PC going through thermal shutdown but temps are nowhere near the thermal shutdown limits

[prot7587]

Hello, this is my first post, so excuse the mistakes.

My computer is going through processor thermal trip apparently but the temps are near 45-60 degrees on load, I tried changing power options in the settings to balanced,power saver and high performance. High performance used to crash as soon as I clicked it before but now it stays on for 3-4 minutes before shutting down, balanced on 80% max processor limit stays on for about half an hour without load and with load it goes shuts down within 10 minutes and power saver stays on for quite a while but it's very laggy. The temps are around 44-50 on all cores in about 70% load and the processor is an Q6600. The computer just crashes out of the blue and for no reason at all. I have a 500W PSU as well, and a new aftermarket fan with thermal compound installed and tried with case open and adding an extra case fan but nothing works.The motherboard is Intel DG31PR.

Earlier, the computer was also shutting down but it was another error it was PXE ROM failure I believe and I fixed it through turning off booting in network cards in the BIOS. The thermal shutdown problem still existed back then but the main reason was the PXE ROM one but after disabling that one my PC shuts down after every 3-5 minutes because of the thermal trip.

Any help would be appreciated. Thank you.

 

[jay32267]

Perhaps it's not "thermal shutdown".

Perhaps a voltage is going out of spec when you pull more current from the PSU under load.

I would be checking PSU voltages with a voltmeter or software like HWInfo or I'd swap in another PSU.

 

[gasaraki]

Probably the PSU is going bad.

 

[prot7587]

Perhaps it's not "thermal shutdown".

Perhaps a voltage is going out of spec when you pull more current from the PSU under load.

I would be checking PSU voltages with a voltmeter or software like HWInfo or I'd swap in another PSU.

I'll look into the PSU matter but another thing was this problem was also happening a couple of months ago and the pc would shut down randomly the guy I took it to fix told that it was some sort of a bent capacitors issue or something, so I don't know if I have to change both the motherboard and the PSU? How will I identify that the bent capacitors isn't the issue?

 

[prot7587]

Probably the PSU is going bad.

I'll look into the PSU matter but another thing was this problem was also happening a couple of months ago and the pc would shut down randomly the guy I took it to fix told that it was some sort of a bent capacitors issue or something, so I don't know if I have to change both the motherboard and the PSU? How will I identify that the bent capacitors isn't the issue?

 

[jay32267]

Bent caps aren't going to cause any issue unless they are bent into something and touching.....and even then...only the leads would matter.

I would look around and see if I could see what he was talking about.....but it sounds strange.

 

[prot7587]

Bent caps aren't going to cause any issue unless they are bent into something and touching.....and even then...only the leads would matter.

I would look around and see if I could see what he was talking about.....but it sounds strange.

I don't know is PSU the case? Because the PC sometimes is shutting down at the windows GUI Boot screen right after the bios one and is also showing different sorts of boot drive errors as well..

 

[Karadjgne]

Post some of the errors. Look in Windows Event Viewer, you'll have a bunch of red-flagged critical errors. These will be what's causing the shutdowns.

Some can be fruitless, like ntkernal errors, but you'll prolly find just as many that'll actually point to a specific reason.

 

[prot7587]

Post some of the errors. Look in Windows Event Viewer, you'll have a bunch of red-flagged critical errors. These will be what's causing the shutdowns.

Some can be fruitless, like ntkernal errors, but you'll prolly find just as many that'll actually point to a specific reason.

PC is not even booting right now, it shows "Processor Thermal Trip" and it shuts down right after the message.

 

[kerberos_20]

start PC
once it shutdowns, touch with your hand heatsink on cpu
if it feels really hot, then its overheating
do some vacuum cleaning of the fan and the sink, apply paste

 

[prot7587]

start PC
once it shutdowns, touch with your hand heatsink on cpu
if it feels really hot, then its overheating
do some vacuum cleaning of the fan and the sink, apply paste

It's not hot at all, it's never hot, my motherboard is more hotter than the CPU according to BIOS, the motherboard is like 34-35 degrees while cpu is always at 24-28 degrees when this loop of thermal trip occurs.

 

[kerberos_20]

so u didnt touch it

 

[prot7587]

so u didnt touch it

I did multiple times it's not hot.

 

[kerberos_20]

then you have an airgap between the bottom of the heatsink and the processor heatspreade or you have used either too much of too little thermal compound

 

[prot7587]

then you have an airgap between the bottom of the heatsink and the processor heatspreade or you have used either too much of too little thermal compound

Don't think fan installations the issue since at 50-60% load when the PC boots up it is around 50-55 degrees and near a 100% load around 65 degrees, the TJ max is 100 degrees so it should not set of any processor alerts for damage.

 

[Karadjgne]

Is that every core individually? Or just the single cpu temp. If you have air/paste issues above 1 core it'll hit max but the others will just read normally or slightly elevated. Do your load test but monitor all core temps instead. If you are seeing a spread of more than @10°C from lowest to highest, there's an issue with either your paste on the cooler, or the Tim inside the cpu.

 

[prot7587]

Is that every core individually? Or just the single cpu temp. If you have air/paste issues above 1 core it'll hit max but the others will just read normally or slightly elevated. Do your load test but monitor all core temps instead. If you are seeing a spread of more than @10°C from lowest to highest, there's an issue with either your paste on the cooler, or the Tim inside the cpu.

Here is a picture of event viewer:
http://prntscr.com/kug5l8

I hope it's the right one, I'm pretty nooby with these sort of stuff and here is a screenshot with core temp of four cores. I changed the power options to power saver and now I notice it's running on three cores?

Core temp screenshot:
http://prntscr.com/kug6ik

Also, the loads fluctuate between 50-60% to 2-3% and the temps remain the same no spikes.

