1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

AMD K8 CPU overheating

Discussion in 'Hardware' started by itbookham, Feb 7, 2010.

  1. itbookham

    itbookham Nibble Poster

    94
    0
    11
    Hello,

    I would be grateful for some advice. My computer is a generic x64 running a 64 bit dual core AMD K8 processor.

    There appears to be an overheating problem with the cpu. After about 15-20 minutes of running the computer will automatically shutdown. After it had cooled down, I loaded Core Temperature CPU monitor. I ran a couple of programs, just to load the cpu. One of the cpu core temperatures quickly reached 125 degrees and subsequently shut down.

    I unclipped the fan/sink assembly and removed all the existing thermal paste and applied a fresh coat. On booting up the computer it shut down again after about 20 minutes.

    The PC has working PSU, case and CPU fans? Perhaps it is a faulty CPU? Has anyone else experienced a similar problem?

    Many thanks,
    Mark
     
    WIP: A+
  2. derkit

    derkit Gigabyte Poster

    1,479
    54
    112
    I haven't experienced it but a couple of thoughts:

    - faulty CPU (as you said)
    - faulty sensors reporting an incorrect temperature. Normally I've seen motherboards moan at around 85 degrees C and power off, but not at 125 (unless its recording in Fahrenheit). Even so, at 125C my guess is its probably gone to CPU-heaven!
     
    Certifications: MBCS, BSc(Hons), Cert(Maths), A+, Net+, MCDST, ITIL-F v3, MCSA
    WIP: 70-293
  3. SimonD

    SimonD Terabyte Poster Moderator

    3,463
    397
    199
    Whilst you cleaned the CPU of TIM did you also clean off the fan of all the dust? Have you checked to see what speed the fan is spinning at? you can use programs such as Everest to check various temps\speeds etc.

    Before assuming it's the cpu (possible) I would check the fan itself (more likely).
     
    Certifications: CNA | CNE | CCNA | MCP | MCP+I | MCSE NT4 | MCSA 2003 | Security+ | MCSA:S 2003 | MCSE:S 2003 | MCTS:SCCM 2007 | MCTS:Win 7 | MCITP:EDA7 | MCITP:SA | MCITP:EA | MCTS:Hyper-V | VCP 4 | ITIL v3 Foundation | VCP 5 DCV | VCP 5 Cloud | VCP6 NV | VCP6 DCV | VCAP 5.5 DCA
    WIP: VCP6-CMA, VCAP-DCD and Linux + (and possibly VCIX-NV).
  4. dazza786

    dazza786 Megabyte Poster

    758
    30
    67
    Amd .
     
    Certifications: MCP (271, 272, 270, 290, 291, 621, 681, 685), MCDST, MCTS, MCITP, MCSA, Security+, CCA(XA6.5)
  5. itbookham

    itbookham Nibble Poster

    94
    0
    11
    Hello,

    Thanks for the suggestions.

    The fan and heat sink were cleared of dust and thermal paste before reseating.

    I have just checked the system and cpu temperatures/current cpu fan and system fan speeds:

    39 degrees
    84 degrees
    2350 rpm
    2600 rpm

    I think the fan speeds look sound, but I am concerned at the cpu temperature - 84 degrees celsius in bios.

    Mark
     
    WIP: A+
  6. SimonD

    SimonD Terabyte Poster Moderator

    3,463
    397
    199
    The BIOS will always read higher than the OS. Does the board support Cool'n'Quiet? if so what happens if you enable it.

    Also depending on the TIM it can take as much as 200 hours for it to cure. It's also important that you didn't put too much on and you did apply it correctly, ideally if it's a dual core processor you need approx 3 - 4mm round dollop in the center of the heat spreader.
     
    Certifications: CNA | CNE | CCNA | MCP | MCP+I | MCSE NT4 | MCSA 2003 | Security+ | MCSA:S 2003 | MCSE:S 2003 | MCTS:SCCM 2007 | MCTS:Win 7 | MCITP:EDA7 | MCITP:SA | MCITP:EA | MCTS:Hyper-V | VCP 4 | ITIL v3 Foundation | VCP 5 DCV | VCP 5 Cloud | VCP6 NV | VCP6 DCV | VCAP 5.5 DCA
    WIP: VCP6-CMA, VCAP-DCD and Linux + (and possibly VCIX-NV).
  7. itbookham

    itbookham Nibble Poster

    94
    0
    11
    Hi,

    CoolnQuiet is enabled already in BIOS.

