Sandia National Lab Starts Licensing Out Super Quiet, Hyper-Efficient CPU Cooler

29

Comments

+ Add a Comment
avatar

drcrazyrich

seriously if they need a tester i m in :) I hope it s as quiet as it looks and kool of course . Hurry up I wanna see the results cause my cpu fan is noisy !!

avatar

Strangelove_424

I want it in copper. So bad. Hope it works as advertised.

avatar

Demonspank

I wonder if this thing runs on blood as well. Can't count the amount of times I accidentally bumped my hand into the casing of my heatsink.

Just trying to think how they would put a frame around this that would stop the potential booger-picker loss while not interfering with the heat dispersal.

avatar

John Pombrio

Sandia National Labs. This is THE nuclear bomb facility in the US. I was offered a job there (as well as Bell Labs, took a job with HP instead) many moons ago.
Its funny how both Sandia Labs and Bell Labs now have as part of their mandate to develop real, commercially viable products. No more pure research, like developing the transistor, lasers, nuclear power plants for spacecraft... This is a waste of some of the finest labs in the world.
Notice how much of a PR stunt this is, showing how Sandia now works closely with industry. A multi billion dollar government lab creating a new kind of fan..

avatar

Derek Conklin

A multi billion dollar government lab creating a trillion dollar fan... okay? I suppose you haven't thought about how important this fan is to our economy.

The years it's taking them to push it to market clearly shows their dedication to industry.

avatar

szore

Hey John, they needed someone to mop the floors?

(sorry, I couldn't resist...)

avatar

John Pombrio

Heh, no. Lab tech working on one of three Star Wars projects (remember them?). I had 4 job offers at Bell Labs (hel'a'day of job interviews but my first limo ride!) working on many things, one being a lab tech on the SOSUS underwater acoustics line to catch the sound of Soviet subs in the Atlantic. HP offered me a job close to home as a Test and Measurement System Support field engineer (mostly at UT, government, and DOE sites in the NE) which I did until I was bumped up to national second tier TM support working from home for the field engineers. Retired at 51 so I could spend more time posting to web sites!
Sandia DID offer 6 week paid vacation STARTING. I guess they wanted well rested nuclear bomb engineers. Just too far from family.

avatar

szore

burp

avatar

Rift2

I just replaced my Socket A AMD 3000+ XP computer with a new Rosewell fan man are those a pain to attach the socket A that is. I stocked up on them incase they stopped making them which is bound to happen.

avatar

Ed Littell

This same new design has been in use for probably over 100 years and is well proven in a few different applications. Best example is the blower fans in almost every gas or electric furnace in homes, apartments and stores. The next most used is in internal combustion engines. Think about the flywheel on top of millions of lawnmowers and other small machines. Yea it works the "new" part is the 1 point of contact and liquid dynamics mumbojumbo.

avatar

Danthrax66

It is completely different...

avatar

aarcane

I would like to see some prototype comparisons between this and a 212 evo for heat dissipation at various thermal levels. Similar comparisons against liquid coolers would be welcome, as well as a fluid version of this cooler. Perhaps an oil immersed version in a closed system?

avatar

loozer

If you read the paper, it says that the self-regulating air gap will work in all orientations.

avatar

rwpritchett

You'd need some very secure and tidy cable management. No slop or that thing will grab your cables and chew them up. I wonder if they plan to add a frame or cage around it. How about some graphene TIM to go with the new design too?

avatar

whathuhitwasntme

your cables should be tidy anyway!
muwahahaha

avatar

Jecht

I wonder if it will work as good our better than liquid cooling.

avatar

stradric

Probably not given that the specific heat capacity of water is over 4x that of air. Other coolants provide even better efficiency. Also, liquid cooling is inherently quieter. Where this design is probably going to rule over liquid is in cost and ease of setup.

I imagine this solution probably comes at least close to maximizing the cooling potential of air, which is significant in its own right.

avatar

Baer

Perhaps this can be used to cool the liquid in the radiator instead of the presently used fans. The best of both worlds.

avatar

thetechchild

Usually you have fans on the radiators anyways in a liquid cooling setup... Even for those, wouldn't this new design be a great improvement?

