My Subscribers Turned MIT Award Winning Toroidal Propeller Into A PC Fan | Fan Showdown S5E5
ฝัง
- เผยแพร่เมื่อ 14 ต.ค. 2024
- New Customer Exclusive - $100 Off Creality Ender 3 Pro 3D Printer: micro.center/dv5
Shop Micro Center’s Winter Savings: micro.center/2d2
Check out Micro Center’s Custom PC Builder: micro.center/ulm
Submit your build to Micro Center’s Build Showcase: micro.center/5wa
Ok the reach of last weeks video was unexpected to say the least. However I could not think of a better follow up than to test some of your MIT inspired Toroidal designs. This week for this new episode of the fan showdown I looked for 4 Toroidal inspired designs. Can you produce a toroidal fan that outperforms mine.
My Twitch
/ majorhardware
Material Used Ziro PLA
amzn.to/3HICToF
Fan Showdown Instructions
• EVERYTHING you need to...
Fan Dims
www.thingivers...
Submit your design (at least .STL file)
fanshowdown@gmail.com
Cameras used.
BlackMagic Pocket cinema 6k pro
amzn.to/3m2RLVw
Reddit
/ fanshowdown
With these submissions got me thinking: we may need a new measurement index, a "noise-to-airflow" coefficient of some sort to see which design can output highest static pressure with comparatively lowest noise. AWESOME job as always, and keep it up Fan Showdown Community!
Yo dawg, you have a pretty high NTAFR
@First name Last name I think we'd go with "hey bro, your fan has good NTAC"
Damn good idea
Noise normalized pressure/airflow. AKA set RPM so noise level is the same for every fan, and then test performance.
@@KellyWu04 An interesting idea, I'm really curious if doing things that way would actually change the NTAC rankings, or just be a different way of measuring the same thing?
The toroidal propeller is not optimized for a ducted application. Its features are a comprise for certain performance characteristics in open channel flow.
Very much true, the toroidal propellers are quieter because their design eliminates the noise cause by the tip of the prop going near super sonic
Excellent point.
exactly!
the whole point of a toroidal fan is to have a ducted fan but without the ducting.
hence why the standard fan on the A12X25 severely out-performs them as it takes full advantage of the duct, and why all the fans which accentuate that duct are the ones which perform better (at the cost of form factor and noise).
personally I'm just waiting for the cheater 2 electric boogaloo which keeps the same low noise of the standard fan but with improved performance as the whole point of this show it to make a better PC fan, and noise plays a huge part in that, he should really put a performance-to-noise graph on there and rank the best fans based off of that as it would clearly show the true best fan and prevent future instances of the cheater's reign.
Yeah came here to say this.
I've been screaming this at my phone since the first toroidal fan. The close fitting duct minimizes tip vortices already.
you should print the A12x25, that way it also has the ridges that everybody else has.
yep, that would be fair.
There's only a difference in noise and it's a very minor one. Print quality and smoothness makes very little difference. An A12x25 would register about 46.3-.5 db printed rough as these.
@@KaiserTom he just did the vid where he printed the A12x25, their was a 2.6% diff in airflow. which if you add 2.6% to all the submitted fans changes the leaderboard. the angel fan plus the 2.6% out preforms the A12x25.
OH! finally a comment nobody has ever made ever before.
Imagine if your phd depends on how good your fan does in the fan showdown
That blows
Lol, academia!
As an added challenge, you have to do your dissertation defense while shouting over the sound of all of the toroidal designs that made it onto the show.
"Imagine if your phd depends on someone else's crappy 3d-printer and lack of ability to use it, and not your own work."
Yes, I'm that fun guy in the party that gets everyone laughing by pointing out how jokes are stupid. At least based on how they all laugh and point at me.
@@anteshell it's an even playing field then
One thing people kept forgetting is fan design is coupled with Reynolds number, which is a dimensionless number predicates flow properties. Fluid (including gas and liquid) acts a lot differently depend on speed, density and object size.
MIT’s paper specifically targeting water. So similar design to that paper may not work well in air.
A classic case is 3-blade propeller on a boat is a lot different compare to a 3 blade propeller in your ventilation fan, or on a 3 blade propeller driving a plane.
