Adding sockets to a ring circuit

[diyman34834]

I am currently renovating the living room on the ground floor and would like to make some changes to the socket layout in this room.

There are currently two double sockets in this room. The first socket is directly connected to the ground floor ring circuit, while the second one is spurred off the first.

I would like to have three double sockets in the room, so effectively adding one more double socket to the circuit. (While I am at it I thought I might also `unspur' the spurred socket so that it is directly part of the ring circuit.)

Just wanted to double check that this will be okay to do on this ring circuit. I have a EICR from when I first moved into the property, so can provide information as needed.

TAI

 

[ericmark]

Assuming the ring is not already at maximum length then yes you can extend it. We are allowed 106 meters in a single ring. This equates to 0.61Ω difference in the ELI between Zs and Ze assuming no parallel earth paths and 2.5mm line and 1.5mm earth.

Really you should measure the line - neutral impedance which would be 0.44Ω between the incoming readings and the mid way socket on the ring. However although we should measure both often only the Zs is recorded so you should see on the paperwork Ze something like 0.35Ω and Zs something less than 0.94Ω it is the difference that matters not the reading. At least until 1.37Ω it was 1.44Ω.

Also all new cable and sockets must be RCD protected.

Without that EICR I would say use either a RCD FCU or FCU to extend with fused spurs as once you use a fused spur you are very unlikely to exceed the limits.

To split the ring to extend either replace the back box for a twin instead of double then you can terminate each end of ring into single socket and use the two points to extend from or use a 4 way grid plate with two sockets which will fit a standard double socket box but that way you loose the switches on the sockets.

Of course once complete you will need to re-test and inspect and fill in the minor works certificate as with any job but knowing the start readings you should know it will pass (or fail) rather than cross your fingers.

 

[Iggifer]

I think the chances of your ring being even remotely close to 106m if you only have two sockets in the room is between 0 and 0.001% so I wouldn't worry about that concern.

I guess it depends how many rooms this ring feeds. I'm assuming maybe two or three rooms tops. Do you have a separate ring for the kitchen?

You are more than likely fine to add a couple of sockets and bring the spurred socket back into the ring. Just make sure you maintain it as a ring and connect the correct legs to the spurred socket.

I'd probably add a couple more sockets than you think you need - future proof!

 

[deadshort]

Perfect answer ! And not reg number in sight


Regards,

DS

 

[ericmark]

I remember a chippy asking me for advise. A year latter I visited his house which was a rambling affair and I realised my advice was rather flawed.

As an electrician using meters every day we often estimate and will look at a job and realise it will likely be fine, on the edge or likely already over the limit and when we realise unlikely to exceed will take a chance.

So yes likely no problems. Even when the volt drop is over the recommended figure in real terms it's unlikely to cause a problem. With RCD protection even exceeding the ELI is unlikely to cause a problem.

However the posted asks the question is it OK and to answer the question we tend to quote what is allowed with regulations rather than what one can get away with.

 

[diyman34834]

Thanks for the responses.

Sounds like I should be okay to go ahead then.

I don't think that the circuit is anywhere near 106 metres in length. The circuit services 3 rooms and has 9 sockets on it.This does include the kitchen, but not the cooker, which is on its own circuit.

The circuit is protected with an RCD (I had a new consumer unit installed when I bought the property).

In terms of the EICR, the circuit has 2.5mm live and 1.5mm wiring, and the measured Zs is 0.21 and the maximum permitted Zs is 0.87. (Couldn't find any Ze readings on the report)

What did you mean by

connect the correct legs to the spurred socket.

 

[ericmark]

You are more than likely fine to add a couple of sockets and bring the spurred socket back into the ring. Just make sure you maintain it as a ring and connect the correct legs to the spurred socket.

I will guess he is referring the the figure of 8 problem. To ensure load is shared on the ring it has to be one ring not two. If you guess and guess wrong where a spur is fed from you can end up with instead of one ring a figure of 8 arrangement. It is one of the things we should test for but it is often missed.

The inspection and testing course was for electricians and ran for 12 weeks at 3 hours a week. You can't expect to learn how to inspect and test from a forum. About the best you can do is plug in one of these

and follow the instructions that come with it.

Since the meter shown reads 0 ~ 1.7 Ω with a single light and you really want to measure between 0 ~ 1.5 Ω it clearly will not find all faults but it should highlight any blatant errors.

 

[JohnW2]

I will guess he is referring the the figure of 8 problem. To ensure load is shared on the ring it has to be one ring not two. If you guess and guess wrong where a spur is fed from you can end up with instead of one ring a figure of 8 arrangement. It is one of the things we should test for but it is often missed.

Interestingly, as has been discussed before, although it presents some nightmares in relation to testing, a "figure of 8" circuit is, in most senses, probably actually 'safer' than a single ring. In particular, any loads in the downstream of the two rings will (assuming a symmetrical arrangement) be shared more-or-less equally between the two legs of the upstream ring, even if those loads were very badly distributed (e.g. all on one side of) the downstream ring. There is also even more 'redundant protection' against broken CPCs.

