On Sun, 9 Jan 2000 06:02:06 +0100, Samuel W. Heywood wrote:
> OK, thanks a lot for the suggestion Ron. I'll certainly check this out
> when I get around to it. I would suppose that the "power good" signal
> is OK because my hard drive always spins up immediately, and it remains
> spinning even after a crash condition occurs.
>
I might worry about assuming that "power good" is above
suspicion. If, for example, the +5V line was at 4.5V (all
figures ficticious, but you'll see the idea!), then the "power
good" line might be saying "power NOT good" but the HDD might be
quite content to run up and do its thing, since not everything
in life is as fussy about spot-on correct supplies as a
motherboard is. On the other hand, if the "power good" detector
in the PSU is actually duff, it might be saying "power NOT good"
when power is just fine. That'll still reset the CPU but not
worry peripherals.
What you really would need to do is to monitor the signal on
"power good" and see what it was up to when things got chilly!
Since it's easily accessible and measurable, maybe that's a good
place to start? It's just possible that the signal may only be
INTERMITTENTLY wrong when cold (noise/ripple on line) so that an
oscilloscope would be more likely to spot the problem than the meter
would.
Hope I haven't fogged the issue more!!
Ron.
