[SOLVED] General build-up of heat in system causes immediate shutdown ?

[Redshade6]

My PC has recently started to shutdown unexpectedly. Today I discovered this is because of an overall overheating issue within the system. By shutdown I mean total loss of power. My guess is that one of the components is triggering an overheat, causing the system to shut itself off. The weird thing is that there is really no need for it to do that.


After doing days and days of debugging, I can state the following facts:

• There is an overheating issue. I know this because I sometimes restart the pc immediately after the shutdown, and it will shutdown again when loading windows. It will keep doing this unless I give it a few minutes to cool down, after which the pc works fine again.
• It's not the major components that are overheating. I run OCCT's "Power" stress test, and I can see that all the temps are within bounds of the component's max temps. See the table below
• Windows Event Viewer reports a critical event after each shutdown. The event ID is 41, and there are no BugcheckParameters.
• Even after cleaning the PC of dust, the same issue occurs.
• Resetting the BIOS doesn't help.

Component	Model	General max temp in C
CPU	Ryzen 5 2600	85
GPU	GTX 1660 super	75 (Hot spot 85)
RAM	Corsair Vengeance 2x8GB @ 3000mhz	?
SSD	Samsung 860 EVO 1TB	TBA
HDD	Some old Hitachi 1TB drive	?
PSU	Cooler Master MWE Gold 750w	?
Motherboard	AsRock B450M Pro4	41

None of the above temperatures seem too high.


Special dates and events:

12 september:
- I reinstalled windows 10, and I'm now on version 20H2.

24 september:

• I decided to remove 1 of the monitors of my setup.
• Immediately after this, the shutdowns started.

How the removal of a monitor can start causing these overheating issues, I don't know.
I am at my wits end. My knowledge of PC's is vast, but this problem really goes over my head. I think there's a small chip on the motherboard overheating, causing this shutdown, but I wouldn't know how to detect that.

Thank you to anyone who wishes to help me out!

 

[RodroX]

I did not apply new thermal paste, because the old stuff wasn't dried up.

I just let it sit on desktop for 45 minutes. The average temperature is 34C.

I would get Cinebench R20, set it to run for 600/900 seconds loop (10 to 15 mins) and see what temp you get there (https://www.guru3d.com/files-details/download-maxon-cinebench.html). I would also get hwinfo64 instead of hwmonitor.

And yeah I would re-apply new thermal paste.

As for the tower cooler, is never a bad idea if you use your CPU for more than simple office tasks (gaming, rendering, streaming, or a mix of those.)

 

[RodroX]

This are also good alternatives:

https://uk.pcpartpicker.com/product...ck-edition-420-cfm-cpu-cooler-rr-212s-20pk-r1

https://uk.pcpartpicker.com/product/PnPKHx/arctic-freezer-34-esports-duo-cpu-cooler-acfre00061a

Remember to make sure it will fit inside your case and it wont have conflicts with your mobo/RAM.

 

[Redshade6]

I would get Cinebench R20, set it to run for 600/900 seconds loop (10 to 15 mins) and see what temp you get there (https://www.guru3d.com/files-details/download-maxon-cinebench.html). I would also get hwinfo64 instead of hwmonitor.

And yeah I would re-apply new thermal paste.

As for the tower cooler, is never a bad idea if you use your CPU for more than simple office tasks (gaming, rendering, streaming, or a mix of those.)

I looped Cinebench R20 for 750 seconds and did that twice. CPU temp was about 81C (used HWinfo64), and it didn't crash at all.

Gonna put new thermal paste on tonight.

 

[RodroX]

I looped Cinebench R20 for 750 seconds and did that twice. CPU temp was about 81C (used HWinfo64), and it didn't crash at all.

Gonna put new thermal paste on tonight.

This is getting more and more strange. So you were able to run 2x 750 seconds loop of CR20 without incidents... Intersting

Well, really hope the new thermal paste helps lower temps s little bit. And please let us know if and when it crash again.

 

[Redshade6]

This is getting more and more strange. So you were able to run 2x 750 seconds loop of CR20 without incidents... Intersting

Well, really hope the new thermal paste helps lower temps s little bit. And please let us know if and when it crash again.

God damn...

So I first wanted to reapply thermal paste. I was pulling on the cooler, and it decided to yank out the CPU with it. Luckily no damage to the CPU. The thermal paste was holding the two together better than cement. I was eventually able to seperate the two with a heatgun set to low.

