Overshot — find the TikToks breaking out, and what they're selling

On-screen text

I'M STILL BUILDING A DRONE FOR THE DARPA LIFT CHALLENGE PART.1 I'M BUILDING A DRONE FOR THE DARPA LIFT CHALLENGE IT'S POWERED BY THIS 2 STROKE GAS ENGINE AND I GOTTA SAY IT SOUNDS PRETTY INCREDIBLE WHEN IT GETS UP TO SPEED PART.1 PART.1 THERE'S A LOT I TODAY SO I'VE ORGANIZED MY THOUGHTS INTO AN INTERNET FRIENDLY PART.1 NUMBERED LIST PART.1 QUICK RECAP WHAT IS THE DARPA LIFT CHALLENGE PART.1 IT'S A COMPETITION TO BUILD A DRONE THAT WEIGHS 55 POUNDS PART.1 LEAST 110 POUNDS OF PAYLOAD CHALLENGE? PART.1 FOR FIVE MILES THIS PART.1 SOUNDS SIMPLE AT FIRST PART.1 BUT ONCE WE DIVE IT TURNS OUT THIS IS REALLY REALLY DIFFICULT CHALLENGE? PART.1 ESPECIALLY BECAUSE THE COURSE CHALLENGE? PART.1 ISNT EVEN A FIVE MILE STRAIGHT SHOT THE COURSE CHALLENGE? PART.1 IS ONLY 1,000 FEET LONG SO THE PART.1 1000FT (300m) 1000FT (300m) Repeat x15 PART.1 30 TIMES SO IT'S A VERY DIFFICULT TASK FOR AN ABSOLUTELY PART.1 MASSIVE DRONE FOR CONTEXT UP UNTIL THIS POINT THE BIGGEST DRONE THAT I'D EVER BUILT WEIGHED JUST 5 POUNDS AND IN THE DRONE WORLD THIS IS PRETTY BIG PART.1 BUT FOR THIS CHALLENGE THE DRONE IS GONNA WEIGH OVER 150 POUNDS WHICH IS ON A COMPLETELY DIFFERENT MAGNITUDE OF SCALE PART.1 THIS INTRODUCES A LOT OF CHALLENGES WITH THE DESIGN SO LET'S GET RIGHT INTO IT PART.1 MY DRONE OFF THE DRONE LOOKED LIKE THIS PART.1 THIS WAS CLOSER TO A MOCK UP THAN A FINAL DESIGN MORE LIKE THIS PART.1 YOU CAN SEE THAT BEEN CHANGED AND FIRST OF ALL IT HAS A REAL FRAME BEFORE OUT OF 2020 ALUMIN EXTRUSION PART.1 THIS STUFF IS LIKE LEGO BUT MADE OUT WHICH MAKES IT REALLY GREAT FOR PROTOTYPING BUT IT'S KIND OF BULKY AND MY ORIGINAL PLAN WAS TO JUST MAKE DO WITH THESE LIMITATIONS BUT THEN THE FOLKS OVER AT CROWN KIRK INDUSTRIAL PART.1 THEY SAW MY ORIGINAL VIDEO AND THEY KINDLY OFFERING UP THEIR MACHINING SERVICES IF I WANTED TO MAKE A REAL FRAME PART.1 UP ON THEIR OFFER AND I DESIGNED THIS MUCH MORE REFINED FRAME WITH A PROPER TOP AND BOTTOM PLATES PART.1 I EMAILED THEM MY CAD FILES AND BEFORE LONG THESE REALLY NICE PIECES OF ALUMINUM ARRIVED AT MY DOORSTEP PART.1 I SPENT THE NEXT DAY DISASSEMBLING MY ENGINE TEST BED AND REASSEMBLING IT USING THE NEW ALUMINUM FRAME PIECES PART.1 HERE'S THE ENGINE ON THE OLD FRAME AND HERE'S ME FORGETTING WHICH WAY THE FRAME GOES PART.1 WAY OR THIS WAY FORGET IT WAS REALLY HOT OUTSIDE SO I TOOK PART.1 DINING ROOM TO BUILD THE DRONE SORRY MOM HERE'S A GOOD LOOK 30 EXTRUSION IN THE CENTER PART.1 THE PAYLOAD WEIGHT THE FINAL FRAME IS OF CARBON FIBER THIS ALUMINUM ONE IS JUST TO TEST THE FITMENT AND IT FIT PERFECTLY SO HUGE SHOUT OUT TO CROWN KIRK INDUSTRIAL PART.1 FOR CUTTING THESE FRAMES FOR ME SO HERE'S WHAT IT LOOKS LIKE THE FRAME HAS THESE TOP AND BOTTOM PLATES SIMILAR TO A NORMAL FPV DRONE RUNNING DOWN THE CENTER IS A 30 30 ALUMINUM EXTRUSION WHICH IS GONNA BE THE CORE BACKBONE OF PART.1 MY GOAL WITH THIS DESIGN REALLY STRONG BACKBONE WHERE THE ENGINE AND THE WEIGHTS ARE ATTACHING BECAUSE THAT'S WHERE MOST GONNA BE PART.1 NOW YOU