I've been thinking recently about the way people apply TIM or thermal paste or glupe.It sorta confuses me why people put it down the way they do and why. Line, rice, cross or whatever else, they just sound stupid. Forgive me if I'm being incredibly stupid or overlooking something.
Also when I say stuff, take that it is always the same TIM (ie; brand and normal one like, MX4 or TH1) and a normal cooler (water loop, or air).
Whenever i see someone putting TIM down whatever the brand, their justification makes no sense. Scientifically the thinest layer of TIM with the most even application should produce the best results, providing your using a conventional cooler.
TIM's purpose is to give an even contact between heat sink and processor and conduct heat, more efficiently than a metal to metal contact, by making sure the surfaces have more surface area.
Wouldn't this mean the spread method is the most efficient? (Providing that both surfaces, are reasonably flat)
In real life of course things don't work like this, because of stuff like mounting pressure and TIM de-grading. Still a big splodge of TIM which hasn't been spread and is quite thick, can not be very good. Doesn't more TIM (thicker) mean a less efficient conduction of heat.
I'm pretty sure someone (quiet correctly) will tell me that few cold-plates are flat and that heat produced on a chip or processor, isn't distributed evenly.
If the cold-plate on a cooler or water block is convex or concave, the Tim should create a flat contact so more surface area of processor or cold-plate is utilised. Therefore giving a better conduction of heat.
Even if heat distribution isn't even each part of heat-spreader will give more heat conduction so why not use it?
Also when I say stuff, take that it is always the same TIM (ie; brand and normal one like, MX4 or TH1) and a normal cooler (water loop, or air).
Whenever i see someone putting TIM down whatever the brand, their justification makes no sense. Scientifically the thinest layer of TIM with the most even application should produce the best results, providing your using a conventional cooler.
TIM's purpose is to give an even contact between heat sink and processor and conduct heat, more efficiently than a metal to metal contact, by making sure the surfaces have more surface area.
Wouldn't this mean the spread method is the most efficient? (Providing that both surfaces, are reasonably flat)
In real life of course things don't work like this, because of stuff like mounting pressure and TIM de-grading. Still a big splodge of TIM which hasn't been spread and is quite thick, can not be very good. Doesn't more TIM (thicker) mean a less efficient conduction of heat.
I'm pretty sure someone (quiet correctly) will tell me that few cold-plates are flat and that heat produced on a chip or processor, isn't distributed evenly.
If the cold-plate on a cooler or water block is convex or concave, the Tim should create a flat contact so more surface area of processor or cold-plate is utilised. Therefore giving a better conduction of heat.
Even if heat distribution isn't even each part of heat-spreader will give more heat conduction so why not use it?
