cam follower mechanism

Yes, do not move it fast. 

It is very easy to loose your position in Creo and everything goes wrong.

Save snapshots so you can quickly get back to a known good state.

For demonstration purposes, set up a motion profile using a servo motor.

If you do an analysis, eg: position analysis that are the faster, it works or not?

It looks as if the cam / follower component's rotation axes are  not parallel. If so I don't think this is going to work.

If the yellow component's motion is linear, try to reduce its follower surface to the planar contact surface, not the entire curved surface. When specifying a planar cam/follower surface, you will have to specify its direction.

Remember a thing: when you work with cams you need more accuracy over the parts in contact.

At the same time also the accuracy of the mechanism analysis and the characteristic length should be adequate.

I think that Yongzhi has left die this thread...

Anyway, in my experience, often moviments actuated "manually via mouse" could cause the break of the kinematic chain; whereas using the servo motors this doesn't happen.

I hope PTC increments the functionalities of Mechanism also with new features.

A better work with the contacts, like cams, and with the 3D contacts in particular, I think should be done.

The idealized surfaces that can be used for cams MUST have their extruded directions parallel to each other. If curves are used, these curves must lie in the same plane in space. This comes from the fact that cams in MDX/MDO are 2D only, so this rule must be followed for cam pairs to work properly. Unfortunately, it does not seem the on-line manual or PLMS instruction points this out specifically, although 2D does infer this constraint.

If the yellow part only slides left and right, and the horizontal surface on the yellow part, in the cam definition above, is removed from the cam definition, and the vertical surfaces remaining on the yellow part's cam definition are split into two separate cams, it should work. 3D contact should work for this problem, but I've found that this option is very compute intensive.

Yongzhi can you share with us the asm file?

cit.: "And Problem is, if I drag the component slowly, it works. But if I drag it fast, it will be run out of work."

Remember that the problem occurs when he drag the part fast; so the cam constraint works, even if it is "unstable".

Christopher I have found this behaviour also with 2d cams (radial cam with roller) or roller with planar surface.

It works well only if you move it with servo motors within the analysis. Otherwise if you drag it manually, the risk there is always.