Also, anyone that never turns off their computer would know that 'stopping' isn't required for dust to build up on the fan blades, dust will quite easily build up on even moving parts.
Many of the questions in this comments thread (e.g. safety, e.g. vertical mounting, e.g. cost associated with the bearings, e.g. the basis of the 7% figure) are already answered directly or indirectly in a year-old Q&A article that was linked from TFA.
Does this heatsink work when it's mounted vertically? The "hydroplaning"/floating effect seems to suggest the heatsink must be resting on the base instead of suspended sideways.
I remember seeing this a year ago. It's a bit worrying that in that time they still haven't gotten around to building a prototype that will be as quiet as the claimed final product. While that doesn't mean it can't be done, it may point to deeper issues in the project such as yet unaddressed technical challenges or lack of funding (which might in turn suggest lack of investor conviction).
Personally I'm curious what the failure rate will look like with these things. Given it will probably be higher than existing PC cooling fans which are moving lighter loads, this may indeed be a problem.
I think what you've seen is this[1] article at extremetech.com or may be the HN post[2] inspired by that article. Both are quite interesting to read (article provided some original research in PDF, HN discussion provided some comments from the fan engineers).
Besides your question about failure rate, I'm also wondering if such a massive thing going at 2000 rpm becomes a dangerous neighbor to other stuff on the motherboard (especially in the case of failure) and to the motherboard itself. I figure the vibration this kind of fan generates may be rather noticeable.
Most larger manufacturing companies move at a glacier pace. I could see Apple making a quick decision to use these and then having them in models within a year, but few other manufacturing companies are willing to move like that.
My impression was that Apple intentionally lagged the cutting edge curve to error on stability and sound user experience. I would be very surprised to see them go in the direction of moving parts - instead I would expect them to continue in the direction of all solid state (no moving parts) as much as possible.
From the article: "This centrifugal force is what gives the Sandia Cooler such massive efficiency, too."
That is not correct. The efficiency boost comes from the very efficient transfer of heat from the CPU to the spinning fan/heat exchanger by, effectively, using the very thin air-gap under the impeller. The rotation of the impeller breaks the boundary layer in the airgap and you get good heat transfer across a very thin gap.
The airflow over the machined aluminum blades and, in general, the rotation of the entire impeller assembly, serve to keep the heat exchanger free from dust accumulation (which restricts heat flow).
Of course, as many have pointed out, it remains to be seen how easily this concept translates to a mass-manufactured low-cost solution in terms of performance and reliability under varying conditions.
I've done a ton of heat flow FEA work when working on various approaches to cool a custom high-power LED array (1,500W power-in). We could get reasonable results with complex forced-air solutions and relatively expensive custom machined heatsinks as well as carefully modeled airflow controls. In these cases the solutions were always very large (volume).
When we switched to fluid-based cooling things changed dramatically. One of the design challenges was to maintain a narrow delta-T across the LED array. This is because thermal uniformity was required in order to have uniform performance across the array. The fluid solution, with some tricks, could easily achieve ten times better thermal uniformity than the air-cooled approach. And, in addition to this, cool the entire array to a much lower final temperature.
A fluid cooling system was constructed using only a small fluid pump and no air-moving fans at all. A passive natural convection radiator could easily handle the heat-load in a normal air-conditioned office environment.
While I have not looked at the specific case of cooling a CPU, based on my experience I have to say that far greater gains can be had by rapidly moving heat from the CPU surface using fluid-based cooling. This, effectively, creates the opportunity for much greater surface extension than can reasonably be applied to the small surface area of a CPU.
Again, I have never studied CPU cooling, but I am not sure that this 150W cooling limit applies to fluid-based cooling. I can see building a systems that can very easily move 150W, or even double that, using a relatively simple fluidic cooler. At some level it is a matter of how many molecules of the fluid you can move across the CPU-side heat exchanger per unit time. The answer to that is "a lot".
I can't see the Sandia or any other pure air-based cooling system used for CPU cooling at the extremes. The assembly would have to be very precisely manufactured and lots of work would have to be done in order to ensure that vibrations and harmonics of the motor drive system itself don't cause damage to the circuit board. If the system needs to have an impeller spinning at 2K RPM or more, lots of work needs to go into making it safe for servicing as a metal impeller like that can shred fingers in an instant.