Guys like you are why I always look at videos with apprehension. Your points are definitely a reason I doubt those who doubt toroidal design are of ANY value. Thanks.
@@uprightfossil6673 There is a video of someone designed torodial propellers for a drone, using a two blade design, but a drone's propellers spun a lot faster than computer fans.
Key point: Yes it may be more efficient, but it only means less "wind noise". The fans may not be pushing more air due to the extra weight weight.
In aeronautics/aerospace engineering, similar principles were tested (loop wings) but the weight and extra drag of the loop out-weighted the benefits of efficiency increase.
@@AaronShenghao I am specifically interested only in the movement of water. I have several pump applications and want to make my system as efficient as possible since it is on solar panels and remote. You mentioned the Reynolds’s number. Can you give me some other terms to shorten my learning curve? So far my biggest project has been the wind powered water wheel using a hose to lift water 2meters/yards.
Aaron is absolutely right about the MAJOR significance of Reynold's number to understanding and optimizing propeller performance.
Further, and I know this is a large barrier for 'printed' models, a blade's trailing edge sharpness has a stunningly large impact on thrust performance. I hope amateur experimentation with these considerations in mind will be performed.
I'll wager US government researchers are well aware of these technical considerations and might be able to turn out usefully quieter drone blades going forward. Or this approach may be a dead end for drone purposes. Design and test!
@@ivarhusa454 agreed. Turbine engines have smaller blades in front for a reason
Been watching this series since the very beginning and it’s still fun to watch. Hope the fan showdown continues on for years to come.
Agreed!! And he needs to do a best of the best every once in a while!
I think most of the advantage you'd get from a toroidal shape are irrelevant in a computer fan, because there's already a wall around the tips of the blades. Also looks like the 7 bladed toroidal and thoroid are working near the hub, but ineffective near the tips, the smoke gets sucked from the edge halfway to the center.
Exactly! I commented the same thing on the last video. The whole point of this design is to reduce wingtip (blade-tip) turbulence, but that's also the point of the fan shroud. Having both is pointless. Using a toroidal fan inside a shroud is simply using a compromising design to solve a problem that is *already solved;* leaving you with a strictly worse result.
@@benjaminmiller3620 Still an interesting shape to 3d model though.
Hear me out.. Who says that PC fans NEED a fan casing ? What if a manufacturer came up with a toroidal fan with the axis mounted directly into a heatsink ? Basically a fully integrated fan / heatsink design.
@@AnonAnono Sounds like the old Zalman coolers.
@@wolfie54321the main benefit is sound. Plus, toroidal fans are not optimized for ducted applications. And 100% efficiency doesn't exist practically speaking, so why even refer to that ?
this guy has a serious fan base
The puns are killing me.....
@@crimsonemperor2219 right? I'm constantly blown away
🤣🤣🤣🤣🤣🤣
@@manguydude287that comeback was phuntastic
It’s great to see all the fan art.
I just know that a fan blades which outperforms the Noctua is going to be found on this chan. It's just a matter of times. It might require a special surface finish though because raw 3d prints are ROUGH.
He should move to resin printing.
He tested surface finishes before. Result was that air flow changes was negligible but noise was increased.
That said maybe certain shapes are more relient on surface finishes than others?
@@ZipperOfficial the edges of the blades should be a focus as well. It can have a noticeable impact on flow. Those ragged edges can create a surprising amount of turbulence.
Sometimes it can even be an advantage to have this little sausage lines on the surface, but 99% of the time resin will be superior and should definitely be used.
@@ZipperOfficial It's not even surface finish it's the fact that he can't make detailed features with any fidelity.
To test them better, you should 3d print the Noctua blade to the same levels and surface finish that you are printing the toroidal fans.
Surface finish on a blade is very important and is often over looked. It is why dust on the fan blade reduces it efficiency.
This is what the Internet is made for. Community coming together to contribute to mutual interest. So wholesome
No. Internet exists to serve as an echo chamber and to reaffirm all the awful ideas I have.
Inverted Sharrow is a good blade design but still well within the realm of "needs more fins." I think it could stand up to a 5- or 6-bladed Noctua clone easily. Either way I don't think these things are meant to be used in ducts like fan frames, but it could work anyway...
