Designing and building a rubber band powered plane for under £5

 My first thought when thinking of an efficient glider is the balsa wood planes with structural techniques similar to real planes, with longerons and airfoil profiles throughout the wings. I started here, 3d printing some basic airfoils to act as struts to get a general shape of a wing through wrapping it in fabric or plastic sheeting.

I then squeezed some 6mm dowels from a nearby art store to form wingspars throughout the wing, which would take the bending and twisting loads of holding the plane's weight in the air. These were definitely overkill, as during testing, these dowels could support up to 250g of weight before breaking, and the plane wasn't projected to breach 100g. This means each dowel only needed to support 50g, so thinner struts could be used here.

A layer of plastic, taken by cutting up plastic wallets usually used in school folders, was used to form the skin of the wing. This iteration of the wing weighed around 45g for a single 30cm wing.

I then modelled a wing box to connect both wings to the fuselage, with holes of 6mm so that the dowels would press-fit tightly into the 3d printed piece without needing glue, meaning the whole plane would be hotswappable and adjustable for centre of gravity shifts.

This same wingbox was used as a spar for the elevators, though in hindsight, a different design could be used since there is less wing loading for the elevator wings, so some grams could be shaved off here. The better alternative that I would use now is to simply make the tail section out of foam, as the airfoil shape is not necessary - in fact, most planes have a neutral airfoil profile for the elevators anyway. This change alone could save 20 grams, and since this weight is at the very back of the plane, it would reduce the extra weight needed at the front to maintain a valid centre of gravity. 

And finally, a vertical stabiliser was printed:

 Leaving us with a first design that looked like this: 

The reason the nose of the plane is extended so far is to make sure the centre of gravity of the plane is in the correct position, near the front of the wings. Having a long nose allows less mass to be used for the centre of gravity to be in the correct place on the plane.

 This plane weighed, in total, around 140g. I wanted to bring this weight down even more, so I started on a new design immediately. One issue I had was needing to have thick airfoil profiles to stop them slipping and rotating on the single wing-spar dowel. To fix this, I used 2 barbecue skewers to hold the airfoils in place, meaning that the 3d printed airfoils themselves could be much thinner since they couldn't rotate in place with two points of attachment. These wings were 50cm in length and 29g in weight, a comparative 159% improvement in weight-to-length ratio.

I could finally implement my own design of the rubber band power unit. This piece was iterated over 20 times, each time improving its efficiency. The difficult aspect of designing this piece was making it strong enough to not break when the plane touched down, but also reducing the friction on the mating surfaces between the shaft and the mount.

The rubber bands would pull the plastic pieces very hard together when tensioned, meaning that at the start of the flight, when energy was needed most, the friction would prevent the propeller from getting the energy it needed. The final CAD model, looks like this:

 

 

I then designed a 3d printable propulsion system for the rubber band, where the adjustable nature of the plane made it possible to keep the centre of gravity in the correct spot, despite the propellor and rubber band arrangement adding weight to the front of the glider.

 

Finally, I remade the wings, using packaging tape on the leading edge to ensure that they didn't buckle under higher speeds, as was evident in the first iterations. The old wing had a tendency to have the plastic shift and crinkle on the leading edge if it was thrown too fast, as the air hitting it would form a 'pocket' running along the length of the wing that would be disastrous for aerodynamic efficiency.

This was the final result. It could easily fly the length of my sports hall in school, which is around 40 metres in length, only being stopped by the opposing wall. The plane came in at 100g, including the propeller and bands. The only improvement I would add to this design is adjustable rudder and ailerons, so that the plane could fly in a circle and not have its flight time limited to the space available. The project was an overall success, only taking around 2 weeks to complete.