Hello, Paul.  We were talking about - -

 PR> With the AT power supplies I used an old MFM FH hard drive as a dummy
 PR> load so it would regulate.

Yes, BIR many of those AT PS's would even turn on and fire up with just
a 5 1/4" FDD attached. So I think an old MFM has prolly plenty of
+5V and +12V load there as required to spin an old clunker up, but it
seems to me (I got no proof) that the key to testing out these
"Questionable" ATX PS's under load is in the interaction of the +5V and
+3.3V rails.

The PS I wrote to Ben Ritchie about had ratings of -

+5V at 30A   (150W)
+3.3V at 14A  (46W)
Combined power on +5 and +3.3V = 160W MAX.
FWIW, those last two are rather puny for a power-hungry system with
something like an Athlon of over 1GHz or so.

A "good quality" ATX 350W PS here has
+5V at 32A  (160W)
+3.3v at 28A (92W)
Combined power at 220W Max.
This is more like what is needed for a system with something like an
Athlon 1.4GHZ chip, which can require 72W max just for the chip.

 JH> What I'd like to have for ATX PS testing is a box with
 JH> multi-lightbulbs, and pin points for reading voltages, which would test
 JH> all of the various voltages at one time, during reasonable power
 JH> loadings.  But - I've never seen such a tester offered for sale.

 PR> If I get a dead board some day I'll torch the connector off it.  With
 PR> that in hand something could be cobbled up.  I like your light bulb
 PR> idea.  Auto parts should work.

I've recently seen that there are "ATX Power Supply Extension Cables"
available, with a M and a F connector at the ends, for about $5, so
maybe available for less elsewhere.  Anyway, one of those could be cut
up and used to connect to the power cable from the PS, to connect to
various loads.

Right now, I'm re-thinking, about Matt's procedure of using progressive
Mainboards and etc. of increasing "value" to test an unknown PS.  That
idea is easier and more do-able for me, mostly because where I volunteer
we got _lots_ of "presumed" and "judged" dead mainboards.

I'm remembering also about what Matt posted in the recent thread
here about mainboards with bulging capacitors, where he had seen bad
mainboards where the PS also had to be replaced.  Seemed like those must
be a case of "murder-suicide" , and I know it can be a costly
procedure to connect a valuable 'test bed' component to a "murderer"
component. (I'm still grieving over my "fried" Duron 700 test chip).

Good luck - - -  JimH.
... Gone crazy, be back later, please leave message.
--- MultiMail/MS-DOS v0.32