LED bulb flashes when off - but only when one of the 2-way switches is used

[^woody^]

I have two light fittings with LED bulbs, one ("the end one") is just directly connected to the other at the fitting, and then these are looped in at a switch - and there is a second two-way switch to operate them.

The end one flickers when off - but only if the lights were switched on at the two-way switch and switched off at the main switch.

I understand that this it's due to slight residual current in the circuit charging the bulb capacitor, but I wondered why this only occurs when the one switch is operated and why just the end fitting?

But the main questions are - is it a safety issue, and is it a fault - a faulty switch or circuit?

The two-way switch had a neon indicator. This can apparently be configured to come on with different switch operation. At the moment the neon is illuminated only when the lights are switched on at the two-way switch and off at the main switch - ie on when the lights flicker. I don't know if that's a clue or a symptom.

 

[ericmark]

Because with three core and earth the earth goes between two cores and one core so it will both have a larger gap switched one way to other and have an earth wire between the two cores.

It is some thing I did look into, but the transmission lines section of my Uni book is rather complex, we should use a ballanced feeder, and if you look at 300 ohm ribbon cable it does look very like twin and earth. The idea is the line and neutral are kept a set distance appart so the inductive and capactive links cancle each other out. However this does not happen with the UK switch cable, the neutral is not taken to the switch.

Before the LED the advantage of having a perment line at the ceiling rose so ceiling fans, and emergancy lights were easy to fit was a good idea, it also reduced the volt drop and loop impedance, so the problem of being unballanced was not really any worry.

With two way switching the lenght of cable tends to get longer, but resistors are added to some LED lamps and not others, so adding a resistor is not really an answer, what we should do is correct the imballance, but to do that you need to measure how far out it is.

With higher frequances i.e. ham radio and CB we measure the voltage standing wave ratio (VSWR) and correct, but with house wiring we just select a lamp which will go out.

 

[bernardgreen]

The "slight residual current" happens because the two wires in the cable to the switches form a capacitor which will pass a very small stray current from Live to Switched Live when the switch is OFF In effect it is a capacitor across the switch contacts.

The amount of stray current that flows will depend on how much capacity there is in the cabling between Live wires and Switched Live wires . With two way switching the wires that are Live and Switched Live can change depending on the setting of the switches and which of the strappers are connected to Live and which are connected to the Switched Live to the lamp.

 

[Spark123]

Can you upload some photo's of what you have including the wiring?

 

[endecotp]

The two-way switch had a neon indicator. This can apparently be configured to come on with different switch operation. At the moment the neon is illuminated only when the lights are switched on at the two-way switch and off at the main switch - ie on when the lights flicker. I don't know if that's a clue or a symptom.

Yes, it's a clue and I think more likely the cause of the problem than conductor-to-conductor capacitive coupling in the wires.
A photo of the wiring at that switch would be interesting.

 

[Spark123]

I don't know of any off the shelf two way switches with a neon. I know you can get grid systems or an additional back plate with neon.

 

[^woody^]

I don't know of any off the shelf two way switches with a neon. I know you can get grid systems or an additional back plate with neon.

It's an external Masterplug/BG Nexus.

It's wired up with 3 core and earth to the main switch, and in the standard way - ie the normal colours to the normal points.

 

[Spark123]

Can you provide photos? I always wondered if arranging the conductors so the earth wire in the 3c+e is between the sw live (to the lamp) and other two conductors will make any odds.

 

[Taylortwocities]

Its not guaranteed, but putting one of these across the offending lamp can cure this sort of problem
http://www.maplin.co.uk/p/rc-contact-suppressor-rg22y#.V9772OARi1A.link

 

[Jackrae]

I used one of Maplin's suppressors a couple of years back with complete success, so well worth a try

 

[Echo the husky]

We use the type with insulated flyleads as they are a bit easier to install.

 

[endecotp]

Everyone, this is NOT due to capacitive coupling, it's due to there being a neon in series with the LED when it's off!

Is it this one ?:
http://www.bgelectrical.uk/public/d...f/WP12_WP42_WP14_WP30_20amp_20AX_switches.pdf

Just to confirm, try disconnecting the neon and see if the issue goes away.
If you have neutral at this switch it can be fixed.

 

[^woody^]

Everyone, this is NOT due to capacitive coupling, it's due to there being a neon in series with the LED when it's off!

Is it this one ?:
http://www.bgelectrical.uk/public/d...f/WP12_WP42_WP14_WP30_20amp_20AX_switches.pdf

Just to confirm, try disconnecting the neon and see if the issue goes away.
If you have neutral at this switch it can be fixed.

Yes that's the one. I'll try and have a look tonight.