THe Sandia Cooler - Breakthrough in Air Cooling design

[imported_ats]

^ This is the problem I see ^

Sandia said in their white paper and in interviews for articles, that this design would function in any orientation.
While that may apply to the heat transfer aspect (depending how the shaft is retrained in a non upright condition), it doesn't translate to air bearings.
AFAIK, Sandia nor any of those coming to market have demonstrated this either long or short term.

Um, this is actually rather easy. Spring, nut, meet shaft. Done. If you look at the various pictures floating out there, you can see the shaft, spring, and nut.

I don't see anyone addressing the startup or shutdown of these coolers either.
Physical contact is a given unless an outside air source (as Sandia used) is initiated during start up and shutdown.
Contact between the 2 metal surfaces produces metal particles (especially on shutdown) and that doesn't go well with such tiny air gaps as are used here.
Commercially, air bearings of this type use an external source to elevate the rotating mass before rotation starts and the gap to the bearing touchdown surface is much greater as the idea is to not transfer heat in those applications.

The initial Sandia prototype used a fed static air bearing. Subsequent designs and production prototypes use hydrodynamic air bearings instead. Both Sandia and commercial partners have demonstrated devices using hydrodynamic bearings with repeated start/stop cycles without issue.

I also think Sandia's thermal resistance and heat transfer numbers are skewed optimistically because of their test setup, but I won't debate that, as it will easily be checked in the next few months when production units are actually tested by websites

Both Sandia numbers are well documented and outside third party designs have thermal resistance numbers inline with the Sandia numbers. I don't forsee a massive gotcha in the thermal resistance numbers coming about.

 

[Idontcare]

Um, this is actually rather easy. Spring, nut, meet shaft. Done. If you look at the various pictures floating out there, you can see the shaft, spring, and nut.

The initial Sandia prototype used a fed static air bearing. Subsequent designs and production prototypes use hydrodynamic air bearings instead. Both Sandia and commercial partners have demonstrated devices using hydrodynamic bearings with repeated start/stop cycles without issue.

Both Sandia numbers are well documented and outside third party designs have thermal resistance numbers inline with the Sandia numbers. I don't forsee a massive gotcha in the thermal resistance numbers coming about.

I agree.

I don't see anything about the technology that raises my eyebrows, outside of the question of production cost.

I'm sure that is what is taking so long to bring this to market. Iterating the design so they maintain performance while making it have a lower manufacturing cost.

But the science aspects of it are pretty cut and dry IMO.

 

[soccerballtux]

clearly we're all awake watching the rocket landing

 

[Idontcare]

clearly we're all awake watching the rocket landing

Did it happen yet?

(it is 7:16PM local time, Taiwan )

 

[Puffnstuff]

I watched that cool chip demo and see that they're using a spindle shaft to keep the mass in place. In this iteration bearing quality is critical to longevity and the shaft seems fairly small in relation to the mass. I'd like more information especially about thermal transfer efficiency. We all keep seeing the horizontal open air demo's now show us the temperatures since air is such a poor conductor. Every demo we have seen is open air with the board laying flat. Lets see the demo of the board in the proper upright orientation in a closed case running a stress test. Last time I mounted a hsf I put enough arctic silver on my cpu to eliminate air in the gaps now we're supposed to rely on it completely for heat transfer. I'm not saying that it won't work, however I think that it will be limited in use.

Also the thought of a small conductive blender on my cpu has a certain air of danger to it. Will they include a free mp3 of Kenny Loggins - Danger Zone? Will they offer different speeds for the fan like mince, blend and puree? Will they get a celebrity endorsement for it from someone like Rory Gallagher?

 

[R0H1T]

Also the thought of a small conductive blender on my cpu has a certain air of danger to it. Will they include a free mp3 of Kenny Loggins - Danger Zone? Will they offer different speeds for the fan like mince, blend and puree? Will they get a celebrity endorsement for it from someone like Rory Gallagher?

I see, so you want the blender test is it
www.youtube.com/watch?v=msErUe2eInk

 

[Micrornd]

Um, this is actually rather easy. Spring, nut, meet shaft. Done. If you look at the various pictures floating out there, you can see the shaft, spring, and nut.

I have not seen detailed pictures of this assembly, can you provide a link(s)?
I don't see any spring in Cool Chips prototype.
Are you saying the shaft/rotating mass is held to the cooler base using a spring to maintain tension to force the rotating mass against the base?
While that will/can keep the rotating mass within the .03mm distance from the base required for the heat transfer effect to work, none of that addresses the side loading of the bearing in a horizontal position.

I'd be leery of the rotating mass (due to it's weight) "walking" or "screwing" away from the base due to side loading of the bearing in a horizontal position, but while spring pressure could prevent this, without an external source of air pressure available, it can also lead to a spiral of balancing lift of the air bearing to pressure of the spring in the horizontal position. (way overly complicated for a cheap appliance such as a heatsink/fan).

And then in a horizontal position you have that mass of aluminum rotating at several thousand rpms, hanging from a bearing at one end of it's shaft, well...
That had better be some good bearing and shaft retainer 'cause that's not some lightweight plastic fan.

