Jump to content
Corsair Community

suspect possible problem with my hx850i


thenamesnick

Recommended Posts

I built the system I'm presently using just over 6 months ago, and for the first 3 or so months the system seemed mostly fine, aside from an occasional issue with windows updates. About the time mid May rolled around I started to notice a slight intermittent buzz coming from my psu. I've looked around on youtube to try to find a video of someone with a very similar sound coming from their psu and did manage to find one. [ame=

]
[/ame]

 

At first I disregarded the issue, because I noticed no other problem with the system when the issue appeared. About a week or so later I notice my system hung during boot up. I decided to simply system restore back a few days and hold off on windows updates for a bit, but the intermittent hang on boot continued to occur. I would also like to add this problem is very intermittent and though I had this occur just yesterday the system had previously gone over 3 weeks without hanging on boot. (problem has happened on cold and warm boots)

 

I continued to live with the intermittent hanging on bootup for a little over a month as the issue was so intermittent it didn't seem worth troubleshooting.

 

About a month ago I purchased HOI4 and while playing that game I noticed my system would intermittently just reboot. Typically it occurs later in the game as more stuff starts happening in game. I checked my system event logs and only see the dreaded kernal power failure message in the logs. No useful system error or bugcheck information appears in the logs and no crash dumps are being recorded by the OS for these reboots. I have confirmed the system is properly configured to create dump files as well. At first I thought perhaps something was overheating. At this point I installed the link software that came with my psu and started monitoring my temps, and I also popped the side panel off my case and started assisting it with a box fan. I've also been checking the digital temp readout on the mobo to confirm its temp is close to the one being recorded by the link software.

 

While doing these things I also decided to run some benchmarks (3d mark) and a few minor diagnostics. I ran the intel cpu diags several times, MS memory diag several times, I ran prime95 for about 15min to see just how hot my cpu would get in a worst case scenario and it never seemed to go over 60c. (I typically run my system in a room that is over 80f. ) I did notice during 3d mark that my video card was pushing just over 80c without the help of the external box fan cooling the case. I figured this might very well have been the cause of my issues and decided to run with my box fan cooler for a while to see if the issue returned.

 

My system didn't have any further issues for a little over two weeks and then it rebooted in HIO4 again. At this point I started really digging at the issue. I threw the game into windowed mode and started watching my temps as the game ran. Once it rebooted again even while seeing safe temps for everything I realized I didn't have a thermal issue. After the system finished rebooting I started the link software up again and went right back into the game in hopes that the issue would reoccur while I was watching the link screen. Sure enough I got luck and the issue occurred again in less then an hour, and at this point I confirmed all my temps were good. Video card was staying under 60c, NB under 51C, CPU also under 50, PSU 39c.

 

The one thing that did catch my eye prior to the last failure was vcore flux seemed a bit dramatic. I was seeing vcore drop as low as 1.08 in game and jump as high as 1.17 in very short time periods. After reading that changes of more then .05 can cause problems including reboots and crashes I decided to manually set the cpu voltage to 1.14. I picked this voltage as this was the high end of what I'd typically notice it running at in game. At this point, I went back into the game and started to really push the system hard. While playing the game from a point that the system failed I also fired up and ran the intel cpu diagnostics in the background several times. I was unable to get the system to reboot in game and the voltage the crazy flux has yet to return. The intermittent hang on boot issue remains however, so perhaps it is unrelated or the reboot problem simply hasn't reappeared yet but will in the near future.

 

I'm not using any OC on this system, and the voltage change I made to the CPU is the first time I've messed with any such setting on the system and was only made to help deal with vcore flux.

 

Here's a condensed list of some of the things I've done.

 

have run intel cpu diags well over dozen times no errors.

ram memtest for 8 and a half hours no errors

ran MS memory diags 7 or 8 times no errors

run virus scans

tried new surge protector

have confirmed temps are all stable highest temp reading ever was video card just over 80c

cpu temp avg during typical loads remains under 50c

mobo also remains below 50c.

ran HD tune error scan no erros found drive health appears normal.

disabled speed step

disabled c state

disabled EIST

set vcpu voltage to 1.14 due to droppage.

 

I will point out if not for the buzzing noise I might not suspect the psu so much at this point if only for the fact that I've noticed no other issues playing other games on this system and reboot has not occurred outside hoi4 which I'd normally consider a software issue, but in my experience software issues typically leave useful error information behind when they occur.

 

Any suggestions I could get on this one would be greatly appreciated.

 

System Specs:

CPU I56600k (not OC'd)

memory 16gb corsair vengeance LPX DDR4 (not OC'd)

mobo MSI M5 Z170A

PSU HX850I

video card gigbit GTX 970 OC edition 4GB

cpu cooler noctura DH-14

Link to comment
Share on other sites

  • 4 weeks later...

Well so far with the vcore set at 1.14 for about a month my system has yet to reboot, and given how windows and driver updates are disabled on this system it is looking more and more like the issue was indeed being caused by vcore fluctuations. (it still fluctuates slightly due to vdroop but is way better then before)

 

In the event the system remains stable for another few weeks should I be RMAing the PSU or mobo over this issue? I really would like to be able to use auto voltage for my cpu, and I'm also concerned that in the future as I upgrade the system or decide to try OC that I'm more likely to run into more power issues.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...