Does anyone test drivers before using them?

I've had two drivers from DX which worked briefly then proceeded to destroy the attached neutral white XP-G LEDs. Is it common practice to test the drivers with a dummy load for a while before building them or was I just unlucky with the drivers I received?

It just depends. Most of the time when I am working with boards I know and trust, I dont. If they require special attention such as programing etc or if there is some unknown factor about the driver like its output. Other wise I hook up to what ever is going to be the light engine the first time and test the unit together before putting it in the light.

What driver exactly are you referring to? Did you mount the LEDs your self?

Why does your post count say 3?

vestureofblood said:

What driver exactly are you referring to? Did you mount the LEDs your self?

Why does your post count say 3?

Click to expand...

The drivers are DX SKU 26110 and SKU 3256. I bought the LEDs already mounted on stars from Shining Beam (I think).

I've bought the components to build a few variable voltage regulators and some power diodes to serve as dummy loads. I'll probably build these up, as I don't want to burn out any more $5 LED's.

DX customer service is very good at replacing bad drivers, but that doesn't fix the burned LEDs. I also had problems with a DX 32954 module. It works with 6 V input, but overheats and doesn't function right with 9 V input. So, I'm wary of DX part quality control at this point.

I've mostly been a lurker for many years.

I test drivers with a salvaged LED before putting them into a light, or light engine. (Finding out that the driver doesn't work after it is installed is a disappointment that I don't want to experience again.) The LED is mounted on a heatsink, and has leads attached for easy testing. I join the leads of the driver and LED with a quick solder joint, and test the board with the proper battery. Never twist the wires together for testing purposes. If movement disconnects the LED and driver, the driver could be destroyed. Many drivers aren't protected against an open circuit. QA/QC on the driver boards from DX, and the like is inconsistent. Don't get comfortable with a type of board just because others like it have worked. To be safe, test each board before you install it.

There is no doubt that things like this happen with DX parts. There is one other thing to consider since you had this happen more than once with two separate drivers. How are you attaching the star to your heat sink? The layer of thermal epoxy between the star and he sink should be less than paper thin. Just a smooth even gloss basically. If there is even a little burr holding the star up this could cause the LED to poof.

What did the led act like right before it quit? Was there any change in the color? Was the light running when it died, or did it just stop coming on one day?

vestureofblood said:

How are you attaching the star to your heat sink?

What did the led act like right before it quit? Was there any change in the color? Was the light running when it died, or did it just stop coming on one day?

Click to expand...

I'm using a thin layer of epoxy and hold the LED down to the pill using the reflector.

I didn't notice anything change with the LED behavior. The lights worked for a few minutes then just stopped working.

I tested two of the 3256 drivers and both fried with 8v and 14v input using a resistive load. These drivers are supposed to be short circuit and thermally protected. I'll have to test all of these before using them I guess.