Leaves fall from trees?

• bsg4

Hello friends!

Over the past couple weeks, I have spent countless hours creating my mechanical machine for my midterm project! Here is a quick preview before I go through my process (and more importantly, some of my more significant struggles) of creating it! 

Also, my apologies for not providing pictures of every step of the process. I often got caught up in the manufacturing and assembling of my machine that I forgot to take pictures, so I will do my best at explaining and describing everything as I go.

Now let’s began!!!

DESIGN & 2D DRAWING 

I began this project my brainstorming possible themes and the purpose of my motion. I decided early on that I wanted to do something with nature. After some time spent discussing with my roommates, I decided that I wanted to create a model where leaves fell from trees. However, I wanted these leaves to reset themselves (ie. go back to the top of the trees) without being seen, so that they can fall again.

My next step was to decide my type of motion. I immediately decided I wanted to create some motion where the user would be able to continuously spin the handle without having to change directions of their motion. This would be my first step to increase the robustness of my system.

With these requirement, I starting looking through possible motions on 507 mechanical movements that could make this kind of motion. After doing some initial research, I realized that an elliptical path would be perfect. I could expose the leaves during one side of the elliptical path as they were falling, and then hide them on the top, bottom, and other side with a combination of trees and grass.

To create this elliptical motion, I initially chose mechanical movement 172 that claims to create an “egg-shaped” elliptical movement.

With this initial movement chosen, my next step was to begin my 2D drawings and design. I began by just sketching an idea on the white board at the OEDK.

I then started my more detailed design, ultimately choosing to create a CAD model, since I am much more comfortable with OnShape (an online CAD software) than I am with Adobe Illustrator. Here is the start to my CAD model using this motion.

While making this CAD mode, I quickly realized that this type of motion would make things more complicated in the long run and was not the most effective way to create elliptical motion. So I went back to the drawing board to determine if there was a better chose of motion for me to use. I eventually found movement 35, described as variable rotary motion produced by uniform rotary motion. I decided that this motion would be more appropriate for my project.  And it had the added bonus that the leaves could rotate while they fell since I planned on attaching them to the axle of the small gear! For the purpose of the rest of the explanations and descriptions, I will use the following terms:

  • elliptical gear, the large gear in the center of the system
  • pinion, the small gear that rotates around the elliptical gear
  • bar, the bar that holds the axis of the pinion and rotates about the center of the elliptical gear
  • spring, the contraption creating tension between the pinion and the rotating axis of the bar that causes the pinion to stay in contact with the elliptical gear

After changing my motion, I began to re-design and create another CAD model.

My first design problem was to figure out how my motion would actually work. I realized that the large elliptical gear needed to be fixed in place so that the leaves could follow it to go on their elliptical path. However, I also wanted to be able to have a bar and spring on either side of the elliptical gear & pinion as to ensure that the two gears stayed parallel to one another. I realized that these two would not work together.

If I were to create a bar & spring on both sides, the elliptical gear would have to float in space, unless I created some sort of two concentric rod system where one is the center axis of the bar, while the other holds the elliptical gear in place. I ultimately decided this solution was unnecessarily complicated, so instead I designed an extra wall on top of the gears that would serve to keep the gears parallel with one another.

The next design problem I had to solve was how to transfer the user’s input of turning a handle to actually rotate the bar in the center of the elliptical gear. I decided that I wanted the handle to be in the front, in an easy to reach location, that would not block the view of any of my motion. I also wanted that rotational motion to be transferred to the elliptical gear somewhere hidden so that it wouldn’t distract from the rest of the motion. My solution to this was to have a series of simple gears (following 507 mechanical movement 24) that were on the back of the machine. The other side of this wall would hold the elliptical gear and be the basis of the pinion’s movement.

After making some of these design decisions, I made the majority of my parts on this CAD file, but didn’t end up putting all of it together into the assembly, in the interest of time. Here is a partly finished version of the complete assembly (both the front and the back).

While creating this CAD model, I realized that the grass and trees can serve to not only block the leaves while they move back up, but they can also serve to support my rotating axis, so I adjusted my parts accordingly.

