I have a marble climbing machine. The balls in the model bounce around even though i have the Bounciness set to zero. I tried transnational and rotational damping but it seems to mask the problem by simply overly damping the normal motion. I have had this problem with other physics model, so i world like to figure out what is the correct way to simulate the problem. I have attached the model.
maybe one thing that could help is to set the balls rigid body`s to spheres and not mesh.
I would also use as much as possible simple geometry for the gear and so on. And would try to avoid the mesh setting .That makes the simulation much faster than using “Mesh” for rigid body shapes… And much more controlable. For the rails you maybe could use a half cuttet spiral tube an set it to mesh for rigid body collision. The simulation steps should be realy realy enough in my opinion to make this nice machine work.