Toroidal Propeller Tutorial in Fusion 360 for FPV Drones
HTML-код
- Опубликовано: 14 июл 2024
- ⚠️ I'll say this again, I'm not an engineer and there are probably better ways to do this. The method demonstrated is certainly not as scientific as I would like, but I think it's a fairly accessible way to design a prop and also be able to make small adjustments to refine it. If you want to know more about why the design is the way it is, I also urge you to check out my previous videos on the Toroidal Topic:
• 10 Toroidal Drone Prop...
• MORE Toroidal Propelle...
• DIY MIT Toroidal Prope...
⚠️ If you're completely new to Fusion 360, you might want to start with one of my older Fusion 360 tutorials (with much worse audio) before trying to take-on the task of complex lofting to create a toroidal propeller, so have a look here at how to make a duct instead:
• How to Design an FPV C...
🔗 Links to my V3 Tri-Loop and V2 Bi-Loop designs on Printables and Thingiverse, including STEP and F3D versions:
www.printables.com/model/394030
www.thingiverse.com/thing:584...
www.printables.com/model/393972
www.thingiverse.com/thing:584...
🔗 Playlist of people who've discussed, bench-tested and flight-tested my designs:
• Community Toroidal Pro...
🔗 Link to PDF document from the MIT people that came up with the idea:
www.ll.mit.edu/sites/default/...
===
👍 If you like what I do, please consider supporting me:
☕️ buymeacoffee.com/strikingfpv
🔔 Subscribe for more content like this:
🔗 ruclips.net/user/strikingfpv?s...
===
💥 STRIKING FPV is a cinematic FPV drone operator based in Thailand, wanting to show the beauty of this country from a new perspective - evidently also likes tinkering and is happy to show you how stuff can be made
/ strikingfpv
/ strikingfpv
===
🕓 Chapters:
00:00 Intro
00:30 @MoppelMat's Successful V3 Tri-Loop Test
01:44 Prerequisites/Disclaimers
02:29 Explaining Varying Blade Angle of Attack
04:30 Create a Hub
05:16 Create Start Profile
11:03 Create End Profile
13:34 Create Edge Profile
15:21 Create Mid Profiles
20:34 Create Base Rails
27:55 Lofting Issue Demonstration
29:25 Create Upper Rails
36:27 Lofting the Loop/Blade
37:43 Post-Operation Sketch Adjustments
40:37 Save Your Work!
40:58 Additional Tweaks
42:52 Fixing Broken Operations after Sketch Changes
44:19 Duplicating and Rotating Loops (Blades)
44:58 Cutting and Combining Loops Together
48:17 Combining the Hub and Fillet Troubleshooting
50:31 Finishing Touches and Flipping the Model
51:16 Modifying Everything to be a Bi-Blade Instead
54:53 More Disclaimers and Suggestions
55:39 That's a Wrap! 
Хобби
A couple comments: instead of creating an offcuts folder, right click on the bodies you don’t want anymore and select ‘remove’ instead of delete. That makes fusion360 hide the bodies for you. Second, you can constrain those points on the rail to the tops of the sketches using the ‘coincident’ constraint. Might make life easier.
And this is why I make tutorials, so that kind commenters can give me tips 😁 - gotta try those, cheers!
Thank you so much you are really good i was not expecting to find a tutorial this specific. I know this is like super niche content but you deserve more exposure
Very nice video! I will try to orint a propeller like this.
Pro tip: a sketch isn't properly dimensioned until all librs are black. Also, midplane is a useful function, saves you a bit of math.
Thanks a lot, your tutorial is outstanding.
Thank you for this, learned a lot!
YOU ARE A BOSS - Thanks for the tinywhoop props and helping advance these in the FPV community - Full review with credits coming soon! -Kenny
My pleasure, crossed fingers it works for you 🤞
We now have a technology to manufacture these propellers in metal and polymers. And we don't use 3D metal/polymer printing. Talk to us, we have a pleasant surprise for all those who wish to get these props made on industrial scale and cost effectively. Pahwa MetalTech Pvt Ltd, India.
can you design them as well?
