Windows 10 randomly shutting itself down

[chrysstreet]

So, about a week ago, i woke up, booted up my PC and started playing some games and browsing the web, when suddenly, out of nowhere, everything closed and windows gave me the “Shutting Down” screen before turning off my PC.

I though “That’s weird” but thought it was just some bizarre anomaly, until it happened again, this time when i booted up the PC, windows refused to load, so i restarted it, and when i did i was faced with a “Recovery” screen, from which i uninstalled the latest updates and rolled back to a system restoration point four days prior.

I imagined that would fix it, but it didn’t, and it continued happening until i finally decided to reinstall Windows.

I did, and for the first day, nothing happened, until today it started happening again.

I have no clue what could be causing this, as it never happened to me before, and i would love to get some insights.

[Bruce]

first thing to rule out is heat.
open up the case and blow out any dust build-up.
remove the memory chips and video card (if add-on) and brush the circuitry and the contacts and the motherboard slots.
clean the air vent and fan blades.

see how that goes.

then you’d be trying to rule out faulty hardware; graphics cards, memory chips, power supply and hard drives.

have you noticed any common factors; time of day, program being used, length of time PC is on for, etc?

give us your PC specs.

[chrysstreet]

Hello, Bruce. Thanks for the reply.

I have already cleaned the case thoroughly, it was my first go-to solution when i first encountered this problem. It yielded no results.

The temps are OK, both at idle, and while gaming, the attached “TEMP.png” file shows the overall situation of the PC after about an hour playing “DayZ”

I have noticed no significant pattern either, this can happen 2 minutes, 2 hours or 20 hours after turning the PC on. It happens while gaming, while browsing the web, watching Youtube videos or simply organizing folders on Explorer. The only time where i haven’t seen it happen yet, is while idle, i have yet to observe it shutting down to itself while left untouched. Though it can be said that the biggest culprit is Opera, but i wouldn’t make that a significant connection, since i think the reason i experience it more while using it, is because i use it so much more than anything else.

The attached “POST.png” image relates to something that i just experienced while booting up the computer to respond to your reply. I have never seen this screen before, nor do i know what it relates too. Prior to this, i had shut down the computer myself, manually, it had not crashed.

The attached “KING.png” image shows the “Kingston SSD Manager” app, i installed it to check if the SSD was faulty, due to the nature of the problem, it doesn’t seem like it is.

Another oddity i need to inform: I have a hard time writing this reply, as Windows has shut down some 8 times while i was writing this, something that has NOT happened before, not at this rate, one after another. It almost seems like someone doesn’t want me to get help.

Thanks in advance.

[chrysstreet]

Quick update.

Never mind about it not happening while idle, it just happened.

Whatever the problem is, it seems to be getting worse, the PC is barely usable now.

[chrysstreet]

Another quick update.

Today my computer just wouldn`t turn on. It would try, the fans would rotate for about 3 seconds and then it would try again, endlessly.

Removing both my HDD`s made it so the computer would turn on, though just disconnecting them did nothing. I had to remove the SATA cables from the Mobo.

With this information in hand i think it is safe to assume Windows 10 isn`t at fault, but hardware itself.

It is still shutting itself down at an accelerated rate which makes it unusable though

[Bruce]

let’s test what we can…

remove all but one memory chip, start it and see how you go.
keep swapping the memory chips, only ever having one inserted at a time, until the issue goes away (hopefully).
if it still occurs regardless of what chip is in use then you either have all memory chips as faulty, or it’s not memory related.

if there is a add-on graphics card, remove it and use the mobo graphics port if you can.

and if the fault is now happening when idle, disconnect all SATA units (SSD, HDD, ODD) and leave the PC on.
obviously you’ll get the POST message - no bootable volume found.

that then leaves the PSU and mobo.
getting your hands on another PSU would be great - neighbour, spare PC, friend etc otherwise you’re up for buying one which may have to happen but it may not be the PSU. suck it and see stage now.

and the mobo - well, not much mere mortals can do with testing those apart from replacing it.

[chrysstreet]

Okay, so i ran a few tests, here is what i have:

I checked the memory chips, one by one. Using both of them, with D.O.C.P enabled or not, i ran into shutdowns and bluescreens. Using only the stick that was originally in the first slot, i got shutdowns, but no bluescreens, and swapping it for the stick that was on the second slot, i ran into no shutdowns or bluescreens.

