Interesting - 434.4 (i) reads very much like 434.2.1 - but without the 3m rule of 434.2.1(i). It would seem odd to say that a single conductor without fault protection needs to be ≤3m long, but parallel similarly without fault protection could be any length, or should we reading 434.4 as specific requirements on top of the general requirements of 434.2.1 - and hence the 3m rule still applies? (which then begs the question of whether each parallel conductor counts as a "branch" from the other).

  -  Andy.

Interesting - 434.4 (i) reads very much like 434.2.1 - but without the 3m rule of 434.2.1(i). It would seem odd to say that a single conductor without fault protection needs to be ≤3m long, but parallel similarly without fault protection could be any length, or should we reading 434.4 as specific requirements on top of the general requirements of 434.2.1 - and hence the 3m rule still applies? (which then begs the question of whether each parallel conductor counts as a "branch" from the other).

  -  Andy.

after re-reading  few times and dwelling:  i take 434.2.1  as about position of device, where as, 434.4 is about when one device is not enough (and multiple might be needed, unless...)  and the device shall be at the supply end (and possible at the load end!).

the two are not related ?

It seems to me that (i) in 434.4  is  aiming to ensure that a multiple paralleled setup is mechanically protected against one of the conductors being 'taken out' and affecting the fault prot dev operation ... ?    (the ii and iii are about adding in more devices)

GKenyon was suggesting that SWA  is the default for direct buried, so perhaps (i) meant that more mechanical protection should be applied to meet (i)  and SWA alone might not do  .   I'm still interested in other opinions on this aspect and why the SWA around a multiple cored cable would not suffice   ... ?

Its not no fault protection- unlike the 3m rule that includes things like a 'self fusing'  5 amp feed to the lights on the multi KVA substation bus-bar, this is  a modest no (N)  conductors in parallel with protection that is sized for the group, and so presumably will operate but not without damaging the cable - just an N-fold overstress, as opposed to a potentially unlimited one.

I agree it is not well articulated, and may be almost by accident, but the 2 cases are not quite the same but get more similar for large values of N, but I don't think I've ever seen more than 4 cables in gang - it gets too unwieldy and 'tail wags dog' otherwise.

M.