Fillet edge grasshopper operation failed

Hi
I can’t figure out why fillet edge is not working, all edges are included
please see attached file
thank you
fe.gh (9.9 KB)

Your target geometry is too complicated:

  1. It is not solid - from this view you can see inside it.
  2. Many of the edges are formed by more than 2 adjoining surfaces
  3. This is an area of strange weirdness.

Basically you gave it something for which it was not designed to handle.

You might be able to get it to work if you made the spheres impinge only on the outside of the box, but I suspect you’d also have to ensure none of the sphere subtractions overlapped.

I made several changes and got Fillet Edge to work.

  • the white group rotates surface planes so that sphere seams are not inside the SDiff result, so they won’t be included in the list of edge indexes.

  • I reduced the PopGeo ‘N’ (Count) from 50 to 5, avoiding overlapped spheres.

  • I changed the Fillet Edge ‘R’ (Radii) slider to REAL so I could reduce it less than 1 but also increased its range to find the limit (which depends on other factors).

As is, it works, though it can be broken with different parameter values.


fe_2021Jan23a.gh (12.4 KB)

It is a “Closed Brep” (solid) with all spheres centered on the surface.

P.S. Here are 20 spheres with Fillet Edge ‘R’ set to 1.4 but notice some anomalies.


fe_2021Jan23b.gh (14.2 KB)

I would say it is more a rhino problem than a grasshopper one.
Fillets in rhino often fail. Even more when a lot of geometry is involved.
If you try the same operation in rhino it will fail most probably.
Since most of the grasshopper methods rely on the methods exposed in the rhino common(not sure in this case) you will probably experience the same in gh.

Usual it is not a bad idea to try things first in rhino before spending a lot of time in gh.

Thank you so much guys!!!