Ice Cube Mold
The objective of this project was to introduce the class to mold making and casting principles. I worked with a partner and we made two ice cube molds out of food safe silicone. I had the idea to make engineering themed ice cubes and reached out to my partner because I knew she would be interested. We decided to make two trays, one that has standard ice sizes, and one with larger gears for cocktails. The idea began with gear ice cubes, but expanded to include multiple sizes and shapes to increase variety and make the project more interesting. We ran into a number of issues in the initial stages of the project, but I think that ultimately my partner and I were very happy with how the design turned out. Both of our molds were cast using Sorta - Clear 40. We had problems with parts shifting in one of our molds but the desired end result was still accomplished.
Gear Design
In the design process for the gear ice cubes, my first thought was to use Onshape's spur gear generator. After making the first iteration I realized that the complex teeth profiles designed to distribute load were not necessary for our use case and that a simpler design could help create a more durable mold and easier de-molding process. The gears were lofted to form a 3D shape with a draft angle of 3.6 for the large gear and 3.5 for the small gear.
Original Mold with large gears
Spur Gear Generator
Custom Teeth
Gear Casting
To cast the larger gears, we needed about 350 grams of our mold mixture. The mix ratio was 100A:10B, so we poured A into our casting container until we felt we had enough, then measured the corresponding weight of B in a separate container and mixed in thoroughly. Because we mixed in the container used for casting, the very bottom of the container had a hard time solidifying, indicating the bottom may not have been scraped enough.
This was the mold that we had casting problems with. I attemepted to weigh down our 3D prints to keep them level in the silicone, but underestimated the viscosity of the material. When I came back to check on the molds, the bars I was using as weights had fallen in, so I had to make adjustments to salvage the mold.
These gears took about an hour to design and 2 hours to print. Additionally, they were painted with XTC - 3D to smooth the 3D print, which took about 4 hours to dry. The mold itself had a cure time of 24 hours. Overall, for supplies and active time, this ice cube tray would have cost about $360. These molds adhere to DFM principles because they are designed for easy removal through gear shape and draft angle and involve little active time in their manufacturing.
Solidified mold with weight
Hammer and Wrench Design
I wanted to practice my surfacing CAD skills for the design of the hammer, so I made the model using a loft, a revolve, and an extrude. Once I had a base model that captured the curve of the handle, I created a sketch of the profile and lofted it with a 3 degree draft angle. The 3D print was about an inch thick, which allowed us to achieve our desired width without needing the 3D print flush to the surface of the mold. My partner designed the wrench with a 5 degree draft angle for easy de-molding of ice cubes.
.png)
Hammer Model
Hammer and wrench w/ draft angles
Practice Mold
Before making our molds with our food safe material, we made a smaller practice mold using OMOO. This mold came out very nicely but highlighted some ways in which we could improve our design to make the molding process easier. We designed our original 3D prints to be exactly the size of the ice cubes that we wanted, but this meant that they needed to be flush to the surface of the mold which was difficult to accomplish. We also had a hard time keeping the 3D prints level which led us to hot glue washers onto them to help with balance.
​
Final Mold
For our final ice cube tray, we laid out the formation of the 3D prints in the container, and I glued washers to the top while Sloan mixed the silicone. I had considered adding 3D print tabs to help with balance but it caused problems with print orientation and the washers worked quite well. After the mixture was poured and the prints placed in I added makerbeam weights to the top of the prints and under the string. This helped to increase tension in the string while keeping the prints pressed into the mold. The mold shown to the right came out extremely well and my partner and I were pleased. We found that the silicone remained flexible in the freezer and the ice cubes were easy to remove which satisfied our project goals.
Practice mold w/ hammer and small gear
Final mold with weights
Reflection and Teamwork
I think my partner and I were both happy with the results of our project. Although the SORTA-Clear 40 had a long cure time, once our 3D prints were placed the solidification process was simple and the prints were easy and de-mold. Our final mold had a lot of bubbles in it, but this was an acceptable result for our use case and we did not see a need to use a vacuum chamber to remove the bubbles. We encountered problems with the 3D prints sinking to uneven levels in the mold, but this was our fault as we did not print them to be a uniform height. The smaller gears were very close to the bottom of the mold, resulting in an extremely thin layer of silicone that we needed to patch after the original solidified. The third time mixing the mold definitely resulted in the best result, and I think my partner gained a lot of confidence through the repetition. To me, the most interesting part of the process was figuring out how to evenly weigh down the 3D prints, and I was happy when the makerbeam resulted in a level ice cube for the smaller mold. My partner and I did the design on our own time but most of the mold mixing together, which was definitely important in the early stages when we weren't sure what we were doing. We forgot to add a mold release agent to one of the molds, but thankfully this didn't cause us any problems.
My partner designed the wrench and I designed the gears and hammer. I would estimate that we spent a combined 6 hours working on the CAD model, including modifications made to the prints after testing. We spent about 5 hours making our test mold because we did not mix the correct agents on the first try. Once the test was complete we spent about hour prepping the final mold and then adjusting the larger gears when they shifted unexpectedly. Altogether my partner and I each spent about 10 hours working on this project. Including labor ($85/hr) and the cost of 3D printing and mold materials our molds would have cost around $1840 as a set.
Large Ice Cube
