SOLVESPACE -- parametric 2d/3d CAD

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

construction lines, constraints, locked constr. lines. (by John Griessen)
I'm working up a cylindric shape with draft angles for vacuum molding release. It will have some bezier curves and radii on internal and exterior corners of a turning.

Is there a weighting built into the solver for dealing with construction lines? They seem to move when I constrain to them as if they had about the same weight, or inertia, or work function as the other elements. So, I guess I will lock points of them before using them as an on grid reference to my curvy off-grid mold shapes desired.

Is there any other lock constraint besides point? If not, it would be a speedup tool to be able to lock a group or all points in a selected set. To avoid lots of work undoing such, to be able to unlock selected set points, (if not grouped), would be good too.

Anyone have hints for speediest methods with solvespace as is when wanting
constr. lines to stay put?

Aside: How do you show grid snap points and change spacing of snap?
Fri Apr 10 2015, 10:47:04
(no subject) (by Jonathan Westhues)
The solver treats construction entities identically to non-construction entities, so your observation is correct.

Can you put the construction geometry in a separate group? All geometry in earlier groups is locked with respect to later groups.

For the grid, see View -> Show Snap Grid, Edit -> Snap Selection to Grid, and the configuration screen (in the browser) for the spacing.
Fri Apr 10 2015, 10:54:08
earlier groups lock, solver small angle workarounds? (by John Griessen)
That earlier groups are locked relatively is what I was looking for, thanks.

I found when changing the draft angle to one degree on a side that the
solver didn't converge. 1.5 degrees, or 3 degrees total is probably fine for what I am aiming for, but this could be a limitation.

Can anyone suggest workarounds for using constraints with small angles like 1 degree?


John Griessen
Fri Apr 10 2015, 13:05:14
solver small angle workarounds? (by John Griessen)
I am using the gtk port from a few days back, not the latest main repo code.
Fri Apr 10 2015, 13:06:29
(no subject) (by Jonathan Westhues)
The way that angle constraints are modeled now makes convergence tricky at small angles. If you constrain the ~90 degree angle with a horizontal line (e.g., to 88.5 degrees) instead of the ~0 degree angle with a vertical one, then the problem should go away. A better definition of the constraint equation could do something like that automatically.
Fri Apr 10 2015, 16:19:49
(no subject) (by John Griessen)
Great! I'll have to turn this part into a tutorial.
Fri Apr 10 2015, 16:22:45
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):
© 2008-2018 SolveSpace contributors. Most recent update Nov 22 2018.