and the only acceptable way to get power on is to split the ring into 2 radials, you could end up with what's effectively 2x 4mm² radials with several sockets spurred off in 2.5mm² without being fused down.

I wouldn't install a ring in this way out of choice either, and I certainly wouldn't install a radial like this. It seems like one of those "no other option" solutions - I don't see why the same size cable couldn't have been used throughout.

 

[7029 dave]

Must admit I have never mixed cable sizes on circuits, and don't see the point of it either.

 

[kingeri]

I have come across this arrangement a number of times recently (although with much more than two double sockets on it).

I would not in any way be happy with this arrangement as it has the potential to overload one leg of the ring unless coming right from the centre of it.

But how can one leg be overloaded if both legs can each take >32A and the OCPD is 32A, if it were a straightforward 4mm radial then a single cable would be taking all the load at all times anyway.

 

[Risteard]

But how can one leg be overloaded if both legs can each take >32A and the OCPD is 32A, if it were a straightforward 4mm radial then a single cable would be taking all the load at all times anyway.

I thought we were talking about a 2.5mm^2 ring final circuit with 4mm^2 conductors taken off it as an unfused spur. Apologies if I have misunderstood.

 

[kingeri]

I thought we were talking about a 2.5mm^2 ring final circuit with 4mm^2 conductors taken off it as an unfused spur. Apologies if I have misunderstood.

The ring is entirely in 4mm and the unfused spur is also 4mm. Cable lengths are short enough that it would comply on Zs and VD even with one return leg removed. Kinda makes the ring pointless.
I totally agree with you that if the ring was wired in 2.5mm and then an unfused spur serving multiple points but wired in 4mm was installed, then that would certainly be an awful design with the potential to overload one leg of the ring.

Risteard, what are your views on my OP, which was a different setup again?

 

[Risteard]

Risteard, what are your views on my OP, which was a different setup again?

If I'm now understanding this correctly, there is a mix of 2.5mm^2 and 4mm^2 legs within the one ring circuit. Both legs from the distribution board are in 4mm^2 and 2.5mm^2 is then used between socket outlets?

It's pretty rough I guess, but assuming the 4mm^2 conductors are of a similar length it shouldn't unduly imbalance the ring although this could still happen to an extent with the smaller csa conductors away from the centre point of the ring.

 

[Peter Monfort]

It would surely have been better to make it into two radials. These could have been supplied by two 4mm^2 radials and the 2.5mm ^2 could have been arranged in a variety of Spurs or radials. I think this circuit is more commonly known as a tree in other countries. It would also simplify testing as well as reduce the number of potential errors that ring circuits suffer from. The circuits could be protected at 20amps or alternatively in a commercial application you might protect at 30amps and use a timer control to zone sections of 4mm^2 radials or ring circuits in 2.5mm^2. There is a really good chapter in the book electrical installation designs which covers the history of the ring final circuit and more modern thinking on the use of tree circuits.