When i swapped out our living room ceiling lamps from low energy to led I found they remained glowing fantly when switched off. I gave up figuring it out after soem hours. If I recall correctly local sparks spent a while fiddling in the ceiling rose having discovered 120v across the bulbs when off…

We have just had a kitchen extension fitted and 2 way LED lights were fitted (two indipendent banks of 4 lights). We had the same issue that the lights would glow dimmly when switched off. (one circuit only) To cure this I fitted a mains neon across the last light in the chain, this glows but is in the celing void so not seen and the LED light switch on and off normally. The two way switching cable (4 core) is acting as a transformer and inducing 80 Volts between adjacent cores of the cable caulsing the LED lights to glow dimmly without the neon.

I think the problem is caused by the capacity coupling between the live feed and the switch wire. Being in a hall I suspect that it connected to a two way switch system so there will be quite a long length of cable where the switch wire and live feed are in the same cable so the capacity will be greater. I have not seen an LED lamp flashing when switched of but I have had problems with them glowing dimly. I have solved the problem by connecting a capacitor in parallel with the lamp. (The capacitor must be rated for connection across the mains. X or Y rated but I can't remember which.) To get a suitably packaged capacitor I have used a contact supressor. (Which is a capacitor with a series resistance.) such as this one. They are terminated with flexable insulated leads so that can be fitted in the ceiling rose or light fitting. This works by forming a capacitave potential divider with the cable capacitance as one part and the added capacitor as the other. This reduces the voltage across the LED when the switch is in the off position so it is too low for the LED to light. A resistor could be used but it would gnerate heat when the light was switched on. There will be no heat with a capacitor as the current is out of phase with the voltage. The resitor in the contact supressor will generate a very small amount of heat but it is not enough to cause a problem. A contact supressor with the lowest value series resistor is the best to choose.

Les.

Glowing dimly tube heater element in older style fluroescents was usually caused by the switch being connected into the neutral and not line conductor. The incorrect polarity at the switch would be discovered when testing the circuit.

Emgee

The problem I have with LED lights, especially the rectangular floodlights, is radio interference caused by them. It comes {I guess} from the oscillator gubbins which drops the 230v to LED voltage, and I wonder if the floodlights, being designed for industry rather than home use, are not designed with shed radios in mind. I will eventually find a solution, but if anyone knows of a quick or easy answer, do let on.

PS I also get buzzes on the radio from my Jaguar inverter, but not so loud.

Cheers, Tim

We had similar issues with the older fluorescent types (with a spiral or folded tube). Eventually we found that some makes flashed briefly every 20 or 30 seconds when off, but a different make appeared to be OK. What was confusing was that it was not consistent. A bulb would flash in one socket, but in a different room, it worked OK. There was no pattern to this – bulbs which worked in one place would flash in another. We eventually got everything to work by 'selecting' bulbs in different rooms. The house is only about 4 years old, and I am sure all the wiring is OK.

Just recently we changed over to LED bulbs instead to reduce our consumption even further, and they all worked OK, regardless of where they were fitted. We did notice that they appear to have a 'soft' switch-off. They grow progressively dimmer, quickly going almost off, but a slight glow is visible in a darkened room for maybe 10 or 15 minutes – this suggests a large internal capacitance to me.

Out of curiosity, I measured the voltage across each socket when they were switched off, because despite the new wiring, I suspected maybe a high resistance circuit when the switch was off, but the switches do appear to be 'open-circuit'. The first meter I used was a high-impedance electronic type, set of course to AC volts. I noticed when the switch was off it appeared to indicate a random low voltage, maybe 2 or 3 volts, which was constantly changing. I put this down to 'noise pickup' aggravated by the high impedance meter – I later tried an old low impedance moving coil type, and got no readings. Don't really know what this proves, especially since we had trouble with the fluorescent type, rather than LEDs.

(Edit) Incidentally, changing connections L/N at the socket had no effect on the flashing bulbs

Edited By An Other on 08/01/2017 18:41:59

I find that some LED lamps come brighter after a couple of minutes. I also have two more which glow for a few seconds when switching off presumably a capacitor discharging somewhere

Roy

Edited By roy entwistle on 08/01/2017 20:08:32

The "electronic transformer" supplying the 5x 20w 12v tungsten spots in my utility/workroom died after many years of service. Taking this as an opportunity to change to LED I purchased 5x 5w LED spots of the LAP brand and a LED driver (30w 12v constant voltage). After wiring up I switched on and all the spots flashed in unison. Carried out fault finding by process of elimination and found that connecting any 3 spots in any position worked. With the 3 spots switched on I connected the other 2 and all 5 worked OK. Switched off and on and flashing started again. Removed any 2 and all OK. Thinking that the total load of 25W was maybe too close to the 30w limit I bought and fitted a 50w driver. Exactly the same happened!

Any advice would be greatly appreciated.

Gordon.