The experimental team's focus in studying the trebuchet was to learn what we could from the launch of a projectile from an actual trebuchet. Scott Teige's K'nex trebuchet was the inspiration for this approach, and we started out by seeing how much could be learned from his model. Unfortunately, as Scott's trebuchet was designed as a toy without specific experimental usage in mind, it was very difficult to learn much from it. Being constructed of plastic, there was a great deal of flucture in the design, especially in the lever arm. Also, the K'nex toys out of which it was constructed are brightly colored and made it very difficult to track on videotape. Thus, in order to learn more about the trebuchet, better apparatus had to be designed and built.
Even in a simplified model of the trebuchet, there are a lot of factors to consider, so we tried to narrow down what we were going to test as much as possible. Going in to the project, we thought that we would be able to feasibly vary the lengths of the throwing arm side and the counterweight side of the lever arm, the mass of the counterweight, and the angle of the release mechanism. In the end, it ended up taking too long getting the trebuchet to fire consistently before we could take data to allow us to vary all of these. We finally settled on varying the length of the throwing arm, with a constant sling length, release angle, counterweight arm length, and counterweight mass. We hoped to be able to find preferred ratios between various and components that would help us discover and optimized design, and we were able to do so with throwing arm lengths. It became very apparent in the testing process that the proportions had a lot more to do with a trebuchet's performance than the magnitude of the counterweight and size of the trebuchet. This was a sad point for several members of the team, when it became obvious that simply adding more mass to the counterweight would not eventually allow us to launch a projectile at the escape velocity!