Unbalanced final ring

[Steve-S]

What has alerted me, to some surprise, is how out of balance a final ring can be, which must be more typical in British homes than we think as heavy current drawing appliances are clustered. This was not a big problem in the 1960/70s. I measured 82% on one leg while Efixx measured 83%. Previously most I measured were 70-30 at most, giving 22.4 amps on one leg - just inside the max for the cable for the majority of installations. Fitting a ring with sockets in round robin (1st socket on leg 1, 2nd socket on leg 2, 3rd socket on leg 1, etc) will solve a balance issue, but this method can be quite impractical and a heavier in use of cable increasing costs.

This points us to that final rings should be all in 4mm, or 4mm on one leg to the clustered heavy current drawing sockets and then 2.5mm for the rest at least, for many reasons. In a normal house/flat I would lean to put as many appliances as possible on a full 4mm ring protected by an AFDD; diversity taken into account of course. A 3kW-ish oven & low current drawing induction hob on a 4mm radial and of course the immersion, if one on premises, on its own 2.5mm radial. Maybe more economical having the immersion on 4mm as final ring and oven/hob will be on 4mm so cable being cheaper all on the same purchased drum. At TLC a 100m drum of 4mm is £147 (£1.47 p/m), with 100m drum of 2.5mm £72 (£0.72 p/m), about twice the price, but £1.03p on a 25m drum. Costs of 4mm could drop in cable costs as only using one final ring in 4mm. Worth thinking about.

 

[Pretty Mouth]

What has alerted me, to some surprise, is how out of balance a final ring can be, which must be more typical in British homes than we think as heavy current drawing appliances are clustered. This was not a big problem in the 1960/70s. I measured 82% on one leg while Efixx measured 83%. Previously most I measured were 70-30 at most, giving 22.4 amps on one leg - just inside the max for the cable for the majority of installations. Fitting a ring with sockets in round robin (1st socket on leg 1, 2nd socket on leg 2, 3rd socket on leg 1, etc) will solve a balance issue, but this method can be quite impractical and a heavier in use of cable increasing costs.

This points us to that final rings should be all in 4mm, or 4mm on one leg to the clustered heavy current drawing sockets and then 2.5mm for the rest at least, for many reasons. In a normal house/flat I would lean to put as many appliances as possible on a full 4mm ring protected by an AFDD; diversity taken into account of course. A 3kW-ish oven & low current drawing induction hob on a 4mm radial and of course the immersion, if one on premises, on its own 2.5mm radial. Maybe more economical having the immersion on 4mm as final ring and oven/hob will be on 4mm so cable being cheaper all on the same purchased drum. At TLC a 100m drum of 4mm is £147 (£1.47 p/m), with 100m drum of 2.5mm £72 (£0.72 p/m), about twice the price, but £1.03p on a 25m drum. Costs of 4mm could drop in cable costs as only using one final ring in 4mm. Worth thinking about.

I have come across plenty of open rings in domestic settings, but have never once found one where the cables were showing signs of aging, let alone thermal damage, as a result of overload. I'm of the opinion that the sorts of overload that are likely to occur are not great enough to cause actual damage, even when the ring is open, or the load heavily biased to one leg.

It's good to explore these things, as it gives us a better understanding of the circuits we work on, but I don't think 4mm is generally necessary for wiring rings.

 

[mainline]

Deja vu comes to mind with this thread.

 

[Steve-S]

I have come across plenty of open rings in domestic settings, but have never once found one where the cables were showing signs of aging, let alone thermal damage, as a result of overload. I'm of the opinion that the sorts of overload that are likely to occur are not great enough to cause actual damage, even when the ring is open, or the load heavily biased to one leg.

It's good to explore these things, as it gives us a better understanding of the circuits we work on, but I don't think 4mm is generally necessary for wiring rings.

Nice comment. Thanks for your input. It is a valued judgment of course. OK a worst case scenario - like the Christmas dinner scenario in cookers. Wife has the washing on in the kitchen. Tumble dryer constantly drawing 13A, washing machine intermittently cuts in drawing 11A, fridge cutting in and out, TV on in the corner. Her three friends pop in, the 3kW full kettle goes taking a time to heat up, and the 1kW toaster. Dishwasher is also switched on.

