For the midterm, I chose to build a model of a system using mechanical movement 212.  The movement converts constant angular motion into discrete angular displacements.  When brainstorming an idea, I tried to think of something that used discrete angular displacements, like a clock.  However, after considering option, I believe it would have either been too simple or too complex.  Instead, I decided to pursue an “elevator/dumbwaiter” system, in which a constant angular velocity is converted into discrete angular displacements, which is then converted into linear motion via a pulley system.  Although this is not how elevators work, I thought it might be fun to try and apply the Geneva Mechanism to an elevator.

First, I had to build a low-fidelity model to just get an idea of how the part will fit together.  In order to do so, I laser cut the device out of cardboard, as it is cheap and quick.  The cardboard fit together well, so I moved on to a medium-fidelity model that was made out of wood.

Now, I wanted to test the mechanism to see if it had enough accuracy and precision to work while rotating as it is meant to do.   In order to test its performance, I created a low-fidelity rig to provide feedback as to how it was working.  The medium-fidelity prototype also worked (and functioned on the rig), so I moved on to a different material: acrylic.  I like acrylic because you can see through it and because it doesn’t burn as easily as wood.  After these prototypes were cut, I had to move on to the rest of the project.

 

Side note: the first parts of this project took way longer than they should have.  I will remember this for our next project.

After the primary mechanical movement was taken care of, I needed to make the frame to hold it up.  I wanted to build the frame out of acrylic so that the mechanical movement would be obvious and visible.  However, it proved difficult to create a structure that would be efficient to make and clearly show the mechanism.  I finally decided on the final design after many sketches/brainstorming sessions.  Then, I realized that I would need to have an idea of how the whole system would fit together before I started drawing the frame in SolidWorks.  I think this is an important lesson in prototyping in general.  One must have a clear idea of how everything fits together before starting.  Hence, it was back to the drawing board for me, laying out all the places where the components would go.

Next, after everything was laid out, I was able to start drawing out the frames in SolidWorks.  After those were completed, I saved them as an AI file.  When I opened this file up in Adobe Illustrator, it was the image of the sketch, almost as if someone had just taken a screenshot and pasted it in.  This was not going to work, as image trace did not come through with great results.  However, thanks to Will Jones’ genius, we discovered that DXF files allowed for easy manipulation in Adobe Illustrator.  After the drawings were successfully input into Illustrator, I cut a sample out.  Nothing fit.

What happened? Well, when you insert a DXF file into Illustrator, it automatically scales the picture in whatever way it pleases.  This then forces you to create extra dimensions to figure out how to resize the picture accordingly (width and height).  After the accidental misprint, everything afterwards was checked and resized.

After the frames were cut out, I decided on a base structure.  The base is made of stacked ovals that provide the frames with a sort of anchor, and the frame slides into the slit in the oval.  Now, I was working with wood, so I finished the wood with a rub of linseed oil (it was a medium-hue wood to begin with, so the linseed allowed more of the dark spots to come out).  After rubbing all the wooden parts with linseed oil, they were attached in the corresponding places.  Surprisingly, the oil did not affect the hold of the glue (from what I observed).

After the glue dried, the frames were attached to the base, and the system started to come together.  The wooden axles were coated with beeswax for slip as well as a shiny surface finish for looks.  The plasma cut crank was angle-ground, sandblasted, and coated with clear coat twice.

All in all, I learned that one needs to have a clear idea or even a picture of the end goal in order to do a good job (or else you will end up redoing a lot of parts like I did).  I also learned to check dimensions whenever you change software platforms, as they may be in different units, or the transfer scaled the image in some way. Lastly, it is good to start early and get as much done early on as possible.