Question Computer crashing while under little to no load, but does not crash under heavy load

[Veggieknight]

Hey folks! I built my computer about 2 years ago, and have no problems since then. In the last two weeks, I would come home from work to find my computer had restarted. Strange, but maybe it was a windows update. I continued using it normally, until it started crashing while I was doing normal idle business like using Chrome or browsing my steam library. At this point I've tried everything I can think of and can get on short notice from Best Buy.
What the problem looks like

• While idle (meaning when running nothing, or just a browser, or just Event Viewer/File Explorer), my monitor will go dark, and all of the fans in my case will spin up. Then, I will be shown the brief booting image, and will be taken back to the login screen. Event Viewer shows a critical failure, which is listed below. When I have ran stress tests to see if this is due to the computer overheating, the computer is totally fine, but as soon as the tests are done, there is an error again. If I run a video game in the background (I've done it with Minecraft and Risk of Rain 2), the computer will not crash while the game is open.
• I was able to replicate the problem only twice by editing the sleep/power settings in Windows Settings. The pc had been running on High Performance Mode, but when I switched it to Balanced Power Mode, it instantly restarted. I was able to replicate this twice, but more attempts showed that might have just been bad timing.

PC Build

• Processor: AMD Ryzen 7 3700X
• Motherboard: MPG X570 Gaming Plus (MS-7C37)
• RAM: 32 GB of G-Skill RAM (2 Sticks)
• Power Supply: Corsair 850x, Fully Modular, Gold Rated
• Graphics Card: NVIDIA GeForce RTX 2080
• CPU Cooler: Corsair H100x (liquid cooled)

I built this computer myself about two years ago, and have had no issued until now. I am not sure what software or physical damage could have caused this issue. Here are some more details about the problem and things that I've tried.
Event Viewer After Crash

• (Source of the crash) Critical Failure, Kernel-Power - 41
• (Could be related?) Error, WHEA Logger - 18

Software Solution Attempts

• CMD -> SFC /SCANNOW (Which ran and was successful)
• CMD -> DISM.EXE /ONLINE /CLEANUPIMAGE /RESTOREHEALTH (completed successfully)
• Updated BIOS for my Motherboard to the most recent version
• Turned off "Link State Power Management"
• Uninstalled/Reinstalled Chrome (assuming it was a Chrome issue, but this did not work)
• Removed Device with a broken driver (one of my peripherals was giving an error in Device Manager, so I physically removed the device)
• Ran OCCT, which is a stress testing software, to see how the computer would handle a heavy load. After two tests that lasted 30 minutes and 1 hour respectively, the computer passed with flying colors. Then crashed after I closed OCCT and was only looking at event viewer.
• Ran Malwarebytes Premium - Antivirus
• Ran Webroot Premium - Antivirus

Hardware Solution Attempts

• Under recommendation from a store near me, I purchased a new SSD, removed my old drives, and installed this new one. I also downloaded a fresh install of Windows to a USB. The computer crashed multiple times during install.
• Reset CMOS
• Reseated RAM (tried running with one stick, and then the other - both resulted in a crash after some time)
• Reseated/cleaned Graphics Card
• Reseated/cleaned Power Supply

My theories (based on internet searches and other recommendations)

• Computer is Browning Out, and the Power Supply is faulty for ??? reason. I have not taken out my Power Supply completely because I am still in diagnosis mode. The only issue is that it only crashes (from what I've seen) under a low intensity load.
• Motherboard has some sort of hardware failure that is causing this resetting problem.
• I DO NOT think it's an overheating problem, especially due to the stress tests being successful

I am at my wits end - I would love to get some input here before I go off buying all new parts. Any and all theories/advice/criticisms of what I've tried is welcome.

 

[Roland Of Gilead]

Hey there,

I suspect the PSU might be involved here. Which exact one it? RMX? Either way, it's a pretty good one, but it may be failing. PSU crashes typically manifest when underload, but also randomly too.

If you have another PSU to test with, then that's your first option. If not, then you could bring to local repair store and get them to swap out the PSU and test.

The kernel 41 means there was a power cut to the system. This is often a PSU. The other possibility is the electricity cables in your house, and if they are up to scratch. Do you have the PC connected to a trailer lead, or UPS or anything?

 

[hotaru.hino]

My guess is there's a problem with the CPU power delivery when it goes down to a lower power state, because things can get unstable when voltages are barely at what's required of the CPU to run. You could test this out by going into the UEFI settings and setting a fixed voltage at around 1.2-1.3V and setting the multiplier around what the base clock speed in MHz / 100 (so in this case 36)

If this "fixes" it, you can either play around with those settings until you find something that satisfies your performance wants or replace the motherboard, because the motherboard's CPU VRM may be faulty.

 

[scout_03]

see this whea 18 https://social.technet.microsoft.co...vent-id-18-microsoft-windows-whea-logger.aspx

 

[Veggieknight]

Which exact one it? RMX? Either way, it's a pretty good one, but it may be failing.

The kernel 41 means there was a power cut to the system. This is often a PSU. The other possibility is the electricity cables in your house, and if they are up to scratch. Do you have the PC connected to a trailer lead, or UPS or anything?

It is the RM850x, yes! My apologies for not being more detailed in the original post. I currently live in an apartment complex, and the PC is plugged into a surge protector that runs my monitor and desk lamp. The power supply is a great lead - thank you!

 

[Veggieknight]

see this whea 18 https://social.technet.microsoft.co...vent-id-18-microsoft-windows-whea-logger.aspx

Thank you! I've seen this link before, I stumbled across it during my initial research. Unfortunately, I've already updated Bios, Motherboard Drivers, Hardware Drivers, and temperature/fan speeds look okay. That includes the power supply fan - it seems to be operational. It looks like the error could occur from poor voltage regulation/power spokes, which could be what's happening... thank you!

 

[Roland Of Gilead]

Thank you! I've seen this link before, I stumbled across it during my initial research. Unfortunately, I've already updated Bios, Motherboard Drivers, Hardware Drivers, and temperature/fan speeds look okay. That includes the power supply fan - it seems to be operational. It looks like the error could occur from poor voltage regulation/power spokes, which could be what's happening... thank you!

Did you clear CMOS after the bios update? If not, you will need to do that first to rule out.

 

[Roland Of Gilead]

It is the RM850x, yes! My apologies for not being more detailed in the original post. I currently live in an apartment complex, and the PC is plugged into a surge protector that runs my monitor and desk lamp. The power supply is a great lead - thank you!

Just check the lead has a fuse big enough. Try it without surge protection, just to see.

 

[scout_03]

it could be a board issue and in bios power settings uncheck restart on power lost and see when you cut power and restart it system load as it should .

 

[matyvan]

Hi, I wonder if you managed to fix the issue as I'm currently having the same issue with quite similar hardware (X570 MoBo and RM850i are the closest ones)
Thanks

Addition: I've had some RAM issues as well but figured it was probably just a dirty slot, that part seems to be fixed now