Ring circuits to be retired?

[LastManOnline]

Could the 30% leakage requirement on rcd, s, combined with the inevitable increase in kitchen appliances with intentional leakage built in, force designers to replace rings with radial circuits?

 

[Lucien Nunes]

One must distinguish between two comparisons of socket-outlet circuits: Ring vs. Radial, and 32A vs 20A. If aggregate leakage is a problem for 32A rings it is also a problem for 32A radials. It would be a shame to abandon the 32A circuit, it does so much more than a 20A circuit. Two, three times more, due to the much greater diversity. And we've put so much effort into making it possible, with our fused plugs.

Functional leakage has always been there; it comes (capacitively) from cables as well as suppressors in appliances. Progress with SMPSU design could ultimately reduce the functional leakage per unit, compensating for the greater number of appliances in use. FWIW, in the specific context of high leakage, rings (used with dual-terminal fittings) actually offer an advantage over radials in providing a high-integrity earth connection by default.

So my personal opinion is no, it's still a viable approach.

 

[timhoward]

(OP) I don't quite follow your thinking. The direction of travel has been dividing installations up into multiple sockets circuits for some time, but I don't see why that would necessarily mean rings would become radials, at least at the design stage.

 

[ipf]

Is RCD tripping due to 'gradual' leakage on the increase or decrease? I would say the latter. Don't seem to come across it anything like as much, these days.
There are more designated circuits these days, also. The general ring circuit is never ending......whichever way you look at it.
Agree with Lucien.

 

[pc1966]

While more "stuff" has electronics in it, the leakage per SMPSU, etc, has generally decreased and is often in the 0.5-1mA region. Also quite a lot is now class II (e.g. wall-wart PSU) so no significant leakage there.

As above that is independent of RFC/radial argument, though often the benefit of the RFC is larger area for given cable, etc, hence likely more stuff on it.

In terms of general trends, using RCBOs per-circuit rather than dual RCD per-floor, for accumulation is probably a bigger factor in less trouble!

Agreed that is is a factor and if you are looking at office/server room something to really consider, but I don't think it is such a problem in real life for the foreseeable future.

 

[pc1966]

Just to add the 30% (i.e. around 10mA) max limit has been in the regs for some years, not sure how often anyone actually calculates what it means in real life!

There used to be some table in the PAT book with typical values, though many of the bigger cases were the likes of oven elements when cold so "so much per kW" sort of figure (from memory).

 

[LastManOnline]

(OP) I don't quite follow your thinking. The direction of travel has been dividing installations up into multiple sockets circuits for some time, but I don't see why that would necessarily mean rings would become radials, at least at the design stage.

Firstly, a pretty definitive response from the forum. Secondly I heard the suggestion when watching a "voltimum" webinair on RCD, s. The suggestion was that two radials replacing one ring would double the available leakage. It sounded reasonable to me

 

[timhoward]

Firstly, a pretty definitive response from the forum. Secondly I heard the suggestion when watching a "voltimum" webinair on RCD, s. The suggestion was that two radials replacing one ring would double the available leakage. It sounded reasonable to me

I see what you mean now and that does make sense. I thought you meant one ring becomes one radial.
I'll be interested to hear what other peoples experiences are but I have most problems with leakage on split load or dual RCD boards, or old TT installs with one main switch RCD on the whole lot, or sometimes or when an entire large house has only one ring circuit. I've found the shift towards RCBO's and (usually) at least up and down sockets being separate, often kitchen too, has made it much less of an issue.
I don't claim to be a representative sample though as I do a lot of commercial work.

 

[pc1966]

The suggestion was that two radials replacing one ring would double the available leakage. It sounded reasonable to me

Of course it would, but the question is it needed?

Once the cost for AFDD for some locations is added in the customer might want a better reason than just it feels right,

As with @timhoward I suspect most problems are in the case of single or dual RCD boards. Again I have a very limited experience in this area but a typical whole flat/floor RFC on a RCBO seems not to have any real problems unless there actually is a water leak in the washing machine, etc.

EDIT: Just to clarify it would allow double the leakage assuming all of the leaky stuff is not concentrated on one radial. Physical layout and room-usage plays a part there.

 

[LastManOnline]

Just to add the 30% (i.e. around 10mA) max limit has been in the regs for some years, not sure how often anyone actually calculates what it means in real life!

Honestly, I have never calculated it. I consider the type of installation and tend to use my "gut". But your point about rcbo populated DB, s really is the solution. It really does"future proof" installations against any issues with nuisance tripping

 

[pc1966]

Honestly, I have never calculated it. I consider the type of installation and tend to use my "gut". But your point about rcbo populated DB, s really is the solution. It really does"future proof" installations against any issues with nuisance tripping

For CU changes you can always use a clamp meter per circuit to get an idea of the present-day situation, but as above I expect many households won't have issues.

Mind you I know at least two people with what is essentially a rack from a data centre with related computers and network switches in thier homes so they might like a dedicated feed just for that lot!