Finally, there's the question of the mass of the spinning impeller. In order to transfer heat into the impeller blades you are limited to certain geometry. If the spinning base and/or the blades get too thin you simply won't be able to move the heat out no matter how well it can move from the stationary plate up to the revolving disk. This is critical and it means that there are certain minimum geometry constraints that are likely to make the impeller somewhat massive. From my FEA work on heat transfer I know that you can only go so thin on blades before they become useless past a few millimeters above the heatsink base-plate. The same is the case here.
What I can see is the use of this concept to create a fluid based solution that uses a liquid to quickly move heat from a CPU to a much larger heat exchanger that uses the Sandia heatsink to move heat into the surrounding air, and, thereby, cool the CPU. Even at that, I'd like to see data comparing conventional forced-air convection cooling of the external heat exchanger and even a comparison to a natural convection solution.
It sounds like you were using liquid cooling, did you ever try phase change cooling? Something like this (http://www.youtube.com/watch?v=Z_X_hgtlJpA&t=36)? I saw a demo of it a couple years ago, really cool stuff.
No. Not flexible enough for what we needed to do and, in some ways, limited in performance. For the right applications phase change heat transfer is fantastic.
EDIT: My comment has to do with phase change cooling of the type implemented with sealed copper tubes moving heat from one end to the other using a phase change fluid inside. This full-immersion phase change cooling setup is a something entirely different. I don't know anything about it.
"That is not correct. The efficiency boost comes from the very efficient transfer of heat from the CPU to the spinning fan/heat exchanger by, effectively, using the very thin air-gap under the impeller. The rotation of the impeller breaks the boundary layer in the airgap and you get good heat transfer across a very thin gap."
You misunderstood the physics involved. When they refer to thinning the boundary air, they are not talking about within the planar air gap binding. They are talking about the air within each of the impeller's channels.
The air in an air gap binding experiences positive pressurization. The proof of this is how the impeller lifts above the bearing surface. The pressure overcomes gravity in this case once the sheer within the gap is sufficient. While there is some mass exchange in and out of the gap, there is no significant volume moving through it. If you think about it, this makes sense, because the volume of flow would be limited by the surface area of the perimeter. Given that the gap is so slight, not much of a window for any flow. It is more like a captured air lubricant, and heat transfer through it is more conductive than convective.
Rather, they are talking about the boundary around the impeller itself, particularly with each vane channel. The air within these channels experiences a centrifugal force. This shrinks the boundary layer thickness over the impeller surface as a whole. If you think about it, the impeller has a lot more surface area than the gap. This is why getting a thinner boundary layer has a big benefit.
To simulate this you'll need not FEA but state of the art in CFD simulation that is aware of boundary layer effects and vortex flows.
Liquid cooling is very neat, especially for very large systems or high head loads. I think an important secondary benefit is that warm liquid is easier to use as an industrial energy input than a mass of warm air. But air only cooling devices have a big cost advantage, so if you can live with the performance constraints they offer, you'll probably be lower total cost.
The goal here is cost effectiveness, not a race to the highest score as a watt number.
"From my FEA work on heat transfer I know that you can only go so thin on blades before they become useless past a few millimeters above the heatsink base-plate. The same is the case here."
The impeller is doing double duty: radiating heat and moving air. While I believe you about the value of fins in a radiating surface, here they're optimizing against two effects, so I think it's plausible that taller fins result in more flow per vane.
"What I can see is the use of this concept to create a fluid based solution that uses a liquid to quickly move heat from a CPU to a much larger heat ex-changer that uses the Sandia heatsink to move heat into the surrounding air, and, thereby, cool the CPU."
I think once you've suffered the costs of putting the heat into a liquid, a passive radiator is going to be better at getting it to the air per dollar, because you're no longer constrained to compact geometry.
