A600 voltage problem

[harmonica]

Hi,
I've recently bought A600 from an auction. It was bare Amiga, only with rebuilt PSU from PC. When I plug it and turn on, it boots with no problems to Kickstart, waiting for a floppy insert. The problem is the floppy drive is dead. So I dismantled Amiga, tried to found out something, but nothing seemed obvious. I plugged out floppy drive from Amiga and tried to measure voltage on onboard molex. I found out there that instead 5 V is only 3.5 - 3.8 V. 12 V seems to be correct, it's around 11.5 V. Then I tried to power floppy drive not from Amiga but from another ATX PSU and it worked like a charm. It reads floppies, boots to game etc. The only issue was the keyboard doesn't seem to work except Ctrl+A+A. Caps lock is permanently on.

My first thought was that the floppy problem was insufficient voltage from PSU. So I rebuilt another ATX PSU, but the problem is still the same. Even if I load the PSU with two HDDs. If I measure voltage directly from PSU, it's 4.8 V. If I measure voltage from onboard molex in Amiga it's 3.5 - 3.8 V and it's just too low to power the floppy drive. So any ideas? I checked capacitors with magnifying glass and it doesn't seem to be leaked.

Thanks for any suggestions.

[harmonica]

Yeah, I probably will try. The interesting thing though is the decrease of voltage. I'm not sure if those capacitors near the power connector are used for some filtration. If they are, they might be going to fail. Nevertheless I'll try to look for some original PSU, but I'm not completely sure if it's going to help.

[harmonica]

I don't know. As I said when FDD is powered from external power source and not from Amiga it works flawlessly. Everything is OK except the keyboard. I'd rather suspect some powering issues. Maybe even the keyboard has insufficient voltage.

he 5v line is too low, may be It affected to the CIAs.
Also, a broken chip could be a "power muncher"...        

Well after a few minutes every chip gets warm a little except Gayle. That is cold, but I have nothing connected to IDE...

EDIT:
If the problem would be caused by those CIAs, is there any way to test it if it's real cause?

[harmonica]

OK guys, thanks a lot for your ideas. I've put together TODO for myself:

1) Get original Amiga PSU
2) Measure if capacitors charge to voltage according to their specification
3) If not, replace capacitors
4) If it won't help, probably so long Amiga

Ad CIAs - as ferix says if it controls FDD motor I'd guess CIAs are OK because FDD motor works properly when it's not powered from Amiga while the ribbon cable is plugged in Amiga. I know Vesalia sells these 8520, but I don't feel myself so handy to replace them.

By the way interesting thing is when FDD is powered from Amiga as it should be, it doesn't work, but when I turn the PSU off FDD clicks and it rotates quarter of a rev. I'm not completely sure, but this behaviour could really imply capacitors problem rather than FDD controler.  I'd wish...

It may take some time to me. Nevertheless I'll inform you.

[harmonica]

So, I've found interesting thing. When I measure voltage on the plug that goes from my rebuilt ATX PSU between +5 V and GND it's 5.11 V. When I plug it to Amiga and measure voltage on the back side of motherboard where the power connector is soldered I measure only 3.8 V (GND <-> +5 V). So it seems it could be caused by the power connector or maybe some problems with grounding. Is it necessary the shield pin of power connector to be plugged?

EDIT: Between GND of PSU and GND of Amiga is 0.6 V.