The initial Sandia prototype used a fed static air bearing. Subsequent designs and production prototypes use hydrodynamic air bearings instead. Both Sandia and commercial partners have demonstrated devices using hydrodynamic bearings with repeated start/stop cycles without issue.

I've seen no indication/documentation of this.
There is nothing that I can find, in Sandia's white paper, indicating they ever attempted to run their prototypes in a horizontal configuration or performed start/stop wear analysis.
Can you provide a link to any of these testing or demonstrations in a horizontal position?

This illustrates some of my concern of the start/stop scenario - https://www.youtube.com/watch?v=EVjBeAzbUMI&NR=1&feature=endscreen

Both Sandia numbers are well documented and outside third party designs have thermal resistance numbers inline with the Sandia numbers. I don't forsee a massive gotcha in the thermal resistance numbers coming about.

As I said, it's not something I wanted to debate, I just feel that they are skewed (prototypes vs production) and will not be as good in actual usage.
Scadia claimed 0.20 C/W for their 1st prototype and projected 0.10 C/W for their proposed second prototype, while Cooler Master/Cool Chip have already backed their prototype up to <0.30 C/W.
That's quite a percentage increase from Sandia's prototype.
We'll see once production units are available for testing what the final numbers settle out on.

 

[Xpage]

I am excited about this, glad that it is making it to market. I wonder if they will make a copper version, i need that slightly higher Emissivity Coefficient

 

[DrMrLordX]

I am excited about this, glad that it is making it to market. I wonder if they will make a copper version, i need that slightly higher Emissivity Coefficient

Forget that, I want to see them make the fins out of CNTs. Lighter-weight, higher k value, what's not to like?

(okay, so manufacturing might be a bit iffy there but . . . )

 

[skipsneeky2]

I am excited about this, glad that it is making it to market. I wonder if they will make a copper version, i need that slightly higher Emissivity Coefficient

Some of the best coolers need a copper version,Hyper 212 Evo,just do it in copper Cooler Master!

 

[Maximilian]

I see, so you want the blender test is it
www.youtube.com/watch?v=msErUe2eInk

Aw man im so disillusioned right now It dosent actually blend

 

[ehume]

ICYMI, Cooler Master seems to be doing something with this idea.

 

[soccerballtux]

Some of the best coolers need a copper version,Hyper 212 Evo,just do it in copper Cooler Master!

i mounted a hair dryer on mine why isn't it cooling????

 

[aigomorla]

i mounted a hair dryer on mine why isn't it cooling????

cuz u have it on heat... instead of cool!

ICYMI, Cooler Master seems to be doing something with this idea.

that to me just looks like a gpu blower...

 

[Haserath]

I'm assuming this would be a lot more effective than the massive heatsinks that barely heat up with ever lower power chips?

Say broadwell uses only 100W and using a big "250W" dissipation unit you still can't get it below 100C because the die area isn't very big and can't effectively heat a big heatsink, but this would effectively cool better with the same 250W dissipation and keep it at 80C*?

I'm thinking this would be right because the smaller heatsink's average temperature would be higher, thus radiating more.

*randomly chosen number out of my invisible hat. It's very handy to have around.

 

[KingFatty]

I think that would only apply for a very short time, before the heatsink has warmed up all the way. When you say the massive heatsinks barely heat up, that's just because they have the big dissipation capacity. But given time, they will reach operating temperature. But then again, the heat change is proportional to the difference in temperature or something.

 

[IlllI]

well there was another article about them, so I guess they are still around http://breakingenergy.com/2015/11/18/sandia-cooler-blows-traditional-cpu-coolers-away/

would have thought they'd have some kind of product out for sale by now.

 

[RampantAndroid]

significantly reducing the energy needed to cool CPUs in data centers.

Huh? It's the same amount of heat going into the air - the CPUs just run cooler. You still need to remove that heat somehow.

 

[NTMBK]

Huh? It's the same amount of heat going into the air - the CPUs just run cooler. You still need to remove that heat somehow.

To be fair, if they can lower the CPU temperatures then they also reduce leakage current, so total power consumption goes down a little. Not sure how much difference that makes in the grand scheme of things.

 

[Shehriazad]

Just yesterday I was trying to remember the name of this thing to see if it's still happening...but is it just me or does it feel like a pipe dream?

 

[coercitiv]

To be fair, if they can lower the CPU temperatures then they also reduce leakage current, so total power consumption goes down a little. Not sure how much difference that makes in the grand scheme of things.

If you can lower CPU temp you can also let ambient temp increase until CPU reaches previous levels, as long as the rest of the components allow it. Allowing increased ambient temp may provide substantial energy gains, but this depends on many more factors than just CPU temp.

 

[LTC8K6]

Just yesterday I was trying to remember the name of this thing to see if it's still happening...but is it just me or does it feel like a pipe dream?

It sure is taking it a long time to get on Newegg for sale.

 

[Magic Carpet]

What does Sandia and AMD have got in common? No need to answer the rhetorical question.

 

[moonbogg]

This thing is a piece of voodoo magic superstitious crap. That's why it hasn't come out yet. It doesn't work.

 

[waffleironhead]

It thought cooler master was liscensing the tech. I remember reading something awhile back.