Bending resistance + twisting resistance?

Dear Group (and Daniel in particular I guess),

I'm trying to use Kangaroo for simulating 3d-bending of a developable ribbon surface.

Kangaroo has proved to be amazingly correct for 2d-bending (and 3d-bending of rods), but so far I haven't been able to get it right in 3d for a developable ribbon surface.

I've made a sketch (image + ghx-file) for a way to simulate bending AND twisting in Kangaroo. Rather than making a network of springs and bending-resistance-joints, this method creates a series of planes and rotates each step in two ways. The first rotation is bending. The other one is the direction of the bending (=twisting).

I'm thinking it would be possible to set something like this up with Kangaroo. Both bending and twisting would be series of "bending resistance" values fed into the Kangaroo.

The definition looks a bit messy, but it's really quite simple. The tapeworm is a scripted version of this.

From physical testing, my impression is that the twisting resistance is much more powerful in a material than the bending resistance. A twisted material will most of all try to become a 2d-bend, and then it tries to become flat. With this in mind it could be interesting to set the amount of bend and twist resistance individually somehow...

Thanks a lot for any help!

Best

/Mårten

http://thegeometryofbending.blogspot.com/

Attachments:

MN-bend+twist.ghx, 792 KB
MN-bend+twist.PNG, 380 KB

Replies to This Discussion

Permalink Reply by Daniel Piker on March 7, 2011 at 5:43am

Thanks Mårten,

This does look an interesting approach and I would very much like to sort out something for torsional stiffness in Kangaroo.

I've found this paper:
http://www.cs.columbia.edu/cg/rods/
and this guy's work:
http://math.arizona.edu/~goriely/
quite inspiring in this area.

The relationship between the geometric torsion and material twisting of a rod is a bit more subtle than it might first appear. In this case of a strip rather than a rod though perhaps it will be more approachable because it has a clearer orientation.

Thanks for sharing your ideas on this, I'll look into it and see if I can turn it into a force.

Permalink Reply by Daniel Dirscherl on October 4, 2013 at 12:56am

Hi Daniel,

I'm relatively new to GH and very new to Kangaroo. Have you had a chance to develop this idea any more since the last post? (I haven't yet read through the posted references- apologies if they answer my question) I am hoping to use Kangaroo to simulate active bending + twisting of a wood member (ribbon surface) into an arch. I am hoping to be able to be able to replicate the lateral deformation that occurs in the ribbon as a result of the torsional forces....

Many thanks!

Daniel

Permalink Reply by David heaton on November 23, 2015 at 3:31am