Thermal Paste Application Techniques

[Kenrou]

Interesting read if it holds true ?

https://www.pugetsystems.com/labs/articles/Thermal-Paste-Application-Techniques-170/

 

[[email protected]]

Its true but why does 2c matter in full load, Even if i had like 80c load i wouldnt change how i applied the thermal paste just for these 2c

 

[BlueNostromo]

Every bit counts!

 

[montaillou]

I read the article and I don't see how you can get bubbles unless the paste is globbed on. I use a credit card and spread a rice sized portion over the entire card and end up probably wiping half off leaving me with a smooth & very thin coating. I've never seen an air bubble.

Is it really time consuming for people to spread paste? What's it take 30 seconds? 60?

 

[[email protected]]

I read the article and I don't see how you can get bubbles unless the paste is globbed on. I use a credit card and spread a rice sized portion over the entire card and end up probably wiping half off leaving me with a smooth & very thin coating. I've never seen an air bubble.

Is it really time consuming for people to spread paste? What's it take 30 seconds? 60?

You can use fingers too!

 

[ATMINSIDE]

You can use fingers too!

I wouldn't... This introduces oils from your skin.

 

[knoober]

I have read a few times that coverage over the entire IHS is not important. Instead that the area over the die is the most important for heat transfer, meaning that any of these is good depending on what the die shape on your processor is like. Ive read articles on Haswell, for instance (sorry no links as it was casual reading) is okay with a line across center of the IHS, where a FX will be fine with a grain of rice method as long as the spread is centered over the die. It was an informative article, but I would have liked to see the issue of different die sizes and placements addressed.

 

[Mr.Scott]

I wouldn't... This introduces oils from your skin.

Finger in plastic bag.

 

[ATMINSIDE]

Finger in plastic bag.

Which isn't what he said

I'll just stick to not spreading paste and letting the heatsink do it... lol!

 

[Mr.Scott]

Which isn't what he said

I'll just stick to not spreading paste and letting the heatsink do it... lol!

I know what he said.
I was just mentioning the plastic bag because this is the old school method.
Sometimes you have to spread by hand. If you do not get 100% coverage on the older CPU's that take 2-4v vcore, they will cook as soon as you turn the power on. Bare die CPU's should be covered 100% also. 1 hot spot will kill it in a second.

 

[ATMINSIDE]

I know what he said.
I was just mentioning the plastic bag because this is the old school method.
Sometimes you have to spread by hand. If you do not get 100% coverage on the older CPU's that take 2-4v vcore, they will cook as soon as you turn the power on. Bare die CPU's should be covered 100% also. 1 hot spot will kill it in a second.

Oh, I agree, legacy hardware is 100% different from our modern hardware. Especially anything bare die, but it's also easier to cover a die than an IHS.

 

[[email protected]]

I wouldn't... This introduces oils from your skin.

Dont think there will be alot of oil on my hands if i wash them before i apply the thermal paste

 

[ShrimpBrime]

Oh, I agree, legacy hardware is 100% different from our modern hardware. Especially anything bare die, but it's also easier to cover a die than an IHS.

The legacy hardware is no different than current. You use the spread method when renewing thermal compound under Intel IHS. I do the same spread method with de-lidded AMD chips and block to die applications. (and have done quite a few Intels too)

 

[[email protected]]

The legacy hardware is no different than current. You use the spread method when renewing thermal compound under Intel IHS. I do the same spread method with de-lidded AMD chips and block to die applications. (and have done quite a few Intels too)

It is different, It was more expensive and it ran hotter. Correct me if im wrong .