Complete freeze shortly after startup, and sudden shutdown

[ToxicXGL]

PC has been randomly completely freezing up after pc startup. Time it takes to freeze is random but seems to follow anything past 10 mins.

After freezing up it suddenly shuts down after ~8 mins.

RAM and HDD have been replaced with new hardware, but this didnt seem to fix the issue.

OS has also been reinstalled, no fix tho.

Temps seem normal except for 1 spot on the motherboard which is sitting at a constant 80-90C upon startup. This pushes me to believe something is overheating on the mobo and forcing it to freeze and shutdown.

Drive health for SSD and HDD are healthy RAM was RMAd so have brand new sticks. Dont know what the issue is. Granted this PC is ~ 6 years old

The odd temp is labeled CPUTIN

 

[Mellon Head]

CPUTIN is usually the sensor located in the socket of your CPU. It is either malfunctioning (which I suspect, based on your core temps), or your CPU is actually getting that hot and freezing and shutting down.

Just to rule out an actual temperature problem, have you checked to make sure that your heat sink and fan are properly seated?

Also, what is the "CPU" temp reading in BIOS? It should be no higher than your core temps show at idle. In fact, it should be a little bit lower. If it is reading 80 to 90C in BIOS, then you either have a sensor problem, which means a new mobo, or you have an actual thermal issue that needs to be addressed before you torch that i5.

 

[ToxicXGL]

CPUTIN is usually the sensor located in the socket of your CPU. It is either malfunctioning (which I suspect, based on your core temps), or your CPU is actually getting that hot and freezing and shutting down.

Just to rule out an actual temperature problem, have you checked to make sure that your heat sink and fan are properly seated?

Also, what is the "CPU" temp reading in BIOS? It should be no higher than your core temps show at idle. In fact, it should be a little bit lower. If it is reading 80 to 90C in BIOS, then you either have a sensor problem, which means a new mobo, or you have an actual thermal issue that needs to be addressed before you torch that i5.

Temps after about 5 mins in bios. Didnt go over 33. So if this sensor is fried why would that cause the pc to freeze and crash?

 

[Mellon Head]

Temps after about 5 mins in bios. Didnt go over 33. So if this sensor is fried why would that cause the pc to freeze and crash?

Well, this is a puzzler.

If the sensor is malfunctioning, the computer can read this as the CPU overheating. CPUs have a self protection mechanism where they throttle the processor speed and shut down if they get too hot. I suppose it isn't unreasonable for the clock speed to go to zero in an extreme case of (perceived) overheating. Although 80 to 90C is extremely hot, Intels can run a little hotter than AMD chips for instance, but still, 80 to 90C is too hot and probably will cause throttling. It is my understanding, however, that the throttling temperature measurement is controlled by BIOS. If your BIOS reading never went above 33C, there is another problem somewhere. Since it crashed in BIOS, that points to a hardware problem IMO, since Windows is not involved at that point.

It could be a couple of things, though I would suspect the motherboard first of all. Your power supply could be flaky, but that wouldn't cause the freezing. The other culprits are the mobo and the CPU and the graphics card. If the graphics card is overheating, it can cause a freeze and reboot. I have had that happen with a failed GPU fan.

Have you checked the temps of your GPU?

 

[ToxicXGL]

Well, this is a puzzler.

If the sensor is malfunctioning, the computer can read this as the CPU overheating. CPUs have a self protection mechanism where they throttle the processor speed and shut down if they get too hot. I suppose it isn't unreasonable for the clock speed to go to zero in an extreme case of (perceived) overheating. Although 80 to 90C is extremely hot, Intels can run a little hotter than AMD chips for instance, but still, 80 to 90C is too hot and probably will cause throttling. It is my understanding, however, that the throttling temperature measurement is controlled by BIOS. If your BIOS reading never went above 33C, there is another problem somewhere. Since it crashed in BIOS, that points to a hardware problem IMO, since Windows is not involved at that point.

It could be a couple of things, though I would suspect the motherboard first of all. Your power supply could be flaky, but that wouldn't cause the freezing. The other culprits are the mobo and the CPU and the graphics card. If the graphics card is overheating, it can cause a freeze and reboot. I have had that happen with a failed GPU fan.

Have you checked the temps of your GPU?

