Overview

I would like to start off this blog post by saying that in my entire time at Rice, I have never spent so much time on one single midterm. Producing a final product, which I could say that I was proud of, was tedious, arduous, time-consuming, but most of all, rewarding. Although I may have spent too much time working on this midterm relative to the coursework in my other classes, putting the pieces together and watching my creation finally work in front of my eyes is what made the entire process worth it. As soon as I put my final prototype together and saw it work, I immediately forgot about all the struggles and modifications that I had to make throughout the course of the project. All of that being said, I can truly say that this assignment was extremely beneficial to me. Not only did I refine my plasma cutting, laser cutting, and post-processing skills, but more importantly, it taught me two important things. The first being how to revise succeeding prototypes after I had determined what was wrong with the current prototype that I had made, and the second — that producing a final product requires precision, planning, and patience.

In this blog post, I will walk you through the entire process that I went through in order to produce my midterm project. I will begin at the moment where Dr. Wettergreen advised me to change the mechanical movement that I wanted to model. Then I’ll talk about my very first prototype, where I came to the realization that I was not only missing some mechanical components to the project but also that I had to add more to my project if I really wanted to produce something that I would be proud of. Next, I will talk about the edits that I made to the project after the first prototype and several of construction practices that I used to ensure that my project is built well. Then comes the construction of the final prototype, which includes the finishing process.  Lastly, I’ll finish the blog post off by talking about what I learned through this project.

The Beginning

I walked into class three weeks ago, eager to begin working on creating a model of mechanical movement #139, the movement which I recreated on Adobe Illustrator the week before.

On that fateful day, Dr. Wettergreen told a few students that their chosen movement might be too difficult for them to make a real model with it. Sadly, I was one of those students. Although I knew I was going to have to look for a new movement and create a new drawing on Adobe Illustrator, I was grateful to Dr. Wettergreen for his insight for two reasons. 1) I knew that this project was going to require a lot of time and effort, so making one more drawing in AI would be nothing compared to what was coming, and 2) I probably would’ve spent a much greater amount of time working on the midterm if I had chosen to keep the original model that I had chosen. Following Dr. Wettergreen’s suggestion, I went back to the 507 mechanical movements web page and chose mechanical movement #100. Unfortunately, in a lapse of awareness of due dates, when the time came for the first check-up I had not created the drawing of my new mechanical movement. Realizing that I had fallen behind, I immediately began my new drawing so that I could begin prototyping.

Although I may have not achieved that check plus that I was aiming to receive for every part of the check-in process, I wasn’t going to let this bring me down. If anything, I used this experience to motivate me to produce the best midterm project that I possibly could.

From Drawing to Real Model

After finishing my Adobe Illustrator file, the time had finally come for me to attempt to create the physical mechanical model. Beginning with a low-fidelity prototype (because in all honesty, I had no clue whether or not my model was going to work), I grabbed a piece of cardboard and headed straight to the laser cutter to cut out all the pieces. Once I had the cardboard pieces, I haphazardly put them together using hot glue.

As I had mentioned earlier, planning is one of the most important aspects when one is trying to create a physical, working model. As soon as I tried to piece everything together, I realized that I didn’t even have a base on which to put them on. This was one of the mechanical components which was missing in my first prototype. In order for the arm to work, the two components which it lays on must be fixed to a bottom structure so that the dowels may rotate and so that the arm may move up and down as a user pulls on the crank.

One of the first revisions that I sought to make had to do with the parts which contain the dowels. In the low-fidelity prototype, the circular components under the oscillating arm consisted of two cylinders. The bigger cylinders were to be fixed to the support structure so that the smaller cylinders could be inserted into the empty space of the bigger cylinders, but still be able to rotate. Originally, I thought that this method would work, since I thought that the bigger cylinders would stop the smaller cylinders from moving out of place. However, in practice, this method does not work well. Realistically, when a user tries to use the crank, they may subconsciously apply a downward force to the crank as well, which creates a moment on the base of the dowel, causing the smaller cylinder on the left side of the arm to lift from the bigger cylinder’s space, not being able to rotate.

Other than including a base and fixing the structures which support the dowels, I knew there was something else that was missing in my project. I felt underwhelmed by the simple oscillatory motion of the arm from pulling the crank. I sought to add something more exciting and enticing to the user. After sketching and brainstorming, I had decided that I wanted to add make-shift pistons to my midterm project, so I went back on Adobe Illustrator and began drawing the edits to my model.

Moving Forward

 

 

 

 

 

 

 

Above are the Adobe Illustrator files that I created in order to not only fix some of the errors from the previous prototype but also add the piston components to the model.

