well two spurs from the same point is fine from a cable rating perspective but may need a chunky junction box to fit the total wire diameter.

A daisy chain spur is not so clever, at least in 2,5mm as the cable may be overloaded - but if you know for sure the loads are only low current not that big a deal. Problems may arise later when the house ownership changes.

The whole 'ring' thing is advisory -  there is nothing to stop you doing a 40A ring in 4mm or a lasso/lolipop final, or a spur in 4mm from the middle of a 2,5mm ring but you need to think to design it for the situation, and be happy in your conclusions.

Most real sparks are quite risk averse, and prefer to do more of the same as they always have.

Generally, rather like the crab-claws broken ring that is more like a centre fed radial,  I think I'd consider recommending  fusing down the whole ring to 25A or even 20A  unless there are good reason to be sure that in the specific case an overload is not credible.
Only if evidence of overheating would it be immediate danger.
Mike

Thank you for the response. 

If one was looking at it from an EICR point of view and based on your comments above, the 'recommendation' to be made   is based on the 'possibility of overload' (by the fact that there is more than one unfused spur socket in either arrangement I mentioned) ... so what is that then:  is it a potentially dangerous situation (based on 'it might happen one day'), or simply an improvement  ...   or is it just an additional comment with no severity rating as such ?

Why again is  one socket + one unfused socket connected to it,  not a similar concern  to make a recommendation on ?

(edit: is the presumption that no more than 26A will be taken from both the socket + the unfused one - 13A each - , and thats the basis)

Thank you again

" Such circuits are deemed to meet the requirements of regulation 433.1.1 if the current -carrying capacity (Iz) of the cable in not less than 20 A and if the under intended conditions of use, the load current in any part of the circuit is unlikely to exceed for long periods the current carrying capacity (Iz) of the cable."

433.1.204

Z.

I agree with Zoom, that’s the appropriate regulation. 
interesting what Mike says about “real” sparks being quite risk averse. I concur, and this is being reflected in the EICRs they produce. Any deviation with the regulations seems to sprout a code 2. I don’t blame them but I would encourage them to stand back, draw breath and explain, if only to themselves, how a particular issue is potentially dangerous. It takes courage, confidence and solid experience in periodic inspection and testing to make a determination that an issue, whist not complying with the regulations or custom and practice, can be dismissed as something of no serious concern.

most will have read that of course.

regards

Thank you for the contribution.

Well as I replied to Zoom, most will be already aware of that Reg wording (or one would hope).  One might conclude that the Reg is aiming at mainly known connected loads in a RFC design, rather than the possible use of a general ring final socket circuit so to speak where that most likely would  not be the case.

Your inference(s) was part of the reason I posted the 'musing' on the scenario I detailed.

Of course the EICR has a purpose and the 'signature' implies certain purposes have been met.

In a domestic there is also a consideration that the installation is under the operation of general ('ordinary') users.

All that aside...and to comply with BS7671 on the EICR matter:

I wonder what a selection of EICR's would produce for the scenario I detailed - as generally it doesn't require sight or further info really in my view (it is what it is),  to be able to offer an observation for it.

Without any other obvious issues (like too many conductors in the back for the terminal spec, or heating affects etc), for me it is no more than 'room for an improvement'  as a belt and braces.  That's my take.

433.1.204 applies to B.S. 1363 accessories. They may be supplied from a ring final circuit with or without unfused spurs.

Z.

You might struggle to get 4 x 2.5 mm² conductors into an ordinary 13 A socket - the BS specifies min 8 mm² , but some manufacturer's terminals will accept 12 mm² (which is quite handy if you want to branch a radial).

Fig 15A does not preclude multiple spurs from one point, but they must feed only one socket, so daisy chains are out unless you start with 4 mm².

I suspect that most sockets nowadays feed very little - table lamps, computers, screens, phones, etc., but we still must allow for that 3 kW heater on a cold day.

yes.

One cant predict the loading on a general rfc circuit where there are no known [e.g. fixed] loads.

regards.

one manufacturer's product I noted specifies max. of 3 of 2.5mm conductors .