One important thing to note before we actually get into the CNC machining, molding and casting process, is that starting earlier is sooooo important. It allowed Gentry and me to have more time to fix our mistakes, and we were able to produce a higher fidelity model at the end of the project.

Here is what the fox looks like in the end.

And here is what we started with, the stl file of fox.

We divided our process to three steps: CNC machining, molding, and casting.

CNC machining took us the most time. We split the fox in half in Fusion 360 along the plane containing the two ear tips and lying perpendicular to the bottom surface. This split allowed us to retain as much of the the original shape as was possible. We then followed online tutorials to create G-code for machine processing. We were warned beforehand that line M7, M8, M9 and G54 would need to be deleted. We also disabled the coolant to avoid unnecessary code and possible machine code-reading problems.

We watched closely during milling for anything that could go wrong. The three view perspectives were pretty helpful in determining where exactly the tool bit was and where it would go next (like this picture below, after the G-code had run for a while).

Here is a video of the CNC machine running.

These two wax blocks below are among our first cutting trials.

However, although there were no problems associated with milling these two blocks, they were not suitable to use for casting. One problem was that the stock offset distance between the fox shape and the pocket walls was so small that there would have been unwanted holes in our OOMOO mold (thank you, Dr. Wettergreen, for pointing this out!) At the same time, because the blocks were very small, a lot of the fox’s features were lost. We therefore decided to scale up our design and use larger blocks of wax.

Here is a picture of an OOMOO mold we made from our original, small wax blocks. You can clearly see that there is a hole in the bottom of this mold, a result of the stock offset distance being too small.

For our second attempt, we scaled up our model, created new toolpaths, and generated new G-code in Fusion 360. (pictures below)

We poured OOMOO 25 over these two blocks to make the mold. To prevent the OOMOO from leaking out of the mold while it was solidifying, we placed wooden panels around the open sides of each block.

And here are the two mold halves.

Since we did not make a key to keep the two parts perfectly aligned, we had some difficulty matching them. We minimized any inaccuracy by poking small dowels through two opposing corners of the mold. We also cut out a pouring hole at the very bottom. Before we poured our casting material, we stretched rubber bands around the two halves to ensure no leakage occurred.

We first made a clear-colored fox using Smooth-On Smooth-Cast 325. Below are the pictures of front view (left), side view (middle) and back view (right) of this fox. From the front view, it is clear that although the ears and bottom of the fox did line up with minimal deviations, there is one section in which the front and back do not match up completely. Gentry and I think this happened because the fox is an irregular shape and there is no way to cut it into two halves without one of the pieces requiring an “underneath” cut, impossible to perform on a mill. When we split the fox into two parts and created the toolpath, the CNC machine was not able to recognize this cut and would simply only trace the outer surface.

Although we 3D-printed a fox beforehand to gain a better perspective for where to place the split plane and pour hole, we did not discover the hidden negative cut angle, which resulted in this unnatural look. If we were to do the project over again, instead of 3D-printing a whole fox, we would print each half separately, allowing us to more easily discover mistakes.

This picture of our second fox shows the same problem:

We solved this problem by using sandpaper to remove the flat areas.

Here are all the pictures of our finished foxes:

We had a fantastic time working on this project, and in the entire class. We’re both looking forward to ENGI 300!