Leaving Thermal Paste on Cpu over night without putting on fan Help!

TehRoxxo

Honorable
Mar 12, 2013
77
0
10,630
So due to some personal problems I had to stop putting the fan on and I still had the thermal paste on the cpu and I cant come back and finish putting the aftermarket cooling fan on atleast for two days and I was wondering If it would dry up and i would have to clean it and reapply or if it was fine?
 

zyky

Distinguished
Sep 12, 2006
174
0
18,710
With most pastes you should be fine as long as you didn't leave it exposed to collect dust/etc. Sounds like you probably didn't have time to take those precautions so I'd redo it.
 
if you have spread the paste thats a bad method of applying thermal paste to begin with, if you left it in the middle as a pea sized blob just put on your heatsink and it should be fine.

if you did spread the paste i would clean it off and start over with just a pea blob of paste onto the centre of the cpu die and let it spread itself when you attach the heat sink.
 

zyky

Distinguished
Sep 12, 2006
174
0
18,710


Myth perpetuated by people bad at applying thermal paste.

 


It is not a myth that spreading thermal compound creates air bubbles in the thermal compound when applying the heat sink, you can even look for videos or do tests yourself.

i even find visible temperature differences when using the 2 different methods, lower temps with the pea in the middle method and lower load temperatures because there are no air bubbles, do some experiments yourself.

here is a good example video
http://www.youtube.com/watch?v=EyXLu1Ms-q4
 


Oh, my. I've been doing it wrong for over 25 years.

Thankfully, I haven't lost a CPU...



 

cmi86

Distinguished


OH NO !! I just did that on my CPU and my 4.5Ghz OC max's at 45c, is that bad ? << sarcasm, kinda hard to tell over the web lol.
 

JimF_35

Distinguished


Speaking from personal experiance I have noticed a 3-5 degree change by not spreading the TIM than spreading. I know not much but if you are trying to eek out more over head. The other problem with spreading it is getting an even layer on your IHS. I have noticed that if you don't get it right one or two cores might see higher temps then the others. So will spreading hurt anything, probably not but if you are OCD about your temps it is better not to spread.


 

cmi86

Distinguished
The point that really needs to be made is there are different application methods for different styles of CPU cooler. stock coolers or zalman cnps any other cooler with a completely flat surface on the bottom the small blob in the center is a fine method. However when installing coolers with direct heatpipe contact such as the hyper 212+ or Evo the spread method needs to be used because there are small gaps between the heat pipes and the base of the cooler that need to be filled in otherwise the cooling system will not transfer heat properly. It all depends on what type of cooler and nothing else.
 

JimF_35

Distinguished
True but, I have used the line method on direct pipe coolers and seen lower temps than spreading. With direct pipe you have to put a line on each pipe. Then when you push it down the TIM squishes into the cracks forcing the air out. Spread can still cause air pockets with direct pipe because if you dont get and even spread and push the heat sink straight down, you can trap an air pocket.

P.S.

I should have added to my other post that there are times when the IHS is not level from the manufactuar which can cause uneven heat distribution. I you find that your IHS is eaither Concaved or Convexed you can either lap your IHS or take it back for an RMA. I don't recommend lapping because it will void your warenty. If you bought your CPU on eBay and want to lap it, then I recommend you look up CPU lapping on Youtube. Many videos out ther on lapping. You can also lap your coollers which I almost always do because I buy used coolers which can be scratched to h3ll.

The whole point behind TIM is to fill in scratches and uneveness to increase your surface area. Once you look at it from this point of view it gives you a better understanding on how to apply it.
 

hdeezie80

Honorable
Jul 18, 2012
205
0
10,710


Yeah this is true I used the spread method until I watched a video showing how different applications settled under a plexiglass sheet, spreading caused air bubbles to form on every test. Now I just do a pea sized dab directly in the center, for socket 2011 I put 2 thin lines in an X after trying the dab and line method and got the best results.
 

zyky

Distinguished
Sep 12, 2006
174
0
18,710


If you get better temps using the pea method, then you're one of the people who don't know how to apply thermal interface material, and you should stick to that method.

The ideal amount of TIM with perfectly flat non porous materials is none. Your thermal paste is only meant to fill these micro gaps in what aren't perfectly smooth surfaces.

With the pea method, most of your contact will be 0, and way too much in the center.

When people who don't know what they're doing apply thermal compound, they get way too much everywhere.

a very very very VERY tiny amount spread evenly, just to fill these pores, is what you want.
 
PhII945_wTIM.jpg


This is actually a bit too much but something similar will work just dandy.

Heat shields are concave in nature. The TIM fills in the ""cave""

P4210126.jpg


You can see where the perimeter of the heat shield was literally embedded in the copper with the TIM 'fill' Interestingly enough, the area of missing paste in the center of the copper represents the location of the actual CPU die underneath the heat shield .... and the 'missing' TIM paste itself remained on the processor.