I'm debugging a design for a project that we've already prototyped and tested. I know that the design works. Anyhow...
We soldered the first one on Friday. The only difference is that we didn't solder the DAC and buffer (which will be used for analog control). Instead for testing we are going to use a power supply. When we began testing it with our LED array, it didn't appear to work. I probed the pins, and they all seemed about right. The only thing I noticed is that that duty cycle was very low. I've attached a screenshot further down. Then, when about to measure something else, the LED array flashed, and we got a sizzle and a puff of smoke from the top right-hand side of the board, around Q2 or Q3 (far away from the IC). We then turned it on again, and the LED array lit up. The only thing I noticed at this point is that the LED array would shut off when Iadj reached about 1.2 V, not 2.25 V. I would then need to bring Iadj back to 0, and ramp it up again.
So this morning, I go to continue testing on it, and it doesn't work again. I occasionally get a crackle and the LEDs flash, but it is otherwise non-functional. I checked the gate of Q1, and again, the duty cycle is really low.
I know the design works. I've prototyped it and tested it, so I don't think it's a design issue. Since we soldered it by hand, it sounds like a soldering issue to me (perhaps a layout problem?), but I cannot find what isn't soldered properly!
You can see the schematic, layout and a screenshot of the gate signal below.
