urethane mold simulation

I'm a student at the university of Texas at Arlington. I'm working on simulating a urethane mold in kangaroo. Our goal is to be able to coordinate the geometry of adjacent panels.

I'm having a hard time getting the right simulation out of kangaroo. Iv'e been working with springs but I'm not sure they are what I need. I'm new to kangaroo and pretty new to grasshopper. if anyone has recommendation on tutorials or way to approach it I would appreciate it.

I attached a rhino file that has the idea of the setup were looking at making. The two rods would deform the surface of the cylinder and would give a digital control over aperture.

_best

Attachments:

Cylinder_Deformation_Test_01.3dm, 944 KB

Replies to This Discussion

Permalink Reply by Bohmwave on October 24, 2013 at 10:00am

I just started Kangaroo and am having a real rough time, too. I couldn't get anything to work properly from the example files (except the ones without gravity, haha). AFAICT, Kangaroo is made to work with closed objects, which you have to explode into points to get to work. My suggestion: Start a new project. Make two mesh objects, then begin using different forces. See if you can get the two objects to interact. Once you do that, you'll see what's going on.

From one newbie to the other: the idea is to convert a mesh or brep into points, then feed those points into forces, then feed those forces to Kangaroo. (I'm still unclear on how to pipe Geometry.) Also, turn off Preview for your grasshopper objects once you have the Kangaroo piped. That way, you'll only be watching what Kangaroo is doing.

I made a box and a sphere, and was able to use one to be the boundary of the other, and get gravity going. Now I'm slowly puzzling the rest out. This helped me a bit:

http://www.grasshopper3d.com/video/bending-simulation-in-kangaroo

***what is below is pure guessing on my part***

For your needs, I'd suggest the rods need to be piped as Unary forces. I'm pretty sure the cylinder needs to become a giant list of points, which you then connect as springs. You could make springs that run the length of the cylinder (vertically) as well as outer & inner perimeter rings of springs at each interval you want to deform (two cross-sections). That way, you can set up material plasticity through the wall thickness of the urethane cylinder. Once you do this, make the bottom and top surfaces of the urethane into anchors, and you should be able to start tuning the response of the urethane. Be warned, however, that since everything is dimensionless, you're going to need to build some way of accurately simulating the response of the material. This is nontrivial.

Permalink Reply by Nathan Barnes on November 1, 2013 at 4:59pm