I'll have to play with this. I don't doubt what you are saying at all. I've been lazy, I said FEA when I meant FEA and CFD, we have both available and yes, boundary layer effects and vortex flows are taken into account. Fun stuff to play with. For some problems you could spend your entire life simulating and never reach a conclusion!
I'll repeat my warning that I am not an aero engineer and what I've learned has lots of holes here and there.
As I understand it, boundary layer control is difficult. I remember going through NASA papers that talk about techniques used in centrifugal superchargers to control the boundary layer. These are impellers running at speeds way beyond that of the proposed heatsink. If I remember correctly, one of the problems with various techniques is that of flow separation. This would cause large portions of the blade surface to, effectively, not exchange any heat to speak of with the separated flow. Airfoil choice is important here.
That said, the proposed heatsink does not, as its primary design intent, have the requirement to be a good pump. The primary design requirement is to opimize heat transfer to the surrounding air. Things can and probably do change when you are optimizing for that.
What I would really like to see is the performance of a complete design. One that includes a casing with suitable inlet and outlets as well as the required safety devices. In my limited experience, that's when things can start to change. For example, the nature of the intake airflow can greatly affect what happens when the airflow hits the vane leading edge and beyond. Also, noise levels can go up.
It'd be interesting if someone in aerodynamics could pitch-in and talk about some of these effects and how things can change outside of simulations and free-air environment prototypes.
I think this design works well only in the context of computer casing or similar. The warm air coming from Sandia doesn't seem to get very far away? So it would eventually disrupt the cooling process if it's not blown away by the regular fans of the computer housing.
Isn't there a problem in fluid cooling that you can't really get the fluid as cool as the air where heat ends up eventually? My professor told that cars usually use fluid cooling because it makes engine temperatures more predictable, and therefore slight loss in efficiency is acceptable.
My professor told that cars usually use fluid cooling because...
Fluid cooling is more stable, because you have the added thermal mass of a couple gallons of fluid. It is easier to regulate temperature, through the use of the thermostat. It also reduces problems with hotspots, and multi-cylinder engines can be made much more compact.
Really, the question of efficiency is practically not even a concern- in modern times it is typically not until you have a racing engine, that you begin to have cooling problems. The real concerns are reliability and added complexity of the cooling system. The biggest advantages are improved longevity and performance of your engine.
It is true that an old car going uphill on a roasty day with the A/C on may overheat, but even then it is often a problem with poor maintenance or the wrong mix of coolant. (When your coolant boils, your cooling system cannot cool the engine effectively. Vapor has far, far inferior heat conduction properties. This is one of the reasons why the system is typically pressurized)
Yeah, I'm curious too, not just in what it's for, but how you provide the power to that - assuming a ~3V forward voltage drop on the leds, you need ~500Amps to light that up.
Each of those squares was a ~2m x 2m array of individually addressable tricolour leds (they could play video on them). Individually addressable means the can't run any of them in series - the whole panel needs to run at the forward voltage of the blue leds, or ~3V. I'm assuming they're built out of the individually addressable led strip like I can buy at Adafruit, so 32 tricolour leds per meter, for a total of 64 x 64 leds each with 3 emitters =~ 12,000, which at 20mA for each emitter requires ~250A to drive it all to full brightness white (which'd "only" be ~750W). They had 12 of those panels on stage. How the hell do you provide 3000Amps at ~3V? I know that's only ~9kW, which isn't a lot of power (in the context of a concert lighting rig), but doing it at 3V and 3000A is quite a different thing to 110V or 220V lamps drawing "only" 80A (or 40A @ 220V) in total. A 3000A power supply must have some impressive amounts of copper leading out of it. Even if it's 12 individual 250A power supplies, that presumably implies thumb-sized or thicker copper wire to pass those currents (and even thicker if the cable runs are any sort of length, the I squared term in I^2 x R power losses mean R needs to be _very_ low at 250A to stop things catching fire…)
A digitally controlled light source for a, well, shall we say, special application.
The light source was about 24 x 16 inches with LED's packed as tight as you can imagine. For best thermal transfer and uniformity the assemblies were bonded to a 0.5 inch thick aluminum base-plate in a vacuum fixture.
