3D scanning & printing

Week 4

Printing

Grid of print tests
Top view

I wanted to find out more about triply periodic minimal surfaces, so I looked into generating them in Rhino / Grasshopper using the excellent plugin called Millipede. I found a tutorial and several tips online as to how to start going about creating these mathematical surfaces, and I had to tweak the definitions to find either a symmetrical (rather than random) outcome or a single module.

Gyroid for sale on Shapeways for $28

Millipede makes isosurface generation very quick and simple, and at first I was just playing around with the parameters to test randomness versus order as a starting point.

Random isosurface
Symmetrical minimal surface
Offset mesh for Zcorp printing
Cleaning up Zcorp prints
First round of Zcorp prints

I kept the meshes constrained to a 4" x 4" x 4" cube with a 4 mm wall thickness, but some of them were scaled down to half that size. The walls still held up (quite thin for this geometry!) and the prints all came out fine. I had to remesh some of them (I did multiple passes of smoothing in Grasshopper) to reach the output geometry, but some of them still had less smoothness than what I was hoping for. My second pass of prints had slightly different minimal surface geometries with color.

Gyroid
Color print with faded colors
Another color print
Same print after finishing with superglue

Scanning

Scanning stages (without and with small reference objects)

I tried using the Sense scanner at RPL, and I quickly realized that I had to test different methods for improving the scan. Initially, I placed my small 3d-printed object (from above) on the floor, but the scanner would keep losing track of it and stop short of scanning (basically) anything. Paloma told me that I could try placing small objects around mine so that the scanner has some reference items to detect orientation / location, and this method worked much better (as you can see in the last two images of the GIF).