Solid-state fans

Jump to latest Follow Reply
CuriouslySane

CuriouslySane

Ars Praefectus
3,736
Subscriptor++
  • #1
    Saw this posted elsewhere. It's CES-aganda about a product called AirJet from https://www.froresystems.com/


    View: https://www.youtube.com/watch?v=YGxTnGEAx3E


    The TL;DW is that it's a solid-state/MEMS device that creates high pressure, directed airflow within a similar 1W power budget as current laptop fans. A lot will come down to the balance between cost, power budget and effectiveness, but it certainly sounds promising. Even if it's pricey, people will pay for silence, and if it's really effective then it could be a shake up.
     
    U

    uno2tres

    Ars Scholae Palatinae
    964
  • #3
    Inside AirJet® are tiny membranes that vibrate at ultrasonic frequency. These membranes generate a powerful flow of air that enters the AirJet® through inlet vents in the top. Inside AirJet® the air flow is transformed into high velocity pulsating jets.
    Click to expand...

    I’m a little skeptical that these will actually be silent once integrated into a real device. There’s a ton of opportunity for weird resonance that leaks audible sound.
     
    • Like
    Reactions: continuum
    A

    Astrallionheart

    Ars Scholae Palatinae
    787
  • #4
    maxresdefault[1].jpg

    I mean the SP7 i5 magnesium chassis that Microsoft gave up on in 2019 (because Intel CPUs were going nuclear with power consumption and they couldn't keep up with the increasing power consumption every generation) was already cooling a ton of heat passively, and it was literally silent. These things will not be completely inaudible if you have to put 12 to 16 of these in the chassis to cool 50-65W.

    It probably makes more sense to wait for Intel/AMD to try to catch up with Apple's efficiency before launching standard fanless designs again.
     
    CuriouslySane

    CuriouslySane

    Ars Praefectus
    3,736
    Subscriptor++
  • #5
    According to their site...
    AirJet Mini removes 5.25 Watts of heat at a silent 21 dBA noise level, while only consuming a maximum of 1 Watt of power.

    AirJet Pro removes 10.5 Watts of heat at a silent 24 dBA noise level, while only consuming a maximum of 1.75 Watts of power.
    Click to expand...
    That's not silent, but it's pretty darned quiet as long as it's not hitting some weird, whiny resonance. We'll see how long it takes for someone to figure out how to cram some into their SteamDeck.
     
    • Like
    Reactions: Crusaderking04
    ScifiGeek

    ScifiGeek

    Ars Legatus Legionis
    16,351
  • #8
    CuriouslySane said:
    Saw this posted elsewhere. It's CES-aganda about a product called AirJet from https://www.froresystems.com/


    View: https://www.youtube.com/watch?v=YGxTnGEAx3E


    The TL;DW is that it's a solid-state/MEMS device that creates high pressure, directed airflow within a similar 1W power budget as current laptop fans. A lot will come down to the balance between cost, power budget and effectiveness, but it certainly sounds promising. Even if it's pricey, people will pay for silence, and if it's really effective then it could be a shake up.
    Click to expand...


    Interesting, but inlets are tiny and seems like they would easily clog with dust in short time.
     
    ScifiGeek

    ScifiGeek

    Ars Legatus Legionis
    16,351
  • #12
    CuriouslySane said:
    They claim it's "dust proof", and the mockups for an ultrabook show several in an array, though that doesn't help if your hotspot exceeds the envelope for a single unit. $100M of startup funding suggests they're onto something though.
    Click to expand...

    In itself it isn't dust proof. It's compatible with certain class of restrictive dust filters. I think this is brought up because it would easily clog without restrictive filters. This also means the device will have to be completely airtight sealed except for that dust filter.

    Right now this is low heat removal solution, 10 watts for each big unit. So it seems more for thin tablets like the surface pro, more than conventional laptops.
     
    ScifiGeek

    ScifiGeek

    Ars Legatus Legionis
    16,351
  • #14
    Doomlord_uk said:
    THey mentioned it having an IP68 dust filter, and significantly (about 10x or more) pressure/suction than a normal laptop cooling fan. And given the air speeds inside, why would it clog up?
    Click to expand...

    It doesn't have IP68 filter.

