INTRO

This is Liam McConnico-Blanchet and Mustafa Latif, and for our Final Project, we chose the king chess piece from Code Geass to make our polyurethane casts.

Gate 1:

For the first gate, we were tasked with choosing a piece and 3D printing it to the scale of our final piece. We chose the following file on Thingiverse: https://www.thingiverse.com/thing:943198/files.

With the final piece needing to be ~3in., we scaled the file to a final height of 3.1” and 3D printed it on the Prusa. Here is the 3D printed model of the piece.

Gate 2:

Now with our chess piece selected and printed at the correct scale, we needed to process the STL file, cut it in half, and attach it to the mold base we would be using to create the silicone negative molds. This final file would become the template for 3D printing the first positive mold half.

First, we uploaded the STL file into Autodesk Meshmixer to process and cut the piece in half. We modified the file to ensure that the polygon composition of the piece contained less than 5000 triangular faces so that the printing process would not take too long. Next, the piece was cut in half down the vertical symmetrical plane. This shape was exported into SolidWorks CAM.

Next, we downloaded the mold base files from Canvas that corresponded with the sizing of our piece. By following the simplified “plug and play” method, we created three coincident mates to perfectly align our chess piece half with the mold base template. Now that this file was complete, we exported the STL file and printed it on the Prusa. Our first mold half was now complete.

In preparation for making the silicone negative mold half from the 3D printed half, we needed to calculate how much silicone was necessary. Below are the calculations.

Gate 3

We then got to forming the other half of our positives by using the CNC machine to try and mimic our 3D-printed positive through a wood block. To do this, we had to first import our 3D printed file into VCarve so that the Shapeoko CNC machine could properly cut it. This ended up being a pretty arduous process as the video tutorial showing how to use VCarve used a slightly outdated version of VCarve, meaning that there were a few differences in the processes that were challenging to fix. However, after some Googling, we were able to form our wooden positive.

We intended to use a 1/8″ end mill for roughing and a 1/16″ ball nose mill for finishing. However, due to a mistake during the final VCarve step on downloading the g-code, we ended up accidentally combining the code for the roughing and the finishing into one passthrough. This meant that instead of having one step of roughing and one step of finishing, the Shapeoko used the 1/8″ end mill for the entire process. As such, not only was the wooden positive not detailed enough for us to use, but since the 1/8″ end mill had a larger radius than the 1/16″ ball nose mill, it had accidentally shaved off 1/16″ off of every corner, edge and surface of the wooden block everywhere it went. That meant that we couldn’t just go over the new wooden positive with the 1/16″ ball nose mill since the piece was already much too small. The image below shows how the 3D printed positive is smaller than the wooden positive.

So, we had to restart and do the whole process over again, ensuring that the g-code didn’t combine, and 2 hours later we had a complete wooden positive.

We then got to silicone molding the negatives by using the positives. To do this, we hot glued vertical cardboard slabs to the edges of the positives, to form an airtight space for the mold to go in above the positives. We then combined equal parts of two chemical reagents, at 80ml each, to form a silicone molding mixture, that we measured out so that when we poured, it would finish 1cm over the height of the original positive, so that we would be able to take it out.

After waiting a while longer, we got our silicone molds ready! One snag we ran into during the silicone molding process was figuring out how to remove the mold from the wooden positive, since it had essentially formed a vacuum within the hole of the wooden block, making it very difficult to pry out. Additionally, the sides of the wooden positive for whatever reason led to the silicone mold having sides that almost exhibited the properties of glue, making it very hard to remove. After a little pulling and tugging, we were able to get it out and get to polyurethane casting.

Gate 4

We then got to polyurethane casting, which involved setting up quite a complicated contraption. We had to figure out a way to get the molds to stick together without them being too squished as to warp the actual chess piece we wanted to make. However, we couldn’t squish them too little or the polyurethane would leak through the molds. To start, we poured a mixture of 25ml of both 300 A and 300 B to pour into our molds, combining a small amount of white dye into the 300 A to make a white chess piece. Luckily, we were told that the pot life of the 300 mixture was more like 30 seconds as opposed to 3 minutes, so we made sure to act very quickly.

Our initial casting method was to have two wooden blocks clamped onto the table, with the molds in between, to create light pressure on both sides. When we casted, all looked like it was going well, except for the fact that we had entirely too much mixture, leaving us with a huge hot puck of 300 at the base of the mixing cups. When we removed the molds, we noticed that the mixture had leaked through and we had to trim the edges, so we decided to tighten the molds the next time around.

We then tried to clamp the wooden blocks together with the molds in between. While this was a tighter fit, we ended up with a lot of holes in our final piece, and we figured this was due to us not being able to stir around the mixture in the mold since it was clamped down. Luckily, our amount of polyurethane was much better, at 7.5ml of both A and B.

For the black chess pieces, we instead used a rubber band approach to combine the molds together. We also opted to just hold the molds in our hands instead of place them down on a surface so that we would be able to stir the polyurethane mixture around in our molds. This worked a lot better, and made sure that we covered all the points in the crown of our chess piece and minimized the amount of holes.

What we learned:

We learned how to properly silicone mold and cast using polyurethane. We also learned that it’s important to double-check that instructions were followed all the way to the end (in the case of the CNC machining). We found that having points in our pieces was a major challenge as the polyurethane would sometimes not properly coat the points. We also learned some post processing techniques for getting the polyurethane pieces to look like they didn’t come straight from a mold.

COST ANALYSIS:

Labor cost ($15/hr) = $165 for 11 hours of work
PLA Filament ($0.02/g) = $1 for ~50g printing
1.5”x3.5”x6” wood block = $3
Silicone ($26/48oz) = $5.86 for 320 mL used
Polyurethane ($30/pint) = $7.5 for 100mL used
Urethane colorants ($30 kit) = negligible cost for our 4 pieces
Plastic mixing cups ($0.30/cup) = $1.20 for 4 cups
Popsicle sticks ($0.05/stick) = $0.20 for 4 sticks