The most powerful, most efficient 5" drone motor ever made // AOSRC SUPERNOVA 2207
HTML-код
- Опубликовано: 5 авг 2024
- Uncertain about what to buy? Check out THE ULTIMATE FPV SHOPPING LIST: www.fpvknowitall.com/ultimate...
SUPPORT ME ON PATREON at: bit.ly/2Ep5Sia
THIS IS MY FULL TIME JOB. Here are other ways that you can support me: www.fpvknowitall.com/support-me/
---
Buy the AOS-RC Supernova 2207 at:
* Aliexpress (Global) - bit.ly/3PzBNPP
* DefianceRC ( USA) - bit.ly/3RFqvfQ
* HobbyRC (UK) - bit.ly/3RFnR9z
* DroneIT (EU) - bit.ly/3rnWSVD
---
00:00:00 - Chris Rosser: Aerothermal Engineer
00:02:09 - Most drone motors are all the same
00:03:50 - Surprisingly high variance in motor performance
00:05:41 - Can we magnetically optimize the motor design
00:08:20 - The value of a "digital twin" motor
00:10:49 - I ran this tool for over 20,000 iterations
00:12:10 - Innovation 1 - Thicker flux ring and thinner magnets
00:14:32 - Innovation 2 - Hollow stator core
00:17:09 - The most powerful AND most efficient motor ever tested
00:22:28 - What is a "smooth" motor really
00:24:28 - The hard questions: price, durability, independent verification
00:28:48 - Will other manufacturers clone these motors
00:31:14 - Chris's new patented feature, revealed
My interview with Chris and Ryan about motor fundamnetls: • FPV Motors round table...
Watch Chris's release of his new motor! ruclips.net/video/WO1D0krt24E/видео.html
What's the opensource software Kenneth! 🤖📻💾
Two legends, one video
Four links, none in stock 😩
NSFW
For some reason i read "two legs one video".😂 sorry
I'm glad Chris Rosser is using his engineering super powers for good and not evil. Awesome, amazing work!!
Well, he could work a bit on anti gravity as well... 😬
Patents are evil, he went down in my regard by going that route. It's ridiculously expensive and doesn't even protect you unless you have millions to defend your patent with. Is Chris this rich already?
You should read The Case Against IP written by IP lawyer Stephan Kinsella. You can read it online for free.
lol what is one example of sm that would be evil
At the risk of sounding like a terribly scratched record… we are so fortunate to have Chris. Science is my life, so his systematic approach to optimization is just such a welcome, needed, fresh breath of air in this hobby. I love Kabab and I’m genuinely grateful for his contributions as well.
To each their own, but this is the level of engineering expertise that I have always wanted to see. Thanks for helping showcase and share the knowledge, JB!
Like a Rhinestone cowboy......
Those of us who know what I'm referring to like it 😁
@@jameshatton4211 crap I don't have the reference
as an engineer with a somewhat similar background to Chris, I couldn't have said it better myself!
Dude, I love this guy so much. It is incredible to get engineering insight into how our gear works in a way that manufacturers don't provide. Thanks for going in depth!
Ya, braking ability was something not very focused on for optimization until now, yet it means everything in propwash handling from a latency reduction.
Ok i m gonna play devils advocate. How major companies developing motors for years, with engineering teams and equipment worth hundreds of thousands of dollars, never heard of Nelder Mead optimization (which is a very simple Matlab optimization algorythm I might add).
Chris is a legend, but I am sceptical on this one.
I cross my fingers and hope it's the legendary, groundbreaking, physics bending motor he claims it is though :)
There’s some good scientific skepticism.
I suspect Rosser’s response would be something to the effect of “let’s let the data speak for itself.” He set out to maximize performance and consistency and he claims to have done it. I haven’t known him to be a bullshitter but I guess we’ll see…
I think you overestimate the amount of money those companies are willing to spend in r&d of a toy product. Which is zero. 😅
@@andreabucciotti I like your name, it s kinda like my gamer tag "Bruschotti"
you mean like dji@@andreabucciotti
I don't know that those guys actually pay what it would take to attract someone of the calibre required to achieve world's best in class results for an application like this. Once they have viable competitive products the pure r+d budget might not be as large as you think.
