[ QUOTE ]
PeterB said:
Thanks for your inputs! Unfortunately I couldn't resolve the problem.
Here are my new findings:
They are both Madmax+ adjustables (red epoxy).
The pots have the same setting (from an optical point of view).
I've checked and cleaned all surfaces. Everything looks ok(also the solder blob is ok).
At three volts I have the same current with both modules (970mA).
At 1.5V I have 100mA input current for the bad module and 50mA for the good one (the brighness is lower, if I measure the current with the fluke. This is probably caused by the series resistance). As already mentioned, the LED voltage for the bad module is 1.58V in this case.
My theory: The converter is working, but there is a relatively low resistance path between the plus and minus output of the madmax. At low input voltage, the converter is relatively weak and reaches only the 1.6V and therefore the LED is dark. The output reaches higher voltages, at higher input voltages => the led is working at 3V input.
The lower output voltage also explains the higher input current (at 1.5V input).
This "short" (should be around 20Ohm from the measured characteristics) could be caused by a defective LED or by the converter itself.
I hope you have new suggestions for me /ubbthreads/images/graemlins/smile.gif
[/ QUOTE ]
Peter, your theories sound reasonable to me. This sounds like it might be ESD damage to the LED. You can check for this by using the diode function of your DMM. It is OK to do this with the LED on the module [but not energized, of course]. If it reads under 2V, I would take this as evidence of ESD damage. From what you have described, I expect that it may measure less than 1.5V.