Alright so i left the system on in the bios idle for ~ 2 hours and nothing happened, no crashes.

Checking the GPU temps, they stay under 31 for the most of it.

CPUTIN now goes anywhere from 50-90c and when it hits 80+ the CPU does throttle to about 1.5ghz

Also yesterday i was able to run a benchmark without any problems, but 30 mins after it froze and crash.

 

[Mellon Head]

I'm beginning to strongly suspect that your CPUTIN sensor is flaky. Especially since it is causing your CPU to throttle. CPUTIN is even kind of high in the first picture when it is not throttling.

What were you doing when the temp climbed to 92C? If you were idling, either your CPU heatsink/fan is not properly attached, or you have a problem with the mobo.

Also, your AUXTIN temp is rather high. What have you got for case cooling? You might need some fresh air intake to bring that temp down, and it will help your CPU temps.

Do you have the internal speaker connected? If so, you should hear some alarms (usually with most motherboards, though yours may or may not) when the CPU gets over temperature, and especially when it shuts down from over heat.

 

[ToxicXGL]

I'm beginning to strongly suspect that your CPUTIN sensor is flaky. Especially since it is causing your CPU to throttle. CPUTIN is even kind of high in the first picture when it is not throttling.

What were you doing when the temp climbed to 92C? If you were idling, either your CPU heatsink/fan is not properly attached, or you have a problem with the mobo.

Also, your AUXTIN temp is rather high. What have you got for case cooling? You might need some fresh air intake to bring that temp down, and it will help your CPU temps.

Do you have the internal speaker connected? If so, you should hear some alarms (usually with most motherboards, though yours may or may not) when the CPU gets over temperature, and especially when it shuts down from over heat.

Idle when 80-92.

So i replaced the thermal compound and re-seated the CPU cooler which brought down the temps to a stable 60-70c, but this only lasted for about 2 mins. After that it went back to the usual 80+. What i did notice tho is that my cpu cooler mounting pin is kind of busted, but the cooler still sits well and firm on the CPU. Fan is also working. CPU still throttling.

Ive got 3 intake fans. 1 in the front panel and 2 in the side window all blowing intake.

CPUTIN seems to be staying at 80+ all the time no longer ranging from 60-90, constant throttling

Going to try the mobo speaker right now. Just need to find the sucker

 

[Mellon Head]

Going to try the mobo speaker right now. Just need to find the sucker

I've asked some other experienced users to look in to offer a second opinion. There is definitely an issue here if it is reaching 80 to 92C at idle.

Bear with us as we examine the thread.

 

[ICIT2LOL]

Ok now just my two cents worth an it is always one of my first things to look at with sudden dropouts / crashes etc is the PSU
Can you read though this and send back the PSU info please you can also go into great depths with other components but please start with the PSU
Using HW Info
PART A:
You can test the volts on the PSU with HW Info HWiNFO - Download < download the right bit version and close the right hand window select Sensors and scroll down to the power section where you will see what the volts are doing see my pic. In my pic the section (Nuvoton) with VBATT as a dead give away you are in the section for the rail voltages. There are other section titles and one that pops up often is ITE (sometimes the usual one for Gigabyte boards)
Now the voltage on the different rails have to be within 5% =+/- of what is required or the machine will not work properly if at all.
Limits +/- (minimum - correct - maximum)
12v1 = 11.4 - 12 - 12.6v
12v2 = 11.4 - 12- 12.6v
5v = 4.75 – 5 - 5.25v
3.3v = 3.135 – 3 – 3.465v
-12v = -10.80 - -12 - -13.20v
+5VSB = 4.75 – 5 - 5.25v
The Power good signal voltage at pin 8 on the 24 pin plug (grey cable) should be the same as the 5v rail reading/s
See this for the rail voltage info
PSUs 101: A Detailed Look Into Power Supplies (Section 2.)
The original right hand window shows the machine running and is handy for that but for looking at the components in some detail close it and use the main left hand side panel
FOR OTHER COMPONENTS
PART B:
Open each small square with + in it on the section the components are in and then click on the individual component/s (it will highlight in blue) - in the right hand side will appear all sorts of details including brands speeds and other essential info that particular device. See pic for example.

