Quadrocopter 2013 Update

The last time I posted about my quadrocopter, I was about to have a ton of problems that would keep me from getting it working. That was still in 10th grade. That summer I mostly gave up on the project since I simply couldn’t get it to level properly.

At the end of senior year in high school in AP Physics I had the opportunity to do a physics project instead of taking the final. I had thought many times about revisiting my quadrocopter, but I never got around to it. Even though I didn’t touch it for the better part of 2 years, I was planning ways of making it work and learning new things in school and online.

After spending months and countless hundreds of hours trying to get it to work before, I got my quadrocopter working in a week. All I had to do was cushion the sensor boards from vibrations better and change a few lines of code. At first it didn’t work. Then I tried a vastly different PID tuning and suddenly it started flying reasonably well.

I had gotten so incredibly close to getting it to work years ago. There was just that last little push, along with a much clearer understanding of math, that I needed.

I made a powerpoint presentation for the class that I’m quite proud of. It explains the theoretical implementation behind quadrocopters, but not the practical physical part. I’ll upload a video of me presenting it soon. I also flew the quadrocopter for the class and it worked even better than when I had flown it at home.

Then I flew it more later that day. And I crashed it. I was trying to “bounce” it, by giving it a lot of throttle for a short period of time. Something happened in the software that made it go directly sideways at full power instead of up. It went directly into the side of a jacuzzi (not in the water luckily). At least one of the motors was completely ruined and all of the propellers were smashed and cracked. Luckily the other electronics didn’t seem to be damaged in any way to compromise their ability to work.

I decided to use this as an opportunity to upgrade the motors and propellers. Both of which were ridiculously cheap Chinese parts. Real motors should’ve cost at least $20, but these were only $5. I decided to go for $10 motors this time and real APC (not APC-style) propellers. The old propellers weren’t even APC-style. They were flimsy bendy pieces of plastic that I seriously doubt could effectively support the weight of the quadrocopter.

IMG_1137I’ve barely flown it since upgrading, mostly because I need more battery life and a better charger (the charger was also $5). It seems this motor/prop set is less responsive and uses more battery life. The motor rotors and propellers are much heavier and the propellers are larger in diameter. That equates to a much larger inertia. I think if I re-tune the PID then it will work just fine, and probably be more stable because of the nicer equipment. But right now it is not terribly stable.


The quadrocopter lives!

I finally have the quadrocopter approaching completion! I have the frame completed out of 3/4″ aluminum square tube and Lexan. Each motor is held on the frame with a clamp mechanism made from Lexan. Each propeller is mounted to it’s motor with a prop saver. The electronics are all mounted and have mostly been all soldered together. Each of the ESCs (electronic speed controllers) is getting power from a circle of 10 gauge wire in the center. The Arduino is mounted in the center on stand-offs. Each speed controller is connected to the Arduino through a custom made connector. The battery (2200mAh 20-25C LiPo) is mounted to the bottom with Velcro. The battery connects to the power distribution system through an XT60 connector. The Arduino will connect to power through the on-board power connector (right now it’s running off of USB power).

Here’s a picture:

Right now I really just have to work on the code to make it run. So far I have code that gets an attitude (angle) measurement from the IMU. I have PID code from another project so I just have to implement that into this project. After that I just have to tune the PID and it’ll be working! The code is really the interesting part of this project. I’ll post an in-depth post on exactly how my code works and how you can change it to make it work for you once its finished. Also, I’ll post all the code, links, design drawings, and anything else that went into building this.

In the meantime, here’s a video of all four motors powering up a bit from computer->Arduino control: http://www.flickr.com/photos/49325300@N07/5698514204/

Quadrocopter Progress

Earlier this year in December I saw a bunch of videos of this new type of RC flying machine called a Quadrocopter. After some research I decided that I could build one for significantly less than the price of a commercial one. I was determined to build one of these. After many many more hours of research I found the code to be the most difficult part because of all the math that I haven’t learned yet. This project is finally starting to come together and I have most of the components needed to build it. This post will document my progress in building my very own quadrocopter.

Here is what I have so far:

I have 4 high power brushless outrunner motors with special Gaui quadrocopter propellers attached. I am using prop savers to attach the props. In the middle I have a 2200mAh 3S LiPo battery and an Arduino with a Sparkfun Razor 6DOF IMU. Also, not pictured, I have 4 18A ESCs to control the motors. The motors, ESCs, and battery + charger cost me about $80 with shipping directly from China. I already had the Arduino, but that normally costs $30. The IMU cost $60, which may seem like a lot, but it should give me an exact position reading of roll, pitch, and yaw.

Here’s a video of me testing one of the motors with a ccw rotation propeller attached: http://www.flickr.com/photos/49325300@N07/5548752629/in/photostream/

These motors are extremely powerful. This single motor was able to lift the heavy vise enough to let it move from vibrations. And that was only at half speed! Just think what four of these will be like at full speed! This will probably be able to lift 5 pounds (made up statistic alert!) extra!