The ring can reach the 32A. it will not trip as it will take more than 32A. The kettle and toaster then cut out taking it below 32A.

a) The point is the ring may be on max more often than you might think.

Then if the ring is out of balance by 80-20 one leg will draw 25.6A, which may be over the max rating of the cable taking derrating into account.

Again:
b) The cable on many occasions will be over its limit more times than we might think it would be.

c) The cable insulation would deteriorate quicker than we might think being heated more times than we might think.

My prime point is that if the ring looks like being heavily unbalanced, run 4mm on one leg to the cluster of heavily loaded sockets. Belt and brace and certainly will ensure no, or minimal, insulation breakdown. 4mm easily cope with 25.6A without and adverse effects in 95% plus of situations, taking into account derrating A valued decision of course.

Yes, it does give us a better understanding, and well worth doing the maths and understanding.

 

[Pretty Mouth]

Nice comment. Thanks for your input. It is a valued judgment of course. OK a worst case scenario - like the Christmas dinner scenario in cookers. Wife has the washing on in the kitchen. Tumble dryer constantly drawing 13A, washing machine intermittently cuts in drawing 11A, fridge cutting in and out, TV on in the corner. Her three friends pop in, the 3kW full kettle goes taking a time to heat up, and the 1kW toaster. Dishwasher is also switched on.

The ring can reach the 32A. it will not trip as it will take more than 32A. The kettle and toaster then cut out taking it below 32A.

a) The point is the ring may be on max more often than you might think.

Then if the ring is out of balance by 80-20 one leg will draw 25.6A, which may be over the max rating of the cable taking derrating into account.

Again:
b) The cable on many occasions will be over its limit more times than we might think it would be.

c) The cable insulation would deteriorate quicker than we might think being heated more times than we might think.

My prime point is that if the ring looks like being heavily unbalanced, run 4mm on one leg to the cluster of heavily loaded sockets. Belt and brace and certainly will ensure no, or minimal, insulation breakdown. 4mm easily cope with 25.6A without and adverse effects in 95% plus of situations, taking into account derrating A valued decision of course.

Yes, it does give us a better understanding, and well worth doing the maths and understanding.

I hear what you're saying, but I really think if it was a problem, we'd occasionally come across rings that were deteriorated significantly, or even slightly. I think the margin of error is enough to absorb this sort of thing.

This shows approximate life expectancy of cables for various temperatures, starting at the operating temperature of 70 deg C. Even running at a continuous 115 deg C (that's 45 deg hotter than it should!), it would take about 2 months before PVC cable deteriorated (that's 1460 hours!). I don't know exactly what overload would cause the cable to heat to that temperature, but I would have though it'd need to be significant. In a domestic setting, it seems unlikely to me that overloads such as you describe would be sustained for long enough to cause any real damage.

 

[Steve-S]

I hear what you're saying, but I really think if it was a problem, we'd occasionally come across rings that were deteriorated significantly, or even slightly. I think the margin of error is enough to absorb this sort of thing.

This shows approximate life expectancy of cables for various temperatures, starting at the operating temperature of 70 deg C. Even running at a continuous 115 deg C (that's 45 deg hotter than it should!), it would take about 2 months before PVC cable deteriorated (that's 1460 hours!). I don't know exactly what overload would cause the cable to heat to that temperature, but I would have though it'd need to be significant. In a domestic setting, it seems unlikely to me that overloads such as you describe would be sustained for long enough to cause any real damage.

View attachment 118381

Thanks. Interesting table that says a lot. T&E can take some abuse for sure. Ten years ago I was involved in the renovation of a 1966 house that was super modern at the time. Spiral staircase the lot, like you saw in the Avengers. Built by an architect for himself who lived there alone until he died. It had original brushed steel flat plate sockets. I was amazed they even existed then - must have been a very expensive special order and designer at the time. The place was gas cooking and gas forced air heating, with a 3kW electric immersion for the DHW, the heaviest electrical appliance.

The quality looking wiring passed an insulation test. As much of the old wiring was used as possible as it would be a big messy job getting new cable to the new CU located in an externally accessed cupboard away from the main house. A gas hob and high flow combi was installed heating the new forced air unit with an electrostatic air filter, so the only high current appliance was the 3kW oven. It was built when there was town gas so the gas pipes were 1", nice and big for the new high power natural gas combi.

