This month in EDES 210, Gabriel and I completed our midterm project. The goal of this project was to implement a few of the 507 mechanical movements into a working mechanical model, using some of the machines and techniques we have learned recently to build it. We chose to create a compliment generator, which turned out really silly and fun! For the model, our mechanical movements were a simple gear mate, a geneva drive, and an unnamed translational oscillatory motion (movement 93). The compliment generator would be driven by the simple gear assembly, where the smaller gear drives movement 93 on the same axle and moves a decorative sign for our model up and down. The bigger gear, which is mated to the smaller gear, would be on the same axle as the geneva driving gear. Then, the geneva driven gear would rotate iteratively and have a wheel with compliments at each iteration attached to the same axle.

Figure 1: Movements used in model
The first step in this project was to create a plan for our model. We created an illustrator file with all the shapes we needed to laser cut. This file included all the functional pieces of our model, including a few gears from the gear generator site provided in class, the geneva drive which we created in SolidWorks, and the oscillation piece which we also created in SolidWorks.

Figure 2: Excerpt of Adobe Illustrator file for component
Our second step was to create a low-medium fidelity prototype; we chose low. We used the laser cutter to cut out our two gears, geneva drive and oscillation piece. Before cutting the gears, we tested hole diameters using the laser cutter to see what size hole was best for press fitting the dowels into. We determined that a size of 0.31” was best for press fitting.

Figure 3: Press Fit Test
We used a piece of cardboard as a base to assemble our gear train with dowels to test the mating of our components. Using our hands to support the prototype, we tested the motion by revolving the driver dowel and ensured that it would work as expected.

Figure 4: Low Fidelity Prototype
Our third step was a lot lengthier than the previous steps and involved creating a higher fidelity prototype than before. We began by building a case for the assembly using the box maker website from the laser cutting assignment. Once we had the file for the box, I measured the ideal position for each of our three dowels and superimposed them onto the file for the box walls. For the case, we only used two walls, a floor, and a small top with a slit that I manually created to hold the oscillatory piece. We wanted to be able to press fit bearings for the dowels into the holes on the walls, so we first experimented with laser cut diameters to fit bearings. We found that the perfect diameter cut to press fit a bearing was 0.86”, a value we ended up sharing with a few teams (we felt very helpful and nice). We then assembled the case and placed the gear train in it using the same configuration from the low fidelity prototype, except this time the oscillatory piece was threaded through the top and stood on its own. One issue we ran into in this phase was the press fit of the dowels loosening, as continuous rubbing of the dowels against the hole of the gear caused the dowels to ultimately wear down. We temporarily fixed this by taping the dowels to the gears so it would hold for demonstration. Another issue we ran into in this stage was depth-wise spacing of the gears. We began experimenting with spacers on the dowels to prevent gears from becoming dislodged with use, and this was a very promising solution to our problem. Additionally, it seemed our movement 93 caused more of a waving motion than a linear motion, and we decided to take advantage of that in the next phase (no spoilers)! A small nitpick we found was that we had an extra hole in our case across from movement 93, since the dowel for it cannot span the whole assembly or else it would block the movement. We planned to address this in the final phase. We were able to do some practice runs of our mechanism that worked well, but there was still much to be done to ensure consistent success.

Figure 5: Medium-Fidelity Prototype
Our last step was to create the final product. The first thing we did in this stage was create a vinyl sticker of a smiley face, stick it on a 2” laser cut circle, and add it to one of the dowels outside of the case for good vibes! This part will spin around as we power the machine. Next, we addressed the waving motion of our movement 93 assembly by laser cutting two identical wooden hands and sandwiching them to the oscillating piece. This made a fun waving hand that we’re pretty happy about. We then laser cut a new compliment wheel, which included 5 evenly spaced messages etched on it for each turn of the geneva drive (also no spoilers for the messages, you have to come try it out for yourself). We also made sure to engrave the text in Professor’s favorite font, Comic Sans! Next, we needed a crank for our assembly which we decided to construct out of metal. We created the profile of our crank on SolidWorks and uploaded it to the Waterjet for some smooth metal-cutting action! This was exciting because it was our first time being able to use the waterjet for this class, so we included some detailed images below. After cutting the crank out, we used the sandblaster to smooth out the edges and clean up the surface.

Figure 6: Prepping the waterjet for the first time

Figure 7: Finished Crank
Then, we laser cut and engraved our nameplate to go on our assembly. Now that we had all of our components, we completely disassembled our model to perform some post-processing. We sanded the surfaces of all of our wood and applied tung oil to the case and nameplate to make it look nice. We also trimmed the geneva driven gear to have a little more tolerance for the nub to slip in when in use. Once this was done, we glued the nameplate over the hole we weren’t using in front of movement 93; a subtle fix!

Figure 8: Nameplate
Finally, we assembled everything back in the case, this time using small circular spacers of the same press fit diameter that we laser cut to ensure that everything stayed in the right place. We even used the spacers on the outside of the case to hide the bearings.

Figure 9: Inside of the assembly, with spacers

Figure 10: Me happy because we’re almost done!
This final assembly worked great, but we also made sure to wood glue the parts to the dowels and seal the case with wood glue for good measure.

Figure 11: Gabe, holding our design together while the glue settles
Overall, we were going for an au naturale look that we are very pleased with. It definitely turned out to be quite a goofy mechanical model. The assembly works pretty smoothly and we’re excited to have people use the Compliment Generator.

Figure 12: Final Product
Cost Breakdown:
Materals:
4 x 1 in 0.25 in thick steel -> $15 (home depot)
1/4″ 2’x2′ Plywood x 2 -> $16 (lowes)
48″ 3/8 wooden dowel -> $1.38
6 pack ball bearings -> $10 (amazon.com)
Cricut Premium Vinyl – $5 (walmart.com)
Labor:
14 hours in OEDK x2 people: $10/hr*14 hr*2 people -> $280
Machine Costs:
We used alot of different machines for this project, so lets say we were able to use these machines for a fair price of $5/hr on average, and we spent about 10 hours of work time on machines.
$5/hr*10 hr = $50
Assume all OEDK tools, such as calipers, are free of charge!
This brings the total cost to make our mechanical model to $377.38. Most of this cost comes from labor and machine costs, although the material costs were high compared to past projects since we used so many different things. However, for a piece as timeless as the compliment generator, it’s value far exceeds its cost.

Figure 13: Last used workspace