I began my adventures in 3D printing with a cinquefoil knot from mathgrrl on thingiverse. I highly recommend her collection if you are looking for impossible objects! Though this project appears to be a lot more waiting and a lot less building than normal, it was actually the most iterative prototyping process so far this semester. It was very fascinating experimenting with so many parameters and 3D printers.
Prototypes 4-8! I learned from each iteration.
I started off using the MakeGear printers. I tried 4 times and got varying degrees of failures. One of them looked like an actual tumbleweed and another time, the fan fell off and crushed everything. Even with a skirt and the best support settings, the failure would always happen early on with the support at the base.
Dr. Wettergreen suggested the file may be bad, so I moved on to the Ultimakers to see if the issue was the file or the accuracy of the printer. 
It worked! In 31 minutes, my hope was restored! I could keep this design, but Dr. Wettergreen recommended I make it bigger to better fit the gumball capsule. I decided to enlargen it by 50% and attempt using the MakerGear again, since the knot would be larger. I also tried the Formlab. The prints took 57 minutes and 2 hours 31 minutes, respectively.
The supports for both knots turned out much better than my initial tries. Scaling up resulted in a huge improvement for the MakerGear. It was very interesting to see that the formlab printer’s optimal support was more vertical. Next, I ventured into the wetlab and used the Fortus printer. I was surprised that it was not as smooth as the MakerGear, but I was pleased with its clean support. I realized the main advantage of the Fortus was that you can remove the support with a chemical bath.
This print had the least evidence of support out of the three knots shown below, but it is also the one that has the most “3D print look.”
Artifacts of support on the base of the knots
The final two knots that I selected were the Fortus and MakerGear prints. While these are lower resolution than the FormLab knot, in my opinion they looked cleaner. Of the two, my favorite would actually be the MakerGear knot because it has a higher resolution and is more smooth.
The Final 2 Prints: Fortus (left) and MakerGear (right)
Cost Breakdown
While I spend many more hours than anticipated while working on this, it was for learning purposes and largely unnecessary. Knowing what I know now, I’d estimate labor to be 1 hour.
Fortus: $10.20 1 hour 43 minutes
MakerGear: $0.12 57 minutes
Labor: $7.50 1 hour
Total Costs: $17.82 – Not bad for 50 cent gumball capsules sold for a quarter.
