Surface area of heatsink = Performance?

hawkeye177 · December 6th, 2001, 12:43 AM

Today in science class we were talking about surface area of rocks and i was thinking that the more surface area the more air can touch the metal. Do you think that surface area effects the performance?

storm2k · December 6th, 2001, 12:46 AM

i'm not sure how you mean. basically, the more surface area of a chip you expose to the heatsink, the more heat can be drawn off, not the other way around. a good heatsink should make even contact with the entier chip to ensure maximum cooling.

KenKun · December 6th, 2001, 12:46 AM

Yes i think the surface area affect, i have seen it and tried it with bigger HS ... thats why the number of fins actually affect the over all performance ...

And also the surface mast be veeery smooth for better contact with the CPU, other point ...

Regards

crystaldragon · December 6th, 2001, 01:31 AM

Right on target. Thats why all the fins

to increase the radiation area (surface) When you add air movement you increase the transfer away from the source

Hellmund · December 6th, 2001, 04:15 AM

Well I'd say it's more a combination of things, primarily the conductance of heat is most important, then having a large surface area helps to conduct more heat simultaneously and also combined with active cooling help to remove it faster, it's a combination.

hawkeye177 · December 6th, 2001, 07:50 AM

If you gain perfomance then why don't overclockers cross cut their heatsink? Like this
original
----------------
----------------
----------------
Cut like this
+++++++++
+++++++++
+++++++++

Spartan · December 6th, 2001, 07:55 AM

I'd guess that the cross cut might restrict airflow. I think that would counteract any potential benefit gained by the increased surface area.

KenKun · December 6th, 2001, 08:04 AM

I think IMO tthat it might cost more on production .. then it about seems the same to me .. even if you cross cut it ..

<edit> also how about rounded columns HS .. it also gives good results ..

gwinters · December 6th, 2001, 08:17 AM

Surface area does determine how fast heat will be transferred between two systems. The amount of heat that can be transferred into the surrounding air, for example, will increase with a greater amount of air touching the heatsink. There are two extremes though. If your heatsink has lots of surface area but little mass then you will not draw heat from the CPU fast enough. If your heatsink is larger than your CPU then there will be areas of it that do not get hot because the heat will dissipate before it gets there.

Air conducts heat very slowly in comparison to the CPU to heatsink. That's why there is a fan on top of the heatsink. It increases the volume of air that touches the surface of the metal by pushing away air that has been saturated with heat and replacing it with cooler air that will more readily take heat from the heatsink.

I'm sure that there are physics websites that can explain this better than I can, and they probably have formulas to back it up.

That's my 2.5 cents.

Graham · December 6th, 2001, 08:59 AM

I thingk gwinters has it about right, yes a larger surface area will improve the HS/fan assemblys ability to move heat, but only up to a limit.
If your HS is only warm to the touch, then you probably have a large enough area.

G