Tap to unmute
Creating the World's Most Efficient Drone
Embed
- Published on Mar 13, 2026
- Try Onshape for FREE - Engineers Get 6 Months of Pro: geni.us/Jabsgt
Ever since I have been building drones I have always wanted to build the most efficient and longest flying drone possible. In this video we will see if I can break the current world record of 3 hours 11 minutes and 54 seconds held by SiFly Aviation.
Bambu H2D 3D Printer: ($300 OFF): geni.us/BONdJCp
Best Mid-Range 3D Printer ($509 OFF): geni.us/2bh0uoq
Best Budget 3D Printer ($80 OFF): geni.us/ekn0LD
Form 4 Resin 3D Printer: geni.us/Ba2162
T-Motor Antigravity MN1005: geni.us/22qeIB
T-Motor 40 Inch Props: geni.us/M6FOde1
Tattu Semi-Solid State Battery: geni.us/nOXL
Hargrave Nanodrive ESC: geni.us/4BGZjbt
Flight Controller: geni.us/Wihbn
My Instagram:
/ lukemaximobell
AIRSHAPER CFD: www.airshaper....
Airshaper CFD RUclips Channel: / airshaper
SiFly's World Record Flight: • BVLOS, Unlocked: SiFly...
SiFly Q12 Drone: geni.us/DW06
#worldrecord #drone #technology 
Science & Technology
Try Onshape for FREE - Engineers Get 6 Months of Pro: geni.us/Jabsgt
Wouldn't it make more sense to use separate escs and use individual batteries on each arm to save weight on all that copper wire?
These kinds of batteries only come in very large sizes. And the weight of the wire isn’t that big of a deal in the bigger scheme of things. Having 4 smaller ESCs would likely weigh more as well
Great project, well done! Please drop me a line to my channel email, I have an invention, patent written, I'd like to show you it, I promise you will like it. 😊
Use Onshape for work, my company where one of the first large scale customers, loving it, the version management and collaboration capabilities are👌
Could it be partially solar powered? Like 200W panel?
Bro's out here collecting world records like they're pokemon.
Gotta catch em all 😂
Quite a fated campaign I'd say.
i mean, all he did was put parts together from things he didnt make, like playing lego
@moosehead4497youre almost making the connection of how all engineering is simply taking subcomponents and assembling something. Very rarely, if ever would someone invent something which uses exclusively things they invented or created as subcomponents
@guns21111boatbuilding not quite what I am getting at, but an engineer who builds his one parts, some of these parts maybe taking 5 or 6 iterations machining them out of a custom chosen metal alloy, writes the code from hand builds controller firmware etc... engineers typically professionally will work on pieces and not do everything, because it is inefficient expensive and time-consuming. But for all those same reasons it is impressive. Another you can also call an engineer is someone who buys the part from china from aliexpress, solders together some wires, downloads some 1 step to fly firmware and calls it a day, sure both are engineers but they aren't doing the same thing, completely different universes of technical ability. Without the war in ukraine, one set of engineers would still be working, while the other set would run out of cheap parts on aliexpress
This will sound pretty wild, but you're no stranger to reckless optimization: Use Aluminium wire in the arms to the motors. Al is 30% of the weight of copper, but 60% of the conductivity, so you should be able to match copper's conductivity using thicker aluminium wire, but at roughly half the conductor weight. You'll just need to use crimped connections on both ends to transition back to short copper pigtails. If you want to be extra unhinged, your wire insulation is probably quite thick and over-rated at 300V or 600V. You could strip and heatshrink the conductors to save a significant portion of the insulation weight.
What about putting batteries and ESCs directly next to the motors? Then long heavy and lossy power wires could be avoided entirely.
Endit: on second though, there is a big issue with multiple seprate batteries: keeping the charge levels balanced. There would need to be small wires connecting the batteries together to keep them balanced.
IU also thought it would be better to flying it at low altitude on the coldest day of the year. It looks quite hilly where he is.
@TildaAzrisk and they would block the airflow of the props
Concur, thought of both of these ideas too. Electrical contractors know about this but you would have to be in the trades. There is almost NO NEED for insulation AT ALL, they run along non conductive tubes, I am certain the two wires per tube can be separated and save almost ALL the weight of the insulation.
could just use no wires and use the arms as conductors. one thinner conductive CF or aluminum arm in compression, and supporting a bit of twist, with two aluminum guy wires below it that're structural as well as the other two legs of the motor. Just don't crash lol
I LOVE that you show the charts that ground your decision making!
