'Bridged' Ring Final Circuits

[JohnW2]

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).

I think it may be you that is being 'too logical' this time

Whilst what you say is true, in theory the OPDs should limit the current to within the cable's CCC in either case - and, in any event, there's always that lurking Devil's Advocate who will postulate a full 32A load right at one end of your ring!

Kind Regards, John.

 

[Paul_C]

there's always that lurking Devil's Advocate who will postulate a full 32A load right at one end of your ring!

And that is a situation in which a "bridge" across the ring could improve the distribution of current considerably, even to the point of preventing the short leg of the ring from being overloaded.

Take that situation where there is, say, a ring running for a total length of 200 ft. with a double socket in a utility room located at a point 10 ft. around the ring. Plug two 13A loads into that socket simultaneously, and the 10-ft. leg of the ring will carry 24.7A while the long 190-ft. side of the ring will carry just 1.3A. Now assume it's convenient to connect a bridge between that socket and another which is on a point 10 ft. from the other end of the ring, and that the length of that bridge cable is 20 ft. The current in the short leg which originally carried 24.7A has now been reduced to 19.2A, while the current in the other leg back to the distribution panel has been increased to 6.8A (rounded). Of that 6.8A, the bulk - 6.12A - will flow through the bridge connection, with the remaining 0.68A flowing through the 180-ft. remaining length of the original ring.

So in this case, the addition of the bridge has possibly eliminated the chance of an overload on the original 10-ft. length of ring from the panel to where the 26A load is connect

 

[EFLImpudence]

Post moved.

 

[JohnW2]

To aid clarity, some exchanges shifted from the spin-off thread ....

Athough this should really be in the original thread ....

I'm beginning to wonder what exactly you think may be the outcome of 'recommending' bridges.

As you know, I have never recommended, or suggested that anyone should recommend, bridges. I merely asked why it was that everyone seems to regard bridges as 'a problem' when, as far as I could make out (by application of engineering logic), a ring with bridges is no less safe than one without bridges (and, indeed, on the contrary, it seemed to me that the existence of bridges would tend to have beneficial, rather than detrimental, effects). You are about the only person who actualy answered my question - and your answer was that you could not fault my logic - but you did add that the 'mess' would create inconvenience in terms of testing and circuit modification.

Obviously, It would make the shorter ring, created by the bridge, less unbalanced and the sockets within that shorter ring better protected but to what end? Would we be limited to one per ring or as many as we wanted - ending up with the new 'Ladder Final Circuit'? It would appear to be making safer things which were satisfactory already or only a means to counteract an initial poor design.

As above, I was not suggesting/recommending - merely asking why it actually seemed to be regarded as 'a problem', perhaps even 'dangerous',since I couldn't understand why that was the case.

... and no-one has yet answered my question as to how (if at all) electricians would 'code' a bridge which they detected, and what parts of BS7671 they would cite in support of that coding.

I merely asked why it was that everyone seems to regard bridges as 'a problem'

I would say because it makes testing difficult.

... and no-one has yet answered my question as to how (if at all) electricians would 'code' a bridge which they detected, and what parts of BS7671 they would cite in support of that coding.

There is a reason for that, as you well know.

Obviously, if coded at all, it would have to be the new C3.
However, if after asking on here for the appropriate regulation and ten pages later no regulation has been found then we would deduce that it did not need coding (or was not able to be coded).

Or, if after such a problematic test procedure, would the electrician not just remove the bridge?

I merely asked why it was that everyone seems to regard bridges as 'a problem'

I would say because it makes testing difficult.

The 'everyone' I was talking about wasn't primarily those who do the work, for whom it obvioulsy is 'a problem' (well, inconvenience, provided they are competent). I was thinking more about the 'books' - many of which say or imply that a bridge is a 'serious problem', some of them citing the 'broken ring' scenario as being the reason a bridge is dangerous (even though you and I agree that it is no more dangerous than a break in a ring which doesn't have bridges). I strongly suspect that a substantial proportion of electricians believe that there is some significant, perhaps major, engineering problem with the presence of a bridge.

