Thanks Forc5, A very interesting question.
Congrats on your server. OK, I'll speculate that psu 'testers' just can
not load a psu the way(s) a m/b does. Sp, when I
supect a psu is OK, I blame the m/b. I believe that m/b's are really
complex; and, lots of things can cause trouble and
frustration. PSU'sare less complex; well, in their basic designs, but
there are lots of places that their regulation can
go wonky. We have all seen this I am sure. And, I could be on very thin
ice with this opion!
I do not know how much 'communication' goes on between a m/b and its'
psu, but I suspect that any/all psu
testers only 'test' via some static load to check the psu's regulation
of its' voltages. M/b's put real, active demands
on the psu; something I suppose a tester just can not do, perhaps. I
also accept that there are 'testers', and, then
there are other 'testers. I recall when psu testers became availableand
folks started reporting about them. I always
questioned how do I check-out, and/or prove that my tester is not wonky?
Never did solve that, so, I never bought one.
Best,
Duncan
On 04/19/2013 12:30, FORC5 wrote:
My server went Dead last week, PSU tested OK with two different
testers so I *assumed* MB infant mortality.
New MB in still DOA, BUT works with a different PSU. My first guess
was accurate but why would a PSU fire up with a tester but not run on
a MB ?
Suppose I needed a spare MB I guess. Problem with that is I messed up
the pins putting the cover back in the old MB CPU socket. Would not
admit that to just anyone :'(
Not sure if the PSU is under warranty or not but not going to worry
about it. Would probably test good for them as it did for me. very
strange.
Only good news is server is back up and running. <:-|
fp