I'm glad you guys find them as satisfying as I do!
@LukeMaximoBell The mathematics and the optimization using that math and calculus is brilliant. Well done. Fantastic content!
18:35 its funny that you can see the props spinning at different speeds because of the fps
One thing that always stuck with me from my aerodynamics elective is that a cylinder is one of the least aerodynamic shapes. The flow can’t stay attached, so it sheds vortices, creating a lot of drag. A flag flaps the way it does largely because of the vortex shedding from the flagpole, and it’s also one of the reasons early biplanes with all the bracing wires struggled so much with drag.
Seeing your frame design made me wonder how much drag is being generated by the round carbon tubes directly below the props. I can’t help thinking a small 3D-printed fairing or aero shroud in that area might reduce drag enough to be worth the weight penalty. Curious if you’ve thought about that?
You are definitely thinking along the right lines here! If you look at SiFly’s drone they have an airfoil on their arms for that exact reason. The only problem is that is much more complicated to make than a tube. I could add a lightweight 3D print to add on an airfoil. But this only helps if it reduces drag enough to counteract the extra weight, so I will need to look into that and run tests for the forward flight version. The good thing is at low speeds drag is pretty low.
@LukeMaximoBellmake the air foils out of thick paper with thinned epoxy painted on. Will be light
@TheDarkrad3rthat’s actually really heavy. Paper is highly absorbent. CF does not actually absorb the epoxy. It coats it.
@PresidioDelux I was thinking sourcing oblong/elipsoidal carbon tubes, somehow. The load is mostly vertical and so is the rotor wash...
That's a good point, but this thing is optimized for large slow props and that means the airflow should be pretty low. While they could make a more aerodynamic frame, they would need to do so without compromising strength or adding too much weight. Maybe the round frame could be the leading edge of an aerofoil type design? Wrap it in a thin layer for an inverted teardrop shape?
For the radios that were not working: You have long AC cables that also experience high dv/dt due to MOSFET switching, this creates changing electromagnetic fields which could be coupling into the carbon frame (which is conductive) basically making a large antenna that creates a large amount of noise that your smaller radios cannot be heard over.
1. To combat against this you could connect the carbon frame to the negative battery terminal over a parallel resistor and capacitor circuit. The resistor would be something like 100kOhms, and the capacitor something like 10nF. Basically the capacitor lets any high frequencies in the frame absorb into the battery (simplified) and the resistor bleeds any DC potential. Connecting chassis and the negative battery terminal would be the ideal situation but it would create a fire risk so I wouldn't recommend. Speaking of fire risks, make sure your capacitor is rated for the voltage you are running multiplied by 2, and make sure it is an MLCC which is much better for high frequencies due to low ESR and ESL.
2. Another suggestion is instead of using a 4in1 ESC just use single ESCs that are all closer to the motors. Doing this means your long wires are approximately DC (given you have appropriate capacitors at each of your ESCs) meaning much lower/no dv/dt and change in current during flight. Another benefit of going DC is that your wires do not suffer from the skin effect or inductive reactance as much, meaning you would need less copper to transmit the power (okay yes you would need a long thin cable to also transmit the control signal lol)
3. Pro tip if you want to use the above, by reducing the amount of inductance in the cables, we can reduce the amount of capacitance we require at the ESCs (lighter). This also feeds into how battery packs are designed. Inductance basically relates to stored energy in a magnetic field (similarly capacitance relates to energy stored in electric fields). SO, the less magnetic field we can store in a cable the less inductance. Basically all this to say keep your positive and negative wires as close as possible, even mechanically fixing them to make sure they don't separate and create more inductance (larger loop area = more inductance).
Hope this helps you! Keep it up
I feel #2 is going to produce another graphing event to see if 4 separate esc's with 2 dc cables each is lighter than 1 esc with 4 trios of ac power cables.
Dude, you make this 70 year old toppie young again. I just LOVE what you are doing. Can't wait for the official attempt!
So glad you are enjoying it 🙌🏼 thanks so watching!
One day, someone is going to invent the perfect battery and we will have drones flying everywhere
Sounds like a world record wasn't enough for you Luke you clearly want another one!
I won’t stop till I have all of them! 😂
@LukeMaximoBellwell your clearly addicted to record breaking!
@RobertGracie 100%! 😂
@LukeMaximoBellplease try a 24/7 flying "drone/plane/glider"
Solar powered
@LukeMaximoBell Yep clearly signs of someone who wants to be remembered as holding the most records for any South African in the GWR books!
nickname for your drone...after seeing what it did to your laptop, i think "the decapitator" fits lol
I think "droney mcDroneface" is better.
