Electrical leakage on timer switch & diode use

[HoHoHo]

Here's one for those who understand these things...........

I have three sets of exterior lights at home all on electronic switches and I've just changed the outside fittings which now include GU10 LED bulbs.

Unknown to me my lighting circuit uses a live circuit with no neutral (not uncommon so I understand) and also unknown to me electronic switches that don't use neutral leak voltage constantly at about 120v

The LED bulbs are actually 80-240v so I now have the scenario that one bulb on one of the three circuits flickers about every second or so.

This appears to be a common issue on this type of circuit with this equipment and the only answer is to use a diode on the circuit after the switch to kill the leakage.

What I don't know is what diode should I use?

Any help would be appreciated

Thanks.

[hairyben]

Dunno about diodes. Conventional switches not an option? You can try a shunt resistor which might work, switch the lights through a relay, or use at one filament lamp per switch as a quickfix.

[HoHoHo]

hairyben said:

Dunno about diodes. Conventional switches not an option? You can try a shunt resistor which might work, switch the lights through a relay, or use at one filament lamp per switch as a quickfix.

I can change it to a conventional switch but that then means I have to be about to switch them on which defeats the object of a timer.

I can always use GU10 bulbs and they won't flicker but they are yellow and if I swap the flickering bulb I may have another that does and I quite like the white light I now have.

I know it can be fixed, just not sure how

[motco]

Electronic switches that use triacs will have a few milliamps of leakage. The answer is either to suppress the leak by putting a filament lamp in circuit to shunt the leak or to use switches with contacts (relays, points, whatever) that have a physical break in the circuit. A shunt resistor may be fine but you need to use a suitable value and wattage to be on the safe side.

[hairyben]

HoHoHo said:

I can change it to a conventional switch but that then means I have to be about to switch them on which defeats the object of a timer.

I can always use GU10 bulbs and they won't flicker but they are yellow and if I swap the flickering bulb I may have another that does and I quite like the white light I now have.

I know it can be fixed, just not sure how

Oh, timer switches

You could try the below but Id try just the one first to see if it works.

http://www.timeguard.com/products/light/automatic-...

[HoHoHo]

motco said:

Electronic switches that use triacs will have a few milliamps of leakage. The answer is either to suppress the leak by putting a filament lamp in circuit to shunt the leak or to use switches with contacts (relays, points, whatever) that have a physical break in the circuit. A shunt resistor may be fine but you need to use a suitable value and wattage to be on the safe side.

Not being an electrician I'm assuming my electronic switch (or any other) uses triacs hence the 120v leakage?

What is also interesting is that I mentioned that I have three switches around the house, 2 currently with 2 lights the other with more and it's only one of the two circuit that has a problem.

I'm going to move switches around to have a play and see if they have an influence on the issue (I have two different types of switches on both two light circuits)

[motco]

HoHoHo said:

motco said:

Electronic switches that use triacs will have a few milliamps of leakage. The answer is either to suppress the leak by putting a filament lamp in circuit to shunt the leak or to use switches with contacts (relays, points, whatever) that have a physical break in the circuit. A shunt resistor may be fine but you need to use a suitable value and wattage to be on the safe side.

Not being an electrician I'm assuming my electronic switch (or any other) uses triacs hence the 120v leakage?

What is also interesting is that I mentioned that I have three switches around the house, 2 currently with 2 lights the other with more and it's only one of the two circuit that has a problem.

I'm going to move switches around to have a play and see if they have an influence on the issue (I have two different types of switches on both two light circuits)

It's not really a 120v leak but it will measure out at something like that dependent on the impedence of the meter. It's actually got a 'face voltage' of as much as 240v but the act of measuring draws enough current to drop the apparent voltage to 120v. Use a different meter and you will probably get a different reading. Don't try it, but you could possibly stick your fingers across it and just feel a slight tingle. It is a very high impedence source but sufficient to make a lamp flicker if it's an LED or even neon. Of course I may be talking boswellox and it is half-wave rectified and is really 120v and will kill you!

