The problem is how to assemble the piston, the ball and the circular CAM ring ?
The working seems so simple : u rotate the CAM ring anti-CW to a degree, it pushes the ball and the ball pushes the plunger/piston...punching done...
Can someone help me out?
I will have a look later today on the assembly.
As for Mechanism, cool feature but I am no expert at it. It is extra work that I am not sure is useful in most of my efforts but I can see occasion to use it, so I do need to learn it better. It is highly effective if motion is a key element in your day to day work.
I do remember taking some helpful tutorials on this with the e-learning module. For some reason, it is not part of my current e-learning license. I have found some helpful You-Tube videos on it however. I am still having trouble just setting up the motor 
I try to stay within the bound of the basic package but I can see how it can quickly fall into shortcomings with plastics design and maybe even some highly dynamic assembly models. Then again, I am old world... I can develop an entire system with non-parametric wireframe modelers and be perfectly happy. So do not use me as an example. It is always good to learn as much as you can about the software you are using. Knowing its capabilities is -as- important as using it.
Attached is the file with a single piston/ball contrained. I moved the cam to the front and gave it an angle contraint. This angle can be varied for seeing the motion. Not like mechanism, but it does change the piston as the cam changes. It can be driven to error once you exceed the contraints.
Mechanism can do a much better job as it adds contraint limits to the cam. Also note that the piston is shown with degrees of freedom although it lists as fully contrained. It is the ball that says it is partially contrained yet nothing can move. PTC still has some work to do here. But the ball/piston is contrained and will follow the cam between 38.5 and 54 degrees.
In this case, I used the simple tangent contraint and alighned axis. Also controlling orientation with planes, which is good to do. Normally, radial components will allow loose rotational contraints to be considered "assumed" contrained.
I've never tried zipping up a complete assembly, so I hope it came out correctly. Creo 2.0 M020.