I can't find the paper again, but a few years back someone built a hexcopter that tilted ever other rotor clock-wise and the others counter-clockwise by 5-10 degrees. (I.e. rotated around the support tube so each motor had some sideways or yaw thrust component) This allowed the drone to produce horizontal thrust without pitching or rolling. Not a big deal for small drones, but with something as big as your flight time record drone, eliminating the power-surge to pitch and roll into gusts might help efficiency.
We used some large quadcopters that had that as well. As your drone gets bigger, the torque you can use for yawing doesn't scale as fast as the moment of inertia, so you get really poor yaw control. By pitching the props at a slight angle you can get more yaw authority at the expense of slightly decreased efficiency
overland power lines use aluminium wire because the current capacity per weight is actually better even though you need thicker wires at the same current.
I thought it was because the strength to weight ratio of aluminum is so much better than copper that the slight decrease in efficiency was worth aluminum cables being able to span much larger distances so you need fewer towers.
@massimocole9689 Fun fact, aluminum doesn't have the tensile strength to hold its own weight over long distances like that, especially in wind, snow, or rain (added force/weight). They use a composite (carbon and fibreglass?) core as the tensile component, and an aluminum outer as the conductive component. Google "ACCC wire" :)
Carbon fiber can be made conductive or non conductive.
and i thought that they do it because Al is way cheaper for the same current carrying capacity. BTW, Aren't they adding steel wire to make the cables stronger? Pure Al is not strong at all, and alloys probably aren't as conductive.
@massimocole9689 Yes, it's the strength to weight ratio.
If the standoffs are a permanent part of the build, can you make them the bottom part of the motor mount bracket? That would save even more weight. Instead of using metal bolts, there are nylon bolts available that should provide enough support to prevent the rotating of the booms. Even more weight savings!
Yes for sure! But they aren’t part of the final design. The drone simply lands on the batteries 😂
@LukeMaximoBell Use wooden Popsicle sticks or toung depressors, very light.
Aluminium bolts are also possible
@scalty2008 Magnesium also, just not very widely available
You can save more weight by making the legs the bottom piece of the motor mount clamp instead of two separate assemblies, half the number of screws
Exactly right, those legs are super heavy and have nothing to do with the drone once in the air.
Great idea.
To add to yours: some screws could be eiminated by incorporating snaps into the bracket design.
Additional weight savings could come from trimming corners and edges from brackets and drilling as many holes in them as possible while maintaining necessary critical structure capabilities.
What about a separate take-off and landing stand, like a circle that is raised by half a metre, for example?
This piece would be left on the ground, saving weight on the flying drone. With this circular stand, the hanging battery would not touch the ground and the two frame sticks would lie on the stand. It would have a fair amount of landing precision tolerance.
Also the motor clamps can be optimized, by making it a single part with screws on just one side
I remember Nicholas Rehm made a tri-rotor drone that had a mode to spin the entire drone like a windmill. The arms acted like giant propellers and dramatically increased the efficiency of the drone while hovering.
Yeah but they only increased the efficiency because the base efficiency with the tiny propellers was terrible
sounds cool I'll have to go check that out
If it spins camera spins as well so that drone not very useful in real life, but you can make drone with only 3 props to cut down power consumption, or even more just make helicopter drone with single motor and counter rotating rotors
Auto gyro drone probably good idea
Whoah that's such a cool idea! hahaha
i got a laptop low battery message while watching this guys drone fly for 3 hours
Hahaha brilliant!
AirShaper looks like a Wind Tunnel in a can
Masterclass of engineering ! Awesome work !
Thanks!!!😊
Best look at the carbon fiber dust from the holes through a microscope. I'll bet you'll use a vacuum cleaner and much better face mask afterwards. Be careful. And viewers too.
You guys are obsessed with carbon fiber dust! I do vacuum every time after I cut carbon fiber and I have an air purifier running on full next to me all the time. I can't show everything! haha
@LukeMaximoBellthat's amazing
what's scary about it under a microscope?
@turkeyphantthen you can see how sharp and jagged the dust is
There's a life.
5 seconds in and I knew this was shot in Cape Town.
Darpa has a heavy lift drone challenge. Maybe something to consider :)
That’s all I can think of watching these. How all these will be turned against us.
@Reanu_Keaves Start building drone jammers and interceptors
@Reanu_Keaves bird shot 😂
Not a bad idea 😅😅😅
That is very interesting. AI says rotary engines are choosen by most Teams.
