• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

does da paste degade? hotter cpu temps suddenly

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

germanjulian

Member
Joined
Apr 28, 2002
Location
Frankfurt/London
dont know but suddenly my cpu temps are 4C hotter? as well my cpu temps are warmer then when I first applied my thermal paste and heatsink, which is weird cause i use artic ceramique (spelled correctly?).
and got an slk 947U heatsink
mother_slk947u_amd8.JPG


any ideals? why would it get hotter?

can i also take the heatsink off somehow without taking my mobo out because of the screws? and the back plate which is behind the mobo?
I think I can as long as I make sure the screws without the springs dont unscrew. right? dont remeber how the think works :)

thanks for the help
 
you shouldnt need to reapply in between cpu changes, some people change the paste regularly, but I recall a post not too long ago that said changing the paste with the ceramique and AS5 shouldnt really be needed.

[from old thread]:
There is no need to reapply AS5 unless the thermal interface has been disturbed by physically bumping the heatsink or rough transport.

Colin Thompson
Arctic Silver, Inc.
 
If anything I notice my temps getting lower after I leave the paste on for a few weeks. I would think the longer you leave it on the better it will work, But at an exponentially more difficult rate, it will be 70% effeciant when you put it on, the next day it will be 85%, the next day 90%, the next day 91%, the next day 91.5% and excetera (I made up the numbers)

I think you might have dust building up on the HS, try blowing your system out with an air compresser.
 
Yeah, just these past few days my cpu temps increased 4C too. I also used ceramique on a SP-97. I applied it almost 3 months ago, I guess I'll reapply again when I have some time.
 
Maybe its because you guys are running the heat all day and making your room temps higher than they are in the summer.
 
craig588 said:
Maybe its because you guys are running the heat all day and making your room temps higher than they are in the summer.

That and maybe there's some Shmootz(dirt and dust) clogging up the heatsinks vanes. Maybe shoot some compressed air in the heatsink.
 
My SP97 + Panaflo H1B just did the same thing today. The build is only about 3 days old. At first, at this speed I'd be at about 39/45-46, but now I'm at 41/50ish. No clue what the deal is. Cant be dust or anything....its like new! Also using ceramique.
 
germanjulian said:
room temps. i really need a thermometer!

i clean my comp this weekend. see how it goes.

You really need to know what the room temp is to get an idea of what your cpu temp should be. That should be a fairly constant thing. Here's an example. If your room temp is like 22C and your cpu temp is 42C. Then your room temp rises 4C, your cpu temp will most likely rise 4C also.


Maybe its not the thermal compound for those that are using the SP97. Maybe its the heatpipes. Maybe there is something wrong with the fluid inside or the pipes are defective. Just ideas.
 
Last edited:
Okay, I just reapplied the ceramique and the temp increase is still there. Room temps have stayed the same and dust is not a problem. The only thing I can think of is humidity, it just began to rain again. Also, I leave the computer on 24/7/365.

Before Idle:
Socket 25C
Diode 20C
Case 17C

Now Idle:
Socket 25C
Diode 25C
Case 17C

100% CPU Usage still increases diode temps by 5C so I think the HS is working properly. Well, I guess approaching 30C for diode at full load won't harm a stock 2500+.
 
after you apply fresh TIM, you need to run it at load, turn it off and let it cool completely, and repeat several times. AS needs a thermal cycling period to reach its optimal transfer capability.
 
Trent said:
after you apply fresh TIM, you need to run it at load, turn it off and let it cool completely, and repeat several times. AS needs a thermal cycling period to reach its optimal transfer capability.

Thermal Cycling doesn't decrease my temps but I do it anyways :D
 
rxc said:
Okay, I just reapplied the ceramique and the temp increase is still there. Room temps have stayed the same and dust is not a problem. The only thing I can think of is humidity, it just began to rain again. Also, I leave the computer on 24/7/365.

Before Idle:
Socket 25C
Diode 20C
Case 17C

Now Idle:
Socket 25C
Diode 25C
Case 17C

100% CPU Usage still increases diode temps by 5C so I think the HS is working properly. Well, I guess approaching 30C for diode at full load won't harm a stock 2500+.

4bidden I don't know how to check the pipes it was just a thought. If they look OK and there are no holes in them then they are probably fine.

rxc there is no way your CPU diode temp can be lower then your socket temp something is wrong there. The socket temp is taken under the CPU externally, the diode is inside the cpu which is hotter usually by around 10C.
Also unless your using water cooling I find hard to believe your cpu diode temp is only 3C higher then case temp. I know your running it stock, but 3C thats a little bit hard to believe. Especially when the diode temp is lower then the socket temp. Who are you David Copperfield?
 
Diode lower than socket is possible. I believe that the socket being underneath the cpu has no airflow whatsoever so it becomes like a mini oven there with the cpu generating heat above it whereas the cpu has the heatsink to cool it off.
 
rxc said:
Diode lower than socket is possible. I believe that the socket being underneath the cpu has no airflow whatsoever so it becomes like a mini oven there with the cpu generating heat above it whereas the cpu has the heatsink to cool it off.

Sorry but its not possible. The source of the heat can't be at a lower temp then the air around it. Diode temp is usually around 5-10C higher then socket temp. The heat is transfered to the heatsink and the surrounding air. They can't be hotter then the source. That would be like having the air in an oven hotter then the flame or coils that created the heat.

Here try some reading at this link

http://www.vanshardware.com/articles/2001/december/011209_AS_Interview/011209_AS_Interview.htm



Scroll down until you see this question;


"VHJ: Even if what you say is true, can’t an external thermistor still be accurate, but low? For example, if a core is running 15 degrees Celsius hotter than the thermistor reports, can’t we assume that a 3 degree drop in reported temp equals a 3 degree drop in actual temp?"
 
Last edited:
Good article.

"current motherboards add 10°C to 15°C to the measurements they receive from the thermistor" can't that not make the socket higher than diode ;)

I think I'll just stop using socket temps.
 
rxc said:
Good article.

"current motherboards add 10°C to 15°C to the measurements they receive from the thermistor" can't that not make the socket higher than diode ;)

I think I'll just stop using socket temps.

The thermistor is in the socket. You have it backwards. The thermistor is in the socket and the diode is in the cpu. Because the thermistor doesn't read high enough the actual temp of the cpu can be around 10C to 15C higher. Higher....higher...and higher. The diode inside is more accurate and will always be higher then the thermistor so your readings like I said are wrong.
 
i seem to be having this prob also, the last 2 days i have seen my temp go 2 C higher. this is when im not running anything either.

i have as5 applied
 
Back