I would love to see some experimentation with various fan parameters. Explore the space of different numbers of fins, different fan pitch, scallops or serrations on the blades, etc.
The RPMs on all these toroidal fans were pretty high. They could use more aggressive angles or more blade area?
I agree, the duct is already doing the job of the toroidal design and dampening the tip vortices, still interesting though.
You turned me into a fan. Haven't missed a video since.
wow he really can make anything into a fan
What was your flow rate and DB level?
@@GamerPro132 I usually run out of breath and black out before the testing is complete. I'll get back to you.
I get it
This entire sub thread is gold
I gotta wonder how some of these would do w/ a light hand-sanding vs the rough lines left by the additive manufacturing process
A quick way to smooth 3D prints is to dip them in acetone.
Depending on what plastic is used, a couple of seconds could be enough to leave perfectly smooth finish.
One of the early tests on the channel was resin printing and vapor smoothing. Decreases noise, but airflow is pretty much unchanged.
Also, resin is less stiff than PLA, so resin-printed blades tend to flex.
Resin printing for less sanding and it's easier
@@louisvaught2495 What Resin are you using? My experience from Resin printing been very stiff. For tough prints I often have to pay 2x my normal stuff. I usally buy Elegoo (For parts that don't deal with stress), and Siraya Build for functional parts. But they are quite stiff still. More so than PLA or PETG.
The only time I get flexible prints are resin that are designed to be like rubber (AnyCubic flexible). Or Engineered resin where I pay a fortune (Forgotten the name but it's 4x my normal stuff, but it's quite amazing stuff)
Louis Vaught it decreases performance and noise, nothing is free.
Just want to put my two cents. I live in a country without micro centers.
I would definitively check them out, especially for hobbyist electronics and 3d print if I were in the States.
Continue sponsoring this channel please, it's the right audience.
agreed, if only Microcenter was here in indonesia
Love that I waited to watch the first video until like 10 mins ago, and then check to see other channel videos to see that this one was only an hour old. We love to see creative growth and problem solving in THIS SHORT OF TIME.
I've just binged this entire series and I love watching the production quality and testing methodology steadily improve. I appreciate that you still test every variety of fan - from the simple to the needlessly complex.
This design is one of the most important things to ever happen in fluid propulsions systems. This is going to save so much money in fuel and it's going to make all of our subs quieter and make our weapons stronger. And that's before we talk about trade.
I say its dead in the water - wanna bet?
This has been around long enough the military has already exploited the benefits
@@mchldpy yeah but the propaganda is so good.
Crazy question, what would be the results of a 3d printed version of the Noctua fan blade for a reference? I saw the stock Noctua fan slaughter the toroidal, but was thinking that in all fairness Noctua is all around a high-quality build
This is actually a really good question! We should probably have both, the original Noctua, and the 3D printed one on the list for reference. Then we'd (probably) be able to see the tiny bit of potential going from 3D printed to mass produced with proper hardware and materials.
the mobius did the same regardless of orientation because your anemometer doesn't differentiate between wind directions. in one orientation it was pushing air through the tube and in the other orientation it was pulling air through the tube.
If you flip the orientation of a Mobius strip, you get the same shape, thanks to it being a non-orientable surface. You gotta flip either the chirality or the motor direction to get an "other side" result
I loved when the fan said "it's morbin' time" and morbed all over the test
The one pushing in the wrong direction would get the same in either direction because it is PULLING the air through your airflow meter rather then pushing. The rounded props are likely going to do nothing advantageous in a framed fan because there are no vortecies on the outer edges like a plane wing, or boat prop. A jet boat impeller would likely do nothing, because again, there are no outside vortacies . :)
Sounds right
Pulling with a fan is less efficient than pushing though, so not "the same".
glad to see another viral for you. I found you way back when you watered down your hyper 212 (aggressively) and have been thoroughly enjoying since. Thanks for the great content and can't wait to see more.
I would be interested in "unducted" measurements of your best fans compared to the toroidal fans.
In my mind the value of the toroidal shape is to overcome drones not having ducted fans (for weight purposes).
First time I've watched an ad intentionally. Part compatibility matcher is an amazing idea, way better than the pain of scouring through marketing docs to find it.
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦.........