When you see no ads in a video. The Video is the Ad
now try combining it with the solar drone!!!
You are thinking the right way! Unfortunately for the Guinness record no solar is allowed.
@LukeMaximoBellshould do it as a project then
@A_You-tuber Yes I will in the future. I am thinking trying to make an EVTOL that gets charged during the day and can fly throughout the night
@LukeMaximoBell record shmecord.
Subscribed. Really fascinating build process. And that mountain makes for a stunning backdrop. Good effort my man. Bravo!!
Thanks so much 🙏🏼 yeh Cape Town is beautiful!
Awesome once again!! Next record- altitude for a drone!
Thanks 🤩
There are videos of drones flying up mountains like k2 or Everest
And jail for that
Developing a flightcontroller would be a lot fun.
12:25 Tearing down a monster lipo pack like that in your living room is a brave move!
I was very careful 😂
it is not. the individual cells are well packed and protected as good as standalone battery. The only real risk is short circuit , but this is applicable for everything even working under the hood of a car or near capacitor of camera flash. I still would advice it doing it when the battery is deeply discharged. Than even if there is mistake it usually wont happened anything sudden and dramatic.
it can explode like a handgranade. do your research
Lipo batteries aren't the boobie-traped battery suprise commonly attested to, it's just the common man has lost an ability of common sense for the most part of the generalized population. Curiosity killed the cat, as to did the cars upon the highway in which the simplistic dumb non-adventurous had to cross to get home!! To prevent risk by determination of danger you've just got to retain some common sense of your abilities & intention! Personally I've been messing around & tinkering with lipos for ages however as yet (touch wood) have never struck a failure problem or error causing a lipo misadventure to proceed & conclude in seconds!! It's not just LUCK either because in general I'm extremely limited on that front without fail. Just saying.
@s1lv3rbordeaux47 one time i wasn't thinking and i chopped the positive and negative wire of my lipo both at once causing them to short and nothing happened aside from some sparks. that's the kind of stupidity that causes lipos to explode and of course thousands of people are doing stuff like it lol
Close enough, welcome back quetzalcoatlus
I wonder if you 3-D printed a paper thin air foil for the motor struts, it may offset its weight with more thrust and efficiency
Now you gotta make a drone that flies upright like the thumbnail
that first test flight at 10:12 looks like rolling a decapitation dice 😅 lucky you
Of many "inventors" designing crazy stuff, this is probably the most professional approach I have ever seen on youtube. Keep up the good work
4 individual li-ion packs and 4 escs under the motors
Cool idea but one drawback (not saying not worth it) is rotational inertia would be much higher .. not sure if that matters a lot
@askyellowhousecorrect!
amazing project & video
Thank you! 🙏🏼
You can save some more wire weight by going to solid core wire enamel, like what's used in the motor windings, along the lengthy arms to your esc.
I was going to suggest stripping the insulation off of one of the wires in each boom.
Skin effect may be an issue if using solid core wire, the higher the frequency the worse the losses are from skin effect.
@Tokenincdc no frequency at all
@michiel1362BLDC motors that are used for drones are in kind of a grey area between DC and AC when considering motor control, the ESC sends a pwm signal throughout each wire in a similar pattern to three phase AC. BLDC motors can operate at pretty high frequencies where the skin effect becomes a serious problem. If it was a constant DC current that would be used with a regular DC motor then no there would be no skin effect.
@Tokeninc No because all conductors are touching each other in silicone insultated wires used for drones. It is the same as solid core.
Didnt realise this video was south African until i heard your friend say 'Jissou'😂
Would work well as a repeater drone or for observation on the battlefield
Congratulations on this great achievement!
About 20 years ago, we set the same record, but for an electric RC helicopter. It was a team effort organised on the Ikarus forum, which no longer exists. We managed to fly the helicopter for just over three hours. Doing all that work and research by yourself makes your achievement even more remarkable.
At that time, we didn't have all the UAVs electronics and software tools that are available now, so everything was done using traditional maths and lots of thinking. The full flight was piloted; we had two pilots ready, but ultimately one pilot was able to complete the entire flight.
I really understand how difficult projects like this are, as it is my daily job to work on improving the efficiency of vehicles (both land and air) for either energy saving or power improvement.
Some tips in next comment.
Great work, my friend!
Please understand that these notes are not criticism; on the contrary, your work is already excellent. They are just some suggestions that may help you in your endeavour.
