Seem to be on the increase.
Makes me wonder if we should be testing samples before installation ?
A full and proper test requires relatively elaborate lab facilities, but perhaps a crude test involving say a vehicle battery and a 100 amp fuse.
This wont reveal all faults, but would at least weed out the most blatant "no trip" types.
An MCB tested thus might of course have been damaged, so perhaps test the odd sample and only install untested examples from the same batch. Rather expensive though.
youtube. fake MCB dismantled