    When you say 3-4mm round dollup of thermal paste, do you mean in diameter or height, or both?

    Thanks,
    Mark
     
    WIP: A+
  8. beaumontdvd

    beaumontdvd Kilobyte Poster

    487
    3
    32
    Try uninstalling core temp I had a smilar problem mate. Let the bios control the temp. Has it only starting overheating when you installed it. Or did icore temp come with the pc?
     
    Last edited: Feb 7, 2010
    Certifications: 070-271, 070-272, (MCDST)Level 1,2,3 NVQ
    WIP: 070-270, A+, N+, S+,MCDST 7 Upgrade
  9. SimonD

    SimonD Terabyte Poster Moderator

    3,463
    397
    199
    Ideally both, it needs to be about half the size of a pea, when you then put the hsf back on give it a slight twist, then the force of screwing down the hsf should spread it evenly over the two cores.
     
    Certifications: CNA | CNE | CCNA | MCP | MCP+I | MCSE NT4 | MCSA 2003 | Security+ | MCSA:S 2003 | MCSE:S 2003 | MCTS:SCCM 2007 | MCTS:Win 7 | MCITP:EDA7 | MCITP:SA | MCITP:EA | MCTS:Hyper-V | VCP 4 | ITIL v3 Foundation | VCP 5 DCV | VCP 5 Cloud | VCP6 NV | VCP6 DCV | VCAP 5.5 DCA
    WIP: VCP6-CMA, VCAP-DCD and Linux + (and possibly VCIX-NV).
  10. simonp83

    simonp83 Kilobyte Poster

    254
    4
    32
    With the computer side off and after having the machine on for a good 15 minutes, touch the heatsink on the cpu gently (might be very hot so be careful) and see how hot it is. If it's too hot to touch, there is something horrible going on with the chip, if it's warm or even cool, it means that the temp reading is wrong or there is a bad contact between the hsf and the chip.
     
    Certifications: A+, MCP, MCDST, MCTS, MCITP
    WIP: 70-291
  11. itbookham

    itbookham Nibble Poster

    94
    0
    11
    Hello Simon,

    With the side cover off, I ran the pc for about 15 minutes without any programs running. The heatsink was cool. I then ran a couple of programes to increase the temperature. The core cpu temperature peaked at about 116 degrees celsius, the heatsink was still cool to touch.

    I don't think the temperature reading are wrong, as both the bios PC Health Check and the Core Temperature program running in Windows indicate that the core temperature is very high when multi-tasking.

    It is possible that there is a mis-aligned contact between cpu and heat sink, but I have now cleaned, re-applied and seated the heat sink/fan twice now without resolving the problem.

    I just wonder if it is the cpu itself but there are no other symptoms to indicate this.

    Most frustrating. Thanks for your suggestions and also to others who have contributed.
     
    WIP: A+
  12. simonp83

    simonp83 Kilobyte Poster

    254
    4
    32
    How is the hsf attached to the cpu socket, is it the retention clip or is the screw in one? If you're cpu is hitting 116celcius, you'll be getting major issues with reliability and system crashes, not just with the system shutting down. If you're getting no crashes at all, every looks to be seated correctly and the hsf is only warm, i do think it's a faulty sensor probe, but it'll be hard to tell without seeing the system.
     
    Certifications: A+, MCP, MCDST, MCTS, MCITP
    WIP: 70-291
  13. itbookham

    itbookham Nibble Poster

    94
    0
    11
    The hsf is attached with a retention clip rather than screws.

    The only pc component that would force a shutdown due to temperature is the cpu, so I am not too sure where to go from here.
     
    WIP: A+
  14. simonp83

    simonp83 Kilobyte Poster

    254
    4
    32
    I've seen powersupplys, cpus and motherboards that can cause this. If the cpu shuts down even when idle, i don't think it'd be the powersupply, might just be a faulty cpu.
     
