4
$\begingroup$

I'm trying to print a supporting base which will house the spindle for an electrostatic rotor. It's basically just a truncated cone with a hole down it's center to house the spindle.

For reasons that I cannot fathom, Ultimaker Cura keeps on adding an unrequested top/bottom layer (color-coded yellow in the screenshots) inside this hole, so instead of a single hollow cylinder of 10 mm depth, the result is a hole only a few millimeters deep with another hollow cylinder behind it.

Here is the intended model, note the open space for the hole at the top.

enter image description here

Here is the inner view of the hole being printed as expected:

enter image description here

Finally, here is a layer view of the print a few millimeters from the final top layer with the unrequested top/bottom layer that covers the spindle hole:

enter image description here

The STL file is on Github (with a built-in viewer).

Can anyone help me understand why this is happening?

$\endgroup$
  • $\begingroup$ Did you try to heal the stl in a program like MeshLab and slice the healed model? $\endgroup$ – Davo Aug 22 '19 at 16:23
  • $\begingroup$ No, I created a VTK file using Pygmsh and then converted it to STL using ParaView $\endgroup$ – Gearoid Murphy Aug 22 '19 at 17:07
4
$\begingroup$

It would appear that your model does not conform properly to STL standards. I base this conclusion on a couple of factors. When I loaded the model into Simplify3D slicer, it displayed fine, but when sliced, displayed nothing. Using the onboard repair feature, it presented the entire model as being composed of non-manifold surfaces.

Meshmixer's Analysis/Inspector feature also highlighted the entire model as flawed.

Another observation is that there is an extraordinary amount of facets/triangles/faces to this model. Nearly three-quarters of a million triangles for something that should be much simpler.

The most recent version of Prusa Slicer 2.0 presents an error message indicating that no layers were detected. This is peculiar indeed.

All of the above points to a problem with the source file or the software used to create it.

Please consider to add to your post the program you used or the source of the model.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I created the mesh using Pygmsh which saved it as a VTK file. I then converted it to STL using ParaView. I can provide the source later this evening after work but it does seem that my STL file is indeed abnormal and that I should probably use a canonical CAD program. $\endgroup$ – Gearoid Murphy Aug 22 '19 at 17:09
  • 1
    $\begingroup$ From my brief bit of research, I see that Pygmsh is a facilitator for gmsh, which appears to be founded strongly in coding. To that end, consider to check out openscad.org for a scripting type 3d modeling program. Your cone might be three modules, maybe four at the most. Two cylinders (base and cone) and another cylinder subtracted from the union. $\endgroup$ – fred_dot_u Aug 22 '19 at 17:44
  • $\begingroup$ OpenSCAD is perfect, so much easier than my previous approach! $\endgroup$ – Gearoid Murphy Aug 23 '19 at 15:06
  • $\begingroup$ I'm happy to hear that you like it. It's my first attack program, unless the part/model is severely organic in shape. $\endgroup$ – fred_dot_u Aug 23 '19 at 17:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.