I noticed something strange. To reproduce it, make a 10x10 box (height doesn't really matter) and two 5x5 boxes inside of it so that the corners touch:



Then subtract the smaller boxes from the bigger box. The resulting object isn't a "Solid" but a "Joined srf":



Is this a bug?

Send your 3DM files for test!
That win some times to modelize!

Yep, this happens. Probably because the central point is going crazy with two boolean ops on top of eachother. Might be a problem with the operation and not the program per se. You can recreate this even if you union the two boxes and the base instead of subtracting them. This might not be a watertight volume as the vertical edge in the middle is being essentially removed by the boolean sub. Might be wrong though!

> Send your 3DM files for test!

Sorry, I thought this was trivial. Here it is. :)

> You can recreate this even if you union the two boxes and the base instead of subtracting them.

Oh, right, that's even easier and it leads to the same result.

Hi Andre, so that is not a bug exactly, it's more of a type of situation that you should avoid.

To have a well formed solid, each edge of the solid must be connected between only 2 surfaces.

In your case the location here there are 4 surfaces touching each other at a single edge:




That forms what is called a "non-manifold" edge in that location. That's problematic for various reasons, it basically makes points on that edge ambiguous for volume oriented calculations.

So you would need to give either a little bit of space or some overlap between your cutting pieces rather than having them exactly graze each other in that manner.

- Michael