Link to slides: https://docs.google.com/presentation/d/1ohyCrOGuuJfOTNp_0BUA9uAjlpCWDYoD1cdVHRuY6Io/edit?usp=sharing
Link to video:
https://drive.google.com/file/d/1D1XSbGCl2Yor2iBMoeFmeMOyoNeaVM8p/view?usp=sharing
https://engi210.blogs.rice.edu/2022/10/18/kacey-and-esther-volcano/
To start, we decided that the creative idea or setting we wanted to use was to be a volcano. We began by browsing through the 507 Mechanical Movements website online to see what was achievable, as well as previous ENGI 210 blog posts. We wanted to have our volcano bubble and erupt mechanically by turning a hand crank, so we made some preliminary sketches of a design using mechanical movement 96 as we saw in an example class, as well as some gears operating like mechanical movement 24. We really liked mechanical movement 36, which used two gears with varying teeth distances from the center to vary the speed at which one of the gears turned in response to another gear. Thus, our initial design was to have dowels that go up and down on different shapes attached to an axel turned by gears, and put pieces of rock or flames on top of the dowels to make the volcano look like it is “bubbling”. There would be two axles of these shapes with an axel in-between holding a large gear that turns them, as well as one movement 38 gear, interacting with another movement 38 gear above it. Preliminary drawings can be seen below (excluding movement 38 which is the first image below). We decided to make the volcano outside structure cross sections that slotted into rings around the system so that the user could see inside the system and admire the mechanics driving the movement.
Below are also our preliminary gears, as well as shapes and flames for the design discussed above.
We started with low-fidelity prototyping. The purpose of our first prototype was primarily to test how well our gears functioned (we designed them from scratch, without a gear generator). For this prototype, we laser-cut the gears, placed them onto dowels as axles, and mounted the axles in an existing cardboard box to test their effectiveness. By far the most difficult part of this process was getting the spacing of the axles exactly right (3”, with a tolerance of about +/- 0.1”).
We found these gears to be functional and so moved on to testing the next aspect of our project: the vertical dowels would support the flames or bubbles at the top. This prototype was still entirely in cardboard. We added hot glue to the ends of the teeth of the gears to make them more resilient, as cardboard has a tendency to delaminate with repeated force.
We laser cut a few of the shapes that drove the dowels up and down to create the bubbling movement and mounted them on wooden dowels inside a cardboard box. This box was vertical with a couple of compartments and was also slightly tapered so we could test how the system worked while the dowels were sloped at an angle. We used binder clips on the ends of the dowels where they touched the shapes to simulate the pieces we planned to build (runners).
We were confident in the moving parts of the system, and so moved to the frame for the entire system. This was composed of an axle holder, supports for the axle holder, two rings that fit flush with the outside of the axle holder to nail down the length of the axles, two discs with holes to support the vertical dowels, and a series of vertical that would form the outer “mountain” of the volcano and set the spacing for each of the horizontal layers. We utilized some iterative prototyping for the structure, where we first modeled each aspect in cardboard to ensure a good fit before moving to wood. For example, we first cut the axle holder in wood to ensure proper spacing. Then, we modeled the first two support rings out of cardboard. Once we had a good fit for those rings, we cut them out of wood and used them to test the fit of the vertical supports. This process was somewhat intensive just due to the amount of laser cuts we set up for ourselves, and the labor involved in repeatedly constructing and deconstructing the entire structure each time. The pictures below are in chronological order and give a rough idea of the progression of our prototyping.
However, this process worked as intended in that it helped us work out some practical issues before finalizing our model. One example is that we initially cut the ring supports (the two bottom supports) as complete circles, which were impossible to install because they needed to slide and click in from two, exactly opposite directions at once. When we cut them out of wood, we made them semi-circles instead. It also helped us identify some helpful features- such as adding slits for the frame in the upper two solid circles so that we could consistently control the angle. Here’s the final structure, in pieces and assembled.
For the runners, we wanted to make sure they were strong and would not snap under the friction and forces since they were conducting most of the mechanical movement, but also make sure that they were as low friction as possible since there were so many being driven by one gear. We came up with the following design:
We sanded a small piece of dowel down to the point where it looked like it was cut in half laterally and was flat to wood glue to the bracket. The reason why we added the dowel was because its curvature added smoother running on the shapes and it would have less friction since it lessened the contact surface area with the shapes.
For our metal pieces, we decided to plasma cut steel rocks and flames to go on the other end of the dowels because we liked the rust on the steel in the machine shop and thought it looked very volcano-esque. We also wanted the slag from the plasma cutter to make the rocks look lumpier like they just came out of a volcano. Below is a file of the rocks with a flat part in the middle to use a screw to attach it to a dowel.