LOW FIDELITY PROTOTYPE

After creating this CAD model, my next step, was to start my low-fidelity prototype. This step of the process had two separate prototypes for me. First, I laser cut my elliptical gear and pinion to check if those worked together. I initially designed my pinion to be in the shape of an elliptical, as this would cause some side to side motion of the leaves while they were falling. Using the rubber band and bolts to act like the bar and spring, to ensure that the teeth of both gears stayed in contact with one another at each time.

However, after trying them out, I realized those gears did not mesh well together and were going to prove difficult to ensure that they meshed well. I adjusted my pinion to be circular instead, and created                      many different versions of my pinion, with different teeth number and height until I got some that meshed well together.

The other side of my low fidelity prototype was my cardboard model. I used my CAD files, created drawings (dxf files) of them, and imported them into Illustrator. Here is an example of how the drawing files of some of my parts looked in OnShape.

I then laser cut all my pieces out of cardboard, and assembled the model with mostly hot glue and some dowels, and a rubber band. Here is the illustrator file of some of the pieces that I cut out. Most of the time, I would just create individual illustrator files for each part from OnShape, so I won’t include all of those pictures for the sake of brevity. 

This version of the project was mostly used to confirm that everything fit together as desired. I also was able to confirm that my gears on the back of the model worked as I expected them to. However, my main mechanism was not functioning in the cardboard model. I initially assumed this was due to the cardboard, but I will discuss later how that was not actually the case. Anyway, here is a picture of my cardboard model… not the prettiest thing, but it definitely satisfied it’s purpose of proving that everything fit together as expected.All things considered though, this cardboard low fidelity prototype was pretty easy to make, but ultimately was not super super helpful for me as it led me to ignore some blatant problems of my design that I later discovered while creating the wood version. I did learn that my gears on the back (which I made by hand with OnShape’s sketching tool) meshed beautifully however which was quite exciting!

 

MEDIUM FIDELITY PROTOTYPE / FINAL PROJECT

As I have pointed to, my next step was to create the wood version of my project. Initially, my plan was to create a working model, take it apart, stain & spray paint my pieces, and then put it back together. Clearly, that didn’t happen. I will get into the detail later on, but long story short, getting my model to actually function took way longer than expected. Additionally, by the time I got it working, I couldn’t take any part of the system apart without fear of breaking the entire thing. To provide some additional, foreshadowing into how painful this experience was, here is my stock pile of wasted wood by the end of the project (it was way better than it looks in that picture!).Anyway, I began creating this version of the leaves falling from the tree by importing all of my CAD files to drawings (dxf files) and then into illustrator, like I did before. However this time, each of my pieces required some additional work before they were ready to cut. I decided to add finger joints to each piece to make assembly easier. Similarly to how I created my finger joints for the laser cut box, I used a combination of rectangles. The blue rectangles are used as cut-outs, the red as inserts, and the gray for spacing.With my Illustrator file done, I created my first set of pieces using the laser cutter. I assembled everything together, and my mechanism wasn’t working particularly well. At this point, I was primarily focused on getting the elliptical system to work, as I had already proved by other gears worked with my cardboard prototype.

However, I will point out right now that I decided to double up on all my gears, meaning each one was double the thickness of one sheet of plywood. This was probably the best decision of this project because it ensured that the gears would continue to work even if they were slightly misaligned.

Now back to the elliptical system! The rubber bands and springs that I was using was pulling the pinion’s axle and the pinion forward towards the center. The wall helped keep them parallel but was introducing too much friction so the pinion was not able to travel around the elliptical gear very well.

With some help from Scout (a wonderful lab assistant), I realized that the rubber band spring system on my bar was not going to work with the rest of the design of my project. To be effective, the system really needed a second bar and spring system on the back, but if you recall from earlier, that would cause additional complications and would force me to re-design the entire project.

Instead, I ended up trying to rotate the pinion about the elliptical gear without the rubber band and it worked much better. I realized that the elliptical path on the forward most wall was kind of a little bit effective in keeping the gear in place. It was by no means perfect, the pinion would often disengage at some points and just freely float through space. Despite these imperfections, it gave me the idea to create another elliptical path on the back wall for the bottom of the dowel to be guided in. These two paths together would force the dowel to stay the correct distance away from the elliptical at all times, ensuring the pinion would stay engaged with the elliptical gear, while getting rid of the need for the spring at all. I know this idea meant that I was deviating from my initial movement choice, but I decided this was a necessary adjustment to ensure that my project would function correctly.

