Here is a good resource about trebuchets, how various components work and a few styles:
There are many trebuchet styles from which to choose. Below is the short list with a few pros and cons of each.
ASOK – Arm slides over CAM (I don’t know why they use a K for cam)
This design is pretty efficient and shouldn’t be too complex to build. Few moving parts and we could play around with an interchangeable cam. Bonus, no rotating pin joints, which I think will be hard to print.
Here is a big one that throws over 1000′ (330 meters)
Cardan Trebuchet
This is the design from the initial proposal. I think it is the simplest to build and operate. I have never seen a full size one and I don’t know what sort of efficiency we could get out of it. Maybe that makes it better for us to build.
Fiffer trebuchet (Tom Pfeifer might like this one)
Named after the team who developed it, First in Fright
I think this one can be tuned up to be pretty efficient. It has a few more moving parts and rotating pin joints, which will make a small scale machine more difficult to operate.
Here is First in Fright throwing 2200 ft (670m)
Merlin Trebuchet (rhymes with Hurlin)
This machine has few moving parts and I think we could use a CAM instead of spokes. I think tuning it in will be the most difficult of all the machines
Here is the big one throwing 2900ft (880m)
Or we could go with the old fashion suspended counterweight machine. A classic, recognizable machine. Downside, not very efficient.
Although, as Yankee Seige demonstrates, with enough counterweight you can over come any inefficiency. With 8 tons of CW (8,000 kilos) they can get a pumpkin to go 2900ft also.
Decision
In the end we decided to go with the ASOK trebuchet. This style’s low number of moving parts coupled with its proven efficiency made it the best candidate in our eyes. We feel the flexibility in design that 3D printing enables will be very convenient for designing a highly efficient cam profile. This design also doesn’t need a wide gap in the middle for a counterweight box. We are hoping that without a wide gap between the legs we will be able to use a 3D printed axle instead of needing to find a steel axle. This keeps us inline with our goal of printing as many parts as possible.