You can also use PART B: to see in detail and of the other components data ie temps speeds frequencies etc etc

Edit: This is an old shot the format of HW Info is very slightly different now.

 

[ToxicXGL]

Ok now just my two cents worth an it is always one of my first things to look at with sudden dropouts / crashes etc is the PSU
Can you read though this and send back the PSU info please you can also go into great depths with other components but please start with the PSU
Using HW Info
PART A:
You can test the volts on the PSU with HW Info HWiNFO - Download < download the right bit version and close the right hand window select Sensors and scroll down to the power section where you will see what the volts are doing see my pic. In my pic the section (Nuvoton) with VBATT as a dead give away you are in the section for the rail voltages. There are other section titles and one that pops up often is ITE (sometimes the usual one for Gigabyte boards)
Now the voltage on the different rails have to be within 5% =+/- of what is required or the machine will not work properly if at all.
Limits +/- (minimum - correct - maximum)
12v1 = 11.4 - 12 - 12.6v
12v2 = 11.4 - 12- 12.6v
5v = 4.75 – 5 - 5.25v
3.3v = 3.135 – 3 – 3.465v
-12v = -10.80 - -12 - -13.20v
+5VSB = 4.75 – 5 - 5.25v
The Power good signal voltage at pin 8 on the 24 pin plug (grey cable) should be the same as the 5v rail reading/s
See this for the rail voltage info
PSUs 101: A Detailed Look Into Power Supplies (Section 2.)
The original right hand window shows the machine running and is handy for that but for looking at the components in some detail close it and use the main left hand side panel
FOR OTHER COMPONENTS
PART B:
Open each small square with + in it on the section the components are in and then click on the individual component/s (it will highlight in blue) - in the right hand side will appear all sorts of details including brands speeds and other essential info that particular device. See pic for example.

You can also use PART B: to see in detail and of the other components data ie temps speeds frequencies etc etc

Edit: This is an old shot the format of HW Info is very slightly different now.

After windows updates was able to play Warframe on max for 2 hours without any issues. PC is still running as we speak.

 

[ICIT2LOL]

That's ok mate the rail volts are fine that's the main thing and to be honest the PSU is often overlooked as a cause for problems and as I said it is always something I like to check early on.

 

[Anak]

Hi, just thinking out loud here:

First this:

What i did notice tho is that my cpu cooler mounting pin is kind of busted, but the cooler still sits well and firm on the CPU. Fan is also working. CPU still throttling.

Then this:

That means that HW Monitor is looking for a sensor where there isn't one, for some reason it has CPUTIN sensor mapped to the wrong location for your motherboard and since its not getting any data at all its reading as its max possible value (128 C)

Check your temps with some other programs, and check in BIOS to see what it says, you may have a faulty CPU temperature probe.

Source: http://www.tomshardware.com/forum/327738-28-cputin-temp-128c#r3045959

Did you ever stop to consider that, that mounting pin that is kind of busted is part of the CPU temperature probe system? It doesn't matter if it sits well and firm, it's kinda busted and until you test with a known good cooling unit you won't know for sure.

Think of it this way; How can the cooler know how the CPU is doing if a probe is busted?

 

[ICIT2LOL]

Think of it this way; How can the cooler know how the CPU is doing if a probe is busted?

Very true and I am guessing that maybe a purely mechanical method might be needed as in one of those infrared guns - they are not very expensive and I have found them really good even if you can only get half way close enough shot at the CPU.

Now in my experience too say with a liquid cooled unit the reading from the gun even off the radiator or cooling pipes if the unit is an air cooled one the difference between the actual CPU temp and the cooler is within approx 10C of the CPU proper sensor reading.

As I said the guns themselves are really cheap and very accurate.

 

[Mellon Head]

After windows updates was able to play Warframe on max for 2 hours without any issues. PC is still running as we speak.

Well, now I'm really baffled.

 

[ToxicXGL]

So a little status update.

Looking for a new cpu cooler.

The pc has been running smoothly since yesterday no crashes. Able to play games such as PUBG and warframe.

Still will buy the cooler regardless. Thanks for the replies!

EDIT: As typing this the pc crashed. Idle.

 

[ICIT2LOL]

Yes well I did notice back in post #7 you said that the cooler mount was kinf of busted just what did you mean by busted?? bent or the thing has cracked away from the board ?