MIGHT HAVE NOTICED THESE REALLY LONG TUBES COMING OUT THE SIDE OF THE FRAME THAT'S THE NEXT BIG CHANGE IN MY DRONE DESIGN PART.1 I'M GONNA BE USING SMALL QUADCOPTER ELECTRIC MOTORS TO CONTROL THE ATTITUDE OF THIS DRONE SO THE PITCH AND ROLL PART.1 MY INITIAL PLAN WAS TO HAVE A PURELY VEIN CONTROL DESIGN BUT THERE PART.1 WERE A NUMBER OF ROADBLOCKS IMPLEMENTING THAT KIND OF IDEA THAT'S WHY I'M SWITCHING TO THE MORE FAMILIAR QUADCOPTER STYLE MOTOR LAYOUT PART.1 THESE MOTORS ARE JUST GONNA BE USED TO CONTROL THE PITCH AND ROLL WHILE I'M KEEPING MY PARAMOTOR ENGINE AS THE MAIN SOURCE OF THRUST SO IT'S KIND OF LIKE A HYBRID DRONE A GAS ENGINE FOR THRUST AND THE ELECTRIC MOTORS FOR CONTROL PART.1 I'M STILL GONNA USE FIXED VEINS OF PASSIVE ANTI TORQUE PART.1 THESE WILL BE ANGLED TO DEFLECT SOME OF BUT THE VEINS WON'T BE ACTIVELY CONTROLLED AND TO SAVE WEIGHT I'M PRINTING THEM JUST ONE WALL THICK PART.1 TO USE REALLY STRONG FILAMENT LIKE CARBON FIBER INFUSED PETG AND TO PRINT PART.1 KINDLY REACHED OUT AND SENT ME THEIR CENTAURI CARBON 2 3D PRINTER IT COMES STANDARD WITH THE ENCLOSED PART.1 PRINT CHAMBER AND HARDENED STEEL NOZZLE SO IT'S READY TO PRINT ANY CARBON INFUSED FILAMENT RIGHT OUT OF THE BOX PART.1 AND SETTING IT UP WAS EASY BECAUSE IT RUNS A FULL SELF PART.1 TURN IT ON SO I WAS UP AND PRINTING RIGHT AWAY WITH NO DELAYS PART.1 OH AND IF YOU'RE PRINTING CARBON INFUSED FILAMENTS LIKE ME JUST MAKE SURE YOU HAVE A FILAMENT DRYER PART.1 I HAVE MY DRYER TO THE PRINTER AND IT PULLS FROM THEM WITH NO ISSUES PART.1 SO BIG THANKS TO FOR HELPING ME OUT AND I'LL LEAVE LINKS TO THE PRINTER IN THE DESCRIPTION BELOW PART.1 SO THESE FIXED VEINS FOR A LITTLE BIT OF ANTI TORQUE AND I'M NOT USING ANY PART.1 VARIABLE VEINS WHY DID I MAKE THIS CHANGE TO GET INTO THAT WE HAVE TO TALK ABOUT MONOCUPTER DESIGN THEORY PART.1 MONOCUPTER DESIGN THEORY THERE ARE A NUMBER OF SUCCESSFUL MONOCUPTER DESIGNS OUT THERE PART.1 OF HARD TO FIND HAS COME UP WITH UNIFIED NAME FOR THEM PART.1 I'VE SEEN THESE CALLED MONOCUPTERS WHICH IS MY PERSONALLY PREFERRED TERM BUT I'VE ALSO SEEN PART.1 THEM CALLED SINGLE COPTERS BALLICOPTERS AND ALL SORTS OF OTHER DIFFERENT NAMES PART.1 BUT ONE COMMON DESIGN FEATURE ABOUT THESE MONOCUPTERS IS THAT THEY ACTUALLY HAVE TO BE TOP HEAVY PART.1 IT'S A COMMON MISCONCEPTION TO THINK THAT FLYING DEVICES SHOULD HAVE MOST OF THEIR WEIGHT TOWARDS THE BOTTOM PART.1 THAT'S CALLED THE PENDULUM FALLACY AND IT'S A COMMON FALLACY WITH ROCKETRY AND DRONES THE FALLACY PART.1 DRONES THE FALLACY IS THE ASSUMPTION THAT SOMETHING IS GOING TO STABLE IN FLIGHT IF THERE'S MORE WEIGHT PART.1 TOWARDS THE BOTTOM JUST LIKE HOW A PENULUM NATURALLY STABILIZES ITSELF PART.1 THE ISSUE IS THAT FOR A FLYING DEVICE THE PENDULUMS DON'T REALLY APPLY AT ALL THE REASON PART.1 PENDULUMS ARE STABLE IS BECAUSE THEY HAVE A FIXED PIVOT POINT FLYING AIRCRAFT THEY DON'T HAVE A FIXED PIVOT POINT SO THERE'S NO ADVANTAGE PART.1 MOST OF THE WEIGHT AT THE BOTTOM AND FOR A MONOCUPTER STYLE DESIGN PART.1 IT'S ACTUALLY MORE ADVANTAGEOUS TO HAVE THE WEIGHT ALL THE