PROCESS:
For my midterm, I decided on creating a machine representing Mechanical Movement 92. I was specifically drawn to this movement because I thought that the translational aspect of it was pretty unique, and would present an interesting challenge:
The first step of the process was to plan on paper. I decided to approach the process in a pretty comprehensive way, where I dissected the movement into its individual parts in order to understand how all components worked together. This would not only help me determine which parts I would have to 2D model in Illustrator, but would also give me insight into the dimensions of these parts and the reaction forces/moments that would need to be present in order for the movement to operate as expected.
Once this was done, and I felt comfortable with my understanding of the movement, I moved on to modeling in Illustrator. This is the 2D Drawing that I presented to Dr. Wettergreen on the date that it was due:
It’s quite hard to see the text in the picture with the resolution I was able to get on it, but the text above each part numbers each part and dimensions said part in pixels. In hindsight, I definitely should’ve dimensioned the parts in inches from the get-go, but the dimensions in pixels were still useful for proportionally relating the size of each part.
The main components of this 2D model are:
- The three rectangular pieces with holes in them, that, when combined, represent the sliding bock in Movement 92. The circle in the middle was initially not dimensioned with any specific axle in mind, but would eventually be dimensioned to match the low fidelity straws available in the OEDK, when it came to low fidelity prototyping.
- The 3 thin rectangular pieces that represent the component of the mechanism in which the block would slide between. The dimensions in this piece are related to part 1 by the fact that the height of the smallest rectangular piece matches exactly with the height of the smallest rectangular piece of part 1, which would be the part sandwiched between the other 2 rectangular components of part 2. The height of each long rectangular piece in part 2 is also exactly the height difference between the taller rectangular components of part 1 and the shorter component, both above and below the center circle.
- The wheel that, when spun, induces translational movement in the sliding block (part 1). This wheel requires two holes, one for the axle that holds it in place and that it rotates about (direct center hole), and the connector between the wheel and the sliding block (lies on one of the spokes of the wheel). Dr. Wettergreen advised me to revise the design of this wheel, as the way that the spokes led into the outer rime of the wheel would be troublesome for the laser cutter to cut (zero thickness error). As for the intent behind the dimensions of the wheel, I made it so that the diameter of the wheel did not exceed the length of the long rectangular components of part 2, as this would make it so that the sliding block would not be sandwiched between part 2 when at its furthest position from the wheel. As for the dimensions of the inner holes, I initially planned for the supporting axle to be larger than the connector hole, but this changed as the fidelity of my models increased (and was also limited by the sizes of the wooden rods in the OEDK).
- The connector between the spinning wheel and the sliding block. This piece is a rounded rectangle with two holes, one on each side of the shape. This piece’s length was determined by the distance between the connector hole of part 3 at its furthest position from part 2, and the sliding block at its closest position to the wheel (with around 20 pixels of wiggle room).
- The preliminary base of the mechanism. I wasn’t quite sure how I was going to support everything, so I made two large rectangles in this version of the 2D drawings.
- Washers that would serve to connect several parts of the mechanism, and restrict translational motion and allow rotational motion about some axles.
I made a separate Illustrator File to demonstrate how the pieces of this drawing would fit together, and to confirm the dimensions I had made:
Looking good so far! However, there were some things that still needed to be addressed.
The parts in this preliminary drawing were definitely not all of the parts necessary for building the mechanism, though, and many of them are not finalized in their design. For example, I had not included any of the aesthetic parts of the project, and I also failed to realize that there would be some sort of lever required to spin the wheel. I also redesigned the wheel per Dr. Wettergreen’s recommendation:
Before adding these parts, I decided to create my first low fidelity prototype out of cardboard. I did this to get a better feel of where I had to restrict rotational motion, and also to better comprehend the how the parts would fit together in 3D (in and out of the page). I resized all of the holes in the original Illustrator to fit the size of the low fidelity straws and began printing out the parts in the Illustrator File from before:
Once all the parts were cut from the cardboard, I assembled them using hot glue, using a thick cardboard piece that I found in the bottom floor of the OEDK as the base. Here’s how it looked:
I did not glue anything down with the intention of testing the actual movement, but I did get a feel for how the mechanism would be oriented on a flat base. However, the dimensions here were a bit misleading, especially in terms of how long my axles needed to be, as I went on to alter some of the design as the fidelity of my prototypes increased.