I then proceeded to clean the setup using Arctic Clean, after which I applied CM Master Gel v2. This reduced the temperature by a few degrees, which is always handy. It then crashed again.

After this happened, I sighed and began replacing the motherboard with a brand new identical one. Finally, I thought, this must be the solution to my problems.

I started my usual stress test, and after a bit it crashed. It is showing the exact same symptoms as before.

Honestly, I don't know what to do anymore. I guess I can try fixing the airflow, but it shouldn't matter with a clean cooler and fresh thermal paste. The CPU never went above 81C.

Guess I'm gonna be returning the new motherboard... Yay.

 

[RodroX]

God damn...

So I first wanted to reapply thermal paste. I was pulling on the cooler, and it decided to yank out the CPU with it. Luckily no damage to the CPU. The thermal paste was holding the two together better than cement. I was eventually able to seperate the two with a heatgun set to low.

I then proceeded to clean the setup using Arctic Clean, after which I applied CM Master Gel v2. This reduced the temperature by a few degrees, which is always handy. It then crashed again.

After this happened, I sighed and began replacing the motherboard with a brand new identical one. Finally, I thought, this must be the solution to my problems.

I started my usual stress test, and after a bit it crashed. It is showing the exact same symptoms as before.

Honestly, I don't know what to do anymore. I guess I can try fixing the airflow, but it shouldn't matter with a clean cooler and fresh thermal paste. The CPU never went above 81C.

Guess I'm gonna be returning the new motherboard... Yay.

Darn, I do not know if you want to keep trying things this is what I can tell you:

Go into the BIOS, load optimized/default settings + Save and Exit. Turn off the PC, remove the power cord from the power wall socket, remove the cmos battery and press the power button a few times (and you can keep it press for 30 seconds) to drain any residual energy.
Now open the case and: remove one stick of RAM (leave only 1), unplug all the case fans, and any USB / Audio front headers, unplug data and power cablesfrom the HDD (Im guessing the OS is in the SSD right?).
After that, while keeping the least amount of components, put the cmos battery in, plug the power cable back and turn ON the PC and test again if it crash.

The idea is to add components, cables, and outside cables (usb/audio headers) little by little, trying to figure out which one may be the culprit.


Other things to do:

Load optimized/default settings in BIOS + Save and Exit. Update BIOS and Chipset drivers (be carefull theres always some steps you need to follow to update BIOS if its too old, so read the BIOS page!). Chipset can be donwloaded from here: https://www.amd.com/en/support/chipsets/amd-socket-am4/b450. Pay atention to donwload the BIOS for your current mobo, sometimes there are different versions.

If everything fail, I would then build the PC outside the case over a wooden table or the motherboard cardboard box.


Please be carefull, and if you are not sure about doing this take the PC to some technician.

 

[Redshade6]

Darn, I do not know if you want to keep trying things this is what I can tell you:

Go into the BIOS, load optimized/default settings + Save and Exit. Turn off the PC, remove the power cord from the power wall socket, remove the cmos battery and press the power button a few times (and you can keep it press for 30 seconds) to drain any residual energy.
Now open the case and: remove one stick of RAM (leave only 1), unplug all the case fans, and any USB / Audio front headers, unplug data and power cablesfrom the HDD (Im guessing the OS is in the SSD right?).
After that, while keeping the least amount of components, put the cmos battery in, plug the power cable back and turn ON the PC and test again if it crash.

The idea is to add components, cables, and outside cables (usb/audio headers) little by little, trying to figure out which one may be the culprit.


Other things to do:

Load optimized/default settings in BIOS + Save and Exit. Update BIOS and Chipset drivers (be carefull theres always some steps you need to follow to update BIOS if its too old, so read the BIOS page!). Chipset can be donwloaded from here: https://www.amd.com/en/support/chipsets/amd-socket-am4/b450. Pay atention to donwload the BIOS for your current mobo, sometimes there are different versions.

If everything fail, I would then build the PC outside the case over a wooden table or the motherboard cardboard box.


Please be carefull, and if you are not sure about doing this take the PC to some technician.

Just did a longer run of Cinebench than before, and it crashed. I can confirm the problem lies with either the CPU or the PSU overheating.

I'm taking the system to a repairshop this week, to have them measure which one is bad. Thanks for the help, and I will keep this thread updated as I figure out more.

 

[Karadjgne]

I would love to read Karadjgne opinion on this but (airflow wise) I would try this:

Oh, lol. Ok.

