Space Frame Mesh Modelling

Hi All,

I tried this for a long time in many different ways but I failed to achieve the result I am anticipating.

I need to model a Bucky space frame based on 2 triangular grid meshes (see image).

This is what I did in the definition:

1- Pick every other mesh face using David Stasuik's VB script

2-Project a pyramid mesh face on each of these

3-Get the edges.

The last step would be to connect the vertex of each of the pyramids in a triangle grid mesh that is 'slipped' from the original grid.

Any ideas?

Definition attached. Your help is appreciated.

Chap.

Attachments:

Replies to This Discussion

Permalink Reply by peter fotiadis on May 21, 2015 at 2:29am

BTW: Just before Seppuku > Plan Z > Skype

Permalink Reply by peter fotiadis on May 20, 2015 at 11:19am

BTW: there's some nasty bug in the ballpivot code: see the direction of the primary edges (near by the left upper GH "window" corner). Nothing to do with the truss ...on a per se basis ... but aesthetically rather unacceptable.

better visible in the flipped mode: WDir towards -Z (mesh out of sight):

Permalink Reply by David Rutten on May 20, 2015 at 10:26am

I meant a GH file without the weaverbird components. I need to see how the points are actually organised within Grasshopper. If they are properly structured then the Relative Item(s) components can probably be used, if they are not, then you need to take Peter's suggested route and actually deal with the mesh topology.

Permalink Reply by Chapuline on May 20, 2015 at 10:33am

Sure David,

Here I replaced the WB Prism component with a face normal SDL and an end point.. Which is the same thing for me really. If these points were connected, the job is done.

Chap.

Permalink Reply by Chapuline on May 20, 2015 at 10:39am

The file..

Attachments:

spaceframetetrahedraleveryotherNOWB (1) (1).gh, 76 KB

Permalink Reply by David Stasiuk on May 22, 2015 at 1:33am

Hi Chap-

Attached you'll find a script that should do the trick. The sorting process works with regular triangulated grids (as you have here: all interior vertices are valence 6)...once you break that topology, though, it won't work.

hope it helps!

Dave

Attachments:

TetTruss.gh, 3 KB

Permalink Reply by Chapuline on May 22, 2015 at 5:26pm

Hi Dave. That's perfect! Its actually exactly what I'm looking for. Is there a way to remesh my topology so that all the interior verticies are valence 6? Then I assume your script would work on the mesh I am working on (the piece Is sent is a small patch)

Best;

Chap

Permalink Reply by David Stasiuk on May 23, 2015 at 12:10am

It depends on the geometry. There are many configurations where it isn't possible (for example, a sphere)...and even in many instances where it is possible, it is extremely difficult. If your source geometry is surface like, then you can distort a valence 6 mesh over it...but the more doubly-curved it is, the more distorted your mesh will become. What does your original geometry look like?

Permalink Reply by Chapuline on May 23, 2015 at 10:49am

Hi Dave,

Thanks again for your helpful insights. My geometry is a tent-like structure, generated through a mesh relaxation.
I'm afraid its very much a doubly curved surface.

Any way that could be resolved?

Chap.

Permalink Reply by Chapuline on May 23, 2015 at 11:14am

Another issue is.. That I am feeding a grafted list of mesh faces, and a grafted list of offset distances. Your script doesn't seem to take data trees.. I need this because the structure gets shallower near an attractor in the design I am working on.

Thanks!

Chap

Permalink Reply by David Stasiuk on May 24, 2015 at 3:27am

It's always easier if you post files...:) If you're using mesh faces as a grafted list, then you lose all of the connectivity information embedded in the mesh (which really is the main reason to use it!).

I also think that if you're working with more freeform geometry, then the general approach that Peter is outlining - basically, using the dual graph of the base mesh (bottom members wireframe) to create a polygonal approach to the top member wireframe, and then creating a tetrahedron for each bottom face. This will negotiate any type of base mesh, so long as it is triangulated...if your base mesh has some mix of 5,6,7 valence faces, then it will be able to represent your double curvature well while maintaining similar member lengths.

Is there a particular reason that you want/need to use the specific truss typology you've shown?

Permalink Reply by AJ on May 20, 2016 at 8:43am