What I'm interested in is how energy efficient (CFM/Watt) the designs are. If you could measure the milli-amp draw for the a few key blades along with these new innovative designs and add a column in you spreadsheet for current draw &/or CFM/W that would be Awesome.
I understand that energy efficiency in a PC fan is not a big deal, but it is a key factor in the interest around these kind of designs.
I may be mistaken, but I thought you had gotten a resin msla printer. If so why haven't we seen you printing these fan blades on it?
Likely answer is that resin is expensive and hard to work with. Not that the results can't be worth it, but I'd expect most prototyping to be done with FDM first, for ease of use - Even with support material involved.
@@Woodledude it's also pretty fragile. Even after a couple of days in the sun, resin prints are just very soft and crumble easily.
I've seen a lot of these vids and it's crazy to think so many people have only seen last weeks video, made some surface-level judgement and just moved on from your channel. TH-cam is a crazy place. Keep on keeping on, I loved getting reminded of your vids
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦........
@@afamhohm7802 Thanks for the reply "afamhohm7802" but I know for a fact you don't have a gift for me
Wanted to ask a question. Could you make a video on the push/pull configuration using different sized fans in the front and back of the radiator? I like the way you run your testes and I wanted to see what the difference is to having the same size fans on both sides of the rad, two different sized fans on the rad or one normal fan in push or pull config.
I like that this kind of technology is really old. I still have my 15-20y old Science illustrated where they talk about using these kind of "continual shape" propellers to cut down on noise emission from prop planes (drones were not a thing back then).
Good video. One problem that I think might be diluting these results is the layer resolution of the prints. I think you really need to have a resin printer or to be making moulds if you want to honestly compare these designs with Noctua et al.
Also I don't think we got a noise test for the flipped Mobius design? Maybe I missed it
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦........
What makes the toroidal good for marine and drone applications is the reduced tip vortex, on a computer fan, the fan shroud takes care of that already, so their main advantage is discarded, and a less efficient but larger blade will move more air usually so having a large blade and the shroud taking care of vortex is a best of both worlds deal, making it sort of a ducted fan
Hello bearded bald man with cool pc fans
I think it would be cool to include print timelapses of the fans printing while you're explaining what they are.
that's a whole lotta footage.
that thing about thoroid giving the same FPM bothsides is really intriguing, with there was an aeridynamic study of the fan
I am very curious to see the static pressure for the Mobius. That is really interesting behavior.
Yes, fan showdown is an open source fan test/optimization community and i'm really happy about it
My biggest hope going forward is seeing him improve the print quality by either changing to resin/UV printing or spending time tuning this current machines and sanding the parts. The stringies and jagged edges are so prominent on some of these that I know they are changing the airflow properties of the fans. Also, I think resin would be great because the parts will come out smoother, stiffer, and closer to an injection molded fan like the original Noctua. The concept of toroidal fans is cool however I don't think it's going to do well in this specific type of tests (pressure optimized vs airflow) and the noise benefits are much lower in part by the fan having a frame or being enclosed part way vs being completely open on a drone support.
You know I got to thinking about it. If you look closely at the stock a 12 X25. The blades kind of resemble just a toroidal shape that's been cut in half. It's like the front half of a toroidal shape where the back half has just been cut off.
The Mobius probably did the same because it was pulling air the opposite way through the tube when it was facing the other way, leading to similar performance
He is measuring pressure so it would be zero in that case.
my favorite part is when the mobius fan said "its mobian time" and mobiused all over the place
They certainly look nice, but it would seem that their real advantage is the sound quality.
Hey man, thank you so much for this series. A couple of years ago I started watching and tried my hand in solidworks. I kinda got addicted to designing and 3d printing stuff and I’m now working as a manufacturing designer full time.
The whole idea behind the toroid is about the interactions at the wing "tips"... which are being blocked by the housing... So i imagine you lose a significant portion of the benefit
This why the airplane came from two bicycle tinkerers and not some governmental think tank. Gorgeous work everyone!
Have you ever compared the performance of a real NF-A12x25 to a 3D-printed one?
I remember a past vid there was one made to the same specs and yeah the 3d print did not give the same output but was very close.
I think whats missing here is to now take the A12 and curl the tips into a toroidal hybrid and use some putty/filler and sand it to a smooth finish surface.