You are just a boss thanks for this video
Hey thanks
Danke das ist die erste perfekt erklärte, aber auch nicht langweilige Demonstration von fusion 360 mit sehr vielen Funktionen die für diese top construktion
Danke
Bitte, glad it helped you out! I'm not a pro and definitely made many design mistakes, but if showing a bunch of the functions helped you then still happy I made the video, haha
Like other's have mentioned, constrain the rails to the profiles. Also do less deleting( like the chord) to keep it constrained so you're less likely to have fusion "freak out". Instead, switch things to construction lines instead of deleting.
My first SLA props using SyriaTech Blu/Black exploded. I'll probably try some different settings, plus I'm working on flattening my build plate( stupid Anycubic,...). Also have an even "tougher" resin on the way. But they printed ohhhh,. sooooo ,... pretty, and smooth. That was printed on supports because of my build plate issues. It really didn't come out bad vs building on the plate. Made me think that with that method, we can set the trailing edge super slim..just enough to attach supports. Might help the noise/efficiency. But for now, I'm going to work on printing from the build plate and getting the props to not explode.
Nice to see others having fun with F 360 and their hobbies…..
Fun and headaches, can be a pain getting a design to work sometimes. There are reasons why this tutorial only happened recently and not a week ago 😅
Wow I'm really good at blender this just took my ability to design to another level, I have so many great ideas I want to bring to real world applications. Thank you
My pleasure!
When I was a noob to the 3D printing world I tried to use Blender to model stuff and it was a disaster. When I picked up Fusion 360 all of a sudden things started working. It really is the right tool for the job.
@@mjodr In my younger days I had access to 3ds Max and did modelling there for things like helmet GoPro mounts and random things. I still remember dragging vertices around 😅.
Of course back then I don't think Fusion was around, and other CAD products were obscenely expensive, so I'm very thankful for how open (relatively) the software space is nowadays.
@@STRIKINGFPV Yeah I started in 3D Studio Max in 1999. Things have changed a bit, lol.
Are you also a fusion guru?
Need to clean up and then get a blender model into fusion or onshape... wonder if you can help?
Thank you!
Good job. I am a Solidworks guy, but the method is the same. I appreciate your insight!!
No worries, Solidworks is a pretty solid option. I'll see myself out 😅...
Muchas gracias. Ahora debo hacerlo en solid edge de siemens, es el programa que elegí usar (es gratis)
If you use the remove command, instead of delete, you can get rid of futures you dont need. Nice video btw, trying to learn something from you :)
Yes, one of our other commenters pointed this out as well. For a while I was wondering what to do, but I figured if I ever needed those bits later-on (I didn't, lol) then having them in the folder wasn't the worst idea. I can't believe I didn't notice that there is "delete" and "remove" though. So similar, lol.
was good for me
Could you share the 3dprinter setting, like infil, layer height, etc?
I'm going to try these on a rocket design, subsonic of course.
I've been looking at the Sharrow marine prop. Obviously the Reynolds number is different so they're not directly comparable, however, they do something different which the MIT design didn't do. The Sharrow prop gets thrust from the tips because the blade is angled to scoop the water in whereas your design is flat to the rotation tangent. Once it's scooped in it can only go backwards and generate lift. Whereas your design has surface friction, weight and no lift from that part of the blade. That can't be good. Doing it like the Sharrow could be a lot harder to print and design though.
My OCD is yelling "Fully constrain your sketches!!"
I would if I were a master at the constraints but a lot of them still behave oddly (to me) since I'm not used to them.
There is a reason for the "not an engineer" disclaimers 😅
@@STRIKINGFPV Fair enough. I enjoyed the tutorial even so.
@@timlong7289 Haha, cheers
That one looks good
I'm not quite happy with the rounded shape from the top-down view, but the 1.5mm high edge profile has resulted in a tighter transition which could actually be better for the AOA further out. Tempted to lower to 1mm just to see, but then we really compromise the strength I think.
@STRIKINGFPV you don't think the strength can hold up to full throttle? Worth testing.
