Design question - chamfers / rounded edges

Is there an easy way of adding chamfers, bevels, fillets or rounded edges on objects inside BS?
I mean other than using other objects as negative parts to cut the edges etc.
Maybe a built-in feature I haven’t seen yet?

@BambuLab, any chance this feature is on a to do list for new features?

I don’t think it’s possible in bambu studio at the moment.

Do you have any specific examples where it would be a good idea to do it in bambu studio, instead of doing it in another 3d program?

1 Like

The short answer is no.

This question comes up once or twice a month on this forum. Features like chamfers, bevels, and fillets belong in CAD, and that division exists for sound technical reasons. These operations rely on mathematical definitions that are straightforward to apply to true geometric entities. CAD systems work with analytically defined geometry - lines, arcs, radii, and surfaces with exact curvature.

Slicers do not work with geometry in that sense. They operate on meshes, which are collections of triangles connected to approximate a 3D shape. You can see this clearly if right-click and object in the slicer and use the simplify model function and enable wireframe view.

A common follow-up question is: if a slicer can import a STEP file, which is a geometric format, why can it not modify it geometrically? The reason is that slicers do not retain the STEP geometry. On import, the slicer immediately tessellates the model, converting all mathematically defined edges, arcs, and surfaces into a triangle mesh.

In the STEP file below, the geometry consists of clean edges, and the fillets on the top edge and the inside bottom of the cup are explicitly defined.

After importing that same STEP file into a slicer, the model is converted to a mesh. Those clean lines and arcs are now approximated by many small triangles.

This is also why converting an STL back into a usable STEP file is difficult. The original geometric intent is gone, replaced by triangle data that must be reverse engineered. That process is computationally complex, which is why tools that do it reasonably well command a premium. Fusion 360, for example, requires an upgrade to its higher tier license to access its prismatic conversion feature, and even that approach is imperfect.

2 Likes

I can see several situations.
-Quick edits on objects downloaded from MakerWorld. Lots of nice designs that are missing that last touch to make the designs almost perfect.
-I do simple designs directly in the slicer, would be nice to round off edges
-Not everybody who are comfortable in using a regular CAD SW.
-Using negative parts are easy if the object you edit have straight edges, but not worth trying on curved objects.

Thank you for the short and the long answer. Appreciate it.

But people are already using negative objects to create chamfers, rounded edges etc. and they get good enough results. So instead of adding “real mathematical operations on true geometrical entities” maybe we could get a library of objects that can be used as negative parts to help people out? Not looking to replace CAD SW but to get good enough results in some situations.

Personally, I am not a fan of bambu studio getting too many cad functions built in. But I can easily see other users could benefit from it.

I sometimes use Meshmixer to change small things on models I download from MakerWorld.

Load the .stl file into MeshMixer and use brushes to soften lines for example. Incredibly easy and and it works on all lines.

2 Likes

In addition to the comment above, Blender is also an alternative. Yes, it’s complicated, but you only need one function. The whole thing can be done in 15 minutes with a tutorial.

On the left is the original, on the right is the rounded version.

It’s best to get this plugin, which can also check your model for errors and has some useful functions for 3D printing. It saves you a lot of work.

I’m not a big fan of trying to cram everything into one program. Above all, there are technical reasons for this, and what seems simple may be difficult to integrate. Bambu Studio, for example, can’t cut properly, except for a flat plane cut, because that’s also an extremely complex topic. At some point, you end up with a program that can do everything, but nothing properly.

It makes sense that there is separate software for slicing, CAD, and meshing. What sounds like “rounding off” actually requires some complex calculations. Ultimately, the triangles and polygons are manipulated and, if necessary, additional ones are added.

1 Like

That is a fair point. The difficulty is not the concept, but how such objects would be applied in a reliable and general way. A frequent request here is the ability to align objects to surfaces, centers, planes, or edges. Without those primitives, reusable negative objects are fragile and depend heavily on manual positioning.

If such alignment existed, it could theoretically be extended to bounded negative shapes with parameters like fillets or chamfers. Herein lies the real challenge. Applying these operations to simple boxes or cylinders is straightforward. That is not how users actually want to work. They want to select an edge and let the software determine the affected boundary. At that point, the problem moves from simple Boolean subtraction into edge detection, topology analysis, and complex geometric logic.

This brings the discussion back to scope. A slicer’s primary job is converting a model into toolpaths, which is already nontrivial. Adding CAD-like feature construction significantly increases development and support complexity without improving print quality or output fidelity. This makes it a quality-of-life feature rather than a core functional improvement.

It is similar to asking a plumber to move electrical wiring. They might be able to do it, but it is outside their core role and adds risk for limited benefit. These capabilities already exist in CAD tools with mature geometry kernels and workflows. Given that, duplicating them in a slicer is hard to justify beyond very limited, tightly scoped cases.

1 Like

Thanks, I will consider Meshmixer and Blender for simpler tasks.

The idea was to avoid exporting an .stl, editing it then importing to BS. Then slicing and checking “quality” and perhaps going back to Meshmixer and to do some changes again, then check in BS again etc.
Depending on situation, I will either use negative parts or export to edit in other applications.