Figure 7 illustrates the changes that I made to the dowel-housing structures which will fit into the base of the model. If you look back at figure 4, you will see that each dowel-housing structure consists of two differently-sized cylinders. As mentioned earlier, this method did not work well for turning a crank which lies on top of these structures. To fix this problem, I made the structures in figure 7. Different from the ones in figure 4, these only consist of one cylinder for each dowel, which is actually glued onto the base of the model. This is a more appropriate method to achieve what I was going for. Since the cylindrical structure will be glued to the base, inserting the dowel and attaching it to the crank will allow the user to turn the crank much more easily.

As you can see in figure 5, one of the first changes that I made was actually including a base in the file. You can also see the two circles cut into the base, which will allow the new-and-improved circular components to function much better. In the actual model, I stack two bases with holes cut in them on top of one more base without the holes. This would allow for the circular components which house the dowels to be properly fixed to the base structure, ultimately improving the movement of the entire model.

In figure 6, you can see the components which make up the piston pieces. The pistons are made from many circular cuts being stacked on top of each other and the housing for the piston is made up of the square cuts. I  also needed two more arms which would attach to each piston. In both figure 4 and figure 5, you can see the holes that I cut out of each arm. This will be where the dowel will go to connect all three arms together. The main arm, or the longest, will connect to the two shorter arms, which will be connected to the two pistons. Thus, as a user turns the crank, the pistons will shift up and down as the main arm oscillates.

After making these edits, I was more than excited to begin putting my medium-fidelity prototype together to see how it runs. Other than switching to wood glue from hot glue, one good practice that I picked up while working on my medium-fidelity prototype is using clamps to ensure that the wooden pieces are glued together well. I found that this process is much easier than holding them together for extended periods of time.

While working on the medium-fidelity prototype, I also practiced sanding down pieces and making them smooth and uniform. One of the pieces that I practiced on was the housing for the piston.

In order to make the housing of the pistons uniform and smooth, I would use the belt grinder and sand off the burn marks from the laser cutter. Afterwards, I would use sandpaper to further smoothen these pieces.

Figure 10 illustrates the medium-fidelity prototype. As you can see, the piston system is included here. This model works much much better low-fidelity prototype. With this model, you can actually pull the crank well, and the piston will go up and down. Something that I must fix about this prototype though, is how well the pieces move together by lubricating them. I will elaborate on this later in the blog.

On to the Final

It is finally time to build the final model of my mechanical movement. Once again, I go cut out every piece from the laser cutter, glue them together using the clamps, and then sand them until they are nice and smooth.

The first step that I took to complete the final iteration was making the required metallic piece. I chose to make the main arm out of metal. Personally, I really like how it looks on the model. With every part being a similar shade of wood, the metallic arm in the center of the piece provides a nice contrast and adds a nice weight to the model, making it feel more premium.

Taking the Adobe Illustrator file to the plasma cutter, I cut out the main arm out of metal. Afterwards, I used similar post-processing techniques as the previous metal assignment to glue these together and give them a nice finish. First I used the angle grinder to take off the slag off of each piece. Next, I filed each piece so that each face was smooth. Then I glued them together before continuing to sand off the edges and the faces.  Lastly, after sandblasting the glued piece, I applied a layer of protective gloss coating, leaving the metallic arm with a nice grey color and a smooth surface.

The next step to finishing my midterm was applying a mahogany stain to all the wooden pieces. I personally love the color that it gives the wood and how it makes the grains jump out at you.

 

 

 

 

 

 

 

 

After finishing all the pieces individually, the last part left was putting it together. Very carefully and precisely, I made measurements to ensure that I put every piece in its proper position so that the crank, as well as the two pistons, would move easily. Then, after gluing all the pieces in their proper places, I used beeswax to lubricate all the dowels that I used in my project. Finally, after connecting all the arms together and attaching the two smaller arms to the pistons, my project was finally complete!

Click on this link to watch the model in action: IMG_5734-1xsypcy

 

 

 

 

 

 

 

 

 

Conclusion

There are several things I could change about my design if I were to redo my project. One of the first things that I could’ve improved is the way that I attach the piston housing to the frame. While I used wood glue to glue each housing to the bottom frame, it would’ve been much more stable and stronger as a whole if I had done something similar to what I did with the dowel housings. I should’ve added some slits into the bottom frame and onto the piston housings so that they would interlock. Another way that I could’ve been more efficient about the completion of my project also has to do with the pistons and their housings. Rather than cutting each slice of piston and housing on the laser cutter and then gluing them together into the entire structure, I could’ve easily done the same thing using the CNC machine and drill a hole through blocks of wood. However, I chose to laser cut each circle and each slice of the piston housing and stack them together so that their grains would pop out more after applying the mahogany finish. This appearance where the entire surface of the piston housing is etched with the wood grain boundaries is very pleasing to me.  This assignment has definitely been my favorite in the semester, possibly in my entire time at Rice. Although there were some struggles and setbacks, I am extremely proud of and happy with what I produced in the end.