Question PC randomly shuts down after closing games

[SgtWireless]

Intel i7 9700k
RTX 3080ti
32gb corsair vengence RAM
1000W PSU
tuf z390-plus gaming motherboard

As the title says randomly when closing any game my pc will freeze up and then restart itself, no mention of a bugcheck in the event viewer just unexpected shutdowns, games themselves and everything else run just fine until then. I have no idea what could be causing this, initially i thought it was was just with battlefield 2042 and forza horizon 5 but now its spreading to other games.
Whilst making this post i discovered one of my GPU fans was jammed but i dont think that was the cause. thermal issues would surely shut the PC down mid game.

 

[Aeacus]

1000W PSU

PSU make and model (or part number) is? Also, how old the PSU is, and was the PSU bought new or used/refurbished?

 

[Ralston18]

Look in Reliability History/Monitor and Event Viewer.

Either one or both tools may be capturing some error code, warning, or even an informational event when the games are closed.

Another thing you can do is to, just before shutting down the game, is to open either Task Manager or Resource Monitor. (Use both but only one tool at a time.)

After opening the tool window simply watch for a couple of minutes. Then shutdown the game.

Look for what happens or changes.

 

[SgtWireless]

i might as well post what my event viewer and reliability manager looks like after the last time it happened. But event viewer doesnt tell me anything specific about the shut down, just listing it as an unexpected shutdown and then a bunch of errors about secureboot. reliability manager shows that the game i was playing had crashed when i closed it normally and then windows had a bugcheck. I really dont know what else to investigate. Everything i play runs as expected until i close them, then theres a chance this will happen. its complete nonsense

 

[SgtWireless]

PSU make and model (or part number) is? Also, how old the PSU is, and was the PSU bought new or used/refurbished?

the PSU is a corsair HX1000 platinum grade and it was new out of the box and around 3 years old.

 

[Aeacus]

Random reboots or shutdowns are mostly caused by 2 issues:
1. CPU/GPU overheats and to prevent any damage, system shuts down.
2. PSU fails to deliver enough power to the GPU or fails to keep smooth enough voltage for PC's operation.

First check your CPU/GPU temps, both at idle and under load. If temps are within reason then it's safe to assume that it's the PSU who is acting up.

Now, while Corsair HX is good quality PSU (initially released back in 2017), there are lemons even among the best. Here, i suggest that you try with 2nd, good quality, known to work PSU and look if you still get the random reboot issue.

Either RMA your PSU (it has 10 years of warranty) or get 2nd PSU to test it out. Good PSUs to go for, are: Seasonic Focus/Vertex/PRIME, Corsair RMx/RMi/HXi/AXi, Super Flower Leadex Gold/Platinum/Titanium.

 

[SgtWireless]

As i said in the post, whilst i was writing the post i discovered a jammed fan in my 3080ti and the GPU itself was lifted too much by its stand that it came with, i will monitor temps and power draw. can you explain why its all fine until i close a game, does closing a game cause a power spike or something? im doing my best to learn how hardware works so i dont have to keep making posts like this.

Edit: just remembered id seen some visual glitches on both my monitors recently, unsure if its related

 

[SgtWireless]

I understand enough to be able to investigate the issue, i just need a good place to start looking, looking over the logs in the managers doesnt tell me enough, stress testing didnt force the issue to occur, ive installed the latest drivers for everything and fixed corrupted system files.

The only saving grace is i can still use the PC for what i want to, just having to hope it doesnt shut down after i close games. Hoping by fixing the GPU jam will have sorted it but i have doubts. I usually have a tech guy on hand for this but hes on holiday so i came here to learn more.

 

[Aeacus]

If your software-detective skills are minimal, as with most people's, start with the obvious: back up your data, and clean install Windows, or reset to factory image. You want to tacklet the software first, as it doesn't cost anything. Only when that's exhausted should you explore HW troubleshooting.

While formatting the OS drive and making clean Win install is free, it only gets rid of software issues. However, random reboots/shut downs are to do with power delivery and there aren't many suspects regarding that. Either high temps (CPU/GPU forces the reboot to save itself), PSU is acting up (there is a reason why PSUs have a warranty) or MoBo VRM is acting up (rare to see, but still a possibility).

Hence why i suggested to test with 2nd PSU. Doesn't have to be brand new one just bought. Can be 2nd good quality PSU from another PC, that is known to work. Or can be PSU RMA, which also is free.

can you explain why its all fine until i close a game, does closing a game cause a power spike or something?

