PC shuts down while gaming or kernel power 41 (63) – possible power supply problem?
PC shuts down while gaming or kernel power 41 (63) – possible power supply problem?
Hey there! I’m trying to figure this out, and your situation could really help save you some money if we can avoid replacing parts. - I’ve been facing problems for about a month now. During gaming sessions, my PC keeps restarting unexpectedly without any BSOD. After the restart, Event Viewer shows a Kernel Power 41 (63) error code. The Bugcheck codes and parameters are all zero, even with the ID 70368744177664 and keyword (2).
Usually, this kind of issue points to a power supply problem since I had a failing PSU before that same issue. But this one feels different. I can play games like Hades for hours, but once something heavy like Dying Light 2 kicks in, it can take anywhere from half an hour to several days just to restart.
To try to find the cause, I’ve done a lot of things: reinstalled and updated drivers, re-seated RAM, ran MemTest86, updated BIOS, disabled C-states, enabled XMP profile for RAM, checked SSD and HDD health, even tried forcing a BSOD in settings. Still, nothing came up.
Until Friday night, I used my PC last. I ran an OCCT stress test using Power Test, and it restarted in just 10 minutes. I logged the results with HWINFO and ran the test again—since OCCT usually hides errors after a restart. If anyone has time to review this log, it would be really helpful!
Also, my PSU is the second one I’ve had since 2017, and I don’t want another replacement. It’s from the same brand and model as before. Here are my specs:
- CPU: Intel i7-8700k (3.7GHz)
- Motherboard: Gigabyte Z370 Aorus Gaming
- RAM: 32GB Corsair Vengeance DDR4 3000MHz (2x16GB)
- GPU: 10GB Windforce Eagle GeForce RTX 3080
- Cooler: Corsair H60 V2 Hydro
- PSU: Corsair 850W RMx Series Modular 80 PLUS GOLD
- Monitor: ASUS TUF VG289Q 28-inch 4K HDR
- SSD: 2TB Samsung 960 Pro
- HDD: 3TB Toshiba HDWD130 (7200 RPM)
- Thermal Compound: Arctic MX-4
- Sound Card: Asus Xonar DSX
- Wi-Fi: 802.11 AC1300 867Mbps/5GHz, 400Mbps/2.4GHz
Thanks a lot!
(Attached is Test1Log.xlsx in Excel format as HWiNFO saved it)
Also, the log ended when the PC restarted again, so hopefully someone can see what went wrong.)
Have you checked for memory errors? It's uncommon—only twice in my life have I encountered a defective bit in a chip that could be repaired. Also, try figuring out the highest power usage your PC components have and make sure your PSU is around 75%. Otherwise, consider upgrading it. These are just suggestions based on what I've seen in my PCs. Libi
I mentioned in my first comment that I ran all memory tests with MemTest86 twice and found nothing. I also confirmed I have a sufficient power supply—probably around 850W, which matches the specs I listed. It really surprised me, though. I tried the Repair Reinstalling tool and got a black screen but still saw some mouse errors. I’m completely puzzled right now. I don’t know what’s going on, and if it’s related to the power supply, it’s definitely something new I haven’t encountered before.
With a heavy load it resets. The main temperature during OCCT testing was around [insert core temp]. Other readings included [list additional temps].
It's true. I experienced random restarts while idle, though only once. I logged this in Excel during an OCCT test. The file ends when my PC restarts and stopped before any Temp problems. Since I've had no issues before, I didn't suspect a Temp issue. My pump supports CPU cooling, I use GPU fan curves, and I raised the temp limit in MSI Afterburner to avoid throttling. If you can't view Excel, I could share it in another format, though it might be less clear. HWiNFO saved the log as CSV or a file so I kept it as is.
I noticed you checked the situation carefully. The VRMs are getting very hot, often above 95°C—ideal temps are around 80°C or lower. It’s usually best to keep them warm for a short time before issues arise. I recommend placing a fan directly on the VRM area to cool them down. If their memory isn’t showing a temperature reading, adding a spot fan could help prevent further problems.
The VRM MOS is overheating significantly. This seems unusual. Have you considered any possible causes? I installed the GPU last year and replaced the PSU before, but I haven’t noticed any temperature issues previously. Could you tell me where the VRM is located on your board? Also, by the way—do you know if your motherboard includes a RAM temperature sensor?
96c performs well under load, though it may struggle with prolonged use without proper cooling. Adding a fan is definitely worth considering. A budget-friendly solution can make a big difference. Temperatures have risen over the past few years, especially during hotter summers. The case design directs heat away with large heat sinks—definitely noticeable. Some high-end memory modules include temperature monitoring sensors. Generally, more advanced components come with built-in temperature tracking.