Timeclock Tripping RCD

[earthboy]

I have a 6A C Type RCBO feeding a Horstmann 20A Timeclock and the switched-line operating AC1 on a 4P 24A contactor (NO).

There are three lighting circuits going into the contactor, all fed from individual 6A B Type MCBs (on the RCD side of the DB).

Why should the RCD sometimes trip when the timeclock switches off? I say sometimes because when I try to reenact the tripping by either rotating the dial or operating the override switch, the RCD (most of the time) does not trip. I've yet to experience the RCD tripping when the lighting circuits are switched on.

The lighting circuits have low loads (2 x 58W fluorescent on one, 2 x 9W CFL on another and 3 x 20W CFL on the third).

All cabling and accessories are new, and all test results show no issues - the RCD operating times all fall within regulations.

Any suggestions?

 

[Jurassic Spark]

Could it be something to do with the neutral connections? If you have an RCBO presumably fed from the main switch side to control the clock and contactor and your circuits fed from an RCD, there might be a slight imbalance. Can you wire the control circuit from the same RCD side of the consumer unit?

 

[Spudmiester]

You have very low loadings, why do you need a contactor, or need to individually breaker the circuits ?

Try running the lot off the timer feed, switch only the line and see what happens.

Could be the RCBO don't like the contactor, back EMF from coil etc. It shouldnt trip, but might just be one of those weird ones. Might also not like capacitors in fittings discharging.

 

[montybaber]

My moneys on the coil

 

[Jurassic Spark]

A contactor is sometimes required for automated lighting systems. I do showrooms for a certain company that leaves selected parts of the showroom lighting on controlled by t/switch & contactor. At a selected hour when the staff have gone home, the t/switch knocks the lights off.

Also, I fit contactor controlled lighting in their offices where one light-switch turns the whole of the lighting off.

 

[earthboy]

Thanks for all your replies.

I think I will try swapping the timeclock feed RCBO and put the circuit back on a MCB on the RCD-side of the board and see whether this resolves the tripping.