For our final project, we had to chose an object and then create negative and positive molds. We chose to make a icosahedron, a 20 sided dice.

To start, we had to make the model in a 3D modeling program. We started with a file that already had a  icosahedron shape. We decided to mill into a block to create a positive mold, then use that to make oomoo negative models, then use the oomoo negate molds to create multiple final objects. We would be creating a 2 part mold due to our shape.

We modified the given 3D file to add walls to bound the CNC machine, added holes and pegs to help us line up the 2 halfs of the mold, and added a funnel to help up pour material into the final oomoo mold.

Then we created GCode from that 3D file for the CNC machine to use. Fusion 360 was not working well on my computer, but we found an alternate program that suited our purposes. Some hand modification of the output GCode was required so the CNC machine would accept it, mostly line deletions.

We used a flat end mill bit for our code. Additionally, we performed 3 passes to create our mold. Pass 1 simply excavated material in a step down fashion. The results of that pass can be seen below. We used wood at first just to make sure the code worked. This took about 20 minutes.

Pass 2 was to quickly run the bit over the surface of the mold and increase the resolution of the mold. The results of pass 2 can be seen below. This took 5 minutes.

Pass 3 was another surface run, but the bit passed over itself multiple times to create smooth sloped surfaces. The results of pass 3 can be seen below. This code took 1 hour and 45 minutes to run, but could be optimized to take about 45 minutes.

We decided that the wood mold looked fine, so we used that to create our oomoo mold and did not rerun the code on a block of wax. We created temporary walls using play-dough, and poured the oomoo in, as seen below. Once the oomoo set, we removed the play dough and peeled the oomoo out of the wood mold.

The resultant mold can be seen below.

Soon we had 2 molds.

Then we started our process of making icosahedrons. We used fast setting plastic to create our shapes. We ran into a few problems when casting our shape. Our built-in funnel entrance was a little too small, and the molding plastic was a little too viscous, so we managed to trap air inside of our mold and ruin a few casts. The air could not bubble out like it could with water. By wedging a larger funnel into our mold, and slowly removing the funnel after pouring the plastic, we solved the problem. A pour can be seen below.

A timeline of our casts can be seen below, from left to right.

The oomoo molds after casting can be seen below.

Overall, this was a successful project. We made enough dice for us and for Matthew Wettergreen.