The next low fidelity prototype was also made out of cardboard, but this time was created with the intention of moving. I used the same parts as when I printed out the first low fidelity prototype (at this stage of prototyping, I wanted to make my mechanism move before I printed out the rest of the new parts I hadn’t designed in Illustrator yet). However, this second prototype turned out to be a total failure for a multitude of reasons. Not only was the place in which the block slid between too flimsy, especially being made of cardboard, but I also had messed up in some of my gluing, and in turn restricted rotational motion in a couple places that I hadn’t intended to.
Therefore, I decided to be thorough about my last and final low fidelity prototype. I started by creating the rest of the non-aesthetic parts of the mechanism. This included:
- A lever
- A new place in which the block would slide between
- A superior support for the wheel, that would actually allow for effective rotational motion.
They resulted in the following 2D drawings, along with the other parts of the mechanism:
As you can see, the shape of the area where the block slides between drastically increased in surface area, resulting in it being less flimsy. The support for the wheel also became a equilateral triangle with a rounded top, allowing for greater stability, and a hole in the middle around which the wheel would rotate.
I then cut these parts and formed my final low fidelity prototype:
Here’s a downloadable link to a video of it moving: IMG_5886
Some of the key takeaways from constructing this low fidelity prototype so many times were being able to better understand what the most ideal order of operations was in constructing the mechanism, and fully understanding which washers needed to not only restrict translational movement, but also rotational movement.
In order to move up to a medium fidelity prototype made from wood and acrylic, there were a couple of things to address first:
- The axles would have to be a material more durable than the low fidelity straws. I went with the ~.3 in. diameter wooden dowels in the OEDK. I would cut these using a hand saw, and attempted to use the belt sander to flatten the cut ends (although this proved to be sort of ineffective, as they turned out a little crooked).
- The triangular support holding up the wheel would have to be more sturdy, as there’s a lot of torque being applied to the body via the lever. I decided to double the thickness of this support by combining two separate laser cut pieces.
- I needed to figure out what pieces of the mechanism would benefit from being made of acrylic rather than wood. I decided that the sliding block would heavily benefit from being made of acrylic, as Adolfo informed me that having a different material for the sliding block and the structure is slides between would allow for smoother movement overall. I also wanted to pick a cool colored acrylic (like blue) and leave the back colored sheet on to give the block a nice effect.
- There would have to be a more reliable way to secure the mechanism to the base. For this, I decided to make 2 layers of 1/4″ wood, with small sections cut out that, along with the help of some adhesive, would help the wheel support and sliding block support stand firmly.
With these changes/improvements in mind, I altered the dimensions of the holes in the 2D drawings to be the same as the wooden dowels I planned on using, and got to laser cutting.
Once cut, I began to realize that this prototype would require different kinds of glue than just hot glue. I decided that I would use acrylic glue to join the parts of the sliding block together, wood glue to join wooden members that had a lot of overlapping surface area (such as the base and wheel support), and hot glue for other connections.
Before I could assemble all the parts into one complete mechanism, I had to undergo a lot of kerf testing to get the base of the mechanism to fit snuggly and firmly:
Once I got the correct Kerf on the smaller piece of the base, I replicated it onto the larger, bottom piece of the base, and placed my constructed mechanism into it, with fresh wood glue applied to the inside (I left the smaller one laying around for my final prototype):
Here’s a link to download the video of the medium fidelity prototype working: IMG_5906
The next step was to determine which pieces of the final prototype would be water-jet cut, and which would be plasma cut. I decided the following:
- The connector between the wheel and sliding block would be plasma cut with 1/8″ aluminum. This is because the rigidity of aluminum (as compared to wood) would be appropriate on a member like this, as it experiences a good amount of pushing and pulling. Also, the plasma cutter would make sense here, since the accuracy of the cut is not essential to the functionality of the part (aside from the axle holes). For this piece, I also utilized the practice cuts within the machine shop to select the speed of my cut.
