New build crashes randomly after starting Windows

[Kroopniock]

Hi all,

 

I have recently built a PC and it keeps randomly crashing from few seconds after windows login screen to even hours after working in Windows. There is no blue screen, the monitor goes black, PC loses HDMI signal and the fans are still spinning. When in BIOS this does not happen, only when in the system. Holding down power button does not turn it off, I have to cut the power in the PSU to turn it off.

 

My specs:

Gigabyte B450M DS3H V2

Ryzen 3 3100

Kingston DDR5 HyperX 2x4GB 2666MHz

Samsung 860 EVO 250GB M.2 SSD

Kingston A400 480GB SSD

Gigabyte GeForce GT710 1GB

Seasonic S12III 500 PSU

 

BIOS version is from August 2020

All drivers are recent

I tried to use only Samsung SSD and reinstall the system, the same with Kingston and it didn't help. I tried resetting BIOS, disconnecting everything and putting it back, didn't help.

 

Anyone can help?

[Radium_Angel]

Look in the event viewer and the subsection for Windows logs then System, see if there are any Red ! marks relating to the video card/drivers etc. If so, either post the full error message here, or a screenshot of the error message(s) 

[Kroopniock]

5 hours ago, Radium_Angel said:

Look in the event viewer and the subsection for Windows logs then System, see if there are any Red ! marks relating to the video card/drivers etc. If so, either post the full error message here, or a screenshot of the error message(s) 

I checked the event log and there is a warning a second before the screen went black. It says: source DistributedCOM, id 10016. This happened 3 times at once. It mentions SecurityAppBroker, WscBrokerManager and WscDataProtection. It also happens when in clean boot mode. There is also a critical error Kernel-Power 41 (63) but this one happens when I switch the power of on the PSU. Sorry don't provide a screenshot, PC turns off now minutes after system starts and on top of that my system is not in English so it would be hard to understand the error description.

[Radium_Angel]

49 minutes ago, Kroopniock said:

DistributedCOM, id 10016

I do not believe those are related (DCom errors are fairly common things)

A few things we can try:

(1) MemTest86. Just to ensure your RAM is good

(2) If you are using XMP, turn that off. I see more problems with XMP than it seems to solve.

(3) A LiveUSB install of Linux Mint. What this will do is verify your hardware is either at fault (you still get the problems) or it doesn't crash and therefore it's a software issue.

(4) It could be a bad PSU, but eliminating that as an issue would require swapping it out with a brand new one, and that may not be possible.

 

 

 

[Kroopniock]

1 minute ago, Radium_Angel said:

I do not believe those are related (DCom errors are fairly common things)

A few things we can try:

(1) MemTest86. Just to ensure your RAM is good

(2) If you are using XMP, turn that off. I see more problems with XMP than it seems to solve.

(3) A LiveUSB install of Linux Mint. What this will do is verify your hardware is either at fault (you still get the problems) or it doesn't crash and therefore it's a software issue.

 

 

 

 

1.I have already run the MemTest86 and memory passed it.

2. I do not use XMP.

3. I haven't tired that yet. I will try now and update the result.

[Kroopniock]

I tried with Linux Mint and the same issue happened. Screen went black and the fans keep spinning. I guess it points to either PSU, graphics card or the motherboard.

I can swap the PSU with a used one tomorrow. Maybe I'll be able to swap the GPU as well.

[Kroopniock]

On 4/18/2021 at 8:40 PM, Kroopniock said:

I tried with Linux Mint and the same issue happened. Screen went black and the fans keep spinning. I guess it points to either PSU, graphics card or the motherboard.

I can swap the PSU with a used one tomorrow. Maybe I'll be able to swap the GPU as well.

I tried with different PSU and it was going good in the beginning, the PC was on for about 2 hours but then the same thing happened. Tomorrow I will have a borrowed GPU to test this more. I have not taken out the CPU yet as there was no indication that it might be overheating. The CPU temp monitor was showing temps below 50 C. At this point it's either GPU, motherboard or CPU.

[Kroopniock]

I have tested the PC with another GPU and I happened again, but it was running longer than with the previous GPU. I am running out of ideas. I will take off the CPU to see if something is wrong there.

[Kroopniock]

I cleaned the CPU, applied new thermal paste and installed again. The same issue happens. I thinks it's the motherboard, anyone has other ideas? Is there a way to confirm that the mobo is at fault?

[Kroopniock]

I have managed to fix the issue. A BIOS re-flash with the newest version helped. Weird IMO that this was BIOS' fault but after the flash all the issues went away.