General Discussions

[NyanTech]

After a recent system upgrade that led to a motherboard RMA replacement, I've been encountering unrelenting instability issues with my PC. After roughly 20-30 minutes on the dot from startup my system randomly reboots with no bluescreens, no errors, no hints to point me in the right direction. Any attempts to hunt down the problem with Event Viewer only show instances of Kernel Power error 41. This is consistently happening even when the system is idle, whether the BIOS settings are on default or the DOCP profile on my kit is enabled.

I've already ruled out thermal trip points being the problem because CPU and GPU both were within spec for operating temperature, with the GPU not exceeding 65 degrees Celsius on GPU die and average of 90 to 92 degrees Celsius on the hotspot sensor. The CPU was operating at maximum 60 degrees Celsius in every single game I tested. These temps were pulled from HWInfo. Testing the PSU using OCCT PSU test shows the PSU is operating within spec for voltages on 3.3v, 5v, and 12v rails.

All system drivers and Windows 11 itself are up to date, including the GPU drivers.

These are my system specs post upgrade:
CPU: AMD Ryzen 9 7950X3D

Motherboard: ASUS ROG Strix X670E-E Gaming Wifi

CPU Cooler: Thermalright Frozen Warframe 360mm White

GPU: AMD RX 6900XT, MSI Gaming Z Trio 16GB

RAM: G.Skill Z5 Royal Silver 64GB 6400MHz CL32

PSU: Superflower Leadex Platinum SE 1000w

[johnnyenglish]

Hi,

Kernel Power Error 41 mostly points towards sudden loss of power, this makes me point towards your PSU.

Although 1000Watt is plenty, that is not a very known manufacturer. 

Have you tried safemode to see if it still shutdown/reboots?
Leave it at BIOS screen for sometime and check if it does to.

At this time you need to discard any software issue because it could be Board/CPU/PSU.

To troubleshoot the graphic card, you can use the iGPU.

Good Luck

[NyanTech]

Super Flower is the original OEM for several of EVGA's PSU models, and the Leadex Platinum SE is one of Super Flower's high end units.

But with that being said some updates on testing:

• PC still reboots in safe mode in the same 20-30 minute timeframe as when it's in normal boot mode.
  
  
• I did come across occasional QCODE 99 (which usually means failed I/O initialization) on my motherboard, but deleting and reinstalling my webcam drivers (one logitech C922 Pro Stream) fixed that problem.
  
  
• The random reboots still happen even in BIOS, and the time limit before it triggers is completely random, but seems to be within a 2-7 minute time limit when it does happen.
  
  
• Testing with iGPU only leads to the same time limit before reboot, which leads me to believe it's a board or CPU fault over a PSU fault.

[johnnyenglish]

Were narrowing down.

Have you tried removing and re-inserting all cables, especially the ATX and CPU AUX? I've seen some crashes because of CPU AUX cable not fully inserted.

I would give it a try. Since you mention a board RMA, is the one you have brand new? Refurbished?

A hardware problem is there, shouldn't be rebooting inside BIOS

Good luck 

[NyanTech]

I'm almost certain the motherboard ASUS sent me as an RMA replacement was a refurbished unit, because the pads for the M.2 slots looked like they had some use, and there were other indications that lead me to believe this is the case. CPU, original motherboard, and RAM were bought brand new so there is still a possibility for the CPU being the culprit.

I inspected the two 8-pin EPS cables and they did appear to be fully seated, but I reconnected them just to be safe. I repeated the process with the ATX main cable as well. The PCIe power cables were fine because the GPU was able to consistently draw around 280 watts when it was under heavy load.

I also re-seated the RAM as a precaution.

[johnnyenglish]

Starting to look like board issue.

They problably tested it for 5min and gave it a clean bill of health, then its back to stock.