Electric shower trips RCD - Help!

I have recently fitted a new electric shower, rated at 10.5 kW, using 10mm2 twin and earth cable and a 45A MCB.

My consumer unit has a 100A main switch, supplying split load sections - one supplying the lights and cooker, and the other the sockets and now the shower, via a 80A RCD (30mA trip current).

I believe this set-up to be reasonable, however the RCD trips as soon as the shower is turned on. I cannot see any fault with the new shower unit, in that I have measured the resistance between live and earth when the water is flowing (with the power off, obviously!) and it is at least 500 MegOhms.

Can anyone with more experience please advise why the RCD may be tripping?

Cheers.

You would need to conduct tests with an insualtion tester, not a multimeter. Test from L to E and N to E. You may not be able to test every part of the shower due to flow switches etc. The most likely part to be 'leaky' electrical wise is the heater.

To be honest, i highly doubt it is the shower.

Try disconnecting all the neutrals on the RCD side of the CU except the shower one, then try the shower. I suspect that it will work. VERY often, a small N to E fault in an installation will sit happily with out tripping the RCD, and then when you put a heavy load on the RCD, more current can 'leak' through this small fault and cause the RCD to trip. I suspect a N to E fault on the sockets somewhere is to blame.

Thanks Lectrician, your suggestions are much appreciated.

I have since checked out the shower with a proper insulation resistance tester and L-E and N-E both appear to show 'infinity' resistance, with and without water flowing through the shower (it does not need power to actuate a valve to start the flow, and therefore all the flow switches are activated properly and the shower is in the electrical state it would be when working)

Before I try removing all the RCD protected circuits neutrals from the CU, as you suggest, I have noticed that all the neutrals are connected together along one busbar - that is from both the non-RCD and RCD protected circuits. The same goes for all the earths.

In addition to this the neutral 'feed' to the RCD comes from this same neutral busbar, whilst the live feed for the RCD is taken from the switched side of the main 100A switch as I would expect. On the switched side of the RCD, the live obviously feeds the MCB's for the sockets and shower, whilst the neutral is left unconnected. Is this the normal arrangement?

This may seem a very stupid question, but should I have connected the shower cable neutral directly to the RCD? The reason I haven't is that the neutrals of all the other RCD protected circuits go to the neutral busbar.

Thanks again for your input.

The Neutral 'out' of the RCD must go to a seperate Neutral bar. These often look like one bar, with the middle bar seperating the two halves.

Sounds like your CU needs sorting. Surprised it hasnt tripped prior to the shower being installed though.

Just a quick follow-up.

The CU was seriously in need of a good sort out. Although two seperate neutral bars were indeed fitted, they had been bridged by a copper link underneath the terminal screws, on top of the bars, making it into a single neutral bar for all circuits, and looking like one continuous bar. The Neutral 'out' of the RCD was not connected in any way.

Removing the link and sorting out the different circuits neutral wiring has solved the problem. I should have spotted this earlier, but just assumed that as the circuits protected by the RCD all worked OK, and did indeed go dead when the RCD was tripped, that all the previous wiring was OK.

The only mystery remaining is how the RCD did not trip when its neutral 'out' went nowhere. Hmmm - think it might need testing!

Thanks again for your help, Lectrician, and for your advice and suggestions.

Cheers.

Maybe a dodgy RCD, I would get a spark to have a look, and possibly do a full PIR report.

Good Luck
__________________________
moderator

post locked to prevent hi jacking