Buying best or cheapest stuff other people offer can look pretty cheap as a known quantity, compared to r+d costs with unknown time frames and outcomes. Plus to be fully vertically integrated as a manufacturer is very, very far from the norm these days, at least for consumer grade.
So all around, while I am very impressed, I am not that surprised.
Does anyone know what DJI spends on research in a year (as opposed to product develoipment)?
And I wonder if maybe the people who might really care (the military) have not already arrived at a similar place, but it's classified.
The only other people that might really make serious effort are the automotive industry, but the size, desired characteristics and acceptable manufacturing parameters don't neccesarily align as closely to small drone motors as might be thought on a casual glance.
See my other comment here re optimisation of inductors and transformers for aerospace applications that I did for my university in 1993.
Stack mounting; resonance cuts that we use in saw blades.
It’s a good idea, with a lot of science behind it.
I like it 👍🏻👍🏻
TIL what those cuts are in saw blades....
@@RobertLeclercq they are for tuning out resonance frequency vibrations, exactly the same purpose as Chris’s application.
I have actually tuned such systems in the past, and can tell you first hand how effective they are. 👍🏻👍🏻
Chris’s implementation is great, I’m actually super excited! Very cool!!!
@@lemonsquareFPV Yea I thought the motors we're cool, but them damn slits are cooler IMO.
@@lemonsquareFPV I am absolutely loving Chis's damping slits, can't wait to see how they stack up. (🙄yes, pun intended) But I always thought the slits on saw blades were to keep the saw blade from getting too hot. I think it was one of those tool review channels that put that heat dissipation thought in my head about those slits on saw blades.
@@Inertia888 love the pun 👍🏻
Look for winding/curving laser cut marks (basically what Chris is doing) or holes. Maybe Look at a few Freud circular saw blades and you’ll see what I mean.
Chris is just such an amazing contribution to this hobby in total. Love flying those AOS frames... let's see what the motor will feel like!
WOW Chris and JB! You guys are so great for our hobby and I love what both of you guys do for all of us! I'm grabbing a couple sets of those motors boys! Cheers!
I'd love to see him also apply this process to a motor for sub250 builds
Sick. Bardwell, Rosser & Ian are downright the best trio for improving almost everything about our hobby. Keep it going.
For 3d optimized motors they should start using hall sensors and depend less on the backfeed, maybe that could be useful for further ESC optimization.
Sensored motors are only necessary if you need low-rpm torque, or if you're trying to do FOC, which we aren't. Back EMF is fine for the time being.
@@JoshuaBardwell Thanks, that makes a lot of sense. Getting more systems on am32 and optimizing that open source ESC seems to be a better choice given blheli32 hasn't had the rosser treatment. Cheers.
So it really only makes sense for the massive cinema rigs, 3D however could provide a whole new cinematography perspective. Sensored motors should improve things when it comes to 10" props with all the additional rotational mass & flux.
That's a fascinating topic for me. I mean to simulate stuff and then try out in practice. Because then, if you find discrepancies... You make discoveries! Which can make our models and theories even better. 😃
A perfect example of it is cosmology. There's always a huge simulation running at some super computer, while we have things like the JWST making observations that we can compare to. 😃
Thanks a bunch for the interview, Joshua! And thanks for your work, Chris!!!
Stay safe there with your families! 🖖😊
The way Chris explains the technology, it makes perfect sense, AND he has the numbers to back it. I already see a few posts where people already have them…
Two kg per motor is nuts, so exciting to have Chris applying his engineering skills!
Two brilliant minds, leading the quadcopter world to a better place. Man, this is exciting!
Incredible interview! What wonderful knowledge and information. Appreciate it very very much. Thank you!
These are totally going to be in my next build!
Excellent Interview I hope it works out well for Chris he seems like a really nice man, thankyou Joshua👍👍👍
Chis Rosser's innovative work is absolutely incredible
This new motor sounds awesome! I usually stay away from expensive motors, but I'll have to get these at some point! I hope they spend enough time in stock to grab them in the future.
