20
votes
Piezoelectric fan to potentially replace all traditional fans in electronics
Link information
This data is scraped automatically and may be incorrect.
- Title
- The Most Exciting PC Hardware in YEARS.
- Authors
- Linus Tech Tips
- Duration
- 16:55
- Published
- May 30 2023
This can potentially make traditional fans within electronics obsolete, and is moving well past concept stage. Thoughts on this and where else this can really change things up?
Im thinking of AIO coolers where the air flow moves across the long axis of the cooler.
Keep in mind thse are also more efficient than traditional fans. If you have a server farm with 1000 fans, dropping from 7W per fan to 2W is a 5000W savings, which translates to (at $0.15/kwh) $6,500 a year. Not even factoring how with modern chips better cooling translates to better performance.
Floorspace is also a premium. If you can shrink a 2U server in half courtesy of better cooling tech, you've doubled your computing capacity.
I could totally see these things hitting the enterprise space if they get unit cost down.
It sounded like cost was a matter of scale of production and more RnD to increase manufacturing efficiency. The Thermoelectric cooling always seemed more of a novelty or a specific use case to me due to how power inefficient it was. Peltier generators (if I remember correctly) were briefly covered in engineering school with the equations shown in the performance section of the wiki so the inefficiencies were demonstrated to me on a fundamental level before I even knew what an AIO was.
The paper filter intakes feel like a weak point vs fans as they seem to be prone to be clogged by dust and only require a small amount before performance is affected. While a regular fan that can take a fair amount of dust before efficiency really drops. I suppose a robust and easy to clean pre-filter can address that however.
Im also curious as to the lifespan of the fan is - if it's made by laser etched/cut single material it may be prone to thermal and mechanical cyclic stress, depending on the range of motion the material is put through.
About where in the video do they mention reversing airflow? My understanding of the tech in its current configuration is that it's one way.
It's at about 7:40. Though maybe it's not that the the fan itself can blow backwards. Rather it might some separate thing that allows for air to be blown in reverse.
Edit: even at 10:34, Linus says "one-way valve operation." So yeah it's definitely something else.
Linus mentioned himself on the WAN show last night that he doesn’t see these replacing fans in desktop computers anytime soon, given the cost. But, in high-end small form factor electronics, these could allow for more efficient, slimmed-down cooling.
Personally, I think it’s pretty neat tech. I could see this being useful in handheld game consoles (Steam Deck, Switch, etc) and thin-and-light laptops. I would imagine the lack of (macroscopically) moving parts would help with durability, too.
I suspect it'll more remain something for specialized applications where either the size or the shape make traditional cooling difficult.
Unless this changed (sorry, didn't have time to watch the entire video) then the cooling coefficient of these devices is still overall rather poor. So from a power-perspective they'd be used where "better" cooling cannot be mounted.
Although, then again, everything is a tradeoff. We don't all liquid-cool our PCs because unless you're an enthusiast you probably don't want to mess with liquid-cooling systems. This might further decrease maintenance for consumers, which could make it win out in the end.
Is there any data on the expected lifetime of these devices under "normal" use, btw?
Unfortunately no cooling capacity was provided. No information on longevity either.
I've expressed my concerns with the paper pre filter, which with my limited design experience, seems like it would be susceptible to dust clogging. Also the longevity of the piezoelectric actuators as they'd be susceptible to mechanical and thermal cyclic stress. I believe the amplitude and therefore the magnitude of the piezoelectric actuators can be controlled via voltage control, if I'm remembering the equations correctly. This means the material can stay under the fatigue limit for cyclic mechanical stress and have an indefinite lifespan.
Cleaning a paper filter sounds easier to me than trying to de-gunk an entire fan. In my experience every moving-air cooling system needs dusting occasionally.
I imagine they've stress tested their actuators. And all moving parts are susceptible to failure so I don't know how that's more or less of a worry than with a traditional fan. My last laptop went through two fans because of mechanical failure.
My intuition, which could be wrong, is that with the higher static pressure and small area of intake, cleaning would be more frequent and more essential to performance.
They've definitely stress tested the actuators. However fans sit on bearings in a designed system for longevity, experiences frictional wear. The actuators on the other would experience cyclic bending stress which can potentially induce microcracks, leading to cracking failure. Due to the nature of the mechanism, the actuators probably have a very high rate of cycles. But I'm sure they've done their due diligence in design the system namely, the material chosen, the amplitude of actuator displacement, and shape of actuator.
I'm not saying these are definite or even likely, just pointing out areas to consider and look out for,
My last two fan failures had nothing to do with the bearings and everything to do with dust gunking everything up to the point the motors burned out. As Linus points out in the video, higher static pressure means you can filter on the device intakes too, which makes cleaning even easier. I wish I could just run a vacuum along my laptop's air intake once a month or so instead of taking the bloody thing apart for deep cleaning a few times a year..
Feels like a solution without a problem for 90% of consumers. These things are more expensive than traditional PC fans, push less air, and for most people a traditional PC fan will last the lifetime of the computer. And even if it does break, they're cheap enough that its not a big deal to replace them.
Just because a solution exists doesn't mean it can't be improved on.
Flat screen TVs were expensive at one time, but because of increase in efficiency of scale and manufacturing methods, theyre a dime a dozen now.
While these push less air, they have a higher static pressure and carry more heat per volume of air (how I'm interpreting when they say saturating the air with heat). High volumetric flow rate fans and high static pressure fans have their own specific applications.
This could prove to be not economic, but its exciting as a new technology. Desktop PCs aren't really applicable but I was just spitballing an idea. Thinbooks, tablets, and other small devices that put out significant heat can benefit from this. High end production cameras can definitely utilize this, and it was mentioned above this could be put to use in enterprise servers. I'm sure there's many other industrial applications that this can be put to.
i don't know whether its because i'm getting older or if i'm just getting bitter, but i have found as of late LTT's videos have been increasingly difficult for me to look at, causing what i suppose you'd call "content sickness"
too much moving text and wacky effects, its just tiresome
admittedly it is less noticeable here due to it being an expo video
It's definitely been following the mainstream with its effects and scripts, but they have overall great content so I put up with it. I only actively watch their videos that interest me tho, which isnt as often. This was one of few recent ones.
Also, they don't let themselves be beholden to investors and outside influences.
I think I saw this tech being demo'd at someone else's coverage of CES. The amount of air being moved was surprising. I think this tech is better suited for mobile platforms, but I'd want to know more about thermal characteristics.