It's been an expensive night so far. :sadcrying I got everything glued onto the heatsink. The first joint which I thought was too thick was actually OK, but I somehow killed that LED anyway.

First, I set the Mastech to full volts, no amps. I just touched the probes to a red LED. It lit briefly, not super bright, but when I tried it a second time, nothing. I could see one of the 4 emitters was darkened--it was fried. It was almost like the power supply gave it a capacitive discharge.

I fired up the other reds, one a time, with no problems. Let them burn in for awhile; they were fairly consistent 10.8V giving 1.2A and 11.2-3V giving 1.5A. Then, on one of many go-rounds, that first one I glued failed to light. On examination, that one also had one cooked quadrant.

The wire I was using was so stiff it was really manhandling the LEDs. I tried to hold off on attaching the wires to the sink cause I hadn't had a chance too really think about they'd be routed: through a hole, around the side, etc. I managed to rip a solder pad each off two of them. Fortunately, I was able to scrale down to the trace and get a lead soldered to them. That's when I decided to go ahead and just use some thermal epoxy to secure the wires near the LEDs for some strain relief.

Gets worse. Still online?