CPU overheating or PSU failure?

[flloutboy12]

I believe my CPU is overheating (50°C while 20% use and 40° while idle, it reboots after 5 min of gaming (Superhot)) and causing reboots, although, I cannot find why is it overheating.
I already changed the thermal paste (Wich I believe I applied correctly).
The stock fan runs just like when I bought the PC.
The case fan also works normally.(Generic one)
Note: when I touch the power supply through the case I can feel it sligtly hot.

Intel Pentium Dual Core G3220
Nvidia GT 640
8gb Hyperx
Motherboard MSI H81M-P33
Generic PSU 500w

 

[teslacoilftw]

The Thermal Limit for this processor is above 72 Celsius ... your computer shouldn't be rebooting..
"If" you feel like this is a cooling issue there are plenty of 3rd party Coolers you can.

How much paste are you using?
There's a bios setting that changes how your computer regulates the Fan when it comes to cooling and triggers a shutdown.
Normally Heat Triggers have a sound.

You might try downloading the MSI Command Center and see if you can tweak the fans from Windows to be turned up to cool down Quicker...

You might check your Windows Event Viewer...
Screen shot it and post your criticals..
Administrative Tools
Event Viewer.
Middle Screen says "overview and Summary..
+Critical
+Errors...

Screen shot your Criticals.. this will show you what's happening just before it reboots.

 

[flloutboy12]

The Thermal Limit for this processor is above 72 Celsius ... your computer shouldn't be rebooting..
"If" you feel like this is a cooling issue there are plenty of 3rd party Coolers you can.

How much paste are you using?
There's a bios setting that changes how your computer regulates the Fan when it comes to cooling and triggers a shutdown.
Normally Heat Triggers have a sound.

You might try downloading the MSI Command Center and see if you can tweak the fans from Windows to be turned up to cool down Quicker...

You might check your Windows Event Viewer...
Screen shot it and post your criticals..
Administrative Tools
Event Viewer.
Middle Screen says "overview and Summary..
+Critical
+Errors...

Screen shot your Criticals.. this will show you what's happening just before it reboots.

Sorry about the Spanish.
Aplicación: Aplication
Sistema: System
Critico: Critical

About the paste: I googled how much to use (videos) I think I used enough, neither more neither less.

 

[SirSub42]

+1. Your issue may not be a heating issue at all. After viewing your event viewer, you can alter the default windows settings that automatically force a reboot when an error has occurred.

 - In Windows, search for and open View advanced system settings.
 - In the Advanced tab, under Startup and Recovery, click Settings.
 - Under System Failure, remove the check mark next to Automatically restart, then click OK.

The safety mechanisms in the BIOS that force a shut-off if the CPU reaches a certain temperature are still in play, so there will be no danger there.

 

[SirSub42]

Ok. So an Error 41 means the computer shut down unexpectedly. Most likely due to a hardware issue. I would start by resetting your BIOS settings to default. I have not exactly heard of a power supply over heating, but if it is being overloaded, it may cause a shut-down. What other components are you running ? (HD, CD drives, lights etc).

Also, can you install this program CPUID HardwareMonitor and give us a screen shot.

 

[lumineZ]

Does it only reboot when you put the system under load or does it also happen if you stay in desktop just idle or light web browsing?

 

[flloutboy12]

Ok. So an Error 41 means the computer shut down unexpectedly. Most likely due to a hardware issue. I would start by resetting your BIOS settings to default. I have not exactly heard of a power supply over heating, but if it is being overloaded, it may cause a shut-down. What other components are you running ? (HD, CD drives, lights etc).

Also, can you install this program CPUID HardwareMonitor and give us a screen shot.

Just a Western Digital Blue 1 TB Hardrive and a case fan.

 

[flloutboy12]

Does it only reboot when you put the system under load or does it also happen if you stay in desktop just idle or light web browsing?

Actually, while i´m web browsing (maybe 2 or 3 windows opened) it reboots.

 

[flloutboy12]

+1. Your issue may not be a heating issue at all. After viewing your event viewer, you can alter the default windows settings that automatically force a reboot when an error has occurred.

 - In Windows, search for and open View advanced system settings.
 - In the Advanced tab, under Startup and Recovery, click Settings.
 - Under System Failure, remove the check mark next to Automatically restart, then click OK.

The safety mechanisms in the BIOS that force a shut-off if the CPU reaches a certain temperature are still in play, so there will be no danger there.

I did this just in case btw

Edit: it just rebooted.

 

[lumineZ]

You dont have a temprature problem that I can tell you.

You say you have an "Generic PSU 500w"... can you give us a part nr. or the name of the PSU?

I am thinking of two things. PSU or MOBO failure at this point.

 

[SirSub42]

That solution was more for system errors that sometimes force a reboot. But given the code 41 in the kernel, we eliminated that as a possible cause. All of your temperatures seem within reason. The temperatures all seem within normal operating conditions. Here are a few other options to try out:

- Disconnect any extra devices in your tower (CD drives, 2nd hard drives etc, anything you don't need and can test without)
- Check the BIOS settings regarding temperature thresh holds that force a reboot

On Power Options window, find and expand Hard disk > Turn off hard disk after, then set it to be Never.
Proceed to find and expand Sleep > Sleep after, then set it to be Never as well.
After it’s done, click OK to save your settings.

