have you ever seen one of these little helicopter toys that you give them a spin the thing just flies away I love these things growing up so I got my kids some although I think I played with them more than my kids did which got me wondering how high could you make one of these things actually go and being a competitive male with access to CNC machines I took that idea way too far making basically the Formula 1 version of one of these helicopters it looks pretty different but it works the same way you spin it up and let it rip although spoiler alert I did not find out how high you can make one of these go because I made a gigantic design error with this thing let's just say it's really good at uncontrolled flight into terrain oh go but even though I didn't achieve my goal it's still a really cool helicopter chalk full of cool mechanics interesting physics and egregious design errors so join me as we see how far we can push a cheap novelty made for children these helicopters are really simple you spin them between your hands which stores up kinetic energy in the propeller which blows air downwards generating thrust and they fly up and obviously if you spin them faster they're going to go higher but there's actually a pretty firm limit to how fast you can spin these things if you think about the tip of the propeller spinning around it's tracing out a circle really fast and if you spin it fast enough you can make it go supersonic which would make a sonic boom which is super cool but it also makes an enormous amount of drag and the propeller kind of stops working at that point but even if you spin it below that limit it doesn't do nearly as well at least as I would have thought because Air drag increases like this spinning 10 times faster is about 100 times the drag think about it this way the spinning propeller has a certain amount of energy like a gas tank which it can use to go up or to fight Air drag spinning it faster gives it more fuel but it ends up spending 99% of that fighting Air drag so what if we ran the propeller a lot slower where it's going to be more efficient because it's not fighting air resistance and instead we spin a flywheel really fast the propeller would be spending a lot less energy overcoming Air drag but it would also have a lot less energy stored in it but we'd make up for this with a fast spinning flywheel which lets us store a lot of energy without all the Air drag of a propeller I ran some simulations of a fairly slow propeller a propeller at the supersonic limit and then a propeller augmented with a flywheel which does this it should be able to go about five times higher but there is a really big problem with this design that we're going to have to solve we're going to have a big propeller with a flywheel attached to it which is basically a chunk of Steel if it goes way up into the sky it's going to run out of energy and then just fall down like a brick which is super dangerous and it'll also pretty much destroy everything when it hits the ground my original plan was to have a parachute that it would deploy when it topped out but if you think about it this thing is going to be spinning and doing who knows what it's hard to imagine the parachute not getting tangled up and everything and it still sending into the ground just with a parachute wrapped around it but I really wanted this to work I went pretty far down the rabbit hole of a pyrotechnically deployed parachute basically shooting the parachute out of a cannon so it gets really far away from the blades quickly but you can still imagine scenarios where it wouldn't work like if it was tumbling and it shot it straight down the helicopter would just fall into the parachute so I decided to go with another strategy which is a lot harder than the parachute but I think it could work let's go back to our fuel tank analogy it takes energy to make the propeller go up but this process works in Reverse when the helicopter tops out and runs out of energy you can reverse the angle of the blades and the air blowing past them as the helicopter descends will spin them like a windmill which refills our kinetic energy fuel tank and then right before we hit the ground we could change the angle of the blades back to normal which generates downward thrust with the energy we just stored up which slows the helicopter down so that it can land the downside is that it's hard to do and we have to make blades that can change their angle and I didn't come up with this it's called an auto rotation and it's what real helicopters do if their engine goes out in theory we can now land it but unfortunately the design still doesn't work so the flywheel is going to be trying to turn the blades like this but there's an equal and opposite reaction that's going to try to spin the flywheel this way but that won't actually happen because the flywheel is going to be acting like a gyroscope basically when you spin something really fast it resists any changes to its orientation so there will be a torque trying to twist the flywheel but it's going to resist this seems like a good thing because we want the blades to spin not the flywheel but the fact that we're trying to spin the flywheel and it's resisting will make something very non-intuitive happen that is easier to show than explain so I made this little thing with a motor spinning a flywheel it's sort of like a mini version of the helicopter with a flywheel sticking out of the side so I'm going to spin the flywheel up and watch what happens when I try to turn it trying to twist it this way results in a torque that twists it this way and I'm not going to get into the specifics of why this happens but it's called procession and it's really bad because it'll just make my helicopter flip over and crash but there's a trick that we can do if we put a second flywheel on the other side of the helicopter spinning in the opposite direction it'll process in the opposite direction and cancel out the twisting on the helicopter the last big question is how to connect the fly wheels to the propeller I decided to use gears but they're kind of weird because the