[M100] Power supply troubleshooting

[Scott McDonnell]

So, if you recall, my newest M100 would not initially power on but I had 
found that if I left it on for a long time, it eventually would power 
up. Suspected caps or nicad.

Earlier this week, I replaced the nicad and recapped it. It wouldn't 
power up after that but I discovered a dirty power switch. If I wiggled 
it, the M100 would power up. So I cleaned the switch with deoxit and 
confirmed the switch is working perfectly now.

But, again, it would not power up. So I sat down with a DMM and an 
Oscilloscope to troubleshoot. Of course, I checked over all of my solder 
joints, looked again for board damage. There was some damage under C85 
from leaking, but continuity checked out.

I noticed that the switching frequnecy was completely dead. Probing 
more, it eventually came up, but was less than a volt. Reflowed joints 
in the oscillator portion of the power supply even though they looked 
fine and suddenly I got a nice 5V squarewave and it powered up. Worked 
for a few power cycles and then nothing again. I am suspecting cracked 
ceramic caps now, but thought I would ping the group with more experience.

Oh, I also noticed the transorb (D27) was getting very hot. I removed it 
and it measures 47ohms. Not certain if that is normal, but for now I 
left it out.


Scott