STEP Reflection: The Show Must Go On!

The purpose of my STEP project was to utilize the tools and concepts I am currently learning in the Aerospace Engineering major at The Ohio State University as well as outside experiences to construct a 6-motor video drone. Next, I would take various photographs and video footage of the city of Pittsburgh for use in a media presentation about my hometown city.

STEP_graphic

The project first began with education. In order to design and build a 6-motor video drone, I needed to learn more about how drones work, design practices, construction techniques, recommended materials and parts, and troubleshooting. After performing roughly a month of research, I began designing my drone based on a “plug-n-play” strategy. In this method, the designer outlines the goals and constraints of the project in order to choose components to be combined to create the finished product. I knew how much overall thrust I needed, my weight and maneuverability constraints, my cost constraint, my time constraint, and my video and picture quality goals. I used all of this information to decide which components to purchase for the hexacopter. With so many options on the market, this required me to sacrifice a large amount of my project time in order to make the best decisions possible.

me_and_copter

Picture of the GoPro Hero 3+ mounted on the Zenmuse 3D Gimbal

Picture of the GoPro Hero 3+ mounted on the Zenmuse 3D Gimbal

Throughout the construction phase, I performed several different tests on components to make sure they were performing as expected. I tested the electrical system, propulsion system, flight control system, and the video capture and transmit system, as well as several individual parts. During testing most components performed optimally to expectation. However I discovered that I was having a much greater voltage drop than expected in the system while running the motors above half power capacity. This was a major concern because the aircraft required the motors to be at a power capacity >65% in order to hover, and between 55-80% for maneuvers.

Aerial picture of my backyard in Murrysville, PA

Aerial picture of my backyard in Murrysville, PA

I did the math for my first build, and I should have had about 18 minutes of hover time, 11 minutes of flight time, and more importantly 14-15 minutes of mixed flight time. This would have been more than enough time per flight to accomplish my project goals. However this was not accounting for the major voltage drop above 50% power. After initial flight testing, I discovered that I only had 10 minutes of hover time, 2 minutes of flight time, and roughly 4 minutes of mixed flight time. After extensive trouble shooting, I learned that in order to successfully get my flight time up, I would need to make major design changes, changing more than half of the aircraft. Before going ahead and spending significantly more money to essentially construct another hexacopter, I corresponded with other hobbyists facing similar issues as well as professionals in the field of aerial drone photography.

My Aunt and Uncle's front lawn in Pensacola, Florida. Notice the shadow of the copter.

My Aunt and Uncle’s front lawn in Pensacola, Florida. Notice the shadow of the copter.

After encountering various struggles and troubleshooting problems with the hexacopter, I’ve come to the realization that this project truly has been a transformation. Time and time again I have been faced with adversity and issues without easy fixes. Throughout the project as a whole, I have learned that troubleshooting can be a very long process. Almost never did my first solution to any problem become my final solution. In most cases it took anywhere from 3 to 7 different possible solutions, built and tested, before I found the best final solution. From this, I learned how to more effectively troubleshoot problems.

14 in. carbon fiber propeller mounted on a Turnigy Multistar motor.

14 in. carbon fiber propeller mounted on a Turnigy Multistar motor.

I also learned first hand about the cost of failure. As the project progressed and I ran into more unexpected issues, the amount of costs began to go up exponentially: Purchasing incorrect parts, purchasing replacement parts, time to construct or reconstruct, etc. The more mistakes I made or issues I ran into, the more frustrating the project became. Each time I would solve a problem, I would celebrate the fact that I no longer needed to spend more time or money on that issue, only to discover that I now had 2 more problems requiring my attention!

It is also important to consider safety when designing, constructing, testing, and maintaining a video drone. A hexacopter is a complex machine consisting of several different systems made up of hundreds of different working components. If any one component does not behave properly, the encompassing system will not behave as expected, causing the behavior of the hexacopter to be unreliable. Because erratic drone behavior is so dangerous, it is absolutely critical that before each flight, each system of the hexacopter has been tested and will behave reliably. Negligence on the part of the pilot could result in serious injury or even death! I can’t even count the amount of times a systems check failed before a test flight, forcing me to cancel the test flight to figure out what was wrong.

One time I took my hexacopter down to Florida to show my extended family and do a small demonstration. Before the demonstration I made sure all systems were operating properly and that the hexacopter would behave as expected. During the test flight, all systems were normal and the hexacopter flew very smoothly. I was satisfied that it would perform well and that I was ready to show my family what it could do. While flying during the demonstration, the hexacopter suddenly stopped responding to my commands on the transmitter. The carbon fiber propellers are 14 inches in diameter, and honed to a sharp outer edge spinning at over 400 RPM. This meant my hexacopter had been transformed from a well-functioning, predictable video tool, into a 12 pound deadly, uncontrollable mass of metal flying directly towards my family.

A still captured from the crash described above.

A still captured from the crash described above.

Luckily, I regained control of the vehicle and executed a crash landing before it reached my family. No one got hurt (except the hexacopter) but the experience sticks in my mind as a reminder: As an aerospace engineer, my actions and decisions could have serious negative consequences, including the unwarranted death of innocent people. Every decision we make must be very carefully thought out and analyzed before action is taken. We must do the best we can to keep others safe, including preparing for the unexpected. It’s not enough to hope nothing will go wrong; as engineers it is our job to go extra lengths to be certain that nothing will go wrong and prevent harm to others in the cases when we are wrong.

My hexacopter next to my puppy Toby, a jack russel terrier/chihuahua mix.

My hexacopter next to my puppy Toby, a jack russel terrier/chihuahua mix.

At this moment in time, the project is incomplete. The surplus of learning I have experienced throughout the course of this endeavor has been paramount to my success as an engineer and a problem solver in the modern world. I may have learned a lot, but I’m not finished learning yet. Even now, after rebuilding the hexacopter with different components and the expert knowledge from professionals and other experienced hobbyists, I still have been unable to get my mixed flight time over 8 minutes. In order to safely fly my hexacopter around the large crowds of people I will find in Pittsburgh, I need to have approximately 15 minutes or more of mixed flight time. For the time being, the project is on hold, but not forgotten! I will successfully get my mixed flight time over 15 minutes, and I will successfully make a presentation of Pittsburgh using media gathered with my hexacopter. Adversity has struck at me, but to adversity I say: “The show must go on!”

Leave a Reply

Your email address will not be published. Required fields are marked *