This on itself, though, would probably not come close to water cooling efficiency. However, air cooling already is the most mass adopted due to cost and ease of use. This is likely cheaper to manufacture than a traditional fan+heatsink combo, not to mention its dust resistance, better cooling, lower power consumption, and less noise. It's easy to see this becoming the new mainstream cooling solution, if it lives up to the tech specs to the researchers are sharing.

avatar

Ilander

So, current radiators depend heavily upon linearly-impelled air impacting the thin fins in the radiator to cool them. This works, but is actually the weakest link in the chain of cooling; the volume of liquid used is actually more of a factor when talking about actual temperatures.

Now, if you wished to use this fan/heatsink (it's really both in one unit) to cool the radiator, you'd have to completely redesign the radiator, as the air it throws off is moving roughly tangentially to the cylinder's curved surface. To maximize the impact of that air on a radiator, you'd have to have a cylindrical radiator with a mesh that orients the cross-section in that line, and that's not an easy task. You could cheap out and have the mesh oriented tangentially, but that would decrease efficiency as it would create turbulence and impede flow, thus acting a bit as an insulator.

All this said, using this fan in that application effectively removes the heatsink-aspect of the device, and that is one of the innovations; there is no reason not to use a cheaper "squirrel cage" fan in that instance, and they are pre-configured to throw the air radially, which would be much simpler.

So, mounting this thing on top of a vapor-chamber that was in direct contact with a CPU might be a good solution, but combined with a radiator, you lose many of the advantages.

avatar

LilHammer

Just one problem with implementing this design in computers that has yet to be addressed. This cooler design works great, as shown, in a horizontal configuration. The hydrodynamic gas bearing is created by the relationship between the motor plate and cooler/fin plate - that relationship is changed completely when you rotate things ninety degrees on the vertical plane (the way this would be mounted on existing motherboards).

Sure, I get these are crazy smart people who will figure something out but this isn't even close to being ready to implement in the current consumer PC environment. Rackmounted servers maybe...

avatar

drcrazyrich

what so ever for me my mobo is laying low at the bottom as shown ,so send it up lets give it wurl :)

avatar

kiaghi7

google "gyroscope"...

avatar

szore

I am no engineer, but I think the centripetal force that disk generates negates any orientation issues.

avatar

Ilander

So, not to be too pedantic here, as it's kind of a lesser-understood area of mechanics, but it's the rotational inertia that you're talking about. The centripetal forces in this case are the intermolecular forces that hold the metal together and keep it from disintegrating under high stress (such as caused by a rapid spin).

The rotational inertia is that property of the spinning object to maintain its direction of spin, and is why a hard drive that's on doesn't like to be turned.

Now, you're clearly familiar with the concept, so now you have the right word to impress your friends with, and complain about my need to tell people this whenever it comes up. That's why it sucks to be a physics guy. Always trying to exactly define everything.

avatar

szore

Thank you Mr. Wizard. Point is this; orientation isn't a problem.

avatar

chronium

Yeah I noticed that as well.

avatar

LatiosXT

Wait, is the spinning part both the heat sink fins and air moving unit? I think my head exploded.

I guess the only other question is if it's orientation dependent. But seeing how hard drives operate on a similar principle (the whole lift off thing), probably not.

avatar

Ilander

So, because gravity is so relatively weak compared to the other forces, when you have a metal object spinning at 2,000 RPM, the orientation issues become very small. Here's the calculation: a point on the extreme edge of this thing, guessing it's about 4 inches in diameter, will undergo an acceleration of 2200 m/s/s (meters per second squared), or about 225 times the acceleration of gravity.

So, if it's on its side, spinning, at the top, that point (from it's frame of reference)is pulling 224 g's of acceleration, and at the bottom, 226 g's. That's not much of a difference to a piece of metal.

Makes you appreciate how strong modern materials are, eh? A human couldn't undergo that kind of force, well, ever!

Log in to MaximumPC directly or log in using Facebook

Forgot your username or password?
Click here for help.

Login with Facebook
Log in using Facebook to share comments and articles easily with your Facebook feed.