solved: OCC bug - subtractive spheres fail in certain curcumstances

[uwestoehr]

During my work I frequently stumble over the problem that subtractive spheres suddenly fail. Now I could reproduce the problem with the attached file. I have no clue what triggers this and how this can be avoided. Maybe anyone has an idea?:

- open the attached file
- the tip is at the feature "Problem-Sphere" and as you can see this sphere is not correctly subtracted

To fix this I even created a new sphere but as soon as I set for its position the reference to the sketch dimension, it gets broken as well.
The strange thing is that my file dates back from weeks and all the time the sphere did not make any problem. Today I changed in the file only a helix parameter - so something that has nothing to do with the sphere - and it got broken and I cannot fix this. Even more severe is that my whole document is broken (all following steps that I omitted in the example file).

The only workaround I found is to set for the sphere a z-position of 0.001 mm. This fixes the sphere subtraction. This is weird. I think it is a bug but maybe it is a problem of OCC and if so maybe anyone has a hint for me how to workaround these OCC issues?

OS: Windows 10
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.18.16110 (Git)
Build type: Release
Branch: (HEAD detached at upstream/releases/FreeCAD-0-18)
Hash: f7dccfaa909e5b9da26bf50c4a22ccca9bb10c40
Python version: 3.6.6
Qt version: 5.6.2
Coin version: 4.0.0a
OCC version: 7.3.0
Locale: German/Germany (de_DE)

[openBrain]

Maybe some clues in this other thread...

[openBrain]

After opening the file, it is only the well-known tangency issue. Just add 1 µm to the radius of the sphere and everything runs like a charm.
AFAIK this is an OCC issue.
There has been another post with a very similar part & issue some times ago. Let me try to find it.

[uwestoehr]

After opening the file, it is only the well-known tangency issue. Just add 1 µm to the radius of the sphere and everything runs like a charm.

Hi openBain, many, many thanks!
It was hard to realize this. I thought I am clever and set the radius of the sphere using a formula. I added now a micron in these formulas and the issue is gone. It helps a lot since it also repairs my full document. (I have 7 different subtractive spheres in there).

So let's cross fingers that these kind of bugs are fixed in the upcoming OCC release

[Vincent B]

Surface intersections seem to disturb freecad with your design. I removed 0,01mm to the sphere to check that and no bug come up.