Low 3.3v causing reboots under load

Uncle Petey · 2 May 2016 at 16:53

Hey guys, I wonder if any of you could help with further diagnosis on this issue?

So I had a Superflower 1200w platinum running the rig in my sig (3970x / Rive) until it started randomly rebooting when under gaming load.
I then installed a couple of things in AI suite to see if anything looked dodgy and it turned out that the 3.3v was dropping to anywhere from 2.8 - 2.9v before the system crashed.

I then replaced the PSU with a Superflower 650w Platinum a few weeks ago while I get around to RMA'ing the 1200w.

At first, it all seemed fine (if still not reading the full 3.3v), but now it's dropping again and reading around 3.072v at idle in bios and dropping to around 2.9v under load before rebooting.

These load readings are with AI suite monitor, not a multimeter, but I can get my hands on one if needed.

Reckon my mobo is going belly up, or is it just a bit of bad luck with PSU's? Any ideas for other testing methods to eliminate/diagnose the issue?

Cheers.

pastymuncher · 2 May 2016 at 17:12

Software isn't very good at reading psu voltages so if you can get hold of a multimeter use that as it will give you a accurate reading.

Uncle Petey · 2 May 2016 at 17:14

Yeah, I know

Fwiw, it's readings do tie in with the issues though...

stuartharrison · 11 May 2016 at 10:30

I'd check with a multimeter
(My nzxt software probably isn't very good but it's giving my 3.3v as 2.03v - if I put my multimeter across I'm sure it will read 3.3v)
It would be unlucky to have 2 PSU's with the same problem

Uncle Petey · 11 May 2016 at 13:22

It was the mobo

I've now changed cpu/mobo and everything's reading and working as it should.

When I changed PSU, it was reading slightly low, but worked ok initially.
It then started to do the reboot under load thing after a couple of weeks, and then began to randomly reboot when browsing or at the desktop.
It then rapidly got so bad that it could rarely even get to the desktop before rebooting and would usually end up on a boot loop.

I initially thought I'd worked around it by connecting the extra 4 pin power cable on the RIVE, but that was short lived, despite 3.3v reading ok @~3.2v.