Ready for 3D printing

Replies to This Discussion

Permalink Reply by Birk Binnard on March 20, 2016 at 1:52am

It depends on the shape of your surface and the shape of the pattern. You pretty much have 2 choices - you can either extrude the pattern above the surface, (Boolean addition) or submerge it below the surface (Boolean subtraction.) I find extruding works better, but it really depends on the specifics of your geometry.

You will also need 2 surfaces - inside and outside - so you printed part has some actual thickness to it. And you have to be sure to join the edges of your surfaces to avoid so-called "naked edges" - slicing software will typically choke on a naked edge.

Permalink Reply by Ricky Huang on March 20, 2016 at 5:33am

can you show me how you add some thickness on it ? 

i have attached the file

Attachments:

• 20150617_Surface_BoxMorph_BWD3BA.gh, 24 KB

Permalink Reply by Birk Binnard on March 20, 2016 at 12:29pm

You have created an interesting piece of geometry, but I don't think it can be produced using a standard 3D printer. The reason has nothing to do with thickness - although there is a way to resolve this. The real problem is the geometry has several problems that would make it extremely difficult to actually print:

1. There is not enough contact area between the geometry and the print bed, so the part would fall over during printing.

2. The depth (front to back) is large enough that the part would need a huge amount of "support material" to hold up the loops of plastic, and this would result in an exceedingly difficult cleanup process.

3. There is not enough contact (actually none) between the rectangle and hexagon pieces, so the whole part would probably just fall apart.

The attached screenshot shows what your geometry looks like after being sliced for 3D printing. To make this image I baked your geometry as-is, exported it as an STL file, ran that through the 3D Builder program to correct errors, and then sliced it with the Craftware slicing program.

The hexagon and rectangle pieces are made solid because the all have naked edges and the 3D Builder software thinks they are supposed to be closed "tops" and "bottoms". Without the 3D Builder corrections the resulting geometry is still sliceable, but the results are nothing like your original geometry.

The way to resolve the naked edge issue is somewhat complicated. You need 2 congruent shapes for each hexagon and box object - an "inside" and an "outside". I usually use the Scale function to generate the second shape. Then use the BRep Edge function to find the surface edges for both the inside and outside shapes. Use the Naked Edge outputs to make a Ruled Surface (flat) front and back surface that connects the front and back naked edges.  You want to end up with a closed BRep (no naked edges) for each such part, so it may be necessary to use the Boundary Volume function to get that.

The attached GH file shows the general procedure for what I described, but for a far simpler object. Since the part's geometry is generated directly from an edge curve I did not have to find the naked edges - I already knew what they were.

I have never used the SBox function to generate geometry, but I will see if I can find a way to come up with inside and outside surfaces for that.

Attachments:

• Capture.JPG, 213 KB
• Polygon Twist.gh, 16 KB

Permalink Reply by Birk Binnard on March 20, 2016 at 2:31pm

Here is a "fix" for the thickness issue. It is messy, partly because you used different methods to define your 2 shapes, and partly because there are aspects of "curve direction" I do not understand. The Loft function is very sensitive to curve direction, so I took a brute-force hit/miss approach to deal with that.

However, these are minor problems compared to what happens when the results are given to slicing programs. I have 3 different slicers (Craftware, Simplify3D, Kisslicer) and they all fail miserably with the results from this GH file, even after 3D Builder's corrections are made. I'm not sure what the exact problem is, but I'm sure it has to do with the overall curvature of the part and the fact that the individual shapes simply touch and have no physical connectivity.

To actually build this I imagine bending flat stock and tack-welding the individual pieces together. But that is just a guess of course. 

Attachments:

• 2.gh, 28 KB