Am asking because as you probably know the CPU spreader and the cooler heatsink have to have as much contact as possible and to do that the two surfaces have to have as much even pressure between them plus the compound to make sure the microscopic scratches and pits on those two surfaces have the optimum level of contact to carry away the heat produced by the CPU.
Now it stands to reason that if the two surfaces are not at 180 degrees to each other as is physically possible then the heat transfer is not going to be even at best and useless at worst.

If you do get a new cooler then I don't know how well you know how to do this and I am not being presumptuous then you need to not only clean the two surfaces but prepare them first before applying a reputable brand compound.
Personally I use Arctic Silver products and my method is to clean the two surfaces (even if they are new) with the AS cleaning solution and then prepare the surface with the prep solution and then using one of my fingers inside a vinyl glove - a latex one will do to keep any body oils from myself from contaminating the two surfaces and genrtly spread an even thin film of the AS thermal compound over each surface.
I then place a spot (rice grain size) of the AS compound on the centre of the CPU heatsink and then fix down the cooler plate / spreader. The compound will on heating up "bleed out" across the two surfaces which have been "primed" with compound. This way I am sure that the compound is in contact with the entire surface of the two components.
I do not subscribe to just using alcohol and/or just putting a blob of compound on the CPU an just putting back the cooler only because I am very pedantic and OCD about doing this because of problems I have had I the past.
The products are in these links Arctic Silver Incorporated - ArctiClean & Arctic Silver Incorporated - Arctic Silver 5 The products are a little more expensive than other s which might be good but I use only this now.

Edit: I should have said though that the choice is yours of course. Plus use clean pieces of cloth or tissue for each solution do not mix them.

 

[ToxicXGL]

Yes well I did notice back in post #7 you said that the cooler mount was kinf of busted just what did you mean by busted?? bent or the thing has cracked away from the board ?

Am asking because as you probably know the CPU spreader and the cooler heatsink have to have as much contact as possible and to do that the two surfaces have to have as much even pressure between them plus the compound to make sure the microscopic scratches and pits on those two surfaces have the optimum level of contact to carry away the heat produced by the CPU.
Now it stands to reason that if the two surfaces are not at 180 degrees to each other as is physically possible then the heat transfer is not going to be even at best and useless at worst.

If you do get a new cooler then I don't know how well you know how to do this and I am not being presumptuous then you need to not only clean the two surfaces but prepare them first before applying a reputable brand compound.
Personally I use Arctic Silver products and my method is to clean the two surfaces (even if they are new) with the AS cleaning solution and then prepare the surface with the prep solution and then using one of my fingers inside a vinyl glove - a latex one will do to keep any body oils from myself from contaminating the two surfaces and genrtly spread an even thin film of the AS thermal compound over each surface.
I then place a spot (rice grain size) of the AS compound on the centre of the CPU heatsink and then fix down the cooler plate / spreader. The compound will on heating up "bleed out" across the two surfaces which have been "primed" with compound. This way I am sure that the compound is in contact with the entire surface of the two components.
I do not subscribe to just using alcohol and/or just putting a blob of compound on the CPU an just putting back the cooler only because I am very pedantic and OCD about doing this because of problems I have had I the past.
The products are in these links Arctic Silver Incorporated - ArctiClean & Arctic Silver Incorporated - Arctic Silver 5 The products are a little more expensive than other s which might be good but I use only this now.

Edit: I should have said though that the choice is yours of course.

Haha no worries I'm not a noob. I have a tube of good Ole artic mx4. And I prefer the pea

So on the lga 1155 it has those weird fastening pins and one of the 2 plastic ends broke off completely. May be able to get a new stock cooler tomorrow.

 

[ICIT2LOL]

Haha no worries I'm not a noob.

Got you my friend and I was not insinuating you do not know what you are doing just offering you a different method of reapplying the compound as I have tried all manner of methods and find mine ensures that tight contact without contaminating the surfaces and every square millimetre of the surfaces are covered with compound.

But as I said too it's your gear and choice.

Yes those fixers are a piece of work too personally I have had many battles with them and why they make them so flimsy (in my mind) is crazy when one is plating around a piece of equipment that costs hundreds of dollars.

Again I now would only choose liquid coolers because most of my issues have been with those enormous air types.