How to properly apply Thermal Paste on Core 2 Duo?

[rawr1234]

Hi guys, Im an AMD user who recently switched to Core 2 Duo and I noticed that the Core 2 chips dont have a die to put Artic Silver on ... so what do you guys think is the best way to apply thermal compound to the chip? spread a thin layer all over the surface, a little dot like AMD chips or some other method? which cools best?

 

[rawr1234]

I typed in "applying thermal paste to core 2 duo" in google and came across a pdf that had this from artic silvers website

picture

heres the pdf

http://www.arcticsilver.com/pdf/appinstruct/as5/ins_as5_intel_dual_wcap.pdf

has anyone tried this method?

 

[John]

Since the C2D has a heatspreader, I don't see the benefit of an elongated thin line application vs. to a "bb size in the center" since the weight of the heatsink will spread it out evenly and over the entire core anyhow.

This is how I apply thermal compound to all of my C2D, s754, s939, and other cpu's with heatspreaders

I also use Goof Off or Prestone brake cleaner to remove old grease and pads

 

[Atty]

Rawr - I used the dot in the middle technique suggested by Arctic Silver and am getting 34c idle 40c load temps with my CNPS9500.

E6600 (Stock Speed)

 

[Capt Caveman]

No matter what the cpu, I've always gone with a thin line and then spread it out with a credit card. As has been proven by many Intel users, the headspreader isn't always completely flat. With the dot in the middle technique, I believe you leave open the chance of not having the thermal paste to completely spread.

One minute of extra work, is worth it imho.

 

[rawr1234]

doesnt the spreading method actually increase temps? which method cools best?

 

[Ayah]

I've always used the dot technique.

 

[cmdrdredd]

Originally posted by: Capt Caveman
No matter what the cpu, I've always gone with a thin line and then spread it out with a credit card. As has been proven by many Intel users, the headspreader isn't always completely flat. With the dot in the middle technique, I believe you leave open the chance of not having the thermal paste to completely spread.

One minute of extra work, is worth it imho.

No, it's the opposite. If you have a non-flat IHS and spread your TIM then your HSF will only touch part of the IHS and leave an air gap where it doesn't make contact. So if you mash the HSF onto a dot of TIM you eliminate this gap and hit the core flat in the middle underneath the IHS.

 

[Capt Caveman]

I disagree. Spreading the AS5 will fill in any gaps. Leaving a thin layer over those gaps will allow for complete contact over the complete IHS w/ the HSF. However, I don't doubt the dot on the center of the IHS method can work also.

 

[Continuity28]

Originally posted by: John
Since the C2D has a heatspreader, I don't see the benefit of an elongated thin line application vs. to a "bb size in the center" since the weight of the heatsink will spread it out evenly and over the entire core anyhow.

It will spread into a circle that may not cover the entire surface above the core. In this case, since the core is twice as long as normal, an oval would be better suited... hence the line vs dot.

The whole purpose of having a dot method was to match the shape of the core without using excess paste, and to fill the gaps of the metal without creating a thick couterproductive cake layer. Now obviously when the shape of the core is completely different, you alter your method a little. You're still accomplishing the same overall goal, just changing the spread pattern to match your core. A line the same width as your dot will just spread into an oval the same width as your circle, but longer to match the longer core. What's the problem? If my core was shaped in a huge "L", I would change the grease pattern a bit as well. Ideally you want a small amount right above the center of the thermal area... in this case the center is a line instead of a single point, because of two cores side by side that put out equal heat.