barrowhill wrote:
Paul,

Thanks for this reply. Performed re-test using both PSU's having first
checked, off board, both working and all connectors showing correct
voltages. Tested both with data and power connectors to 2 x DVD
connected, then again with data and power connectors disconnected.
Original fitted PSU powers up (well fan and fan lights at least) when ON
button pressed in both test cases. Spare PSU doesn't !!! (again in both
cases)

I can't think of any reason why this is the case as both tested fine
off board.

I've run out of ideas now in trying to resolve original problem other
than coming to conclusion that the moth board is faulty.

I hesitate to suggest this, but the PS_ON# signal can be activated,
while the main connector is connected to the motherboard. Since
PS_ON# is open collector, grounding it can turn on the supply and help
verify the rest of the motherboard is operational.

The problem with doing this, is it defeats safety systems on the
motherboard. For example, say your processor heatsink/fan fell of
the processor. On a normal computer, that would trigger THERMTRIP
(sent from processor to motherboard). The motherboard would
respond, by turning off PS_ON#. If you ground PS_ON# (effectively
jamming the setting in the ON position), then a THERMTRIP event
will not result in the power supply being turned off, and the processor
would be "cooked".

So using that test case, grounding PS_ON# while main power is connected,
can be used to verify the computer actually boots, but it does not
solve the problem of why the motherboard logic is not currently
driving the signal itself. As you say, it could be a defective
motherboard, but you also have to look at the circumstances, and
whether there were similar behaviors (difficult to get system
to turn on), before this happened.

*******

There are various versions of power supply standards, to help
you determine pinout on the cabling. While cable colors are
relatively standardized, I have at least one extension cable
here, that doesn't follow the rules on color. And when I'm using that
cable, I have to "count pins" to find the right signal.

http://web.archive.org/web/200304240...12V_PS_1_1.pdf (page 27)

http://www.formfactors.org/developer...X12V_1_3dg.pdf (page 30)

http://www.formfactors.org/developer...public_br2.pdf (page 37)

Paul