Kind Regards, John

 

[diyman34834]

OK, I see. That makes sense.

I can actually see most of the wiring for the downstairs ring circuit from the basement, so I can see the layout of the circuit directly.

Thanks again.

 

[JohnW2]

OK, I see. That makes sense. I can actually see most of the wiring for the downstairs ring circuit from the basement, so I can see the layout of the circuit directly.

Does that mean that you are satisfied that it is a real, single, ring?

Kind Regards, John

 

[diyman34834]

I had a EICR done when I first purchased the property. That stated that it was a ring circuit, and I've not made any changes to the circuit since.

I can see all of the ring circuit relating to the living room and the dining room (I can't see parts of the circuit in the kitchen). The living room is where I am making the changes to the circuit, so I am confident I will not introduce a figure of 8 into the circuit.

 

[JohnW2]

I had a EICR done when I first purchased the property. That stated that it was a ring circuit, and I've not made any changes to the circuit since. ... I can see all of the ring circuit relating to the living room and the dining room (I can't see parts of the circuit in the kitchen). The living room is where I am making the changes to the circuit, so I am confident I will not introduce a figure of 8 into the circuit.

Fair enough - and ("between you and I" ), as I recently wrote, I'm not really convinced that a figure-of-8 is that much of problem, anyway - except for people trying to test it!

Kind Regards, John

 

[ericmark]

I will guess he is referring the the figure of 8 problem. To ensure load is shared on the ring it has to be one ring not two. If you guess and guess wrong where a spur is fed from you can end up with instead of one ring a figure of 8 arrangement. It is one of the things we should test for but it is often missed.

Interestingly, as has been discussed before, although it presents some nightmares in relation to testing, a "figure of 8" circuit is, in most senses, probably actually 'safer' than a single ring. In particular, any loads in the downstream of the two rings will (assuming a symmetrical arrangement) be shared more-or-less equally between the two legs of the upstream ring, even if those loads were very badly distributed (e.g. all on one side of) the downstream ring. There is also even more 'redundant protection' against broken CPCs.

Kind Regards, John

As you say likely a figure of 8 will not present a danger.

I have like yourself often wondered what is the problem in both figure of 8 and taking spurs using cable rated at 32A (i.e. 4 or 6 mm²) but to use the British ring system we are required to abide by the rules.

We can only guess what was going through the minds of the regulation writers? I will guess the problem arises where the point where the connection is made is where the loading would be unequal on the legs of the primary ring.

Taken from the centre of the ring as you say be it a figure of 8 or a 6 mm² cable feeding for example a garage it will likely not cause an overload.

But the rules for conductors in parallel are different to that of a ring final and if we want to take advantage of the ring final system without doing all the complex calculations then we must follow the rules laid out.

If you are prepared to do the calculations then I suppose nothing to stop a figure of 8 but I remember doing Norton's Theorem and Thevenin's Theorem in University and I would hazard a guess these are beyond the ken of most DIY guys and in real terms they have to follow the rules for the ring final and simply accept them as I for one would not like to try and explain the Theorem's to the normal DIY'er.

If you want to explain them I will not stop you but even if you do what are the chances the normal DIY'ed will understand them?

 

[JohnW2]

As you say likely a figure of 8 will not present a danger. ... I have like yourself often wondered what is the problem in both figure of 8 and taking spurs using cable rated at 32A (i.e. 4 or 6 mm²) but to use the British ring system we are required to abide by the rules.

The 'rules' don't actually say that you can't do either of those things. OK, the guidance in Appendix 15 does not include them, but that is only 'guidance', and one can't expect one page to cover all possible situations - they are just 'examples'.

I will guess the problem arises where the point where the connection is made is where the loading would be unequal on the legs of the primary ring.

As I said in my earlier post, a figure-of-8 will, if anything, improve the equality of loading in the two legs of the upstream ring.

If you are prepared to do the calculations then I suppose nothing to stop a figure of 8 but I remember doing Norton's Theorem and Thevenin's Theorem in University and I would hazard a guess these are beyond the ken of most DIY guys and in real terms they have to follow the rules for the ring final and simply accept them as I for one would not like to try and explain the Theorem's to the normal DIY'er.

I haven't 'sat down and done it', but I don't think one needs any particularly clever maths (or necessarily any maths at all) to demonstrate that a true 'figure-of-8' can never result in a worse situation (in terms of currents or current balance) than the equivalent single ring. Whilst 'the BS7671 ring final' involves assumptions/dispensations that only 'they' can make (since standard electrical calculations would preclude such a circuit), but if one can demonstrate that one's design is 'no worse' (in terms of currents) than their design, then I reckon that's good enogh!

Kind Regards, John