... and no-one has yet answered my question as to how (if at all) electricians would 'code' a bridge which they detected, and what parts of BS7671 they would cite in support of that coding.

There is a reason for that, as you well know. Obviously, if coded at all, it would have to be the new C3.
However, if after asking on here for the appropriate regulation and ten pages later no regulation has been found then we would deduce that it did not need coding (or was not able to be coded).

Quite - and I really don't think anyone should code it unless, if challenged, they could cite a regulation to support the coding.

[Or, if after such a problematic test procedure, would the electrician not just remove the bridge?

That's an interesting one. I think you and I are agreed that differences between ring finals with and without bridges are in the direction of the ones with bridges being marginally 'safer'. Could an electrician justify making the installation arguably a little bit less safe in order to make life for the next electrician a little easier? As I said above, convenience should not really take priority over safety.

 

[JohnW2]

there's always that lurking Devil's Advocate who will postulate a full 32A load right at one end of your ring!

And that is a situation in which a "bridge" across the ring could improve the distribution of current considerably, even to the point of preventing the short leg of the ring from being overloaded....

Indeed so. I've thought quite hard about this and, so far, I can think of no situations in which the presence of bridges in a ring binal circuit make the circuit any less satisfactory in engineering terms, or any less safe - whilst, as you say, there are certainly some scenarios (and fault situations) in which the converse is true.

Kind Regards, John

 

[EFLImpudence]

Reposted in the correct position

[Or, if after such a problematic test procedure, would the electrician not just remove the bridge?

That's an interesting one. I think you and I are agreed that differences between ring finals with and without bridges are in the direction of the ones with bridges being marginally 'safer'. Could an electrician justify making the installation arguably a little bit less safe in order to make life for the next electrician a little easier? As I said above, convenience should not really take priority over safety.

I suppose.
Less safe it may be, strictly speaking, but nevertheless as safe as originally intended. As I previously said anything can be made safer by adding unnecessary additional cabling.


What about 523.7 BGB (conveniently 523.8 BRB) Conductors in parallel.

Stating that conductors in parallel must be approximately the same length and run together and not be branched,
it then states that this does not preclude the use of ring final circuits with or without spur connections.

Does that mean bridges ARE allowed or not?

 

[JohnW2]

Could an electrician justify making the installation arguably a little bit less safe in order to make life for the next electrician a little easier? As I said above, convenience should not really take priority over safety.

I suppose. Less safe it may be, strictly speaking, but nevertheless as safe as originally intended.

As I said, it's an interesting one. However, I would not like to be the person who had to justify having made a circuit (slightly) less safe, even if it was still 'as safe as it needed to be' - particularly if my only 'reason' was to make life easier for the next electrician who had to deal with the circuit. Would you be comfortable doing that 'justifying'?

As I previously said anything can be made safer by adding unnecessary additional cabling.

Indeed, and my response would be the same. If someone had made an installation 'unncessarily safe', could you happily justify making it less safe, even if still "as safe as it 'needs' to be"?

What about 523.7 BGB (conveniently 523.8 BRB) Conductors in parallel.
Stating that conductors in parallel must be approximately the same length and run together and not be branched, it then states that this does not preclude the use of ring final circuits with or without spur connections.
Does that mean bridges ARE allowed or not?

I certainly don't think that it means they are not allowed. I think you have overlooked the 'or' option [ 523.7(ii) in BGB ] which essentially says that any parallel conductors are OK provided that CCCs are not exceeded - which will presumably always be the case with 'redundant' bridges.

Kind Regards, John.

 

[EFLImpudence]

As I said, it's an interesting one. However, I would not like to be the person who had to justify having made a circuit (slightly) less safe, even if it was still 'as safe as it needed to be' - particularly if my only 'reason' was to make life easier for the next electrician who had to deal with the circuit. Would you be comfortable doing that 'justifying'?

Why would you have to justify removing a belt whilst leaving the braces in position?