    Certifications: A+, MCP, MCDST, MCTS, MCITP
    WIP: 70-291
  15. itbookham

    itbookham Nibble Poster

    94
    0
    11
    I tried another psu. The cpu temperature still rose sharply peaking 114 degrees. I shut the pc down after this. At idle the system seems to cope. The system shuts down only when you start to load the cpu.
     
    WIP: A+
  16. beaumontdvd

    beaumontdvd Kilobyte Poster

    487
    3
    32
    yes, but what if its software thats causing the shutdown. I had a temperature monitor program installed that was set to default. But the defaults were not right for my particular pc. The software can force a shutdown, it can also decrease/increase fan speed so maybe the default fan speed has been decreased and hence the reason the cpu is overheating and coretemp is shutting your pc down. Try uninstalling it or system restore if you have recently installed it. This worked for me?

    Regards,
     
    Certifications: 070-271, 070-272, (MCDST)Level 1,2,3 NVQ
    WIP: 070-270, A+, N+, S+,MCDST 7 Upgrade
  17. greenbrucelee
    Highly Decorated Member Award

    greenbrucelee Zettabyte Poster

    14,283
    254
    329
    You may have a faulty cpu but it is always wise to explore all avenues first.

    for the record a psu can get hot and shut off, faulty ram can also cause the system to shut of unexpectedly too.

    since your issue is temp related and you have tried another psu although I don't know if it was a decent one or just some generic one then I would suspect you either have a faulty mobo sensor or cpu is on its way out.

    it is very rare to see a cpu die unless it has been abused either by overclocking or constantly pushed but its not impossible for a cpu to give up.

    your cpu should be shutting down long before 100 degrees c, if I remember your cut of should be 72.2 degrees c, so you may have a faulty sensor.

    do you have another cpu you could test?
     
    Certifications: A+, N+, MCDST, Security+, 70-270
    WIP: 70-620 or 70-680?
  18. SimonD

    SimonD Terabyte Poster Moderator

    3,463
    397
    199
    How cool is that HSF? by rights it shouldn't be cool to touch with the machine on for 15 minutes, the idea is that it transfers the heat away from the cpu, ergo the HSF should be hot but not as hot as the cpu.
    If the hsf is cool to touch after the pc has been on for 15 minutes your hsf isn't in contact with the cpu well enough and that being the case the cpu is using the thermal cutout to shut down.

    I would honestly look at another hsf, one that actually screws down rather than clips.
     
    Certifications: CNA | CNE | CCNA | MCP | MCP+I | MCSE NT4 | MCSA 2003 | Security+ | MCSA:S 2003 | MCSE:S 2003 | MCTS:SCCM 2007 | MCTS:Win 7 | MCITP:EDA7 | MCITP:SA | MCITP:EA | MCTS:Hyper-V | VCP 4 | ITIL v3 Foundation | VCP 5 DCV | VCP 5 Cloud | VCP6 NV | VCP6 DCV | VCAP 5.5 DCA
    WIP: VCP6-CMA, VCAP-DCD and Linux + (and possibly VCIX-NV).
  19. itbookham

    itbookham Nibble Poster

    94
    0
    11
    Thanks for the all information.

    The cpu heat sensor software was installed after the problem arose, just as a double check on the cpu temperature.

    It is possible that there is a fault with the hsf, or that it is not seated correctly. I will check this again, but I have already checked it twice.

    Faulty memory a possibility - could be the memory stick or motherboard ram slots?

    Another possibility the temp. sensor on the motherboard is giving an incorrect cpu temperature reading and forcing the computer to shutdown. If so, this will probably be a replacement motherboard.
     
    WIP: A+
  20. beaumontdvd

    beaumontdvd Kilobyte Poster

    487
    3
    32
    Have you checked the bios temp settings if you have them? are they set to default?
     
    Certifications: 070-271, 070-272, (MCDST)Level 1,2,3 NVQ
    WIP: 070-270, A+, N+, S+,MCDST 7 Upgrade

Share This Page

Loading...