Hello !
I am currently doing a simulation of a 3D cylinder falling on to the ground (analytical rigid surface) with an initial velocity of 5m/s from a height of 2m and allowing it to keep bouncing until it comes to rest.
The problem is that after the cylinder bounces in the middle of the rigid surface, it slides to the edge and falls down from the edge. I just want to have only the bouncing of the cylinder without having sliding and falling down from the edge.I have used surface-to-surface contact. I have attached the displacement graph which clearly shows the sliding and falling part after the bouncing stops.
I have tried using surface-to-node discretization and got just the bounce of the cylinder without sliding and falling. But, it would be better if someone can help in solving this case for surface-to-surface contact.
Thank you
Pravin.
Pravin, we need a lot more information. Most important is to understand the motivation for your whole simulation. All simulations involve assumptions or approximations, and one of the most important simulation skills is selecting those assumptions and approximations that are appropriate for what you are trying to learn. So what is it that you are trying to learn? For example, is this an academic exercise that you were assigned? If so, what is the "deliverable" (required product of your work)?
What simulation tool are you using?
Is gravity orthogonal to the cylinder axis, i.e. in the -y direction? So the curved surface of the cylinder is hitting the plane surface? If so, the horizontal direction of the axis is unstable. At least after one bounce, one end of the cylinder will hit first, and the impact will become uneven, and more so with each impact.
You wrote that the cylinder "slides", but the figure suggests rolling. Which is it?
Please give us all the details. (And next time, please start by saying "this is a mechanical simulation" (if that's true), not, e.g., electromagnetic.)
Andy
As m. Miller says, the Devil is in the details. I once passed a few days just to "debug" the behavior of a physical simulation I did with the Bullet framework. Sorry but I do not think your problem can be diagnose "remotely" on a forum like this...
You need to provide more detail to exactly understand the simulation environment.
I made a similar simulation with explicit solver for free fall of ball and had rolling of ball where I expected just bounce, but when I made the similar simulation using the implicit solver I got rid of that.
I guess that in surface to surface contact the problem is in smoothing of mesh data and point where your cylinder makes contact with surface, if your cylinder doesn't make contact exactly on mesh node there will be a numerical error and you will tend to have rotation in direction of the far node, also considering fact that your cylinder is flexible there will be a micro deformation at time of impact with surface which will most probably alter mass matrix to induce the rotation. There can be many possibilities, to go deeper full simulation description is necessary.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Computer Graphics and Animation
- 3D