Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

My computer shuts down during intense use of the processor, but the sensors show that temperature of CPU is less than 78oC.

By intense use, I mean any of the following: playing games, converting/encoding video, even simultaneously using tabs in the browser.

Hardware Information:

  • CPU: AMD Athlon 64 X2 6000+
  • Chipset: nVidia nForce 3 Pro
  • Video: GeForce 7300 GT

Hardware Monitor dialog:

cpuid hardware monitor

Why would this be happening and what could be the root of the problem?

share|improve this question
    
When does it shut down? While playing a game? Please define the intense use. –  Mehper C. Palavuzlar Oct 26 '09 at 8:23
    
Intense use defines: Playing game, Converting/Encoding Video, Even simultaneously using tabs in browser. –  user15475 Oct 26 '09 at 8:32
1  
"My computer shuts down during intense use of the processor, but the sensors show that temperature of CPU is less than 78oC." I would assume you meant Fahrenheit? 78 C = 172.4 = cookin'! –  Nathaniel Jan 22 '10 at 5:05
add comment

6 Answers 6

up vote 5 down vote accepted

I had a computer case without fans once, and my Pentium III processor would overheat easily (it's not that hard in the Israeli summer...). Eventually I opened the side panel of the case and left it open, aiming a 16" fan directly at the motherboard. This was the only thing that cooled down the computer enough to keep running.

share|improve this answer
    
I did this right now and it had effect but computer still restarts while converting/encoding video (this is more processor using feature than gaming). –  user15475 Oct 26 '09 at 10:11
1  
this is excellent advice from Yuval -- but if opening your case and pointing a giant fan at it does not help, I would 1) check the CPU heatsink to make sure it is properly attached with thermal paste and after that is checked, 2) run a Prime95 torture stability test overnight to see if the memory is faulty. –  Jeff Atwood Jan 8 '10 at 0:56
add comment

You need to follow the following steps:

  1. Clean your computer. Open the case, and dust it. Vaccume clean or blow air to remove the dust accumulation on Motherboard and Fans.

  2. Apply a cooling paste between your CPU and CPU Fan. Cooling pastes are excellent heat conductors, Intel's paste should be available for around $1.

  3. If you don't have any additional fans already in the cabinet, install them. Add one to the front, one at the back. The front fan should suck air in while the back should throw it out - promoting ventilation. They should be of the same make and size to facilitate airflow.

  4. Change the PSU if it is old. I switched from a 450W local PSU to a 390W Cooler Master for one of my PCs - and immediately observed amazing difference in cooling.

While we are on the topic, this is an excellent free software to monitor the temperatures (and also has a horde of other features): Speedfan.

share|improve this answer
    
Agreed. Sounds like poor heat-sink contact. A burst of CPU usage can quickly push the CPU temp above shutdown temp if the heat-sink is not very well connected thermally. Power supply would be the next thing to check-on. –  DanO Oct 28 '09 at 4:17
add comment

I've found that when you're really overheating your computer won't start up immediately after shutting off, or if you do, will shut off again before making it to the desktop. OTOH when I found that my MSI AtiXpress200 motherboard's bios wouldn't support the GeForce card I had, during a game it would shut off at a random point. When I had bad ram, windows itself was pretty good about mentioning that it thought I must have bad ram after it was started from a crash, and that I should scan for ram problems. Personally I suspect you have an issue where you PCIe BIOS isn't quite up to date with the geforce's use pattern.

share|improve this answer
    
I'll try updating BIOS and if that will help solve the problem you have my vote :) –  user15475 Oct 26 '09 at 10:14
add comment

Ok, I had the same problem as the first post. As you will see from my comment above.

I cleaned the fan assembly on top of the heat sink... but what I didnt check was between the fins of the heat sink...

I will admit it, I smoke in my office... bad habbit, what can I say.

So, on a second inspection of the heat sink (paying a little more attention this time) I noticed there was build up between the fins - which no doubt was stopping air flow over the sink.

I gave it a damn good blow out and no more shutdowns.

I hope this helps anyone having the same issues.

share|improve this answer
add comment

In your screen, it shows that the nforce3-pro CPU is 85 degrees. Is the sensor malfunctioning in the software ?

share|improve this answer
    
Probably, but I haven't seen it less or more than 85 degrees. But when I touch it has temperature of a cup coffee. :) –  user15475 Oct 26 '09 at 8:43
add comment

Software based CPU temperature sensors rely on the BIOS settings. Check your BIOS for power management settings to see if the settings have been changed from the default. Otherwise, I suspect a bad or under-powered PSU.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.