Student Teams

For the final project, the class of 30 students were randomly placed into six teams of five students each.

Group 1

For this clock the overall goal was to provide a compact and modular design. This led to the stacked-layer configuration. The inspiration for our machine is wrist watches which feature a cylindrical stacked component design which makes them compact. We also decided on this configuration to keep all the clock components rotating about an axis perpendicular to any platform motion. We felt this gave us the best chance at avoiding interference from the movement of the platform. Some key design choices are the swiss escapement type and planetary gearbox. We chose to use a planetary gear system because it offers significant reduction in a compact fashion. In our final design we achieve a gear reduction of 25:1 through two gear boxes. The swiss escapement design was another design feature aimed at increasing the functionality of the clock. The tooth shape allows for low friction interaction between the gear teeth and pallet.

Group 2

Although we were technically allowed to use provided fasteners and bearings, our group decided to design our clock entirely out of custom designed 3D printed components. It took some creativity to design and print our own fasteners, but we were able to achieve this with considerable success. Our clock was modeled after a boat to emphasize the history, purpose, and oceanic theme of the marine chronometer project. We incorporated this idea into several aspects of our design: the chassis forms the shape of a boat, the balance wheel has added masses to look like the spokes on a ship’s steering wheel, and we etched a lighthouse into the flat stand face to add more to the overall marine look. We mounted the Arduino directly onto the chassis in a rectangular mount on the side that looks like a life raft. Some of us printed parts in different blue, sliver, and white color combinations to further play up the aquatic theme.

Group 3

Our chronometer design is modeled after a grandfather clock. The power is supplied by a falling weight, which is a 3D printed bucket filled with supplied hardware. The falling weight is guided down a track system and as the mass is lowered, it turns a spool. The escapement mechanism, featuring a graham deadbeat design, is attached to this spool. The falling weight provides a series of impulses to the pendulum, via the escapement subsystem, in order to keep the pendulum swinging and overcome frictional energy losses in the system. The pendulum, a harmonic oscillator, is the timekeeping device within the system. The pendulum was designed to swing with a 1000ms period, so that the output would record 1 pendulum period every second. A critical design constraint was that the clock must fit within the dimensions of a Ender 3 printer. This affected the overall size of the clock, limiting both pendulum length and the overall height that the falling mass could descend. In order to maximize the height available to the falling mass, the clock was mounted to the front right corner of the print bed. In this configuration, the weight could fall past the bed of the Ender 3, and all the way down to whatever surface the printer is atop. This allowed us to extend the life of our clock by adding more track for the falling weight.

Group 4

To tackle this exciting seafaring challenge, our team designed a fish-themed marine chronometer, outfitted with two fins and a tail, in order to swim through the stormy sea! From the very beginning, our team’s primary design goal was simple: build a fully functioning mechanical clock that optimizes accuracy and precision. We designed for a hybrid horizontal and vertical orientation, in order to dampen the motion effects of the “stormy sea” on our clock’s precision, as well as to fit within the size constraints of 200mm x 200mm x 250mm. Within this greater assembly, we designed each subassembly keeping specific goals and constraints in mind:

The Mainspring

"The m(itochondria)ainspring is the powerhouse of the c(ell)lock” (Ram 2020)--it supplies the energy necessary to run the clock. For our design specifications, we had in mind 45 sec. - 1 min. This meant a thick, large spring to store as much energy as possible (and naturally a large ratchet + barrel diameter to go with this design). Ultimately, the mainspring provides enough power to run for ~2-4 min.

The Gear Train

With the gear train, three primary points drove the design process: reduction of meshing points to reduce energy lost to friction; minimizing spatial occupation to meet design constraints; and most importantly, gear meshings. Ultimately given the design metrics this project prioritized, we decided to reduce the total gear meshings to just one: between the mainspring and the escapement sub-assemblies. Gearing down was outsourced to the pallet fork on the escapement, so the final gear train could ultimately be designed around its new main design spec of energy transfer. This outsourcing improved the accuracy of the clock design, and optimized the duration of the clock running from a single winding of the mainspring.

The Escapement

The escapement has the role of transferring the purely potential energy of the mainspring and gear train to the carefully tempoed motion of the balance wheel and hairspring. Thus the main consideration in designing the escapement was that it would harness the power of the mainspring as effectively as possible to keep the balance wheel ticking as long as possible. It was decided to use a Swiss lever escapement for two reasons: a Swiss lever escapement has geometry easier to effectively machine on our 3D printers than that of other common escapement types which rely upon precisely formed spiked teeth, and Swiss lever escapements are employed in over 98% of mechanical watches today. In the end, the escapement was efficient enough to keep the balance wheel ticking throughout the entire operating region of the mainspring.

The Balance Wheel + Hairspring

For the balance assembly, the primary goal was to sustain the power coming out of the mainspring for a range of 45 sec - 1 min. With this in mind, we decided that the hairspring would need to be strong enough to last multiple lengthy test runs, while not being so strong that it wastes the power from the mainspring. The ability to adjust the moment of inertia was also necessary, so holes, in which screws could be added for added weight, were added in the balance wheel. Ultimately, the hairspring was able to provide a 1 tick/sec rate for 2-4 minutes.

The Housing

Lastly, our chassis was designed to fit within the overall size constraints and be highly modular. The size constraints were placed upon us from the very beginning. High modularity allowed us to overcome the differences in our team’s printer tolerances and fine-tune our clock with ease. By creating slots on our base plate and using press fits in conjunction with holes for the provided fasteners, we designed a chassis on which we could place each subassembly exactly where we wanted it to be and adjust its spatial orientation, in respect to other subassemblies, on the fly, without having to print any new parts or go back and repeatedly change the CAD model.

Group 5

The main concept for our clock was the horizontal design. This decision made it possible to clearly see each component and subassembly, and how everything works together. After a few iterations, we decided to change our escapement design to a Swiss lever fork, in order to maintain a consistent oscillation and tick. The final assembly was longer than the printer bed, so bed clamps needed a little more creativity. Thanks to Juan's advice, we decided to print the clamps directly on the bed with the clock. But thanks, especially, to all teammates for designing a clock that works (for the most part)!

Group 6

Our team focused on having a long running clock for the competition, with accuracy as the secondary goal. Because of this, we focused on maximizing the amount of revolutions the escapement wheel would undergo and the time it took per revolution. This involved making a 18:1 gear train, a club-tooth escapement that minimized friction, and a hairspring with an optimized low frequency. The power source was also a point of improvement we looked at, and strongly considered adding a falling weight to the mainspring design, but decided not to avoid errors stemming from increased complexity. With tweaking that minimized friction, the clock was able to run more than 5 minutes when testing.

Page updated

Report abuse