PC powers off like pulling the plug or freezes solid (clock/mouse/screen frozen in time). Thoughts?

[XClayX]

I have this weird issue that has been getting progressively worse the last month or so. My PC has been freezing (either ERRR noise or simply froze in time) or shutting off like someone is pulling the plug out of the wall. Most stressful thing this does is watch HULU / surf the web. It will freeze about once or twice binge watching a show. If your lucky you'll get through a show or two.

Circa 2014 build in a Corsair 200r case
AMD a-10 5800k (not over clocked) with Hyper 212 cooler
8gb Ram (Should be good)
Gigabyte mATx Board (Latest Bios - Reset to defaults)
HD 7870 2gb card (same issue using onboard video or card)
Seasonic 750w PSU (brand new replaced original PSU same issue)
CMOS battery (Brand new)
Lots of HD's/SSD's (HD's pass windows/ SMART / diagnostic tools I can throw at them)

Drivers
Video (replaced with latest)
onboard Network card (replaced with OEM drivers)

PC to TV connected via 25ft HDMI cable.


Windows logs show "improper shutdown occurred" but I guess that's all I see. My only guess is the system is overheating, but I don't know why or how. Fan's (4 - 120mm) all work and its been cleaned properly configured airflow direction. The computer sits in a cabinet with the full back removed so there should be plenty enough airflow.


Software recommendations to track temps?

Any other thoughts on what to try?

[A.Wilder]

CPU-Z verify voltages are withing spec?

[XClayX]

Running a stress test now. Currently reading 88c (190f) on the cores and stable. I'll let it run for a while and see.

Sent from my SM-G965U1 using Tapatalk

[A.Wilder]

Running a stress test now. Currently reading 88c (190f) on the cores and stable. I'll let it run for a while and see.

Sent from my SM-G965U1 using Tapatalk

Bios and all drivers up to date? Maybe one was updated just before you started having problems? HDD diagnostics? Ram ok? Anything particularly old? I kept reusing some ram I had since it was high-end when I got it, but it had corrupted at some point years later and was giving g me fits trying to find the cause of my shutdowns.

[BRealistic]

That CPU runs hot.
The original is probably crumbling talc by now.
Reapply the thermal paste.

If that's not it.. idk.
Try completely disabling Windows Sleep Mode.
I know that caused me some issues in the past.

Edit.
Have you not done a clean Windows install yet?

[XClayX]

Bios and all drivers up to date? Maybe one was updated just before you started having problems? HDD diagnostics? Ram ok? Anything particularly old? I kept reusing some ram I had since it was high-end when I got it, but it had corrupted at some point years later and was giving g me fits trying to find the cause of my shutdowns.

Running latest Bio's. Circa 2014 for the board.
HDD's pass every test I throw at them. Smart, Standard Windows Test, Seatools etc.
Ram is original.
I haven't really changed anything since it was built. I replaced the PSU and the CMOS battery 2 weeks ago thinking it could be the issue.

That CPU runs hot.
The original is probably crumbling talc by now.
Reapply the thermal paste.

If that's not it.. idk.
Try completely disabling Windows Sleep Mode.
I know that caused me some issues in the past.

Edit.
Have you not done a clean Windows install yet?

The CPU seems to run hot (54c idle, 89c load), I wouldn't think they would be that high with an Hyper212 cooler. Thermal paste might be an issue. I have not done a clean installation of windows.


Test Runs with Prime95
Core voltage are 1.396v (1.400v target per bios) DEFAULT BIOS, temps 89c with prime95. Errors hardware failure detected in P95, around 2mins mark, 2nd attempt around 45minutes.

Lowered core Voltage (-.0500) and HWMonitor reads core voltage now at 1.287v, temps are 74c w/ Prime95 full load.

[XClayX]

According to this the max temp is 74c for my CPU. I'm thinking I might be hitting a limit.

https://safetemp.blogspot.com/2016/0...-temp.html?m=1

Sent from my SM-G965U1 using Tapatalk

[XClayX]

I'm going to replace the thermal paste, I think it will help. Apparently this stuff doesn't last for ever.

Sent from my SM-G965U1 using Tapatalk

[dr_spock]

Do you have power fluctuations in your area? Do you have a UPS?

Do any of your electrolytic capacitors looks bloated?

[XClayX]

Do you have power fluctuations in your area? Do you have a UPS?

Do any of your electrolytic capacitors looks bloated?

Not sure on the power fluctuations. No UPS on it. Capacitors look and feel fine.

New thermal paste dropped it a few C. CPU fan has a slight tick to it but spins freely.

I checked the error logs and its been locking up / restarting 1-4 time a day. Its a htpc so it runs most of the day.

I'm reinstalling video drivers to see see if it helps. The sudden OFF and long post/restart seem like this might be a hardware issue.

Undervolting the CPU slightly seemed to help temps and stability although I havent ran it that long like that. Side effects are stunted performance like watching your keystrokes t...y...p...e...

Honestly I'd like to rebuild this machine. The AMD a-10 APU is slightly faster then a core 2 duo so anything would be miles faster.


Anyone try the Windows 'Start Fresh' option? Any think to watch out for? I'd need to make sure all files are kept safe, programs not so much but photos etc.

Sent from my SM-G965U1 using Tapatalk

[BRealistic]

Do you have an extra hard drive?
Slap that in there and install windows 10.
See if it does the same thing with only a fresh windows 10 install.
I'm telling you, weird things can happen when drivers and such get screwy.

-If you were close, I'd give you one of mine. I have like 5 or 6 extra 500 GB seagate platter drives sitting around I bought used/refurb cheap back when new harddrive prices were crazy. Now... I guess I should figure out how to put them all together to at least use them for storage (since everybody has gone ssd/m.2).