Wing Crossbar Mount

The 2024–25 AIAA Design/Build/Fly rules asked for an aircraft that could fly a delivery lap, carry external fuel tanks, and release a test vehicle in flight, with a ground mission timing how fast the whole configuration could be assembled. Columbia's answer was Carrier Pigeon: a single tractor motor, a constant-chord high wing, a conventional tail, and tail-dragger gear.

The team chose to build no central fuselage at all. A 1.07 in carbon fiber rod runs the length of the aircraft and a set of printed parts clamp to it (motor mount, battery mount, electronics compartment, empennage adapter, payload assembly) each doing a job a fuselage would otherwise do. That decision saves weight and manufacturing time, and it concentrates responsibility: the wing mount becomes the one part carrying the entire wing into the airframe.

The mount has to receive both wing spars and hold them in alignment, transfer the lift and the bending moment they carry into a round carbon rod, survive the impact of every landing on a tail-dragger, and still come apart by hand (the ground mission is scored on how quickly the aircraft can be configured, so the wings have to go on and off fast).

It also has to be light, and it has to be small. Weight added here is weight taken away from the fuel payload that Mission 2 scores directly, so mass and stiffness were traded against each other on every iteration. Keeping the envelope as small as possible was the other constant target, both to leave room for adjustment along the rod and to keep the printed tolerances tight enough that the spars and the clamp fit the way the drawing says they should.

The part splits into a top section and a bottom section. Where they meet there is a square cutout that slides onto the central rod, and the two halves pull together on 8-32 nuts and bolts. Clamping force is what holds the mount in position along the rod, so nothing has to be drilled through the structural member, and the whole assembly can still be slid fore and aft to trim the center of gravity.

The wings enter from either side. Each wing is built around a 1 in square carbon fiber spar, and the square profile is what keeps the wing from rotating in its socket. The spars press-fit into the mount and are locked with cotter pins and 4-40 hardware, which is a joint a ground crew member can make in seconds.

Between the spar sockets and the clamp faces sits the geometry that does the real work: a trussed web carrying wing load down and outward into the clamp. Material was left along the paths that see load and the volume between them was hollowed out, which is where the weight came from.

The load case came from the aircraft, not from a rule of thumb. In Mission 2 configuration Carrier Pigeon weighs roughly 10.7 lb, and past flight results put the maximum acceleration it sees at about 6 g, which gives 64.2 lbf applied along the direction of gravity.

Printed PLA is anisotropic and FDM parts are genuinely difficult to analyze, so rather than using bulk PLA properties the simulation used simplified values derived from the actual infill density, infill geometry, and print orientation, cross checked against the filament datasheet. That uncertainty is also why the target factor of safety was set high.

The two halves were analyzed separately because they fail differently. The bottom section peaked at 0.985 ksi against a 3.771 ksi yield strength, a factor of safety above 3 on the von Mises criterion. For the top section the bottom face was fixed and 66 lbf was applied upward at the top inner face of the spar cutout, which captures the moment the wings put into the joint rather than just the vertical force; that returned a factor of safety of about 4.

The mount was load tested physically by supporting the whole aircraft from its wingtips and hanging weight at the center of gravity, which puts the wing bending moment through the joint the same way flight does. An earlier iteration failed there: at 13.38 lb of applied load, the mount broke.

It broke at the interface between the wings and the mount, which is exactly the region the top section analysis had been probing. The redesign thickened that interface, and the FEA on the revised part returned a factor of safety of 3.94. The rebuilt mount was put back on the rig and held.

That sequence is the useful part of the project. The analysis said where to look, the physical test said the part was not good enough yet, and the second analysis confirmed the fix before the aircraft flew with it.