System shuts down repeatedly shortly after booting up
System shuts down repeatedly shortly after booting up
I recently replaced the motherboard, processor, and graphics card in my PC. When I power it on, it displays a black screen and immediately returns to the Windows 10 login screen after about five minutes. The fans and lights remain active during the restart. In Event Viewer, I see a Kernel Power Failure (41) error. I've searched for solutions already and tried several fixes, such as upgrading the power supply from 550W to 750W, using a different outlet, re-seating RAM and the graphics card, reinstalling drivers, running sfc /scannow and dism.exe, restoring health with dism.exe, reinstalling Windows on a new SSD, switching to an older graphics card (GTX 1070ti vs GTX 970), and even booting in safe mode. None of these steps resolved the issue. I'm hoping someone can assist me. The attached XML file from Event Viewer is included for reference. Running Windows 10 in safe mode seems like the only way to access logs or drivers without a full reboot. Current specifications: CPU – Ryzen 5 2600, GPU – ASUS STRIX GTX 970, RAM – 16GB (2 x 8GB), SSD – 256GB SK Hynix M.2. New specs: CPU – Ryzen 5 5500, GPU – ASUS STRIX GTX 1070ti, same memory and SSD as before.
Here are some useful tips from the article: https://www.minitool.com/backup-tips/ker...er-41.html
This situation seems to point toward a power issue. Occasionally you notice a click when the device stops functioning. Your system dropped significantly from low to minimum levels. I recommend using a higher PSU—possibly 850 with that unit. Or consider a 750W ATX 3.1 model. The concern isn’t just about raw power, but about sudden spikes causing shutdowns. A 970 model wouldn’t be affected in the same way. You might want to check the heat output; maybe the TIM has degraded or the mount is faulty. Heat and PSU performance are similar except for that particular model. The unusual part is the old hard drive size described—less than a modern phone. It could be overstuffed. I’d consider adding extra mechanical drives for cheap, slow storage and keeping unused items on the case. Drives often behave oddly when they’re near full capacity (over 80%). I’m someone who enjoys memory testing—blue screens or crashes happen gradually. It’s a gradual process that stops you automatically. You just run it while doing another task. Someone suggested using memtest64 instead of memtest86, possibly because of an update. I’m not familiar with memtest64 yet.