Indeed, and my response would be the same. If someone had made an installation 'unncessarily safe', could you happily justify making it less safe, even if still "as safe as it 'needs' to be"?

Same answer.

I certainly don't think that it means they are not allowed. I think you have overlooked the 'or' option [ 523.7(ii) in BGB ] which essentially says that any parallel conductors are OK provided that CCCs are not exceeded - which will presumably always be the case with 'redundant' bridges.

Yes, true.

I'm still trying to think of a situation where the intentional inclusion of a bridge would be advantageous rather than better initial design, i.e. using two radials.

 

[ban-all-sheds]

A bridged ring is harder to build. Testing is part of the "building", i.e. it's done before the product is considered finished, i.e. ready to use.

If a fault during the assembly results in a continuity break in a simple ring that will be detected during testing. A bridged ring could easily have a break in it which will not show up when testing at the origin - all the tests there would look OK, and might be very hard to detect when testing at each socket.

Or impossible to detect - I don't have time right now to try and work it out, but would you not lose all predictability of readings at each outlet?

 

[JohnW2]

Why would you have to justify removing a belt whilst leaving the braces in position?

As I implied, I think it's a personal thing. If you would be happy to explain/justify/whatever that you felt that reducing safety was OK, since it was still 'safe enough' (at least safe enough to satisfy regulations), then fair enough. As I said, I'm not sure I would be that comfortable. I wonder how others feel?

I'm still trying to think of a situation where the intentional inclusion of a bridge would be advantageous rather than better initial design, i.e. using two radials.

As I've said, this didn't really start with my suggesting that anyone should ever intentionally include a bridge in a designed circuit - I merely asked what was actually wrong with the situation which arose if someone had done it (virtually always unintentionally).

As for your question, 'safety wise' the only advantage (over radials) which comes immediately to mind is even more potential protection against a broken CPC than is afforded by a ring without bridges. However, in the spirit of my question, I think we really should be comparing a standard ring final with a ring final + bridge(s) [not comparing a ring final + bridge(s) with radial(s)] - in which case there are seemingly more advantages.

In terms of 'assured safety' in relation to CCC of cables, radials will, of course, always win over rings of any sort provided that the circuit is used within its design parameters. However, in relation to (16A/20A) 2.5mm² radials, I would suggest that they are, in practice, more likely to be 'moderately overloaded' in relation to their Ib (and In of OPD) than a 32A circuit (ring or 4mm² radial) - but that's only because of the lower design current of the circuit.

Kind Regards, John.

 

[Spark123]

It would be a nightmare to design, given different amounts of current will flow each way around the ring to the point(s) of utilisation (owing to the impedance of the cables), i.e. if you only draw 32A at the a single point (for simplicity - you aren't meant to do this) you are not guaranteed to have 16A flowing equally through the two legs. You may have 20A down one and 12A down the other.
Fire a bridge in there somewhere and there and the impedances will change hence you may end up with more current flowing through one leg than designed.

 

[EFLImpudence]

However, in the spirit of my question, I think we really should be comparing a standard ring final with a ring final + bridge(s)

Fair enough.

I suppose then the bridge can be regarded as a 'mistake' which paradoxically makes the circuit electrically safer, but this no more than a DIYer using larger cable than necessary because he knew no different.

However, should there be a couple (or more) of these mistakes - not necessarily 'straight across' the ring as we are probably imagining. I feel the testing is more of an issue than you seem to acknowledge.
It would be mandatory on checking before changing a CU, for example, and would increase the time taken considerably.

So, that the bridges are not dangerous does not detract from the fact that they are mistakes and, as such, should be rectified.

 

[JohnW2]

A bridged ring is harder to build. Testing is part of the "building", i.e. it's done before the product is considered finished, i.e. ready to use. If a fault during the assembly results in a continuity break in a simple ring that will be detected during testing. A bridged ring could easily have a break in it which will not show up when testing at the origin - all the tests there would look OK, and might be very hard to detect when testing at each socket. Or impossible to detect - I don't have time right now to try and work it out, but would you not lose all predictability of readings at each outlet?