I tried taking 2 gigs of RAM stick out of my PC and somehow I was able to boot into windows this time unlike going through auto shutdown after BIOS screen? What's wrong, I'm confused af.

 

[Karadjgne]

The top 2 events are a network error where windows can't verify its a registered copy.

The rest of those event 65's are where you have programs that use registered certificates and not being validated.

Neither one should result in a crash. So windows is crashing with nothing pointing towards a crash.

There are ways to fix both those events, but do require some technical expertise as they deal with the registry.

These issues have seen major increase since the last few major windows updates. I'd make sure of 3 things. 1) Absolute latest bios revision. 2) Absolute latest motherboard drivers, especially LAN, audio, USB family, Sata etc. 3) open up a CMD window with Administrator permissions type sfc/ scannow . This should fix any system files that might be corrupted.

Yes, running only 3 temps is odd, as is that cpu running 1.6GHz, unless that was done during an idle period. Normally a cpu is good for over 20 years if temps are good for the duration, but OC can significantly shorten that if the cpu is subjected to near max voltages. With just 3 reported temps, I'd start maybe looking into checking to make sure that your cpu hasn't cooked a core and is yelling obscenities at you by crashing..

 

[larrycumming]

Dude, is that windows 10 on an LGA775 socket t mobo?

If so try going back to an older os or try it with a Linux distro, windows 10 doesn’t play nice with some of the older hardware due to driver compatibility.

 

[prot7587]

The top 2 events are a network error where windows can't verify its a registered copy.

The rest of those event 65's are where you have programs that use registered certificates and not being validated.

Neither one should result in a crash. So windows is crashing with nothing pointing towards a crash.

There are ways to fix both those events, but do require some technical expertise as they deal with the registry.

These issues have seen major increase since the last few major windows updates. I'd make sure of 3 things. 1) Absolute latest bios revision. 2) Absolute latest motherboard drivers, especially LAN, audio, USB family, Sata etc. 3) open up a CMD window with Administrator permissions type sfc/ scannow . This should fix any system files that might be corrupted.

Yes, running only 3 temps is odd, as is that cpu running 1.6GHz, unless that was done during an idle period. Normally a cpu is good for over 20 years if temps are good for the duration, but OC can significantly shorten that if the cpu is subjected to near max voltages. With just 3 reported temps, I'd start maybe looking into checking to make sure that your cpu hasn't cooked a core and is yelling obscenities at you by crashing..

It's running at 1.6 ghz because I capped it at the power settings , otherwise my PC will crash after 30-40 minutes of use and after crashing the first time randomly crash after 5-10 minutes or even less, same "Processor thermal Trip" error every damn time but with the cap at 70% of the processors max usage the cpu can run for a few hours and not face any crash, sacrificing some performance that is, maybe the 3 cores are because of that?

And, the errors I can fix but I'm not sure about how to do a BIOS update and when I check the drivers through device manager they always say "The best drivers for your device are installed." Should I try going back to Win 7? Or would that be of no good? And the CPU was never overclocked btw.

 

[Karadjgne]

To check the drivers, look in Device Manager, select a component/driver and under properties it'll list the driver version and date. Compare that to the drivers listed on the motherboards support page.

Windows almost always says it has the latest drivers since windows itself has generic versions included which are newer and upto date, the downside being they are generic and not specific to your mobo. Meaning they'll work, just not to the best ability.

Bios upgrade isn't hard. Just 2 rules. 1) follow the instructions exactly. 2) do not deviate, don't do something stupid like turning off the pc or pulling power cuz you assume there's a problem. If in doubt refer to rule #1.

If you download and save everything to desktop, keeping it all in a group off to one side, makes life easy to deal with. Many of the drivers will have self executing updaters.

 

[prot7587]

To check the drivers, look in Device Manager, select a component/driver and under properties it'll list the driver version and date. Compare that to the drivers listed on the motherboards support page.

Windows almost always says it has the latest drivers since windows itself has generic versions included which are newer and upto date, the downside being they are generic and not specific to your mobo. Meaning they'll work, just not to the best ability.

Bios upgrade isn't hard. Just 2 rules. 1) follow the instructions exactly. 2) do not deviate, don't do something stupid like turning off the pc or pulling power cuz you assume there's a problem. If in doubt refer to rule #1.

If you download and save everything to desktop, keeping it all in a group off to one side, makes life easy to deal with. Many of the drivers will have self executing updaters.

I'll try and post a reply, thank you.

 

[prot7587]

To check the drivers, look in Device Manager, select a component/driver and under properties it'll list the driver version and date. Compare that to the drivers listed on the motherboards support page.

Windows almost always says it has the latest drivers since windows itself has generic versions included which are newer and upto date, the downside being they are generic and not specific to your mobo. Meaning they'll work, just not to the best ability.

Bios upgrade isn't hard. Just 2 rules. 1) follow the instructions exactly. 2) do not deviate, don't do something stupid like turning off the pc or pulling power cuz you assume there's a problem. If in doubt refer to rule #1.

If you download and save everything to desktop, keeping it all in a group off to one side, makes life easy to deal with. Many of the drivers will have self executing updaters.

PC doesn't even boot up properly and keeps hitting the infinite loop of processor thermal trips unless I take a 2 gig ram out of my PC and restart but the problem again pops up when I put it back on so is that the problem?

 

[Karadjgne]

Could be. Check that stick individually.

 

[prot7587]

Could be. Check that stick individually.

Both stick seems to work on both sockets individually but when combined crashes again... why is this happening? But when I use one stick it doesn't cause any crash/error shutdown in my half an hour to an hour of surfing the web or mild gaming since 2 gig ram cant run crap.

Edit: It is sometimes causing memory management issues while browsing in chrome when ram hits 1.7 gigabytes of 2 gigabytes limit or when I stress it.