Q9550 High Idle Temps

So I just build my first PC, and I had some problems at first with the hardware but afterward I finally settled install Windows XP and decided to see what my processor can do, so I decided first to check the settings... the settings at idle are pretty high. At first they were from 48-53 and I was already kinda weird-ed out so I came reading lots and lots of forums to find out what was the problem and some said it was the thermal grease so I actually went and disconnected the Zalman CNPS9700 110mm (which was a pain and I cleared the bottom of it and the top of the Q9550 very well and applied a small coating (maybe I need to even less?) and the temperature had changed to 43 43 37 41 which was a big changed I guess... So yeah not much to worry about regular but I was planning on OC-ing. Through a RealTemp test they blasted up to 60s in those like 3 secs, before removing the Thermal Grease I tried Prime95 and they sky rocketed so I soon stopped it.I've tried many temp reading programs and they are all about the same no real difference if any at all. I would like to go down at stabled 3X idle, please help.

I've read everywhere but so many forums and I've found no answer, and hoping this will quicken my search for a solution.



Specs:

Mobo:ASUS P5Q Pro

CPU FAN: ZALMAN CNPS9700

CPU: Q9550

Case: COOLER MASTER Centurion 5

PSU: CORSAIR CMPSU-550VX 550W ATX12V

Any extra information needed just ask.
24 answers Last reply
More about q9550 high idle temps
  1. Prime95 is a stress test to display the peak temps. The temps are supposed to go up. How high did the temps get on prime95?
  2. If Prime95 pushes your CPU into the 60s at stock speeds, the HSF probably isn't on right or you've used too much thermal paste (blob in the middle, size of a rice grain is enough, the HSF pressure spreads it evenly and thinly).

    aford10's right, Prime's going to make your temps skyrocket very quickly, a CPU going from idle to 100% activity across all cores is going to get hot. Think of it like a car, start it off, engine's pretty cool. Rev it to the redline for a minute and it's going to get hot very fast.
  3. Umm I understand 60s but it was more like 80s... HSF is on fine, trust me it helps fine. I might have used too much as I just cleaned it a bit more and the temps settled down more.

    However I didn't really put drops it was like a nail paint just a brush. I took out brush cleaned it a lil in the top of bottle and just painted the CPU grey on top.

    Edit: I just ran Prime95 again now and they actually took seconds increasing little by little not just bam 80... it still got to 80 thogh and beyond as timed passed.
  4. note to all

    let your system burn in bios for the thermal compound to even out! 24-72 hours

    do not go right to 100% testing
  5. blade2577 said:
    Umm I understand 60s but it was more like 80s... HSF is on fine, trust me it helps fine. I might have used too much as I just cleaned it a bit more and the temps settled down more.

    However I didn't really put drops it was like a nail paint just a brush. I took out brush cleaned it a lil in the top of bottle and just painted the CPU grey on top.



    what case?

    is the air flow right?

    is the hsf pointing the right dirction?

    you sure you have enought compound?

    did you allow time for it to settle?

    did you use the bb - method? dumb! takes more time to spread


    need more details!
  6. dragonsprayer said:
    note to all

    let your system burn in bios for the thermal compound to even out! 24-72 hours

    do not go right to 100% testing


    Thanks I'll take a note of that. So before checking the temperatures I should just not let the computer do lot of workload so the Themal paste evens out? I guess if I have the same problem in 3 or 4 days I'll re-ask. I was just concerned for it being so irregularly high.
  7. dragonsprayer said:
    what case?

    is the air flow right?

    is the hsf pointing the right dirction?

    you sure you have enought compound?

    did you allow time for it to settle?

    did you use the bb - method? dumb! takes more time to spread


    need more details!


    Air flow is alright... the HSF is... Hmm cleaning it lowered the temperature so I'm guessing I do, or I should just apply again because I basically took it out.

    Uhh Well I didn't just place it together and made it run, not the first time at least but I cleaned it and put some again and then I just cleaned it again. Basically right now it's all kinda gone.

    Edit: Oh and the case is in the first post.
  8. You haven't OC'd yet have you?


    Those temps, on stock speeds, are a little high in prime95. I really don't like seeing a cpu go over 70C. You may have a defective fan. Are you using the mobo backplate for your zalman?
  9. Nope not at all, don't even want to looking at those temps. Well in Prime95 the temperature just kept increasing and increasing til it got to 90 something and I just stopped the test. Again I doubt the Fan is the defective one I mean I haven't even put it all the way up with Fan Mate 2, and without it the CPU was pretty high so I'd say its doing it's job. I wonder whether to re-apply the thermal grease since I basically took it off last time I cleaned it, then just not making the PC do any workload for a day.
  10. The Zalman has an adjustable fan. Do you have it turned all the way up?
  11. Normally I don't, not even 1/4th but I did for a while(meaning 30 mins at least now) now - all blast, spinning really fast making a loud noise, and no difference in temps.
  12. That tells me the fan may not be installed correctly or is defective. It should make a difference.
  13. Well can I make sure it's defective. I might have installed it incorrectly, however as long as it's on top of the CPU does it make a difference? it's PLACED correctly at least, facing the right direction, idk. AND it does get faster with FanMate 2. If it's defective how would I know?
  14. Look carefully at the fan manual. It will show which direction the fan should be pointed on top of the cpu. Then make sure it's secured properly with the mobo backplate. It should be snug on top of the cpu. It shouldn't wiggle at all.