Let me start with a couple of observations. First, Op, You can do better, much better. That's a pretty dirty pc. Its doing you and it no favors. Secondly, cable management. its a rats nest in there, which doesn't affect Air, but does affect air Flow.

From what I see you have fans fighting. Front intake on that case is pretty miserable, and what air is being dumped into the case is being channeled out by the drive bays, but then runs into the side fans output, which is a mess and not an even flow, since it has to wind its way through all that cabling. The fan on the psu is on Top, not underneath, so airflow in that is compromised by the top venting and being outside air fed, isn't being used by the pc as exhaust. It's self contained.

What I would do is take the pc cabling apart, almost if not fully, and start from scratch. After a very good cleaning. With that in mind, route cables neatly where they'd not be obstructions to airflow. IF (like my old CM 690 II Advanced) you can flip the case sides and put a solid side facing you and the vented side panel behind the mobo, I'd do so. What that side does is allow the rear exhaust and cpu cooler easier access to fresher air, that's not filtered, but more importantly, not the intake air. Enclosing the side will force the rear exhaust to pull air from the front of the case which then creates airflow. When air flows, heat moves. Right now, the heat in the case isn't really moving much, its supplanted by air directly from that big gaping hole in the side going almost straight through.

Turn the psu so the fan is facing downwards. Its free exhaust and won't affect the operation of the psu at all.

Move that front intake to the back side so you'll have both fans there as intake in the same area, adding to the force of what comes in instead of bumping into each other .

The above temps (photo) were from HW Monitor, not HW Info, two different programs. With HWM you get a bunch of gibberish, what TMPIN really are is totally unknown, you can't tell if that's a sata controller at 80+ (totally normal) or your SSD/HDD at 80+ (totally disastrous).

There's nothing intrinsically unhealthy about the stock coolers, AMD (CoolerMaster) has done a fantastic job there, considering the old FX coolers or even Intel's best (feeble) attempts. Before spending money on a Band-Aid, fix the bones of the pc so that any cooler upgrade actually stands a chance of showing a worthwhile improvement. The best coolers in the world are absolutely useless if they cannot get enough air to do the job.

View: https://youtu.be/2GXNgPSBq8g


Surprise, surprise.

Any decent tower cooler is going to be a bonus. Not only to cpu temps, but because the fan(s) pull air from the front area, actually helps with airflow too.

 

[RodroX]

Oh, lol. Ok.

Let me start with a couple of observations. First, Op, You can do better, much better. That's a pretty dirty pc. Its doing you and it no favors. Secondly, cable management. its a rats nest in there, which doesn't affect Air, but does affect air Flow.

From what I see you have fans fighting. Front intake on that case is pretty miserable, and what air is being dumped into the case is being channeled out by the drive bays, but then runs into the side fans output, which is a mess and not an even flow, since it has to wind its way through all that cabling. The fan on the psu is on Top, not underneath, so airflow in that is compromised by the top venting and being outside air fed, isn't being used by the pc as exhaust. It's self contained.

What I would do is take the pc cabling apart, almost if not fully, and start from scratch. After a very good cleaning. With that in mind, route cables neatly where they'd not be obstructions to airflow. IF (like my old CM 690 II Advanced) you can flip the case sides and put a solid side facing you and the vented side panel behind the mobo, I'd do so. What that side does is allow the rear exhaust and cpu cooler easier access to fresher air, that's not filtered, but more importantly, not the intake air. Enclosing the side will force the rear exhaust to pull air from the front of the case which then creates airflow. When air flows, heat moves. Right now, the heat in the case isn't really moving much, its supplanted by air directly from that big gaping hole in the side going almost straight through.

Turn the psu so the fan is facing downwards. Its free exhaust and won't affect the operation of the psu at all.

Move that front intake to the back side so you'll have both fans there as intake in the same area, adding to the force of what comes in instead of bumping into each other .

The above temps (photo) were from HW Monitor, not HW Info, two different programs. With HWM you get a bunch of gibberish, what TMPIN really are is totally unknown, you can't tell if that's a sata controller at 80+ (totally normal) or your SSD/HDD at 80+ (totally disastrous).

There's nothing intrinsically unhealthy about the stock coolers, AMD (CoolerMaster) has done a fantastic job there, considering the old FX coolers or even Intel's best (feeble) attempts. Before spending money on a Band-Aid, fix the bones of the pc so that any cooler upgrade actually stands a chance of showing a worthwhile improvement. The best coolers in the world are absolutely useless if they cannot get enough air to do the job.

View: https://youtu.be/2GXNgPSBq8g


Surprise, surprise.

