What hardware fault can cause random Windows BSoD?

[Crow846]

Hi,

Lately random BSoD errors started to happen on my PC.
Windows will just display an error that Critical process died but won't give additional info.
The PC is not shutting down but just stuck on BSoD.
I made a Windows error check and it didn't found any thing wrong with the Sys files.
Also all hardware tests I made returned clean bill of health.
The errors are completely random. So it has to do something with a hardware failing.
The question is which hardware can cause such inconsistent errors?
RAM, Disk drive or the mother board has a fried component? Some capacitor or resistor maybe that cause random errors?

I don't have a lot of time on my hands to troubleshoot the errors so will appreciate any thought on the problem.

Thanks,

 

[mackerel]

Is it the system in sig? I'm not familiar with that CPU family but in general, undo any overclocks as first step. If official ram speed is lower, try dropping that also.

 

[Crow846]

The system is running at default. the BIOS is returned to default settings.
But the system fails continue.
The SSD is running slow but I don't know if it cause by the mother board or a fault of the SSD.

 

[mackerel]

My top 2 suspects would be unstable ram or unstable CPU, especially if it is running hot. How are CPU temps in normal use? More interested in peaks than idle.

 

[Crow846]

The CPU is running at 30C idle and 50C under stress test.
I don't play any video games or run any 3d rendering software.
So it has nothing to do with the temperature of the CPU. (it crossed my mind as well btw)

I suspect that some component has failed.
I just hope it is not the motherboard. as I will have to build a new PC .

 

[trents]

To be clear, you say the system continues to blue screen even with the overclock removed?

What hardware troubleshooting tests have you tried so far? Did you run memtest86? Alternatively you can use the built-in Windows Memory Diagnostic tool.

You can use CrystalDiskInfo to check for system and storage drive issues. It reads the SMART info on the disks and reports problems that can indicate failure trends.

 

[Crow846]

I ran System check via CDM.
I ran Memtest 86
I ran SSD manager tool and ran Crystal Disk Info. and it is healthy. IMO it is rubbish but it won't detect the SSD as broken so it must be fine. (right?)
I removed the overclock as first aid. still BSoD.
What other troubleshooting tool can I use?

 

[trents]

Have you stress-tested the system say with P95?

 

[Crow846]

I never stressed the system for a long period of time but as the BIOS is returned to default I don't see any reason to do a stress test.
Also my PC is never under stress. So it shouldn't effect anything.
I think the SSD might faulty or the mother board has finally failed after 8 years of service.

 

[Mr.Scott]

What's the code on the BSOD?

 

[Crow846]

I don't think there is a code. only one text line "Critical_process_died" then a message asking me to show that error to a system admin.
I think a random process dies and then windows 10 shuts down if that process was important.

BTW I checked for viruses with AVG and Malwarebytes and it checked out clean.

 

[Mr.Scott]

That sounds like OS failure to me, more so than hardware failure.
Have you tried repair install or a fresh install?

 

[Crow846]

Can OS really fail? It is Windows10 we are talking about and I never had a Windows that actually failed.
Sure that viruses and a lot of trash cause issues but my OS is clean and barely something was installed on it.

But I did made a System check via CMD command, scannow.. and it returned negative result.

 

[trents]

I have a feeling this is either your motherboard or your RAM.

Even though you have tested RAM with memtest86, one thing I would do is remove all but one stick of RAM. I would try each stick of RAM in each RAM slot you have been using, one at a time. I had a problem like yours a few years ago and it turned out to be a burned trace between the CPU and one of the RAM sockets. After I took the motherboard out I could see it on the backside of the board.

 

[Mr.Scott]

Can OS really fail? It is Windows10 we are talking about and I never had a Windows that actually failed.
Sure that viruses and a lot of trash cause issues but my OS is clean and barely something was installed on it.

But I did made a System check via CMD command, scannow.. and it returned negative result.

Yes. Happens all the time no matter what version of windows you run. All it takes is one bit of corrupted data to screw it up.
Run a repair install. At least your data won't be lost.

 

[EarthDog]

After a repair install, if it happens again, I'd look at a hardware cause of RAM or perhaps the mobo as trents said earlier.

 

[Mr.Scott]

After a repair install, if it happens again, I'd look at a hardware cause of RAM or perhaps the mobo as trents said earlier.

I agree.

 

[trents]

Since your system was overclocked to begin with it is possible there has been degredation of the components over time. Have you tried giving the CPU a little extra vcore and the RAM a touch more voltage above stock default voltages, even in at default frequencies?

 

[don256us]

Memtest86+ was run and I assume no errors were found. I would say that the RAM is fine as is the memory bus on the MB. When spinning HDD fail, you could typically see the system hang and then blue screen. I don't much experience with SSDs failing even though I have more than a few in operation. I think that a stress of the CPU is a valid test to see if the CPU is failing.

How old is the PSU? I see that you have a 900 watt PSU which is well above what you should with that setup. That's not bad but you are probably running it at the 50% level which is usually not the most efficient. If the PSU is old, it may be failing in the 3v, 5v or 12v legs which could be an issue.

 

[Crow846]

Sorry for the long response.
I read all your suggestions but none gave me a definitive answer at what hardware is failing.
As the only possible way to rule out hardware components is to replace them with a working one and hope nothing will fail.
(Sure, with proprietary diagnostics tools it is possible to check hardware for errors but I don't own such tools)
My PC is old as it is and some hardware part is most likely the cause. the PSU will be my best bet.
For now it seems to be stable and for several days I have not experienced BSoD.
So I will see how stable it will remain and when the time comes I will build a new PC. Hopefully with your assistance.
Thank you all for your effort.
I really appreciate you all and this forum.