Regarding weight:
- You can probably use aluminium countersink screws and nuts in many places instead of steel counterbore ones.
- You could use XLPE-insulated wires, which are much lighter than silicone/PVC-insulated ones. Alternatively if you want to go to the extreme you could use enameled copper wire.
- You could use tapered CF tubes, but these are probably difficult to find in the appropriate lengths and diameters.
Regarding electric efficiency:
- While direct drives are a great option due to their simplicity, they are not by far the most efficient option in this case. If possible, I suggest replacing them with higher KV motors with a reduction gear. This will reduce weight and significantly improve electric efficiency.
- I did not see you check if the propellers were balanced in your video. You probably have, but if not, check it out, especially if you have done any repairs to them.
Aerodynamic efficiency:
- A nose cone and a trailing cone for propeller/motor hubs will improve the downward airflow efficiency, even if it means adding a few extra grams.
- Adding some kind of trailing edge to the frame tubes may help as well on the downstream propellers airflow when flying stationary however it will may add lot of drag when dynamically flying. But may worth considering.
Tappered tubes with small diameter at the motors ends may be the way to go as suggested before; they will reduce weight and reduce this aerodynamic loss same time.
- Propellers; if you really want to improve the overall efficiency, this is probably the main factor. The propellers you are using are not suited to this application at all. They are designed for much higher rotational speeds, differente flight speeds and much higher weight loads. And this is not the case.
Your propellers are your wings and should be designed for their intended purpose. I suggest you take a look at F1B propellers; their shape and characteristics, are probably more suited to this application than conventional drone propellers.
I hope all this helps.
Keep up your great work.
I know its more easy to repair with the screwed in mounts, but you could save some energy by gluing them on and removing the mounting points where the screws go through
Excellent video. Thanks for sharing, and I appreciate your effort.
Thanks!!
Its amazing that one person with no more than some 3d printers and a computer is able to get a longer flight time than a group of aerospace engineers with financial backing and access to more sophisticated tech. I think that its a testament to what you can do when you have an idea and want to make it a reality. So many people think that they cant do something because they need the right tools and money etc.. RUclips creators like Luke are proof that that isn't the case though.
I like his video, but to be fair thats both because its a commercial product and secondly (most importantly) becuase the level is currently relatively low for what the record is vs what can theoretically be achieved. Few resource-strong have taken an interest in this specific field/category as of now. 10+ hours of flight time could be possible with current technology in my estimation, leaving a large gap for hobbyists such as him.
@GoldenCroc I think your right that the bar was on the lower side, but this is just one example. there is also the fastest drone record that different people are breaking and rebreaking all the time.
This drone used a solid state battery - the previous record holder did not.
This is just incredible to watch as a man who love technology. Good job man!
I have one tip for you. Try flexi circuit board intead of wires. It can be printed on special printers. its like sticker around the tube. I used them many times you can order them from china.
I've flown high-efficiency drones. Your design falls perfectly into the design space and has excellent optimization. Large disk area, low disk loading, and low rpm to maximize rotor efficiency. Minimal structure to keep empty weight down and maximize the battery weight fraction.
The low rotor rpm and relatively low torque required to spin the propellers results in these type of drones having rather anemic yaw response. Canting the rotor slightly to tilt the thrust vectors can significantly improve yaw control. In a props-out configuration like your drone the rotors should be canted slightly inwards towards the longitudinal axis of the drone (front-right tilted left, front-left tilted right, etc.). A 10 degree tilt will decrease the vertical component of the thrust vector by 1.6% but will redirect 17% of the thrust sideways to help with yawing. In your X configuration tilted rotors will also have some interesting additional benefits: pitching forwards will directly generate forward thrust, rolling right will directly generate right thrust, etc.
A tip for flying your drone: keep your descents slow. The low downwash velocity of quads with low disk loading makes them susceptible to entering vortex ring state during descents and suffering a sudden loss of lift, something that generally doesn't occur with typical quads due to their high disk loading. In addition, a high efficiency drone generally does not enough excess power available to be able to power out of vortex ring state. To avoid vortex ring state, try to keep your descent rate below half of the downwash velocity.
clearly this guy could get 20% better flight times if he studied and learned the more technical quad setup info i.e. the pid settings and all the other adjustable rates in betafilght or beyond
he is using dji (closed source) flight controlling?
