Gary Cavie wrote:
In article ,
says...

Having recently changed my CU, on a 1980 house, I found the wiring to be
in excellent condition. No insulation resistance problems, visually the
wiring looks fine. The only point to note is that the cpc on the 2.5mm
T+E is 1mm, not 1.5mm. MCBs disconnect in the event of a fault current
much faster than fuses - if the cable is protected with a fuse, then the
disconnect time and resistance is such that the cable will get too hot
in the event of a fault, possibly causing damage (as I understand it, it
*isn't* a fire risk). Since I now have a shiny new CU with MCBs, no problem.



If the total Zs of the circuit is within the specified limits, the fuse
/ MCB will go within the time required (0.4s or 5s generally). For a 30A
BS3036 fuse, the max Zs is 1.14 ohms, for an equivalent BSEN60898-B MCB
rated at 32A, it is 1.50 ohms. If your circuit is within the BS3036
figure, changing to MCBs is going to really achieve anything, certainly
not on safety grounds. It just makes it more convenient to reset.

Those are the maximum impedances to achieve the disconnect time, but not
to avoid thermal damage to the cable.

t = k^2S^2/I^2

Zs is 1.14ohms, so I is 240V/1.14ohms == 210A
t is therefore 0.299s, and if the disconnect time is any longer, the
cable may be damaged by heat. A type B MCB will disconnect much faster
than that because I/In 5, but looking at
http://www.tlc-direct.co.uk/Figures/3.13b.gif
a BS3036 30A fuse will just meet the 400ms disconnect requirement, but
certainly not 299ms.

If we required that Zs was no more than 0.69ohms (plucking a figure from
the OSG), then t becomes 100ms, and S is 350A. At that point, a BS3036
30A fuse still doesn't disconnect fast enough - but a BS88 does - just.
Which is, as far as I can see, why BS88 is okay with 1mm cpc, but BS3036
isn't.