I began using S3D last year and except for the problem being discussed, I'm generally very happy with it. I'll continue to recommend it for the things I believe it still does better than Slic3r and Cura, but it is very frustrating to see that S3D v3.0 still hasn't addressed the issue of poor infill structure design. It really does detract and leads to otherwise unnecessary modes of failure in some printed parts.
To those who say, "design for manufacture" - I couldn't agree more. However, I believe the manufacturing process itself
can be improved by improving the nature of the infill that S3D is generating. That S3D enables the ability to tweak infill is some mitigation; but for this particular problem S3D then fails to live up to its name of simplifying the printing preparation experience - especially when the slicers already mentioned generate ideal internal structure without any effort on my part, despite their other short-comings such as poorer quality of automatically generated support structure.
Here's a good example of the problem; a bicycle light clip, max xyz size: 31x33x18mm.
The problematic feature is the horizontal spring tab (2.1mm thick) which protrudes from the vertical wall (2.5mm thick):
In the next screen capture, the image to the left is indicative of how S3D does not extend any of the top (3) and bottom (3) layers from the tab into the wall. On the right, the single central layer
is extended into the wall, though to little overall structural benefit:
After only a very few flexure cycles of the tab, the lack of significant infill cross-over from tab to wall results in the tab breaking off quite cleanly, right along the wall-tab interface boundary.
By comparison, In those parts printed using Slic3r or Cura, when the spring tab fails it does so further out from the wall, in the tab itself:
That is something I can overcome, (designing for manufacture) by making the tab thinner.
- Hamish