I'm still building a drone for the DARPA Lift Challenge P.1 #challenge #DIY #diyproject #drone #trending #foryoupage #fyp #viral
The attitude of the drone is to do what I'm familiar with, which is electric motors in a standard quadcopter motor layout. These motors are relatively small and they only produce a couple of pounds of force each, but for my use, that's fine since these aren't producing any lift for the drone. The drone isn't going to be very responsive with only a few pounds of thrust, but this isn't a freestyle FPV drone, it's gonna stay close to level. So a few pounds of thrust to control attitude should be okay. My design now has 1 meter long carbon fiber tubes extending from the center of the frame. The motors are located at the end of the booms, away from the main downwash of the engine. Now, the huge advantage with using these motors for control is that I can tune the PID values of the drone without the gas engine ever being on. This makes testing a lot safer and easier to accomplish. The drone is gonna sit on a balancing stand that pivots about the center of mass, similar to the monocopter test stand which I showed earlier. And a huge thank you to Burnhild Industries for helping me with the engineering of the test stand. This stand still has to support the full 165 pound weight of the drone and have it pivot as smoothly as possible. So there was a lot of work involved in this design. Anti-torque. So we have electric roll control. What about countering the torque of the main gas motor? The paramotor engine, it spins the propeller counterclockwise, which means the main body of the drone wants to spin clockwise. My drone needs some way to counteract this torque. A lot of people in the comments have mentioned counter-rotating propellers, which definitely would be the best choice. But there's no off gearbox that I could just bolt onto it. Engineering a counter-rotating gearbox is an entire project of itself. And so if you're skilled enough, it's possible. But for me, I'm sticking with my single propeller and figuring out some other way to counter. Now, before I came up with a solution to counter the torque, I need to know exactly how much torque there actually is to counter. This is a common calculation you need to make with full-size helicopters, since many helicopters have a large main rotor and then a small tail rotor to counter that torque. The issue is the actual calculation is more of an estimate than a single set in stone number. I'm gonna spare you the math, but what's important is that there's an efficiency constant, which is how efficient the engine is at doing work. And that efficiency constant is just an estimate until you collect real-world data. As a baseline using ballpark estimates, this equation gave me a value of 12 pounds of anti-torque force needed at 1.1 meter away from the center of rotation. But I really needed a way to verify these calculations, make sure that they were correct. And unlike pitch and roll testing, this torque testing needs the engine to be running. There's no getting around that. Air Conception R200 Xtrem. It's time to dive into my drone, the engine. This is a two-stroke 190cc gas engine for paramotoring. The reason I chose this specific engine was because of its thrust numbers. The manufacturer has videos of it producing over 200 pounds of thrust. That gave me the confidence to choose their engine over any others, most of whom didn't even publish any thrust values. This is because the people who make this engine are high-performance engine enthusiasts. So they care about making the most powerful and lightweight engine possible and not catering to the more everyday pilot. And as I would soon find out, that means that there's barely any beginner information on how these engines work. The manual is bare bones with not a lot of information. The internet isn't that much better, since most people are flying pre-built paramotor setups and not buying just the bare engines themselves. So join me on my long journey to get this engine up and running. So first up, let's tour around the paramotor engine to get familiar with all the parts. There's the main cylinder head where the main piston is. The spark plug goes from the cylinder head to a magneto on the crank. That's what generates the spark. The magneto has an electrical spade connector on it, which is for the kill switch. Ground that spade connector to ground. The spark can't spark and the engine shuts off. To turn the engine on, my engine has a starter motor. Paramotor engines can also start, which ends up being like a full pound lighter, which was pretty enticing. But there's no way I'm pull starting this death trap. So electric start it is. The electric starter motor is a brushed motor. So there's one large terminal for power, for ground, this engine uses chassis ground, so the entire engine block is the ground connection. There's a big heavy exhaust pipe and since this is a two-stroke engine, the size and shape of it is actually critical for performance. So that has to stay. There is a small muffler at the end, which I theoretically could remove to save weight, but it's made out of carbon fiber anyways, so not a whole lot of weight to save. The