Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

Suspected GPU or CPU getting hot & crashing


  • Please log in to reply
15 replies to this topic

#1 officerchops

officerchops

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 03:58 AM

Hello,

 

I suspect a fan or my water cooling is broken and I need to know what's broken.

My screen goes black and the pc crashes after about 10 minutes of gameplay.

I can hear the fans ramping up to full duty cycle it's very loud

 

I am logging GPU & CPU temperatures in a program called 'SpeedFan' the logs are stored in .xml format (which I can't make much sense of because I can't use spreadsheeting software to create a graph).

I am running windows 7

I have a corsair H50 water cooler on my CPU

 

Please let me know if I'm going about this in an overly complicated way, perhaps there is an easier solution to work out the health of my hardware and what exactly is happening?

What information can I give? pictures of my setup? how can I tell if the water cooler is working?

 

Thank you so much for any help!

 

-Charles



BC AdBot (Login to Remove)

 


#2 Kilroy

Kilroy

  • BC Advisor
  • 3,461 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Launderdale, MN
  • Local time:02:02 PM

Posted 27 January 2015 - 06:20 AM

Most likely your BIOS will tell you your CPU temp, can you boot up after the shutdown and check what your CPU temp is?

 

That will allow you to confirm your CPU is overheating.  The problem is that it won't tell you which part of your cooling solution isn't working.



#3 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 07:31 AM

at shutdown CPU temp was 105 and GPU 115

degrees celcius



#4 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:03:02 PM

Posted 27 January 2015 - 08:22 AM

Both temps seem high if that is in C... I would dust out the computer:

 

To dust out a pc I have recently been using a shop vacuum in "blower" mode with a small nozzle (blowing out rather than sucking in) mine has the option to move the hose to the exhaust port for blower mode. I like it better than an air compressor for the fact that no liquid will be in the air coming out. Air compressors can be used but I would recommend you have a water filter in the line and use no more than 40psi with a nozzle, also be careful not to overspeed fans, they can break quite easily. You will want to dust out the heat syncs and fans. Since you have a water cooler you will want to dust out the radiator as well, they hold dust well and take a bit more air to clean but that should fix any heat problems unless the water cooler has a problem or your connections are lose etc. Don't forget to dust out your power supply. PSU's may cause crashes and shutdowns as well.

 

FYI: CPU overheating shutdowns are generally an actual shut down with no restart/reboot after the machine instantly turns off. The mainboards typically shut off the pc entirely and you would have to push the power button to turn it back on. GPU's on the other hand generally just crash and reboot, often with a BSOD.

 

If there are no obvious hot temps while sitting idle you might want to try different benchmark tools to stress the components 1 at a time. Prime95 for the cpu stress test and furmark for the gpu stress test.

 

If prime95 crashes within 5-10 minutes every time, you may have a cpu cooling issue or cpu related issue, if that passes and furmark crashes every time then you have a gpu problem.

 

I generally do not let furmark run longer than 15 minutes. Most times I have had to deal with a failing video chipset furmark crashed within the first 10 minutes.

 

You can also monitor your temps while doing all of this using HWmonitor. If your temps are reaching near the max recommended, consider opening your computer case and dusting out your computer. I dust out my gaming rig about every 3 months. If you live in a dusty environment like myself and don't dust out your machine frequently, you will start to have overheating issues. Overheating chipsets even once can damage them.


Edited by zingo156, 27 January 2015 - 09:20 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#5 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 08:27 AM

thanks zingo, downloading them now.

can I just log the temps in SpeedFan? or is it a bad program? I thought it was the equivelent of HWMonitor.

those benchmarking programs are just what I need.

I'll go about cleaning the GPU now. I've already done the cooler radiator, it was terrible. (I'm still crashing and BOTH CPU and GPU are heating up, so maybe I've broken the CPU cooler?)



#6 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:03:02 PM

Posted 27 January 2015 - 08:32 AM

It is possible to break coolers but typically it is the connection that fails or the brackets that hold the cooler to the cpu. Watch the temps while running a stress test. You can use any software you want, I typically use HWmonitor because I can also watch min/avg/max voltages which sometimes helps catch a bad psu.


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#7 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 08:38 AM

FYI: CPU overheating shutdowns are generally an actual shut down with no restart/reboot after the machine instantly turns off

yep that's me!


I have yet to run prime95, because I can't work out how to use it.



#8 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 08:41 AM

trying a 'blend' stress test



#9 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:03:02 PM

Posted 27 January 2015 - 09:14 AM

Blend should be enough. Whatever the default settings are usually work.


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#10 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 09:20 AM

nah it just got up to 85, ill try fur for GPU now.

 

when I'm playing games, the CPU gets to 105 and SOMETHING breaks



#11 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:03:02 PM

Posted 27 January 2015 - 09:27 AM

If you do not have good case cooling, the gpu can heat soak the cpu/cpu cooler. In that case running both at the same time will result in higher temps. 85C is still pretty warm. What cpu do you have?


Edited by zingo156, 27 January 2015 - 09:30 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#12 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 09:31 AM

i5 750

 

Fur just heated my GPU to 115 and CPU to 105 then crashed.

 

So maybe the GPU is still clogged up...

or

the cooler is not circulating/out of water/broken

 

Any other things you can think of before I tear my GPU apart?

Also is there a quick way to do it without needing to reapply thermal paste or take out 15 screws? I have a GTX 480 GPU it has a black plastic shrouding (casing) that I don't know how to remove



#13 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 10:01 AM

There was a WALL of dust between the heatsink and fan on the GPU.

Loading it up now.

 

Thank you temperature sensors and automatic cut-outs!

and of course you Zingo.



#14 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:03:02 PM

Posted 27 January 2015 - 10:43 AM

It definitely seems heat related. Are the temps being reported in C not F? If so that is hot. I tried to find solid max temps on the cpu, from intel the Tcase max temp is 72.7°C.

 

I am not certain what temperature your bios is set to automatically shut down at but that is something worth checking, don't change it.

 

What case do you have? Are all of your case fans working and dust free? It seems heat is being built up in the case. Definitely clean all of the heatsinks and fans etc.


Edited by zingo156, 27 January 2015 - 10:44 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#15 officerchops

officerchops
  • Topic Starter

  • Members
  • 22 posts
  • OFFLINE
  •  
  • Local time:06:02 AM

Posted 27 January 2015 - 11:01 AM

problem solved. thanks guys






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users