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

Q6600 sky high temps

Discussion in 'Tech Support' started by DE.223, 30 Aug 2008.

  1. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    All, I am looking for a second opinion as to whether I have fried my CPU permanently . . .

    Currently my q6600 is averaging about 70° C idle . . . At load it has been known to get up to 92°. These readings are consistent across realtemp, coretemp and the hardware monitor in the BIOS. I would find this odd under any circumstances . . . however, considering I am using a Thermaltake IFX-14 and a Delta 120mm focussed flow fan pushing 190CFM it is more than a tad frustrating :wallbash:

    I am wondering whether this is down to me frying my last motherboard (Asus P5N-D) after abusing it a little too much (one too many romsip updates . . ). The q6600 temp reporting seemed fine in the P5N-D averaging around 37° idle with a xilence redwing. Is is possible that the CPU temp reporting was somehow fried in the motherboard death?

    I have since obtained a maximus II formula and have been looking forward to installing a watercooling system and going hunting for a decent overclock once I scrape some cash together. However, since the second I put the q6600 in the new board it has been giving me very high temps . . . I have tried remounting the IFX-14 to no avail. Updating the maximus bios (against my better judgement) has also had no effect.

    The IFX-14 is always cool to the touch and I have no stability issues at all. Is there any guru trick I can try to find out what is really going on? :confused:
     
  2. Kurayamino

    Kurayamino As long as the Raven flies

    Joined:
    5 Sep 2005
    Posts:
    542
    Likes Received:
    0
    Someone might correct me if I am wrong but I believe there to be a temp diode on the cpu maybe that got frazzled in the incident which then any software that is finding the temp would get the info from that?! Maybe that is the problem?
     
  3. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    Urrgh :sigh: Is there anyway to check if a diode is fried? When I run the "test sensors" test in RealTemp it gives me a good amount of movement for the sensors. If the diode was fried would the sensor movement be zero?
     
  4. outlawaol

    outlawaol Geeked since 1982

    Joined:
    18 Jul 2007
    Posts:
    1,935
    Likes Received:
    65
    Try leaving the fan off for bit with the heatsink on and see if the bugger gets really hot. If its 'warm' then its not frying, if its hotter then hell fire then its cooking something. And by 'off for a bit' I mean a short time, nothing prolonged. Touch it around the pipes and close to the CPU, anything over 80c should be very noticeably hotter.

    I think the diodes are really small, and as far as I know there is no way to 'physically' check them. And no, if a sensor dies it usually is in a high or low number. A fried sensor does actually still read, just not within a tolerance that is within bounds, so the numbers will still bounce about. Another way to take a check is with a infrared temp sensor gun. It detects and reports temps of surface areas without contact. Not a common tool, but works nicely.

    :)
     
  5. mm vr

    mm vr The cheesecake is a lie

    Joined:
    18 Nov 2007
    Posts:
    2,968
    Likes Received:
    84
    Also, check that you're using a correct TJmax in the programs you monitor the temps in. Nobody really knows what the TJ is for the Q6600 but people's best guesses are 95-105C. :)
     
  6. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    Good idea . . I wish I had thought of that before. :duh:

    So I unplugged the Delta and switched the beast on. Wow, nice not to have to listen to a typhoon raging in my case. BIOS gave me a CPU temp warning but I ignored it and booted to windows anyway . . . fired up p95 and RealTemp. I watched the temp climb steadily upward over 10 minutes until it hit 100° . . . then the machine switched itself off as expected.

    What is worrying, is that far from being cool to the touch the heatsink was very hot. Not burning hot, but enough that it would sting fingers touching it.

    Does anyone own a similarly sized heatsink and is using it to passively cool a quadcore? Is such heat normal? I am really starting to think I have somehow mutated my mild mannered q6600 into a raging thermonuclear fury . . .

    I currently have the Tjmax in RealTemp set at 95.
     
  7. 7755matt

    7755matt New Member

    Joined:
    31 Aug 2008
    Posts:
    23
    Likes Received:
    0
    do you have a temp sensor you could actually fit near the cpu socket? Remount the ifx with the sensor near by. It could also be that its not mounted right? I know you've said you remounted it, the backplate cooler hasnt slipped has it?
     
  8. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    Sadly, I do not yet own a temperature probe . . . on my wish list though. :lol:

    I think I shall try and tame it with the watercooling system I have been meaning to put together . . . if that does not work I'll try flogging it somewhere . . . :sigh:
     
  9. Malketh

    Malketh New Member

    Joined:
    22 Jan 2008
    Posts:
    143
    Likes Received:
    0
    You know I was having the same issue but not quite as hot, my Q6600 was running around 50-60 at idle which was alarming me quite a bit as that's the same temps my gpus were running at idle. Then I remembered I'm using the stock intel heatsink and fan, so I ordered the Thermalright Ultra-120 and a hopefully nice thermometer. Hopefully the stuff should get here tomorrow and I can bring those temps down a good dozen degrees or so.
     
  10. badders

    badders Neuken in de Keuken

    Joined:
    4 Dec 2007
    Posts:
    2,635
    Likes Received:
    72
    Have you tried reseating the heatsink, using new TIM?
    It could well be that you've cooked your thermal paste.
     
