Question What could be causing random shut downs and reboots?

Globber

Distinguished
Jan 30, 2014
203
1
18,695
I built a fresh system about two months ago. About a week after building the system, I had a day where my PC was randomly shutting down and rebooting. I pulled my cables out and re-seated them all, and switched my power strips and tried just straight plugging directly to the wall. It was good for a couple weeks, then had another day where it was doing it again. It seems to be very inconsistent with these reboots and shutdowns. Windows Event Reader is giving me event ID 41 for kernel power loss. I have tried about everything I can think of for troubleshooting, and am at a loss other than thinking the CPU or motherboard is defective. I am running an X570 Taichi with Ryzen 3600XT, could me having both the 4-pin and the 8-pin ATX CPU power plugged in be causing this as well?

Thanks!
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
Sorry for lack of info. I was using an EVGA 1300 G2, thought it was PSU so I pulled it out and switched to a Corsair RM850X PSU and made sure I pulled all my cables out and replaced them when I dropped the RM850X in my PC.
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
name and model of psu and you could test it with occt to see if it hold test for about 10 min note voltages before and on test you could use hwinfo sensor page t o follow them .

I do run HWMonitor and HWInfo while using my PC. I will try a 10 minute OCCT test tonight when I get home, but having the same issues with 3 different PSUs. I also have a SuperFlower 850w PSU.
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
i would try the breadboarding process to see if the is any short from board to case and in the same look carefully on both side of it for any abnormal parts or broken link on the back .

Was really hoping not to get this deep into it. I guess that's the price I pay for having my GPU watercooled as it is quite a process to drain my system, and pull my mobo back out. I think I have all but decided if I have to pull the mobo out, I'll go back to air cooled GPU and just simply get a different mobo. I really wanted to like this mobo, and have had zero issues with any of my 7 previous mobos(3 of which in this same tower), but at some point it's time to call a spade a spade.
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
since you remove it from case check if the watercoling did not leak on the board or gpu .

I check my watercooling system once per week. It is bone dry on all my fittings and PCB's. I installed specific drain points that are well below and away from all of the components since I have a huge PC case. But yes, I will be pulling it apart and and going back to the ACX cooler on the GPU once I have some time to pull it apart. The strange thing is that the Event 41 instances only happen every few weeks. Stress tests have all come back fine, so I am convinced there is a short somewhere. Thanks for the assistance!
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
could be a cable or under board where standoff are the other thing could be a loose gpu cable that the outside shell touch the case and ground .

That is what I am thinking, however I have inspected my cables at least 4 times, and have replaced them and used now 3 different sets of cables for GPU, 24-pin, CPU ATX power and SATA cables. My cables are all in great shape, but will check and re-seat them again. Once I get my system drained and my motherboard pulled out, I'll let you know what the underside of the motherboard looks like. I will also check the chipset fan cable to see if that might be pinched between heatsink and mobo PCB.
 

Globber

Distinguished
Jan 30, 2014
203
1
18,695
I have been racking my brain like what the heck is wrong with my PC? I take a lot of care installing my hardware and take pride in taking good care of my items, care with my coolant and not dropping or scratching my hardware. A bit of a perfectionist if you will. I finally came across a reddit thread that was LOADED with folks having the same issue as me, with the same BIOS version 3.40. MY board came with version 3.40 on it, so I never thought that could be the issue. They all have stated that they rolled back to version 3.00 and have not had an issue since. Now there is a new version out, 4.00 which is the new AMD agesa 1.2.0.0. Wondering if I should try that version, or just roll back to 3.00 and see what happens. I also notice that versions 3.20 and 3.40 are over 17MB while all other versions are around 14MB. MAkes me wonder if they overbloated something in those 2 BIOS versions. Any thoughts?