Cool Fan Idea

[Newtun]

A redesigned heat exchanger could speed up your home computer, while reducing fan noise and energy use. Air-cooled heat exchangers haven't changed much in 40 years, says Jeff Koplow, a researcher at Sandia National Laboratories.
Link

This Popular Science article claims potential wider applicability, to refrigerators, air conditioners, etc.

 

[ddrueding]

I still don't understand how the heat is effectively transferred from the static base to the moving heatsink. First it says the static air of the boundary layer is a problem, then it says that the heatsink is separated from the base by a cushion of air?

 

[Bozo]

Somewhat better explanation.

 

[Will Rickards]

The original paper is from Jan 2010.
Why don't we have these already?
Saw this on reddit yesterday and read the original paper.
The air cushion is small, they compare it and its mechanism to that of a hard drive and the heads floating above it. So we are talking very small cushion of air and because of the movement supposedly even it isn't a limiting factor.
What I found interesting about that is they said it could be mounted upside down or sideways and still work. I don't understand the fluid dynamics there to know how that even works.

Noise should be minimal as they use a brushless motor I think.
They mention low noise in the paper but not specific numbers.

 

[Howell]

The boundary layer in fluid dynamics is by definition close to the physical object and has very low velocity compared to the fluid farther away. I'll read the article later and I may have something to add to the understanding of it.

 

[Howell]

It is difficult to understand how they the heat efficiently off of the base plate.
Additionally it looks like there will either be a dead-air spot at the center of the fan or the deadwood will be so small so as to restrict airflow.

 

[LunarMist]

The original paper is from Jan 2010.
Why don't we have these already?

Perhaps they are not efficient or reliable enough in real-world use. I'd want to see some durability tests at high temp, high humidity, med-high concentration of non-viable airborne particulates, vibrations etc. :queen:

 

[Mercutio]

I read somewhere, a couple years ago, that Intel put some engineering talent into developing a low cost, self-contained water cooling setup that could be easily installed, and that the final product of that development was under $25 per unit.

I want to know what happened to THAT.

 

[ddrueding]

I read somewhere, a couple years ago, that Intel put some engineering talent into developing a low cost, self-contained water cooling setup that could be easily installed, and that the final product of that development was under $25 per unit.

I want to know what happened to THAT.

I remember that. I should see if I can find someone who knows...

 

[LunarMist]

If it is not on the market either it is not practical (too expensive, unreliable, etc.) or nobody really needs it compared to other products that satisfy the same market. That happens all the time with "great ideas."

 

[Bozo]

I think they were devolpeing that when the P4 roomheaters were out.
Now that the 'Core' series is out, you don't really need water cooling. So why bother.