I am making some undercounter lights and have encouneterd a problem. I am using 3 Seoul p4's powered by a Xitanium 350mAH driver. The light casing is aluminum. I used heat compound on the back of the stars for heat transfer. Since the compound wasn't setting up enough to keep the leds in place, I put some clear epoxy over the star for the sides I wasn't going to solder. I made two bars. One has one led, the other has 2. I had no problem with one of the two sets of light bars when they were separate. I attached the two light bars and turned them on. I am running them in series (that means Positive to Negative, right?). The center led had a dim glow for a minute then stopped working, the two outer ones never came on. I disconnected the driver and tried them on each individual led. None came on. I have another driver and tested each led. Nothing. I tried a spare luxeon on both drivers. Lights up without problem. With the Seouls, some give a tiny spark when I touch the wires to the contacts. What is wrong? I can explain further if this isn't enought to help me out. Thanks in advance.
Issue with driver or leds?
- Thread starter mitaccio
- Start date
Help Support Candle Power Flashlight Forum
uk_caver
Flashlight Enthusiast
If you try connecting a Luxeon in series with one or two of your Seouls, does it still work?
If it does, could you measure the voltage drop across the Seouls while the Luxeon is lit?
Unlike Luxeons/Crees, the Seouls include a reverse-connected protection diode as part of their static electricity protection, which starts conducting at a low voltage, which means that if power is applied wrongly, there's the possibility of damage, though I'd have thought that wouldn't be likely to be instant with a 350mA driver.
When you say the Seouls give a 'spark', is it correct to interpret that as meaning a brief flash from the main emitting area?
If it does, could you measure the voltage drop across the Seouls while the Luxeon is lit?
Unlike Luxeons/Crees, the Seouls include a reverse-connected protection diode as part of their static electricity protection, which starts conducting at a low voltage, which means that if power is applied wrongly, there's the possibility of damage, though I'd have thought that wouldn't be likely to be instant with a 350mA driver.
When you say the Seouls give a 'spark', is it correct to interpret that as meaning a brief flash from the main emitting area?
uk_caver
Flashlight Enthusiast
If by the 'contact point', you mean the blob between the die and the external negative connection, that's the place that can glow if reverse voltage is applied, and where I assume the protection diode is.
From limited experimentation, reverse voltage application can cause the protection diode to fail open, leaving the LED working, or can fail in a way that leaves the LED broken, and (at least in the way I experimented) it seems to be something like a 50/50 chance of either result happening.
From limited experimentation, reverse voltage application can cause the protection diode to fail open, leaving the LED working, or can fail in a way that leaves the LED broken, and (at least in the way I experimented) it seems to be something like a 50/50 chance of either result happening.
uk_caver
Flashlight Enthusiast
OK - I misread what you'd written.
There's obviously some current path if you're getting a spark, but unless there's something particularly odd about the stars and the wiring between LEDs is OK, the current must be going through the LEDs, in one direction or another.
If the current is going in the right direction, but nothing is lighting up, that would seem like the LEDs had failed short-circuit.
It's certainly possible for a string of LEDs to be killed by a driver failure, with all LEDS failing short-circuit, (I've seen it happen to strings of 6 or 8) but I'm not sure if that's at all likely with a Xitanium.
It's also possible for the odd LED to just die, but multiple spontaneous failures seem highly improbable.
Are the stars original Seoul ones?
I seem to remember hearing about some third-party stars which had some LEDs on backwards, but then you said you'd tested out the LEDs before mounting, so that doesn't seem a usable explanation.
There's obviously some current path if you're getting a spark, but unless there's something particularly odd about the stars and the wiring between LEDs is OK, the current must be going through the LEDs, in one direction or another.
If the current is going in the right direction, but nothing is lighting up, that would seem like the LEDs had failed short-circuit.
It's certainly possible for a string of LEDs to be killed by a driver failure, with all LEDS failing short-circuit, (I've seen it happen to strings of 6 or 8) but I'm not sure if that's at all likely with a Xitanium.
It's also possible for the odd LED to just die, but multiple spontaneous failures seem highly improbable.
Are the stars original Seoul ones?
I seem to remember hearing about some third-party stars which had some LEDs on backwards, but then you said you'd tested out the LEDs before mounting, so that doesn't seem a usable explanation.
If you got your Seoul stars from DX or Kaidomain, I would check to see that the LED is oriented on the star correctly. There have been a few isolated issues in the past where the LED was on the star in a reverse polarity orientation. This would kill the LED very quickly.
There are small tabs next to the leads. One is solid, and the other has a hole in the middle. The solid tab indicates the positive (+), and the tab with the hole is the negative (-). You might want to check that before you proceed much further.
If the LED's were placed on the boards incorrectly, it would be a good claim to have them replaced.
There are small tabs next to the leads. One is solid, and the other has a hole in the middle. The solid tab indicates the positive (+), and the tab with the hole is the negative (-). You might want to check that before you proceed much further.
If the LED's were placed on the boards incorrectly, it would be a good claim to have them replaced.
