"Calvin" wrote in message
oups.com...
I too think you're seeing the effect of capacitive coupling (aka
crosstalk or pickup) between the individual wires in the cable. It's
interesting that you observe it only in one of the two "off" states - I
think I can see how that can happen - not that I can offer a fix
though.
The cable is probably 3-core and earth which is usually arranged as
colourA-earth-colourB-colourC across the cable. I don't think that the
coupling will be between colourA and one of the other colours as the
earth will supress it. That means that colourB and colourC are
coupling. My guess then is that when you don't see the effect colourA
is connected to the lamp with colourB and colourC as live feed and
connected to the switch. In the other off state colourB and colourC
are live feed and connected to the lamp, hence the coupling.

Other posters have suggested that more expensive CF bulbs don't do this
- I wonder if they have bleed resistors?
[In fact maybe there's a DIY solution there - a bleed resistor across
the terminals of the fitting].


I measured the voltage using an analogue meter from the light side to earth.
The voltages were 40v and 75v in the two "off" positions (light removed). I
think the cable
arrangement is
123e. 2 and 3 connect live in one state ( capacitance is 2 - 1) and 1 and
3 in the other (capacitance 1 - 2 and 2 - 3).