I'm understanding that you have a ring final circuit, run in 4 sq mm T+E and it's grouped with 5 other cables, and it is all under 50 mm of insulation.
It's fairly simple until you get to the grouping.
Ci - 50mm insulation of 0.78
No. Table F2 is
length in insulation, not depth of insulation.
However, read the Note on first page of appendix F, Ci is 1 if you use the tables in that chapter to get the CCC.
Then I get stuck here:
Ib < In < It < Iz
For a ring final circuit, reg 433.1.204 is a special case that overrides 433.1.1.
From 433.1.204
Minimum cross sectional area of 2.5 sq mm (tick)
Such circuits are deemed to meet the requirements of 433.1.1 if the current carrying capacity of the cable is not less than 20A and (paraphrasing) no bit of the circuit will pull > 20A for a long time.
So you need the tabulated current carrying capacity of 4 sq mm PVC (I think you had the wrong table above). It sounds more like reference method 100, which gives you 27 amps. This exceeds 20 amps and is deemed to meet 433.1.1 if no further derating considerations apply.
So far so good. (Why are you saying 103 as that is for inside a stud wall?)
The complication is the grouping as you have 6 cables together for a long distance.
I agree that you are interpreting section 7 of the OSG correctly, the three exemptions don't apply and it refers you to Appendix F. (Personally I make flipping sure I don't end up in this situation! Couldn't you have used more than one hole?!)
This is where it gets complicated. The lb for a RFC is 32 amps but each cable doesn't have to carry 32 amps because of 433.1.204.
If you proceed anyway and correct Lb to 56 amps, then half it (28 amps) then I think that is a reasonable attempt.
In reality it won't be a perfect 50/50 split.
28 amps is slightly over 27 amps of course. Maybe an ambient temperature correction could save you?!
I think the real solution to be honest is re-route a couple of them in a 2nd hole.
I'd be interested in the views of some of the forum legends on this one.
@davesparks @pc1966
Out of interest, if you do the grouping derating for the other circuits passing through these joists do they stack up?