To answer the other question, power was provided by a set of 48V 500W AC to DC power supplies feeding purpose-built current and voltage control boards.
It produced an output luminance of about 60,000 cd/m2 (sixty thousand candelas per square meter). In other words, it was actually dangerous to look at it directly, all measurements had to be taken through stacks of attenuating filters. Fun project but about as dangerous as working with lasers.
I would have though the efficiency comes from spinning the heatsink and so having a very efficient metal-air heat transfer compared to blowing air over a stationary heatsink
It's weakpoint is getting the heat from the CPU across the air gap into the spinning part
I don't see the super-thin air-gap as an issue. In fact, it is probably an asset. As an example, when we build the 1500W LED array we used a very thin insulating film to transfer heat from the LEDs to an aluminum clad circuit board. If you have surface area (the stationary disk in their case) a thin gap can be surprisingly efficient at transferring heat from one element to the other. Ad to this breaking up the boundary layer within that gap and now air becomes a "better" conductor.
There is no difference between spinning the heatsink and moving air at the same velocity across stationary heatsink fins. It's about molecules of air at a lower temperature going across a surface with at a higher temperature. N molecules per unit time "absorb" some of the heat and come of of the process at a higher temperature. The surface temperature drops by a similar amount.
What kills you in both cases is a very thin boundary layer of air that sticks to the surface and has a velocity profile that goes from zero at the surface to whatever the air velocity might be. This boundary layer isn't good for heat transfer from the surface to the moving air molecules, which takes a toll on the overall efficiency of the process.
There are techniques to help break-up the boundary layer in stationary fin heatsinks. Some break-up airflow to introduce turbulence. Others use raw velocity to break it up (air impingement cooling).
I don't know how well the rotating aluminum impeller breaks-up the boundary layer surrounding the fins. It won't eliminate it completely but it certainly could be much improved from the case of stationary fins with low-speed air going across them. If you had a heatsink with stationary fins and air moving at the same speed as in the case of the rotating impeller (by using fans, of course) you'd probably get similar boundary layer effects.
What I learned is that there really is very little magic in heat transfer. I have seen some really wacky ideas brought to market that always seem interesting but have never proven to beat the simple solutions. Funny enough, other than materials, most of the focus is always around dealing with the boundary layer, an issue that disappears (in terms of its significance) once you move into fluid-based cooling.
Caveat: I have not studied rotating aluminum impeller/heat-sinks, so I could very well be missing something that I simply don't know. My opinions are based on thousands of hours of research and FEA work in looking for solutions for the aforementioned project.
I think you failed to actually read TFA. Unlike you, I know almost nothing about heat dynamics, nor about FEA, so my comment is predicated entirely on the Sandia staff being truthful in their video (and being quoted accurately in ars's Q&A article). I claim zero authority on the subject.
Your top-level comment claims that the breakthrough in this is in the thin-air gap, rather than in the centrifugal force. Respectfully, Sandia's Jeff Koplow specifically claims otherwise, in detail.
Koplow claims the boundary layer is the key problem (a claim with which you appear to agree). He further claims that when the radiator is spinning (or otherwise accelerating) that the boundary layer thins, and that in their application it thins by roughly 10x. See the video at 1:25 or so. And while I don't trust my under-educated intuition particularly heavily, it's very easy to imagine how accelerating the fins is fundamentally different from blowing air across them, in terms of how it affects air molecules in the boundary layer. Again, note that you specifically claimed "There is no difference between spinning the heatsink and moving air at the same velocity across stationary heatsink fins", and Koplow has specifically claimed that this is not true.
The fluid dynamic bearing only becomes relevant as a secondary problem: if you're going to spin your radiator but not your heat source, obviously there's a transfer problem. Apparently this is easy enough to solve, I guess? Koplow did mention (a year ago) that they were considering adding roughness to the revolving surfaces to perturb the air in the gap to improve transfer.
Oh, and you repeatedly say "they should use fluid". Well, first of all my little pedantic nit-pick is that air is a fluid. But more interestingly, the article contains a link to Q&A with more technical details, in which that very subject is discussed! The short version as I understood it is this: viscosity kills you.
