LED lights not turning off.

[thehighlander]

Hi folks,

I was wondering if anyone has come across the problem of LED lights not completely turning off when switched via a 2 way/Intermediate circuit.
They work fine and switch off completely when wired as a one way lighting circuit, but when in a 2 way/Intermediate circuit they dim significantly but do not turn off.
There is about 10 volts supply going to them when they are meant to be switched off which is obviously enough for them to light dimly.
I would think this 10 volts is being induced from the other strapper in the circuit and would not be enough to light a normal incandesant or halogen bulb.
I did not wire the lights but have checked the wiring and everything seems fine, I am not sure what wattage the LED's are but the lights are actually meant for decking.(240 volts)
Does anyone know of a solution that does not involve changing the lights for halogens etc?

 

[Marvo]

I would think this 10 volts is being induced from the other strapper in the circuit

If there's several lamps in the circuit and you're getting 10volts then this is a fault current. Usually if there's any type of load on the circuit and the voltage remains then it's unlikely to be an induced voltage from another cable or wire.

If you isolate the supply to the circuit at the point of entry (rose or switch) does the voltage disappear?

 

[topquark]

If it's two way controlled and it's off, what current do you think could be flowing in one of the strappers to induce a current in the other strapper (there should be no load to draw a current)? Unless you had more than one cable together, assuming that the circuit is wired in three core+E.

More likely to be a wiring fault (accessory in series or two circuits interconnected). Time to get the test gear out

 

[thehighlander]

Hi Marvo,

There is several lamps on the circuit, I was getting the voltage at the switch.
When connected as a one way circuit they function correctly as the cables linking the switches had been disconnected hence no induction to the switch wire. Their is actually 2 lots of these in the house that are 2 way/intermediate and both have the same problem. They are also fed through there breaker via an RCD.

 

[Marvo]

I also don't see you getting sufficient induced voltage between the live wire and the strapper to illuminate or even slightly illuminate several lamps. I would start by isolating the entire lighting circuit and disconnecting the wiring at both switches and doing a mega test between them all, especially the live and the strapper. If you get anything less than a few Megs then go hunting for the fault.

 

[thehighlander]

Thanks for your time marvo, I will need to look into it in more detail. The spark who wired it would have done a full test on the wiring before he could sign it off as it was a complete rewire. His boss has also been to the house to investigate it and could not come up with any solutions.
It was the boss who left them working as a 1 way switch. It is actually my brother in laws house and he mentioned it the other day when i was round and all i had was a multimeter, I'll need to dig out my test gear. I did find people online who have had the same problem when they have changed incandesant bulbs out for low energy bulbs on 2 way switched circuits. A few of the guys who offered explainations have also thought that it was to do with an induced voltage as these bulbs do not require that much energy to flicker or light slightly.

 

[Marvo]

A few of the guys who offered explainations have also thought that it was to do with an induced voltage as these bulbs do not require that much energy to flicker or light slightly.

They don't consume much energy to glow dimly but if there's several of them in the circuit then it's definitely a fault none the less.

If you're not a sparky you'll need to get one in to find this kind of fault. Also you'll need the proper test equipment, finding this fault isn't a job for a DIY type multimeter.

 

[ytpower]

i come from china,we produce led light. yt power company limited can help u

 

[thehighlander]

I am a spark,and take it when you talk about mega you are referring to insulation resistence tester. I do have one of these which is manufactured by megger.
I will test the circuits but find it hard to believe the same fault would exist in two seperate circuits and that it would clear when used in a one way switching arrangement.
I would not be so quick to discount that there is some capacitance coupling going on in the 2 way 3core & E cables, as having been on DIYnot forum a lot of people have been reporting the same thing happening when changing out halogens for LED's. They seem to work ok when they connect them as a one way, have these people also got the same fault? Some of the advice has mentioned using contact suppressors to fix the problem. Has anyone else come across this,was it a fault current that was causing it?

 

[Richard Burns]

There are various other threads on here about this, Markie Sparkie has given a full explanation in this thread http://www.electriciansforums.net/electrical-forum-general-electrical-forum/46668-led-glow.html

 

[thehighlander]

Hi Richard,

Thank you very much for that information. That has answered the question for me, marky sparky has written a very good explanation of the cause of this problem.
I will get the relevant resistors from RS and hopefully that will cure the problem.

Cheers.

 

[Marvo]

There are various other threads on here about this, Markie Sparkie has given a full explanation in this thread http://www.electriciansforums.net/electrical-forum-general-electrical-forum/46668-led-glow.html

From the post you linked to that was written by markiesparkie

Once more lamps are added the loading on the cable capacitance is such that the individual smoothing capacitors do not reach "firing" potential.

The op stated there were multiple lamps on this circuit so the phenomenon described is very unlikely.

I'm still inclined to think that there's an actual fault. Either way it would be best to get someone to professionally test the circuit.