Be nice someone has a thrust stand to test power & then test amp drawn at given thrust vs simular power prop. Then test real world.
If can be close to regular props and a lil quieter it's a win
Can we use that kind of propeller in bigger(like x-talon uav) planes?
That CAD system seems like a pain. You should try Solidworks instead.
Regarding the prop, don't you want the outer edge of the "loop" to be "open" in the direction of rotation? As you did it it was perpendicular to that.
Thanks for the video, I will give it a try too, in SW though 🙂
Hey Ashton, I wish I'd had notifications on so I could have watched this before tackling my own fusion design it may have made my process faster.
If you want to make your designs more scientifically start with your chord profiles like I did to ensure than the pitch angle remains constant. I haven't watched this video yet other than to scan in FF to see if you changed the design from your original design, because I have a couple of comments on your original design after having modeled a few of my own designs.
When I looked closely at your original design I noticed that the pitch angle was not constant, ie it lacks twist to lower the angle of attack as you move towards the blade tips which will compromise your efficiency.
When I started trying to figure out how to model the toroid propeller I did they same thing, I used the same angle on my airfoil profiles. I just wanted to get something that lofted the manner I desired at start. Once I had that figured out I calculated the proper airfoil angles and modified my sketches them to optimize the lofting of the blades to assure that I have a constant pitch angle along the blade length. Then I did a bunch of section analyses to ensure that the loft and rails kept the loft following a the path to maintain the pitch angle.
The propeller pitch angle decreases from the blade root to the tip in order to maintain constant pitch. The relationship between Pitch & Pitch Angle
is dictated by the formula: Tan a = Pitch / 2P r
where: a = pitch angle and r = radius and P=Pi (3.14159)
Since the blade is curved the radius of the blade should be measured from the the middle of the chord to the center of the hub.
I also believe that a triloop design will be more efficient if the loop sets meet at the root rather than along one of the blades. I have one on my thingiverse illustrating what I believe will be more efficient
I will watch this video in great detail shortly because I really wanted to see how you sketched the rails for the leading and trailing edges. My initial design I failed to pay close attention and made a single curved plane to project a 3d line onto and as a result when my edges were reversed on the return portion of a loop the leading edge on the return loop at the loop was lower than the other side, resulting in a blade that would have unbalanced thrust.
Want to test my 5040 biloop toroidal prop? It is on thingiverse under TorontoBuilder
Your designs look pretty good, only issue I see is on the leading blade near the loop point the base of the blade which would be on the print bed actually juts out forward, so it looks a bit like an issue with the bottom rails, if you used bottom rails.
Funny you mention testing though, I literally just finished testing my own tri-loop design along with a standard hex-blade that I designed so there can be a more apples-to-apples comparison, in terms of having the imperfections of a 3D print.
Might be my last video on the subject for a while, but that depends on the more casual testing I'll do in the future, now that I've got my print settings somewhat dialed-in. Hope my tutorial helps you out - it's not perfect, but there are some ideas in it that I'm sure others like yourself might find useful and run with.
@@STRIKINGFPV Did you do flight test only or have you made a thrust test rig? I'd love to see any of your results
I did use rails for both the leading and trailing edges on every blade. Yes the base of my trailing edge does jut forward near the blade tip as the loft makes a sharp compound curve the face gets pulled backwards slightly away from the edge before sweeping forward again. However the swept back portion is only about 1/4" long and will be removed in my post printing operation.
I intend to sand the front of the base of my trailing edges to fair them a little to be closer to the ideal airfoil. I thought about trying to tweak the design, but I really wanted my base to maintain 2 walls widths all around to provide good bed adhesion to minimize potential printing issues.
I've been thinking more about how to improve the strength and durability of 3D printed propellers. I did a bit of experimentation using the annealing of PLA embedded in powdered salt in the past when a batch of pot hooks I made for my wife started breaking over the course of a few weeks. I plan to try that method to treat my propellers, then wash off any salt residue and then coat the parts with epoxy resin mixed with a mix of milled glass fibers of between 0.009 to 0.03" long followed by sanding after the epoxy cures.
