Adding sockets to ring main/add another ring

[Bill Payer]

Our house was built around the early to mid 60s (so has "modern" PVC twin & earth cabling) and has just one ring main feeding the whole house. We want to add in some extra sockets but obviously I don't want to overload the one circuit so I've had an idea that I want to run past you peeps here.

Most family/friends houses seem to be wired with an upstairs ring and a downstairs ring. My idea is to go to all the sockets in the upstairs rooms and basically pull the cables out of the sockets, back down to under the floorboards and joint them there (either with "traditional" junction boxes or with crimps/heatshrink sleeve), so that the continuity of the ring is preserved but it is now just serving the downstairs sockets, then run a new ring for the upstairs sockets, and adding new sockets to each ring where needed.

Is this OK?

TIA, Bill

[Tim W]

Bill Payer wibbled on Monday 19 October 2009 12:07

In principle it sounds fine. Crimps would be an excellent way to joint the cable provided that you have a proper crimping tool and are happy with how it's enclosed. Another way, if you are concerned about the accessibility of the joints is one of the new Hager Ashley maintenance free junction boxes:

have spring contacts and good cable clamps and IME (of the smaller lighting ones) *very* well made.

The only 3 flies in the ointment AFAICS (other may see more):

a) Is your cable in good condition after 40 years? Even PVC has a finite life.

b) Depending on how well you can see the cables being run, you have to watch out for the possibility of creating an incorrect ring by mistake (eg 2 radials or a figure of 8 ring). it would be worth testing as you go to ensure that the cable you are about to joint is going where you think.

c) Part P - whatever... But for something like this, it would be worth borrowing a Megger or similar and actually testing the resultant ring for peace of mind at every socket outlet. Then you can prove there are no insideous wiring errors (perhaps resulting from the original circuit, like a broken or loose conductor somewhere you can't see) and that the insulation is still good.

Another approach could be to break the original ring deliberately into 2 or more 20A radials - this might be simpler if it leaves you with a reasonable distribution, though your kitchen/utility room might scupper this with a large presence of high loads in one place?... There're no real restrictions on radial topology if you can be sure that 20A is enough for each circuit.

HTH

Tim

[Bill Payer]

Thanks for the quick reply Tim! :o)

You make good points there. I've taken some of the sockets off and all connections look to be clean and tight with no signs of burning/brittleness/overloading and I've had a couple of the floorboards up and can see no signs of any damage. Visually, everything looks OK.

Noted

Will do. I'll also test the existing cabling before starting any work to supplement the visual as in your (a) above.

Hmm, that's a thought. Thanks Tim.

Bill

[Owain]

Sounds like too much work.

Why not just find the approximate mid point of the ring, split it there, and run 2 cables from the CU to there, so you have 2 rings?

A new radial or ring for the kitchen might be enough though, most bedroom circuits are very lightly loaded assuming you don't have space/ water heating on them.

Owain

[NT]

Its entirely pointless, unless of course you run a lot of large loads upstairs, for reasons I cant imagine.

Splitting the ring into 2x 20A circuits doesn't gain you anything much, and only worsens its safety performance.

The only place your average 2 bed house would benefit from a 2nd ring is the kitchen, where the heavy loads live. But even then, as you already know, in practice one can run a whole house on a single ring.

NT

[Jim]

We've got a similar problem in that we have solid floors downstairs so all of the ring mains will need to run in the suspended ceilings. The trouble being that there is a 50m cable length limit on the ring main, and you use up about 4.5m each socket coming down from the ceiling then back up.

Unfortunately the windows come too low down on the wall to run across underneath them, and there are doors/chimney breasts to get in the way as well.

I was thinking it might be possible to run a thick cable out to some kind of junction box, then spurs out to each socket rather than a ring - I don't know if that's allowed under the wiring regs?

[Dave Plowman (News)]

I'd say it's making too many unnecessary joints. Do you know where the ring goes upstairs and back down again? You might well find they are together. Also if you intend keeping the existing socket positions upstairs it's making more work.

Other thing is the ECC on older TW&E is smaller than currently. Not quite sure how that effects radical alteration. But my gut feeling would be to keep the old ring as short as possible.

[Andrew Gabriel]

In my view, these are unsuitable for inaccessible connections and do not comply with the regs for this purpose.

Ashley's use of "maintenance free" would appear to mean they provide no means to perform any maintenance, not that none would be required. Indeed, with the absence of large surface area gas-tight contacts which are present in all the proscribed [maintenance free] methods for inaccessible connections, I would expect a shorter lifetime from this product than from standard screw terminals, and as such, thoroughly unsuitable for an inaccessible connection.

Not used one myself though.

[David Hansen]

On Mon, 19 Oct 2009 13:21:07 +0100 someone who may be Jim wrote this:-

The Wiring Regulations offer a variety of standard solutions, for want of better words. Many people think they are all that can be used.

However, one is not restricted to using these standard solutions, in fact they encourage innovation. There is a space on the forms IET offer which allows one to list departures from the standard solutions. However, if one does use something non-standard then it has to provide at least an equivalent level of protection to the standard solution. If it comes to it one may need to be able to demonstrate this in a court, so may people sensibly stick to the standard solutions.

If someone wanted to use say Schuko sockets in a building they would have to wire them up to provide an equivalent level of safety. No ring final circuits, suitably small floor areas, RCDs, double pole circuit breakers and double pole switching, off the top of my head, would provide an equivalent level of safety, but that is only off the top of my head. There are few places where it makes any sense to do this though.

