I've lost track of what's going on in BS7671 these days, but shouldn't the question be whether there is anything which expressly forbids it? If there isn't something which explicitly says that it's not allowed, then you don't need an exception which permits it. Anything which is not forbidden is, surely, permissible by default?
two ring mains joined
- Thread starter visionset
- Start date
I've lost track of what's going on in BS7671 these days, but shouldn't the question be whether there is anything which expressly forbids it? If there isn't something which explicitly says that it's not allowed, then you don't need an exception which permits it. Anything which is not forbidden is, surely, permissible by default?
Sponsored Links
What do those sections say, specifically?
Quite - that is the (one and only) point I've been making. Unless I'm very much mistaken, there certainly is nothing in BS7671 which explicitly forbids cross-connections in rings. Indeed, there is only one short paragraph about ring finals in BS7671.
That would be my view - provided, of course, it was compliant with all relevant regs.
BAS is now (as so often) now relying on an essentially semantic argument. He is right in saying that any situation in which cable CCC is less than the In of the OPD protecting it is, in general, (explicitly) not allowed - and that ring finals are only allowed to exist because of the specific dispensation in 433.1.204. However, it only makes that dispensation (for Iz < In) for "ring final circuits" (defined in Part 2 as "a final circuit arranged in the form of a ring and connected to a single point of supply"), and BAS is arguing that what we are talking about is not "a ring", and therefore cannot enjoy that dispensation. I don't really agree with that, although he will undoubtedly produce dictionary definitions to support his view that what we are talking amount is not 'a ring'.
Kind Regards, John
OK - as I said earlier to John, you may have that figure-of-eight circuit if you want. But then there is a regulation which expressly forbids you from using 2.5mm² cable and a 32A breaker - 433.1.1.
Maybe it would have been better if this topic had started out like this:
Not if it's using 2.5mm² cable and a 32A breaker - if it is you need to change the breaker to a 20A one.
You are not mistaken about that.
But you are very much mistaken if you think that with those cross-connections you are exempt from 433.1.1 and can use 2.5mm² cable and a 32A breaker.
Change the MCB to a 20A and it will be.
No - I'm not going to do that.
As soon as you start in with your claim that somehow being concerned with what words mean make an argument invalid, and as soon as you start claiming that when the regulations say "a ring" they actually mean "a ring or any other sort of shape which I have decided is also a ring", I'm out, as I have learned from past experience that you will refuse, for ever and a day, or until you declare that you really don't care what the regulation say (whichever comes first), to listen to reason.
If it is not a ring then it is not a ring, and that is all there is, and all there can ever be, to it.
Sponsored Links
As I said, it's essentially a semantic argument about 'what is a ring?', and whether the reg was worded in a manner which correctly and fully reflected the authors' intent. Neither if us can know what was in the minds of the authors but we do know that the primary (essentially only) purpose of 433.1.204 was to 'dispensate' a ring circuit which had Iz<In - which, in the absence of the dispensation, would be in violation of 433.1.1.
The regs themselves are silent on many aspects of ring finals - for example, although they talk in terms of "with or without unfused spurs", they don't even say that those unfused spurs are limited to supplying one socket, and even the 'informative guidance' in an Appendix does not address the possibility of an unfused 4mm² spur. In the absence of such details in the regulations, IMO we can but attempt to apply electrical common sense in relation to those issues about which the regs are silent. Since adding cross-connections to a (compliant) ring would make the circuit even 'safer' in the situation in which Iz<In, I would have thought that it is reasonable to conclude that this would be within the spirit of what the authors had in their minds.
Kind Regards, John
Really? I wonder when that change happened, as they always used to have such restrictions. Under the 14th edition a non-fused spur was limited to supplying one double socket, two single sockets, or a single fixed appliance via an FCU. Later on it became one single or one double socket.
Are you sure that was actually ever "in the regulations" in the days of 14th ed., or subsequently? Very reasonably, there is still the 'guidance' about only one (single or double) socket on an unfused spur, but that is in ('informative') Appendix 15, not in "the regs" themselves. As I said, all there is in the regs themselves about ring finals is just one small paragraph, essentially just saying one can have them ('with or without unfused spurs') under the stated conditions.
Kind Regards, John
I forgot to add - supplying a fixed appliance via an FCU is obviously not a "non-fused spur". Not unreasonably, the 'guidance' in Appendix 15 'allows' one to connect as many sockets as one wants, or whatever, via a fused spur.
