Our latest mission was to 3D print small, impossible objects.

We defined impossible objects to be designs that would have been impossible or extremely hard to make with traditional manufacturing methods, such as milling, lathing or molding and casting. For example, designs with separate internal parts, parts that interlocked, or parts with extreme details, especially in hard to reach places.

With this definition in mind, I started by looking through Thingiverse for objects with either internal portions or intricate details.

Before long I found this design:

https://www.thingiverse.com/thing:19104

I chose this design because the internal portion would be hard for traditional manufacturing processes to reach and the overall cellular design style is also challenging to replicate with traditional manufacturing processes.

As part of this challenge was to fit the impossible designs into small containers for a gumball machine, when I imported the stl file into 3D Printer OS, my first step was to try sizing down the piece. Initially, I did not make it as small as it needed to be, as this was my second time ever 3D printing. I thought trying to make something so complex, small, may need to wait until after I got the hang of printing something like this. Next, I made sure it was on the bed and selected to have no supports, as the design didn’t seem like it would need any, and they would have added a lot of extra material that would have been hard to take off such an intricate and fragile design. Finally, I sent it off.

My first two attempts at printing, I ran into some issues. The first try ended up with my print moving around the bed, so after this, I changed the skirt to a brim. Then my second attempt was ended because it had the ‘wrong printing speed’ and I realized I couldn’t adjust the layer height.

Eventually, on my third attempt, I was able to print something, but apparently, on this one, I forgot to select ‘no support material’ and I ended up with a print that had “an incredible amount of support.”

The next print looked promising, as it finished and looked to be printed cleanly.

However, this piece seemed big, so the next piece I sent was significantly smaller.

This version printed well and fit inside the gumball, so I repeated this print 3 more times.

After this stl file proved successful with the PLA 3D printers, I took it to the resin printer.

After the autogenerated print file, I edited it to have less supports, as the design is very intricate and extracting them would be hard. The printer didnt love the idea of less supports, but I decided to test it anyways as the PLA printer hadnt needed supports at all.

Thankfully the print came out okay! The supports were very similar to the design itslef, even after the acetone bath and curing process. So I had to use sidecutters to remove them, wich was a pretty dextrious process. At the end, It came out great and I was able to remove all of the supports without harming the piece. The piece also fit into the gumball container! 🙂

This assignment was very challenging to do without access to the OEDK, and I want to thank everyone who looked at my prints for me, whether they were failed or not, including Ruben and Dr. Wettergreen. The time the remote situation added to the iterative design process was immense and made any turnover challenging. I think this speaks to how important fitting several iterations is into design and manufacturing, and also how important physically inspecting your prototypes is to improve the design. I was able to accomplish so much more by spending time in the oedk and immediately editing my files after failed prints. Eventually I was able to do all the prints, even if I was not able to find all of the gumballs.

The costs for this project so far have worked out to be:

PLA material for working prints (FDM):

$0.01 x 4 = $0.04

PLA material for failed prints (FDM):

$0.05 + $0.60 + &0.05 = $0.70

Resin material (SLA):

Lab assistant labor

$10/hr x 2 hr = $20.00

Gumball Containers:

$0.10 x 5= $0.50

Total: $21.24