WAY AT PART.1 THAT'S BECAUSE THESE MONOCUPTERS HAVE CONTROL FINS AT THE BOTTOM THESE PART.1 CONTROL FINS AT THE BOTTOM DIVERT THE AIR WHICH THEN TILTS THE DRONE AND WHERE THE DRONE TILTS ABOUT IS ABOUT PART.1 ITS CENTER OF MASS SO WE CAN THINK OF THIS LIKE A LEVER WITH THE FULCRUM OF PART.1 THE LEVER BEING THE CENTER OF MASS AND THE END THE LEVER BEING WHERE THE FINS ARE PART.1 SO BY HAVING MOST OF THE WEIGHT AT THE TOP AND THE FINS AT THE BOTTOM THE MONOPATTER HAS THE LONGEST PART.1 LEVER IN WHICH TO ROTATE ABOUT WHICH GIVES IT THE MOST CONTROL AUTHORITY IN MY LAST VIDEO PART.1 I MENTIONED THAT THERE ARE GIANT SCALE RC AIRPLANES THAT CAN HOVER JUST USING THESE CONTROL SURFACES AND SO PART.1 IF WE DRAW THE DIAGRAM FOR THIS THE CENTER OF GRAVITY AIRPLANES IS LOCATED WAY TOP WING AND THE CONTROL SURFACES ARE FAR BELOW IT PART.1 SO THIS RESULTS IN A VERY LONG LEVER WHICH MAKES IT EASY FOR THE AIRPLANE TO MAINTAIN CONTROL IN A HOVER PART.1 NOW LET'S TAKE A LOOK AT MY INITIAL MOCKUP DESIGN THE ENGINE LOCATED RIGHT IN THE MIDDLE AND THE FINS ARE LOCATED ALMOST EXACTLY ON THAT SAME PLANE PART.1 THAT WOULD MAKE FOR A VERY SHORT LEVER WHICH MEANS THAT THOSE CONTROL SERVICES WOULD BE DOING VERY LITTLE TO CONTROL THE DRONE PART.1 SO IN THAT EXACT CONFIGURATION IT WOULDN'T REALLY WORK TO FIX THIS DESIGN I WOULD HYPOTHETICALLY NEED TO MOVE THE CONTROL SURFACES FARTHER AWAY FROM THE CENTER OF MASS WHICH WOULD MEAN PUTTING PART.1 THE DRONE ON GIANT STILTS ADDS A LOT OF COMPLEXITY BECAUSE NOW THIS LANDING GEAR PART.1 HAS TO SUPPORT THE FULL WEIGHT OF THE DRONE JUST BUILDING LANDING GEAR THAT IS LIGHTWEIGHT AND CAN SUPPORT 1 HUNDRED AND SIXTY FIVE POUNDS THAT'S A PRETTY TOUGH CHALLENGE PART.1 TO SAVE WEIGHT WHAT I REALLY WANT TO DO IS I WANT TO USE MY 110 POUNDS OF DEAD PAYLOAD WEIGHT AS PART OF THE LANDING GEAR PART.1 THIS WAY THE PAYLOAD IS ESSENTIALLY RESTING ON THE GROUND AND ACTING AS ITS OWN LANDING GEAR SO MY DRONE JUST HAS TO SUPPORT ITSELF PART.1 SO THAT'S THE FIRST STRIKE AGAINST USING VEINS FOR CONTROL THE SECOND STRIKE PART.1 AGAINST USING VEINS IS FLIGHT TESTING ALL DRONES USE PID CONTROL ALGORITHMS TUNED FOR STABLE FLIGHT PART.1 TO TUNE A MONOCUPTER THE EASIEST WAY TO DO THIS IS TO USE A STATIC TEST STAND THE TEST STAND PART.1 IT HOLDS THE DRONE IN PLACE ABOUT ITS CENTER OF MASS WHILE THE DRONE IS FREE TO ROTATE PART.1 THEN YOU POWER UP THE MOTOR AND USE VEINS TO DEFLECT THE AIR AND TEST OUT THE CONTROL ALGORITHM PART.1 THIS IS A SAFE WAY TO TEST THE CONTROLS ACTUALLY TAKING OFF AND POTENTIALLY CRASHING PART.1 BUT THE ISSUE IS FOR A VEIN BASED CONTROL SYSTEM LIKE THIS THE MAIN MOTOR HAS TO BE SPINNING IN ORDER TO CREATE PART.1 THE AIRFLOW FOR THE VEINS TO THEN DEFLECT THAT'S PRETTY EASY AT THIS SMALL SCALE BUT FOR MY DRONE THAT ADDS A LOT PART.1 OF RISK MY DRONE SPINS A 55 INCH PROPELLER AT 3,000 RPM PART.1 JUST MAKING A STAND THAT CAN PIVOT BUT STILL HOLD THIS MONSTER IN PLACE WOULD BE A MASSIVE UNDERTAKING PART.1 SO INSTEAD OF DEFLECTING AIR TO CONTROL THE DRONE THE SIMPLEST METHOD TO CONTROL THE PART.1