I guess what I'm saying here is this: you started off by saying the article was wrong (about the centrifugal force), and went on to question all the design choices involved, but I think you actually skipped the bit where you read the article in enough detail to know whether or not it was wrong. Even though you obviously have a much better background in the material than I or most other commenters.
This news isn't new. Articles and at least one paper came out, if my memory serves me, two to three years ago. Back then I went through the available data in detail. I didn't need to dive into the article posted to HN to know what they were doing. Still, I did read the entire article and watched the video before posting.
As for the boundary layer issue. A fluid (OK, liquid) based cooler has virtually none of these problems and does not require having MULTIPLE metal masses inside your computer spinning at 5,000 RPM (per the paper on the Sandia site).
Remember that you need to cool memory, graphics cards and other elements in a typical design. The air-based CPU cooler moves air around the CPU cabinet and out the back or top. All of it serves to cool other elements. Dust or not.
You can't put a bunch of 5,000 RPM coolers inside a
In a typical data center you have reasonably-clean air available. Dust should not really be a problem except for the most neglected portions of an installation. I have seen systems in service for years with no indications of dust accumulation at all.
Dust can be an issue in office or home environments. Even then, from personal experience (and only from personal experience) I have never seen a problem.
I do think that the Sandia heatsink might have interesting applications as part of the heat-exchanger in a liquid-cooled system.
If you have a centrifugal fan turning at 5,000 RPM you are going to move a lot of air radially out. So far all of their experiments seem to show the device working well in the context of pretty much an open air environment. You can't have this pump simply circulate hot air inside a computer cabinet. Because of that you will need to surround it with an intake structure as well as an exhaust structure. This device sucks a lot of air at 5,000 RPM. That, without a doubt, will be noisy.
There's also another element here that is not being compared. How many cubic feet per minute of air is this device moving? How would a stationary fin heatsink perform if you moved that much air through its fins.
A few years ago we modeled and built a custom heatsink that consisted of a centrifugal fan mounted at the center of a field of fins located at the exhaust of the fan. Put another way, fan mounted at the center of a flat plate, intake is at the center, exhaust is radially outward. The fins where located to "grab" and channel the exhaust flow. They were also "ducted" meaning that the top of the fins had a "roof" so no air could escape without bathing the entire fin. This heatsink performed very well. Expensive to manufacture, but it did very, very well. We could custom machine boundary layer control elements into the fins and do better yet. The fan was an off-the shelf plastic DC brushless centrifugal fan with no thermal properties other than moving lots of air. And it was quiet.
I am not necessarily putting down the Sandia fan. I am simply saying that one should be careful not to be attracted to new shiny things without a little critical thinking. I have seen companies waste millions by jumping into technologies that sounded great on paper an were later found impossible to commercialize due to a million real-world issues.
>I don't know how well the rotating aluminum impeller breaks-up the boundary layer surrounding the fins. It won't eliminate it completely but it certainly could be much improved from the case of stationary fins with low-speed air going across them. If you had a heatsink with stationary fins and air moving at the same speed as in the case of the rotating impeller (by using fans, of course) you'd probably get similar boundary layer effects.
I think the point is that it's a lot easier to get a piece of solid metal moving very quickly with little noise than to get air to do the same.
It has to get that air moving at comparable speeds to be as efficient as the traditional way. The trick here is to get the air move faster with less sound. It's possible because of larger airfoils, so they can get more laminar flow and with more laminar flow comes less sound.
Right, if you were building a wing. The problem is that here we are after heat exchange and, as I understand it, the last thing you want is laminar flow due to the accompanying boundary layer.
wow - that is a surprisingly thorough introduction to thermal transfer. Based on your experience, to increase thermal transfer would you want to dimple the surface or add protrusions? (I know, at the extreme these are almost the same thing). I was reading golf ball dynamics and not sure if it applies or not.
Both of those ideas work to varying degrees. You can "trip" the airflow at the leading edge of a heatsink fin by introducing notches and protrusions of different geometries. This will make the airflow past that point turbulent for a certain distance. Depending on geometry you might have re-attachment, which leads to rebuilding the boundary layer. Dimples and surface roughness can help too.