Great video, crazy that the mobius pushes as well as it pulls.
It got perfect symmetry. The result is logical.
I think it will work well in an old OEM Intel style heatsink and fan without a shroud
It measured the same inverted because you're measuring the amount of airflow not the direction.
If there's an intake and an exhaust it will move air either way and the reader will measure it either way as far as I'm aware
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦........
I would love to see a design finnaly offer at least a good noise to airflow ratio and inspire real futur products. Giving you and anyone else involved credit would be like free marketing and it could be a huge seller early on that way.
It was either season 1 or 2, but that did happen. I forget the name of the fan, but James was contacted in regards to getting in touch with the original designer of the fan. I'm not sure how it went, though. You might be able to find something about it in the Fan Showdown playlist, or on Reddit.
Love the work you're doing here, and access to the fan designs is awesome. I'm wondering about scaling these up to room-size fans, for other applications, like whole-house and AC compressor, etc.
I'd be interested in seeing the static pressure generated by each fan. Have you thought about a testing methodology to include this in the future?
The fan is mounted on a radiator for the tube/airflow test
I watched a video of a guy who 3d printed a toroidal blade setup for his drone and tested it against another set of blades which looked more like a PC fan. The PC fan design outperformed these new fans in every way,
Speed, sound, manoeuvrability etc.
Had to watch the one before this, these are great ideas! I would love to see you print one out for a drone.
So much hype! The toroidal propeller can certainly be improved.
The thoroid fan sounded calming!
Great initial testing!! Same blade area, same diameter, pitch optimized for the same rpm, no duct, same blade finnish - and go again!
Solving tip vortices might be great for open propellers, but the design trades are a pretty different in the world of ducted fans. The toroidal design reduces interaction outside the diameter of the rotor whereas a ducted fan leverages the interaction with the duct wall.
Exactly. This stinks of skimming a paper on toroidal blades without actually under standing how fans work.
This is the Earliest Ive ever been for a Video!!!
Heya! Long term Fan Show Down Fan here! Suggestion: A printability score could be a cool addition! (Something simple like: Easy / Medium /Hard?) And when presenting the fans, an addition to the people's names, so we can see if they already had another fan on the show. I know you usually say it, but I still sometimes wonder if the person has already sent something in.
Nice video, thanks.
One point to note, The advantage of toroidal propellers is largely in eliminating tip vortices in non-ducted propellers. The close fitting circular housing on computer fans is a big help with noise and tip vortices, the toroidal end connecting one blade and the next kind of gives this same effect, so it's not that surprising that a ducted toroidal is not much better than a ducted fan.
i think toroidal fan blades are specifically designed for open-blad propellers to reduce wing tip vortices and everything these vortices causes, like extra noise, efficiency loss etc. they are not designed for propellers with housing, like these desktop fans.
9:59 It's like the Mobius has only one side or something, fascinating!
Who would have thought 3D printed PC cooling videos would have so many fans.
He can turn ANYTHING into a fan, Some one needs to turn the Text ANYTHING into a fan.
I had a gut feeling you were from where I thought you were from, I was watching your Fan Showdown videos, you posted a picture of you in the Air NG and I've flown out of
Rickenbacker way more than a few times, I spent 4 years Active Duty Army and 8 years Army NG, great videos Brother, I especially liked all the Fan Showdowns, can't wait for more. Oh and the Hoppes gun oil also gave you away lol.
Yay. My fan kicked ass. Always steal your designs kids. Wooop. Thanks @majorH
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦............
The thing is that the value of toroidal propellers comes from the elimination of blade tip vortexes which are resolved with the enclosure present on the pc fan. Really interesting testing though
Two things:
-Clean the edges of the printed blades(lots and lots of drag, causing noise that would'nt be present when ejection molding parts)
-You should also measure the power consumption, to calculate efficiency.
Bro I do respect you so much.. all ur fan video are so genius! Keep making content like this please!
It seems that 3rd, 6th & 9th placed fans on the chart are the top three. Fascinating video.
Would’ve been cool if you had a pic of the fan along side the examples so we could see which fan black was performing in which manner. Good vid
The Mobius performs same flipped around because regardless of the direction, the fan is either pushing into or sucking out with the same pressure differential which should generate the same mass flowrate. The design "flaw" with the mobius is just that the mounting hub was printed the wrong way around.