With this information, i decided to physically check the stick i removed, i noticed some oxidation on the electrical contacts, i cleaned that up and put it back in its slot, after doing this, i managed to reconnect all the HDDs and get the PC to turn on, no more problems with it not wanting to boot, or POST screen, but i was still getting shutdowns and BSODs.

This made me worried, as either the chip decided to fail suddenly, after a year of usage with no problem, or the motherboard killed it. So i decided to run memtest86, with both sticks connected and DOCP enabled to see if i would get any errors, and i did. But the error message that i had left me even more confused, as i am now not sure if the issue is with RAM, CPU or MOBO

Hopefully, you can learn something from the data i collected, because i am even more confused. BSODs and memtest86 results attached.

I couldn’t test the Graphics Card, as i have no integrated graphics.

[chrysstreet]

Okay, so i ran a few tests, here is what i have:

I checked the memory chips, one by one. Using both of them, with D.O.C.P enabled or not, i ran into shutdowns and bluescreens. Using only the stick that was originally in the first slot, i got shutdowns, but no bluescreens, and swapping it for the stick that was on the second slot, i ran into no shutdowns or bluescreens.

With this information, i decided to physically check the stick i removed, i noticed some oxidation on the electrical contacts, i cleaned that up and put it back in its slot, after doing this, i managed to reconnect all the HDDs and get the PC to turn on, no more problems with it not wanting to boot, or POST screen, but i was still getting shutdowns and BSODs.

This made me worried, as either the chip decided to fail suddenly, after a year of usage with no problem, or the motherboard killed it. So i decided to run memtest86, with both sticks connected and DOCP enabled to see if i would get any errors, and i did. But the error message that i had left me even more confused, as i am now not sure if the issue is with RAM, CPU or MOBO

Hopefully, you can learn something from the data i collected, because i am even more confused. BSODs and memtest86 results attached.

I couldn’t test the Graphics Card, as i have no integrated graphics.

Something i forgot:

I also noticed that with the “bad” memory chip, i get lots of stutters and all-around weirdness while using the PC, this happens if i use it alone, or with the other stick. This doesn’t happen if i use the “good” chip by itself.

I have to admit i may have become biased against this memory chip, as i am kind of hoping it is the culprit, and not the motherboard. What leaves me worried is the fact that it suddenly decided to “pop”, i am worried the motherboard caused this, and could potentially damage other components, as i am not sure a memory can just suddenly develop oxidation and die like this by itself, without the MOBO frying it.

Eager to hear your opinion on this. Thanks.

[chrysstreet]

Update:

The memory chip that i had previously described as safe and shutdown-free, has indeed caused a shutdown, after being left idling for a couple of hours, followed by POST screen when i turned the computer back on.

Now i am truly at a loss.

[Bruce]

so you have 2 x 8GB 3000MHz Corsair chips?
are they overclocked?

have you disconnected all bootable drives except for the OS drive?

yeah, having 2 faulty chips at the same time is rare.
I’d be getting the power supply tested at this stage.
might be time to take it to a computer shop as they can test the PSU and swap the memory in store and see if that helps - unless of course you have spare memory and a PSU lying around.

[Rustys]

Thread closed lack of response from original poster

[chrysstreet]

So, i have recently opened a thread here asking for help in solving a problem i am having on my computer.

https://pchelpforum.net/t/windows-10-randomly-shutting-itself-down.56932/#post-104912

Unfortunately i caught COVID-19 and had to be hospitalized for a week, and thus could not reply to that thread, and it was closed. But now i am back home and i am still trying to troubleshoot that problem, so i once again ask for your help.

Summary: Recently my computer started shutting down by itself, it happens at random, sometimes while playing games, sometimes while browsing the internet, sometimes while doing nothing at all. It can take 2 minutes, or a full day, with no pattern at all, i can not force it to happen. After it happens, my computer just won’t turn on afterwards, it will try, the fans start spinning, but it turns off again, and keeps turning on and off until i remove it from outlet, sometimes it will not do this, and instead simply won’t turn on, when it happens i have to leave it for a few hours and try again.

