Im having a hard time getting the _shell command to create a proper, fully formed solid. When i run the command (using a thickeness of 2), selecting the faces at the end of the prongs for removal, i get a “shell result is not solid” message, and a prong and the top talon are not properly shelled.
I dont understand why I am getting these results (still new to this). My guess is i am not select correctly the “faces to remove from polysurface”, but i cant think of any better faces to select. Anyone have any tips on how to approach this? Let me know if you need any more information.
That’s going to be tricky. If you just run offsetSrf on it, you can see a number of spots where it messes up, the issue is not the faces you pick to remove from the shell operation. There are a number of surface connections that are close-but-not tangent,and as offsettting will emphasize the slightest error it’ll just be impossible for Rhino to get clean intersections. There are also spots near the ends of the “prongs” where the curvature is too small, the offset surfaces will fold over on themselves. So you’re looking at using OffsetSrf, then seeing what you need to clean up manually. What is the end purpose, is the work worth it? If a mesh would suffice OffsetMesh produces a result with 15 self-intersections but that might not actually be a problem.
ultimately, this is going to be sliced in half and exported for 3d printing. my (rough) understanding is that meshes aren’t valid for 3d printing. I could get away non piping it, though if i slice it in half I will end up with a wasted surface where the halves would join (and htis is a large project, so it will add up), but you are right, it may not be worth it.
actually I may just go with printing it as a solid, hollow object, and then cutting it in half (though i worry about the crudness of that.)
Meshes ARE what you normally export for 3D printing, so there’s no problem, it just has to be smooth enough to achieve the finish you need given the limits of the printing process. And depending on the way it’s processed, the errors in the mesh may not actually be a problem…but don’t quote me on that. To “shell” shapes like this I’ve had to resort to essentially constructing the offset manually, maybe using the results I get from offsetsrf as a reference and to take section curves from.