For our midterm we were tasked with bringing one of the 507 Mechanical Movements to life. I chose number 100 because it seemed useful for real machines, and because I am interested in machines that convert one type of motion into other types.

The first step toward my final product was thinking about what layers were necessary. Obviously the pendulum cannot swing through the drive wheel, so it had to be on a different plane, which would necessitate having a spacer underneath the pivot point. And all of this had to be supported, so it would need a base. And the axis of rotation and pivot point would need to be rigid, so the base should probably be at least 2 layers. Plus I would need to turn the drive wheel, so that would require extending a handle out in some direction. All of a sudden this super easy looking mechanism was getting more difficult, so I threw together a 3D model using SketchUp:

Once I had a the general idea down, I set to work in Adobe Illustrator.

For the low fidelity prototype I planned on using some 1/2″ dowels, some 3/8″ dowels, and some 1/4″ dowels. This made things overly complicated and lead to some mistakes, so my first attempt did not work well because some of the holes were the wrong size.

I knew early on that I wanted to use bearings in all the parts that rotated, so I ordered some 1/4″ wide bearings and used them in multiple places throughout my design. Because the plywood is really only 0.20″ thick, they were not completely flush with the surface. They also had metric dimensions, which complicated things a bit. Fortunately though, a 3/8″ dowel or bolt fits well into the 8mm bearing inner diameter.

With all the math figured out, I was able to make a successful medium fidelity prototype:

This one was actually quite good, but still had a few flaws:

1. There was nothing keeping the crank shaft (the axis of rotation) from moving in and out a bit, so when you turned the crank you also had to apply pressure in or out to keep the drive wheel from pressing too hard against the base or the pendulum.
2. The base was not supported in the middle, so the sides were free to bow in or out. Further reinforcement was necessary.
3. When operating the machine, you had to hold the base down with your free hand to keep it from falling over, so I decided it would be a good idea to make a wide stand for it.
4. The drive wheel looked pretty boring.

To solve the structural problems, I made box joints (aka finger joints) in the sides of the vertical base, and just cut out 2 more of the top/bottom supports to fit into them. I also made the bottom support much larger to keep the whole thing from falling over. To solve the crank shaft problem I decided to use spacers and stops along the crank shaft to hold it in place. This might not have been the best idea. I also engraved my name and some info about the project on the front and back, and added an owl to the pendulum. Finally, after going over various designs, I settled on making a drive wheel in the shape of the Rice R.

       

Let’s talk about this R for a bit. First of all, it consists of multiple pieces, so I had to add in supports for those pieces in Illustrator. This worked really well with the laser cutter, as you can see above. However, as part of the assignment we had to make at least one part out of metal, and I thought it would be cool to make the R with the plasma cutter.

Didn’t go so well. So for plasma cutter Take 2, I just cut out the outline of the R. Another difference between the plasma cutter and the laser cutter is the kerf size. Turns out that if you try and cut a 0.35″ hole on the plasma cutter, it comes out closer to half an inch. So for plasma cutter Take 3,  I set the hole size to 0.21″, and this ended up being manageable, although only barely. One complication that I was not able to overcome was that whenever the plasma cutter made the holes, it used a male lead in. This resulted in a bit of a defect in each of the 2 holes. In the end though I think the R turned out pretty well.

After cutting it, I used the belt sander to clean up the dross, then a 5/16″ bit in the drill press to clean up the holes, and then a 3/8″ tap to thread the holes to accept 3/8″ bolts. Afterward, I sanded it using different methods until I was content with the finish.

Next came painting the R. If I were to do it all over again I definitely would not paint it. It took a lot of time to dry (very high humidity in Houston this week) and even when dry the paint still is not very durable, so it gets messed up easily and rubs off onto everything it touches. It looked pretty good without the paint, but I thought it would look even better if it was blue (because Rice is blue, and for contrast).

I also though it would be cool to add some texture to it, to make it more visually interesting, so my first iteration looked like this:

However, it looked a bit too unintentional, so I decided to just go with a solid color instead.

This turned out looking much more intentional and smooth and less “artsy”.

Once all this was put together, it still required holding it all down with your free hand, so I used a super useful function in Illustrator to make the stand bigger: you can choose a path, then go to Object-Path-Offset Path and enter in how much of an offset you want (can also use a negative number). It turned out like this:

To get everything to rotate smoothly, I used spacers inside the base to adjust the position of the drive wheel relative to the base. I took way too much time figuring out exactly how many of what kind of spacers made the fit just perfect. It ended up looking like this the for the first iteration:

Eventually I ended up cutting out more spacers all of the same size, which made it look much better:

With all the final pieces made, it was then time for finishing them. I liked how the unfinished pieces already looked bright and light and shiny, so I chose Tung Oil to add some shine and protection without distorting the colors too much. Before wiping on the oil, I hand sanded all the parts with 400 grit paper, being careful to sand in the direction of the grain. I applied the oil with the grain as well.

After that had dried, it was time for final assembly.

To glue in the supports for the base, I used epoxy. I also used it to attach the owl to the pendulum. Other than those parts, nothing else is glued. Because I know that many people will be interacting with my mechanism (hopefully), I wanted the project to be as modular and as sturdy, but yet adjustable, as possible. By loosening or tightening the bolts, you can change the smoothness and ease of actuating the mechanism. I have it set right now to be as smooth as possible, while maintaining rigidity and keeping the pendulum tight enough that whenever you release the handle it does not fall to the side.

Here is the result of minutes waiting for the parts to get cut, hours waiting for access to the tools to cut the parts, and weeks planning and designing the parts to be cut by the tools.