Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Big Red lines even if "force NURBS surface to triangle mesh" and export stl #1298

Closed
fredOnGitHub opened this issue Oct 7, 2022 · 5 comments

Comments

@fredOnGitHub
Copy link

System information

  • SolveSpace version: 3.1 70bde63
  • Operating system: win11

Hello,

from this it is ok
bug1

but from this arc version
bug2

i get
bug3
and it is ok when i do "force NURBS surface to triangle mesh" => no red lines
but when i export triangle maillage
i get
bug4

I am furious and disappointed (LOL).

Thanks

@ruevs
Copy link
Member

ruevs commented Oct 7, 2022

It is a known problem #1291.

Make the straight parts (corner) of the cutting ark sketch stick out of the part by a bit (instead of coinciding with the corner) - it will solve your problem. In other words the 13mm line segments and the ark should not meet tangent to each other at the edge of the part. Instead there should be two short line segments perpendicular to the arc ends (and the part) sticking out.

@ruevs ruevs added the NURBS label Oct 7, 2022
@fredOnGitHub
Copy link
Author

Hello,

if i understood i have done that
done1
that
done2
that
done3
that
done4

to get
done5

1st question :
Is it the good way ? More simply ?

and finally with Windows 3d viewer
done6
But I see polygons...
2nd question :
Are 'chord tolerance (in percent)' and all other parameter good ? I don't remember the default values.
How can I get nice shape without polygon mesh ?

Thanks

@ruevs
Copy link
Member

ruevs commented Oct 8, 2022

You did it in a way that is more complicated than needed. Here is what I would do:
FilletTestsFailuresExplined

The failing case asks the software to make two "complicated" (for an algorithm) decisions:

  1. Does an infinitely thin (zero thickness) "sliver" of material remain in the spots marked in yellow?
  2. The material being cut approaches zero thickness when approaching the points where the line segments and the arc meet tangential to each other. this poses a numerical problem.

To avoid this I just used two construction lines to position the arc and moved the two lines forming the sketch to the outside.

Another hint - make the extrusion with the "difference" operation "two-sided" to avoid asking SolveSpace to solve a similar singularity problem where the "front" faces of the two solids coincide (it does usually handle it OK).
FilletTestsRotated

The test model:
FilletTests.zip

Some previous explanations on the topic:
#470 (comment)

The issue collecting all such problems:
#738

@ruevs
Copy link
Member

ruevs commented Oct 8, 2022

Are 'chord tolerance (in percent)' and all other parameter good ? I don't remember the default values.

The export chord tolerance (in mm) and export max pieceise linear segments parameters control how fine the exported meshes are. Make the former as small as you want and the latter as big as you want to get the desired precision. After an export operation SolveSpace displays the exported mesh until you press ESC. Turn on the mesh MESH to check if it is acceptable.

Read these two sections of the reference:
https://solvespace.com/ref.pl#Chord
https://solvespace.com/ref.pl#Known

@fredOnGitHub
Copy link
Author

fredOnGitHub commented Oct 8, 2022

More simply !
Thanks !
I have a nice NURBS into the form now.
good1

compared to rendering before
good2

About the exportation, I have done that :
good3
to get
good4

Thank You very much !

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants