Experimental UAV Glider

Following my 3D Printed Aircraft Competition, I decided to take initiative and explore a different design for a 3D printed glider. Researching a simpler design that arguable had better performance.

Despite its simplistic design, it boasted superior performance when in flight, and it gave me a viable platform to test and iterate the aircraft for improvements. Looking forward, I intend to further mature the design by improving aerodynamics, control surfaces, and integrating electromechanical components. My ultimate goal is to incorporate sensors and a flight control system allowing for semi-autonomus flight.

Prototypes

Prototype #1

The first built prototype was made early in the semester to test proof of concept. It used a basic foam airfoil and tail, attached with a rubber band, along with a simple boxy fuselage printed from PLA Aero. No power, controls, or electronics. The goal was to test the basic design and see how it glided when flown. If it flew straight, the concept had legs.

Prototype #2

The second prototype took this concept further by adding rudder elevator controls and propulsion. It included a thinner fuselage design, and a similar foam main airfoil, except with dihedra wing tips to provide rolling stability. The most important design choice was a movable main airfoil which allowed the CG to be adjusted my moving the airfoil, allowing us to fine tune the aircrafts stability.

After testing, the main design idea behind the aircraft worked as intended, however the control surfaces where two small, and the fuselage was deemed too thin and would often break on landing.

Ptototype #3

The next prototype included the same main airfoil, except with a redesigned fuselage that instead of being a uniform diameter the full length, it tapers and gets narrower towards the rudder, which allows the use of friction fittings, and the majority of the weight moved towards the front. The main section was printed from PLA Aero for light weight, and the nose printed from PLA Basic for durability during landings. The fuselage proved to be stronger and better suited.

Prototype #3.5

A big leap came for this fuselage as a second version was created but instead with a fully 3D printed main airfoil, rather than foam. The airfoil was printed from PLA Aero for light weight, and included internal wing spars printed from PAHT-CF for structural rigidity. The airfoil was overall slightly heavier than the foam version, however did not significantly affect performance.

Prototype #4

The latest prototype came about once the airframe shape and sizing have been verified with the only thing left to do is lose weight. For this I innovated and re-designed the fuselage to be printed from PAHT-CF at one wall in vase mode. This created a monocoque style structure where the skin itself carries the load, and due to the PAHT-CF improved strength to weight ratio, the fuselage weight dropped by %15, drastically improving performance, verified by flight tests.

Flight Controls

The aircraft uses a three-axis control system built around passive stability, and simplicity for light weight. This deliberate design reduces mechanical complexity and is designed to be as light weight as possible for enhanced performance. Pitch is controlled via a moving horizontal stabilizer, yaw is controlled by a single rudder, and roll stability is provided passively by 20 degree dihedral wingtips. All surfaces are actuated by SG90 servos connected to the pilot's RC transmitter. The main benefit of this design is it only requires two servos, as oppose to need 4 servos with ailerons. This decision further reduced weight on the aircraft that improved performance.

Pitch Control

Rather than a conventional elevator, where a portion of the trailing edge deflects, for this aircraft the entire 16 x 4in horizontal tail surfaces pivot along a central hinge. A similar design to supersonic fighter jets, and high performance gliders. This approach was primarily chosen due the simplistic design and the need to save weight, as integrating a hinge into the elevator would increase weight and complexity. An added benefit is a greater pitch authority per unit area than a conventional elevator.

A conventional elevator usually has a pitch effectiveness ratio around 0.45. Where 45% of the surface area contributes to the pitching moment. A stabilator, by contrast, has an effective ratio of 1.0, with the full surface generating a pitching moment at every deflection angle.

We can then calculate the horizontal tail volume coefficient as a measure for pitch stability and control authority.

V_H = (Sh × lh) / (Sw × c̄) = (64 in² × 22 in) / (576 in² × 8 in) = 0.306

This value falls in the typical glider range of 0.3-0.6, meaning the tail is of adequate size and stability without excess drag. Meanwhile a moment arm of 22 in from the center of lift to the stabilizer, gives sufficient leverage without a long tail boom.

Yaw Control

The rudder measures at 8.5 x 3 in, with a surface are of 25.5in^2. This then gives us a vertical tail volume coefficient of 0.0142.

V_V = (S_v × l_v) / (S_w × b) = (25.5 in² × 23.1 in) / (576 in² × 72 in) = 0.0142

This is slightly lower than a typical glider range, this reflects the tradeoff on a rudder only aircraft with no ailerons, where the rudder generates both yaw and role, through adverse yaw coupling with the dihedral.

Dihedral Wingtips

In the photos you may notice the outer 12 inches on each side of the wing are angled up at 20 degrees. This creates passive roll stability without ailerons or servos. This is the core aerodynamic innovation of the design as it eliminates the complexity of ailerons completely, relying on geometry.

