I just created geometry including a set of four "corners" which are constructed of two perpendicular straight spans closed by an arc. The job include two instances of these. One of the 8 "corners" has a different and not intended toolpath applied to it. I believe that all 8 "corners" are equivalent. Is this a bug, or have I created some unintended geometry? I have attached a screen capture of the preview with the unexpected "corner" indicated with an arrow. I have also attached the .crv3d file that produced this result.
Attachments
Image of preview with arrow pointing to unexpected "corner".
Thank you, highpockets! I ran the vector validator and it identified a number of "overlaps" where the arcs were tangent to straight line spans. Curious that this was the case for corners that toolpathed as expected as well as for the one that did not. I did not find any extra spans, and I was not able too identify anything that appeared to be an overlap.
For the troublesome corner vectors, I cut the three spans apart in node edit, then joined them using the join tool. When I created a VCarve toolpath with the rejoined spans the result was as expected.
I expect it's all to do with tolerances. We can see that it isn't actually overlapping but as the gap runs to less than 0.0001" of an inch the internal algorithm is probably seeing that as overlapping and given that there's no way a tool can fit in that gap the v-carve toolpath is probably treating it the same way.
Moving the start point to the centre of the curve sorts it out.