He says he did test the insulation and recorded all the columns as 1000mohms because all the readings were over 1000.
Quite unusual - I have never yet found an installation old enough to need a replacement consumer unit which returned IR values on all conductor combinations in excess of 1000Mohms - but let's give him the benefit of the doubt. (He should have recorded the values as greater than 1000, though.)
The RCD works by monitoring the currents in the line and neutral conductors that supply it. Any current 'leaking' from a circuit supplied via the RCD will result in an imbalance and if this imbalance is greater than the (quite sensitive) rating of the RCD it will trip.
So, in effect, the current flowing 'in' to the top of the RCD through the brown wire must equal the current flowing 'out' of the blue wire.
One of the most common faults masked by non-RCD consumer units is a neutral-to-earth fault. This would be found by properly testing insulation resistance.
Then there could be a borrowed neutral, which any qualified electrician should always be on the lookout for and problems would become apparent as soon as things got switched on, which is/should be part of the testing process.
Then there is faulty connected equipment, but you would have been able to narrow this down quite quickly. (And, if there was equipment which was not able to be disconnected, this would also have been discovered during IR testing!)
Crossed neutrals, the problem most respondents have suggested, should also have been evident as soon as the installation was re-energised (during the post-operative cup of tea stage). But this problem is simple to identify and correct.
And a faulty RCD should have also been discovered during testing, although it is just possible that the fault developed after completion.
We tested each circuit on its own, with the other MCBs switched off and the one MCB on still tripped the RCD
.
Is this only on one
particular circuit, or does it happen with
any circuit?