Kind regards, John
Positive. Sorry, I can't get at my copy of the 14th edition at the moment (boxes of books still packed away from the recent move!), but it definitely limited a non-fused spur to a maximum of two socket outlets.
Of course, not sure why I mentioned an FCU, but I recall the regs. did refer to feeding one fixed appliance on a non-fused spur.
After writing the above, I had a feeling I'd had a similar debate elsewhere some years ago and probably quoted the appropriate regulation. Sure enough, a search turned it up - Regulation A.40 of the 14th edition:
Interesting. As I said, in the 17th (and maybe earlier) the near-equivalent of that is certainly no longer in the 'regs proper' but has been moved to just being 'guidance' in a non-normative Appendix.
Kind Regards, John
'bye, then.
I think it's often assumed by many that the regs. have become more strict over time, but it's not always so, as this example demonstrates (and I had no idea that the non-fused spur limitations had been reduced to mere guidance until now).
Other things which come to my mind on the spur of the moment (no pun intended!) are:
1. The voltage warning labels we discussed before - Required for anything over 250V, then only required if >250V would "not normally be expected," now apparently not required for 240/415 or 240/480V at all;
2. Sockets on different phases or poles in the same room - Used to be prohibited in dweillings and only permitted in commercial situations if sockets on the same phase/pole were grouped together and in no case was a socket on one phase/pole within 6 ft. of a socket on a different phase/pole, now apparently there's no restriction whatsoever;
3. Relaxation of supplementary bonding requirements in bathrooms.
4. Increase in rated current-carrying capacity of cables.
5. Number of 13A sockets allowed to fed on a 20 or 30A radial - Used to be limited depending upon whether in a kitchen or elsewhere, then numbers were increased, now just any number within a certain floor area.
No doubt we could come up with more with just a little thought.
I believe I saw a comment somewhere that a stationary cooking appliance is no longer required to have a local isolation switch; if so, then that's another relaxation of the rules, since they always used to be required to have an isolator within 6 ft.
Last edited:
Indeed - and, although some of the changes in that direction do not seem unreasonable, some are difficult to understand, particularly the one we've been talking about. The restrictions on unfused spurs clearly should be part of the regs - since, otherwise, the regs might as well also allow 2.5mm² 32A radials.
It's actually 'worse' than you suggest. Not only do comments about floor area only appear as guidance in a non-normative Appendix, but what it actually says for 20A and 32A radials is "Historically, the floor area served has been limited to 50 m²" (for 20A, 75m² for 32A) (and a corresponding statement for ring finals mentions a historical limit of 100m²).
AFAIAA, there is no requirement in the regs for anything to have local isolation, certainly in domestic installations. There is a requirement that there should be 'a means of isolation' for everything - but that, of course, could be thge main switch in a CU!.
Kind Regards, John
I hadn't realized the area limits had gone - I wonder when that happened?
Under the 13th edition 30A radials were restricted to six sockets maximum; 20A radials could serve three sockets so long as all were within a single room of 200 sq. ft. or less and not a kitchen, otherwise they were restricted to two. And a twin socket counted as two sockets in all cases.
The 14th edition allowed a 20A radial to serve up to six sockets within one room of not more than 300 sq. ft. and not a kitchen, otherwise restricted to two as before.
Did the unlimited sockets and floor area limits come in with the 15th edition, or later? I'm not sure, and there may have been some other changes in between.
It's interesting that 30A radials (using appropriate 7/.036, later 4 sq. mm, cable) were restricted to six sockets while a 30A ring could serve unlimited sockets within a floor area of 1000 sq. ft. Did that ever really make much sense? And while the radial specifications changed multiple times over the years, the ring final has remained pretty much the same for decades, other than 1000 sq. ft. being rounded up to 100 sq. m with the change to metric in 1970.
Come to think of it, does the 100 sq. m ring final limitation still stand, or has that now been reduced to a "historical note" in an appendix too?
By contrast, if you want to compare with the NEC here in the States, the basic formula for determining the number and size of circuits required hasn't changed for over half a century, although admittedly it's slightly different due to major appliances each being given their own dedicated circuits, so we don't have dryers, washers, and other heavy power users all plugged into general-purpose branch circuits.
DIYnot Local
Staff member
If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.
Select the supplier or trade you require, enter your location to begin your search.
Are you a trade or supplier? You can create your listing free at DIYnot Local
Sponsored Links
