Development of A Fuller Moon

The 5 polyhedra

The 5 polyhedra

Fuller Moon, like most of our projects at SDS started as an exploration that had nothing to do with light, lighting, or any kind of product at all. I had a growing interest in geometry before I was familiar with the work of Buckminster Fuller.

In the course of my work over the years, I had often run into the problem of making 3D shapes from 2D planes.  Maybe it was just the way I usually went about making models, but I’m sure other industrial designers, and architects out there have found that using flat things (usually matte board or foam core) to make anything other than a cube, gets confusing pretty quickly.

Study 1: A tape + foam board icosahedron and dodecahedron, a tape and dowel tetrahedron, and a large icosahedron with 3D-printed connectors

Study 1: A tape + foam board icosahedron and dodecahedron, a tape and dowel tetrahedron, and a large icosahedron with 3D-printed connectors

That lead me down a long mathy road starting with platonic solids.  Tetrahedron (4 faces), Cube (6), Octahedron (8), Dodecahedron (12), and Icosahedron (20), these are the only 5 solids that can be made all of one shape of the same size where the same number of those shapes meet at each vertex.

I’m a very physical learner.  I need to do things with my hands to really understand them, so I went back to the straight edge + compass and constructed all the triangles and pentagons for the icosahedron and dodecahedron by hand high school geometry style.  I soon realized that assembling the last face of a shape is very hard if you don’t plan ahead.  Where do you put the tape if you want it to seal up nicely?  If I was having this problem with the paper version, it would be a problem with anything bigger and made of a less forgiving material…

The cutout pattern from a dodecahedron and icosahedron

The cutout pattern from a dodecahedron and icosahedron

Onto the next version!  I decided I could deal with something a little more complex.  I had build a geodesic dome for Burning Man before, so I knew a bit about the work of Buckminster fuller.  I attempted a v2 geodesic icosahedron with the same dowel and connector construction style as the big icosahedron so I wouldn’t run into the “closing” problem on the last face.  A v2 geodesic icosahedron is constructed by dividing the icosahedron’s edges in half and “pushing” those midpoints out onto a sphere that encloses it.  You can keep doing this over and over again, but v2 means you did that twice.  The result is a little closer to a sphere than an icosahedron, and the more times you do it, the closer the shape approaches a perfect sphere, but the more faces and slightly different triangles you’ll have to deal with.  V2 is a good starting point.  It only has triangles of two different edge lengths.

icosahedronSubdivision
V2 Geodesic Icosahedron with connectors 3D printed on a Makerbot

V2 Geodesic Icosahedron with connectors 3D printed on a Makerbot

This v2 sphere is very strong, but printing each connector is time consuming.  For my next experiment, I wanted to see if I could come up with a simpler construction method.

For the next sphere, I went back to matte board and tape.  This time put a gap in between each face so I could reach in and used paper strips to connect each edge.  Because geodesic spheres (and all enclosed shapes made of triangles) have self-supporting geometry, I wondered if using a flexible connector would allow the faces to bend to the correct angles and hold itself in shape only once they were all taped together.  That way I wouldn’t need the accuracy of the 3D printer.  The material would do the work for me.

And it worked!  It was a little floppy, but at least the shape resolved into a sphere.  I was proud.  It was late.  I tried to go to bed.  When I turned the lights off, there was still a small LED light on an old project nearby and I noticed SHADOWS.

Industrial designers now are always taught to 3D model by default, but I’m still a big advocate for physical prototyping and model-making.  It might be slower, but in the computer you’re ideas are isolated in the land of the screen where things don’t interact with them unless you tell them to and your sense of touch is eliminated.  Physical models lead to discovery over visualization of ideas you had before you even started.

The jump into explorations with geometry and light flowed easily.  The next one obviously needed to be made of mirrors.  Lots of mirrors.  All the mirrors.  The result was a little hard to comprehend.  Infinite reflections bouncing from face to face through the center of the sphere.  The gaps helped let light in which also meant it would let light shoot out.


Christopher Yamane is an industrial designer and Co-Founder of Super Duper Studio