Once they were cut, we used a mallet and vice to fold the pieces so that they were double-sided almost for more presence of the design and then decided to add some more spray paint onto them because the rust had washed off a little bit in the powerful plasma cutting process. Then, they were a vibrant red which we were happy with.
For the middle, Kacey used some red and orange wire from the elec lab and used artistic creativity to create an “eruption” of lava that we mounted to the middle dowel. Originally, this dowel would have been attached to the movement 38 gears and had more variable movement, but after medium-fidelity prototypes of that gear were not promising enough, being short on time, and already having a complicated design, we decided to change the gear design to one similar inspired by the whack-a-mole team. We had originally cut the movement 38 pieces and hoped to get them to work by sanding the wooden pieces down (this way we would also save material without doing many test cuts) since they were just barely not working, but as time went on we decided this would be too risky. At the time of this decision, the gears and shaped had been cut out of wood, wood glued in doubles, and already glued onto the axels, so we had to be creative with our solution. Mechanical gear 38 was left in between the glued driver gear and a glued shape, so we could not remove it or add any other shapes. We decided to instead cut a shape to fit around the existing gear that had the outside we envisioned (similar to whack-a-mole). A visual of this piece can be seen below.
After cutting two of these pieces, we glued them onto the gears and dremeled the sides. This worked fantastically and our gears and shapes were finally complete and installed in the frame.
With the number of moving parts and contact surfaces in our design, our post-processing time was severely limited due to the amount of troubleshooting that was necessary to ensure that each runner ran smoothly on its shape, and threw two layers of the frame. We ran into three major problems. One, the axles being bowed and so the shapes translating instead of purely rotating, and so the runner would get stuck at an angle. Two, some of the shapes were not straight on their axles, and so as the axles rotated, the shapes wobbled on their axis and jammed against the side of the runner. Three, the vertical dowels all ran through the two solid circles of the frame at an angle to emulate the slope of a mountain- over about 14 inches in height, the ends of the dowel moved about an inch inwards toward the center of the circle. Thus, we were trying to translate vertical motion at an angle, which caused some difficulty as the dowel hit the side of the holes in the two solid circles of the frame.
We spent a lot of time sanding the shapes down after they were glued together with a dremel, and then sanding even further with a sanding stone bit on the dremel, which “polished” the sides so they had much lower friction. After sanding the frame together, and post-processing the metal pieces by bending them and painting them, our design was complete but just needed some stain and color. We were able to accomplish some of the stain, but hope to return and utilize our full artistic desires on the model now that it is fully working. On this project, we prioritized functionality over aesthetics but were still able to include the most important aesthetic aspects: a frame with jagged cross-sections to create the outline of a volcano, vibrantly painted flames, and rocks that bubble when the hand crank is turned, a vibrant and textured plume of lava in the middle, and clean, visible shapes that can be seen through the frame as they work.
We are very proud of the volcano that we made because it is fully functioning and has some sick decorations! After we complete the staining and add a little bit more lava (dripping paint) to complete our aesthetic visions, our device will be ready for OEDK users to explore mechanical movements!
Reflection:
We received some feedback about our dowel with the hand crank being too thin which is a consideration that segues well into what we learned from this project. First, we learned that although the device fully works with the existing hand crank, we could have facilitated stronger movement with a thicker dowel. Second, one instance of the movement working smoothly does not preclude complications if you repeat that over and over again. Repetition exponentially increases the number of complications that can arise. Third, it’s very difficult in the middle of a project to predict the behavior of higher-fidelity materials. We did so many cardboard iterations that we were confident that everything would fit in wood. While this might be true of static structures (like the frame), the moving parts were never guaranteed. The amount of low-fi prototyping we did made us overconfident in constructing the final, wood version. Finally, this has been another valuable lesson in time management within a team. We worked together fairly effectively- we used a “plan together, execute independently” which was excellent in that both of us stayed on the same page, and were able to contribute to the project as time allowed. However, in the end, when we were both executing together, we didn’t have the experience to delegate as efficiently as we could have.
Cost Analysis:
We each spent about 40 hours on this project, so by using minimum wage, that alone totals to $580. With regards to wood, since it is $10 for 2ft by 2ft, and we estimate to have used around three sheets, that would be $30. For small pieces like the screws, steel, and dowels, we estimate around $30 because we split quite a few dowels and used them all throughout the prototyping process. Finally, we estimate around $10 for the adhesives (hot glue, wood glue, super glue), as well as the stain and paint. Fortunately, the OEDK provides free access to laser and plasma cutters which do not need to be replaced after use like adhesives, stain and paint do, so we did not include that in our cost estimate
All in all, this totals to $650.