fly wheels are at a 90° angle to the propeller and here's how they work imagine you have the shaft for the flywheel and the propeller and then you draw two imaginary lines to where they intersect then you make two cones going from that point to each shaft the cones will touch and if one spins so does the other one if we looked at this from the side it would look something like this then if we delete the tips of the the cones and add some gear teeth we have the gears that we're going to make these are called bevel gears and I need them to be very light so I have to make them myself and every other part of the helicopter so let's do it the gears start as a blank cut out on the water jet then they go into the 5axis CNC which can turn the part in any direction which makes Parts like this a lot easier here's the newborn gear before his teeth came in I'm using a special cutter with the shape of the gear tooth in it to cut each tooth one at a time that turned out a lot better than I was expecting it's pretty much the same process for the fly wheels but overall I just had to turn the tunes and crank out a bunch of parts because unfortunately the entire helicopter is metal I'm pretty sure I used every metal working machine in my shop on this and now the fun part putting it all together and seeing it not work oh yeah I waited so long to see this thing spin finally all the parts are made I got them put together and it looks really cool the most important thing is the rotorhead and the blades which are on a pivot so we can change their angle by pushing and pulling on them and this is kind of tricky because we have to be able to do it while they're spinning the secret is this little Servo on the bottom which is stationary when the blades are spinning but it has a little control rod that goes up through the main shaft and can push and pull in the blades to control their angle and then the the lower part is designed entirely around the bevel gears they have a 5 to1 ratio meaning the fly wheels will spin five times faster than the blades and they were tricky to make because they're going to be spinning up to 24,000 RPM so they have to be balanced really well and then the space in the middle holds all the electronics for controlling it measuring altitude all that stuff so I just finished installing all the electronics but I was looking at the flywheels and it occurred to me that I'm in idiot remember I was excited that the two flywheels will cancel out each other's procession do what else they cancel out their gyroscopic effect as well which this design depends on the Flywheel spinning the blades forward will make an equal and opposite reaction that spins the body of the helicopter backwards if I try to fly it the body's going to start spinning backwards at 1,000 RPM it's going to start shedding Parts it's going to crash I guarantee it's not going to work I have no idea how I missed this they usually do a demo of this on the first day of physics class they even talk about it in the physics book I got from my kids when they were babies I think I just had tunnel vision thinking about procession and didn't look 2 in to the right and the right thing to do would be to redesign this but we put too much time into this I think it's time to embrace the sun cost fallacy we're going to let it flow through us and we're going to transition to a different kind of working called hacking so here's the plan airplanes have these things on their wings called control surfaces which move like this when air flows over it it's redirected which makes a force that pushes the wing down if I move this one up and this one down it'll generate forces like this that will twist the wing what if we stuck something like this on the bottom of our helicopter we could use the air flow from the blades to twist the wing and keep the body from spinning let's find [Music] out and I know this is ugly and just kind of bolted on but remember we're hacking here it is all together and the air flow from the blades will go down over these control surfaces and hopefully generate enough torque to keep things from spinning and I'm just going to gloss over the second Wing because it just isn't important right now you can probably guess what I'm trying to do I think we're at the point where we just have to try it but before we can do that we need something to get these flywheels spinning really fast this turned out to be a bit more complicated than I was expecting and it kind of turned into its own whole project and the result is this dock thing which the helicopter sits on like this there's a motor on the side with a special coupling to interface with the flywheel and then there's a little computer that controls how fast it spins everything up and it will slowly ramp the fly wheels up faster and faster and when everything's going fast enough it releases a trigger which quickly pulls the motor out of the way so the helicopter can fly without hitting anything it communicates with the helicopter over this little Breakaway cable the little computer tells the helicopter that it's good to fly and if it's flying autonomous ly it'll take off to the heavens or I can start flying it with an RC controller so it's time to do some testing but I really don't want to I'm I've just been putting it off because I'm too afraid of what's going to happen but I came up with a scheme that's relatively safe I'm going to have the helicopter held down with these guides they'll kind of let it fly but they'll keep it from tipping or doing anything totally crazy I have it set up so that when I press this button the helicopter will spin up and I've got this piece of polycarbonate to cow behind because just seems like a lot of ways it could violently disassemble itself directly into my face so let's see what [Music] happens there's something really weird happening here I can only get it to like 30% of the speed that I want and even at that level our motor is completely maxed out we're melting wires burning up speed controls my guess was that there's too much friction in the system so I swapped the bearing grease for light oil which