 

[brianmoooore]

In fifty years, the only cable I've seen damaged by excess current was a 6mm2 earth bond to the mains water pipe on a TNC-S system. The first glimpse I caught of it, under a floorboard in the bathroom, I made some mutterings about plumbers and their blowlamps but when more boards were lifted I could see it was well cooked along its entire length.

 

[Steve-S]

In fifty years, the only cable I've seen damaged by excess current was a 6mm2 earth bond to the mains water pipe on a TNC-S system. The first glimpse I caught of it, under a floorboard in the bathroom, I made some mutterings about plumbers and their blowlamps but when more boards were lifted I could see it was well cooked along its entire length.

Sounds like a Neutral dropped out with all going via the earth cable, even next doors.

 

[brianmoooore]

Sounds like a Neutral dropped out with all going via the earth cable, even next doors.

Only a small village, and only one more property 'down the line', but that's probably happened at sometime in the past.
Insulation on the wire was properly bubbled up and blackened all along, but removed and replaced by a 10mm2 as part of the rewire/extension I was doing at the time.

 

[Steve-S]

Only a small village, and only one more property 'down the line', but that's probably happened at sometime in the past.
Insulation on the wire was properly bubbled up and blackened all along, but removed and replaced by a 10mm2 as part of the rewire/extension I was doing at the time.

John Ward did an excellent demo rig explaining your scenario. Woud have been best to convert to TT earthing in the renovation.

View: https://www.youtube.com/watch?v=fen8ekrhlBI&t=1s

 

[Pretty Mouth]

By chance this popped up on youtube earlier, it shows T+E overloaded to destruction. The experiment is a little flawed IMO, because he doesn't really allow enough time for the smaller overloads to reach their peak temperature, but it still may be of interest.

View: https://www.youtube.com/watch?v=IF2UvUZoO2E

 

[PS Electrics]

By chance this popped up on youtube earlier, it shows T+E overloaded to destruction. The experiment is a little flawed IMO, because he doesn't really allow enough time for the smaller overloads to reach their peak temperature, but it still may be of interest.

View: https://www.youtube.com/watch?v=IF2UvUZoO2E

Popped up in mine as well, looks like the algorithm is finding us!

 

[brianmoooore]

I have an old Pichill 180A stick arc welder that gets dragged out occasionally for some heavy duty welding. On a few occasions, when it's been out and I've been bored, I've connected the odd metre or two of T&E or singles between its clamps just to see what happens at various currents.
General conclusion is that it'll take considerably more current than you might think before damage occurs.

 

[KungFuTurkeyLeg]

Why not rebalance the ring by extending it in one of the cavities? It's an old ideology but in theory works. Calculate the length to the two sockets feeding kettle and dryer, for example, by splitting the ring. Then calculate the distance from the other side of the split ring socket back to the CU. L1 = 10m and L2 = 15m, add a 5m figure of 8 onto L2 and reconnect the ring. 15m/15m each side of the ring to the heaviest loads

Diversity wipes out the above though. But something like this I would look to do at design stage and make it a none issue

 

[Steve-S]

Why not rebalance the ring by extending it in one of the cavities? It's an old ideology but in theory works. Calculate the length to the two sockets feeding kettle and dryer, for example, by splitting the ring. Then calculate the distance from the other side of the split ring socket back to the CU. L1 = 10m and L2 = 15m, add a 5m figure of 8 onto L2 and reconnect the ring. 15m/15m each side of the ring to the heaviest loads

Diversity wipes out the above though. But something like this I would look to do at design stage and make it a none issue

Thx, but:
"L1 = 10m and L2 = 15m, add a 5m figure of 8 onto L2 and reconnect the ring. 15m/15m each side of the ring to the heaviest loads"

Did you mean L1 not L2?

 

[mainline]

Thx, but:
"L1 = 10m and L2 = 15m, add a 5m figure of 8 onto L2 and reconnect the ring. 15m/15m each side of the ring to the heaviest loads"

Did you mean L1 not L2?

I think he means Leg 1 and Leg 2.

 

[KungFuTurkeyLeg]

Thx, but:
"L1 = 10m and L2 = 15m, add a 5m figure of 8 onto L2 and reconnect the ring. 15m/15m each side of the ring to the heaviest loads"

Did you mean L1 not L2?

Yep! Edit button was disabled by the time I seen it. Hopefully the 15m/15m clarified what I intended