Success! After weeks of redesign (see this post for redesign info) and unforeseen challenges, we have achieved our first flights over the weekend! Over the past week, a core team conducted nightly builds with morning and evening flight windows to test new iterations and fine tune both test setups. Our many accomplishments this weekend included vertical takeoff, hover, powered forward flight, maneuverability testing, aerobatic maneuvers to push the plane’s limits, and stall speed testing.
One goal of this weekend was to prove our hover concept. This meant we had to prove that our motors have enough power to lift the plane and that our autopilot is able to stabilize the plane. We outfitted our built-up model with our three Electric Ducted Fans (EDFs) and conducted a series of tests with the plane on strings. These tests showed that the EDFs could provide enough lift; however the strings interfered with the autopilot’s stabilization too much, so we moved the plane off the strings. We conducted a series of tests that showed the plane had enough lift to hover and that the autopilot could keep the plane stable in roll and pitch.
The following video shows a compilation of our takeoff and hover tests. One item to note is that this version of the plane does not have the hover yaw control surfaces (vanes on the back EDF) installed. This means that we had assist the plane to keep it from yawing too much. If you watch closely you’ll see that the people assisting the plane are not holding it up and are instead giving it slight nudges to keep it in the testing space.
We’re very proud and excited to have obtained successful hover, but we must keep pressing forward. The hover tests have shown that we have very little margin in our pitch control. We are currently sizing a slightly more powerful motor for the back EDF. Last, we need to tune our autopilot to the plane and test position keeping while hovering.
A CNC cut foam model was used as an easily reconfigurable test bed. This method alows for quick build, easy repairs, and flexibility of component placement. Even though the plane was designed to be hand-flown, forward flight was assisted with standard RC helicopter gyros for general stability assistance for the new vehicle design. Our final landing gear consisted of a tricycle-configuration with carbon fiber main gear attached with nylon bolts to Birch-plywood support plates embedded in the foam wings and a Birch nose box for the forward steerable gear. This plane was weighed down to 7.6 pounds to attain proper CG location and confirm flight capabilities.
We have created a video compilation of some of our forward flight capabilities! As you can see from this montage, the aircraft is quite maneuverable and could easily perform basic aerobatics, displaying a comfortable performance margin in forward flight.
Further tests to be conducted with this airframe include:
- Practice way-point control with autopilot
- Weigh aircraft down to discover max takeoff weight
- Test without gyros for full performance characterization
- Test performance with lift fan holes to determine necessity of fan covers
Our next step for forward flight will be to integrate forward flight capabilities into the hover vehicle and practice transition flight.
Lessons Learned from this Integration Test
- Strong landing gear is a must for testing purposes.
- CG location is important
- Structural fuses are great but need to be strong enough to work
- The more re-configurable your prototype, the better.