How to calculate ampacity of 3 phase 4 wire singles installed as quads spaced 2De not trefoils?

I note multicore 3 core and 4 core (and 5 core including earth) have the same ampacity

That seems to be the general rule - just count the loaded conductors - and of course in conventional 3-phase circuits the most onerous is when all three phases are fully loaded - and so N carries zero current, so still just 3 cores carrying current. In other situations any N current (due to imbalance) is compensated by an equal reduction in current in one or more of the line conductors, so the overall heating effect isn't increased.

Granted the presence of an extra (unloaded) conductor will make some difference - it's solid presence may restrict heat loss by convention, but at the same time it may provide a extra path to take heat away by conduction and provide a larger surface area for overall heat loss. All cable ratings are approximations, and in the great scheme of things, unloaded cores/cables seem to make a small difference.

If it's a datacentre, with presumably lots of switch-mode PSUs, my worry would be more about currents not cancelling out in the N and you actually have 4 loaded cores rather than the 3 that usual 3-phase theory predicts.

   - Andy.

Thanks for the response. My concern is that Table 4D1A Ref F ONLY refers to 3 cables, 3 Phase. Ref A, B, C refers to 3 or 4 cables. Table 4D2A Ref A, B, C, E refers to 3 or 4 cables. What ampacity should one use 4 for say single quads, separated by 2De on a cable ladder. I accept your comment about harmonics and, in fact, PF may be leading or lagging 0.95. I Dont think Ref F trefoil can be correct. I see Ref F 3 cables touching flat has a higher ampacity!

Be very aware that the current ratings in the appendices to the regs are not an exhaustive set, and they include a number of assumptions that are quite sweeping in their effect - for example in most cases cables routed vertically and horizontally may have the same rating-  clearly a moments thought about convection shows that cannot be right. In the same way cables heated by the rising hot air from cables below should have a lower rating than the same size cable at the bottom of the stack. aAlso if there is a ceiling or a solid shelf above the top of the ladder. that will restrict the free passage of the hot air.

If your PF is 90-95% then your loads are substantially resistor-like and when all 3 phases are equally loaded the neutral will be cold. When they are not equally loaded, at least the one of the phases will be running colder. So your quads will always behave more or less like a trefoil with a freeloader alongside.

How are you arranging your bundles in the vertical/ horizontal -  A 3 wide stack 2 high will cool better than a 2 wide stack 3 high, for example.

Do you have a copy of the old ERA tables? (vol  5 I think, but I will check, is the one closest to what you describe) They have some design rules that may be helpful

Are your cables AWA and if yes,  how are you terminating the armour - there will be armour currents, and these can add a few percent to the ratings - not serious usually, but if you are right on the edge of a jump to the next size up it is worth looking at tricks to reduce that.

BS7671 ignores that and tells you for XPLE  96% or 93 % of 625amps per phase depending on the stack-up.

Mike.