ring mains - balancing load

[Jim Ingram]

Hi,

Could someone please expain to me what the requirements are for ensuring a balanced road on a ring main.

As part of some minor building work I'm planning to install a new ring main for the kitchen. As is often the case dishwasher and washing machine are next to each other and would be relatively close to the start of the ring. I would estimate about 4 m in from the cu with a total length of circuit of around 17m. If the two appliances were on together this might be around 6.5 kW giving a current of 27A of which I estimate 76% would travel down the shorter leg. This would be getting fairly close to the cables capacity. An additional load near this point could take the cable to above its rating (~24A for

2.5mm2) and not trip the 32A mcb. I would have thought this situation to be just about acceptable, as at least some of the loads will be fairly short lived (minutes rather than hours).

Alternatively I could lose some cable under the floorboards to lengthen the tail or put the wm/dw on a separate radial circuit.

The existing ring main is much much worse - serves the whole house with a potential load of 40 A (helped by an illegal spur) at the second socket 7m from cu on a 70m ring circuit! not good.

Thanks

Jim

[Lurch]

On Tue, 4 Oct 2005 19:14:39 +0100, "Jim Ingram" scrawled:

Well, 'losig some cable under the floor' isn't the way to balance a ring.

Easiest thing to do if you're worried is put in 2 16A radials for the WM and TD, or 1 32A radial for both.

[Andy Wade]

The essential requirements are in Reg. 433-02-04 and can be summarised thus:

- min. conductor size is 2.5 mm^2 (with one exception: 1.5 mm^2 is permitted for 2-core MI cables);

- the as-installed cable rating (Iz), taking into account the installation method and grouping factors, etc., must not be less than 20 amps;

- under the intended conditions of use, the load current in any part of the ring should be unlikely to exceed, for long periods, the actual as-installed value of Iz.

The last few words are significant. Short term loads such as kettles and toasters (assuming normal domestic use) can be disregarded. The actual Iz can vary between 20 A (the minimum allowed) and 27 A (clipped direct, no grouping).

In this example there's probably nothing to worry about, unless you have other long-term loads. Even dishwashers and washing machines aren't particularly long term loads: thermostats and diversity will conspire to reduce the 6.5 kW quite quickly.

If you are worried, other design options might be to take off the DW & WM feeds from nearer the mid-point of the ring, running unfused spurs back to the kitchen (but watch the grouping factors and circuit impedances), or to wire the whole ring in 4 mm^2.

HTH

[Andrew Gabriel]

or spur one of the loads directly from the MCB as it's close by.

[Andy Wade]

[Andy Pandy]

This is all well and good in theory but what happens when new occupants move into a house, with various load appliances ? Are the regs soon going to tell us where we can place our appliances? Have there been cases of houses being damaged by misplaced appliances on properly installed ring mains ? This issue seems on the surface, to be getting out of hand.

Andy

[David Hansen]

On Tue, 4 Oct 2005 19:14:39 +0100 someone who may be "Jim Ingram" wrote this:-

In most cases, test that the ring actually is a ring.

Not a problem. These things were thought through some time ago.

I have a vague recollection of a figure of 20% going the "long way" being the worst case. That is for a standard circuit, which has a maximum floor area for a number of reasons one of which is to do with loads.

Cable ratings depend on the installation. They are also time dependent. 2.5mm2 cable will safely carry many hundreds of amps for a short period, which is what happens in a short circuit. In an earth fault, in flat twin and earth wiring, the smaller protective conductor carries the same current as the live. A 32A MCB will take around 30s to operate at 100A, which the circuit wiring will take if it is designed, installed and tested correctly. A lot of thought has gone into the Wiring Regulations over a long time and co-ordination of protective devices with cable capacity. 100% safety is not possible, don't use electricity if you want that, but it is a good compromise with the competing factors.

[Fred]

I find it surprising, will the IEE and authorities hell bent on reducing risk - Part P and all that, that a ring main based on 2.5mmT+E is still allowed. In most environments 2.5mm is not up to passing 32A which is an every day possibility. Excusing this on the basis of diversity really isn't good enough unless the MCB rating suits the rating of the cable, ie worst case scenario.

The very fact that this is not seen as a problem gives me opinion it's quite acceptable to exceed a cable working temperature of 70 deg C.

[Jim Ingram]

Hi,

Thanks to all for the replies. I don't think I need to worry as I haven't got any other large loads.

Cheers

Jim

[Andy Wade]

Unless they alter the kitchen layout, similar appliances will end up in similar positions - no problem. If the kitchen is radically altered then wiring work is likely and the Part P registered electrician will of course re-check the design :-)

There are certainly cases where cables have been damaged due to failure to consider the requirement to avoid the likelihood of long-term small overloads [433-01-01]. Hence the clarification added to 433-02-04 in respect of ring circuits in Amendment 1 to BS 7671:2001.

Nothing's getting out of hand, but more emphasis is now placed on ensuring that the regs are actually complied with, and on making installers realise that there is a design process (however trivial) involved in wiring work.

[Lurch]

On Wed, 5 Oct 2005 09:13:33 +0100, "Fred" scrawled:

When? You sure you've not got ring and radial mixed up?

[Fred]

From a current perspective, what's the difference. Suppose we have two sockets close to one end, say nominally 1/10 into the ring. 9/10 of the MCB breaking current will pass through the shortest path. So yes a radial circuit can then take 10% less current.

I accept that probability is small, nevertheless it's still there, and in our present nanny state I'm surprised it's still allowed. Wait until the

17th Edition! It'll be the 18th or 19th before they insist all light fittings be 12V or to have "finger proof" fittings. After that 115V balanced around earth.

[Lurch]

On Wed, 5 Oct 2005 16:25:55 +0100, "Fred" scrawled:

Hmmm, don't think it's quite as simple a calculation as that.

[Bob Eager]

Oooh, I dunno....when it gets overloaded enough, its resistance will increase and it'll even out... :-)

[Kalico]

[snip]

Would it be possible to run the short length or the ring, as far as these high loads, in 4mm^2?

Replace 'spam' with 'org' to reply

[Fred]

Therefore for a 30 degree increase the resistance will go up by 12%. So instead of 90% of current flowing through the short leg in the example I gave, it would be just under 89%.

[Fred]

Short leg 10% of resistance of ring. Long leg 90% resistance of ring. The voltage drop is the same across both legs. Simple ohms law - short leg carries 9 x current of long leg. Add the two legs together to get (90% +

10%) = 100% Where do you think I'm going wrong?

[Bob Eager]

(a) I did know that - it was tongue in cheek! (b) I was thinking of a glowing cable!

[Fred]

Apologies, I don't normally see tongue in cheek until it hits me in the eye!

[Richard Conway]

That's not tongue in cheek, that's tongue in eye :)