Houston we have a problemPosted: October 4, 2011
Yesterday I set about connecting all the jumpers and wires inside the computer, all was going well. Indeed I was feeling rather pleased with myself. This was short lived when I attempted to power the machine up and discovered the LED flasher board was not working. On closer inspection, I discovered two wires has come loose. Quickly out with the soldering gun, I had the wires reattached in no time at all. So back to powering the board. After a little tweaking of various connections and testing for conductivity before I powered up again, I felt satisfied that this time the board would work.
Flashing in to life, it was a sight to see all the LED’s on the front panel light up and working for the first time and being power from the internal ITX PSU as well.
My joviality again was not to last for long as i noticed one of the LED’s had stopped blinking then another, until all of them where out. Taking a look at the PCB, I went to place my finger of the microchip to test if it where getting to hot. Only to find a large glob of hot soft glue come off on my finger. The chip had not just got warm, it had fried itself.
At first I couldn’t fathom why this had happened. During all my bench tests, the pcb had run fine through the regulator I had build. The 2.8v feed should not have killed at all, that’s when it struck me. I had accounted for the constant voltage, swapping from the universal psu to the ITX psu. But I had not taking in to consideration the current, the level of amps from the test psu was 1.5mah, how much was it from an ITX power supply? I wasn’t sure, but if i had to hazard a guess. I would conclude in excess of 1.5mah’s. Add to this that the PCB for the blinkers is a very cheap and simple circuit that does not have a current regulator. So nothing was there to prevent the circuit from drawing more current then was needed. Effectively committing circuit suicide.
So today I am out again to out beloved poundland in the hope i can sauce another flasher unit, if not Thrashbarg has offered to design a circuit that will run from the 2.8 volt regulator i have made.
Fingers crossed i shall have it working this time, fourth times the charm I hope.