  11. outlawaol

    outlawaol Geeked since 1982

    Joined:
    18 Jul 2007
    Posts:
    1,935
    Likes Received:
    65
    Is this all stock or you OCing? If OCing run at stock speeds/defaults. Also be sure your core voltage isnt skewed real high (I think the default on a Q6600 is 1.2-1.3v) Almost sounds as if the thing is getting to much power, could even be a faulty PSU. My Q6600 dosnt get 'stinging hot' even under OC, warm at best. And my fan is only pushing 45 cfm. With the fan on does it still feel really warm?

    :)
     
    Last edited: 3 Sep 2008
  12. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    No it feels pretty cool to the touch with fan on, but I would expect that given it is pushing 190CFM! :lol:

    I thought it was getting too much juice as well when I first saw the temps, especially since I was "overjuicing it" when my overclocking on the P5N-D went badly wrong. However atm cpu-Z is reporting an average of 1.2-ish.

    Correct me if I am wrong, but the voltage is controlled by the motherboard . . . so if :worried: I was overvolting the cpu to 1.5 :) when the motherboard died the cpu would not frozen at that level correct?

    *grasps at straws* I think I'll try changing the cpu voltage in the BIOS . . . see if it un-sticks something . . . .
     
  13. SPQQKY

    SPQQKY Evil Modder

    Joined:
    7 Jul 2007
    Posts:
    509
    Likes Received:
    0
    Good question gone unanswered. Also going unanswered is whether or not you are using the stock heatsink or an aftermarket one?

    More info is needed if you truly want some help. And turning off the fan and running the cpu when you already know you are having temp issues is a really stupid idea.
     
  14. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    As I stated in my first post I am using an IFX-14 and a Delta 120mm focused flow.

    I did try re-mounting the heatsink, this had no effect, though tbh I am running out of options and am probably going to try this a 3rd time . . . :waah: Given how hot the heatsink gets though I am 99% sure that I have a good connection with the processor.

    Regarding the TIM, I am using arctic silver 5 which is supposedly superior to the Arctic Cooling MX-2 which I had been using (see here: http://benchmarkreviews.com/index.php?option=com_content&task=view&id=138&Itemid=1&limit=1&limitstart=4). While it is true I have not tried a new thermal compound, I assumed that the TIM was not the issue given the new one is supposed to be better than the old and should have improved my temps. I was not aware that one could "cook" thermal paste. I think Arctic Silver needs time to cure before it becomes fully effective, however I thought that the normal on/off cycle of the machine was enough to do this correctly?

    Changing the voltage setting in the BIOS manually to 1.2 volts has had no impact on my temps.
     
    Last edited: 7 Sep 2008
  15. klutch4891

    klutch4891 New Member

    Joined:
    4 Feb 2008
    Posts:
    693
    Likes Received:
    25
    You're right about Arctic Silver 5 needing time to cure, but I think the recommended time is something like 1-2 weeks. I could be wrong - but even if I am that only lowered my temps about 5c
     
  16. bullseye969

    bullseye969 New Member

    Joined:
    8 Sep 2008
    Posts:
    16
    Likes Received:
    0
    about 20 boot ups is the usual time taken for the paste to "settle" as it were
     
  17. outlawaol

    outlawaol Geeked since 1982

    Joined:
    18 Jul 2007
    Posts:
    1,935
    Likes Received:
    65

    I would assume the HS is a aftermarket if its pushing over 190cfm. Also, running a CPU without the fan, for testing purposes, is not a stupid idea, it'll passive cool through heat-draw (granted it will get warmer, but not metal-melting-super-hot). Now Running the CPU without a heatsink is a stupid idea. Lets get our stupid ideas clear here! ;)

    This is a an odd issue man. Perhaps it would be best to use a thermal gauge that isnt being read by the computer. Something stand alone so you can get an unbiased read from that (and controlled). I'd also say if the thing is running stably without issues, no BSOD's artifacting etc, then its thermal sensors are just screwed up. Run memtest, PI fast for a few hours and let it stress it. If it locks up and so on, then there are other issues for sure, if not. Well, sensors do lie (or are borken).

    :)
     
  18. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    OK, update time:

    Regarding the "settling" of the arctic silver my machine has been operating like this well beyond the curing times stated above, and as klutch4891 pointed out even if this did have an impact it would be somewhere in the region of 5° C.

    Yesterday I ordered some new TIM (OCZ freeze) and should have it mid-week, once I do I will clean the heatsink and processor and remount the heatsink . . . again :wallbash: . . . hopefully this will make a difference . . .

    I would like to get a temperature probe, but am a little unsure where to start, my usual online etailers do not have one, and googling has brought up a range of products which do not look up to the task . . . does anyone have any recommendations?
     
  19. leexgx

    leexgx CPC hang out zone (i Fix pcs i do )

    Joined:
    28 Jun 2006
    Posts:
    1,353
    Likes Received:
    8
    is the heat shink hot when its reporting 80c?
     
  20. DE.223

    DE.223 New Member

    Joined:
    30 Mar 2008
    Posts:
    30
    Likes Received:
    0
    When the fan is on, not in the slightest, cool to the touch. However, I would be tempted to put this down to the power of the fan which I think would easily dissipate 80° C making the heatsink cool to touch.

    If the fan is off (see above) - the heatsink gets hotter than hell - which is to be expected . . however I could swear I heard of someone passively cooling a stock q6600 with the same heatsink somewhere. . . That is why I am slightly leaning towards the "processor is actually producing massive amounts of heat" hypothesis as opposed to mis-reporting. However, against this is the fact that the machine is rock stable.

    Really need a probe! :lol:
     

Share This Page