A Ring Main with a Ring Spur, anyone heard of this being done?

[ElectricsRanger]

Hello to everyone,

I was thinking after i looked at some wiring under the floor of a house rewired in 1970-72, that there are two junction boxes off two twin an earth cables, one wire "A" going towards a socket ive yet to check, and the other wire "B" going around possible swinging around back to that socket.

Its not conclusive at this time, ive only just started tracing the cables, but has anyone heard of people wired ring spurs on to ring mains?

It sounds like fiction, but we all hear of strange decisions made by people, because it was easier, because it did not seem to matter. This was a long time ago, and the house is due a certification.

I know the socket is wired like its on a ring, but this wire "A" going towards it where nothing else should be.

I find it hard to believe it could have a ring spur on main ring main, but i just wondered if anyone has heard of it happening. Ill trace the cables and note the facts.

ElectricsRanger

 

[Taylortwocities]

Figure of 8 rings are not uncommon. They often occur as a result of additions to an existing ring final. They are not correct and shoujld be sorted out. They give very strange results when carrying out R1+R2 tests on the ring!

You need to confirm that the sockets are actually part of the ring. What you could be looking at are two separate spurs coming from that one JB. Two minutes work with a continuity tester would determine that, and will tell you what you need to do to sort it out!

 

[ElectricsRanger]

Thank you for your reply Taylor, ill post back when i know more.

 

[ElectricsRanger]

Okay...

So i was able to continue my trace. After performing a safe isolation on the downstairs ring main, i disconnected the cables from the downstairs socket, and disconnected the wire from the junction box. I then used a long meter lead to link the two ends, and it turned out as i expected the wires were not connected.

I then lifted a floor board on the other side of the room as i originally planned, and it turned out that the cable was probably a radial spur installed after the original electrical installation. It has been neatly placed under the floor boards, with spaces cut out of the beams for the cable to sit. Luckily no nails had punctured the cable, but i imagine that installations like this, could be the ones someone might hear of a nail being live, from puncturing a cable.

Before i did my lead test, i could see when i removed the junction box cover, and tested it to be sure it was also not live, that the cables were older than first expected. The earth was a kind of dull silvery multi strand wire, possible aluminium i don't know. Actually all the cables were multi stand. Who ever installed the spur had carefully removed grey outer sheath, then removed enough insulation on the live, neutral and earth and placed each wire conductor neatly in to each individual junction, then added the newer cable with the more modern, red, black and green & yellow. It looked good and kept the ring wires continuous.

Seems the installation team in the 70s left all the cables in place. They removed all the sockets that have been added previously, but left the cables in the walls and in the celling. When i lifted the floor board above the socket below, the old twin & earth was emerging from the wall as if it was still in use. I would have thought they would have at least cut the cable before installing a new ring.

The new ring cables looked good in the celling above the socket i was tracing, and had been installed in a straight line. Ill have to take another floor board in the room next room so i can continue the trace. I expected the cables to be a mess, as i was getting strange cable detection readings through the celling over half a meter away, which ive found to be rather odd, and i expect the EMF's are not right. The field completely disappears when the ring is off, and everyone feels better and sleeps better, but as soon as its back on, the room and most of the house feels cursed, sleep problems and irritability common issue.

Thank you for your reply Taylor. Your figure-8 was a good suggestion, apparently that was a common feature of the 1940s ring main.

****
Update 9/11/2015 22:10

Errr.... When i said it was a good suggestion, i was talking about the possibility of my circuit being a figure-of-8, because of the age of the wiring. Seems a whole new discussion just exploded about it. I apologise if it looked like i was making a point about a sensitive topic.

ElectricsRanger

 

[ericmark]

Not sure of date but house building stopped during the war and pre-war is was 5 and 15 amp sockets typical a house would have 2 x 15 sockets one upstairs and one down stairs. Post war it increased to around 5 sockets now using the ring system I think one socket up stairs and one in each room down stairs was the norm.

The big change was 1966 when we started to earth the lights. Not sure of date when we went metric with cable 0.029, 0.036, and 0.044 seemed to be thickness of strands there was also a 0.064 and we had 1, 3, and 7 strands with fixed cables. For the ring it could be 3/0.029 but more common 7/0.029 and as a rubber rather than plastic cable it could take a lot more current than the metric cable which replaced it.

So having a unfused spur from a spur was not that much of a problem, although overloading could happen, it was less likely to damage the cable than with modern 2.5 mm².

I think there was about one year where aluminium was used. Somewhere between 1965 and 1979 as that was when Ian Smith was in power in Rhodesia which caused the copper shortage.

I have many times had problems with old dead cables, it is hard to prove dead, the problem is all switches and thermostats and relays must be on to prove dead, it is specially a problem with central heating cables, which can seem dead until some timer or thermostat switches on. Using proving units is useless as they can't prove dead, they and only prove live. I have had cables proved as dead becoming live when something has switched on.