It would be awesome if Chris made a motor like this for a 3.5"!
I bet they'll be a lot of prop deflection, I wonder if we'll need a new line of props in the future to handle both these motors and the general trend of heavier more powerful quads. Carbon fiber props would be interesting, but definitely too dangerous for freestyle.
Great interview JB, this is my favorite type of fpv video! I really like the deep discussions about fpv products and fpv theory, as well as the science behind it all! I've watched all of your interviews with Ryan Harrell at least twice.
What a way to look at it Chris yes your motors may make other motors better thank you in advance
Chris's motor is going to do very well! He gives alot of time testing for us. People will support him!
1st batch was sold out when you typed the comment I am replying to. The web page said the 2nd batch is expected to ship early Oct. I would be surprised if the 2nd batch is not sold out before they start the 3rd.
@@Inertia888 as much testing and knowledge he has people know it's going to be quality and standards. Yeah this with be a hot sellers
That frame idea is amazing
This was a learning class. thank u both of u.....................
Loved the video! Very informative!!
definitely gonna be checking them out
Well done interview JB 👍👍
Very interesting video, I have learned something today 🙃
Excellent approach to engineering a better motor!
This is very informative
for chris, vibration isolation:
think about mounting gyro/stack not to a bottom or a top plate. mount it in the middle by using for example the standoffs -sidewise. so its "floating" but stiff connected to the sides. the metal standoffs wont bend.
OR:
use the stackscrews as standoffs aswell. make them screwd to the bottom AND top plate additionally. it wont be likely that this area will bend if its connected in a "cube" bottom and top. its isolated for itself
what do you think?
Guys this show was fantastic!
I would say the influence of the ESC is just as important as the motors. The choice of components and the matching of mosfets plays a major roll in what happens to the motors. Fake components are also a major problem. Drones being as good as they are in general is something to be grateful for. Nice to see Chris being so enthusiastic on motors I'm sure he'll improve things.
Waiting for the day when AOS branded ESC and FC are released.
Also bit of nitpick but according to the data that Christ showed the AOS motor is only the second most efficient motor that he has tested. The most efficient is the MEPS Space (row 28).
meps hiệu suất cao nhưng lực đẩy yếu òm, tính ra thì nó lại ko ngon, chỉ phù hợp longrange ông chriss cũng đã nhận xét
@@minhhungnguyen7867
Great interview ~ 10~30% variability is huge.
I learnt several things today!
Legit the first time im excited to try AOS and rooting for Rosser... in the past ive had opposing experiences to his test result's including his first frame design and motor analysis. Causing me to disregard him completely. I hope I get a chance to see and experience these motors and frame designs for myself
the Flight controler cutouts is pretty sweet idea.
I could see where the gap would allow it to float. and could control the contact points.
depends on the material. more flexible softer crap will eat more vibration. if its rigid as fuck vibration will transfer like its not even cut.
@@derrickcripe507 It seems so simple and obvious, once Chris explains it, and shows how it works. I wonder if the direction of the carbon fiber weave would affect the travel of standing wave vibrations that make their way to the flight stack?
I think Gal Kremer can independently test those motors. He has a Tyto Robotics thrust stand
can’t wait for these motors to ship
Awesome. I love how Chris dives deep and truly innovates. That spreadsheet though... showing 6 decimal places / 10 significant digits for thrust measurements?
Great talk. There's not a more solid guy in FPV.
So interesting! I hope he applies this same core concept to smaller motors, too!
I love rocking motors with a 5mm hub on my 3.5” but motor choices are so limited. T mount sucks (hats off to those of you that can consistently get the damn screws into each hole for each prop, but a single 5mm shaft is just so much easier).
I’m rocking xnova, but would love more options!
Can't wait for Joshua to test these 🤩
The flight controller mount is a neat ass idea.
sounds like the fpv will be perfect with this guy products. will see
Chris is amazing.