Transcript

I'm building a drone for the Darpa lift challenge. It's powered by this two-stroke gas engine and I got to say it sounds pretty incredible when it gets up to speed. There's a lot I want to cover today, so I've organized my thoughts into an internet friendly numbered list. Quick recap, what is the Rift Challenge? It's a competition to build a drone that weighs 55 pounds, but then carries at least 110 pounds of payload for 5 miles. This sounds simple at first, but once we dive into the details, it turns out this is really, really difficult. Especially because the course isn't even a 5 mile straight shot. The course is only 1000 feet long, so the drone has to fly 1000 feet, turn around and repeat 30 times. So it's a very difficult task for an absolutely massive drone. For context, up until this point, the biggest drone that I'd ever built, weighed just 5 pounds. And in the drone world, this is pretty big. But for this challenge, the drone is going to weigh over 150 pounds all up, which is on a completely different magnitude of scale. This introduces a lot of challenges with the design, so let's get right into it. Where I last left off, the drone looked like this. This was closer to a mock up than a final design, which is going to look more like this. You can see that a lot's been changed and first of all, it has a real frame. Before, the frame was made out of 2020 aluminum extrusion. This stuff is like Lego but made out of metal, which makes it really great for prototyping, but it's kind of bulky and my original plan was to just make do with these limitations. But then the folks over at Crown Kirk Industrial, they saw my original video and they kindly reached out offering up their machining services if I wanted to make a real frame. So I took them up on their offer and I designed this much more refined frame with a proper top and bottom. plates. I emailed them my cad vials and before long these really nice pieces of aluminum arrived at my doorstep. I spent the next day to assembling my engine test bed and reassembling it using the new aluminum frame pieces. Here's the engine on the old frame. And here's me forgetting which way the frame goes. Well, I mean this way or this way. Forget. It was really hot outside, so I took over my mom's dining room to build the drone. So m. Here's a good look at the 3030 extrusion in the center. This holds all of the payload weight. The final frame is going to be made out of carbon fiber. This aluminum one is just to test the fitment and it fit perfectly. So huge shout out to Crown Kirk industrial for cutting these frames for me. So here's what it looks like. The frame has these top and bottom plates similar to a normal FPV drone. running down the center is a 3030 aluminum extrusion, which is going to be the core backbone of this drone. My goal with this design was to have a really strong backbone where the engine and the weights are attaching because that's where most of the forces are going to be. Now, you might have noticed these really long tubes coming out the side of the frame. That's the next big change in my drone design. I'm going to be using small quadcopter electric motors to control the attitude of this drone, so the pitch and roll. My initial plan was to have a purely vein-based control design, but there were a number of road blocks implementing that kind of idea. That's why I'm switching to the more familiar quadcopter style motor layout. These motors are just going to be used to control the pitch and roll while I'm keeping my power motor engine as the main source of thrust. So it's kind of like a hybrid drone. A gas engine for thrust and the electric motors for control. I'm still going to use fixed veins for a little bit of passive anti-torque. These will be angled to deflect some of the main thrust, but the veins won't be actively controlled. And to save weight, I'm printing them just one wall thick, which means I need to use really strong filament, like carbon fiber infused P-ETG. And to print it, Elagu kindly reached out and sent me their Santari carbon 2 3D printer. It comes standard with the enclosed print chamber and hard and steel nozzle, so it's ready to print any carbon infused filament right out of the box. And setting it up was easy because it runs a full self-check when you first turn it on, so I was up and printing right away with no delays. Oh, and if you're printing carbon infused filament like me, just make sure you have a filament dryer. I have my dryer set up right next to the printer and it pulls from them with no issues. So big thanks to Elagu for helping me out with this