Kernal power
Kernal power
Hi everyone. I'm facing problems with two systems after the holiday updates. The first is my partner's work PC/home file server, which handles graphic design and photo editing tasks. It also serves as our file server. Its hardware includes a Gigabyte Z790 UD AX (demo), Mobo I7 14700k (new), a Noctua NH-L15 (from previous setup) with two 32GB GSKILL Trident Z5 DDR5 modules, an RTX3060 12GB (used), a Kingston SKC3000D 2TB M.2 (new), a 1TB M.2 (reused from old system), and four XSATA drives (from previous setup). The PSU is a reused 750W M.2 PSU from three years ago.
The second system is my media/gaming PC, used in the living room. Its specs are an Asus Strix H470-i Gaming (used), an I7 10700K (used), a Noctua NH-L12S (new), two 16GB Corsair Vengeance LPX RAM (new), a Kingston SKC3000D 2TB M.2 (new), a Western Digital 1TB M.2 (from old system), an RTX3070 8GB (used), and an Corsair SF850L 850W PSU (previously reported as new).
Both systems are experiencing kernel power problems, causing random restarts on the first and BSODs on the second. Despite both being fully updated with the latest BIOS, I can't find any event logs for the BSOD and haven’t managed to capture a photo of it. I used Windows 10 after a recent update, which became stuck in a boot loop when an update was applied. I checked the Windows support pages and they listed possible causes, but the Event Viewer seems to indicate power interruptions without clear details.
This is really frustrating. I've been assembling computers since I was 16, and this has never happened before. Any additional info would be greatly appreciated. Thanks.
Ensure the newest chipset and disk controller drivers are up to date. Check if the timing of these kernel-mode events aligns consistently or varies unpredictably across systems. Determine whether this issue has recently appeared or has been ongoing for some time.
Thanks for the update. It seems these setups are quite inconsistent. My partner's PC handled it multiple times in an hour before dropping off, while HTCPC stayed idle for days without errors. Both systems haven’t been used much due to problems and holidays. The biggest strain on my partner’s machine was running Plex Media Server and downloading programs for work. I’ve only recently set up both systems over the past three weeks. After switching to Windows 10, things got worse—she got stuck in a boot loop after an update. Initially, Windows 11 worked fine for a few days, but after a Windows update, it became unresponsive. I tried installing Windows 11 on her Kingston M2 drive, but it didn’t install properly. My partner’s PC has the latest drivers, while I downloaded ASUS drivers and only half of them worked. The main issue I’m facing is getting my partner’s system to function reliably.
I don’t really aim to just throw parts at them since I’ve already invested more than usual in an upgrade. But perhaps I’ll pick up a 1000-watt PSU for my work PC and see if that resolves the issue. Is there anything I can do to perform a fault check further? I’ve run the hardware monitor to verify the PSU power delivery, but on the work PC it doesn’t display a 5V or 12V reading. The HTPC shows strong Power On signals for 3V 5V and 12V.
It seems both systems share the same problem, possibly linked to a common cause. I’d follow the trace to find a single underlying issue. Have you experienced similar issues with other devices like TVs or lights? Probably, you’ve already ruled those out. I’d examine the memory timing—your larger system uses DDR5 while the smaller one uses DDR4. If XMP is enabled on both, consider lowering the speeds or disabling it, and see if stability improves with just one stick. If that doesn’t help, try the other stick. Kernel power might be affected by external factors or insufficient supply quality. Since you’re concerned about your partner’s PC, test their 850W PSU in that setup. Make sure cables stay properly connected to the power supply, and avoid swapping pin configurations. Also, verify that the memory matches the CPU and mainboard specifications before purchasing replacements.
I recently replaced the safety switches in my home due to a power problem on another circuit. Because of age, I had them all updated. I verified compatibility on the Gigabyte website and Parts Picker, which suggested using at least an F8 BIOS; I chose F12 instead. I’ll double-check later. I’ve disabled XMP, cleared the CMOS, and reset the BIOS. I plan to purchase a new Silverstone 1000W PSU and swap out the power strip and cord to see if it helps. Most cases online turn out fine once a new PSU is installed. In a recent twist, the HTPC provided bug codes last night—something I’ll review after a long evening at work.
I just verified everything again and all systems seem to work. My HPTC is listed correctly, but the RAM isn't showing up in the supported list. It looks like I might have overlooked something during the check. The RAM I received is CMK32GX4M2E3200C16, while the supported version is CMK32GX4M2C3200C16. A single letter difference changes the timing slightly. I have some extra RAM from an older system; I'll test it to see if it fits.
It seems there was a change in the power unit, but the problem persists. I managed to reset the system by using the snappy driver installer, though the moment I launched it and began downloading it, the system would reset again. I tried this three times last night after entering the PSU information.
Took a bite of the bullet and restored Windows 11. It seems the boot loop problem has been resolved after the update.