@@STRIKINGFPV OH and one final comment, well maybe final that is... it may be of interest to your viewers to know that MIT tested the thrust and the mechanical and electrical efficiency of their propeller and motor combinations with the series 1580 Drone Test Stand from Tyto Robotics located here in Canada in the province of Quebec. IF I had a youtube channel or patreon supporters I'd appeal for donations to buy one of these things since they're so cool.
Being poor and without a youtube channel of patreon supporters I'll just build my own
I wouldn't worry about a very thin trailing edge. Set a brim and it will be stabilised.
I've since tried to do the loft command with a fine edge and it always fails - maybe creating a very tiny radius curve at the end could make it work, but considering the 3D-printing nature of it all I figured having the flat edge on the print bed would be fine.
Ideally, though, using a NACA profile with a fine trailing edge would theoretically be the best way, if I could get the loft command to not fail, haha.
What about Winglets on then tip of a normal prop? Like the airliners have on the wings? I think this would be an awesome design?
The DJI drones did this with the first Mavic's Platinum refresh with the swept-back top to reduce noise (along with the sine wave commutation).
A few FPV drone props do the sweep-back, but I haven't tried them nor do I know if they are any quieter or less annoying.
If you're actually thinking of an upward wing tip (or downward like on some helicopters), then it becomes more complicated to print and also becomes weaker. At least with flatter designs we benefit from the strength of single extrusions from the hub out towards the tip, but an upward (or downward) wingtip would be relying on the layer-layer adhesion which isn't as strong, more likely to snap off under load.
Resin could be fine in theory.
G'day
Could you copy and paste the bottom rail for the top rail and move it? Instead of drawing it again.
Referring to the top rail. When you were drawings in 3D and moved the points by dragging it down, there's a faster and more accurate way.
On the move tools click point to point instead.
Great way to do the design 👌
Greg
In theory yes, that would be a good way to replicate the same shape top and bottom, then it would only need a little bit of minor adjustments whilst maintaining the same control handle positions.
As for moving the top rail in 3D, yes I now realise there's a constraint that would help me make sure the rails stick to one another 😅. I'll have to experiment a bit because I'm also concerned about how the control handles will behave with their constraints as well, but overall it does indeed make a lot more sense.
I can confirm that the bi-loop does not work as a boomerang even when split and re-formed at a 63 degree angle.
Can you try to design a rc boat toroidal propellers?
Ho ho, actually that's quite a tough one. My local makerspace has a trash collecting drone boat project where one of our guys designed and 3D printed custom props which worked remarkably well, so I've been wondering about doing a Sharrow style prop for the boat as well, but the geometry is quite a bit different, looks a lot more challenging.
I actually need a taller hub to accommodate the Sharrow design, but the shaft cannot be lengthened so I might have to leave a space for the lock nut and then have the rest of the prop shaft floating above the lock nut. Then the actual geometry looks like a major headache, I'm still not 100% about how to rail the shape.
With these toroidals for drones it's much flatter and luckily I can create the blade profiles at 90 degrees to one another, but the boat prop looks like I will have to actually wrap profiles around the cylinder which makes the hub, and this becomes much more complex for me, so will take time to work out how to do it. Not even sure I'll be able to.
someone should dig into the NACA prop design reports (WW2 era and before) to help define the blade profile
you should be able to add constrainst to the curves that attach them to your blade profiles so they move as you tweak the profiles (I use onshape, not fusion 360, which isn't available for Linux)
Actually I had a gentleman on the toroidal propeller designers group on Facebook point me to airfoiltools.com which allows you to reference and modify known airfoil designs which is nifty.
One can export the shape and pull it into the program of choice to mess with, so that looks kinda promising.
But hey, if you're on Linux, there's an open source fluid simulation thing which runs on Linux which could be fun for analysing the designs that you make. OpenFOAM I think.
Hi, I am interested in your work and I need the quitest propeller. Among all your designs, which one is the quitest? Btw, I don't care much about efficiency. Thanks a lot.