Can't wait for the next episode!!!!
use polystyrene blocks as legs. Engine mount blocks - add a plastic pin into arm to movement. Then add a clip mechanism to get rid of metal screws
amazing ingenuity i cant even make my pavo 20 pro take off flight
Carbon fibre is conductive. As long as you keep the arms separated, you could replace one of the wires with the arm itself. To save even more weight that is.
I was thinking ! Copper tape could also save with the wires plastic covering.
And with seperate ESC's under the motors you only need 1 positive wire as a supply to the motors.
I'm surprised that blade didn't destroy your laptop. Haha
Good work, I got 92 mins as my best flight time back 15 years ago :) 320Kv ftw back then
92 mins 15 years ago is crazy, with the batteries that were available back then
13:55 the shutter effect is so cool. Great benefit from the slow drones
This drone with these very large props would definitely be a great contender for altitude record!
it's my understanding that drones are legally altitude limited, so that attempt would clearly be preceded by some kind of permission application.
@meetim6271 So! why not?
@AuzFrog you cant take a drone into plane airspace in case a plane runs into one. its not like you have 360 degree visibility to see whats coming, you can only see where the camera points.
Brilliant! I can’t wait to see the total forward flight results and Guinness guys present.
It would appear that you have some work to do on your speed record drone. I suppose it’s always going to happen, the nature of competing. I haven’t watched to other guys video, but the title said it all. You are still our champ!
We could definitely get more speed by overcharging the batteries. But we don’t want to bore viewers will small gains! We are working on V5 now which will be a biiiig improvement 🤩
I love videos like this, where people design, build and improve custom RC stuff be really cool if you could test if adding light weight solar panels / capture would increase range or decrease cause of weight
I rarely enjoy watching videos like this
Amazing! so cool to see it fly. one thought: putting a hole for the wires in the root of the tube is not ideal; you could do the splice between the tubes at the actual main body hub, that way you have 2 wall layers where the hole needs to be made.
That was some serious engineering. Great job! It was entertaining to watch
Very cool! One potential upgrade I could see would be to use titanium bolts on the prop mounts. Rotating mass makes a huge difference in efficiency! I didn't see you mention the bolts so they might be already, but something to note.
But the props do just barely change speed on this drone. So i don't see real potential for improving efficiency with that. But maybe I'm wrong.
@Tobi-as but doesn't air resistance constantly decelerate the props so in my head it makes sense that light props = less energy to spin?
Congrats on another world record! Looking forward to the forward flight record attempt.
Congratulations! Never expected 3h is possible! Ever calculated to exchange the main wires from copper to aluminum? It may could save some gramms?
Thanks! Quite a lot of people have suggested it, so I will look into that for V2
This man single handed beats a team and professional company. This was awesome! Thanks for the video!
A couple of weight savings ideas to consider are to make the motor mounts single split ring clamps that slide over the ends of the arms and clamp down with a single bolt. Have a key piece in the clamp/mount that slots into a notch on the end of the CF tube to keep alignment. Also just get CF tubes that are the right length to avoid additional bolts and overlap material.
Do you have any numbers on how much extra time you would get for every 100g reduction?
Using a Drone to video a drone is crazy 😂
Guys, it seems you're a little late. In Russia, they have been flying a mass-produced quadcopter "Fenix-3M" for 3 hours and 56 minutes :)
Здарова)
@Sosiski_litivieпривет нонамчик
The power of Russian Vodka
@Aviator_2_2_9А Аргентум где?)
I can confirm this, Veles Labs is telling the truth👍👍👍
I love engineers doing real engineering. Great vid
@lukeMaximoBell Hey Luke, random idea for your endurance record attempt:
what if you treated batteries a bit like rocket stages?
For example, fly with two (or even three) batteries, and once one is fully depleted, you drop it to reduce weight and keep flying on the remaining one(s).
It would be interesting to run the numbers - whether two identical batteries make sense, or maybe different sizes, and whether shedding the “dead weight” mid-flight could actually extend total flight time.
You could even imagine a tiny parachute for the dropped battery so it lands safely.
Probably not practical, but as a thought experiment it feels fun to analyze 🙂
I'd assume the record stipulates that the entire drone must be airborne for the duration of the flight, putting any drone built to jettison batteries into a different class.
Imagine the lithium battery hits a sharp rock and suddenly grandpa thinks the Viet Cong found him
Definitely the eastern front can use staging
I like it!
@Flesh_Wizard Fair point of course, but fully discharged batteries won't make that much snap-crackle-pop.
I find it fascinating how professionally you approach your projects. Another new record, congratulations! I’m already looking forward to your next video.
It's called engineering....
