Grasshopper

algorithmic modeling for Rhino

Constant strain tetrahedral elements in Kangaroo

Testing some new finite elements in Kangaroo. Note that these are not just networks of springs, but properly volumetric 3d elements, using material property inputs of Young's modulus and Poisson's ratio.

The second part shows the contrasting effects of setting a positive or negative Poisson's ratio.
Thanks to Professor Chris Williams of Bath university for sharing the theory.

Views: 810

Comment

You need to be a member of Grasshopper to add comments!

Join Grasshopper

Comment by Daniel Piker on March 31, 2014 at 6:15am

Hi Rasmus,

You can already use actual material parameters such as Young's modulus for the 1d elements in the current version of Kangaroo, as described here.

While you can simulate surfaces and volumes as networks of 1d springs (triangles or tetrahedra respectively), this is not really accurate in most cases, because in reality deformation in one direction affects the other directions.

What is shown in the video above is a tetrahedral element, which calculates the nodal forces based on the strain over its whole volume, not just for each of the edges in isolation. I've also added a similar constant strain triangle.

..and yes, I plan to include these features as part of the next version (maybe along with the 6dof nodes too, we'll see).

Comment by Rasmus Holst on March 31, 2014 at 5:47am

Hi Daniel.

Look really promising. Is this part of a new version implementation? With actual material parameter inputs instead of "imaginary" numerical values? You say that this is done as a solid, and not as a spring model? Can you explain that? Or will that follow with a new release? :) Looking forward to it. Any news on 6 DOF's?

Comment by to] on March 27, 2014 at 1:29am

Great, it is really awesome to see your work and thoughts. I still cannot stop watching the elephant :)

Comment by Hrvoje Petrovic on March 26, 2014 at 1:17am

Great and superb as usual!

Comment by Mateusz Zwierzycki on March 25, 2014 at 5:37pm

Superb. I bet that the commercial FEM software is not as far as you think Daniel.

Comment by Ángel Linares on March 25, 2014 at 4:28pm

Yeah, sure, but what you show us here are some very nice results; It could be nice to compare them with some commercial tools. We are heading, slowly but confidently, to a pleasant and nice future of parametric structures design ;)

Comment by Daniel Piker on March 25, 2014 at 4:21pm

Haha, thanks Angel - let's not get too carried away just yet though ;)

More validation is needed before we can be confident of the results.

Comment by Ángel Linares on March 25, 2014 at 4:16pm

Awesome work and results!!!! Let's make architecture schools teach Kagaroo as default FEA tool :)

Comment by David Stasiuk on March 25, 2014 at 2:16pm

stunning!

Comment by Andrew Kudless on March 25, 2014 at 1:59pm

Nice work Daniel!

About

Translate

Search

© 2019   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service