h a l f b a k e r yIt's as much a hovercraft as a pancake is a waffle.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
The best way I can explain this idea is to lead you through the thought process.
I was looking at the fan on a CPU cooler. Current fans are plastic, and just move air over the heat sink. I thought it would be better if it had metal blades instead of plastic, so that the air moving over the blades
would help the cooling.
Then I thought that if you are going to have a metal fan to disperse heat, why not get rid of the whole static heatsink, and have just a moving metal fan. But, instead of fan blades, have some 3-D fractal structure that maximizes surface area to have more cooling.
So now imagine a spinning metal evergreen tree over the cpu. I don't know how the joint connects to the CPU -- probably some thermally conductive grease. I don't know if friction would be a problem, but there already is friction in current CPU cooling fans adding heat. So there.
Here's a real working spinning-heatsink-as-a-fan (apparently)
https://www.extreme...hats-also-a-cpu-fan
The heat transfer from the stator to the rotor is accomplished by an air gap, they claim! [notexactly, Apr 25 2019]
[link]
|
| |
Since the fan doesn't connect to the CPU (and could not possibly connect to the CPU), it doesn’t matter what it's made of. The design of the heat sink is what you should be concerned about. |
|
| |
[ldischler] I just mean that there is a continuous mechanism of heat transfer, rather than empty air. |
|
| |
Anyways, I think you're missing the point. In this, the fan *is* the heatsink. There is a single spinning contraption that draws heat away from the CPU and transfers it into the air. |
|
| |
Even if you could mount the shaft of a fan to a CPU, the heat transfer would be greatly limited by the bearings, and by the relatively thin shaft. Much better to mount a heat sink to the CPU and blow air over that. |
|
| |
OK, how about a thick shaft? What about a fan stalk that 'floats' in grease instead of spinning around ball bearings? |
|
| |
Never mind the 'fractal', it seems the 'spinning' is giving you the problems. How about just a fractal heatsink? Regular fan, cool metal cooling block? |
|
| |
Where would you put the fan on a fractal heatsink? Whatever the exact shape of it, I'm certain it won't have any flat spaces if it truly is maximized for surface area. |
|
| |
Poop. Even a thermal grease has a conductivity 400 times less than aluminum. And you don't put the fan on a heatsink. You put the fan so that it blows air over the heatsink, which works best if bonded directly to the CPU. |
|
| |
Dischler. You *do* put a fan *on* a heatsink, and yes, I'm well aware of its role in blowing air over the heatsink. |
|
| |
Maybe then we mount the fractal heatsink on the chip and have the entire chip spin? |
|
| |
I'll jump straight to the next step, since maintaining the connections between the board and a spinning chip would be more than a little bit difficult. Spin the whole computer, minus the case. Even have the monitor spin (it has wires to the computer as well) and set the refresh rate accordingly. |
|
| |
Short-range radio communication between the chip and the board. No need for spinning motherboard, or monitor for that matter. |
|
| |
Perhaps the fan could be mounted in a ring-shaped bearing, rather than a shaft shaped bearing, to maximize thermal contact with the CPU. |
|
| |
Hot tip: you don't have to cool using air. |
|
| |