LED downlight remaining on

Joined
23 Feb 2015
Messages
48
Reaction score
0
Location
Hampshire
Country
United Kingdom
Sponsored Links
Even with properly earthed system the capacitively coupled power from Live to Switched Live in the cable to the switch is enough to "glow" an LED lamp when the switch is OFF.

A capacitor in series with a resistor across the supply to the LED will in almost all cases solve the problem. The capacitor absorbs the capacitively coupled power without the voltage being high enough to affect the LED.

0.1 microfarad 400 volt capacitor polyester and 100 ohm 1 watt resistor have proved sucessful. I would also fit a 1 amp fuse in series as well

400 volt as it has to cope with the approx 320 volt peaks of 230 v AC It cannot be an electrolytic or polarised capacitor.

Contact suppressors such as the ones below provide the necessary circuit in a single convenient package. The use across LED lamps is not their intended purpose but they do provide the function required.

http://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=0209241
[URL='http://click.linksynergy.com/fs-bin/click?id=HSFXGpFPZRE&subid=&offerid=403983.1&type=10&tmpid=18154&u1=XF&RD_PARM1=http%3A%2F%2Fwww.maplin.co.uk%2Fp%2Fcontact-suppressor-rg22y%3F']http://www.maplin.co.uk/p/contact-suppressor-rg22y[/URL]
This has a 0.1uF capacitor which is a bit of an over kill but better too large than too small.

There are other suppliers.
 
Thank you for your reply, I am unsure of " capacitively coupled power" but can understand the principle of "induction" in the cables causing this to happen. Is this a close enough assumption?
 
Sponsored Links
Yes, I wondered about that, too. The Maplin one does, indeed have a 0.1μF capacitor. However, the RS one has a 0.22μF one, so I suspect that bernard perpetrated a typo, and was intending to say that the 0.22μF one was 'a bit of an overkill'.

Kind Regards, John
 
The problem is many bulb manufacturers include resistors and/capacitors to remove this problem, so adding exterior ones means every time a bulb is swapped one should check if still required.
 
Yes, I wondered about that, too.
For the average domestic length of switch cable 0.1μF+100Ω is adequate to absorb the power capacitively coupled from Live to Switched Live.
The 0.22μF+100Ω ( RS device ) was used on a very long switch cable so I quoted that one and then remembered the 0.1μF+100Ω Maplin device was a cheaper and easer to obtain option
 
For the average domestic length of switch cable 0.1μF+100Ω is adequate to absorb the power capacitively coupled from Live to Switched Live. The 0.22μF+100Ω ( RS device ) was used on a very long switch cable so I quoted that one and then remembered the 0.1μF+100Ω Maplin device was a cheaper and easer to obtain option
Fair enough, and I agree with all you say. However, I still think there must have been a typo (which BAS picked up on), since you wrote:

Kind Regards, John
 
The problem is many bulb manufacturers include resistors and/capacitors to remove this problem, so adding exterior ones means every time a bulb is swapped one should check if still required.
This reply solved the problem. By simply changing to a different brand bulb resolved the issue.
Thank you.
 
Wouldn't a double pole switch get rid of this?
Break the neutral and no current flow.
But that involves re-wiring probably.
 
I changed a load (18) 50W halogens for LEDs. Six were in a sitting room- no problem. The other 12 were six each upstairs and down in the hall and landing. These ones had the glow issue. In the end, advice was (from here, I recall) to change one lamp up and down for a low wattage halogen. Problem solved, though I don't know why. Two way switching is the only reason I can see for the landing lights glowing when the sitting room ones were fine.
 
Wouldn't a double pole switch get rid of this? Break the neutral and no current flow.
It presumably would but, as you go on to say, it would probably require a fundamental change to the wiring.

If one tried to 'improvise' the wiring changes (i.e. adapt the current wiring), I don't think it would be guaranteed to solve the problem since, just as there can be capacitive coupling between permanent and switched live conductors, so (with some wiring arrangements) could their be capacitive coupling from neutral conductors to CPCs - and the combined impedance of those two capacitive paths might possibly be low enough to allow at least some flashing of the LEDs.

Kind Regards, John
 
Wouldn't a double pole switch get rid of this?
Break the neutral and no current flow.
But that involves re-wiring probably.
You are on the right lines, it is the fact that the switch does not have the neutral taken to it with the line. The problem is complex, it's called transmission lines and any cable is considered as a series of capacitances and inductors and if the line and neutral are the correct distance apart then they cancel each other out. However at 50 Hz we really do not worry, once you get into the MHz range then very important.

Radio people measure the voltage standing wave ratio and go to great lengths to ensure the circuit is balanced, but with lights it really does not matter if the light stays on dim, adding an inductance or capacitance to correct may be the theoretical cure, but in real terms we simple leak away the excess bypassing the lamp, be it CFL or LED. Only real cure would be DC.

The war of the currents was interesting, USA had some DC supplies right into end of 20th Century and the pros and cons DC v AC are now leaning more to DC with modern technology. But the ability to transform to high voltage for long runs then back to low voltage for local use means even if there are some losses the net result is AC wins.

If you can light fluorescent tubes under power lines then clearly some power is being radiated, however when compared with power lost with conversion equipment to use DC it is still best way to send power around the country and your house.

It does not have to be bulbs lighting dimly, silly LED and Neon lights on routers, TV's printers, etc can illuminate a bedroom, I spent ages sticking bits of black tape over them all the other night.
 
You are on the right lines, it is the fact that the switch does not have the neutral taken to it with the line.
I don't really think that is the issue that ColJack was thinking about. If all that goes to the light is an L/N pair, and if that pair is DP switched, then it's all but impossible that the LED would even glimmer - since the only paths through which current could flow would be the tiny capacitances across the switch contacts. However, as he said, to achieve that would usually require a total change of the circuit wiring.
The problem is complex, it's called transmission lines and any cable is considered as a series of capacitances and inductors and if the line and neutral are the correct distance apart then they cancel each other out. However at 50 Hz we really do not worry, once you get into the MHz range then very important.
We've been through this many times before. Transmission line theory is totally irrelevant at 50Hz, unless one is considering cable runs of at least hundreds of kilometres.

Kind Regards, John
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top