I'll be interested to get my hands on a set of these to try them out. Chris Rosser definitely has his stuff together and is very knowledgeable, I always fear that efficiency and durability are inversely proportional - maybe he solved this one? Can't wait to find out.
I bought a set 😁 just because they sound awesome 🤯 looks like I'm building a fpv 😁 i have a mini 3 pro 😉 thinking about an 03 air unit to go with my DJI rc. I have been looking at the hobbyking orca as well. Couple months I've been seeing that around. Also saw a speedy bee frame 🤔 master 5 v2 i think. That looked interesting 😁 guess i need to go back and watch some more videos. Thanks for everything guys 😁
Super cool channel here... Keep this up!
Ive heard some people in the hobby speak ill of Chris rosser, one even actually referred to him as a charlatan. I must say though, after watching his motor testing video, i ordered the rcinpower WASP majors (because they had excellent test results)... and they REALLY are the best motors i have used yet. People can say what they want, but my own real life experience backs up Chris rossers testing. I am very grateful for the work that he has done for FPV, almost as much as JB. I trust both of their opinions over everyone else.
Yea, this is a tough one - I just posted something similar. However even though he is a bit cocky he probably does have some of the best results when it comes to technology in the drone hobby arena. His frame is cool, these new motors look awesome but very expensive, but what should the "best" cost? So I just hope he is a good dude and that our hobby is better off having him doing what he does for us.
By the way, there was that one moment in this interview (at time 30:33) when Chris said "they are likely to copy design ideas of my motor and that will make their motors better..." and Bardwell kind of laughed - I took it as Bardwell saying 'man, you are full of yourself". How would you interpret that laugh that he gave?
@@genjitsu7448 they know a patent/copyright means shit when China clones everything with no repercussions.
This is why I love using this guy's frames. So easy to get jello free video even with the different prop options, including dual blade.
I'll be trying out my first frame of his when I build the UL 7 soon. Can't wait.
The aos 5 v2 made me stop trusting his stuff. Hope to be proven wrong here..
Excited 😆
Ok, that vibration dampening idea is great - brilliant. One of those things that once you've seen it it's "obvious."
Wow! Electric motor designer uses magnetic analysis tools to optimize a motor design! Extraordinary! Whatever will he do next?
And yet nobody in FPV has done it before. It was so obvious, how did they overlook it? Maybe he deserves at least a little credit?
Hmmm . . . @@JoshuaBardwell
I would only have liked to have a little more images to accompany the words, for those of us who are not native English speakers it would be better. But it is spectacular to see how far this can go, thanks to people like you. Thank you so much
Cool!
wow! what an amazing salesman!! love it love it heres my CC!!! take it lol seriously , such a good topic!
That's very interesting, especially the frame stuff - to me, anyway.
I fly small quads up to 4" props. Recently, using a commercial one piece frame (2.5mm CF - carbon fibre) I have had big problems with loop stability with aggressive props, had to back off gains a damaging amount. A friend with another similar commercial frame (3.5mm CF) didn't. But, his frame was 33.5g as opposed to mine at 23.5g - a big difference in a 200g craft, 5%.
On the basis that the biggest contribution to stiffness comes from mass furthest from the centre of gravity, I then made a frame (I have the CNC tools) from 2x 1mm outer skin CF, with 2.5mm core of 3D printed polycarbonate/CF. Total was actually about 4.2-4.3mm as the 1mm skins were slightly less than 1mm. Big improvement, no stability issues.
Carried out some relative deflection measurements and it showed the arms (same XY plane shape) were a lot stiffer. Relatively, my original 2.5mm frame under a set load deflected 0.33mm, my friend's 3.5mm frame 0.11mm, my 4.25mm 'sandwich' frame 0.06mm. not only that, it only weighed 25.5g, less than his. So, a big stiffness to weight improvement.
It seems to me therefore that stiffness to avoid coupling into the gyro is very important. In the past on smaller quads (2") I've also got measurably better gyro/accel performance by AV mounting the motors.
Of course the downside of the sandwich approach is likely damage in a crash, the arms will be less strong as the 3D printed material does not contribute much to the overall strength. Depends how much one values performance against robustness.
