Fault finding series LED outside lighting (collingwood)

[mattg4321]

Got a regular customer who has just bought a house with some series wired Collingwood LED lights wired in series in the garden. Sparks who has done it hasn't done a bad job to be fair, but clearly didn't think too hard about those behind him!

All 9 lights are on 1 switch and 1 driver (constant current). None of them currently work. I'm assuming (perhaps wrongly?) the reason for this could be any of the 9 lights/the driver having failed, thus breaking the circuit to all lights.

Lights are all joined to the SWA's running round the garden in adaptable boxes, then filled with magic gel, as is the driver.

Has anybody got any bright ideas as to how I can locate the faulty light without just selecting lights to link out at random. Obviously I'd check the driver first but if its not that I dont really want to clear out all the gel out of all the boxes until I find the faulty unit.

 

[Deleted member 9648]

Almost certainly the driver. I've worked on a number of these Collingwood series systems and the drivers commonly fail. Check that the number of lights the driver can handle isn't exceeded as well.

 

[mattg4321]

I've fitted similar collingwood gear on a few jobs but it's never failed yet! The driver can take 3-12 lights so isn't overloaded.

Will obviously check the driver first, but if it's not the driver, I assume I'll just be selecting lights at random to link out?

 

[fall-apart-dave]

I know nothing about these systems... But is there a way you can either take another cable as a temp ground? If so, go to light 5, take off the ground there, and return the ground of 5 to the driver rather than to the live of 6, if you follow? Or is there no way to know which order the lights are in?

Again forgive my ignorance, and sorry if I am teaching you to suck eggs, but if you can return at light 5 and it still doesn't work, that tells you the problem is at 5 or before (do the same with 6 onwards to confirm). That'll be a bit quicker than randomly linking. Then ust keep cutting whichever is the offending half in half again and you'll get there in no time.

 

[mattg4321]

Thinking about it I could enter light 5 and rewire it so the first 4/5 lights are connected with the rest disconnected and cut in half and in half again etc from there.

Bit of a pain though, downside of series wiring I suppose.

 

[fall-apart-dave]

Thinking about it I could enter light 5 and rewire it so the first 4/5 lights are connected with the rest disconnected and cut in half and in half again etc from there.

Bit of a pain though, downside of series wiring I suppose.

That works too, the reason I suggested running an extra cable is that would also tell you, if both ends lit up, that the fault was on the ground side. But yeah, swings & roundabouts. Series is a pain... *removed* Christmas lights... Every *removed* year!!! Swapped to LED lights thinking "Yey! No more problems!!!" - WRONG! My LED ones were bloody series fed too, albeit grouned into a dozen or so parallel LEDs series fed.

 

[mattg4321]

Can anybody confirm that if I test a 350mA constant current driver, if functioning correctly I should expect 350mA on my multimeter even with no load connected? If not then the driver is dead presumably?

 

[fall-apart-dave]

You'll only get 350mA under load. Though if testing with your meter that will act as a load though will almost be a short. No idea how these drivers handle a short, what impedance are the lights? Can you match that with a resistor?

 

[mattg4321]

I suspected as much. Sounds like I need a known working compatible LED load to test the driver quickly and easily then.

 

[fall-apart-dave]

Dont take my word for it I know FA about these drivers, just applying logic and know that power supplies dont like short circuits.

 

[davesparks]

Can anybody confirm that if I test a 350mA constant current driver, if functioning correctly I should expect 350mA on my multimeter even with no load connected? If not then the driver is dead presumably?

Yes a constant current driver will pass 350mA through your multimeter, but you will need to meet the minimum load requirement if it has one.

 

[davesparks]

Dont take my word for it I know FA about these drivers, just applying logic and know that power supplies dont like short circuits.

This is a constant current supply rather than the constant voltage supply we would normally work with.
So the current will remain the same regardless of the load, the voltage will vary to maintain that current.
So theoretically a direct short will not present a problem as the voltage will drop down to achieve the correct current, connecting too great a load however can result in the current supply attempting to raise the voltage too high.

 

[fall-apart-dave]

This is a constant current supply rather than the constant voltage supply we would normally work with.
So the current will remain the same regardless of the load, the voltage will vary to maintain that current.
So theoretically a direct short will not present a problem as the voltage will drop down to achieve the correct current, connecting too great a load however can result in the current supply attempting to raise the voltage too high.

With ya. Thats why I said not to take my word for it - though is there not a threshold for the output to operate? I know with the current amps in the amplifiers I used to build, the operating voltage of the transistors on the output of that part of the amp had a threshold below which nothing would output (am I making sense?). Wondered if it was the same with these drivers - a short or near short and they output zip, hence the duggestion of sticking a small resistor in matching the impedance of a couple of lights. But as I say, this is all just my logic, I have no experiencecwith Collingwood drivers.

 

[rich.250]

What voltage is the output of the driver?
You could always disconnect the load from the driver and connect the same or lower voltage drill battery and see if they light up. I know it's not a preferred method.
(This was shown to me by a lighting rep and how they always test led tapes before connecting).


Sent from my iPhone using Tapatalk

 

[davesparks]

What voltage is the output of the driver?
k

It's a constant current driver, so the output voltage is not fixed, it varies to maintain the constant current