Choice of Software Link to heading
The biggest challenge was settling on the proper tool to perform CAD-ing on. After significant defliberation, I decided that I should try AutoCad. Of course, oftentimes reality is not as kind and permissive as one would wish. In much more concise and less philosophical terms AutoCad is only supported for Windows and my laptop only has Linux. This is not a problem unique to Autocad, but rather the whole ecosystem of CAD-ing tools is mostly comprised or proprietary tools, with open source community tools being subpar. One of the main exceptions to platform dependence(but not to the proprietary aspect) is Onshape.
Yet, I still decided to stubbornly use AutoCAD by taking advantage of the GIS Data lab in Rotch. Obviously, this decisions delayed my project by quite a bit since the Data Lab is open only during some quite limited hours.
The design process Link to heading
My first idea for a parametric construction kit was to move away from a clasic system based on cartesian coordinates and to move towards something non-standard. At first I was quite intrigues with hexagons and attempted to create a set of hexagonal shapes that could be linked together by joints of different angles.
Yet, after attempting to design something out of the hexagonal tiles, I quickly grew disillusioned with their usefulness. So as to not abandon this idea completely, I decided to pivot towards another quite regular shape: pentagons. My primary reason for choosing pentagons was their ability to construct regular dodecahedrons. As a fun math fact: It is impossible to make polyhedrons from regular polygons with more than 5 edges.
For those interested the cad file can be found here: https://jmp.sh/usnGj9r2
The fabrication Link to heading
At first I tried laser cutting, yet the joints wouldn’t fit together because my sizes got messed up while I was exporting my work. Instead of going back to Rotch to readjust my files, I decided to instead empirically rescale my design over and over with the limited software available. Below we can find my attempts at repeated resizings:
Size iterations
After a while, it finally worked and I started the actual construction. All went okay, until I met the first problem:
Partial build
An experience observer might notice the problem. There is no way to actually place the next pentagon without bending the joints. Fortunatelly, cardboard is quite flexible (and as a consequence allows for poor planning on the part of the designer). I was able to brute-force my way and forcefully bend the joints in order to make the next piece fit.
The final product looked like this:
Finished build