The tool I use for designing this stuff is FreeCAD, and that does have an FEM (finite element analysis) module. Therefore since one is only after relative qualitative improvement, it might well be interesting to experiment with different shapes / cutouts using that.
Get this guy a Nobel prize.
Great interview. I would really love to hear about the entire process of filing the patent, including how he decided it was worth the costs... I was always under the impression that the costs were prohibitive.
Pre-ordered.
Logic and rationality are limited to what we know sometimes assume. Breakthroughs happen unexpectedly often enough thru accident. Therefore restrictions for preconceived efficiency can be counter productive. Many proto types are test for unexpected out of the box or reaffirm your expectations. Great discussion !
This sounds good! My only worry is magnetic dust.. There's a lot of that in the bandos we fly here around Denver, can really kill a motor quick when a bunch of it gets into an air gap. But hopefully the less leaky flux ring will help!
huh maybe someone needs to design a kind of plastic gap wiper thing that somehow scrapes the inside without getting stuck and without blocking too much cooling air. Or, maybe we start sealing the motors up and do cooling some other way...
This all sounds great on paper. More efficient means less heat which is great. Now lets hope the bell design can handle a beating.
Im interested in seeing JB's test of the motors now ahah
I hope for super efficient long range motors. I'm going to get that 10 inch AOS too I am super happy with my AOS 7 UL.
Chris is a treasure
Joshua Bardwell for president!❤
I found a long time ago that switching to 3d printed soft mounts with a firm tpu. Worked way better than purchased silicone soft mounts. Probably for this very reason.
Basically chris is a fpv engineer we all need💀
Chris this is great and thanks for the interview Joshua. Chris, have you tested at the edge of tolerance for all of the combinations of parameters that have effect on performance using the digital twin? Have you established the critical to function tolerances? Is there statistical process control on these critical to function parameters at the supplier? You can see where I'm going with this... When you are trying to achieve perfection, your ability to reproduce that would be the next challenge, requiring a very competent supplier with the latest in equipment and measurement capabilities. Cost versus precision and repeatability becomes the next hurdle.
I had a few years ago same idea virtually, but never tested it in production. In the video the cut around is very simple and I believe it is like very un optimized solution. I have been think to properly create mechanical resonators tuned for specific freq. by cutting the frame in very specific way.
Those cuts are pretty smart. Allows the rest of the base plate to vibrate, but not pass that into the stack, via a ridgid compliant mechanism. The same opposite of the gyro basically.
I dont want to put you on the spot chris rosser: every question puts him on the spot😂❤
do the bells come off easy? i fly Bandos and need frequent cleanings. the RCinpowers ive had dont come off even with the screw off
ello! I have the iflight nazgul evoque f5 with a Frsky reciver in it. The 2 antennas in the reciver 1 got ripped off. Should I be worried about it or will one antenna be fine?
Just ordered the new ones thanks Chris. I hope to see more of this kind of design tailored to long range and other niche kinds of things
Chris: Hyper Parameter Optimization for AI models…
Josh: mmm hmmm 🧐
He is absolutely right. If we were at the limits of motor technology, they they would all be at the same power level, amp draw, and performance. But they are not. As most of us know........
Excellent point! And pretty sure how he did it. Been on my mind for about 20 years........ i hope im right and he made it.
science is alway the key,! great job! I'm looking where to buy the motors in EU :)
Interesting video.
The method Chris has used here is very similar to the way neural networks (/AI algorithms) are trained to do whatever they do so it's not surprising that the tool he used is to wrap the MIT algo is 'borrowed' from that field. I think more and more engineering is going this way actually. Come up with a digital twin and throw millions of tweaked model variables at it.
I just ordered myself a set and I’m excited to put them to the test, i am curious now that they have been out for a bit if there is any feed back on the live flight pros and cons.
What have people seen be the best improvements in performance, compared to what are the negatives (if any) people have run into, and how may we avoid these negatives?
I'd love to see more low KV motors for use in robots, if there's a way to get just a few Nm of torque out of a 200g bldc at 250W for £50 we'd have jumping robots, amazing CNC machines and emobility for everyone!
