SOLVESPACE -- parametric 2d/3d CAD

(you are viewing a thread; or go back to list of threads)

triangle mesh export results in errors (by Garfield)
I often model parts in solvespace and export these as triangle mesh.
When I import these in Slicer (actually PrusaSlicer), I regularly get a message the the STL model had errors that were "fixed". It seems not to affect the output, but I find these messages a bit worrying.

BTW, I don't get any errors with OpenSCAD or FreeCAD (if FreeCAD does not break the model during editing, but that is another issue and why I like SolveSpace so much).
Sat Feb 6 2021, 15:41:22
(no subject) (by Tom)
Try "Show Naked Edges" under Analyze? SolveSpace can do the STL check too, and will highlight the problem areas.
Sat Feb 6 2021, 15:44:27
(no subject) (by Paul)
If you can avoid using "force nurbs to triangle mesh" it should always export a perfect mesh. If the NURBS fail and you needed to use that checkbox to get around it then I guess you found both a NURBS bug and a mesh bug. If you can create a simplified test case, we can add it to the list.

Also, you didnt specify a version number. 3.0 has a lot of those bugs fixed.
Sat Feb 6 2021, 21:51:39
Example (by Garfield)
I use SolveSpace 3.0-02aa9ea8, compiled from github few days ago.

Attached is an example.

With NURBS there are no errors displayed with "show naked edges". However, Slicer reports error with the exported STL.

With the triangle mesh everything is fine.
Mon Feb 8 2021, 03:03:42, download attachment example.slvs
(no subject) (by Paul)
I don't think there is actually a problem with that sketch. What I find strange is that the triangulation is different when you check or uncheck "force to triangle mesh". I don't know why that would happen with a single Lathe solid.

Maybe the slicer doesnt like some of the very thin triangles?

If you're building from source, you could try commenting or deleting the optional triangle fan code. It's in src/triangulate.cpp around lines 392 to 455. It has comments at the start and end of the optional section.

I don't think that's the problem though. I suspect Slicer is just complaining about a non issue. Is this case a fluke or do you get warnings a lot?
Tue Feb 9 2021, 10:20:47
(no subject) (by Garfield)
I get these errors very regularly, often with rather simple bodies. I mostly ignored them because so far they had no consequences.

I commented out the fan code as suggested but this does not change the result.

What I get in Slicr is "Manifold: Auto-repaired (166 errors)"

Out of curiosity I imported the STL into MeshMixer (running under Wine). Also there I get errors when choosing "inspect", see attached screenshot.

blue indicates holes, red that the body is "non-manifold"
Wed Feb 10 2021, 06:10:11, download attachment meshmixer.png
Post a reply to this comment:
Your Name:
Your Email:
(no HTML tags; use plain text, and hit Enter for a line break)
Attached file (if you want, 5 MB max):
© 2008-2020 SolveSpace contributors. Most recent update Oct 30 2020.