Difficult to tell.

Hardware wise, RTX 30-series is notorious in GPU transient power spikes and while it is possible that GPU power spikes, HX1000 has more than enough capacity to deal with GPU spikes. RTX 3080 Ti is 350W GPU and it could spike 2-2.5 times of it's TDP. So, ~700W-1050W or so.

Though, HX1000 OPP (Over Power Protection), while set relatively low, is still set to 1186.13W (118.6% of max capacity) and it should have enough of a cushion to soak up the transient power spikes. That is, IF HX1000 is sound and working as it is supposed to or unless you'll get the 2.5 times spike, that is too much for HX1000 OPP to handle (also consider other component power draw as well).

HX1000 review, protections page: https://www.tomshardware.com/reviews/corsair-hx1000-psu,5214-6.html

And here's a vid of what GPU transient power spikes are:
(At 12:04 is the point where Steve confirms that one of the PSU companies did find RTX 30-series GPUs to spike 2 to 2.5 times of rated power.)

View: https://www.youtube.com/watch?v=wnRyyCsuHFQ

 

[SgtWireless]

Is there a way to monitor the power spikes whilst i close a game, that would tell us everything i need to know i guess, ive had this GPU for like 3 years and only now does this issue pop up from time to time.

i should point out i always had youtube running whenever the shutdown occured as well.

 

[SgtWireless]

Hey now, dont be rude, i have some knowledge, ive been going over the logs my pc kicks out whenever this happens, but sadly its not giving me much to work with here, if you want more details about the exact crash ill tell you right now then.

So i close a game, and about a minute later the pc stops responding at all, then 30 seconds later, both monitors will go black, then the pc will shut down and reboot automatically, no error message, the event viewer just lists unexpected shutdowns and sometimes reliability viewer will record a bugcheck crash. Everything else is as normal before the crash, and sometimes it just doesnt happen at all. The only consistant factor is its always after i close a game.

What more information can i provide you at this time, do you want the eventviewer logs because i have them.

 

[Aeacus]

Is there a way to monitor the power spikes whilst i close a game

Well, there is, but reliable way to monitor it would require oscilloscope. E.g seen at 1:45 in the video i linked.

Software wise, best option, while still falling short, is HWinfo64,
link: https://www.hwinfo.com/download/

In Sensors Mode, it does have Logging feature that you can start recording the system and it should save the log once system reboots. While it doesn't show the millivolts (power spikes), it at least shows the max power draw and PSU voltage regulation on all it's power rails. So, if e.g 12V rail is way out of spec, you can see it from HWinfo64 sensors.

 

[Aeacus]

By attempting diagnosis without sufficient basis, you (and everyone else) are guessing. If OP can't be bothered to give additional context, even when directly asked (as I did), then there's not much to say. Telling him he should buy this or that is not doing him any favor.

This is what the forum is all about. OP describes their issue, we offer OP to test most likely suspects (based on our knowledge in the field) and we'll see if the proposed fixes helped. If not, then other fixes are suggested. No-one knows exactly what the issue is with a PC. And most of the troubleshooting is consumed by finding out what the actual issue is, rather than fixing it.

E.g same is with doctors. You have health issue, you go and see a doctor. Doctor then gives you medicine that works 60% of the population. If medicine doesn't work for you, you go back and say: "good medicine but doesn't work for me". Then doctor gives you more rare medicine that works for 20% of the population. And so forth, until correct medicine is found that works for you.

But regarding this topic, all i see, is you complaining, without helping OP.

 

[SgtWireless]

im gonna have to take some time to understand this software, ive used similar stuff before, but now i have to learn how to interpret all this information, this may take some days. im still hoping that it was a thermal issue and that fixing my GPU fan will make the issue go away, but i know better than that.

 

[Aeacus]

ive used similar stuff before, but now i have to learn how to interpret all this information, this may take some days.

Initial guide here (dismiss the cryptomining talk in there);

View: https://www.youtube.com/watch?v=ZJYsiKh1FcI


Full stress test the PC with HWinfo64 logging guide here,
link: https://rtech.support/docs/guides/hwinfo.html

 

[SgtWireless]

ive figured out on my own how to start logging with this, will come back to this post with anything if it happens again. that guide was talking nonstop about mining, didnt actually give very good information, i think i knew all this already. this software is markedly similar to something else now that ive taken a longer look at it.