Make It Glide

Introduction

Hello!

This hand-launched glider was designed to take advantage of the unique capabilities 3d printing has to offer when it comes to aerodynamic design. This model was created as an entry in the Prusa “Make It Fly” design contest. The guidelines of the contest were broad. My personal design goals for this project were:

• Design a glider that could be printed all at once on my printer (build volume of 220x220x250mm)
• Easy to assemble (no gluing)
• Required minimal parts outside those printed
• No complicated print settings (previous designs of mine required extensive use of support blockers to modify things like infill and wall counts)
• No supports

UPDATE 3/15/23

The forward fuselage and nose cone were the last items designed for this project. Previous test flights were made with a balsa stick with a AA battery taped to it for balance. First flight with the nosecone+FFUS was today: https://giphy.com/gifs/z99bBJSQ7GdbpQd7LE . Other than the landing that I'll get to below, it flew great. Wish I had gone to a park and really launched it hard, but that was fantastic for backyard flying. 

The CFUS snapped apart in that landing. It is now the weak link in this design. I would suggest printing with a higher infill (10%) until I can redesign to be more sturdy. On a positive note, the CFUS was the only component damaged. The wing and tails will not have to be remade.

*Also, the wrong FFUS file was uploaded before, that has been fixed.

General Design

General stats of the design of the aircraft:

• Wingspan: 480 mm
• Root chord: 61 mm
• Wing area: 231 cm^2
• Aspect Ratio: 10.1
• E216 airfoil
  • High lift to drag ratio, with sufficient thickness for structural strength
  • Also has a thicker trailing edge compared to similar airfoils
• Elliptical planform
  • Results in an Oswald Efficiency Factor of 1.0
• Elliptical “dihedral”, wingtip is 45mm above the root
  • Provides roll stability in a unique way
• Elliptical sweep, wingtip is 78mm aft of root chord leading edge
  • Purely done to look cool. Would be useful if flying transonic

Recommended Print Settings

• All parts are already oriented in intended print direction
• 1 wall count
• 2 top/bottom layers
• 2% infill, cubic. As low as you can get away with to provide twisting rigidity, while keeping weight low.
• 10% infill for CFUS, while awaiting redesign 
• Brim
• Layer height 0.2mm
• No supports
• 50 mm/s print speed or lower, especially as you get up to the wingtips
• Lightweight PLA (LW-PLA) 

I would also recommend Cura 4.13.1, as that was the slicer I used when designing certain aspects such as the wing spar cutouts. With the above settings, Cura will add extra material as intended to form an I-beam spar in the wing to provide sufficient strength. Loading into Prusa Slicer 2.5.0 did not give the results I was expecting/intending. That could be due to lack of experience on my part though.

I have included an image of the wing in my slicer, showing the intended result.

Assembly Instructions

1. Join the left and right wings together using the wing keys (print 2, both identical) inserted into the slots in the wings
2. Slide the aft fuselage (AFUS) onto the the notch/groove connector of the center fuselage (CFUS)
3. When fully inserted, rotate the AFUS so the vertical tails point up to lock in place
4. Insert wing into CFUS, using the notch to center in place
5. Secure wing with a rubberband looped over the wing
6. Slide the forward fuselage (FFUS) onto the notch/groove connector and rotate to lock in place
7. The nosecone press-fits onto the front of the FFUS, but wait to do that until you determine how much ballast is needed in the next section.

Balancing

For stable flight, all aircraft need their center of mass to be located in a specified region. For this aircraft, that would be about the middle of the wing. Too far forward, and the plane will nosedive. Too far aft, and the plane will not be aerodynamically stable, and will crash uncontrollably. Too far forward for this design would be balancing near the leading edge of the wing. Too far aft would be balancing near the trailing edge of the wing.

To balance, add pocket change to the front of the FFUS. The compartment hole is large enough to accommodate coins the size of an American nickel. Add enough so that the aircraft balances near the center of the wing. It might be possible to avoid the use of coins as ballast altogether if the nose cone is printed at 100% infill.

Flying

Fly in an open, outdoor space. Preferably with soft grass like a field or park. Hold at the fuselage behind the wing, and throw hard at a slightly upward angle. With minimal infill and a single layer wall thickness, this model is not very durable when it comes to crashes. Expect the fuselage to get beat up. The wing will usually be saved with its non-rigid attachment.

The final weight of the model is going to be a large determining factor in how it flies. The heavier the aircraft, the faster it has to fly achieve enough lift to hold itself in the air. That also means the faster it hits the ground if it crashes. See the below table for the calculated stall speeds (slowest the plane can controllably fly) of various total aircraft weights.

I would highly recommend printing a lighter weight PLA. There are some versions out there that will foam as they are printing, giving greater than 50% weight savings. Foaming LW-PLA is pretty difficult to work with, I still haven't gotten a full model to successfully print with it.

There are some pre-foamed LW-PLA's on the market that while having less of a weight savings (~20%), it is much easier to work with.

Future Redesigns

I really want to cut down on the stall speed for this plane. I plan to extend the wingspan by adding a 3rd section of wing. This will both increase the wing area and the efficiency by increasing the aspect ratio. Both of those will result in slower speeds necessary to fly.

Another goal is to get the wing design to be more friendly with PrusaSlicer. That may be tweaking the slicer settings, or tweaking the size and placement of the wing spar and other things to get the intended result in more slicers.