I am modeling a part which I thought would be easy. I am running Rhino 8 on a MacBook Pro M5 with 27" monitor. After the boolean union I analyzed for naked edges (resin printed part) and found them. I tried to fix them manually w/o success. If you look closely at the edges of the union there is distortion in some views but not others. I am wondering if there is an issue with the halves being mirrored?
I have had boolean issues in the past but was able to fix. Any solutions would be very much appreciated.
where are the naked edges you re talking about ?
how can I / we repeat your error ?
you are using
_showEdges
to show them ?
Did the naked edges only show in the exported mesh ?
is the step of moving and boolean union missing in your file ?
I guess the naked edges could occur if the two adjacent red structures are pushed together and unioned.
But the problem is that all three red structures are invalid polysurfaces and are not cleaned up by exploding and joining. More investigation will be needed.
EDIT: The highlighted ten surfaces are invalid (note the gaps in them):
After deleting the invalid surfaces and capping the two polysurfaces to close them as good, I moved them together but attempts to union the pair fail in both R8 and the Beta. I don’t know why, they appear to generate complete intersections.
However by removing the adjacent faces, leaving open polysurfaces, it was possible to join the two halves into a single, closed, valid polysurface:
When I did a boolean union of the two pieces I ended up with a few voids and some distortion. Naked edges were revealed with analysis. I tried to fit surfaces to the voids w/o success.
How are the invalid surfaces found? What are R8, Beta?
the two parts on the right are to be moved together. After the union the seem in the middle is the big issue. At the top of the seem there is some distortion before and after the union.
analyse
_selbadObjects → will show the objects are bad
_what → to see if edge tolerance is an issue (it isn t)
_explode → ( _join _selbad → not solved _undo)
_selbad on single surfaces will show those invalid surfaces.
repair
(get rid of the bad surfaces)
_explode
_selbadObjects
_delete
_join
_cap (you re lucky its only planar openings)
boolean
_mirror
_booleanUnion fails
why ?
_intersect → will show open curves and the spot where boolean fail:
the gab here is 0.0006057 but your document tolerance is 0.001
the rule of thumb: smallest detail should be 10x document tolerance
as @jeremy5 wrote - looks like joining (“closest edge”) works - instead of booleans work here - and use different approach then booleans that rely on intersection.
fix it:
_extractSrf or sub-Selection → delete the faces
_join
_mergeAllCoplanarFaces to reduce the amount of faces and edges
Apologies if I am stating the bleeding obvious, but as I was wondering how you ended up with those troublesome tiny gaps it occurred to me that one factor might be scaling your model parts.
If you are not doing so already, I would recommend working in full-size units so you can take dimensions directly off plan and/or use physical dimensions. Even if just modelling from photos, estimating lumber sizes makes more sense if you have a range of real-world dimensions to use.
I would also keep beams, planks etc as individual elements as that will make it easier to adjust or correct things as necessary.
Once you have a completed structure you can copy the whole thing to a new file (to preserve the individual elements in the original file for future changes) and then boolean the various parts together, if you need to, before finally scaling the whole thing to its miniature dimensions.