Since the problem first started and i made the previous thread i have:

Reinstalled Windows, along with all of the Drivers available for my components, this did not solve the problem, as it happens even with a fresh windows installation and nothing running at all. So i ruled out Windows and Driver problems.
Tested my computer with another Graphics Card that is working okay on it’s computer, and the problem persisted, so i ruled out the GPU.
Checked the temperatures, and everything is as it should be, even when gaming. So i ruled out temperature problems.

Now here is where it gets confusing:

Around the same time i also started getting some BSODs (screenshot attached), and started to investigate those, believing my RAM was causing them, so i ran MEMTEST86 and i got a CPU error (screenshot attached).

It was at around this time where i got sick and had to abandon the search, but now i came back and ran some other tests to see if RAM or CPU was at fault.
Windows Memory Diagnostic detected no errors (screenshot attached).

Prime 95’s Blend test was giving me “Prime 95 ERROR: rounding was 0.5 less than 0.4 and hardware failure detected” after running for a few seconds, it happened multiple times yesterday, but i forgot to screenshot this, today, i could not replicate this result, the test ran fine.
IntelBurnTest was also giving me a critical failure, said my system was unstable, but this result was also different today as the test ran completely fine, with no errors (screenshot attached). Keep in mind i did not change anything, not in the BIOS, not in the system, not Hardware, so i cannot explain this change.

I suspect the CPU error in MEMTEST86 should be gone as well, but i haven’t had the time to test this yet.

This leaves me with two possible culprits, the PSU and the MOBO, but i have no idea how to test these two.

I tried with the PSU, at first, i would get it to turn on by removing a few components such as the HDDs, which made me think it was not handling everything in the system, but then this would not be enough, and eventually the problem would happen even with only the MOBO, CPU and RAM connected to it, with the GPU and SSD and HDDs all removed. And sometimes, it would not happen at all, with everything connected to it. This randomness makes everything so much harder. Can the PSU be at fault? Or maybe it is the MOBO? How can i test this?

Thanks in advance.

[Rustys]

Combined the two threads and unlocked. Glad t see that you are back and doing better.
@Bruce

[Bruce]

these sort of random issues are buggers to pin down.
certainly sounds like software is not the root cause.
and as you have tested, the graphics card can be ruled out.
that only leaves the memory, motherboard and power supply as likely contenders.

in my experience, those memtest programs don’t tend to give a reliable outcome, as you are also discovering.
their results should be taken with a grain of salt.

the only testing mere mortals can do is the PSU, either the paperclip test (which simply only proves it power cycles really) or use a power tester that shows the condition of all the rails - green for good, red for bad. they cost about $30.

the only real world test you can do with RAM and mobo is the replace them, preferably with ones at hand, but not everyone has that luxury.

as stated before, this is probably the point where you take it into a PC shop - not only will they have some gear there to test things but they can replace parts and pin down where the fault is. any part they try that doesn’t fix the issue isn’t charge for so you only pay for the replaced part. and they test it etc.

so your decision now is, pay someone to fix it or buy a new PSU (for starters) and see if that helps.
if not, then new RAM, again, see if that helps.
then, if not, you are chasing a new mobo which potentially forces you to upgrade your RAM and CPU due to architectural changes.

[chrysstreet]

I forgot to mention, i actually did take the PC into a shop, the same one who built it actually. They could not find the root of the problem, and all they really did was disable DOCP in the BIOS. I am guessing they were thinking RAM was at fault, so they set it to run at lower speeds, that did nothing though.

Well, since the processor gives me different results depending on time of day, and since i have little evidence that RAM is the culprit, i will see about buying a new PSU.

On that note, quick question: Can i downgrade the power supply without problems? My country’s economy is quite broken, and as a result, hardware became quite a bit more expensive. Right now i have a CX650M, but i think i won’t be able to get another one of those since it’s hard to find, and it’s overpriced when available, the next best thing being the TX650 that costs almost as much as my MOBO+SSD, so i am looking at getting a CX550M.

Specs are as follows:

Asus PRIME B450M-A
AMD Ryzen 5 2600X
2x Corsair Vengeance 8GB 3000MHz (CL16)
Asus GeForce GTX 1660Ti
SSD Kingston 480GB
2x HDD Seagate 1TB
3x 120mm Case Fans

Can it reliably and safely handle this setup?