I know adhesive pads tend to be the standard for anything free mounted or lacking pressure holding the headsink down but i cant help but try and get some extra thermal conductivity from a proper thremal grease if maybe its viscous enough or cures hard to keep things in place.
is there a de-facto brand i should be considering in this scenario or is something like the fujipoly pads my best bet?
Heat tranfer ability of different things:
Copper is 400 W/(m·K) heat transfer
Aluminum 210 W/(m·K))
Lead is 35 W/(m·K)
Typical TIM is rated 3.5-10.5 W/(m-K) (GC Extreme 8.5W/(m·K), Chill Factor III 3.5W(m-K) in use this is only 1-2c difference in CPU temp)*
Air is 0.024 W/(m·K)
To put that into perspective, any way you look at it
Metal to metal is 50 times better than TIM:
Copper is 47 times better than Gelid GC Extreme TIM & 114 times better than Chill Factor III
TIM is 333 times better than air at sea level.
generally if there is any type of clamping force available for a single chip like cpu/gpu then liquid ultra is my TIM of choice.
theres K5-PRO for areas where traditionally pads have been utilized to sop up the gap in manufacturing tolerances when spreading a single heat sink surface across multiple chips.
but if there is no clamping force available, and you do have a perfectly flat surface of a single chip available, say in a mosfet or pwm.. i'm curious if just the surface tension that a more viscous thermal paste can provide will offer superior performance vs a thermal adhesive.
hrmm arctic alumina looks interesting, although i'd like to avoid something so permanent if possible.
i think i actually did remove some heatsinks utilizing this stuff once with a razor blade and a lot of patience.
i guess they wouldn't make the stuff unless it was superior to say just their regular arctic silver 5 in some scenarios.
unless its intended for chips mounted on a vertical vs horizontal plane.
going to look into application instructions more clearly and see if any of the conventional pastes specify clamping force is required for proper interfacing.
hrmm arctic alumina looks interesting, although i'd like to avoid something so permanent if possible.
i think i actually did remove some heatsinks utilizing this stuff once with a razor blade and a lot of patience.
i guess they wouldn't make the stuff unless it was superior to say just their regular arctic silver 5 in some scenarios.
unless its intended for chips mounted on a vertical vs horizontal plane.
going to look into application instructions more clearly and see if any of the conventional pastes specify clamping force is required for proper interfacing.
I've used it to attach heatsinks to VRMs on a motherboard before, only reason I mentioned it. It worked very well for me, dropping CPU and socket temps enough on my 1090t to let me get another ~200MHz out of it, and yes it is quite permanent. Pretty sure if I tried to pull those heatsinks off the VRMs would come with it
came across that little gem that makes me think that even with some good initial pressure, once released the surface tension wont be enough to prevent a thicker layer of the TIM than is conducive to conventional performance tolerances.
or worse yet, introduction of air pockets.
so looks like k-5 pro which is advertised K>5,3 W/m.K but i'm unsure how that translates if there is a thicker than usual layer (but thick layers are understood as part of this products specialty)
the arctic alumina and other thermal epoxies sound good at about >4 W/(m-k) and might benefit from some clamping force applied during the initial curing period prior to activating device service.
ultimately i think a 1/4 sheet of 0.5mm fujupoly thermal pad which ranges from 11.0watt/mk to 17.0watt/mk listed spec seems the ideal solution to have around for this use scenario.
can anyone comment on the online store below and whether a better price can be sourced elsewhere?
came across that little gem that makes me think that even with some good initial pressure, once released the surface tension wont be enough to prevent a thicker layer of the TIM than is conducive to conventional performance tolerances.
or worse yet, introduction of air pockets.
so looks like k-5 pro which is advertised K>5,3 W/m.K but i'm unsure how that translates if there is a thicker than usual layer (but thick layers are understood as part of this products specialty)
the arctic alumina and other thermal epoxies sound good at about >4 W/(m-k) and might benefit from some clamping force applied during the initial curing period prior to activating device service.
ultimately i think a 1/4 sheet of 0.5mm fujupoly thermal pad which ranges from 11.0watt/mk to 17.0watt/mk listed spec seems the ideal solution to have around for this use scenario.
can anyone comment on the online store below and whether a better price can be sourced elsewhere?
I too, have wondered about CPU TIM tests showing data of low to higher mounting pressure. As we know, TIM is not even close in heat transfering ability as direct metal to metal, .. and the more pressure applied the more it pushes the TIM out and the closer the two metal surface become. So of course higher pressure gives better heat transfer.
The thing is I don't think it's because of the pressure but because of more metal to metal contact. I'm betting if they backed the pressure off after the TIM was forced out from between the cooler and IHS the temps would remain the same lower temperature.
I know this is hijacking but ...
CPU heat source is much smaller than IHS. The IHS serves 2 purposes; 1/ to transfer heat from CPU chip to cooler, 2/ to supply a base for cooler to set on. My experience is (dependent on size of CPU chip) a dob of TIM the size of a grain of rice is plenty. This spread into a circle that is close to sides of IHS .. well beyond the edges of CPU chip. Other testing results back up this. of course bigger CPU chips like Nehalem, Sandybridge and even bigger AMD need bigger print areas to match chips. Here is link to CPU TIM application and seating guide.
If you have any thoughts about this please start a thread and we can all discuss it.
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
Ask a question
Ask a question
Overclock.net
27.8M posts
541.2K members
Since 2004
A forum community dedicated to overclocking enthusiasts and testing the limits of computing. Come join the discussion about computing, builds, collections, displays, models, styles, scales, specifications, reviews, accessories, classifieds, and more!