In the end you have to deal with the real world. Some of these techniques are really good to grab dust in the airflow. If that happens, you have, ultimately, created a problem greater than the one you started out to solve.
After all I've been trough I have become a huge proponent of fluid based cooling. I don't think there's any way to have forced air cooling even begin to compare with the potential performance gains of using liquids to mechanically move heat around.
Wouldn't the boundary layer thickness be affected by the centrifugal force that presses the air against the fins? I was under the impression that that was the main advantage of this design.
I seriously doubt it. That said, I'll claim ignorance here because I have not studied that part of things in depth. I know that it can take a lot to break-up the boundary layer in any meaningful way. In aircraft design there are reasons having nothing to do with heat transfer to want to break-up the boundary layer. As I understand it this can be difficult to achieve across a large surface (I'm assuming at sub-sonic speeds, don't know). They even resort to such devices as drilling small holes and literally sucking the boundary layer from the surface. This problem is tougher than it might seem. I am not an aero guy, I've learned a lot from thousands of hours dedicated to finding solutions to our specific problems, but I am far from an expert. Maybe an aero expert on HN can fill-in some of the holes.
At some level, think about it this way: At some point, microscopic as it might be, some of the air molecules touching the fins have to "stick" to the surface of the fins. At that point air molecule velocity with respect to the fin surface is zero. Then there are molecules that stick to these molecules one layer above, and so on. After a certain thickness the greater airflow will win out and molecules will move at the average velocity of the bulk air mass moving across the fins. What you have is a velocity profile from zero to the average air mass velocity. That's your boundary layer. You can do things to make it thinner, but eliminating it is very difficult. Techniques like impingement cooling do this to varying degrees of efficiency.
Yes, spinning the heatsink is the whole point, to take advantage of the reduced boundary effect from the centrifugal force. I imagine the force of the air pressing against the fins in this case is considerable at such high RPMS. The other clever aspects of the design like the air bearing are simply to overcome problems that result from having a heatsink that's moving with respect to the heat source.
There is of-course no theoretical difference, except this presumably gets much high air flow rates than a conventional fan + static finned heatsink. and you don't have the problem of stalled pockets of warm air in corners of a fixed fin that don't get flushed out.
One of the problems with small heatsinks is that the bulk convection flow you model at high delta-T/high power don't always work in practice with small fans and small heatsinks - this design should scale down a lot better.
My concern was that in order to get good conductivity across the air gap you would need very close tolerances which are hard to make reliably in practice on cheap consumer gear.
Also, although the fin blades themselves should clear dust - I would worry about an oil/dust/dirt film building up in the gap if it's relying on constantly forcing new air through this to make a cushion
I remember reading the paper (or "a" paper) on the Sandia cooler a couple of years ago. The author seemed to indicate that the disks spinning so-fast-yet-so-close served to break the boundary layer AND keep the gap clean and clear of dust and grime.
I'd be interesting to understand just how precisely matched the surfaces have to be for this to work well. Machining a reasonably flat reference surface on a CNC lathe or mill isn't all that difficult. The question in my mind is more about how flat these surfaces have to be. The cutting tools will leave some grooves, even if almost imperceptible. Do the surfaces have to be lapped (sanded) and polished for this heat exchanger to work well? What are the tolerances? A good shell cutter on a high-quality milling machine can produce a mirror-like surface. It's one thing to do this in small quantities and quite another in mass production (which I now nothing about).
The gap is 1-thou (30um), this isn't challenging machining - you could make the gap 30nm ! (it's a bit more annoying with copper)
Actually some machining marks would probably be good, small surface irregularities will break the boundary layer - like sharks skin and make the air flow mix better.
Agreed. I've made "reference" flat surfaces on aluminum using our Haas VF3-SS vertical milling machine with a good quality shell mill with new inserts and, if you do everything right, you can see yourself on the mirror-like surface that results. This machine is probably the grade of machine you might expect to find in a high-volume production shop.
There is totally a theoretical difference between an accelerating frame and a non-accelerating frame. And a rotating frame is accelerating, and a fan-plus-static is not.
