F5F Stay Refreshed Hardware Desktop Computer powers down intermittently and occasionally restarts on a daily basis

Computer powers down intermittently and occasionally restarts on a daily basis

Computer powers down intermittently and occasionally restarts on a daily basis

F
FireLion973
Junior Member
Find
6
07-19-2024, 03:47 AM
#1
I own the system in my signature and it operates smoothly at a cool temperature. It also functions well under Prime95 (at 95°C). I have PBO turned on and RAM configured with XMP at its rated speed. There is no dedicated GPU. The power supply should be excellent for the components. My main demanding tasks are converting HD videos in Handbrake, which runs at 88°C. All components like VRM and SSD remain very cool. A few months back my wife mentioned the PC would shut down occasionally, then fans would spin up to full speed. She only accessed it remotely via workstation. No CPU overload was observed. This also happened to me. Two times during video conversion at 88°C, and once casually. Usually the PC stays off but fans keep running at maximum. Recently it restarted itself automatically. It's hard to pinpoint because it occurs sporadically every few days or weeks. I've already tried: updated BIOS, chipset, CPU drivers; checked connections; ran Prime95 tests. There might be a memory (MB) issue since fans often run full speed after shutdowns. But I need more concrete steps to confirm. Could this be related to PSU? I could swap out the PSU or turn off PBO and XMP, but since it happens occasionally, I'm unsure if that was the cause. Maybe I should disable XMP and wait a few weeks. Would anyone recommend faster diagnostics?
Reply
Reply
F
FireLion973
07-19-2024, 03:47 AM #1

I own the system in my signature and it operates smoothly at a cool temperature. It also functions well under Prime95 (at 95°C). I have PBO turned on and RAM configured with XMP at its rated speed. There is no dedicated GPU. The power supply should be excellent for the components. My main demanding tasks are converting HD videos in Handbrake, which runs at 88°C. All components like VRM and SSD remain very cool. A few months back my wife mentioned the PC would shut down occasionally, then fans would spin up to full speed. She only accessed it remotely via workstation. No CPU overload was observed. This also happened to me. Two times during video conversion at 88°C, and once casually. Usually the PC stays off but fans keep running at maximum. Recently it restarted itself automatically. It's hard to pinpoint because it occurs sporadically every few days or weeks. I've already tried: updated BIOS, chipset, CPU drivers; checked connections; ran Prime95 tests. There might be a memory (MB) issue since fans often run full speed after shutdowns. But I need more concrete steps to confirm. Could this be related to PSU? I could swap out the PSU or turn off PBO and XMP, but since it happens occasionally, I'm unsure if that was the cause. Maybe I should disable XMP and wait a few weeks. Would anyone recommend faster diagnostics?

Reply
Reply
L
LoliosYT
Junior Member
Find
10
07-19-2024, 03:47 AM
#2
I completed the W11 RAM test before, but only the standard one. Now I ran the extended version that lasted 35 hours in two cycles (it mentioned it would take a few minutes). No problems were found. I've also been using Prime95 blended for a few hours. The highest temperatures recorded were: CPU 95°C, PCH 34°C, PCIEX16 30°C, CRM MOS 61°C, VSOC MOS 55°C, RAM SPD hub 40°C, SSD PCIe 4 32°C. This suggests the cooling and power supply should be fine. The RAM looks normal. Based on this testing, I don’t see any issues. Could you try additional tests? Besides confirming connections are secure, what other physical checks could help? Also, I experimented with using my monitor’s hub—plugging in a mouse, keyboard, and printer via USB to the monitor. I’ve done this for years without problems. I extended the Prime95 run, but it already exceeds typical load times for videos or heavy tasks.
Reply
Reply
L
LoliosYT
07-19-2024, 03:47 AM #2

I completed the W11 RAM test before, but only the standard one. Now I ran the extended version that lasted 35 hours in two cycles (it mentioned it would take a few minutes). No problems were found. I've also been using Prime95 blended for a few hours. The highest temperatures recorded were: CPU 95°C, PCH 34°C, PCIEX16 30°C, CRM MOS 61°C, VSOC MOS 55°C, RAM SPD hub 40°C, SSD PCIe 4 32°C. This suggests the cooling and power supply should be fine. The RAM looks normal. Based on this testing, I don’t see any issues. Could you try additional tests? Besides confirming connections are secure, what other physical checks could help? Also, I experimented with using my monitor’s hub—plugging in a mouse, keyboard, and printer via USB to the monitor. I’ve done this for years without problems. I extended the Prime95 run, but it already exceeds typical load times for videos or heavy tasks.

Reply
Reply