Bug: PlaneCoincident Lock Angle random behavior

[not-a-number-dot-org]

Hello.

I have a very strange behavior with Lock Angle of PlaneCoincident constraint. It seems to work but suddenly (meaning with no reason it stops working. I discovered this while asking for help on Assembly forum of FreeCAD. So everything (including file) is already explained there : https://forum.freecadweb.org/viewtopic.php?f=20&t=65900&p=567699#p567699

Thanks

[luzpaz]

@not-a-number-dot-org is this still an issue on latest Daily ?

[not-a-number-dot-org]

Yes it looks like. Here is a sub example with only the nut block and the screw if you're willing to test. Sorry I wont have time to fully test it until end of July.

test.FCStd.gz

[not-a-number-dot-org]

Just installed 2022.709

from angle 0° to 80° with 10° stepping : it works (<<vis sans fin>>.Placement.Angle is 80°) back from 80° to 0° in one step : doesn't work (<<vis sans fin>>.Placement.Angle is set to 90° instead of 0°)

[realthunder]

With current implementation of the SolveSpace solver, this is an expected behavior. To work around angle ambiguity problem, use Attachment constraint whenever possible. You can change the angle by changing the Offset property of any of the ElementLink grouped under the Attachment constraint.