The video mentions that transferring heat to the impeller is a challenge, but it wasn't clear to me how they overcame this. Is the space between the CPU and the impeller so small that heat can be transferred efficiently?
If that is the case, I wonder how cheaply these could be produced given the incredibly tight clearance.
Their data suggests that the air gap provides minimal resistance to the flow of heat. As for the clearance, they've stated that unusually high-precision manufacturing is not required for the two surfaces.
Re: clearance, while .001" clearance isn't something I'd want in a design particularly, modern machine tools are good enough that you get flatness and surface finish good enough to not have problems at .001" clearance basically free on any machined surface.
Yes it's a radiative/convective coupling in the thin air gap.
Although the thermal conductivity of an air gap is normally low - which is why you have double glazed windows - it gets complicated if the gap is much less than the free path in the air
Because it's a prototype without noise filters or any other attempt to reduce motor noise. His point in turning off the motor was to show that there's little noise caused by the blades chopping through the air - which is where fan noise actually comes from.
If I had to guess I'd say it's because the video is targeted at people for which this will be obvious. They aren't looking to sell this directly to consumers. They want to license the cooling technology. As far as they are concerned, the noise of the motor is probably as irrelevant as the bracket's color.
I think what you actually hear is coil noise from the fast switching of the current in motor's coils. The motor they use looks and sounds more like a HDD motor more than a normal fan motor - optimized for constant speed under quite a heavy load. The fan motors work in a similar way, but they use less power and less coils in a bigger enclosed space, so they can be better optimized for sound.
In a production version the noise is largely caused by the air flow, whereas the motor noise would be cut down/eliminated by casing which is not present in the development version
I'm not sure what problem this is trying to solve. I have a truly fanless heatsink [1] that is ducted to a 120mm case fan at 1000RPM, which is guaranteed to produce less noise than a 2000RPM rotating metal disaster. A plus is that there is no air gap necessitating such a high rotation speed.
According to them, the air gap provides a thermal resistance on the order of 0.02 C/W. The primary resistance is the fin -> ambient air interface.
This device is also very quiet, based on that video. Assuming their goal of 0.05 C/W resistance is reached, it would perform significantly better than the heatsink you're mentioning.
We'll have to wait and see about the "dust immunity". "Centrifugal force" is not by itself enough to prevent dust buildup, or existing fans running at higher RPMs would be spotless.
As for quiet, you'll still have to pair this with a case fan if you want any exhaust, so I'm not sure it compares with a fanless solution for overall noise level.
You could probably use a heat pipe to the back of a laptop's screen. The problem becomes the joint where the screen connects to the case, but a flexible pipe is not out of the question...
if every conventional heatsink in the US was replaced with a Sandia Cooler, the country would use 7% less electricity. For the most part, these savings would come from air conditioning and refrigeration systems
How does that work? Is the same amount of heat not being expelled from the chip into the room? Does transferring the heat with a smaller fan somehow make it smaller heat?
Transferring heat with a smaller fan requires less power to run the fan. For a given amount of electricity, the Sandia Cooler can extract more heat from the processor, so it requires less electricity to keep the processor cool.
Not just CPU fans. FTFA, and the latter part of which was included in the GP's quote:
"an ideal replacement for just about every fan-and-heatsink installation in the world... For the most part, these savings would come from air conditioning and refrigeration systems..."
I know, I read that. The impression I got from the article was that these new heatsinks were going to be used on the CPUs and that somehow that would generate less heat. e.g, if Intel makes a lower power chip, a lot of savings will come from AC, even though Intel isn't making AC systems. It was not obvious to me (and I still have my doubts) that "every fan-and-heatsink" was referring to AC systems as well as CPU heatsinks.
I think they're referring to the heatsinks _in_ the air conditioning and refrigeration systems. The higher efficiency of the cooler itself would allow these devices to operate more efficiently.
That would be interesting. A more efficient way to heat the planet. :) It sounded like they were only talking about CPU/GPU coolers, but I suppose similar ideas could be applied to AC systems.