This is because as the aircraft rolls or sideslips, the lower wing generates more lift than the upper wing due to its angle of attack increasing relative to the incoming airflow. This lift imbalance produces a restoring moment that naturally returns the aircraft to wing level with no control input required.

Effective dihedral angle (area-weighted) = 10° × (24 in / 72 in) = 3.33°

Cl_β ≈ -0.00228 × Γ_eff = -0.00228 × 3.33° = -0.0076 per degree of sideslip

This approach trades active roll control for passive stability, a valid design choice to reduce complexity and weight for better performance.

Servo Linkages

Servos where mounted right behind the main wing to bring the weight as close to the CG as possible. They were then linked to the control surfaces using metal push rods that ran through plastic tubes mounted to the top of the fuselage. A clevis was used at the end of the push rod to link it to a control horn, mounted to the control surfaces. This design was chosen as it was very simplistic and highly adjustable when triming the aircraft for flight.

Testing

Flight Testing

Flight testing was used extensively throughout the aircrafts development. Multiple flight tests where performed for each prototype. Tests where validated by creating a complex testing schedule and procedure. These included a pre-flight checklist, in-flight test maneuvers, and a post-flight review of data.

Cardboard Prototype Test #1

Cardboard Prototype Test #2

Proto #1 Servo Test

Test Flight #1 Prototype #1

Test Flight #2 Prototype #1

Test Flight #3 Prototype #2

Test Flight #5 Prototype #3 TAKEOFF

Test Flight #5 Ariel Tests

Test Flight #5 Landing

Test Flight #7 Prototype #3.5

Filament Impact Testing

An form of testing developed early in the project involved printing samples of each filament used and impact testing with the charpy impact tester. This was a guage on the materials strength, ductility, and impact resistance. Results in the left table reveal PAHT-CF and PLA Basic the ideal filament for impact resistance.

Strength Testing

When deciding to switch to a PAHT-CF fuselage, numerous considerations had to be made. Strength, impact resistance, and layer adhesion where to be tested to ensure this was the correct approach.

To test for strength and layer adhesion, a simple bending test was devised using test pieces braced on the end and a increasing load in the center. Both PLA aero and PAHT-CF were tested as they were the two candidates for the fuselage material. Both had different wall thickness so that both would remain the same weight.

The results of the test seen below reveal the PAHT-CF sample endured much more weight, and more importantly, it expereince deformation, rather than failure, seen in the third photo. The PLA Aero endured less weight and experienced a brittle failure. This test proved crucial as it explained my previous fuselages made from PLA Aero of ten failed on landing, due to the materials brittle nature.

Innovations

Vase Mode Prints

Vase mode, which is also known as "spiralize outer contour," is a 3D printing technique where a single continuous spiral wall is generated in the slicer that follows the outer wall from the base to the top, with no infill and no layer by layer seams. The technique allows a seamless, lightweight shell, with no weak points made by traditional layers, due the print head never stopping to print a new layer.

This printing technique proved very useful for aircraft structures, allowing the part to be as light as possible while preserving geometry. Therefore for this prototype, vase mode was used exclusively for the airfoil and fuselage, minimizing weight but still allowing structure. Lots of testing was needed to create a rigid structure, seen in the left image. Vase mode also provided the benefit of a smoother surface finish that improved aerodynamic efficiency across the surface.

The use of the carbon fiber reinforced nylon fuselage with its high stiffness to weight ratio, gave the fuselage a monocoque-style structure, Where the skin itself carries the structural load, with added corrugation for added stiffness. The result is an over lighter and more impact resistant fuselage rather than the multi-wall PLA-Aero approach used on competitors aircraft.

Ground Sensor

To explore low-altitude situational awareness, a ground sensor was integrated into the aircraft. This was in the form of a ultrasonic sensor mounted to the underside of the fuselage and hooked up to an Arduino microcontroller. This sensor continuously measures the distance from the ground and feeds the data to a segmented LED bar display to give the pilot a visual readout of distance from the ground. Additionally the system also includes a warning light an a buzzer to alert the pilot when near the ground. This systems proves particularly useful during glide and landing phase, where the pilot is given head-up awareness of ground proximity, without needed to visually gauge altitude. The system is still in active testing to further refine it, however it aims towards future developments with more sensors to allow for more semi-autonomous operations.

Bomb Release Mechanism

Additional add-ons to the glider include a payload release mechanism, which was designed and integrated under the wing. Capable of carrying eight 3D-printed munitions from wing-mounted brackets. With the release mechanism triggered directly from the radio controller, giving the pilot full control over drop timing.

With the current design as a functional demonstrator, the mechanism establishes the groundwork for future payload applications, such as aerial delivery or sensor deployment.

Page updated

Google Sites

Report abuse