Regarding the wire length, did you consider putting some of the battery mass, and motor controllers, out at the ends of the arms with the motors and rotors? That way you could keep the thicker power delivey wires much shorter, and just send low power control signals along much thinner wires. Lowering copper losses and overall weight.
Of course, this would increase the angular moment of inertia, and with it reduce attitude authority, but this thing isn't intended to do any rapid maneuvering anyway. Also, the battery packaging might not be as good that way, since you'd have to use multiple cells instead of a singular larger cell, so there will be some weight added by that. It's just a question of how much weight the extra battery casing add compared to the weight and copper loss savings of having shorter wires for power delivery.
i think its not that relevant to just save 20W probably better to keep the mass centered for stability
I was wondering whether the arms could be engineered to house the batteries, similar to a big flashlight. Could that arrangement also replace wiring between the controller and the motors?
i just love videos of people optimising a mechanical/electrical thing
What if you tilt the motors slightly in the forward direction? As we know, propellers are more efficient in forward flight due to reduced induced losses and improved inflow conditions. By introducing a small forward tilt, part of the thrust vector would contribute to forward propulsion, potentially reducing the need for the drone to pitch forward during translational flight. This could improve overall aerodynamic efficiency while maintaining stable hover performance.
The temporal aliasing is cool to watch, even though each propeller has almost the same speed, they look very different
Yeh it’s awesome hey! 😂 I did have an issue where one prop was spinning faster than the rest
8:55 That is definitely not right, the computer is still doing hard calculations, but for small stuff you won´t notice the limits it also have in performance, the more you draw on one file the slower it will get, and it will draw way more power than reading facebook...
The power loss analysis was very smart, good on ya mate.
3V per cell... Isn't that already damaging the battery? 20:05
I haven’t looked but I believe the semi solid state batteries have different (lower) nominal voltages per cell
Depends on how many times he plans to use them. in the end every use of a battery damages it to some degree.
I like this sound of motors an props... very quiet. like this project, looks so simple!
1:58 Next step up: old Wind-turbine blades ;P
_(Yes, THOSE... those you need a special car to move)_
aerodynamically, they are the exact opposite of whats required... they extract energy, they dont apply energy.
Man I thought you were going to build it from scratch thats wild
Elect Eng. here.. I hope your viewers realize that this “engineering” is not easy. Kudos to you for dumbing it down yet keeping it super engaging.
Thats crazy. Next is the distance record.
Maybe using some fine Kevlar strings to increase the stiffness of the frame, pure sine wave driver with pusher motor would make it more efficient?
LOL. watching the accident with the laptop, when I am watching through the exact same laptop 😢
8:59 I understand that video is sponsored, but this is incorrect. Being web based does not mean battery efficient. I can (and did) make a website that will load both CPU and GPU to 100%. Even the opposite, same program would run more efficient natively than web version because abstraction is lower.
Yes and no. The question is only client vs server calculations. If all the calculations are happening server side, it's great for your battery. If math is being done client side, it's battery intensive.
People seem to forgot we are making rocks do math by pumping electricity through them. It's witchcraft compared to 100 years ago.
@bobguy6542 I would expect onshape to perform CAD math client side. I might be wrong, but it would result in visible latency in every action. Of course stuff like CFD is offloaded.
@ivanjermakov Exactly. Even desktop applications offload things like CFD to the cloud (e.g. Fusion 360) but it's still going to have the same general load on the PC as any other software, unless it's coding exceedingly efficiently. However, there's nothing inherent about "web-based" that = more efficient.
@ivanjermakovCSG stuff is also done server-side. Overall it’s probably more energy-conserving for the client machine. But I personally don’t think it’s very professional to do serious cad on a laptop anyway.
Beautiful work Luke! Well done! and Thank you.
love the drone, now add solar charging as well.
It wont give enough to fly and charge but would really slow down the discharge rate on the battery
This is really amazing... 👍👍 We need many more engineers like this !
You started with an uncycled battery. NMC cells typically take about 5 cycles to break in. If you had fresh batteries, you should cycle them a few times before your next test flight - you will probably get around a 10% addition to your runtime.
I doubt it would be 10% as the batteries aren't being compressed. The compression helps get rid of gas (it squeezes it out) that can cause problems with the charge. It can take a couple of charges before the gas bubbles start showing up so he may be better off doing it fresh. If he had compression on the pouches, then a few charges would be good to get maximum charge capacity. I would bet after 5 charges of the loosely packed pouches, gas bubbles would form and bulge the battery and the bulging pouches would be producing less charge.