My company used to sell a flood lighting control that used a triac and installers would often ring and complain that the lights were constantly on. In reality they had yet to connect the (500w halogen) lamps and were simply using a meter to check but extrapolated their readings into "the lights are on". Had they connected the lights then there would no measurable voltage and the filaments would be dumping the superficial current to neutral.

[HoHoHo]

motco said:

It's not really a 120v leak but it will measure out at something like that dependent on the impedence of the meter. It's actually got a 'face voltage' of as much as 240v but the act of measuring draws enough current to drop the apparent voltage to 120v. Use a different meter and you will probably get a different reading. Don't try it, but you could possibly stick your fingers across it and just feel a slight tingle. It is a very high impedence source but sufficient to make a lamp flicker if it's an LED or even neon. Of course I may be talking boswellox and it is half-wave rectified and is really 120v and will kill you!

My company used to sell a flood lighting control that used a triac and installers would often ring and complain that the lights were constantly on. In reality they had yet to connect the (500w halogen) lamps and were simply using a meter to check but extrapolated their readings into "the lights are on". Had they connected the lights then there would no measurable voltage and the filaments would be dumping the superficial current to neutral.

My electrician who has just done some work on the garage measured it at 120v. I'm hoping he knows what he is doing but that said he was unaware these electronic switches leak power and told me it must be buggered.

By chance earlier on Saturday I didn't get a tingle when I changed one light for a new one thinking that the switch was off..............

I got a fk ME WHAT WAS THAT! as I touched a couple of wires at one stage.

I did think it was rather odd at the time

[hairyben]

HoHoHo said:

My electrician who has just done some work on the garage measured it at 120v. I'm hoping he knows what he is doing but that said he was unaware these electronic switches leak power and told me it must be buggered.

By chance earlier on Saturday I didn't get a tingle when I changed one light for a new one thinking that the switch was off..............

I got a fk ME WHAT WAS THAT! as I touched a couple of wires at one stage.

I did think it was rather odd at the time

An electrician should know how these types of switches operate as it's a problem you increasingly come across.

Basically the switches electronics need powering but with no neutral charges itself via the permanent live and the switched live, which will have 230v available when "off", putting a very small current through the "bulb". In filament lamps not an issue, they dissipate the current as a tiny amount of heat, not enough to show any light, in LED's anything can happen, I think with your flashers the voltage builds up untill the lamp tries to fire, discharges the voltage, and the cycle repeats. Higher powered lamps might be less trouble.

[motco]

hairyben said:

An electrician should know how these types of switches operate as it's a problem you increasingly come across.

Basically the switches electronics need powering but with no neutral charges itself via the permanent live and the switched live, which will have 230v available when "off", putting a very small current through the "bulb". In filament lamps not an issue, they dissipate the current as a tiny amount of heat, not enough to show any light, in LED's anything can happen, I think with your flashers the voltage builds up untill the lamp tries to fire, discharges the voltage, and the cycle repeats. Higher powered lamps might be less trouble.

I think that's what I was trying to say - less concisely in my case though! Never use one word when ten will do.

[HoHoHo]

hairyben said:

An electrician should know how these types of switches operate as it's a problem you increasingly come across.

Basically the switches electronics need powering but with no neutral charges itself via the permanent live and the switched live, which will have 230v available when "off", putting a very small current through the "bulb". In filament lamps not an issue, they dissipate the current as a tiny amount of heat, not enough to show any light, in LED's anything can happen, I think with your flashers the voltage builds up untill the lamp tries to fire, discharges the voltage, and the cycle repeats. Higher powered lamps might be less trouble.

Yes, that's as I understand it.

I have also been told by my local electrical shop it's a problem with LED bulbs and it can and can't happen, it's hit and miss They also suggest all LED bulbs are now multi-voltage and they can't supply a higher volt bulb.....is that not the case

@motco - thanks for your help and you had explained yourself :thumbs up:

However, short (see what I did there ) of changing back to GU10 normal bulbs is there anything I can do?

[VEX]

If you google dummy electrical load, this will effectively draw all the leaky power and disapate it as heat, in the early days electricians use to hide a standard incadecant bulb in a ceiling void to draw the power.

Effectively the same as a resister, but I am unsure what the power rating would need to be.

V.