Rotating-Heatsink

This looks like a good idea. I'm not totally sold on the self-cleaning aspect of it, but I'll wait and see.

I have to think that the thermal conductivity of the component/heatsink junction is absolutely horrible. I cannot believe this will be able to move a lot of heat.

Yes. It is relying on convection rather conduction as the initial step in heat transfer, with air no less. Even though it would have a high Reynolds number, it still can't compete with conduction. I cannot believe it either.

I don't have the exact numbers or articles at my fingertips but I remember a previous particle or two where lab tests showed heat transfer using a very thin air layer worked surprisingly well if the layer of air was kept very small, so I think it is theoretically possible and lab bench proven (in concept).

The problem I have (as I have already said), is its tiny heat transfer air boundary between a fixed heat-generating object and the rotating heatsink. Any wobble, imbalance or movement parallel to the spinning heatsink is going to cause the rotating and fixed heat sinks to collide together creating friction, noise, increased motor load, uneven fan speed and in the end premature failure.

It would seem to me that this is the biggest hurdle for long term reliability in a real world product. I would be interested in how they intend to overcome this economically to make it a viable alternative to the proven fan+heatsink, heatsink on its own and heatsink+water combinations.

Interesting premise that a small, thin air gap can transfer heat better than expected. I'd love to see the specifics. I agree though that relying on a thin gap requires some very tight mechanical specs that makes me wonder about support and deflection complications.

I do like the idea that the fan's driving fins becomes the heatsink fins, too. I wonder if a fluid bearing could do the multiple duties of providing sufficient mechanical support, freedom of motion and thermal coupling?

Nanotechnology-based heat sinks with their greatly-increased surface area and no moving parts seem to offer a better alternative (assuming the tiny little carbon forests can be grown economically).

Thin Air within fins or?

I would like to see these articles...

I remember a previous particle or two ...

Up? Down? Strange? Charmed, I'm sure!

Seriously, I would like to see the articles you speak of. Perhaps there is a region of extremely high N_Re that creates heretofore untold quantum heat transfer effects.

Here are a few examples in the relevant fields. Mostly they are related to the use of nanotechnology (either by itself as a coating or heatsink or used in conjunction with other materials). Some are less relevant (found by a search and included as examples) and some I couldn't find again (they were within the last 6 months to a year or so).

So, in no particular order

http://www.physorg.com/news/2011-03-microscale-benefit-military.html

http://www.physorg.com/news195306544.html

http://www.physorg.com/news/2011-04-advance-microchannel-industry-applications.html

http://www.physorg.com/news195306544.html

http://www.physorg.com/news110479667.html

http://www.physorg.com/news133712899.html

http://www.physorg.com/news128797959.html

etc.

There is a strong push at the moment in heat transfer technology, and this is primarily due to the increased attention nanotechnology is getting as (of all things) a surprisingly useful heat sink component. No laws broken but a few are bent a little.

Can't wait. Hopefully with increased cooling they can increase speeds as well.

Sigh, cooling capability does not drive clock speed. Clock speed drives heat load which drives cooling requirements.

yep, and worse the heat is an exponential curve with increased clock speed for MOS devices.