Seeking clipper equivalent on 2-way curved surfaces

[Will_Wang]

picking up on this

Smart offsetting of curves with trimmed intersections

Clipper is working when the surface is planar. Thanks a lot for this plugin. Now if I want to convert the offset network to a surface like pathways, how should I do it? Region difference crashes Rhino after hanging it. [Screenshot%20(114)] What is the way to use clipper so that the offsetting trims work on curved surface as well? I am getting the following result. [CRVD%20SURFACE] The original cellular grid was generated by using paneling tools. I am attaching both my model and grasshoppe…

Has anyone achieve this yet?

I’ve mapped voronoi cells onto a surface like below

How can I offset them on the surface without the corner clipping and intersection problems?

Long story is that I’d like to punch holes on the surface with the offset voronoi cells. The “struts” should be of a certain width. I’ve previously done all this with planar operations and remapped the results but because of the free form surface, I could no longer maintain the dimensional accuracy on the surface.

I’ve also tried multipipe on this but this circular cross section is rather limiting. Self intersection can happen too.

[martinsiegrist]

Check the linked post for an approach on how to eliminate short segments…

Merging 3D Voronoi Network into one unified object

Below is a definition which uses the Multipipe component to create a closed SubD around the edges of the voronoi structure. This definition also removes short line segments. You can bake the result and use ToNurbs to turn it into a polysurface. Converting into a Brep in Grasshopper is also possible but it might take some time. [image] [image] CollagenModel_Wire_16022021.gh (21.2 KB)

This topic could also be interesting for you:

Voronoi Cells Minimum Edge Length - Grasshopper - McNeel Forum

[Will_Wang]

Thanks. Those could solve the multipipe self intersection and some of the offset clipping problem
But I’d still like to know if there is a way to have the cell offset inwards almost to a zero area extreme (where they just disappear like what Clipper does)
In that case no edge is long enough…

[martinsiegrist]

This can be approximated by using local planes.

I evaluated the surface at the closest point of each cell centroid.

The offset curves need to be pulled to the surface.

clipper_offset_local_planes.gh (35.9 KB)