Liquid thermal paste

[TPCEA]

2 hours ago, daygeckoart said:

Conductivity is a property of the material and doesn't vary with thickness. What I'm saying is if you want heat to move twice as fast through the paste, just make it half as thick. That is easier than using a metal conductive liquid with higher conductivity. Many people now are using too much thermal paste and ending up with a thick layer.

That's great, except it's impossible to make the TIM half as thick unless you improperly mounted the cooler/water block in the first place. You literally cannot end up with a thick layer if you mount it correctly and, even if you did, so what? As long as they're not using liquid metal or another conductive TIM, they're usually perfectly safe, although they may not like the clean-up.

[daygeckoart]

On 5/29/2024 at 8:17 AM, Dedayog said:

With coolers having a stop for tightening, how can there be too much paste?  It gets squeezed out if extra.

 

I literally can't make my paste any thinner as my cooler won't tighten any further .  If I put less on, my coverage will suffer and it won't touch the cpu and cooler well.

This isn't true. Every modern cooler has some form of pressure limiting, not position limiting. You screw them in to the stop point, and then springs or metal clips hold them in place. The thickness of the thermal paste is going to vary based on the amount of force but also the amount of thermal paste you put to begin with. Some Youtubers have created the widespread belief that the amount doesn't matter since it just gets squeezed out, but if you watch the same videos where they remove heatsinks, you see how much there is.

 

I spread a thin layer and it works perfectly for all my computers with various sockets and CPUs. There isn't an air gap!

[Dedayog]

18 minutes ago, daygeckoart said:

This isn't true. Every modern cooler has some form of pressure limiting, not position limiting. You screw them in to the stop point, and then springs or metal clips hold them in place. The thickness of the thermal paste is going to vary based on the amount of force but also the amount of thermal paste you put to begin with. Some Youtubers have created the widespread belief that the amount doesn't matter since it just gets squeezed out, but if you watch the same videos where they remove heatsinks, you see how much there is.

 

I spread a thin layer and it works perfectly for all my computers with various sockets and CPUs. There isn't an air gap!

Then by your own admission the thickness doesn't matter as all the pastes are within a few degrees of each other for the most part.  If thickness mattered so much, there would be some video somewhere of thick vs thin paste temps?   I haven't run across one so far yet.

 

[daygeckoart]

1 hour ago, Dedayog said:

Then by your own admission the thickness doesn't matter as all the pastes are within a few degrees of each other for the most part.  If thickness mattered so much, there would be some video somewhere of thick vs thin paste temps?   I haven't run across one so far yet.

 

The thickness does matter because thermal paste is an insulator so the thinner the layer, the better the heat will go through. I've seen experiments where people compare different methods, but they use too much for all the methods anyway. I haven't seen an experiment where thin application is compared to thick application.

 

The idea with Liquid Metal is that it conducts better. So the thermal resistance of the layer of paste/liquid/pad/whatever doesn't matter, why bother?