How can one determine the proper placement for a pneumatic cylinder? Or figure out the proper or optimal geometry?
CAD?
Yes, or on paper.
Kinda like in the other thread, I once again recommend Adam Heard’s 973 Ramp Videos.
He has a great video on linkages that also shows how to model the motion of a piston using line segments.
Wonderful video! That is very similar to what we did this year, only I sort of like that method more. Using lines as cylinders is good.
Our 2015 can grabber (borne between State Finals and World Championships) started with a static mock-up using a variety of cylinders, pivot points and linkages until a there was a reasonable fit of the known physical constraints of reach and angles. This was followed by a powered mock-up using the chosen cylinder, some plywood, 2-by-4’s, conduit, an old ski pole and shop air set to 60 psi. Geometry was changed and tweaked (by drilling different pivot holes in the plywood, until video proved the fastest overall arm deployment by counting of video frames. Once we had a final design, we translated it into machined aluminum, and did very well with it. Total “design” time: a couple of hours.
Pneumatic arm deployment is different from motorized rotational movement. Motors go from stall to stall with little variation in force. With pneumatic cylinders acting on a rotating arm, however, it’s not always so simple: sometimes, you want cylinder geometry to act more normal to the arm at the start of the stroke, and sometimes at the end. There’s always a trade-off of final force and final speed that must be reckoned and evaluated.
There might be software that would have done the same thing, but part of the beauty of pneumatics is that it is so easy to mock-up, test and optimize.
There were faster can grabbers than ours, but our entire grabber mechanism was 2" wide and weighed 2 pounds. We could not have accomplished that with motors, gearboxes and linkages.