I don't think so. They mention the chip has a cavity and some MEMS membranes which push the air forcefully right into the metal heatsink (200km/h pulsating jets), thus breaking through the boundary insulating layer, thus getting better efficiency (it pumps more heat in the same volume of air). Peltier cell pumps the heat from one side to the other. This device moves that heat into the air and has pretty good back pressure from what is shown in the video. And it seems to also be quiet.
I googled mems cooling and saw some references to nasa work. Pretty old idea (2003ish) and is based on sterling engines. Interesting idea. Sounds like someone has found a way to get it down to practice. Interesting and quite practical. Another reference to the current device said only generated 21db of noise. Great for a low noise HT system.
Yes indeed quite interesting. Yet I think they have some scaling limitations. They seem confined on the horizontal plane, the depend on surface more. There might be some room to improve but I don't think too much.
Quite interesting that they managed to cool an over 20W device with a few of these. And they did mention they're looking at 100W as next target.
Also I don't know how this tech fares in time. Clearly you need a particle filter, but how does it deal in time with oils on the internal cavity.
Quite interesting that they managed to cool an over 20W device with a few of these. And they did mention they're looking at 100W as next target.
Also I don't know how this tech fares in time. Clearly you need a particle filter, but how does it deal in time with oils on the internal cavity.
No, it´s a real air pump, to be more precise a copper block with slots, and his SS pump blowing fresh air through them.
Very interesting.
In any case, no matter what technology, it all boils down to air flow in cubic feet per minute, not sure they mention it.
Maybe in the small print.
Very interesting.
In any case, no matter what technology, it all boils down to air flow in cubic feet per minute, not sure they mention it.
Maybe in the small print.
Long term reliability is unknown compared to fans and heat sinks that are common today so time will tell
If the that interface is copper hit by vapours at ultrasonic speeds copper it's probably ok, but if it's using open air vents, the silicon and aluminium oxide found in the air will penetrate and ripp off copper particles from the heatsink so it has limited span life in an open design.Listening to micro ultrasonic booms at particle levels might not be very cool if used in audio systems ...
No. You have to "scrub" through the stagnant layer of air on the fins. Unless you do that, you can have a lot of barely-warm air come out. What cools better, 1CFM at 5 deg rise or 1CFM at 100 deg rise? As someone just said:
Conventionally you improve that with near-sonic air speed, or at least "highway" air speed like in the car radiator. But that is loud.
Piezo flappers have been around for decades with excellent reliability but very tiny air-flow and not cheap. SEMS is another technology with some maturity and deep experience (modified chip-making). It may again be not-enough for too-much, but seems worth considering.
Of course it is all vapor-ware until it is on NewEgg{.}com at $13 and 4.5 Eggs. I'm not building fighter-planes or crypto-diggers.
The question of fouling on heat exchangers is also there.
A dirty surface is a poor heat transfer surface.
This vibration is going to cause mechanical stress to the chips, particularly the plastics and the bonding wires.
I think welding and cleaning when I hear the word 'ultrasonic'.
Like @PRR above, I am deeply skeptical.
A dirty surface is a poor heat transfer surface.
This vibration is going to cause mechanical stress to the chips, particularly the plastics and the bonding wires.
I think welding and cleaning when I hear the word 'ultrasonic'.
Like @PRR above, I am deeply skeptical.
This may fit just well in the whole "buy it and use it for few months then buy a new one" narrative.
If they lose performance by getting dirty I'm curious if they can be revived with an ultrasonic bath in Isopropyl alcohol.
If they lose performance by getting dirty I'm curious if they can be revived with an ultrasonic bath in Isopropyl alcohol.
They had to use 4 air-jet minis to get 20W of TDP, sustained. That's 4W of energy drawn from the laptop battery, just to cool the CPU. I'd expect that two fans would, for the time being, do the same at a lower internal battery drain. Nevertheless, it's close energy consumption-wise, and it's completely quiet. I'd also expect that by balancing the air-jets movement, the vibrations could be annulled... so, yeah, bring it on. Can't wait to see commercial implementations...
Peltier Effect heat pumps were popular at one time.
And gamers keep doing something new to improve performance.
You can look that up.
And gamers keep doing something new to improve performance.
You can look that up.
Peltier coolers are worthless for anything besides cooling a CPU.
I throw out one in a yacht , it draw 7A and managed to cool the fridge to 10degrees over night.
The replacemet compressor used 2.8A when running and managed to cool the fridge to 5 degrees in a few hours, then it run 30% of the time to keep cool.
As i understand it peltier elements degrade by age and get even worse.
I throw out one in a yacht , it draw 7A and managed to cool the fridge to 10degrees over night.
The replacemet compressor used 2.8A when running and managed to cool the fridge to 5 degrees in a few hours, then it run 30% of the time to keep cool.
As i understand it peltier elements degrade by age and get even worse.
Safer than this ultrasonic emitter, could damage components.
There is little space in laptops to put larger cooling fans, maybe a different design, running slower, with more blades in a different shape, for more air flow, and a circuit to fool the controller monitoring it?
That may be a better solution than this or coolers.
There is little space in laptops to put larger cooling fans, maybe a different design, running slower, with more blades in a different shape, for more air flow, and a circuit to fool the controller monitoring it?
That may be a better solution than this or coolers.