'Bridged' Ring Final Circuits

[JohnW2]

Recent discussion about the merits (or otherwise) of the ‘British phenomenon’ of ring final circuits has reminded me of a point I’ve often thought of raising.

Particularly in relation to the testing of ring finals, a lot is said and written about the ‘problem’ of bridged rings, and some of the more tedious aspects of testing such a circuit are undertaken at least partially to detect such a ‘problem’.

However, is this really a ‘problem’? As far as I can see, in the absence of any faults, the effect of one or more ‘bridges’ is, if anything, positive – it helps to equalise currents throughout the ring, and the redundancy it introduces even provides a degree of ‘protection’ against the effects of some possible faults (notably ‘breaks’ at some places in the circuit).

The standard argument appears to be that a break in the circuit downstream of a bridge can leave one with two or more sockets effectively on unfused spurs, hence supplied with under-rated cable. That’s very true, but why is it any different from (or worse than) the situation in which a break occurs in a ring final which does not have any bridges? Indeed, in some senses the latter is a worse situation, since it usually leaves all sockets on the circuit supplied with an under-rated cable, whereas a break downstream of a bridge will have that effect only on some sockets. Nor is there any difference between the two situations as regards ‘identfiability’ of a fault – whether or not there are bridges, the user (e.g.householder) will be unaware of a single break, since all sockets will continue to function normally.

So, I wonder if there really is a logical reason (which I am missing) for us to regard bridged ring finals as any more of ‘a problem’ than a ring final without bridges. If bridges were regarded as acceptable, then there would have to be some changes in testing practices – but that, in itself, would not be a good reason for not accepting them - and, as above, it would even be possible to argue that bridges represented an advantage.

Comments?

Kind Regards, John.

 

[EFLImpudence]

Too logical, John.

I can't disagree with anything you have written but whilst attempting to add further points to your post I could see, whilst writing, disadvantages with my proposals.

The main one of which was that, if allowed, the circuit would end up, because of further mistakes, as a complete mess which would be impossible to test.

Even this may be considered electrically safe by measurements taken but without disconnection at every socket and accessory no one would know which cable went where and whether it was a bridge, a bridged bridge or part of the original design.

 

[ban-all-sheds]

It would make identification of multiple-outlet unfused spurs harder.

Best to get rid of rings altogether, really.

 

[JohnW2]

Too logical, John. I can't disagree with anything you have written but whilst attempting to add further points to your post I could see, whilst writing, disadvantages with my proposals. The main one of which was that, if allowed, the circuit would end up, because of further mistakes, as a complete mess which would be impossible to test. Even this may be considered electrically safe by measurements taken but without disconnection at every socket and accessory no one would know which cable went where and whether it was a bridge, a bridged bridge or part of the original design.

I'm glad to hear that you can't fault my logic. Yes, I agree that it could end up as an almost incomprehensible (unless documented) mess - but so can almost any circuit (we've all seen them!), so I'm not sure that this, in itself, would be a reason to 'outlaw' bridges. Mind you, although it's obviously not mentioned in the regs in relation to ring fnals, I'm not convinced that one could find anything in BS7671 which actually 'oulaws' (albeit not law ) bridges. I'm also not convinced that making life easier for electricians is necessary a good reason for 'outlawing', or even discouraging, a certain wiring practice, if the practice is sound in engineering terms (those who design cars etc. often do not have much thought for the poor devils that have to maintain them!).

I certainly don't intend to rush around my house installing bridges, but it does seem a litle odd that they seem to be regarded as such a big 'problem' - when, as I've said, they arguably have a positive effect in relation to 'safety' (in its broadest sense). As has been illustrated by the Internet, the 'best' arrangement, because of it's multiple redundancy, would probably be an 'unholy mess', with cross-bridges all over the place, including 'bridges between bridges' (arguably making the concept of a 'ring final' more acceptable to some as an engineering concept!).

Kind Regards, John.

 

[JohnW2]

It would make identification of multiple-outlet unfused spurs harder.

It would - but, as I've just written, if you look around the world, ease of maintainenance/testing often doesn't feature much in design processes. Engineering soundness should be the primary criterion.

Kind Regards, John.

 

[ricicle]

Anyone think two properly designed 20A radials are better - 8 amps more capacity, no ring continuity testing and better continuity of supply ?

(Ok - so two breakers to fit instead of one but....)

 

[JohnW2]

Anyone think two properly designed 20A radials are better - 8 amps more capacity, no ring continuity testing and better continuity of supply ?

Certainly simpler, simpler to test and, as you imply (and if I correctly understand you), any increase in the number of circuits will improve 'continuity of supply' (the ultimate being a separate circuit for each socket).

