• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Tell me if this would work...

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.


Apr 23, 2001
Indiana, USA
Ok, this seems too simple, so there must be a reason it's not done...

When I look at my P3 and see that all the surface I have to work with for heat transfer is that little half inch die square, I started thinking. A lot of heat radiates out into the material around the die. Removing heat from there seems like it would lower the temperature of the die. So...

Could you...

Hollow out a space (like with a router) in the center of your heatsink, slightly bigger than the die and as deep as the die is high. When you re-apply the heatsink, you would bring metal into contact with the material upon which the die is mounted, giving your heatsink more area from which to remove heat.

Lemme know what makes this a bad idea...

I could be wrong.... but i think this would have the same detrimental effect that shims have.... being that you would be better off creating airflow between the cpu and heatsink (nearby blower or fan) than sealing it off.... i get the impression that the board that the processor die is mounted on is not a material that conducts heat well... more of an insulator. in fact i am willing to bet it takes a while to heat up (in comparison to the processor die at any rate). a nearby fan would also cool down those gigantic capacitors and the voltage regulators that mobo manufaturers seem to love putting really close to the processor (aside from the obvious heat, i know i am not the only person that has wondered if one of those caps is going to break a lead when removing a heatink or putting one back in).
yeah, what snuffle said is right. The shim effect wouldn't be good. The best idea is a copper base to quickly draw heat away, and then to aluminum which releases heat to the air much better.
Well, I would be afraid of making that notch in the heatsink bigger than the die. If this happens you would no longer be touching the die and cooling it. This is bad! However, if we could make the notch exactly the same size as the die, maybe not quite as deep, then we would be slightly increasing the area of contact between the heatsink and die. The top and sides of the die would be touching. This might give slightly better cooling, but I do not think it would be enough of a performance boost to justify the effort of making an *exact* notch. You would probably need a computer controlled milling machine to do it.
wouldn't recommend it...

the milling is going to be bl00dy hard to do, for a start!

also, the reason "non-conductive" shims are now being marketed is because metal shims were doing extactly what you want to do, and it was causing worse temperatures. the reason: the shim provides a secondary path for motherboard heat to get into the hsf, making the mobo cooler, but overloading the hsf. you would just be doing this directly.
There was a article i read here about a fellow that used JB weld and filled in the area between the non die part of the proccesor and the HSF how he got it in is beyond me but he did say it lowered temps a little this change is permanent before i tried something like this i would invest in a better HSF [top quality] to get temps down a few degrees and if you have a nice HSF i would just let it do its job.