    It sounds like you applied the correct amount of thermal paste. That's why I think it's the fan.
  15. You could also put the stock fan back on and see what the temps are with that.
  16. Well the thing is the fan mentioned two ways of putting it and I just put it the way it made sense to me but when I see it online it's put reversed. Though there are like 4 different ways to put it.

    Using the stock fan, hmm genius idea.
  17. though honestly I hate that fan... to get the push pins in right was a pain.
  18. Funny enough, the airflow was indeed backwards, should have payed more attention to the manual though honestly that did pretty much nothing for me, but I'm getting there, 35-35-30-35. is there a picture on how much thermal paste should I actually put, I dont want to fix the paste without knowing how much exactly.

    Maybe I really don't have enough thermal grease, I've been googling and all I see are plastered CPUs, I just have a light coating... If I take a pic, can anyone tell me if I put enough?

    Also that Zalman has cut me so many times it's not funny. If it ends up not working is because the motherboard is going to be plastered with blood.
  19. Lol, I've got a similar zalman and know exactly what you mean. The fins are sharp as $%^&. I used gloves when installing it.

    You should use a drop of thermal paste about the size of a grain of rice. You can spread it around or use the cooler to do it. Too much and it can overflow down into the socket pins (bad!).

    The pushpins on the stock fans are garbage. But it may give a better idea of if there's a problem with the zalman.
  20. Well I had to sleep on it and well the temps are the same when I woke up.

    Like I said Idk how much a drop really would be. It's not really liquidy so it won't overflow but if it's a grain of rice then the small coating I have now should suffix.

    Even if the temperatures are a lil' lower now I still worry of the Prime95 tests as the temperatures just keeps on increasing and increasing to like the max if you wait long enough.

    I'll try stock cooler and see if it continues.
  21. Ok so I cleaned the CPU again of the thermal paste and put the stock fan (with pre applied thermal paste on top of it, I know it's making connection as it left thermal paste marks on the CPU, Regardless of all that the CPU is still running at the same temps it was running with Zalman, really no difference. Tried Prime95 tests and they keep increasing til 75 celcius then it doesn't change from 77 or rather I did not want to wait til it changed to higher... regardless I'm supposed to be getting only 60C with Prime95 at most.
  22. Well I put back Zalman this time I read the manual without looking at the pictures. Sometimes you just need to use logic. So placed it what I believe to be right this time. and my temperatures only in RealTemp when I set the TJMAX to 95 are 31,31,30,36. I don't know why that last core is 36 but it doesn't bother me too much, Should it? In other programs (SpeedFan, Everest, Core temp, just add about 5 degrees and it's the temperature. Which should I trust? (I haven't found an option to set the TJMax to 95 in any but RealTemp) In the bios the CPU came from 40 degrees to just 32-34 so I'm guessing it's good. and Tests in Prime 95 quickly rise to 55 but then slowly gets higher after 3 mins only in 60-59 or so, I know It'll keep increasing but that's good results. Just in case I'll go to RadioShack and apply Artic Silver 5 instead. Maybe help regulate some more. What I do seem worried about is under load with prime95 the core temps are too different. Is that a problem? I got 60-56-51-50
  23. If you are peaking at 60c during prime95, you should be fine. It still seems a little high, but within the accepted range. You could look into adding a fan to your case, it may help the airflow and overall cooling.
  24. I've always used tjmax of 100 for my q9550. Loads out between 60 and 65 at 3.6ghz with a V8 cooler with p95. (Use linpack if you really want to toast the thing :) )

    I've read that the tjmax is 95.. but frankly I don't know, I've seem many conflicting oppinions.. I'd rather error and be assuming the temps are 5higher than they actually are then OC assuming they are 5 lower. (Let me know if someone is sure of the tjmax actually.. I'm curious again :D )

    On topic, the cores will have different temperatues for various reasons. a) you dont ahve even thermal contact to teh heat spreader, b) one of teh sensors is stuck c) It is jsut the way your chip is. They will almost never be exactly the same. Generally there will always be a +/- 5 difference across them at all times.

    -Cheers
Ask a new question

Read More

CPUs