project and I'll leave links to the printer in the description below. So these fixed fans I'm using for a little bit of anti-toric and I'm not using any variable veins. Why did I make this change? To get into that we have to talk about Monocopter design theory. There are a number of examples of successful Monocopter designs out there, but it's actually kind of hard to find them because no one has come up with a single unified name for them. I've seen these called Monocopters, which is my personally preferred term, but I've also seen them called singlecopters, Boliter and all sorts of other different names. But one common design feature about these monopters is that they actually have to be top heavy. It's a common misconception to think that flying devices should have most of their weight towards the bottom. That's called the pendulum fallacy and it's a common fallacy with rochetry and drones. The fallacy is the assumption that something is going to be more stable in flight if there's more weight towards the bottom, just like how a pendulum naturally stabilizes itself. The issue is that for a flying device, the pendulum don't really apply at all. The reason s are stable is because they have a fixed pivot point. Flying aircraft they don't have a fixed pivot point so there's no advantage to having most of the weight at the bottom. And for a monopter style design it's actually more advantageous to have the weight all the way at the top. That's because these monopters have control fins at the bottom. These control fins at the bottom divert the air which then tilts the drone. And where the drone tilts about is about its center of mass. So we can think of this like a lever with the fulcrum of the lever being the center of mass and the end the lever being where the fins are. So by having most of the weight at the top and the fins at the bottom, the monopatter has the longest lever in which to rotate about which gives it the most control authority. In my last video, I mentioned that there are giant scale RC airplanes that can hover just using these control surfaces. And so if we draw the diagram for this, we can see that the center of gravity on these RC airplanes is located way up here on the top wing and the control surfaces are far below it. So this results in a very long lever, which makes it easy for the airplane to maintain control. in a hover. Now, let's take a look at my initial mockup design with the weight in the engine located right in the middle and the fins are located almost exactly on that same plane. That would make for a very short lever, which means that those control surfaces would be doing very little to control the drone. So in that exact configuration, it wouldn't really work. To fix this design, I would hypothetically need to move the control surfaces farther away from the center of mass, which would mean putting the drone on giant tilt, which adds a lot of complexity because now this landing gear has to support the full weight of the drone. Just building landing gear that is lightweight and can support 165 pounds. That's a pretty tough challenge. To save weight, what I really want to do is I want to use my 110 pounds of dead payload weight as part of the landing gear. This way the payload is essentially resting on the ground and acting as its own landing gear, so my drone just has to support itself. So that's the first strike against using veins for control. The second strike against using veins is flight testing. All drones use pig control algorithms that need to be tuned for a stable flight. To tune a monopter, the easiest way to do this is to use a static test stand. The test stand, it holds the drone in place about its center of mass while the drone is free to rotate. Then you power up the motor and use veins to deflect the air and test out the control algorithm. This is a safe way to test the controls without the drone actually taking off and potentially crashing. But the issue is for a vein base control system like this, the main motor has to be spinning in order to create the air flow for the veins to then deflect. That's pretty easy at this small scale, but for my drone, that adds a lot of risk. My drone spins a 55-inch propeller at 3,000 RPM, just making a stand that can pivot, but still hold this monster in place would be a massive undertaking. So instead of deflecting air to control the drone, the simplest method to control the

I'M STILL BUILDING A DRONE FOR THE DARPA LIFT CHALLENGE PART.1

Selling · Other Product

Why it worked

Summary

Structure

Original caption

More from @shazilarajput

Product placement

On-screen text