As for downsides, that's a least partially a function of the fact that "properly designed" is not completely possible with socket circuits, since one can only guess what is going to be plugged in where. At least theoretically, each of your radials would be unable to supply two 13A loads. With a 'fully loaded' 32A radial, it's quite possible that more than 20A will be (legitimately) drawn by one 'subset' of the sockets. It therefore really all comes down to how well a designer can predict/guess 'what will be plugged in where', and hence design the radials appropriately.

Kind Regards, John.

 

[ban-all-sheds]

It would - but, as I've just written, if you look around the world, ease of maintainenance/testing often doesn't feature much in design processes.

That's utter nonsense - manufacturers make strenuous efforts to design products in a way which makes them easier to maintain because that makes them cheaper to maintain and therefore gives them a lower TCO, which is a competitive advantage.

 

[JohnW2]

That's utter nonsense - manufacturers make strenuous efforts to design products in a way which makes them easier to maintain because that makes them cheaper to maintain and therefore gives them a lower TCO, which is a competitive advantage.

Well, those 'strenuous efforts' must be pretty lacking on occasions, then. Furthermore, the opposite also seems to not be uncommon - i.e. manufacturers apparently making strenuous efforts to ensure that a product is not repairable/maintainable, seemingly so as to increase their sales of replacements!

Whatever, please don't slide this (potentially interesting, at least to me) thread off on a tangent which could easily lead to 'thread degeneration', because I'm getting pretty sick of the manner in which you so often ruin potentially interesting threads in that way!

Kind Regards, John.

 

[ban-all-sheds]

And I'm getting pretty sick of the manner in which you think it's OK for you to introduce something into the discussion and then try to impose limitations on what other people may write in response.

 

[JohnW2]

And I'm getting pretty sick of the manner in which you think it's OK for you to introduce something into the discussion and then try to impose limitations on what other people may write in response.

Fair enough. Please see new thread.

Kind Regards, John.

 

[JohnW2]

Mind you, although it's obviously not mentioned in the regs in relation to ring fnals, I'm not convinced that one could find anything in BS7671 which actually 'oulaws' (albeit not law ) bridges.

In the light of the above comment, it occurs to me to ask those of who who undertake such things, if you were doing a PIR/EICR and encountered a 'bridged' ring final circuit, what clause(s) of BS7671 would you have in mind if you felt that it was a 'departure from BS7671:2008', and what 'code' (if any!!) would you give it (particularly with the EICR C1-C3 codes)?

Kind Regards, John

 

[Paul_C]

As for downsides, that's a least partially a function of the fact that "properly designed" is not completely possible with socket circuits, since one can only guess what is going to be plugged in where. At least theoretically, each of your radials would be unable to supply two 13A loads. With a 'fully loaded' 32A radial, it's quite possible that more than 20A will be (legitimately) drawn by one 'subset' of the sockets. It therefore really all comes down to how well a designer can predict/guess 'what will be plugged in where', and hence design the radials appropriately.

On that note, it's also curious that the Wiring Regs. have never seemed quite able to reconcile potential loads on rings versus radial circuits even of the same rating in a consistent way.

For example, in the 13th & 14th editions a 30A ring in a domestic installation could feed any number of 13A socket outlets serving up to 1000 sq. ft. (or 100 sq. m. in the revised 14th edition), while a 30A radial was limited to a maximum of just six sockets. If 30A was considered enough to feed a whole house with, say, a dozen sockets when wired as ring, why shouldn't the same dozen sockets on a 30A radial circuit be just as satisfactory?

In recent times we've lost the specific limitation on the number of sockets on a 32A radial, but with the stipulation of a maximum area of 75 sq. m, versus 100 sq. m for a 32A ring circuit. Again, why should 32A be considered sufficient to feed up to 100 sq. m in one case, but only 75 sq. m in the other?

 

[EFLImpudence]

I suppose it may be because, however you work it out (even the 20A stipulation), 2 x 2.5mm² has greater CCC than 1 x 4mm² (or the imperial equivalent).

 

[JohnW2]

In recent times we've lost the specific limitation on the number of sockets on a 32A radial, but with the stipulation of a maximum area of 75 sq. m, versus 100 sq. m for a 32A ring circuit. Again, why should 32A be considered sufficient to feed up to 100 sq. m in one case, but only 75 sq. m in the other?

That's actually effectively gone in the BGB, and wasn't even very strong in the BRB. In the latter those figures (in Appendix 15, which is only 'informative', anway) were prefaced with "As a rule of thumb...". In the BGB, that's been replaced with "Historically, the floor area served has been limited to ...."

The inconsistency you mention in the 'historical' figures is obviously crazy - but the figures always were ridiculous and meaningless. As I guess the IET have now decided, it's ridiculous to make statements about how many sockets, or what floor area, a circuit can serve without any reference to the use to which the space is being put!

Kind Regards, John.