"T i m" wrote in message
...
On Sat, 15 Jan 2005 04:55:43 GMT, "BigWallop"
wrote:


Well, I'm sorry too, because I can't find anything, anywhere, that says a
standard RCD in a normal domestic installation will trip the supply
unless
there is a fault from Phase to Earth.

Not sure this helps anything but when I was rewiring Dads garage a
while back I fitted one of those little RCD + 2 x MCB CU's.

I had powered it all up and was using the 'lighting' CCT and had the
live switched out on the MCB. I cut through the T&E on the 'Power' cct
and the RCD tripped (because the RCD had detected earth current
derived from the neutral potential as they were shorted by my
cutters?).

All the best ..

T i m


No it didn't, it detected induced current through what should have been a
neutral conductor, but which was then a phase conductor because the lighting
was still in use when you cut the cable to the socket circuit. I can bet
that the lighting and power cables run along side each other on some stretch
of the wiring plan, and it was either the induced current, even though very
small, or a larger impedance on the earth conductors that caused the RCD to
trip open. It could have either been the earth was not properly conducting
the residual current away from the induced leakage through the cable
insulation, and this created a higher potential in the neutral conductor, or
what should have been a neutral conductor had risen more toward phase
potential from some other means as yet undiscovered. This would cause a
Residual Current Devices to detect an imbalance between phase and earth and
activate the breaker circuitry.

If you have a look at the inner workings of an RCD you see there is no way
that a neutral to earth leak will cause the device to trip open. That's
unless there is an earth loop impedance fault or, to put it another way, a
higher resistance path on the earth conductors than the neutral conductors,
which will then causes the neutral path to become higher in potential energy
than earth and so cause an imbalance between a positive potential and
negative potential at earth (where common current flow specific is used).
This type of fault causes what should be a neutral potential to literally
become phase potential and so again cause an imbalance between phase
potential and earth potential which an RCD is designed to detect. He
http://www.memonline.com/rcd1.html for instance, shows how the device works,
with a diagram of the inner parts.

The whole point of equipotential bonding using larger cross sectional area
(csa') conductors than those used in the supply, is to greatly reduce the
impedance along these conductors. The reduction in impedance should be to
such a level that will not cause what should be a neutral potential to
obtain phase potential. This use of large csa' cables for earth bonding
should be capable of removing all residual current obtained by the neutral
paths, either through inductance or contact with a phase potential
conductor, and should also be capable of preventing the neutral path from
creating an imbalanced current flow between it and earth. Neutral and earth
should always remain at the same potential, or at a very negligible measure
from it, no matter where it is on the whole installation plan. This is the
requirement, stated in all the good guideline hand books and other
literature, for the earth bonding to have proper residual current removal
properties.

An example. Phase and neutral conductors can be created using, lets say,
2.5 mm csa' conductors but, to remove the residual current created by close
proximity between the two supply conductors and any other means of induced
current flow, the earth bonding conductor needs to be of a larger csa', of
more commonly 4 mm to 6 mm, to reduce its impedance further than the supply
conductors and so be capable of reducing induced current flow and prevent
nuisance trips on an RCD. So, larger cross sectional area cables are used
for equipotential bonding to produce a much lower impedance path to ensure
proper earth leakage current detection from the phase supply. This, in
turn, creates a much safer reaction time if such a leak to earth does occur.
It isn't just a small safety thing to use equipotential bonding, it is also
there to keep both phase and neutral paths in balance so that a higher
residual current, commonly 30mA, can be detected quickly.

It is natural to have a current flow of anything up to 12 mA, or even 15 mA
in some installations, between phase and earth/neutral conductors, even when
they are not under load. This is due mainly to induction through their
close proximity with each other inside the insulated outer sleeve of the
cable. The totally bare earth protective conductor is placed in the cable
to try to reduce this natural induced current flow phenomenon as much as
possible. The bare conductor should also help bring the two supply
conductors back in to a balanced state again. It isn't because the cable
makers are stingy with the insulation materials that the earth protective
conductor is left totally bare naked, it's because an insulated coating
would create a much higher resistance and reduce the bare conductors
capabilities in removing the residual current and so be unable to creating
some kind of harmony inside the cable again. Electricity suppliers also use
tests to make sure that the earth bonding at a supply is within certain
limits. This is to create a proper equipotential bond between all the
conductors. To high an impedance on any one of them, phase, neutral or
earth, will put the whole power balance out of sync and cause problems.

So, to sum up, I still say an RCD will not trip open unless the fault
current it detects is between a phase potential and earth mass.

Boy, can I go on a bit when I get started. :-) LOL !!!