I was told when doing my C&G2391 I should test for figure of 8, I accepted this without looking at regulations, however I can't find a regulations to say you should not use figure of 8, there is an overload problem if taken near the end/start of run, but it does not seem to be outlawed.

 

[JohnW2]

I was told when doing my C&G2391 I should test for figure of 8, I accepted this without looking at regulations, however I can't find a regulations to say you should not use figure of 8, there is an overload problem if taken near the end/start of run, but it does not seem to be outlawed.

As we've discussed at length, if one takes any 'ordinary' ring final and turns it into a figure-of-eight by installing some 'cross-connection', one will inevitably decrease the chances of any of the cable becoming overloaded, as well as increasing CPC redundancy.

The main reason for not installing such a circuit, and the reason I would not deliberately install one, is the difficulty and potential confusion it can introduce in relation to subsequent testing and fault finding.

In practice, I'm pretty sure that nearly all figure-of-eight circuits have arisen 'by accident', during circuit extensions/modifications, rather than having been deliberately installed as such.

Kind Regards, John

 

[ericmark]

If the second ring is taken from the mid point of first ring then no problem. However if taken near to the consumer unit then it can overload one leg. Of course the same would apply if you put a row of sockets next to the consumer unit I seem to remember working out with a 106 meter ring only one socket should be fitted within 5 meters of consumer unit. However it was a long time ago that I did the calculation.

I tried to re-do the calculation current rating of cable 21 amp length 106 meters at centre point drawing full 32A would result in 16A per leg. At some point there would be a 21/11 amp mix to get the 32 amp and I would assume it would be 106/32*11 = 36 meters from the consumer unit drawing 32A would put the full 21A load on one leg.

We all realise that the 21 amp is method 100 and in real terms unlikely installed to method 100 plus of course it will stand some overloading. I know where my dad used 2.5mm² to extend a 7/0.012 spur to include extra sockets mistaking it for a ring the 2.5mm² cable was damaged but the 7/0.012 cable showed no signs of stress. Part of it's life with a 30A fuse part with a 32A B type MCB. So I know 2.5mm² will not take load from a ring final protection without damage. But I would guess it will take more than 21 amp. I think damage was caused from using a 4 bar electric fire. However that was only 13A how it got enough load to damage the cable I don't really know. There were 4 socket outlets from the 2.5mm² cable but what could have been used with a total of over 8 amp I really don't know.

In real terms I would think 106 meters is rarely used so that 32 meters would also be less. However if a draw of 32 amp be it a 6mm² cable taking supply to shed or garage or a figure of 8 ring is take to close to the consumer unit one leg of the ring can be overloaded.

I am sure I have made an error I know regulations state current carrying capacity must be 20A or more, there is a statement

If under the intended conditions of use, the load current in any part of the circuit is unlikely to exceed for long periods the current-carrying capacity (Iz) of the cable.

it does not say 2.5mm² ring finals with 30/32 amp protection are always allowed.

I know when we calculate ring final length we use 26 amp not 32 amp being 20 amp at centre and 12 amp even distributed but for current carrying capacity I am sure we use full 32 amp.

As I say I am sure I have made an error in calculations within 36 meters of start/finish of ring final seems out. So interesting to see where I have made the error. However I am sure a second ring inserted too close to the start/finish will cause an overload?

 

[JohnW2]

If the second ring is taken from the mid point of first ring then no problem. However if taken near to the consumer unit then it can overload one leg. Of course the same would apply if you put a row of sockets next to the consumer unit ...

Sure - but, as you say, you're now just talking about having a lot of load connected near one end of the ring, which carries the risk of overloading' the short leg - but that's no more true of a figure-of-8 circuit than a conventional single ring.

What I said still applies. If you take a single ring which is of risk of overloading one leg because of a lot of sockets close to one end of the ring, if you convert that circuit into a figure-of-eight by adding some 'cross connection' (any cross-connection), you will decrease the risk of (or extent of) the overloading of the short leg.

I tried to re-do the calculation current rating of cable 21 amp length 106 meters at centre point drawing full 32A would result in 16A per leg. At some point there would be a 21/11 amp mix to get the 32 amp and I would assume it would be 106/32*11 = 36 meters from the consumer unit drawing 32A would put the full 21A load on one leg.

Indeed - give or take rounding, that arithmetic is correct. More typically, if the installation method is such that the CCC is 27A, that 36 metres reduces to about 16.6m. For a more realistic 30m ring, the figures are about 10.3m when CCC=21A and 4.7m when CCC=27A.

Kind Regards, John

 

[ElectricsRanger]

After my experience with the house i am troubleshooting, which ive lived in for 30 years. I don't think i would vote for a ring main, whether it was a standard ring main or or a figure 8, at least at this moment.