Very interesting approach to the vibration situation. More recently I've been testing a change in the orientation of the main deck and how the arms attach. I'm finding some nice consistent improvements. Combining this with Rossers cut outs as well as bracing for the areas around the carbon bridges to the stack may lead to further improvement.
I'd love to push the limits of the motor software to inspect some of the unique motor designs I've come up with. Especially if it can deal with reluctance forces. Not just for FPV. Rosser and I historically seem to be on the same wavelength so I'm sure I can just wait and he'll come up with similar ideas on his own.
Lately I've been looking into things that there are no answers to. From motor designs, gearboxes, batteries, battery management, carbon integration and manufacturing methods. There's a shocking amount of bad engineering out there across a wide variety of industries...
You should clone both his motor and frame for the lulz.
I have a friend in the carbon fiber industry and when I showed him some of the best frames on market he laughed at the quality of the carbon lol. Just simply changing layups and resins you can get astronomically different results from just the carbon itself. Something I've been looking into lately myself for some non fpv related stuff
Still praying you 3 (Chris, Bob, Ryan) get together and tear the roof off the hobby...
@@Brian-S Most people from other industry is look at our construction materials and think it's comical how bad they are. I have tested some of the The best carbon out there according to the manufacturer and have found that there are actually quite a bit of improvements that we would gain however the cost is also astronomically high so the value proposition just isn't there. Cost is really the crux of everything. If you can't make it cost-effective enough to be used in any reasonable application, why bother developing it.
Yea I was going to say it’s not bad engineering but practical engineering.
We can’t have supper expensive fiber in these kinds of destructible quads. Trade offs.
yoooooooooooo. Now please Mr. Rosser do the same science to all the popular quad sizes! 3.5" next!
Is there going to be a smaller version optimized for 3.5 inch prop?
Very interesting, FYI there's a few minutes of dead space at the end of the video (39:58 to 44:23) that you might want to chop out?
Looks like he did. 😊
I like the approach on the frame, but could a good result also be obtained by mounting, say, three accelerometers at fwd, mid and aft and then applying some averaging/selection/rejection to the signals and at the same time potentially get some redundancy, along with some condition monitoring - eg chunk out of one of the prop blades, or pick up a motor shaft being bent or a winding o/c or maybe even shorted turn etc etc ?
These chips seem pretty cheap these days and SPI bus allows multiples connected without chewing more ports , and so on and so on?
What about the tuning ? If i change to this motors for my aos 5" what should i do with the tuning?
The cool thing about being "just a nerd" in the drone industry is, you're always professionally surrounded by all the engineers and designers etc. that don't have any personal interest in drone flying as such... ;)
Awesome Video like always, where can I get this awesome T-Shirt Joshua?
Hi can anyone help me I’ve been using an Acrobee 75 hd and when I arm the quad I have a warning saying RCSMOOTHING and it stays there through out my flight I don’t know what it is or how to stop it can anyone help me please
Question is: Will these innovations come to the smaller motors i.e. 2004 or under the 2207?
how important is deceleration in measuring responsiveness? and is it changed concurrently with acceleration response?
But what is the tool? The finite element modeller? Is it open source? There is many .?
It would be nice to see these motors for all size ranges of quads. It would also mean you could run the smaller motors meant for a race or freestyle quad in its size category on a bigger class quad for insane efficiency builds. Ie a 1305 or 1404 on a 4 in .
Whooo!
Great stuff! What is the name of the optimization algorithm mentioned at 10:19 ?
I don't quite get it clear...
Can anyone help me get live fpv video feed from my goggles onto a computer or phone for others to view while I'm flying?
I think I been trying to setup digiview on windows but I keep getting an error
I tried the digiview android app but my phone doesn't support video out via the usbc port, I guess that includes video in as well.
IPhone doesn't seem to have digiview available.
🤕
I'd like to know what power level variance is seen with multiple motors from a specific company, say one of the manufactures near the top. Is the variance a fluke of design or production? What about power draw differences?