I would say V3 because it's the most aerodynamic compared to the first two, but like I've explained in my other videos, toroidal propellers are not so much quieter, more like their noise profile is less annoying so they seem quieter. Also, my design is not ideal for a final production prop, it's optimised for FDM and somewhat SLA printing, so it's far from perfect.
IM ABOUT TO TEST A SET IN ABS FILAMENT 2.5 PROP
FOR A WHOOP
GOOD LUCK TO YOU SIR, HOPE IT PRINTS OKAY - I'M FINALLY PRINTING MYSELF IN PETG, TOOK A FEW ATTEMPTS TO GET THE SETTINGS GOOD, SO CROSSED FINGERS 🤞
@@STRIKINGFPV FIRST SET HAD LAYER ISSUES BUT I GOT A SET DONE AND GOING FOR A 5IN SETT WITH SILK PLA OR THIS GLOW PLA I GOT SORRY FOR THE CAPS IT BE CAME A BAD HABIT EVRYONE
@@thebronxgeek2121 NO WORRIES IN THAILAND THE WRITTEN LANGUAGE DOESN'T HAVE CAPTILISATION NOR GRAMMAR NOT EVEN FULL STOPS JUST SPACES USED TO RANDOMLY MAKE THINGS MORE UNDERSTANDABLE BUT STILL KINDA RANDOM ANYWAY HOPE THE PRINTS GO WELL
What‘s about the Surface of printed props. Smooth enough?
To fly, yes. To be efficient, not without some serious post-processing, like MoppelMat did with his prints and test flight
9:08, 24:00
Hello Ashton,
I am Labhesh Chaudhari currently pursuing B-Tech from Mechanical Engineering in NIT Trichy, I would like to use your design in my project where I am testing out agility of a dual axis tilting quadcopter with toroidal propellers. But for that i need detailed specification of your propeller design. Its Pitch,Diameteretc.
I would kindly request you to provide me with such information.
Hi Labhesh, apologies for the late reply - if you download my files from Printables or Thingiverse, I've named the files with the standard propeller naming convention for propellers. 3x2.5x5 for example means the prop is 3 inches in diameter, has a 2.5 inch pitch and has 5 blades. I have named mine similarly, however the design is not perfect, and the effective pitch might not be exact due to how the shape needs to become steeper as the loop comes back in on itself.
Basically I've tried to be as clear and definitive with my design, but at the same time it should be taken with a huge grain of salt due to the experimental nature and unorthodox design.
I can't help thinking these will be like the fidget spinner craze and will be forgotten in a couple of months, are any of the major prop manufacturers making these? Why no commercial aircraft using them ?
I got half way thru and the realized the snap points f'ed me making the rails. I'll have to try again.
Success! Thanks for the tutorial.
Yeah you know I used to manually drag the rails to as close to the profiles without snapping and it did work for the first couple versions I did. Then I tried working out how to get them to snap and that's what the projected geometry thing did for me.
A lot of head banging 🤣. Glad it worked for you in the end!
ruclips.net/video/z58RORCUTao/видео.html has an interesting isight with two baldes prop at 30 degress from eachother and offset on the same axis by (my guess) one blade chord, all counterballanced on the other sde on a short arm with a weigth, and that makes it even more quiet plus manatians the airpush of a two blades, highly effcient.
What if you take the two (double) baldes thoroidal system, eliminate one blade (pair) and replace it with a weight, ballance it with some metal inserts and eventuall screws and measure it against the others?
You are creating your 'pitch' by defining it by the forward advance ratio. Pitch is a somewhat nebulous term that has defined aspects of blade geometry differently at different times since the 1940's. Traditionally, pitch was by the angle of attack at 75% of the blade from the hub and assumed a constant twist along the entirety of the blade.
That's interesting - when I was first thinking about prop pitch I was wondering if it was defined as like an average AOA across the blade, until I did some Googling. Didn't come across the term "forward advance ratio" though, I really like that term, very eloquent.
Reverse engineering a human
Well, I don't think Fusion does soft body stuff, haha