Dodgy lightbulb or wonky electrics?

[MrCheerful]

was it fitted in a different orientation?

[Malcolm Hackett]

Came back downstairs the other night to find that my hall light - which had been switched off - was flashing at slow intervals. Quite spooky.

Took the bulb out and replaced it with another and found that it behaved as you would expect, ie switch it off and it stays off.

Put the bulb from the hall into a standard lamp, tried switchin on for a while and worked fine. Tried switching off and it stayed off.

The questionable bulb is one of the low energy jobbies, a Philips Ecotone

18W, looks like two mini tubes bent into elongated D shapes. Had it a while, so I did wonder whether the strange behaviour was a sign that it was about to give up the ghost, assuming that these fittings can somehow hold a small charge after they are switched off (I'm clearly not an electrician so don't know if this is possible)? But then how come it behaved properly in the alternative fitting?

House is brand new so the electrics *should* be ok and like I mentioned, no problems with the swapped bulb.

Any ideas anyone?

[Lurch]

On Mon, 5 Apr 2004 19:31:37 +0100, in uk.d-i-y "Malcolm Hackett" strung together this:

Knackered bulb. As you summised they can hold a small charge after they are switched off.

[Andy Hall]

Throw the bulb away. These things are the spawn of the Devil.

You might want to check that there is continuity on the neutral and earth at the lamp fitting. You will need a long piece of wire and a test meter. Turn off the mains at the consumer unit and check that the neutral and earth do run all the way back to it by connecting one side of the meter (on resistance range) to earth at the CU and the meter to the neutral at the lamp fitting. Same with the earth.

Other than that, you don't have any children named Damian, do you?

.andy

To email, substitute .nospam with .gl

[Lurch]

On Mon, 05 Apr 2004 20:56:44 +0100, in uk.d-i-y Andy Hall strung together this:

Which type of lampholders have you seen that need an earth to work, and why would there be an earth at the lampholder, and why is it relevant? Not having a go, just curious.

[Martin Angove]

Other suggestions aside, this is often the effect you get if you use a CF lamp in a circuit with the kind of timer switch or PIR which doesn't have its own neutral connection. To work, these things pass a very small amount of current through the circuit at all times. An ordinary incandescent bulb won't notice, but a CF can store up the charge and occasionally flash on.

Quite why you'd have a PIR on the hall light is beyond me, but a timer perhaps? Have you fitted a "security timer" switch? Or maybe one of those "find in the dark" light switches where there is a neon behind the faceplate?

Hwyl!

M.

[Andy Hall]

Considering that this type of lamp is working through a discharge, strange effects can happen.

I wasn't thinking so much of an earth being required at the lamp holder but the possibility of a metal fitting that is not earthed or some other form of pickup if the lighting circuit and switch are miswired - e.g. live present the whole time and a capacitative or inductive pickup to neutral or earth.

Think what happens when you wave a neon screwdriver around a cable....

.andy

To email, substitute .nospam with .gl

[Ian Stirling]

Another CF bulb, or a standard one?

The way CF bulbs are designed is that they charge up a capacitor, to provide a resovoir of energy to draw from when the mains voltage drops to zero 100 times a second. It takes a very, very small amount of current to charge up this capacitor slowly. This can come from various sources, some of which are completely safe. Depending on the current, and the design, it'll flash occasionally.

As a very, very rough guide, the bulb probably takes around a twentieth of an amp. The bulb will probably flash once a second if the current leakage to it is around a twenty thousandth of an amp.

The bulb is certainly OK, it's just that it doesn't work quite like a conventional on in teh case of usually benign faults.

[Lurch]

On Tue, 06 Apr 2004 00:23:47 +0100, in uk.d-i-y Andy Hall strung together this:

Ah, gotcha. Mind now on same track! Although if he tried it on another lamp..... but as you say, strange things happen with discharge lighting.

[Malcolm Hackett]

The bulb in question is a bayonet fitting. Don't see that offers any other orientation?

[Malcolm Hackett]

The light switch is a standard on/off affair, no neon, no security timer. I wouldn't know what a PIR was if it bit me!

[Malcolm Hackett]

'Twas a standard olde fashioned one. Though I have used energy efficient bulbs (CF)? for years and have never seen this happen before...

OK, I suspect I'll regret asking this (non electrician who needs a very, very small amount of techie stuff to make my mind boggle)! but what are some of these sources ... and which are the ones that aren't completely safe?!

It was actually flashing once every few seconds.

[N. Thornton]

And the way to cure it is to fit a suitably rated resistor across the bulb 's N and switched L. Suitably, mind.

Regards, NT

[Ian Stirling]

From memory, a 240V 12W CF bulb that I tested started at about 90VDC, and was completely happy running at that voltage. (I was playing with it on a 150V variable PSU)..

Long runs of cable going to a switch come to mind. If the cable is swiched+live+earth, then when the switch is opened, the AC voltage in the live will induce an AC voltage in the switched cable of around 120V. (if it's got no earth, 240V) The current is very, very limited because of the low capacitance of the cable. Capacitance is something that happens when you get two bits of conductor close together. A change in voltage on one tends to cause a change in voltage on the other. The amount of current that flows depends on the circuit, and how big the capacitance is. (speaking of AC)

The bigger the area between the plates (or wires), the closer together they are, or the higher dielectric constant (all materials have this, it's like mass or conductivity, but varies from 1 for vacuum and air, to several thousand for specially designed stuff, and is under 5 for most common stuff) the higher the current will flow.

You need a meter to measure this though, you wouldn't feel anything from the socket if you touched it across live and neutral when off (don't).

Unsafe conditions would be if there was arcing going on across the switch, or a short in the cabling, and it was that that was causing this. (vanishingly unlikely IMO)

[Malcolm Hackett]

OK, OK - I don't think I understood much of that, except the last sentence! But thanks for trying.

It seems to me that it was most likely to be the bulb causing the phenomenon, that it could have been caused by capacitance but that even if it was I most likely don't need to worry about it.

Thanks to all who answered and confirmed my preference for woodwork and gardening!

[Ian Stirling]

Vanishingly unlikely to be a fault in the bulb as such - there is no way for it to store the energy to flash occasionally. It's just reacting differently to the very low levels of power fed to it by the 'off' switch than a normal one would.

Not even a fault as such, more an incompatibility between the existing wiring standards and the way the bulb is made.

It's likely that changing the circuitry slightly, or adding a 1p component could make it not do this, but if nobody realises that it could happen then it can't be designed against.