Trying to remove portions of a mesh, but Mesh Boolean will not work

I am currently working on a cosplay. I am trying to remove a decal from a gauntlet so I can 3d print its design separate from the rest of the gauntlet which will be built with EVA foam.

Unfortunately, the decal I want to remove intersects with the rest of the model, so I have to trim off the overlapping portions for the design before I print it, or it otherwise will not fit over the gauntlet like it does in the model.

I assumed that in order to do this, I would need to mesh boolean subtract the gauntlet from the decal. (everything in the file is a mesh) I figured it would take off the intersecting portions of the decal, as some tutorial videos I watched showed it working this way.



The mesh boolean subtraction did remove the portions of the gauntlets the decals intersected with, but the intersecting portions of the decals remained.


Mesh Boolean split does not work either, in fact, no observable change actually happens to either object when I use this function, as far as I can tell.

Why isn’t this working? I presume I should be using another function command, so what should I be doing instead?

Sincerely,
Nicholas

Hi Nicholas,
I am sure other people will have some solutions for this tricky task,
Or … you could download the Rhino v8 WIP and try this,… Shrink wrap…it is awesome !

@Nicholas_Watson-Titc to me it looks the mesh object is not closed, if you can share the Rhino file, I will have a look.

Here is a copy of the file I was working in. You may have to explode the model before you can edit individual parts.
Gabe FOREARMS with decals.3dm (3.3 MB)

something like this?

this was done in v8 where the tools have been improved, but I used all standard mesh repair tricks. Nothing fancy…and only about 15 min of fiddling to get here.

boolean split worked better than trim, and it helped to make all the bits watertight first by extruding open edges then capping. I also used patch single face to break up the holes into more manageble regions.

fwiw your gauntlet wouldn’t render right until
Gabe FOREARMS with decals_kfix.3dm (3.9 MB)
I ran rebuild mesh on it.

file in v7 format attached-

this video may help-

1 Like

Thank you! I will watch your video for more information should I have the issue again. My internet was fussy today so I couldn’t check out Rhino 8, but I will soon, it looks really cool.

@Nicholas_Watson-Titc Any chance you’d give me permission to use this model as a subject for a workflow video? It’s a great case with issues that a lot of users repairing meshes would encounter. I think it’d be useful to document the repair process.

1 Like

I have no problem with you using the model in a video. However, this is a portion of a larger model which I did not make, so I would advise asking the actual model maker, who goes by Dot Flare. His Art Station page is linked here.

ah, ok thanks- probably not a good candidate then.

At work we have a new piece of digital fabrication equipment. I am tasked with figuring out the workflow of preparing files for it. Increasingly the files are meshes of unreliable quality from outside clients. Shrinkwrap and all of the re-meshing, analysis and editing tools have helped me so much.

Once the mesh is nice and watertight, the next step in the process is slicing up the model for machining. For that I’ve gravitated toward using booleans. This week I’ve been testing out the new mesh booleans in Rhino WIP. They appear to be more reliable and well organized than they used to be. I really appreciate them. The way MeshBooleanSplit works in Rhino proper is great for me because it caps all the openings.

I would like to be doing this in Grasshopper though, because we often need to try a few iterations and I need to name and organize the parts, which I like to do with Elefront (and now the new Rhino attributes : )) It doesn’t seem to work quite the same way in GH! The split component is reliable but leaves me with many holes to fill. The boolean difference component sometimes does the right thing but I can’t quite predict its behavior.

The file I used to take these screenshots ended up as about 100 pieces so it is meaningful to streamline the steps.

I guess my question is: Should I expect to be able to do this in GH? I’ve looked around the Forum a bit and haven’t quite found a way to understand what’s happening.




if you get weird results from a boolean, typically speaking it goes back the normals…

make sure all your splitting surfaces are set up with the same normal direction.

for instance if the split on the right has normals facing out, and the one on the left has normals in the opposite direction and you use both surfaces to split at the same time, you’d get a weird result like this.

Thank you! I partially understand that from reading another thread and did use a single surface during this split, I also experimented with flipping the normals. Attaching my file in case you have time to notice what I’m missing. :pray:t3:
Boolean Test 2023-06-13.gh (407.8 KB)

@keen can you post a sample that doesn’t require plugins?

@Gijs
Boolean Test keen 2023-06-25.gh (391.4 KB)
I scraped out all the Elefront bits, sorry I didn’t think of that before.

While I was working I accidentally opened this in Rhino 7 and there, without the Elefront components, the mesh differences had results that were what I would have expected, so that’s already a strategy that could help me.

hi @keen thanks for the sample. I tested in both Rhino 7 and Rhino 8, and see there are differences.
RH-76040 Mesh boolean difference gives open result

1 Like