Fanless, spinning heatsinks may be in our future!

[ssjwizard]

Well now I live in a suburb of ABQ and ALOT of people I know work at sandia labs. I can say that whatever research they did is legit. Whether or not the company that paid for the research is misrepresenting it is a different question. Basically we do research on just about everything here at sandia and los alamos. I would classify this as interesting but unnessicary.

 

[Bobnova]

Think you can find someone who worked on this thing and ask 'em about it ssj? That'd be glorious!


I've been thinking about it through the day, and I suspect heat transfer between the two plates is better than we're thinking, given the velocity difference between them that airspace is going to be a turbulent nightmare, and that makes for far better heat transfer than a static airspace. As good as a much smaller surface area with TIM? Hard to say.

I still don't think it'll actually fly as a CPU cooler, but it may work decently in some applications.

 

[Psycogeec]

so that just leaves the 1.7 million doller question.
will it work in a smart bomb for 15 minutes
sorry i live next to its sister lab LLL

 

[Bobnova]

lol, probably not. I imagine a smart bomb doesn't have a lot of airflow through it, they probably just use straight thermal mass for cooling the electronics.

 

[turbohans]

Are we trying to reinvent the wheel here?

 

[Archer0915]

I really think we are over thinking this. I understand this to be more for industrial and equipment. It will work but not so great in processor apps. If however there were a way to reduce the gap and introduce some sort of media with better thermals than air it would serve as a functional alternative to conventional heat sinks on a CPU.

 

[Psycogeec]

then how about a larger hollow copper core shaft, in fluid dynamic base bearings, bearings directally linked to the core, the low thermal conductivity of the motion interconnect, is extended from top to bottom, so although it is slow conduction there is more of it. dump the micropores that will eventually fill with dust and road film and cat hair, then encase it in removable screen filter. Use the inside of the hollow core to forumlate a heat pipe spiral, and figure a cool way to increase the fluidic (i make things up as i go) thermal transfer in the core, based on the centrifugal forces, to increase the fluid thermal transfer.
use the same base for the armature assembly and contain the rest of the assembly under the heat sync. reduce the base size, which looks like it provides 1/2 the cooling anyway, and dump the air ride , because that is only going to work in a lab.
then formulate the worlds first round chip and wrap it around the bearings.
that'll fix it, and only cost them another 5 years
and in 5 years, they will have changed the transister type to low waste , negative thermal action style, that looses heat when it switches in the other direction, and gains heat when it swithes in the other direction, and obsolete the whole idea of needing cooling.

ok so now you know why they dont let me in the lab
.

 

[turbohans]

If linked directly to the core why would it need to move??

I think the efficiency loss in making something that moves conduct heat would be immense. Why not just be simple and make only the heat move?? Convection anyone??

 

[Psycogeec]

If linked directly to the core why would it need to move??

I think the efficiency loss in making something that moves conduct heat would be immense. Why not just be simple and make only the heat move?? Convection anyone??

because passive cooling isnt enough, unless we go back to tubes?

what is inefficient about that vat of greace they cook your fries in its lube , and it is liquid, and it transfers heat well. it is just a metter of getting just the right formula going, a nice freon lube

for passive i am trying to figure out how to make popcorn copper, I still cannot figure out why all the uses of copper are in sheet, they have all this micro this and micro that, but what do we get but only one type of copper? what would happen if you could have the thermal conductivity of copper, with the passive air conductance of ??? . . . ok i never got that far. but i want to see what poofy copper would do.

 

[turbohans]

because passive cooling isnt enough, unless we go back to tubes?

Think outside of the box! Again, reinventing the wheel is no fun!

Notice, I was not thrashing your idea just saying.

what is inefficient about that vat of greace they cook your fries in its lube , and it is liquid, and it transfers heat well. it is just a metter of getting just the right formula going, a nice freon lube

It serves it's purpose.

for passive i am trying to figure out how to make popcorn copper, I still cannot figure out why all the uses of copper are in sheet, they have all this micro this and micro that, but what do we get but only one type of copper? what would happen if you could have the thermal conductivity of copper, with the passive air conductance of ??? . . . ok i never got that far. but i want to see what poofy copper would do.

The long strands of molecules conduct heat, if it is irregular like Styrofoam, or has a lot of "porosity" it will act more like an insulator...

 

[Psycogeec]

The long strands of molecules conduct heat, if it is irregular like Styrofoam, or has a lot of "porosity" it will act more like an insulator...

ahh, that makes sence. what i would want to test, is a usual sheet, on the inside, to move heat well, and some type of special change on the outside, that makes it lighter, and "allows" heat to depart . the copper transfers the heat well, but seems to cling to it also, and not let go.
lots of the copper stuff we can buy is "forged" ok, but where is the extras, the ridging or mini fins and all the stuff that would increase air to metal contact.

or like you were saying some structure that is capilary convection chimneys. right now most of it is just machine shop work. not "special"
mabey they have tried this stuff, in the lab and found it to be useless, but that never stopped them from selling it before
Apple made a chimney model, it did what it needed to do.
.

