I don't think 90 degrees bug in rotation has anything to do with this problem. I think it is because you subtract the head from the hair leaving a knife edge and then union it back with the head. That union has a lot of exactly coincident faces in theory but in practice they are slightly off due to number representation. The differences are too small to represent in floats, so the topology gets corrupted by truncation. I get around the rotation bug by using own version. module rotate(angle) // built-in rotate is inaccurate for 90 degrees, etc { a = len(angle) == undef ? [0, 0, angle] : angle; cx = cos(a[0]); cy = cos(a[1]); cz = cos(a[2]); sx = sin(a[0]); sy = sin(a[1]); sz = sin(a[2]); multmatrix([ [ cy * cz, cz * sx * sy - cx * sz, cx * cz * sy + sx * sz, 0], [ cy * sz, cx * cz + sx * sy * sz,-cz * sx + cx * sy * sz, 0], [-sy, cy * sx, cx * cy, 0], [ 0, 0, 0, 1] ]) children(); } The built in sin and cos functions are accurate in degrees but they are not used in the rest of OpenSCAD. I can fix that if the PR is likely to be accepted. On 1 March 2017 at 03:59, adrian <adrianh.bsc@gmail.com> wrote: > Thanks Wolf, > > I don't know the API yet, but that a 90 degree rotation causes degenerate > faces is definitely a bug. I'll try and take a look at this sometime in > the > next couple of weeks as this is completely unacceptable. > > > > -- > View this message in context: http://forum.openscad.org/STL- > is-unreadable-or-not-watertight-tp20642p20678.html > Sent from the OpenSCAD mailing list archive at Nabble.com. > > _______________________________________________ > OpenSCAD mailing list > Discuss@lists.openscad.org > http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org >