makes them super low friction which didn't seem to make a difference then I spent ages trying to adjust how the gears fit together which didn't seem to make a difference then I gave up and just got a bigger motor which didn't seem to make a difference then I was thinking maybe the coupling between the motor and the flywheel was binding up so I redesigned it but it didn't seem to make a difference then I gave up again and got an even bigger motor which didn't seem to make a difference and then it finally clicked it's the flywheels so obvious I don't know how I didn't see this so as this little flat part of the flywheel spins around it's pushing air out of the way and I just ran the numbers on this and it's astonishing to spin the fly wheels at 20,000 RPM it should take about 20 horsepower which is a lot of horsepower now that we know the problem at least it's pretty easy to fix I'm sealing them up with a super light heat shrink film there's no more big flat surface that we're dragging through the air it's just a smooth cylinder think of spinning a Paddle Wheel in water versus a smooth cylinder it's way better so let's see if it works 3 2 1 lift off we have a lift off 32 minutes past the hour lift off it is finally flying at least an inch but it's lifting itself under its own weight although I can't really touch the control surfaces with these retainers so I'm going to take those off we're going to put on training gear which kind of like training wheels for a helicopter it makes it harder to tip it over I'm just going to fly it manually with an RC transmitter here we [Music] go throttle [Music] over oh no man I am this close to just destroying everything thank goodness I have the training gear on there they are really earning their salary let's try a couple more times the helicopter is not failing to fly I'm reversing the pitch of the blades to generate upward thrust and Shove It Down onto the ground because I can tell it's going to crash so I've got the control surfaces pegged all the way trying to turn the helicopter left and they're just being overpowered by the torque trying to spin it right I don't think there's anything this Wing is going to be able to do so I think we basically just have to try to make a bigger Wing we're doing the same thing we did before just long I think this is about the biggest Wing that we're going to be able to fit into this thing and longer is better because it's going to collect more air flow from the blades which will also be going faster because we're near the blade tips and it will generate more torque because it has more leverage I hope this works because it's pretty much our last chance so here we go [Music] come on the more lift you generate the more it torqus the body of the helicopter so I'm giving it just enough thrust to barely lift the helicopter which is the best possible case and the wing is still being overpowered by The [Music] Twist man that sucks I just don't think we're going to be able to s solve this spinning issue without a complete redesign and rebuild so let's tentatively call this part one but we're still not done here I mean we've flown it what 6 in we got to take this thing out and just go for it I'll get the wife to come out hopefully she'll even be impressed if she doesn't realize how messed up it is it's midnight right now so tomorrow doing it live all right 6:00 a.m. and I'm rideing my bike because I thought it'd be interesting to show you what I do every day my routine is I think the most important thing that I do it always starts with exercise and it is insane how good this is for my brain once I'm done with that I sit down and spend 30 minutes learning something new and it is crazy how much you can learn if you do this consistently so how I've picked up a lot of my skills and it's also why I'm a huge fan of this video sponsor brilliant brilliant is a tool that's designed to help you learn technical stuff and it works really well it's why I've been talking about it for years it'll take a technical subject like Ai and present it as a series of interactive lessons which have you working with real data and solving actual problems so you're learning by doing which is just the best way to learn and it's so much more than AI they have literally thousands of lessons across math science data analysis programming all kinds of stuff and it'll take these complex subjects where it's hard to know where even to begin and break them down into a series of bite-sized lessons that buil build one on top of the other which is really good for learning a bit every day so you can probably see why I like this so much and it is shocking how the knowledge compounds it doesn't feel like you're doing that much dayto day but do it for a month and suddenly you know calculus it's crazy so if you want to learn a bunch of interesting technical stuff that's going to level up your problem solving abilities or just maybe relearn the math you forgot from school you should check out brilliant you can try it for free for 30 days just go to brilliant.org stuff made here you can also click the link in the description or scan the QR code and you'll also get 20% off an annual premium subscription and that's it thank you BL for sponsoring this video and thank you for taking the time to check it out all right the time has come to give this thing the beans so what do you think this thing is going to do I think it'll fly up really high and be awesome that's what you think that's what I think and then it'll land and pieces will break off this pups got to fly man I'm nervous this is a lot of work that's about to be probably destroyed all right hit [Music] it [Music] and baby's purring W watch out oh gosh watch [Applause] out well I don't think the training gear really helped 2.7 pounder all right scale of 1 to 10 with what 10 being scale one to 10 I'd say 10 that Landing uh that's a 10 it was a good Landing I don't know what I'd call really satisfying to finally see it go yeah and then such a epic fail oh it was a nice try it was good it's nice try I think I know how to fix it so part two check back in a while [Music]