In article t,
Chip writes:
On Thu, 07 Jul 2005 17:25:28 +0100,it is alleged that Andy Wade
spake thusly in uk.d-i-y:

Chip wrote:

I just don't like the concept of the protective conductor being a
smaller size than the line conductors potentially feeding fault
current into it.

There's no logic to that. Protective conductors only have to be sized
to withstand the earth fault current for the time it takes to clear the
fault.

That is where I have issues with the concept, I admit my calculations
verge on the 'worst case scenario' but a 1.5xI(n) fault would not blow
a 30 amp rewireable fuse in under a few minutes, and 45 amps flowing
through 2x1.5mm earth wires seems a generally bad idea to me. I am at

45 amps will be generating some 10kW at the scene of the fault.
It would be pretty impossible to even deliberately engineer such
a fault which lasted more than a few seconds, without it either
shorting completely or blowing itself out into an open circuit.
The generation of 10kW will change the nature of a fault site
very quickly.

There just aren't credible scenarios where large currents flow
for an extended period in the earth conductor.

Live conductors [1] have be able to withstand continuous
full-load design current as well as (where relevant) occasional
short-term overloads. The principle of using 'undersized' CPCs is very
well established, both in theory and practice, and has been so for a
very long time. The practice doesn't compromise safety; in fact it
avoids over-engineering and saves copper.

I think the differences are not as major as people at first think, in
the US the neutral on the drop (TN-C-S system before the split) is
often slightly undersized, they just don't extend it to final branch
circuits, we in the UK do, and many other european countries disagree
with this practice.

On a fully loaded US system, the neutral current is zero.
Same is true of a fully loaded 3-phase system in UK, which
is why 4-wire 3-phase circuits do sometimes have reduced
size neutrals (need to watch out for 3rd-harmonic components
though, which do add in the neutral rather than cancel out).

--
Andrew Gabriel