Yeap, thats what I wanted hwinfo too, it always try to show the right "labels/text".

Just did a longer run of Cinebench than before, and it crashed. I can confirm the problem lies with either the CPU or the PSU overheating.

I'm taking the system to a repairshop this week, to have them measure which one is bad. Thanks for the help, and I will keep this thread updated as I figure out more.

OK, I guess a fresh look may help. Hope you get it fix it. Too bad I couldn't nail it from here. Sometimes gets hard to diagnose something you can't see / hear / touch.

Anyways, I will still try to fix cable management and move the fans so they don't collide to each other.

Cheers

 

[Redshade6]

Oh, lol. Ok.

Let me start with a couple of observations. First, Op, You can do better, much better. That's a pretty dirty pc. Its doing you and it no favors. Secondly, cable management. its a rats nest in there, which doesn't affect Air, but does affect air Flow.

From what I see you have fans fighting. Front intake on that case is pretty miserable, and what air is being dumped into the case is being channeled out by the drive bays, but then runs into the side fans output, which is a mess and not an even flow, since it has to wind its way through all that cabling. The fan on the psu is on Top, not underneath, so airflow in that is compromised by the top venting and being outside air fed, isn't being used by the pc as exhaust. It's self contained.

What I would do is take the pc cabling apart, almost if not fully, and start from scratch. After a very good cleaning. With that in mind, route cables neatly where they'd not be obstructions to airflow. IF (like my old CM 690 II Advanced) you can flip the case sides and put a solid side facing you and the vented side panel behind the mobo, I'd do so. What that side does is allow the rear exhaust and cpu cooler easier access to fresher air, that's not filtered, but more importantly, not the intake air. Enclosing the side will force the rear exhaust to pull air from the front of the case which then creates airflow. When air flows, heat moves. Right now, the heat in the case isn't really moving much, its supplanted by air directly from that big gaping hole in the side going almost straight through.

Turn the psu so the fan is facing downwards. Its free exhaust and won't affect the operation of the psu at all.

Move that front intake to the back side so you'll have both fans there as intake in the same area, adding to the force of what comes in instead of bumping into each other .

The above temps (photo) were from HW Monitor, not HW Info, two different programs. With HWM you get a bunch of gibberish, what TMPIN really are is totally unknown, you can't tell if that's a sata controller at 80+ (totally normal) or your SSD/HDD at 80+ (totally disastrous).

There's nothing intrinsically unhealthy about the stock coolers, AMD (CoolerMaster) has done a fantastic job there, considering the old FX coolers or even Intel's best (feeble) attempts. Before spending money on a Band-Aid, fix the bones of the pc so that any cooler upgrade actually stands a chance of showing a worthwhile improvement. The best coolers in the world are absolutely useless if they cannot get enough air to do the job.

View: https://youtu.be/2GXNgPSBq8g


Surprise, surprise.

Any decent tower cooler is going to be a bonus. Not only to cpu temps, but because the fan(s) pull air from the front area, actually helps with airflow too.

Many thanks for your advice. My PC has been long overdue for some love, so I will apply all your tips this week.

I have actually already switched from HWMonitor to HWInfo64, for that same reason. It shows better temperatures, for the right things.

However, I do know the PSU has its fan pointed downwards, so that doesn't need to be changed, right?

The PC is receiving a thorough cleaning no matter what happens this week.

 

[Karadjgne]

When you say it's pointed down, that means if you look inside the pc you can see the fan on the bottom? That's good. To my view, it looked as though the fan was on the top, pulling outside air in.

 

[FlatToes]

I'm noticing that if I use only 1 of the 2 RAM sticks I have, the CPU goes up to 89C and the system lasts a bit longer than normal. I'm sure this is just a coincidence, but that's what I found out. I alternated the sticks and nothing changed.

Oddly I made an account just because this piqued my interest. So I'll toss in my $0.25

If only using one stick makes it last longer, it very well could be the bridge chipset. Since the bridge is only using half-channel mode, in turn generating less heat. Make sure that the chipset cooler is properly attached and not loose, it's located directly beside the ram slots, stamped with the brand logo. If everything looks good, then it's the mobo.

 

[LadyGamer955]

This is the stock cooler, but it only gets this hot during the OCCT test. Something like Discord and Hearthstone at the same time will also kill my pc.

The setup has been running without any hiccups for the past 2 years. All the components are about that old, and were bought new.

You could try running a little freebie program called Open Hardware Monitor. It monitors the temp of everything.