I agree that testing continuity would be a nightmare. As for your last comment, I haven't analysed the situation in detail, either, but (particularly if one didn't know how many bridges there were, or where they were meant to be) I think it probably would be impossible to draw any conclusions about the integrity (continuity) of all conductors without locating and disconnecting the bridge(s) - just look at resistance measurements in the intact circuit would almost certain leave ambiguities and uncertainties.

Kind Regards, John.

 

[JohnW2]

I suppose then the bridge can be regarded as a 'mistake' which paradoxically makes the circuit electrically safer, but this no more than a DIYer using larger cable than necessary because he new no different.

Yep - although I presume that you would not then propose removing the unnecessarily large cable and replacing it with more modest cable?

I suppose However, should there be a couple (or more) of these mistakes - not necessarily 'straight across' the ring as we are probably imagining. I feel the issue of testing is more of an issue than you seem to acknowledge.

Nope. As I've just written in response to BAS, I fully acknowledge that testing could be a nightmare.

It would be mandatory on checking before changing a CU, for example, and would increase the time taken considerably.

That's a fair enough argument in relation to the consequences of deliberately installing bridges, but is not actually very true in relation to the 'accidental' bridges (which are, in practice, the only ones which exist). Your pre-CU change testing will result in your identifying the presence of,and locating, the bridge(s). Having done that, I imagine it would take less time to disconnect the bridges, check the contnuity of the main ring and the bridge(s) and then re-instate the bridge(s) than it would take to re-wire so as to eliminate the bridges (and then re-test), wouldn't it?

So, the fact that the bridges are not dangerous does not detract from the fact that they are mistakes and, as such, should be rectified.

...but only because it's a 'mistake' which has nuisance value to electricians. As above, I don't think you would replace the oversized cable, which is also a 'mistake', would you? I suppose one also has to consider what one means by a 'mistake'. If, as so far remains the case, no-one can find a way in which such a circuit is non-compliant with BS7671, in what sense is it 'a mistake which has to be rectified'?

However, most of this is academic, since you have answered my question. You have agreed with me, several times, that, if anything. bridges increase (not decrease) the safety of ring finals, and therefore that the seemingly widely-held belief (in books, and I suspect the minds of many electricians) that they represent a major engineering problem, and are hence dangerous, is simply an urban myth (aka 'rubbish'!). We seem agreed that the only problem with bridges is the difficulties in testing which they present.

Kind Regards, John.

 

[JohnW2]

It would be a nightmare to design, given different amounts of current will flow each way around the ring to the point(s) of utilisation (owing to the impedance of the cables)...

It would be a nightmare to design, but no-one is suggesting that anyone should design such a circuit - the discussion is about how we should view bridges when they have arisen 'unintentionally'. Mind you, even a standard ring final cannot be properly designed, since one never knows what is going to plugged in, and rare. If one did have knowledge about the size and location of loads, it would not be all that difficult to design an 'optimised' ring with bridges, if one so desired.

i.e. if you only draw 32A at the a single point (for simplicity - you aren't meant to do this) you are not guaranteed to have 16A flowing equally through the two legs. You may have 20A down one and 12A down the other. Fire a bridge in there somewhere and there and the impedances will change hence you may end up with more current flowing through one leg than designed.

With again the caveat that I'm not suggesting that anyone should intentionally include bridges .... I think if you consider this carefully, you will probably come to the same conclusion as me - that bridge(s) can never increase the imbalance between the two legs of a ring. In very specific (and very rare) circumstances, bridges may not affect the balance, but in all other (i.e. virtually all) cases, the effect of one or more bridges will be to bring the currents in the legs closer to balance. In other words, if a standard ring is loaded in such a fashion that the CCC is not exceeded in either leg (or any part of the ring), then I don't think that the addition of bridge(s) could ever result in CCC being exceeded anywhere. However, I need to think a bit more about that before being absolutely certain that I'm right!

Kind Regards, John