Hi all,
I am still on my Bubble House project (mock-up model below)
Now I’m trying to merge the spheres with the “fillet” parts (all well-defined closed breps with no issues whatsoever), but I quickly run into “boolean union failed” issues, even if I lower my precision to an absurd amount like 100 units in distance and 120 degrees in angle.
Is there an easy way to fix this?
Thanks in advance
For_Forum_BU.3dm (6.3 MB)
that depends what your further steps are, if you want to keep working in nurbs you have to find a way that the surfaces can clearly intersect each other which looking at this constellation might be a bit of a work out.
if you skip the nurbs from here. honestly i would not even have started with nurbs, though i am a stiff advocate of nurbs usually.
to fix what you have for printing, for instance use Shrinkwrap, but you will have a mesh from there on.
another option:
_mesh
_shrinkWrap
_quadremesh → to Subd
optional _ToNurbs
(Edit)
like this:
finetune with curves / to get proper circular areas.
There is no easy way to do this. Even programs that are good at making fillets will likely choke on this. Booleans are going to never work. That’s a guarantee. Making the fillets that you have into solids is a mistake and just makes things much worse.
You can make all the necessary fillets with Rhino either by using FilletSrf and FilletSrfCrv. You have the right idea of connecting the spheres with fillets with large radii. Then the fillets around the capped openings can have smaller radii that lap over the larger radius fillets. You can’t put a big fillet over a smaller one so making the big ones between the spheres first is the right idea.
After making the large fillets that connect the spheres and trimming the spheres, you need to make the smaller fillets that wrap around the openings and overlap the large fillets. Many of those fillets will be made with FilletSrfCrv. In the enclose model the red and purple fillets are made with FilletSrfCrv. The light blue fillets are made with FilletSrf. The base surfaces need to be trimmed. I recommend testing to see if things trimmed properly by joining and then undo the join and fix whatever did not join. Leave the model as individual surface until the end for the final join. Joining permanently alters edges so you don’t want to do that until you know everything will join properly.
I did some of the spheres in the enclosed file: The red Fillets are made with FilletSrfCrv. The light blue are made with FilletSrf
For_Forum_X.3dm (832.8 KB)
The rest of your model is hidden.
The biggest problem you will encounter is all the bugs in FilletSrfCrv and FilletSrf. All the fillets made with FilletSrfCrv needed to be fixed. Usually the fillets were incomplete ending just slightly too short or slightly too long. Do not use the trim option in FilletSrfCrv. Most of these fillets need to be matched with the next fillet because the ends are not correct. Since the fillets are incorrect the automatic trimming is also incorrect and just makes more work to fix that later.
FilletSrf also has bugs. Several of the Fillets are supposed to end in a point. FilletSrf fails to do this. This requires fixing manually. If its not fixed then trimming will fail and then joining will fail.
All of this could be done pretty quickly if the bugs were fixed, but since that is not the case its pretty long hard process that takes more than 10 times longer than it should.
