top of page

Dasher's Dinner: Fiberboard Automaton

February 2020

Mechanical Prototyping project to build a fiberboard automaton that uses simple mechanical mechanisms to translate the rotation of a hand crank to another motion.

This project was my first true introduction to using gears and cams to create something that moves. The goal of this project was to create a simple automaton: my dog (Dasher) lowering his head to eat out of his bowl and then looking up again after a few seconds.

Though I was successful in many areas of this project, such as using the cam tool in SOLIDWORKS, designing a piece for 3D printing, and making gears out of fiberboard, the biggest lessons that I learned were as a result of my failures.

Part of the layout used to laser-cut the fiberboard.
An image of a SolidWorks model of the automaton in which some of the sides have been made invisible to show the gears and cam inside, with components labeled.

At this point, I was unused to building things that needed to be both beautiful and functional. I spent a lot of time at the beginning of this project focusing on the aesthetic presentation of my automaton--how it would look, rather than how it would work.

Despite this, the CAD design of my automata went smoothly. I learned how to use complex mechanical mates and the animation tool and, except for a finicky cam mate, the SOLIDWORKS model worked perfectly.

My mistake was neglecting to consider the physical restrictions on fabricating this automaton. I did not plan how I would assemble it ahead of time, which lead to some last-minute quick thinking in order to put it all together.​ I also toleranced my parts incorrectly, leading to gaps between pieces and more setbacks during assembly.

An image of a SolidWorks model of the automaton in which more components have been made invisible to make the cam mechanism more visible.
A close-up view of the 3D printed dog bowl. The gears can be seen in the background.

Worse, I underestimated friction, a mistake that I will take every precaution not to make again. Though my gears meshed smoothly, when I turned the crank handle the cam follower jammed against the cam and the entire automaton locked up.​

Though this project was not as successful as I would have liked, I still consider it to have been an incredibly valuable experience that I will look back on next time I am tempted to focus on form alone and ignore function.

A photo of a fiberboard automaton in the shape of a dog standing on a box, in front of a white background.
bottom of page