Showing all 4 smoke tests at once was really cool. I must not pay close attention when watching because this is the first time I noticed that.
the music for the smoke demo was superb👾
This channel has come so far and I'm really proud to be an OG subby
If you can imagine holding the center wile applying torque to the blade structure, flattening the angle of attack a-bit giving it a more swept wing vibe to it..
The smoke tests are always my favorite part...ASMR for my eyes! XD
Thanks for the feedback.
Expect more videos soon.
Send a direct message...
I have a Gift for you📦.........
The magic happens in the outer loop as the air transitions and is push under pressure.
The leading edges are lowering the pressure to suck the air into the space.
So in my view it's the air transition from lead edge to the outer loop.
Croud sourcing to further the MIT experiment. Love it
What I actually really like about these is the high freq reduction, this would be a boon for server fans when shrunk down to 40mm because of how dang fast those bad boys spin. I think given a re-design that tightens up the dead space around the blades and cage and optimizing the flow via CFD we could see these perform just as well.
Hell yeah just discovered the series and already get to see a fresh video.
The fans from BeQuiet will beat all fans, including noctua fans.
Their Silent Wings and Shadow Wings 2 are non audible at low RPMs.
I use those for ventilation exhaust in residential rooms and boat cabins, where sleep level silence is important !
When I saw Thoroid, thargoids from Elite Dangerous came to my mind immediately. The looks are somewhat similar, and so the name. :)
When they do injection mold plastics for the toroidal fans, you will get thinner blades and more efficiency i think. 3d print is pretty thick and heavy by comparison and you have a lot of clean up to do with sanding, not to mention balance
MIT grads: publishes research.
The internet: lets make a PC fan outta that
Hey MH, one thing you might consider doing for your 3D printing. Have an "acetone baker" to smooth out the rough ridges. I've seen some videos that show people using acetone vapor in a confined space to basically "bake" the rough ridges out of a 3D printed object. "Some experimentation required." :D Hope this helps.
nothing compares to watching Fan Showdown at 3 AM while you are stoned
I love that chart you have there with all the designs tested and compared... and NOTHING's better than reference. They're all louder, move less air, have less pressure etc...
I love that people are goofing around, but there's very little you can do against decades of scientific design and tweaking from multi million dollar company.
It's very likely that fans are as good as they're going to get and there's no way to improve anymore. :D
All the blades on the A12x25 are basically half toroidal. It wouldn't be surprising if Noctua tested these designs but made some concessions for static pressure and total blade surface.
Burst out laughing at the Mobius first test... well done!
I get that this is just for fun, but if you think about it for a second, the whole point is for these to reduce tip vortices, which are basically not a problem at lower reynolds numbers (aka fan speed). An actual comparison is to strap that noctua to a drone motor in an open airflow config and let it rip. That's when you'll see the difference. Otherwise, this is a square peg, round hole situation.
The reason behind a Toroidal Prop is to mitigate cavitations which are little pockets of air that are the reason behind sound on props. Although it is interesting for the performance in a pc they are useless due to the surrounding material which servers the function of a toroidal prop.
You should try to clone the noctua reference fan, just to see how much of a difference the printed version makes. It might also show how much noctua is able to improve the other versions with their manufacturing process.
Get the MIT team that did the original drone design to make a model for you to print. I think that would be most interesting to compare to today's submissions.
Don't forget that toroidal propellers are "quiet" for 20-20khz frequencies. So you should maybe think about the noice on a certain flow rate. Or measure the needed power to generate that flow rate ( to get the efficiency ). The strength (in my understanding) is not the performance but the efficiency
dOH GOD DUDE, i recently watched a vid avoid this efficient and effective fan design in ships and wind turbines, then figured "damn, i wonder what if it was a GPU or CPU cooler" then rmeembered your channel, so here it is 👏
now is necesaary an edition of fan showdown where the fans are mixed with the different accessory inputs and outputs of other fans to find the best combination of low noise, focused airflow and volume of air mobilized
One very interesting question for your designs and tests is the concept of windage. Fans work best when there is very little gap between the housing and the edges of the fan blade.