engine, it drives a 3-to-1 belt reduction to the propeller hub, which is on a clutch. The clutch isn't as good for performance, but it is good for peace of mind. If there was no clutch, then any spin of the propeller could potentially start the engine and take my head off. So I'm fine with having a clutch. For throttle, the engine uses a diaphragm carburetor with a throttle lever. The throttle lever is spring-loaded, so there's no immediate need for a return spring, probably. And because this is a diaphragm carburetor, this engine can operate in any orientation. I suspect this is because paramotor pilots often do flips and roll stunt moves. So it's important that the carb gets fuel in any orientation. It's perfect for my drone because the engine is laying on its back, which is not the usual orientation for this engine. You buy the engine itself, it doesn't come with a throttle cable or a handle. Paramotor specific handles they cost like $200 for no reason in my opinion. So I purchased a generic throttle cable kit on Amazon for 10 bucks and I'm using a servo to control it. The manufacturer sells the official throttle cable for pretty cheap on their website, but I missed it because I just assumed it came with it. Now the set screw in the throttle cable is a 4-40 machine screw of all things and it's not included. I never expected an engine made in France to use American screws. But it turns out the carburetor is an American style Walbro carburetor. So the throttle on the carb uses American screws. I know this isn't but this is just one of what felt like hundreds of tiny small issues that come with operating an engine like this. Dealing with this engine has been an ordeal, to say the least. I could go on for hours about the nitty gritty details of this engine. I'll save that for another video. But here's a cheat sheet of some important information that the manual is critically lacking. Because the paramotoring sport requires licenses and taking in-person lessons, I suspect that most of the information in the paramotor hobby is passed from word of mouth. So there's not a lot of information on the internet about all of these nitty gritty details. Okay, so there's the engine. How do we control it remotely for use in a drone? Alright, so I've just finished up wiring up the heart of the drone. And so before I go ahead and install this into my drone over here, I thought I'd go over what's actually going on inside the brains. So the main flight controller is a SpeedyBee F405 V3. This is an older one, that's a couple years old, but I like it because this has the older BMI270 gyro, but I think it's more vibration resistant than ICM gyros that are everywhere. Um, the O4 Pro is the gyro gate because they're using those cheap gyros now. So that's why I'm sticking to a tried and true BMI 270 from a couple of years ago. In this stack, this is not a speed controller, this is actually a 12S power distribution board. I'm not gonna need 12S, but my previous version of this control box, it only went up to 4S and that wasn't good for my starter motor. So let's see what's going on here. So first of all, so yeah, the speed controllers aren't here. The speed controllers I'm using are standalone, old school style speed controllers with PWM inputs. And so these are wired up to that distribution board down here. And through these cables, and I've labeled them so they don't get mixed up. They only need signal, so I've isolated them to just use signal and ground. So I'm gonna run servo extensions, um, in between these. And so these will just hook up just like, in like that, and I'll run really long extensions to them. Uh, for receiver, I'm going for Express LRS and I'm actually using a Beta FPV PWM receiver. Set up so that it's actually, uh, outputting Crossfire, the standard Express LRS protocol over these two channel wires. So as far as the SPD is concerned, this is a completely normal Express LRS receiver. It doesn't know that there's anything funny going on. But because this is a PWM receiver, I'm remapping the extra servo channels to other servo functions that I need. This is the throttle servo, so it uses a standard servo plug, which can just plug right into that receiver. And I know that Betaflight, you can do servo outputs through like pin IO and stuff, but that's like a software solution and that requires extra headroom on the CPU. So I thought it'd be easier just to go directly to the receiver and skip passing through the flight controller. The throttle servo is going to plug into there. Next, um, one of the servo plugs is going over here to this servo operated relay. So this is a servo level line level relay for small currents. That relay over there is relatively small. This is the kill switch. So the drone has a magneto. Magneto has a spade connector on it, so this will plug into on the engine of the drone. Um, when I activate this, uh, servo signal on my radio, it's gonna go here into this servo switch, it'll close the relay, and that'll short out the magneto to ground. So you can see I have one wire running over to ground over here and the other wire will go to the magneto. And by shorting the magneto to ground, uh, the spark will stop and the engine will stop.