Hey everyone! Check out this new cooler design by Sandia National Laboratory. It's an interesting concept I think :3. There's a technical article here for those of you who want details. What do you guys think?
I think "awesomes" to any new technology or invention. But this one is about cool, quiet CPUs, so I think "turbo awesomes." Let's just wait until they hit the shelves...
Awesomeness personified. One thing about this cooler that amazes me is that the interface between the base and the rotating part is... air. Very cool indeed.
If it works just think how small they can make and what changes to all fans with the PC case. Cannot wait to see the finished product reviews when released
One thing I noticed today after looking more closely at the design is that most of the testing appears to be using the heat sink while it's standing up. In their illustrations the air is being admitted from the top of the cooler then distributed outward by the spinning fins. However, just about every case I can think of mounts the motherboard (and thus the CPU and its cooler) sideways. The airflow on most cases is also front to back (or back to front). This means that in a practical setting cool air from the case's intake would hit one side of the rotating cooler rather than the top. The air would then get redirected randomly by the spinning motion and never reach other parts of the cooler and case. So the top of the cooler isn't really taking in the fresh air coming from the case's intake like it's supposed to, and the spinning motion means that any air that does get directed into the CPU cooler is redistributed in several directions, which would disrupt the airflow in the case. Case airflow is usually designed front-to-back; traditional air coolers have their fans set up so that it compliments this air flow. This new cooler though seems to be designed for air flow to come from the side then get taken out of the front/back/bottom/top of the case. For this cooler to work optimally, I believe that cases would need to be redesigned to allow the cooler to sit upright and have airflow go into the top then out all sides of the case. That may be too much for case manufacturers to really warm up to, especially since all their current case designs would need to be retired. Even if this cooler does become widely accepted, I don't think it would work in an optimal setting until case manufacturers bite the bullet and redesign their cases. What do you guys think?
I don't think it would make a huge difference tbh. The movement of air is not relative to orientation but to the rotation of the cooler, so it would likely behave exactly the same way no matter how the cooler itself was mounted. And I'm noticing a common misconception that "cool" air is required to lower the temperatures of a heatsink - not true; as long as the air is moving, even it is warm (say, 25-30C) it will still effectively draw heat away from a heatsink. If the air is the same temperature as the heatsink, then you have a problem.
Intel has a 'thermally advantaged chassis' design with a 80mm (im sure case makers could change this to a 120) intake on the side panel which is ducted to a stock heatsink. Change the other fans to exhaust orientation and i can see this working. So instead of retooling for new cooler design, just enlarge that design?
Its an interesting design idea and could make the coolers a lot smaller and more efficient. Dependent on the speed of spinning presumably you could do this with a liquid between the top and bottom and potentially increase the rate of heat temperature. Still if it works well then it does not matter. Will it be quiet? Metal on metal is not a nice noise at high speed and those metal fins cutting into the air at high speed might not produce a nice tone.
If you read the PDF file it mentions 5,000 rpm. Depending on the quality of the motor, it's not going to be loud.
Am I the only one that thinks this is basically a blender for our PC...? I've caught my fingers (and hair) in and, bashed them on PCI cards, ect, and got cut up just installing a Du-Orb. This looks like it'd chew your finger off, which is why I think it'll never come to desktops...
It won't eat your fingers any more than an ordinary stock cooler would - it has a really weak motor and won't be spinning fast enough to cause any real damage. Blenders, on the other hand, have 20,000rpm motors which draw hundreds of watts. Now invert your scenario, and put a cooling fan on that bad boy
Ever tried putting fingers in the spinning blades of a delta fan? I did as a bet, I can confirm it definitely hurts, and i couldnt type anything for weeks due to having no skin left on my fingertips after that foray. If the technical document theory is correct at 5k rpm, it will be able to do some damage.
There're some pretty sane comments over on SemiAccurate (yes, SemiAccurate! ) as to why this idea might not really be feasible.
5000rpm is very quick, and will easily cause you some damage if you get your fingers caught, a stock cooler has plastic fins going at maybe 2000rpm? As someone said above, deltas can cause serious damage (cutting through fingers to the bone) and they have plastic fins going at around 5000rpm if I'm not mistaken/
Hey guys! I found an article that has all your questions and more answered by the inventor of the new heatsink design, Jeff Koplow . Thanks to Extreme Tech for this
Silver, gold and copper all have higher thermal conductivity than aluminium... but aluminium is cheap and extremely light, whereas the others are expensive and very heavy
Its a fancy impeller, air is drawn in through the inlet at the top and the heat is then transferd back through the fins. It sits on an invisible cushion of air like an air hockey puck and its also self cleaning because of magical centrifugal forces. Sounds good to me! Probably not really any more efficient than a radiator, but still.
According to the inventor the cooler doesn't do leaps and bounds for cooling efficiency, but it does do a lot for maintenance, noise, and size.