    He said the notebook could have IP68 dust filters. IMO this will be required to keep the tiny inlets, MEM devices in it from dust clogging. Then you will need to keep on top of cleaning the IP68 filters.
     
    cogwheel

    cogwheel

    Ars Tribunus Angusticlavius
    6,691
    Subscriptor
  • #15
    Doomlord_uk said:
    And given the air speeds inside, why would it clog up?
    Click to expand...
    Clogging is more a function of dust and other contaminants in the air and the smoothness of the surfaces involved. For something like this, the edges and corners of the air inlets absolutely will build up dust readily, and the interior surfaces and corners will as well depending on how smooth they are at a microscopic scale. Stickyness of the air contaminants also plays a part (ever seen inside a smoker's computer?), as does electric charges on the air mover's surface (likely to be higher with this than a normal fan).
     
    cogwheel

    cogwheel

    Ars Tribunus Angusticlavius
    6,691
    Subscriptor
  • #24
    continuum said:
    www.anandtech.com

    OWC Teams Up with Frore for 32TB and 64TB SSD Devices

    www.anandtech.com www.anandtech.com

    Seems like a reasonable use case, to cool one M.2 SSD per Airjet.
    Click to expand...
    Not really. The OWC Shuttle is usually closed with a big fat heatsink plate, so thermal pads to that will likely outperform those things. Adding a quality fan to whatever enclosure the Shuttle is inserted in would provide more heat dissipation, a location to install a filter, and easy repairs when the fan dies.

    Doing some quick research and messing with a heatsink calculator, it looks like the AirJet Mini is equivalent to a 130x130x2mm (about 5x5x1/16") flat (unfinned) aluminum plate, with no airflow, facing down and dissipating heat only from that downward-facing surface. That's not much at all, and it's compounded by the fact the plate doesn't need any power to operate, but the AirJet Mini needs 1W. Orient that same single aluminum plate vertically, and it significantly outperforms the AirJet Mini. Add some fins, and the AirJet Mini is left in the dust. Add just a little forced convection, and the AirJet Mini falls off the map.

    It's neat tech, but I'm having difficulty coming up with a use for it. Industrial/cleanroom applications aren't usually extremely space limited. Spacecraft don't have air to run through it. End user devices need to cope with dust. Thin laptops are as much battery life limited as heat disipation limited, so even if this could let the CPU use more power than they currently can (and I'm sceptical the AirJet would do that), you probably wouldn't want to do that.

    redtomato said:
    If this tech doesn't appear in an Apple product soon, then odds are that there's an underlying issue with it.
    Click to expand...
    Agreed, and my guess is that for the MBA it isn't better at heat dissipation than the case itself and uses quite a bit of power (1-2W is significant when your SoC is around 10W max), and for the MBP it has the clogging/reliability problem I mentioned months ago.
     
    • Like
    Reactions: continuum
    C

    Cool Modine

    Ars Tribunus Angusticlavius
    8,539
    Subscriptor
  • #26
    ScifiGeek said:
    He said the notebook could have IP68 dust filters. IMO this will be required to keep the tiny inlets, MEM devices in it from dust clogging. Then you will need to keep on top of cleaning the IP68 filters.
    Click to expand...
    If the filter is accessible this may not be a big downside. As it is, laptop fans do get full of dust and clog as well, and it's a little obnoxious to clean them out.
     
    cogwheel

    cogwheel

    Ars Tribunus Angusticlavius
    6,691
    Subscriptor
  • #29
    CuriouslySane said:
    Hey, made it to the front page. Still seems to be in pitch mode though.
    Click to expand...
    So more of the same. "Interest" from industry, likely being dropped once the manufacturers realize the performance drop-off due to inevitable clogging.

    It's interesting how their math works, too: 3x of those clog-o-trons can dissipate 12.75W of SoC power, but somehow the SoC was using 20W and not melting? Makes me wonder how much of the gains were due to the clog-o-trons and how much were due to modifying the chassis for better thermal conduction and ending up dissipating more heat through the bottom plate than in a stock configuration. I'd love to see where just modifying a MBA to have good heat conduction between the SoC and the bottom plate would fall performance-wise.

    continuum said:
    From the front page article. UmmmHmmmm…
    Click to expand...
    Yeah, that's interesting. The Mini's datasheet still says 1W max power consumption, so where did that extra 2W+ come from? Is Frore overclocking their fans for demos?
     
    ScifiGeek

    ScifiGeek

    Ars Legatus Legionis
    16,351
  • #30
    cogwheel said:
    ... and how much were due to modifying the chassis for better thermal conduction and ending up dissipating more heat through the bottom plate than in a stock configuration. I'd love to see where just modifying a MBA to have good heat conduction between the SoC and the bottom plate would fall performance-wise.
    Click to expand...

    Some people did that mod. It's not a good idea, because then you have the metal case reaching unreasonable temperatures.

    I don't doubt these devices can improve the thermal performance of thin previously passive devices, so they have a niche. But I don't think it's a huge niche.
     
    ScifiGeek

    ScifiGeek

    Ars Legatus Legionis
    16,351
  • #34
    cogwheel said:
    That's a straw man.

    It would be throttled to maintain the same chassis peak temperature at equilibrium as the standard configuration. I grant that the difference might not be that much, but it isn't zero.
    Click to expand...

    It's not a strawman, and that's not what happens. The chassis will get MUCH hotter than it does in the standard config.

    In the standard config, it throttles much sooner, and that limits the chassis heating up.
     
    You must log in or register to reply here.
    Jump to latest Follow Reply
    Jump to latest Follow Reply