Myself and three other Mechanical Engineering students are in our last year of college and we wanted to go out with a bang.... So we though, what would be fun to build as a senior project?
Well, we decided unanimously on building a heavy lift gas powered quadcopter. It will have a payload of 50lbs, variable pitch to allow quick maneuvers and two 12.5hp two-stroke hobby motors running parallel. Each rotor head will have four 435mm rotor blades making this UAV one big quadcopter measuring roughly 40" in length. The four of us are really excited to finish and watch it fly. If you are interested check out the links below.
Kickstarter:
web page:
Facebook page:
Comments
No only should it work well (Yaw included) it shouldn't be much trouble to do with the existing code base!
I look forward to seeing this progress :)
@UAV_Enthusiast: First, thanks for clearing up the terminology. We initially started by looking at aitplane props which are all rated with pitches since they are constant, so sometime I accidentally use the wrong term. Second, while changing the AOA would induce a speed change due to drag, we are going to maintain a constant RPM using PID control loops on each engine. The change in drag (and therefore change in torque exerted on blades) will allow for yaw control, not the speed of the blades.
I want to mention also, that the rendering attached is a somewhat old concept we were working with. There have been some considerable changes since, and we are working on getting an updated design and rendering. We'll be sure to post it as soon as it's completed.
@Chase: Yeah, one of our team mates brought up that point. I'm not sure we will be able to incorporate it since a) as you said, we aren't sure how effective it would be, b) we probably won;t have enough time to incorporate the programming required (I don't think this is a capability of the ardupilot, is it?), and c) I am not sure how we would test it.... If this project ever become more than just a senior design project, then it is definitely something worth exploring further.
Ah valid point...changing the inertia of the blades by changing throttle on the engine probably wouldnt occur facst enough for effective yaw control. Eh just stick a tail boom out there and call it complete :)
Here are a few interesting ideas:
First: I think they are using pitch interchangabley with the term collective. Collective is what changes the AOA in a helicopter blades "collectively" as it rotates around thus producing changes in lift. This is sometimes referred to as collective pitch control. Gabe and other engineers be careful on your terminology.
Yes changing this on opposite pairs of rotors will induce drag and cause the rotors to slow and produce a yawing motion. Nitro/gas helicopters compensate this effect by coordinating the collective with engine control as a tuned RC helicopter should maintain RPM during flight.
But here is an interesting thought....what if you linked each motor on the diagonally opposing rotors (assuming X) or opposing rotor (assuming +). Thus, you could directly change the RPM of the motors by changing the throttle. It would be an interesting control problem but still would effectively create the yawing moment and not have to change the collective on the rotors. Of course the problem in all this is what dynamics are you creating on your system.
You'll also have the ability to autorotate in the, hopefully unlikely, event of a 2 engine failure. Probably won't help much, given the low mass of the props, but it's something.
There are two main reasons behind the two engine concept.
First, HLQ will autonomously dock to its payload which will be immediately under the center of the frame, and would block an engine from being centrally located for balance. We had to move the engine to the side to accommodate the payload. This, however, puts HLQ out of balance, so we halved the engine size and designed it to use two.
Second, we were a bit worried about engine stall. We did not want HLQ, which could potentially weight 100lbs when fully loaded, simply falling out of the sky if an engine stalls or otherwise stops for some other reason. With a second engine, if one fails, we can detect it, possibly drop payload, and maintain enough control to land, though it may be a hard landing. The engines will be connected to the central drive through one way bearings, so if one engine fails, it will not induce additional load to the other bearing. It also allows props to slow themselves down through drag, rather then have a reverse driving force on the blades exerted by the engines slowing down, which we think will cause unneeded wear of the drive train.
@Luke: The pitch will indeed affect yaw. When the two CW blades are low-pitch and the two CCW blades are high pitch, the CCW blades will be putting more energy into the air, resulting in a yaw moment. I'd need to actually break out MATLAB and run some numbers to figure out real yaw authority, but I think you'd be surprised at how effective this strategy is.
Several people have done large-scale, centrally-driven (all rotors the same speed) multi-copters. I've seen a few quads and a few trikes - particularly successful is this T-copter:
https://www.youtube.com/watch?v=VXR-I4InEs4
http://www.rcgroups.com/forums/showpost.php?p=21808835&postcoun...
Obviously, as a tricopter, it has no issues with yaw authority...
I get the variable pitch side - as weight goes up, required motor size goes up, and as motor size goes up, motor inertia goes up, and they can't change RPM fast enough to maintain control. Complexity goes up exponentially with the # of motors, past 8 motors, you start to hit diminishing returns... Variable pitch lets you use as big a motor as you like and still have fast response.
Ditto for gas - if you're going for a BIG 'copter, batteries start to become a hindrance, and again, with variable pitch, you don't need super-fine motor control.
However, I don't understand the twin-motor setup. Why not just a single larger motor, even a multi-cylinder one? synchronizing multiple gas engines sucks, and adds drivetrain complexity to an already complex setup.
Hi folks,
My name is Nick Conover, and I am the team lead for the Incredible HLQ project. Gabe asked me to hop in here and help him field some questions. Neither of us thought there would be quite this much immediate response to the project. I am glad there is though!
As I understand it, the yaw motion of a quad, whether it be electric or gas is ultimately cause by the sum of the torques for each motor not equaling zero. In a electric quad, if you slow down two motors moving in the same direction, you also reduce the torque required to keep it at that speed. Assuming the other two oppositely spinning motors stay at the same rotational speed, an inequality in torques will cause a yaw motion.
In a variable pitch version, the same principals apply. In this case however, we do not change the rotational speed, but rather change the angle of attack of the rotor blades. By changing the angle of attack, you also change the surface area of the blade in the direction they are traveling. This changes the drag of the blade as it travels through the air which also reduces the torques required to keep it spinning at a constant rate. This, similarly to the electric fixed pitch version, also creates an inequality of torques compared to the blades spinning in the opposite direction, and yaw movement is achieved.
This is, at least, our theory on how it will work...
-
2
-
3
-
4
-
5
-
6
of 6 Next