-
Save 15% on ALL SFF Network merch, until Dec 31st! Use code SFF2024 at checkout. Click here!
Cooling What ever happened to Coolchip?
- Thread starter hat1324
- Start date
They partnered with Cooler Master and showed off some prototypes at CES this year: http://techreport.com/news/27637/cooler-master-kinetic-engine-turns-rotating-heatsink-into-a-fan
Haven't heard a peep since then. I suspect we'll see closer to release units at CES next week.
Haven't heard a peep since then. I suspect we'll see closer to release units at CES next week.
Last edited:
I was and am mostly skeptical about the technology from a moving and orientation standpoint. I'm not sure but I've not seen the cooler in a vertical position and what happens if you bump your case while it is spinning. Will you have a 3000 rpm aluminum piece of metal fall and hit you GPU, will it chip against its own base and fail or sound like nails on a chalkboard? Probably all solvable but I'd like to see it demonstrated.
I love the idea of an air bearing but air bearings are usually meant as a precision bearing.
I love the idea of an air bearing but air bearings are usually meant as a precision bearing.
If I understand their design correctly, the bearing occurs along a narrow rod in the center that is locked in place. I doubt any design in which the heatsink pops off would ever leave the drawing board
The problem they are facing is that the main bearing has to be extremely precise, carry a lot of weight in all directions and be suited for high RPMs at the same time in order for the heatsink to be both effective and robust. This is probably the issue QinX is seeing here and I fully agree that this is a complex engineering challenge.
It looks like it is constrained radially on a standard bearing that mounts on the shaft (dark grey piece on the female side):
From my understanding the two grooved surfaces form the air bearing, creating an axial constraint in one direction (looks like there is a nut keeping it from popping off in the other direction.) The thin amount of air has limited insulation and heat readily crosses the air bearing to the fan side where it can be dissipated.
If I recall correctly, an air bearing has tolerances measure down in the millionths of an inch...most typical manufacturing is done in the thousandths of an inch (a good CNC mill or lathe is good to around 1/1,000th in. and a precision lapper/grinder is good to around 1/10,000th in.) The manufacturing process is rather difficult and the surface between the mounted heat sink and the moving fan has a grooved pattern to it, so it isn't a simple flat surface, either.
From my understanding the two grooved surfaces form the air bearing, creating an axial constraint in one direction (looks like there is a nut keeping it from popping off in the other direction.) The thin amount of air has limited insulation and heat readily crosses the air bearing to the fan side where it can be dissipated.
If I recall correctly, an air bearing has tolerances measure down in the millionths of an inch...most typical manufacturing is done in the thousandths of an inch (a good CNC mill or lathe is good to around 1/1,000th in. and a precision lapper/grinder is good to around 1/10,000th in.) The manufacturing process is rather difficult and the surface between the mounted heat sink and the moving fan has a grooved pattern to it, so it isn't a simple flat surface, either.
The worry is that with such a precision fit, even with a precision shaft it's reliant on the air bearing to prevent a touchdown and permanent (a contact at high RPMs is going to squish the mating surface and jam the entire thing into an interference fit) jam. A sharp shock or rotation during operation could well be sufficient to kill the heatsink. For a server, this is not so much of an issue (apart from ruggedised servers), but it may make the heatsink a no-go for laptops, or even desktops.
Remember the disc-scratching issue with the Xbox 360 (not from tilting the console, but from normal operation in horizontal or vertical orientation)? That was a MUCH looser fit than here, but even minor vibrations from people walking nearby were sufficient to cause damage.
Remember the disc-scratching issue with the Xbox 360 (not from tilting the console, but from normal operation in horizontal or vertical orientation)? That was a MUCH looser fit than here, but even minor vibrations from people walking nearby were sufficient to cause damage.
