In article ,
chris French writes:

Heard a sort of chattering sound from the MCB (30A) supplying one of the
ring mains in our house this afternoon. Turn of the MCB, turn it back on
again, all is ok for a little while (maybe 10 minutes?) then it starts
again.

The circuit in question is one of four ring mains in the house, it
supplies the sitting room, one bedroom , a socket in the hall, in the
conservatory and the dressing room. generally lightly loaded - a few
table/standard lamps, TV, VCR, DVD etc.

Whilst investigating this problem I noticed that when this circuit was
switched on the disc on the meter started fair whizzing round - 30
rev/minute up from the say about 3 rev/minute is was doing normally.

My quick calcs indicate that 30 rev/min implies something like 10kW/h
being consumed - around 40A? , or have I made some big boo-boo?

Anyway, there is nothing on the circuit drawing anything like that (no
we don't have any secret electric showers, or unknown immersion heaters
etc.) These symptoms continue to occur once everything on the circuit
has been disconnected.

I've not had time today to investigate this more thoroughly, the circuit
is isolated now via the MCB pending time tomorrow.

I wonder if next door's cannabis plants are now in the dark?

So, my thought so far is that there is some fault in the fixed wiring,
causing a high current draw, and that the MCB is faulty in some way,
hence the noise rather than tripping.

Does the above sound a reasonable diagnosis?

It's very unlikely that a fault in the wiring dissipating 10kW
will stay that way for more than a few seconds. It will quickly
burn out, either sort or open circuit. Besides which, you'd be
quite likely to notice 10kW of heat coming into the house from
an unexpected location.

I would not be surprised at a 30A breaker making a noise at 40A,
and I wouldn't immediately assume something wrong with the breaker
in this case. It's probably right on the virge of tripping.

Seems like you need to find what the high load is, which your
meter is also confirming.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]