- The wheel would be water-jet cut using 1/4″ stainless steel. This would not only allow the part to be more rigid, but also an accurate on the wheel would be convenient, as cleaning the dross out of the smaller cuts in between the spokes would have been a bit annoying.
Here are the results:
There was little to no dross on the plasma cut connector! However, I still filed it down, and then proceeded to sand-blast each part to make them smoother.
Before I got to constructing the final prototype, I had to create the aesthetic parts of my project, as well as the associated supports for them.
- A checkered racing flag, that would attach to the sliding block, representing the start of a race.
2. A clown, that would be situated behind the wheel, which would be the visual cue that the clown was riding a unicycle.
Once these were created and printed, I went on to create the final prototype.
Once everything was fastened to each other with the appropriate adhesives, I finished the majority of the parts with a Mahogany finish (did so after to avoid complications with increasing the thickness of the material with the wood stain). Unfortunately, some of the wood stain got onto the wheel and acrylic. Here is the final product:
I also engraved my name onto a wooden square, stained it, and glued it onto the base. However, I had already taken this picture prior to doing that.
COST:
For these calculations, I will divide the costs into labor costs, raw materials costs, and machine time.
Labor Costs: $280
I spent ~28 (probably more) hours cumulatively designing and (many times failing) constructing the device in its various levels of fidelity (sometimes more than twice!). I am valuing my work at $10 an hour.
Raw Materials Costs: $62.44
I am not going to consider the cost of cardboard in my calculations.
I used about 3 of the 40″ x 28″ sheets of 1/4″ wood, making for 23.33 sq. ft. of 1/4″ wood. We can buy 32 sq. ft. of 1/4″ sanded plywood at Home Depot (https://www.homedepot.com/p/1-4-in-x-4-ft-x-8-ft-BC-Sanded-Pine-Plywood-235552/100063669) for $18.02.
I also used about 24 sq. in. of 1/4″ acrylic. The smallest amount of acrylic I could find sold at once was 144 sq. in. (https://www.homedepot.com/p/Falken-Design-12-in-x-12-in-x-1-4-in-Thick-Acrylic-White-Translucent-50-2447-Sheet-Falken-Design-ACRYLIC-WT-2447-1-4-1212/308669093) for $12.85.
I also used about 75 sq. in. of 1/4″ stainless steel and 20 sq. in of 1/8″ aluminum.
Stainless Steel: I found 144 sq. in of 1/4″ stainless steel (https://www.onlinemetals.com/en/buy/stainless/0-018-stainless-sheet-304-304l-annealed-%234/pid/713?gclid=Cj0KCQjw0brtBRDOARIsANMDykadXwZnLVZuqf-hfkEP4fxsicgdJfRmnyuWl6boeTxjH_u53bcU3PkaAgqlEALw_wcB) for $6.88.
Aluminum: I found 144 sq. in of 1/8″ aluminum (https://store.buymetal.com/aluminum/sheet-plate/5052-h32/aluminum-sheet-5052-h32-0.125.html) for $10.48.
In order to join the members, I also needed acrylic glue, super glue, and wood glue:
Acrylic Glue: $2.28
https://www.homedepot.com/p/Lifetime-10-1-oz-Pro-Clear-Siliconized-Acrylic-Adhesive-Sealant-0866PR/202261776
Super Glue: $1.98
https://www.homedepot.com/p/Loctite-0-07-oz-Super-Glue-2-Pack-1399963/310182672
Wood Glue: $2.97
https://www.homedepot.com/p/Gorilla-4-fl-oz-Wood-Glue-62020/306912991
I also used wood stain: $6.98
https://www.homedepot.com/p/Varathane-1-qt-Red-Mahogany-Classic-Wood-Interior-Stain-339709/305502014
Machine Time Cost: $60
I am going to approximate machine cost using lab tech salary ($10/hr.). The amount of time that I was using the sand blaster, miter saw, laser cutter, plasma cutter, and water-jet cutter is being estimated to 6 hours.
TOTAL COST: $402.44 (don’t spend as much time messing up as me!)