This looks familiar:
Dutch article: http://be.hardware.info/reviews/669...ft-twee-jaar-garantie-op-consumentenproducten
Translated:
To keep the Core i7 CPU and desktop GTX 980 temperatures at bay, Acer uses three fans in the system. Remarkable is that these fans are made of metal instead of plastic; this way they can have more fins/blades than usual and generate more airflow. The three fans are positioned in a way that air moves from the front to the back of the laptop. Additionally Acer (?) has developed a technology which detects when there is too much dust inside the laptop; the fan's rotation is then temporarily reversed to exhaust dust from the laptop. The Predator 17X is expected to arrive to market in July and has an MSRP of 2699 euro.
Dutch article: http://be.hardware.info/reviews/669...ft-twee-jaar-garantie-op-consumentenproducten
Translated:
To keep the Core i7 CPU and desktop GTX 980 temperatures at bay, Acer uses three fans in the system. Remarkable is that these fans are made of metal instead of plastic; this way they can have more fins/blades than usual and generate more airflow. The three fans are positioned in a way that air moves from the front to the back of the laptop. Additionally Acer (?) has developed a technology which detects when there is too much dust inside the laptop; the fan's rotation is then temporarily reversed to exhaust dust from the laptop. The Predator 17X is expected to arrive to market in July and has an MSRP of 2699 euro.
Yup, it's just a more solidly built fan.
It IS nice to see the filter-cleaning flow-reverse technique from industrial dust-extractors used for PC cooling. The ability to periodically reverse fan flow (e.g a 'blow' cycle on system startup) would make restrictive dust filters a lot more useful, as the dust that clogs them would be regularly and automatically ejected.
It IS nice to see the filter-cleaning flow-reverse technique from industrial dust-extractors used for PC cooling. The ability to periodically reverse fan flow (e.g a 'blow' cycle on system startup) would make restrictive dust filters a lot more useful, as the dust that clogs them would be regularly and automatically ejected.
As long as it occurs regularly (e.g. once a day rather than waiting for a filter to clog) then it'll effectively just be keeping the dust already in the room from accumulating on the filter. Most accumulation is due to forced airflow rather than passive deposition, so even if the PC is left turned off for long periods it should not accumulate more dust than would settle on any other surface in that room.
Good point. How dusty the environment is and how long the machine is running each day would be huge factor, too. You would need an additional pin on the fan connector for this sort of thing to be backwards compatible, right? Would it even work properly if the fan blades aren't specifically designed to exhibit sufficient airflow/static pressure when turning backwards?
This reverse spinning thing seems to be becoming quite popular. Here's a series of axial fans that just got announced today:
https://www.techpowerup.com/222061/enermax-announces-the-new-dfr-technology-with-d-f-pressure
And I'm pretty sure Enermax aren't even the first ones. I know the stock ASST_FAN on my Z97 Gryphon can spin backwards, haven't tried with the Noctua I'm going to replace it with.
There's probably nothing special about the motor itself, it just needs a reversed current. I bet there's a dedicated motor controller on board that does this, all the Super IO chip needs to do is change PWM frequency or duty cycle. It's a common approach in other applications (i.e. Cytron MD20C). Centrifugal blower-type fans generally don't rely on blade shape too much, I haven't seen many variations.
https://www.techpowerup.com/222061/enermax-announces-the-new-dfr-technology-with-d-f-pressure
And I'm pretty sure Enermax aren't even the first ones. I know the stock ASST_FAN on my Z97 Gryphon can spin backwards, haven't tried with the Noctua I'm going to replace it with.
There's probably nothing special about the motor itself, it just needs a reversed current. I bet there's a dedicated motor controller on board that does this, all the Super IO chip needs to do is change PWM frequency or duty cycle. It's a common approach in other applications (i.e. Cytron MD20C). Centrifugal blower-type fans generally don't rely on blade shape too much, I haven't seen many variations.
MSI did it with their GTX 580 in 2011: http://www.engadget.com/2011/09/07/msis-reverse-fan-tech-keeps-dust-out-graphic-cards-chilled/
Similar threads
