Attractor Point Auxcetic Structure

Hi! I am doing a project on auxcetic structures, and I am trying to change the thickness of the mesh depending on the distance between an attractor point and points near the structure. But for some reason in my definition, the thickness changes in each bowtie cell (my auxcetic stucture) and not in the whole structure. Does anyone know what may be happening?
Greetings!

BowTie_2.1.gh (22.4 KB)

The reason your mesh does not have a gradual attractor based thickness is because it’s not one mesh but 128 separate meshes. The separate meshes do not join because they do not have common vertices. Use the 2D mesh I created for you in the other thread as a start. It looks best when you TriRemesh it before calculating distances to an attractor point.

mesh_thicken_attractor.gh (23.7 KB)

Actually… the collision works best with the 2D mesh. After the collision you could add the thickness based on an attractor. The collision isn’t accurate but at least it leads to a nice result with an intact quad mesh.

@DanielPiker might have a better idea to simulate the collision with a thickened mesh.

bowtie_mrtn.gh (78.1 KB)

2 Likes

Thank you Martin! For some reason I can not visualize the first definition that you sent me, trimesh shows me an error…
Do you know if I can do it with the first definition I sent you? I need to 3d print this in the same size that I have in the first definition I sent you to, make a comparison between models I printed previously.

which file shows an error ?

This one:
mesh_thicken_attractor.gh (23.7 KB)

I don’t know what’s wrong on your side. But it actually works almost better without the TriRemesh and a Combine & Clean instead. This way you can keep a quad mesh

mesh_thicken_attractor.gh (22.2 KB)

Have you noticed how you can move the attractor point?

With Preview on, selct the point component, then drag a Gumball arrow.

Ok, thanks! And is there any possibility if I can apply the mesh with the attractor point to my first definition? Thats because I have parametrized certain parts and I need to compare the new model with old ones.

And for some reason the mesh adds these cones on the corner of the bowtie, do you know what problem may be?

DUPLICATE (related?) thread:

Dear Tom, as mentioned above, I would not use the thickened mesh for the collision simulation in Kangaroo.

Maybe @DanielPiker has an idea how you could do that.

Thanks! But I am using the defintion that you sent me ( mesh_thicken_attractor.gh) and for some reason I have an irregular mesh like the picture I sent you before

It has to do with the goals…

Here’s a definition that works with a thickened mesh.

kangaroo_bowtie_collision_thicken_attractor.gh (46.7 KB)

Thanks Martin! I kept on working on the defintion, and I am trying to make a shell structure with other components of Kanagaroo2, but for some reason my structure is copied various times, instead of being a shell structure. Do you know what may be happening?

Kangarro_BowTie-Auxcetic_060821.gh (27.8 KB)

Hi Tom, I have been on holiday…

I got rid of Pufferfish some time ago and I’m not 100 % sure what the move to point component does.
I replaced it with an GH native Orient component.

Look at the inputs of your Rigid Body goal by hovering the mouse pointer over the inputs. The path structure does not line up! You have 128 open breps in one list and then planes and points in trees with 128 branches. The Part input needs to be grafted.

Look at my file… I added an Entwine component before the Kangaroo solver. This way you can get the path 0;1 or whatever you need very easily. For some reason your Kangaroo solver did not output the results with a path structure, which makes me guessing you are not using the latest version of Kangaroo and you might have to update your Rhino installation.

Kangarro_BowTie-Auxcetic_060821.gh (37.0 KB)

Instead of piping all segments separately, you could use Multipipe. You get nice interections

Thanks very much Martin! I still have te doubt if it is possible to have the results of the simulation using the extrusion in the begging and not with the multi-pipe component, because I am trying to imitate a analog simulation of a structure that was printed in FDM.

Greetings!