Yeah, the biggest heatsinks are not in your computer. There are big heatsinks in your car (radiator) and A/C units that would benefit from the tech more than your laptop.
Yep. If it can use a lower power fan/impeller, then there would indeed be less heat expelled into the room- a very tiny bit less, but the principle holds true.
The major energy savings would come from large implementations such as air conditioners.
If every car before it needed both an engine and a car, and now there's a car that is its own engine, I would give them some leeway to call it an engineless car. The big draw here is that the heat sink is the same as the fan, instead of needing to buy both a heatsink and a heatsink fan.
"Fanless" is still weird, I'd prefer to call it integrated.
This is a very interesting heatsink design. That said, having a finned heatsink spinning at 2000+ RPM seems potentially dangerous, especially if someone were to touch it with their bare fingers while it's operating.
Given that it sucks the air in through the central part, it must be spinning counter clockwise, with external fin ends pointing away from the direction of spinning. Not 100% safe, but not a finger chopper either.
Last time this was one hacker news someone suggested mounting the chip onto the heatsink and spin the whole thing, without any thermally conductive bearing. Still an awesome idea imo.
I missed the article last time, but I had this thought too, on reading it this time around.
The challenge, obviously, is moving data on and off of the spinning element. I assume one would stack the CPU, the GPU, the mobo, the RAM, etc, up along the axis of rotation. It's certainly easy enough to get keyboard and mouse and audio get on and off the spindle via wireless... but what are the costs (in money and in latency) to get higher-bandwidth signals on and off, like video, and like main storage?
That part looks really expensive to manufacture. It's machined. Can they make it out of molded plastic (like fans) or sheet metal (like CPU heatsinks)?
Most of the original (boxed) CPU fans are machined. The sheet metal fans are 3rd party and they need costly heatpipes to move the heat away from the (machined) CPU base. It will be costly because the machined metal fan will need great balance and a good polish to stay quietly floating above the base and not jump and rip through the computer's case.
You cannot use plastic as it has a lousy thermal conductivity. You cannot use copper because great thermal conductivity (2x better than Al) comes with greater weight (3 times the density of Al).
If this ever makes it to mass market, there will be some bloody knuckles on people who are trying to make small adjustments to the inside of their computer case while it is running. It's basically 100 butter knives spinning at 2000 rpm.
A Peltier element only moves the heat from one side to the other. If you don't have some device to cool the hot side, the element will overheat and burn - so you still need a big heat exchanger to dissipate the heat into the air.
unlikely. if you look inside a macpro, you'd see that they have 2 giant fans that cool the cpu & ram. They pipe the air so that it flows over the cpu heatsink & ram, but no other components. There's another fan for cooling the rest of the system. Apple would have to re-engineer the entire case if they wanted to use the impeller design.
I wonder how cheap and reliable are the bearings needed to run a 2000rpm 'fan' 1/1000" above a heat pipe, while being vibrated by hard drives and PSU fans, shaken around in laptops.
Generally you can have bearings that are small/fast/precise or cheap - generally not all 4.
I was under the impression that this used a similar principle to the one that keeps hard drive heads in place. So some reasonable amount of vibration should not be an issue. On the other hand, assuming a reliability similiar to hard drives, failures will happen. and in worst case, will send a weighty heatsink spinning at 2krpm flying inside your computer case. Given how collision-unfriendly that baby looks, its not something i would want to happen inside my desktop box.
But the heads are mostly kept above the surface by aerodynamic effects - effectively they are held in a thin air film which stops them touching the platter.
This new heat sink / fan is like the heads in the hard drive: it floats on a thin film of air. Consequently, it does not need any conventional bearings. The video describes it as "using an air bearing".
It isn't an air-bearing by any normal use of the term.
It still needs a radial bearing to attach it to the motor - especially if you aren't mounting this perfectly flat and level. It can use an air cushion effect to reduce the planarity demands of the bearing to keep the heat transfer surfaces parallel - but this is very different to 'floating' a 10mg sprung drive head
It's called a "hydrodynamic air bearing" in the text, so presumably there are no traditional bearing parts. The question would then be the cost of machining the surfaces to the right tolerance.
