Ok, get it, but then to make a offset of the original and mirrored it gives Unhandled exception in ViewProvider::eventCallback: makeOffset2D: result of offsetting is null! (Event type: SoLocation2Event, object type: SketcherGui::ViewProviderSketch) Guess that isnt fully implemented yet, hope when i...

I dont know how to put that horizontal line crossing that point. Use Sketcher ToggleConstruction first and then Sketcher_CreateLine . From that bottom point to somewhere on the right from it (just make sure that it's horizontal - auto constraint will take care of this if you position the cursor pro...

Trust me, the lines arent mirrored as you can see int the attatchment, yellow dot pick. Try with a horizontal line crossing that point instead. That's the usual way of mirroring in sketches. mirror.png Also the offset should keep the original lines as we project it, then offset a side and close wit...

Offset is on one side, just leaves the original geometry but you can delete it using a proper checkbox (shortcut U).

What mirror option ? The Symmetry tool creates a mirrored copy of the original geometry by default. There's also an option to delete the original geometry like with offset.

Yeah, it works well when the view of the cylinder is vertical but not when it's horizontal:

To be honest, I don't use Boolean fragments this way. But "fuzziness"/tolerance can be helpful when this boolean operations fails due to inaccuracies in geometry alignment (like other fuzzy booleans available unfortunately only in the Defeaturing workbench). In most cases it's not necessary.

RodMM wrote: ↑Mon Apr 08, 2024 10:09 am Seems like I had set the tolerace for boolean fragments to 0, which broke it. Thanks, it's now meshing fine !

It's 0 by default. What value do you have to use here to make it work ?

What are some potential project ideas for a Google Summer of Code (GSoC) project related to Finite Element Method (FEM) and the use of Abaqus and CalculiX as FEM solvers? We can talk about that on pm. There's still a lot to do to improve FEM. Of course, we can't use Abaqus as a solver in FreeCAD bu...

This geometry is broken so you won't be able to mesh it:

Usually, the agreement with hand calcs is much better for simple problems like this. But we have to keep in mind the simplifying assumptions of those formulas. Here, the circle should be small, according to Roark's. If you make the radius 4 times smaller, you will get a deflection of 0.332 in which ...