Unexpected multimeter results (two identical PSUs)

vois2

Distinguished
Jan 6, 2006
72
0
18,630
I have the Dell Dimension 8400 desktop and so far have had the good fortune to have had absolutely no issues. For this thread, I'm going to denote my stock 350W PSU with the letter 'A'. It's the "good" PSU in this story.

Some time ago, a DellTalk Forums user could not get a 7800 GT nor a 7800 GTX to work on his 8400 with stock PSU. He worked with myself and others through all the usual suspects very diligently, but he could not get the 7800s to give an output with that 350W PSU.

He previously *could* get an X800 to work on that PSU, however. So that was oddity #1.

Suspecting his PSU was the problem, he replaced it with some or such brand of 500W -- he chose a quality make. As soon as he changed out the PSU, he was immediately up and running on his new 7800 GTX.

He was kind enough to send me his old 350W stock PSU so that I could use it for comparison.
I'll call his "bad" PSU with designation letter 'B'.

SO today I finally got around to using a multimeter on both PSUs. I ran the multimeter on my good 'A' PSU first, and then I replaced 'A' with 'B' on the motherboard.

As soon as I hooked up his 'B' and connected to AC power, I noticed that the fan on my 6800 GS did not spin up for the usual brief AC power test spin-up. This was oddity #2. Ah ha! I thought in just a moment I would surely see a fault on his +12V pin or peripheral connection...

But check out the results:
PSU 'A' 4.96 11.67 -11.93 11.69
PSU 'B' 5.06 12.00 -11.98 11.99
----------------------------------/ -------/
motherboard(first 3 quantities) ...(last) all +12V pin conn. on the PCIe P10 6-pin

So the results above was oddity #3.

And here is oddity #4 ... the 'B' PSU fails on a standard power supply tester with a load to the 24-pin ATX motherboard connector. On which rail does it fail? The +12V rail !!

Help? What is the problem here? Why doesn't the "bad" PSU allow the 6800 GS to do its little fan spin up? Why does it fail on a PSU tester -- failing on the +12V rail, no less -- if it shows excellent 11.99 voltage throughout the whole continuity all the way down to all +12V rails on all peripheral connections?

I appreciate any thoughts.
vois2
 

jap0nes

Distinguished
Mar 8, 2006
918
0
18,980
i dont know if i understood right... you measure PSU "B" with a multimeter and it shows 12V line ok... than you use a PSU tester (how's that??) then it shows a problem on that line. It seems that is some kind of bad contact or something, as the PSU is delivering the correct voltage (multimeter).