 

[turbohans]

ahh, that makes sence. what i would want to test, is a usual sheet, on the inside, to move heat well, and some type of special change on the outside, that makes it lighter, and "allows" heat to depart . the copper transfers the heat well, but seems to cling to it also, and not let go.
lots of the copper stuff we can buy is "forged" ok, but where is the extras, the ridging or mini fins and all the stuff that would increase air to metal contact.

or like you were saying some structure that is capilary convection chimneys. right now most of it is just machine shop work. not "special"
mabey they have tried this stuff, in the lab and found it to be useless, but that never stopped them from selling it before
Apple made a chimney model, it did what it needed to do.
.

Yea, I should have rephrased that too, it is the 'density' of the molecules that makes for a good conductor. There are exceptions to that though.

When we are talking metal some of the best conductors are "rare earth" metals that are hard to find. However with other techniques it may be possible to use other 'materials' but that is not easy as gas's and liquids that posses the quality's to be used as a conductor are most likely highly toxic also.

So that leaves not much to play with here, but this is also something that has been evolving into a scientific need. ea "form follows function" and with shrinking electronics, and more powerful processors I do see an increasing demand for a 'different' cooling solution. I would hope more along the lines of a solid state part though.

EDIT:

Thinking about it now I do see how this could work very well though..

The mass of spinning fins could also act as a pump to move fluid through veins on its fins and through something at the base...

 

[Psycogeec]

The mass of spinning fins could also act as a pump to move fluid through veins on its fins and through something at the base...

ok were all set, lets just get the taxpayers to pay us to re-do the idea with liquid vein lube cooling system , with micro flow pump integrated into the shaft, and closed system block cooler , with free power convection rotation, from waste heat. and nano tech copper.
now i will go work up the press releace, and get that money flowing first, how many times should i use the word green? and will it be mercury free?
i am just assuming that we can show it breaking the "thermal brick wall" allowing that cpu to be overclocked beyond 3g, think 4g will sell them on the idea?

.

 

[turbohans]

ok were all set, lets just get the taxpayers to pay us to re-do the idea with liquid vein lube cooling system , with micro flow pump integrated into the shaft, and closed system block cooler , with free power convection rotation, from waste heat. and nano tech copper.
now i will go work up the press releace, and get that money flowing first, how many times should i use the word green? and will it be mercury free?
i am just assuming that we can show it breaking the "thermal brick wall" allowing that cpu to be overclocked beyond 3g, think 4g will sell them on the idea?

.

Yeaa, let me get some more beer first.

 

[psionic98]

I wouldn't aim so high as to 4g. It's always nicer to aim lower and totally blow their minds. I'd shoot for a 3.8 and say under some extreme conditions we could aim for a solid 4g!

Seriously though. I don't think I would use this to cool my old Ti-82

 

[LennyRhys]

The reception is quite different here from what it is at bit-tech forums. The research paper says very clearly that the 0.03mm layer of air doesn't in any way hinder the performance of the cooler, yet all the "experts" here immediately dismiss the viability of the design because of this air gap. Lol.

I think it's a neat idea and it'll be great to see it in action, and perhaps somebody will develop the idea further and come up with something even better. Who knows?

 

[Bobnova]

Lotta experts here

They're working on the gap being small enough that there is an incredible amount of turbulence in the airstream moving heat around. I'm sure there is, but solid metal is going to conduct better.

Frigid gas being pumped through the legion of tiny holes in the bottom section doesn't exactly hurt the cooling power either

 

[turbohans]

Air gap not hinder performance??

I take it they never had a "bubble" in the TIM on their L33t rig!

 

[LennyRhys]

Armchair experts, yeah!

@ hans, The thing is it's not "just" an air gap - the research paper explains in detail that the air is violently sheared by the rotating portion of the cooler which increases its thermal conductivity. Moving air conducts heat better than still air which, incidentally, is why fan ovens are hotter than conventional ovens. It is more complicated than that, but that's the layman's explanation!

 

[turbohans]

Armchair experts, yeah!

@ hans, The thing is it's not "just" an air gap - the research paper explains in detail that the air is violently sheared by the rotating portion of the cooler which increases its thermal conductivity. Moving air conducts heat better than still air which, incidentally, is why fan ovens are hotter than conventional ovens. It is more complicated than that, but that's the layman's explanation!

Yes certainly more complected than I would think. I just don't see how It could be as efficient as they say. But with a quick look at the PDF anyone can see they have put a lot of thought into it.


I would rather see someone convince all major law forces in the US to dump their gas guzzling V8's for turbocharged 4cyl motors... LOL But hey if this guy wants to make my computer run faster go for it!!