The house has had possible had problems with the wiring for a very long time, and because there has been no trips on the main and thankfully no fires or smoke, no one thought there was a problem to investigate. The risk of resistance and balanced current flow problems in the ring, possibly creating EMF seems to out way the advantage of using two 3000w heaters for us.

Two radials circuits for the downstairs is something i am now investigating, or something possibly similar.

ElectricsRanger


Ban-all-sheds, its not that the cables will move about, its if work carried out and the installation is not tested well enough to detect new faults. A ring system could have a larger resistance in one side of the neutral or live, resulting current not returning the same path as it arrived, creating a larger field around the cables than it would have, if the current around each wire was cancelling each other out as they travel to the same side of the ring.

 

[ban-all-sheds]

The risk of resistance and balanced current flow problems in the ring, possibly creating EMF

Really?

You actually think the cables will move about?

 

[ericmark]

I could not find a 0.5 ohm resistor in my electronics work bench so used two 1 ohm in parallel the top shows a figure of 8 the bottom a standard ring loads are the same the ammeters show the load on each leg. The point is with a single ring the cable resistance is more even spread if we assume 60 meters of cable 40 in original and 20 in the figure of 8 with the figure of 8 being 10 meters from the start/finish then the load on that 10 meters will be much more than if at the 10 meter point the cable was split and ring inserted as the sockets would have some of that extra 20 meters of cable before load is taken.

I have debated with my son on the idea of 2 x 16A MCB's instead of a 32A MCB one feeding each leg of the ring. It would still be 32A total but out of balance load would then trip the MCB's with a cascade effect. However he correctly pointed out going away from standard would confuse anyone working on the system. Would be better if a twin pole 16A MCB was used but again one could not ensure anyone working on the system would understand what they were working on.

There would be an advantage using double pole 16 or even 20 amp MCB's as if the ring was broken then there would be no problem each leg would still be protected to the cables working current. It would remove 90% of the objections to using ring finals. However can't see that ever happening!

 

[JohnW2]

... the top shows a figure of 8 the bottom a standard ring loads are the same the ammeters show the load on each leg.

I am having some difficulty in seeing in what sense the top diagram constitutes a 'figure-of-eight circuit. One thing I'm not sure of is whether or not the top of R1 is joined to the junction of R6 and R9.

To illustrate the point I was making, you could try the following ... start with a standard ring with loads applied near one end so as to somewhat 'overload' the cable of the short leg. Now add a cross-connection between two of the sockets (in theory any two sockets, but the illustration will be clearer if you chose ones quite a distance apart on the ring), turning it into a figure-of-8 - and you will hopefully see that the degree of 'overloading' diminishes or disappears.

Kind Regards, John

 

[ericmark]

Taking the example shown one would expect the impedance to be about 0.45Ω total each 10 meter section around 0.056Ω we will assume all sockets loaded equal at around 270 watt each. And the fig of 8 link zero ohms.

With the link in or out the balance point will still be the link between sockets 18 ~ 19 but with link in the cable resistance at the top will be 0.11 Ω and bottom 0.25Ω so with link in we have 21.3 amp top and 10.6 amp bottom and without the link 16 amp both top and bottom.

So even with even loaded sockets the figure of 8 can produce an overload.

 

[JohnW2]

With the link in or out the balance point will still be the link between sockets 18 ~ 19 ...

Am I missing something. Your socket number is a bit confusing (11 is missing and the 26th one is labelled 27) - but don't you mean between sockets 13 & 14 (assuming that "12" was re-numberd "11")? If so, your arithmetic obviously would change considerably, with far less difference with/without the "fig-of-8 link'.

So even with even loaded sockets the figure of 8 can produce an overload.

Even forgetting the above, what you are seeing is surely mainly the consequence of connecting your second loop very close to (12.5% of the circuit length from) one end of the ring? A more realistic/sensible 'fig-of-8' would connect, say, sockets 10 and 17 (the left-hand ones in the third and fourth rows).

Kind Regards, John

 

[PBC_1966]

I then lifted a floor board on the other side of the room as i originally planned, and it turned out that the cable was probably a radial spur installed after the original electrical installation. It has been neatly placed under the floor boards, with spaces cut out of the beams for the cable to sit. Luckily no nails had punctured the cable, but i imagine that installations like this, could be the ones someone might hear of a nail being live, from puncturing a cable.

Running a cable in notches cut in the joists so that it would sit right below the floorboard was very common at one time. So long as the cable was run down the middle of a floorboard there was no real danger from nails, as the floorboard is nailed at the corners and down the long edges.

The earth was a kind of dull silvery multi strand wire, possible aluminium i don't know. Actually all the cables were multi stand.

Probably Imperial cable, which was discontinued circa 1970. The strands were tinned copper.

Here are the cable makeups:

And the current ratings:

Both of the above are from the 14th edition IEE Wiring Regs., 1966.