Sandia Cooler: no fan, just spin the heatsink

Why have a fan on your heatsink when you can just spin the heatsink?

https://ip.sandia.gov/technology.do/techID=66

www.youtube.com/watch?v=uGpV_VPUn8g

Claims are it's 30 times more efficient than your typical heatsink and fan setup, which probably means stock coolers, not the high-end stuff we talk about here. But I still look forward to HardOCP's review.

Edit: and it turns out to be last year's story, so this is probably less exciting than I thought.

Remember seeing this before! Seems all kinds of silly...and having less contact between the base and the fins. Buut if you have to replace the whole HSF everytime a fan dies = $$$.

DeathPrincess said:

Remember seeing this before! Seems all kinds of silly...and having less contact between the base and the fins. Buut if you have to replace the whole HSF everytime a fan dies = $$$.

Click to expand...

I think the point was that that lack of contact between the base and the spinning fins is whats supposed to make it work better, by eliminating/minimizing the stagnant hot air boundary layer or w/e.
Based on what I can take away from the design. It would seem like the the point of failure would be the motor/bearing. So if an OEM is smart, they could design it so that part would be removable/replacable, so if it does fail. You wouldnt have to chuck the whole thing away, just change out the motor.

On another note, it seems like this would only work on horizontal applications. I'd guess it might not work the same in the typical vertical tower case.
EDIT: Nvm, read the pdfs, "Q: Can the device be mounted in any orientation?
A: yes—the air bearing assembly is held together by magnetic attraction (between the stator and permanent-magnet rotor). "

*Edit*

never mind.

I wouldn't feel safe using that.

Sign me up, I'll try it !!
Seriously, I'll buy one as soon as I see it available in stores !!

http://www.youtube.com/watch?v=JWQZNXEKkaU

I just have this small voice in the back of my head that says that type of cooler is going to fling off one day and wreak havoc inside my poor little computer. It is an interesting concept however

It seems to me to definitely be a 'budget' solution. As in, no overclocking/volt modding.

My engineer mind has the feeling that there'll be a critical 'mass' of heat output that this thing utterly cannot handle and the chip dies. So like, it's totally awesome with X heat wattage throughput, but exponentially can't hack Y heat throughput.

So it seems to me that it's best applied to OEM builds. But I for one, wouldn't employ this device in an oem box as it has a too critical a point of failure: if the spinning part stops spinning, there's no significant heatsink to sink heat into and the chip will fry in seconds. Just like how a watercooled cpu will fry in seconds/minutes if you disable the pump.

But yeah, a ridiculously optimised air TIM joint concept is extremely interesting.

evilsofa said:

Why have a fan on your heatsink when you can just spin the heatsink?

https://ip.sandia.gov/technology.do/techID=66

www.youtube.com/watch?v=uGpV_VPUn8g

Claims are it's 30 times more efficient than your typical heatsink and fan setup, which probably means stock coolers, not the high-end stuff we talk about here. But I still look forward to HardOCP's review.

Edit: and it turns out to be last year's story, so this is probably less exciting than I thought.

Click to expand...

If they are comparing the efficiency to Intels stock heatsink design that has been in use for around 10 years.. then yeah, i can believe the efficiency claims.

I saw this when it first came out. It looks like a really horrible and unreliable design to me. Since when was rotating a chunk of metal more efficient than rotating a fan to blow air through a well designed cooler?

metal-air-metal transfer is going to be way worse than metal-TIM-metal unless you are using an insulator for the TIM no matter how you spin it.

cyclone3d said:

If they are comparing the efficiency to Intels stock heatsink design that has been in use for around 10 years.. then yeah, i can believe the efficiency claims.

I saw this when it first came out. It looks like a really horrible and unreliable design to me. Since when was rotating a chunk of metal more efficient than rotating a fan to blow air through a well designed cooler?

metal-air-metal transfer is going to be way worse than metal-TIM-metal unless you are using an insulator for the TIM no matter how you spin it.

Click to expand...

This has a lot more applications than just CPU heatsinks, and there seems to be a lot of interest for commercial applications. It obviously works, and works well, otherwise companies wouldn't be planning to spend money on it just to be different. Like they said, a Heatsink with a fan is the cheapest and easiest way to cool something, not necessarily the best. If it was, water cooling would probably never have been invented. I'm curious to see how this works out.

My only concern is the heat transfer across the junction between the heat source and the moving part of the heatsink, but they seem confident that they've overcome that. The motor itself is loud for my taste, but I'll probably still get one just for the novelty of having it on my shelf.

It isn't just Metal-Air-Metal; it is Metal-TIM-Metal-Air-Metal. Let's remember the fact that air is an excellent insulator. This means there is not gonna be crap for heat transfer from the base plate to the fins except for through the brushless motor itself. Constant heat through Neodymium magnets reduces their effectiveness over time, which actually speed up the heatsink and reduces startup torque. This causes more heat in the brushless controller which will eventually cause it to burn out and release the magic blue smoke.

Basically, the first time I saw this design, I scoffed. I see it again, with an actual prototype and still I scoff. It wont work. Quite frankly, its a horrible idea and will never relieve 150W constant from the heat source. This is horrible inefficient.

StatutoryApe said:

It isn't just Metal-Air-Metal; it is Metal-TIM-Metal-Air-Metal. Let's remember the fact that air is an excellent insulator. This means there is not gonna be crap for heat transfer from the base plate to the fins except for through the brushless motor itself. Constant heat through Neodymium magnets reduces their effectiveness over time, which actually speed up the heatsink and reduces startup torque. This causes more heat in the brushless controller which will eventually cause it to burn out and release the magic blue smoke.

Basically, the first time I saw this design, I scoffed. I see it again, with an actual prototype and still I scoff. It wont work. Quite frankly, its a horrible idea and will never relieve 150W constant from the heat source. This is horrible inefficient.

Click to expand...

Agreed. But I think our understanding of air heat dynamics changes when that air gap is ridiculously small.

That's the magic it seems.

But yeah, I don't think this system would handle high heat loads that we see through overclocking/over-volting.

I don't get how they plan to keep dust from quickly building up at the center of the rotor.