Terry Pinnell wrote:
The Natural Philosopher wrote:

Terry Pinnell wrote:
Terry Pinnell wrote:
Hasty correction!

---------
OK, but this is where I'm still obviously being dense. Given a normal,
properly working setup, can you please summarise what *should* happen
and which switches in the main CU should open for each of the
following events:

1. A device on the LH side (non RCD) develops a short.

My assumption: it opens that MCB, but power remains for all other LH
and RH (RCD) circuits; and it probably blows a plug fuse too.

(The 'main LH switch' is MANUAL only, as you pointed out.)

Not necessarily. See previous post. RCD's may trip as well.

So that could be a possible explanation for the behaviour described in
my opening post?

I have to say I was too busy to read it properly then.

But if you had an MCB and an UNRELATED RCD trip, then yes.

ANY spike on ANY line can trip an RCD if there is enough capacitance
around, and with a few computers and so on, around, there generally is..