USER FORUM

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

Error with the normals (by Albert)
Hy all.

if You make 'assembling' You find the next error (see the jpg-file)
the ordinates (y-axies) are parallel, the abscisses (x-axies) are parallel too (the angle is set to zero).
but the browser window shows 'angle = 2.54 degrees).

Albert.
Wed Jun 4 2014, 23:34:26, download attachment SolveSpaceImage.jpg
(no subject) (by Jonathan Westhues)
I'm not sure which two skew lines are constrained with zero angle between them, but that doesn't usually work very well. The angle constraints are designed for nonzero angles. When the angle is nonzero, a 3d angle constraint removes one degree of freedom (DOF). When the angle is zero, it means the two vectors are parallel, which means that it removes 2 DOF. That's a fundamentally different thing, and the parallel constraint should be used instead.

But, you can't use two 3d parallel constraints, since that would overconstrain the system (3 - 2 - 2 < 0 DOF). You can use:

A 3d parallel constraint, plus a 2d parallel constraint as projected into an appropriate workplane.

A 3d parallel constraint, plus a 3d perpendicular constraint.

The "same orientation" constraint, which is designed specifically for this application.
Thu Jun 5 2014, 22:46:02
Error with the normals (by Albert)
Hi mr. Westhues.
Yes, the angle constraints works not very well around the small angles. At the screenshot one can see the two bases and the constraint 'the angle between abscisses is equal to 1 degrees'. But the measurement shows 2.06 degrees. This example You can get when You will shift the applicate axis.

Albert.
Fri Jun 6 2014, 21:54:08, download attachment AngleBetweenNormals.jpg
Error with the normals (by Albert)
one more screen
Fri Jun 6 2014, 22:02:39, download attachment DefectiveNormalsII.jpg
Post a reply to this comment: