V2.1 Internal support not being created

[cdub]

Something has changed about the way support is automatically generated.

I have parts with voids modeled into them that in the past support material has been added to automatically. The voids are there primarily to control stresses and warping.
Previous versions had no problems with this. The factory file I attached 25Cube_4 is an open cube (4 sided) where support material is generated just fine. The file 25Cube_6 is a six sided hollow cube that no support material seems to be generated for.

This is really frustrating, as I'm trying to slice/print a new version of a part I have printed many times before, and suddenly this little thing doesn't work....

[Festus440]

Why not make the cube solid and reduce the infill percentage to a low values such as 10%, like this:

Am I missing something?

[TenKOhms]

I dont see the difference between wanting supports to be generated in a closed hollow cube you and adjusting the infill percentage on a solid cube? If you're not going to be removing the supports from the inside of the cube wouldn't it be easier to just print a solid cube with low infill?

[cdub]

The parts I'm printing are quite large and are composed of a variety of thick and thin cross sections. I'm printing in ABS, and I want the part to be largely solid (I typically print with 90% infill and extrusion calibrated to be "full"), for a variety of reasons. However, I have found through quite a bit of trial and error that putting these voids in the part greatly aids in reducing the warping. It is essentially a simple way to make a part with very solid/thick walls in all areas, yet having "sparse infill" in a few places where the part doesn't need to be solid.

The point is that this technique has worked for me in just about every version of Simplify3D since version 1.4 and now for no particular reason it doesn't....

Just for grins, I also found that if I add a 1 mm hole to the top surface, it causes the void to be supported. I haven't tried making the hole smaller and smaller to see at what point the slicer ignores it, but I bet it could get quite small.

[jimc]

i tried a couple of you r files and i see what your talking about. i have never actually tried that i guess because i wouldnt think that support would even apply to a hollow cube. the insides of objects are infill and exteriors are support. i guess they just changed how the support works. i dont really have an answer other than you will need to change your workflow a bit.

[TenKOhms]

Sorry, I'm not sure if I quite understand, but, if you wanted some areas of the cube to be printed with less infill, there is the option to print a certain height with lower infill, then print other heights of the object with a thicker infill. Hope that helps, if not nevermind me

[jwiede]

As far as an inverted four-sided cube is concerned, support would be required to hold up the (bridged) top of the cube. Supports aren't automatically "infill" just because they're somewhere within the boundaries of the outer surface of the model.

Consider printing a model of a dining room table positioned as it would be normally in a home: The area under the tabletop and within the perimeter of the legs would need supports, but as the legs don't form a "closed surface" under the tabletop, the tabletop supports aren't really "infill".

[TenKOhms]

As far as an inverted four-sided cube is concerned, support would be required to hold up the (bridged) top of the cube. Supports aren't automatically "infill" just because they're somewhere within the boundaries of the outer surface of the model.

Consider printing a model of a dining room table positioned as it would be normally in a home: The area under the tabletop and within the perimeter of the legs would need supports, but as the legs don't form a "closed surface" under the tabletop, the tabletop supports aren't really "infill".

from what i gathered, OP wanted to print a 6 sided, closed, some-what hollow cube.

[cdub]

The 4 sided and 6 sided cube files are just examples of the issue. They are a highly simplified representation of the problem with the slicer. The real models I'm having these problems on are much larger and much more complex.

There are many situations where you would want the behaviour I'm describing for the purposes of reducing total material usage and controlling warpage. Honestly, I'm surprised more people aren't having an issue with this, I just assumed everyone did that sort of thing.

[Festus440]

With all due respect, what you are describing is certainly not typical as far a I know. Sparse infill is of course designed to do what you describe: Provide internal support and reduce the mass of the plastic to reduce heat buildup and warping.

Automatically adding support to an enclosed hollow volume would be a bug. A bug that apparently S3D has fixed. What if you want to print something that needs to be hollow, such as a hollow ball?

It sounds like in your case, making all the walls thicker is not an option. Your best option if sparse infill won't work (without seeing the model) is to open the enclosed volume with a hole (as you have already discovered) or add the internal support you structure you want in the CAD model itself.