Odd lighting question!?

[sparklymarkly]

Here's a slightly odd one. I have just installed some outside lights controlled by a photocell. Having tested the system prior to 2nd fixing (ongoing building project), using a temporary supply, the lights came on OK but every time the photocell turned the lights off, the RCD in the CU tripped out. This even happened when I reconnected the temp supply from the existing house lighting circuit which is not even protected by the RCD in the CU!
The lights are 9W PL fittings, and I suspect this may be the root cause, as when I substituted them with ordinary lampholders with standard tungsten 60W lamps, all the system worked fine.
Any advice would satisfy my curiosity!

 

[mapj1]

Either the lamps are mis-wired in a way such that you have introduced a live earth impedance or (as it trips when not on the RCD supply) a neutral earth impedance. However, assuming you have eliminated that then you are left with the possibility it is a radio frequency problem, and the inverters built into the PL9 lamps are injecting high frequency hash into the supply and confusing the RCD (unlikely unless it is an electronic RCD type).
If you suspect this you will need to arange some RF filtering between the supply and the lamps. As a quick test, if you bring a MW or LW radio near the lamps when operating can you here any interference that goes on and off with the light themselves? If you think this is the case a suitable filter as a first go would need to not introduce so much L-E leakage through its own caps as to fire the RCD anyway! Probably a bifilar RF toroid and a pair of X rated 100nF caps would be the kind of thing - Shcaffer and others make them as a lump with terminals.

 

[sparklymarkly]

Thanks for your reply Mapj.
I,ve solved the problem - took the fittings back to the shop and exchanged them for ones which take a standard lamp!!
Thanks anyway.