lndm said:
The boundary layer we're talking about is air covering the entire surface of the heatsink.
The roughened surface would be worse at transferring heat out of the device bolted to it because increased roughness reduces physical contact area. Filling the gaps with goo improves heat transfer, but nothing beats metal to metal contact. Best heat transfer between the device and heatsink occurs when the heatsink and back side of the device are both polished smooth.
I_F
As usual, it's not a yes/no thing, but a matter of degree. If the roughness is not too fine for the given airflow, then a boundary layer will not form, and the turbulence will actually improve heat transfer over the laminar flow where only a small portion of the air gets in contact with the heatsink. But roughness on that scale is really far bigger than can be gotten by etching. We're talking about something like crimples, not scratches, on the surface.jackinnj said:
Naturally, there is a size premium within a PC. Where this is not an issue, a heatsink that achieves the same surface area by way of a greater bulk, will handle the same steady state dissipation (notwithstanding access to a more spread out selection of air), but it will have a greater immunity to the effect of dissipation peaks, due to the bulk.
Perhaps this suggests that etching the surface may carry more weight in a steady dissipation situation.
Perhaps this suggests that etching the surface may carry more weight in a steady dissipation situation.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.