I would suggest charging and discharging several times while the batteries are compressed and then taking the compression off after charging them, and then putting the fully charged naked battery on the drone for flight. I have never done this though.
I am a robotics and mechatronics engineer and I used to work on autonomous drones (agricultural drones). I used to take batteries apart and reorganise them, we used to use Anderson connectors for the high current. Batteries are such a frustrating thing for drones, we tried all types of configurations, including 21700 batteries but always went back to pouches because of power/weight.
Looks like the legs were abandoned for the attempt... which makes perfect sense.
Why not incorporate the legs with the motor mounts?
Because for the final design I removed them 🙌🏼
Congratulations, great to watch.
Arms are shading airflow. Convert it to pusher to win some extra efficiently
Clean air into the props is more efficient, puller drones win in this scenario.
The air must come from somewhere so the arms will still shade the incoming airflow
Well done! I like it very much that you show your diagrams and your engineering approach
I have a few suggestions to loose a bit more weight.
1) Wires.
Since they run inside the carbon tubes, you can strip the insulation from one of the wires, the carbon tube should work as insulation just fine. You could even go so far at run one of the wires on the outside of the tube, and potentially save on insulation there as well.
2) Motor mounts.
If you design the mounts as 2 rings, almost shaped as the Ohm sign, and use the mounting bolts going through the motor to screw into the mount to clamp them down, you can save all the middle material in the mounts, and the 4 bolts to screw the 2 halves together.
3) Carbon tube connection.
Instead of running 2 bolts through the rod, just use a plastic strip, or potentially a little 3d printed plug. The twisting forces shouldn't be high enough to tear anything there.
4) Legs.
Print the legs as flat parts with a slot down the center, so you can slide them together and glue in the middle, that would wastly increase their strength.
Just a few suggestions. Good luck with the record!
But carbon fiber is an electrical conductor ¯\_(ツ)_/¯
Rethink the isolating properties of carbon
@geuzeg Yes, carbon IS quite conductive!
carbon fiber is conductive
Carbon might be conductive. The resin is not, at all.
I’m on the same journey!! This will be great info!!!
9:00 i have to correct you on that, running a cad software inside of a browser does drain your battery a lot faster than browsing normal websites, that's because most modern browsers allow using webgl or webassembly which is basically running code in your browser and it will consume way more battery and use way more resources and it's not that efficient so an optimized dedicated cad software will be more efficient
Made my own comment about this, didn't see this one first :D
Essentially, if the same feature-set and level of optimization would be employed in a native application instead of a browser based application, it would actually drain less battery. Current dedicated CAD applications are just very resource intensive to begin with.
@raremc1620yes but the application running in a browser even if it's using wasm results in less efficiency since it's still not as fast as native code and it also has to run a browser, so an optimized cad will be more efficient
@CreeperITA104 I mean, that is what I said..!
I completely agree, my point was that making a native application with the same "thought" put into it would result in less battery drain than any browser version would.
Love the data and presentation!
What about a 5 prop lightweight build where you have one primary larger prop that handles the lifting in the top center and 4 small props that only kick in for auto balancing and steering?
If you have one main prop and one sub-prop, you have a helicopter. Are there any benefits to going from one vertical sub-rotor to 4 horizontal ones, and would the 4 sub-props have enough authority to counter the torque of the main rotor?
Dunno. Probably worth modeling though.
Listen you are amazing! 3H!! Very good job! I am glad for you!🎉
13:39 Wow, it's so efficient that the prop blades barely need to spin! 🤣
The beauty of the camera shutter speed! hahaha
Alternate title: How to keep a LiPo in the air for as long as possible.
- The motor mounts could be a single screw if you just print a tube that slides over the carbon fiber pipe. Add a flat spot for the motor and screw a single screw into into the side to join the mount and the tube. Single part. Weight savings on screws.
- Wire resistance changes with temperature - maybe pumping the carbon fiber tubes with helium might be an option? Weight reduction + wire coolant.
- When designing weight-bearing components like the legs, avoid sharp angles that line up with the layer lines. Do large radius curves. Sharp angles create stress concentrations.
- Cylinders are not very aerodynamic. Add an upward-facing symmetric airfoil to the carbon fiber tubes under the props, so the rotor wash can stay attached to the cylinder. Smoother downwash column = less electricity used to hover. An alternative could be angling the tubes 45 degrees upwards, so they're not perpendicular to the downwash.
Could he take the bolts out and use carbon pins with a dab of glue?