Ideas for modeling motion of disc revolving on angled axis while maintaining ground contact?

[anonymousstudent]

I made this simple model to try to figure out a technique for later use. The goal is for the blue disc to revolve while maintaining constant contact (at a single point) with the ground. In order for that to happen, the red frame has to pitch forward (thinking of the disc as the front end of the model), as shown in the images below.

Is there a way to model that motion? (That is, without faking it by using a motion link, which would not give accurate results.)

Thanks for any suggestions...

[anonymousstudent]

To be a bit more particular, I'm wondering if there is some combination of joints (and contact sets, possibly) that would result in the disc revolving while maintaining contact with the ground at a single point; and where that motion would result when driving the joint that revolves the disc.

[davebYYPCU]

Cylindrical Joint does not work?

Red to blue axis.

Drive the rotation control.  Contact set (unreliable) blue to black.

Might help.....

[anonymousstudent]

Thanks for the suggestion... if I understand correctly, a cylindrical joint would basically have the blue axis part telescoping out of the red frame. However, in the motion that I'm interested in modeling, there wouldn't be any telescoping—the only desired motions would be the disc revolving and the red frame pitching forward as required to keep the disc in contact with the ground.

[davebYYPCU]

Revolve red and blue

Revolve red and black.

Still relies on contact set, they are unreliable.

Might help.....

[anonymousstudent]

Here's another version of the file with the revolute joints and contact set you suggested (if I follow you correctly); however, it's not giving the desired motion. (As in the original file, there's also a slider joint.)

Any thoughts?

[anonymousstudent]

I had another thought and replaced one of the revolute joints with a cylindrical joint, updated file here. Still doesn't give the desired motion, though.

[davebYYPCU]

You have 3 bodies - 2 are components representing ground.

I have removed the cylindrical joint, and contact set.

Grounded the plate that had the existing rev 4.

Set joint limits and Motion link - close enough to get you going, because the disc scoops into the plate if you look close enough.

Animate Model using Rev 5 for - looks ok.

Might help....

[anonymousstudent]

Thanks—yes, the motion link approach is ok for a loose approximation but I'm hoping for something more exact. Thanks for your help though! I'll post an update if I have any other ideas...

[anonymousstudent]

Haven't found a proper solution yet, but came up with a way to visualize the movement in steps (rather than being able to animate a model with joints) by just dragging a sketch line. (This version is restricted to angles between 0 and 90.)