help! my light bulb lights up without electricity.

[bardos]

I don't know if this is going to save the planet or not...

but seriously it's a problem

i put a new bulb in to replace an old worn out one. these are the new fluorescent types. The bulb is in and the switch is OFF. and yet the bulb continues to flicker... i unscrew it and it stops, screw it in and it starts flickering again. as if to start. turn it on and it's normal. turn it off and it won't go off completely. it continues to flicker. what in the name of all that is holy is going on?

 

[JohnD]

A tiny induced current between the contiguous switched and unswitched cables where they go down to the light switch, probably. It happens a lot. The electronic CFL is able to store this tiny current until it builds up to enough to make a flicker.

Bit annoying but not a worry.

If there is an earth wire in the same cable it should happen less.

Some brands of CFL don't seem to do it.

If you can put a slight load on the circuit (like another bulb, or a neon on the switch) it will drain away this tiny current and stop the flickering.

 

[pcboffinuk]

Is it on a two way circuit, or likely to be near the wires in one?

Normally this is caused by induced voltage from the lighting cables. It generally only appears with low energy fluorescent fittings because they have a very low internal resistance, and therefore are affected by this induced voltage.

It may be a case of trying other makes of bulb, or reverting to a normal bulb in this fitting.

Sadly its not going to save the planet though

Edit: BAH! Must type faster next time

 

[plugwash]

try adding a small mains rated capacitor (something like maplin code SC91Y) accross the light

 

[Mosez]

try adding a small mains rated capacitor (something like maplin code SC91Y) accross the light

Put a 220K/0.5w resistor across the lamp wires, maybe at the switch/Rose across the switched wire to the lamp and the neutral.

 

[bardos]

thank you for your suggestions, which seem to me a little complicated just to have a fluorescent light bulb. so i went to the city today and bought a couple of new ones, different brand. came home enthused and replaced the flickering one with a new one.

result: the same exact flickering. hmm...

took the bulb to a desk lamp and screwed it in... no flicker! works normally.

at the original light i switched the two wires around, result: same flicker. hmmm...

 

[Softus]

thank you for your suggestions, which seem to me a little complicated just to have a fluorescent light bulb.

so i went to the city today and bought a couple of new ones, different brand. came home enthused and replaced the flickering one with a new one.

result: the same exact flickering. hmm...

took the bulb to a desk lamp and screwed it in... no flicker! works normally.

at the original light i switched the two wires around, result: same flicker. hmmm...

Ahh - I see now why your 'solution' is so much simpler.

 

[bardos]

ok, so i guess the question is: why does the bulb work normally at one lamp and weirdly at another?

thx.

 

[Softus]

ok, so i guess the question is: why does the bulb work normally at one lamp and weirdly at another?

//www.diynot.com/forums/viewtopic.php?p=541004&highlight=#541004

//www.diynot.com/forums/viewtopic.php?p=541016&highlight=#541016

//www.diynot.com/forums/viewtopic.php?p=541123&highlight=#541123

//www.diynot.com/forums/viewtopic.php?p=541324&highlight=#541324

 

[sm1thson]

why does the bulb work normally at one lamp and weirdly at another?

Normally this is caused by induced voltage from the lighting cables.

so basically in the lamp the switched live wire is running alongside a live wire (common in two way switching) and this acts as a capacitor (a capacitor is an electrical store like a mini rechargeable battery, they are made by having two conductors separated by an insulator hence why your wires can act like this). in the desk lamp the switched live doesnt run alongside the live.

the resistor across the light sounds like the best plan to me. (after checking its all earthed right because that can sometimes help reduce the capacitance also)

 

[JohnD]

I would have said more like a transformer than a capacitor.

I wouldn't have said you'd charge a capacitor with AC.

 

[sm1thson]

your right john, i realised after i posted, but you were two quick as always!

its induced voltage rather than capacitive: so my explanation should have read something like:

basically in the problematic light the lamp the switched live wire is running alongside a live wire (common in two way switching) the live wire produces a magnetic field around it, as the switched (off) wire is in close proximity to it and converts the magnetic field back into electricity. This is similar to the electrical to magnetic to electrical process that’s in a transformer. in the desk lamp the switched live doesnt run alongside the live so the fields don’t interact.

 

[bernardgreen]

your right john, i realised after i posted, but you were two quick as always!

its induced voltage rather than capacitive: so my explanation should have read something like:

basically in the problematic light the lamp the switched live wire is running alongside a live wire (common in two way switching) the live wire produces a magnetic field around it, as the switched (off) wire is in close proximity to it and converts the magnetic field back into electricity. This is similar to the electrical to magnetic to electrical process that’s in a transformer. in the desk lamp the switched live doesnt run alongside the live so the fields don’t interact.

You are both right as the actual circuitry between the live and switched live is a complex impedance comprising both transformer and capacitive elements.

The resistor across the lamp is effective but will get warm (220,000 ohms at 250 volt is 0.28 watts ) in a confined space. Better is a capacitor and resistor wired in series but for the life of me I cannot find the value needed.

 

[Adam_151]

An old post from over at the IET:

The problem is one of the capacitive 'divider' caused by the swich wiring, more length adding capacity from live to fitting, in shunt with switch, and the capaciitance of the wiring from juntion box or rose to fitting, (capacity in shunt with lamp) At ~100pf/metre, for a hall and landing switch , 10m or more is possible, so the capocitance shunting the switch could be >1nF. To guarantee not to trickle charge the electronics, the lower divider capacitor (shunting the fitting) needs to be increased, to be larger - i'd suggest 10 times larger as a start, so the AC present on th lamp holder is < 24V try running a lamp off a variac - its surprising how low they go and still do something can often be less than 50V).
not a problem on loop through switch circuits of course.
Actually, the induced voltage in 'off state' can be measured with a high impedance meter if you take out a normal bulb too, but is normally of no consequence, as the filament bleeds it away, and at a few hundred micro Amps, nothing glows.

So adding 10 to 100nF accross the CF lamp reduces the induced voltage to sweet nothing on all but the largest installation, and a cap is much preferred to a resistor, as the whole point of an energy saving light fitting is to make the heating work harder by a few 10s of watts to heat the house to the same temperature, not to have a resistor heating up when the lamp is on.. If that saves enough fuel to justify the poisenous rare earths and environmentally horrible electronics used to make such a bulb or not depends on the method your heating system - if it is electric heating anyway, then clearly it doesn't, but building regs don't acknowledge this subtelty yet.

[/quote]

 

[bardos]

is 2 way switching a term that refers to a light that has two switches? my light fixture (that flickers) has only one switch. Two wires from mains are heading for the fixture, one of them is "intercepted" by a simple wall switch. so one of the wires is always "dead" when the switch is in off position. plz enlighten me.

thx.