Reference/Inspiration
For this project we were challenged with applying parametric design to the casting and mold making process. I found this uniquely challenging, however I decided to focus on natural mathematical phenomena to center my project around. I decided early on that I wanted to use the Fibonacci sequence, found in countless natural shapes and growth patterns, and eventually decided I wanted to create my own parametric spin on the nautilus, which is found in various animal shells and flowers.
Fibonacci sequence:
1, 1, 2, 3, 5, 8, 13, 21, 34 ...
Process 1 / 3 - Modeling
The initial design process of this model was difficult because of Grasshopper's learning curve. I tried using various other online source codes for building my model, but none of them did exactly what I wanted. None of them were parametric 'enough' for me. Most of them simply used the Fibonacci sequence object built into grasshopper, made it 3D, and then built a logarithmic defined shell around the curve.
For my project, I wanted as much of the object as possible to be defined in some way by the Fibonacci sequence. For this I decided I wanted the height, width, and even resolution to be based on the sequence of numbers. I started as with other examples, with the simple Fib object, which spits out numbers in the proper sequence, and tied it to a curve rotating around a center point. Then I moved that curve up along the Y axis with an adjustable multiple of the original sequence.
 |
| Base Curve |
Next, along this curve, I placed perpendicular frames, which I later defined as circles. Through a complex series of calculations, I based the width on a multiple of the sequence, but adjusted it along a 4 point curve using the mapper function, scaled between the maximum and minimum values. I tried it without scaling, but it didn't look as appealing. The number of cross sections is completely adjustable, as well as the non linear scaling.
 |
| Cross Sections |
Along each of those cross sections, I placed polygon curves, with the number of polygons, and the number of faces per polygon, based on the Fib sequence. The largest ring would have a number of polygons equal to one of the higher numbers in the sequence(34), but a number of sides equal to the lowest (3). The result is layers of polygon curves which get less dense, but higher resolution, as they approach the peak of the structure.
 |
| Circumscribed Polygons |
The final step was to loft each of these circumscribed polygons to create intersecting pipes. This required some adjustment of the radius of the polygons, which ended up being a similar curve to the one used for the radius of the cross section rings. Everything was fine tuned and adjusted to create a solid mass with each "ring" intersecting enough to print and cast as a solid object.
 |
| Loft and complete script |
Along the way I had several iterations, and a very interesting accidental model I chose to save for possibly a future printing project.
 |
| Iteration 1 |
 |
| Iteration 2 |
 |
Iteration 3 |
 |
| Iteration 4 |
 |
| Oops - Grafting the loft data? |
 |
| Final Iteration |
Process 2 / 3 - Printing
The 3D printing process isn't something I usually deal with, but grasshopper ended up being such a hurtle that I wasn't going to have enough time to print from Shapeways, so I borrowed a friends Monoprice Select Mini, which is a $200 3D printer, and tried to print it myself. This also turned out to be a serious learning experience, with finding proper layer height, getting proper bed adhesion, building an enclosure for an otherwise open air printer, and eventually developing a method for getting semi-successful prints. I wasn't the happiest with my final print, but it worked fine for the casting process. It was sealed with Rust-Oleum Triple thick glaze spray-on clear coat.
 |
| Failed test print (filament got caught) |
 |
| Learning about skits vs rafts |
 |
| First Success! |
 |
| Built an enclosure |
 |
| Trying it out full sized |
 |
| Added a blanket to keep in more warmth |
 |
| Support material broke off, but eventually self corrected |
 |
| It was a mess, but 12 hours later it finished |
 |
| Self correcting support material |
 |
| Even with all the glue, the interior support didn't stay down |
 |
| Final Prints |
Process 3 / 3 - Molding and Casting
I decided to try and cast both the interior and exterior of my print, so I could save on casting material, and have a more exact copy. It worked out, after a small amount of trial and error (I ruined a couple molds, but hey, you live and you learn). The final casts are full shell shapes, with unique interior and exterior forms. I also cast the small print for good measure.
 |
| First cast, of the small print |
 |
| The Molding mess. Using a bucket of water to hold my model down, in an attempt to repair a mold. |
 |
| Iterations, getting better at measuring material for just the shell. |
 |
| More iterations |
 |
| Final Collection |
 |
| Makes a nice cup |
.
No comments:
Post a Comment