Strange electrical problem... Low energy lamp..

[minghis]

Anyone cast any light on this?

I've just exchanged my bathroom 60w ordinary bulb for one of those low energy things (courtesy of some freebies from my Gas supplier) but now my extractor fan comes on without the light actually being on - which is strange?

Any ideas would be welcome..... is it safe, or should I just isolate the fan and put up with the steam?

[GTIR]

Er, I know they do not work with dimmer switches....


Have you tried washing in the dark?
Its like someone else is doing it, bit like lying on your right arm until it goes dead then playing with yourself.

[Pigeon]

Try connecting a 100k 1W resistor in parallel with the lamp.

[Dogwatch]

Go back to using a proper bulb. Won't bring the planet to a grinding halt and you 'll have both light and fan.

And, yes I do have 'energy saving' bulbs around the house.

[Deva Link]

minghis said:

Anyone cast any light on this?

I've just exchanged my bathroom 60w ordinary bulb for one of those low energy things (courtesy of some freebies from my Gas supplier) but now my extractor fan comes on without the light actually being on - which is strange?

Any ideas would be welcome..... is it safe, or should I just isolate the fan and put up with the steam?

Is this a known issue with cfl's? I can't imagine why the fan should come on, always assuning the fan isn't wired in series with the light, or something else stupid.

Just guessing - does the fan have an electronic timer?

Edited by Deva Link on Wednesday 24th September 22:50

[Morningside]

I didnt think these CFL's liked being in an enclosure?

I suppose your fan has one-of-those moisture sensors? Perhaps the electronics is over sensitive?

[Pigeon]

Yes, it does. The timer is triggered by an input from the switched live to the lamp.

Switching a lamp off does not render the switched live totally "dead" - there is still a significant voltage on it due to capacitive pickup from other wires, imperfect insulation, and stuff like that. However the voltage source has a very high impedance so any load on the circuit drags it down to zero. The impedance is so high that you can still touch the switched-off wire and not feel anything.

The trigger input on the timer, however, is far more sensitive than it needs to be and is capable of being triggered by this voltage. Normally this doesn't matter because the load of an incandescent bulb on the circuit is way in excess of what is required to drag the leaked voltage down to zero so the timer doesn't get triggered. A CFL, however, will draw no current at all and will look like an open circuit unless sufficient voltage is applied to start the inverter. So the leakage is now able to generate enough voltage to trigger the timer and the fan comes on.

[Deva Link]

Pigeon said:

A CFL, however, will draw no current at all and will look like an open circuit

Hmmm, struggling with that, to be honest. So if you take an incandescent lamp out (or it blows, of course) will the fan start up?

[grumbledoak]

Just put the normal bulb back in and forget about it.

I don't doubt that 'Pigoen' is explaining it perfectly, but this is the easiest solution.

[Pigeon]

Deva Link said:

Pigeon said:

A CFL, however, will draw no current at all and will look like an open circuit

Hmmm, struggling with that, to be honest.

A CFL contains a rectifier which rectifies the incoming AC and charges a reservoir capacitor. The high DC voltage on the reservoir capacitor is then fed to an inverter, which basically comprises a pair of transistors which switch on and off alternately at high speed to produce high frequency AC which is then fed to the lamp via an inductor.

If it is not given enough voltage, the inverter will not start - neither transistor will be switched on and the lamp will draw no current. If the leakage voltage is below the threshold voltage the lamp will draw a little current until the reservoir capacitor is charged to a corresponding level and will then sit there drawing no current at all.

On an installation with severe leakage problems then it sometimes is bad enough that the capacitor will charge up over the course of several minutes until there is enough voltage on it that the inverter will start. This feeds power to the tube, causing a dim flash, and discharges the capacitor. The cycle then repeats. The result is a CFL which is nominally switched off flashing every few minutes.

Deva Link said:

So if you take an incandescent lamp out (or it blows, of course) will the fan start up?

Yes. Of course it varies between installations according to the amount of leakage. But in the OP's case I would expect that if he left the fitting with no bulb in it at all the fan would start up in the same way as it is doing with a CFL in it.

Hmm. It has just occurred to me that possibly the timer is designed to cope with leakage - by having a resistor between the trigger input and neutral to discharge any leakage - but that resistor has failed in this case and become open circuit. Resistors connected directly across the mains often do fail despite an apparently adequate power rating due to the voltage gradient across them causing electromigration effects. It would be worth checking the timer board... And it occurs to me also that the easiest way to connect a 100k 1W resistor in parallel with the lamp is to connect it between the trigger and neutral terminals on the timer, keeping in mind that it will get hot.