Hi,

the error is in a fillet surface.




You could fix it with the ShrinkTrimmedSrf function.

Don't know if the source is the original modelisation software or in the import.

Have fun...

Hi 3image, it looks like it's a bug in the solver where in the bounding box extents calculation it's mistakenly going outside of the trim boundaries and processing some parts of the underlying surface that should not be active, for 2 of the surfaces that are portions of a torus that Pilou shows above.

I've put it on my list to investigate in the future but I expect it to be a high difficulty and high risk area to try and fix.

- Michael

Hello Michael,
Thanks for your reply, I see. May you elucidate why you expect it to be hard to resolve? I ask only out of curiosity. Thank you.

Hi 3image,

re:
> May you elucidate why you expect it to be hard to resolve?

It's because it is a very complex area of code and some things like adjusting tolerance values for what to consider intersecting or not intersecting can produce side effects like causing other things that used to work to now fail.

So it's in an area that would need first a huge intensive effort to properly figure out exactly what is going wrong, and then followed by another huge intensive effort to test a whole lot of other cases to make sure it doesn't cause regressions in other cases.

Basically it's not going to be the kind of thing where it's a typo or simple logic error.

- Michael