wrote:


The new unit behaves differently, but it does have one very, very
annoying habit. Every so often without any rhyme or reason that I can
tell, part of the thing trips and all the sockets in the house go
dead. The lights are fine, but nothing else. I asked the electrician
to test it, but he simply suggested unplugging things until it stops
happening. Yet more blinking 12:00s to play with.

It sounds like you have a nuisance trip issue with a split load CU. This
is not common, can happen and is as you have discovered a pain to fix!

Of the two "master" trips to which you refer, chances are one is just a
main switch, and the other is the RCD (you can have RCDs in both
positions but that is currently not that common unless you are out in
the sticks).

A photo would be handy.

Fine, except that it might happen four times in one day, and then not
happen again for three days, whereupon it might happen another six
times in three hours.

Tracking the cause of these trips can be time consuming. You can read
some background on the subject he

http://wiki.diyfaq.org.uk/index.php?...Nuisance_trips

I don't have the correct terms for all the bits and bobs of the
consumer unit, but it's divided into two rows of eight trips (what
would, I suppose, have been fuses in the old days, what I think are
now called MCBs) with each row having a master "trip" (an RCD?) and an
over-arching master "trip" (a super RCD?) for the whole shooting
match.

Your naming sounds about right. The "super" RCD may just be a switch as
mentioned before.

It's the lower master "trip" that trips - the one controlling the
second row of individual trips. These trips are labelled "Hob",
"Sockets - kitchen", "Sockets - Upstairs", "Sockets - Garage",
"Sockets - Extension", "Immersion Heater", "Cooker", "Hall/Lounge
sockets downstairs".

It sounds like there are some design failings to be addressed there -
the arrangement of circuits is not ideal with various things that
represent low electrocutions risk, but high nuisance trip risk being
connected to the RCD side of the CU.

Given that the rate of failure is so low and MTBF so long (for
elimination purposes), and that the sub-master trip is what goes
(rather than the individual trip for the misperforming part of the
system), which doesn't exactly narrow possibilities down, does anyone
have any ideas how we could go about finding the problem? Electrician

The individual trips will only trip under different circumstances anyway
- when something draws too much current. The RCD trip results from a
different cause.

thinks the prime candidate is the fridge, which is the one thing that
we can't realistically turn off and see if everything carries on
working (plus it didn't trip for the 9 days we were away). Would
using an RCD on the fridge tell us anything - such as this

Not really - even if it was the main source of leakage it might not on
its own cause a trip.

http://www.maplin.co.uk/Module.aspx?...FQaLEgodbE-F2g

http://tinyurl.com/26zt6u?

Could it be a fault in the consumer unit itself?

Not very likely. It could just be a fault in the house wiring though.
(e.g. neutral to earth short)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/