Your proposed circuit would be non standard. Provided it was designed, installed and documented properly no problem. Do you have the skills to do this?

[Bill Payer]

Well, this is rather the point in doing the job in the first place, ie, we don't have enough sockets. It's a three-bedroomed house with just one double-socket in each room and a single socket out on the landing, so doing it the way I originally thought would leave four joints under the floorboards. Is that too many?

[Jim]

[...]

We have a qualified electrician doing the specification and commissioning. I've just been looking at the regulations and trying to figure things out so I can have meaningful conversations with him.

Surely my "proposed circuit" could be considered as a radial with multiple spurs? In which case it could have as many spurs as necessary so long as the cable length limit from CU to socket wasn't exceeded.

[Tim W]

Andrew Gabriel wibbled on Monday 19 October 2009 13:35

Indeed. These things (Ashley in particular) keep getting mentioned on the IET forums in exactly this context and opinion goes both ways.

Obviously Hager are strongly hinting that they are suitable. all I can do is quote:

"The screwless push to fit terminals do not relax, and so do not require inspection." from:

would be interesting to challenge them on the specifics.

You're an engineer and obviously your opinion is worth much and I can see your argument.

I've used them in a semi-accessible location (under floorboards), where there's some flex in use (SELV downlighters, I use high temperature silicone flex) and they are very good for this (because the contacts are suitable for solid or flexible and the cable clamps being present).

I should note that the springs are pretty strong in these. Whether the springiness can be proven not to decline in time or whether the contact pressure is sufficient to prevent oxidation of the cable I couldn't comment.

Cheers

Tim

[Tim W]

NT wibbled on Monday 19 October 2009 13:14

I'm putting 4 rings in my house. 1.5 for the kitchen/diner (the other 0.5 serves utility and one bedroom), one for the rest of the ground floor and one upstairs. It's not necessarily the case that the loads demand it, but it's geographically convenient and povides sensible seggregation in the event of one circuit tripping (ie I don't lose the whole house).

[js.b1]

Single ring does not comply with 314.1, may not be "balanced" re all kitchen loads at one end. Many councils will reject a single ring based on the latter.

Breaking into two 20A radials doesn't gain much.

20A radials are good in quantity - that is one per room (lots of PCs, rented) and multiple in a kitchen (fridge/freezer, washer/kettle, dryer/microwave, fan heater). Quantity provides redundancy and localises any earth leakage fault (cable fault, element fault). 20A radials have a safety benefit in that a 2G socket is limited to 19.5A continuous, whereas downstream protection is 13+13A and upstream CPD on a ring would be 32A. The benefit of a radial is topology - whatever you want.

32A rings are good in quantity - minimum of two unless you know the ring will be balanced (not likely). 32A rings have a safety benefit in that there are 2 paths for earth, so effectively reducing EFLI although that is less critical with RCD protection. The downside of rings is the topology must be a ring, not always easy in some houses (and decoration!).

32A radials are a good compromise using 4mm FTE - still rare in domestic, but practically very useful in a kitchen environment. Note that Grid Switches are limited to 20A circuits, but then I suspect a kitchen could have a single 32A isolator re "emergency isolation" or isolation for all appliances (burning microwave, removal of built-in appliances).

I personally prefer lots of 20A radials re 314.1 redundancy, topological flexibility & expansion flexibility.

However for the OP I would break the kitchen off into its own 32A ring for 314.1, then have the house on its own ring. It might be very convenient physically to do this.

[Bill Payer]

Doh! That should be one double socket in each BEDroom, not, as you may have thought, each room of the house - sorry.

[Mike Clarke]

Bear in mind that adding more sockets isn't likely to significantly increase the loading which is dependent on the number and power rating of the appliances you use. So, unless you intend to get lots more high power appliances, adding more sockets to the existing ring won't be a problem. In fact it will probably make the whole installation much safer if it means using less multiway adaptors and trailing extension leads.

[David Hansen]

On Mon, 19 Oct 2009 13:57:51 +0100 someone who may be Jim wrote this:-

Electricians come in variable forms. Some just want to stick to the standard solutions and select cable sizes from a table. Others are very good and happy to do calculations if you pay them enough.

I see from re-reading your posting that you are not proposing a large cable leading to a ring, as I wrongly assumed.

You could use a standard 20A radial circuit, without the thick cable, provided that the floor areas were complied with.

You could also set up an non-standard radial circuit with the thick cable, but you would need calculations done to see that all the cables were properly protected. It may be the case that say a 32A protective device which protected the thick cable did not adequately protect "normal" 2.5 mm cable. There are a number of conditions to meet when sizing cables and protective devices.

[Andrew Gabriel]

Last couple of kitchens I've done (16th Ed regs), I put two rings into each, one RCD protected for the accessible socket outlets, and one non-RCD protected for stationary/fixed appliances such as fridge, freezer, oven, boiler, etc, which you don't want sharing an RCD with anything else, and don't merit one themselves.

Thus far, I always used 30mA RCBO's per circuit (or 10mA in a few cases), but never a single RCD covering multiple circuits.

[Dave Plowman (News)]

I see that - but are you adding to what's already there position wise? If so you could save some work by using the existing cables - they're likely longer than needed to the next existing socket so can be cut.

IMHO any unnecessary joints should be avoided.

[<me9]

I made sure my boiler was on an RCD protected supply. I want volts off ASAP if there is a water leak.

Why /do/ they always put the electrics /under/ the water parts in boilers?