You mention it reboots after 5 minutes of gaming. How long will it stay on if it sits idle?

 

[flloutboy12]

You dont have a temprature problem that I can tell you.

You say you have an "Generic PSU 500w"... can you give us a part nr. or the name of the PSU?

I am thinking of two things. PSU or MOBO failure at this point.

Sorry the delay, I had to open the case.

 

[flloutboy12]

That solution was more for system errors that sometimes force a reboot. But given the code 41 in the kernel, we eliminated that as a possible cause. All of your temperatures seem within reason. The temperatures all seem within normal operating conditions. Here are a few other options to try out:

- Disconnect any extra devices in your tower (CD drives, 2nd hard drives etc, anything you don't need and can test without)
- Check the BIOS settings regarding temperature thresh holds that force a reboot

On Power Options window, find and expand Hard disk > Turn off hard disk after, then set it to be Never.
Proceed to find and expand Sleep > Sleep after, then set it to be Never as well.
After it’s done, click OK to save your settings.

You mention it reboots after 5 minutes of gaming. How long will it stay on if it sits idle?

The hard drive was set to sleep after 20 mins, fixed it.
Around 15/20 mins while idling, then reboots.

 

[SirSub42]

Well, I was hoping for some clear cut solution, but here are some next steps we can take.
- Bootable Stress Testing
If Success
- I have read that an incorrect driver could be causing these problems. As the driver instructs the hardware, a bad driver could cause a hardware failure. Double check the device manager and check for any errors. (The posts I read, specifically mentioned the audio driver, though I suppose it could be any).

Testing your PSU is a bit more difficult unless you have a PSU tester or a voltmeter handy. But check back in after the stress testing. That will answer some important questions.

 

[CompuTronix]

The Thermal Limit for this processor is above 72 Celsius ...

teslacoilftw,

Respectfully, and for everyone's benefit, almost everyone gets blindsided by Tcase, since few users understand what it actually means.

There are two Thermal Specifications; Tcase and Tj Max. The specification to which you've referred is Tcase, which is much more useful for developers of cooling solutions; but for users, Tcase is a misleading and confusing specification.

Tcase appears on Intel's Product Specifications website and in the Datasheets, but Tj Max appears only in the Datasheets. Intel no longer shows Tcase on their website since 7th Generation, but instead now shows Tj Max. Both specifications are still shown in the Datasheets.

Intel Product Specifications, Pentium G3220 - https://ark.intel.com/products/77773/Intel-Pentium-Processor-G3220-3M-Cache-3_00-GHz

There are 4 very crucial points that Intel does not openly disclose concerning their Tcase specification:

(1) Tcase is a factory only temperature measurement on the external surface of the Integrated Heat Spreader (IHS), which users have no way of measuring.

(2) Tcase is not Core temperature.

(3) Tcase varies depending on which cooler is being used.

(4) Tcase is not the limiting specification; Tj Max is.

See Page 76, Table 27, 6th Row, 4th Column: TCC Activation Temperature, Desktop 4th Gen Intel® Core™ Processors Datasheet, Vol. 1 - https://www.intel.com/content/www/us/en/processors/core/4th-gen-core-family-desktop-vol-1-datasheet.html

Tj Max (Throttle temperature) is 100°C.

flloutboy12,

Your screenshot shows your CPU is idling at 2% load, and the maximum Core temperature was 54°C ... but at what load?

Q: What is the maximum Core temperature at 100% load?
Q: What is your Ambient (room) temperature?

Here's the operating range for Core temperature:

Core temperatures above 85°C aren't recommended.

Core temperatures increase and decrease with Ambient temperature.

Download Prime95 version 26.6 and run Small FFT's for just 5 minutes: Prime95 v26.6 - http://www.mersenneforum.org/showthread.php?t=15504

Unlike later versions, 26.6 Small FFT's is a steady 100% workload. Since Intel thermal tests their processors at steady 100% workload, if your stock cooler is working properly, your G3220 should not reach Throttle temperature.

The #1 cause for overheating and throttling on stock coolers is a push-pin that's popped loose from the motherboard, which may have been the original problem. Since you've reseated the cooler, the same push-pin might still be giving you trouble. It can be deceivingly tricky to get them properly latched completely through the motherboard.

You might want to very carefully double check each corner of the cooler and push-pins using a strong light. If your case allows you to see the back of the motherboard's CPU socket, make sure that all pins extend completely and equally through the motherboard.

You can also put a heavy workload on the processor, then while watching the Core temperatures, firmly press down on each corner of the cooler, holding pressure against the motherboard for about 45 seconds for each corner. If you see the Core temperatures drop, then you've found the loose push-pin.

It's a basic troubleshooting step that I recommend you try first, prior to looking elsewhere.

EDIT: Concerning your power supply, since your case is configured for a top-mount PSU, it will "feel" a bit warm due to it's location, as heat naturally rises to the top of the case. A bottom-mount PSU will of course "feel" cooler.

The fan in some PSU's won't spin up until a certain temperature is reached, while in most "generic" PSU's the fan spins constantly. Since generic PSU's use inferior fans with sleeve bearings, they tend to fail at lower runtime hours than high-end PSU's with quality fans.

Q: Does your PSU fan spin up OK?

CT

 

[SirSub42]

So you're good to go? Top notch response btw Compu. But you stole my 50th solution. I got de-badged