I don't have a copy of the regs here at the moment and if I did have mine here it would be 2001 version. I have 2001 OSG which simply states: "A non fused spur feeds only one single or one double socket or one permanently connected equipment."
and then goes on: "Permanently connected equipment is locally protected by a fuse complying with BS1362 of rating not exceeding 13A or by a circuit breaker of rating not exceeding 16A..." ... So that clearly states a fused spur is restricted to 13A fuse but a piece of fixed equipment to 13A fuse or 16A MCB.
Indeed - but, as so often, the OSG is not telling the whole story (and nor, probably, does the person who writes it even understand 'the whole story'!). What the OSG is talking about is a spur wired with 2.5mm² cable - which, as I said, is the only type of unfused spur which is depicted, or mentioned, in the guidance of appendix 15.
I fully understand there is little electrical difference between a row of sockets on a ring or on the end of a 4mm² spur but it goes against all I've ever learnt and the only document I have available to read. .... I assume the information contained in OSG is correct [yes I understand it is a guide] and, unless the regs have changed, it is clearly wrong to have more than a double socket on an unfused spur regardless of what size cable it is.
No - and that's because "the only document you have to read" is
not "the regulations". Nothing in the regs themselves says anything about spurs from ring final circuits, other than to say that such circuits may have 'unfused spurs' - so absolutely nothing about what may, or may not, be supplied by any sort of spur, and the only stipulation about the cable (by implication, this applies to spurs as much as to the ring itself) is that it must have a CCC of at least 20A and must have a CSA of at least 2.5mm² (unless it is MICC, in which case 1.5mm² is allowed).
Nothing in the regulation which permits ring finals ('with or without unfused spurs') (433.1.204) over-rides the general provisions of 433.1.1, the effect of which is to allow a 4mm² cable protected by a 32A OPD to supply as many sockets as one wants - whether that cable is a spur from a ring final, a radial or anything else you can think of!
Those who point out that an unfused spur supplying multiple sockets which originates close to the origin of a ring could theoretically result in overload of part of the cable are correct - but exactly the same risk exists if multiple sockets are installed 'on the ring' close to one end of it. As required by 433.1.204 t is for the designer to minimise such risks by appropriate selection of the location of sockets 'on the ring', or of the location of the origin of a spur from the ring.
Kind Regards, John