Why Fly Subscale

Learning through the build/fly/test cycle

Peter Kalogiannis
July 29, 2020
·
4 min read

We’ve all heard the adage that “The longest journey starts with a single step.” But what we sometimes forget is that the next step is just as important as the first, and so is the one after that, and the next one, and the remaining thousands until we reach our goal.

In an aircraft development program, especially of a unique type like we’re developing at Airflow.aero there are thousands of individual steps that we have to complete to take our concept from an idea to a certified, flying product. Each step requires us to gather data, learn lessons, and apply those lessons to our design.

Today’s aircraft developers have access to incredibly complex computational tools to help in aircraft design, and these tools make our jobs much easier. But one of the best tools we have predates modern CFD: the subscale aircraft model. By building, flying, learning, tweaking and flying again at the very beginning of our journey, we are able to learn about the unique aircraft we are working to develop.

Subscale models have been used for aircraft development since, well aircraft development has existed. NASA has created an excellent publication describing the vast uses of subscale aircraft in aircraft development. They have been used for a wide variety of purposes, for everything from use in a wind tunnel to observe complex aerodynamic phenomenon that can’t be easily modeled, to exploring new propulsion technologies, to developing new kinds of flight control techniques.

So what are we doing with ours? All of these! The state of model technology is advanced enough that we can build an aircraft with distributed electric propulsion, implement our control algorithms, and start gathering data in a matter of weeks, instead of the years it would take to do so with a full-scale aircraft. Utilizing the plethora of off-the-shelf airframes, propulsion systems, and flight controllers available today Geoff, Thomas, and Don have been flying and gathering data to analyze and feed into our models.

Subscale aircraft, depending on their particular purpose, can have significantly varying levels of complexity. Our first aircraft is relatively simple with the purpose of starting the data collection as quickly as possible. We’re using a Cessna 152 1:5 scale airframe that we’ve modified by removing the standard single motor and replacing it with 8 distributed electric motors mounted on the wings using nacelles that we created and 3d-printed and some custom wiring and power distribution hardware we built.

The control system is designed to support the new characteristics of our aircraft and allows us to start experimenting with approach profiles that take maximum advantage of the benefits of Distributed Electric Propulsion (DEP) to allow for high lift generation at slow airspeeds, a key characteristic that will allow us to fly from extremely short runways.

For the control system we’ve started with a Pixhawk / ArduPilot controller, which has provided us a base platform on which to easily build our custom control system utilizing Froude Scaling Laws and work that NASA has done to define scaling factors to make the subscale behave like a full-size aircraft. Because of this, we can develop and test control laws on the subscale that can be taken and directly applied to our full-scale aircraft.

Geoff preflighting the aircraft

Our first flights were all hand-flown by our accomplished R/C pilot Geoff, who led the subscale development program for Vahana. These flights explored the flying characteristics of modified aircraft and verified that our initial calculations on things like weight and center of gravity, cant angle of the motors, and control authority were correct enough to give us a controllable aircraft. From there, our flights progressed to operating with more and more autonomy, as Don expanded the flight control system to include things like reverse thrust and configurable approach profiles. We’ll take this further and look at things like failure mitigations and the possibility of more advanced flight control techniques utilizing the DEP system.

While we’re learning with this version of the subscale aircraft, we’re planning and designing our next iterations which will more closely resemble the configuration, mass distribution, thrust-to-weight, and other characteristics of the full-scale aircraft we intend to build. This will allow us to explore other variables, that will ultimately affect how we design and operate our full-scale aircraft.

While we see the subscale aircraft as one tool in our bag, one that we will use in combination with computer modeling, wind tunnel testing, systems modeling and ground, and flight testing, we have found in the past that flying early and often in the development program is an incredible accelerator in the aircraft development process. We are proud of what we’ve been able to accomplish so far and excited by what we have coming up ahead for our subscale aircraft.

Take a look at our video of the subscale aircraft in action!