First, I adjusted the base plate to have room for an additional back plate. I then just created a copy of the front wall to use as my additional elliptical path on the back wall to test it out. Due to a combination of assembly and dimensional problems, this path was not effective. The path was too wide, and the pieces were not centered, so this did not help solve any of my problems yet. The pinion did not disengage as often, but it was not smooth, and was not able to rotate around the elliptical gear very effectively. You can watch a video of the struggle of that gear / path system here.

A few tries later, I created these pieces in Illustrator to map out my paths are accurately as possible based off the dimensions of the elliptical gear and pinion. Here you can see the offset ellipticals that align with the two sides of the dowel while enabling the teeth of both gears to stay aligned and engaged. I used this layout to create the paths. I also added an additional alignment hole, offset from the center, to the path, elliptical gears, and the back plate which helped significantly in ensuring that everything was assembled as it was intended to be. 

After creating these new paths, I remade the majority of my parts. To ensure that they worked before glueing everything together, I taped the base pieces together and was able to test the system out. I confirmed that it worked and then permanently attached pieces together with wood glue.

Somewhere in the middle of this process, I decided that the bar of my machine should be made out of metal on the water jet. I made this decision because in my cardboard version of my prototype that was the weakest part that broke when I attempted to turn the gears. I realized that it would need to be pretty strong if this machine were to function. Water jetting this piece was an additional journey because I had designed my entire layout around the thickness of bar being the thickness of the wood. This meant that I needed to find an equivalent sized piece of metal as to. not have to redesign anything. This was much easier said than done, and resulted in some fun time spent in the wet lab configuring my cutting path and the metal piece to align correctly but eventually we were able to get it. And a few minutes digging around in the bath of the water jet never hurt anyone right? After water jetting the metal, I sanded both sides with numerous grits of sand paper to help the bar glide across the wood and limit the friction between the two pieces.

After I reassembled with my new paths, I continued to face problems, because what’s an ENGI class without any problems to handle??? My next problem was that the metal bar was not rotating with the axle of the big gear right behind it. I tried super glue, hot glue, wood glue, and eventually epoxy and none seemed to be effective. Probably due to a combination of the adhesive, my inability to use said adhesive, and my impatience. With the advice of Adulfo, I decided to make a little finger joint thing that would connect to the dowel, and then also have a finger that sticks out into the slot in the bar that would force it to rotate. This saved my project, because at this point it was impossible for me to take any part of this apart, without breaking everything. I created this finger piece on the laser cutter, and then epoxied every open spot between that finger piece, the bar, and the dowel they spun on.

I left that to dry over night, and when I came back, the elliptical mechanism was working! And even better yet, I was able to make it work by turning the handle – it was honestly a miracle. After I got this mechanical movement working finally, I had the time to add the leaves that were actually falling. I attempted multiple things, but ultimately decided to just glue on a set of leaves to the axle of the pinion and it seemed to be very effective. Here are some of my failed attempts, when I tried to put the leaves around the dowel, and then my successful one, with a solid circle that was just glued on to the dowel!

While testing it out, I realized that I had not accounted for enough trees to hide the leaves and mechanical system the entire time that they were resetting back to the top. To address this, I added another smaller tree on the left side of my project that blocked off the rest of the mechanism’s secret inner workings from the user.

The last practical thing for me to add was some spacer / washer things to keep everything in replace. I quickly cut a bunch of little tight-fit washers that I placed on dowels throughout the machine to hold gears and objects in place and ensure adequate spacing. These spacer / washer things looked like this: 

With the basic mechanism working now, I decided it was time to add some pizazz and make the leaves pop since they are the star of the show anyway. I decided to use the vinyl cutter to create the same pattern as the leaves in multiple fall colors. Here you can see the vinyl that I used and the stickers on my leaves!

I then had a bunch of leftover leaves, so I used those to decorate the rest of my the base of my machine with old fallen leaves. Last but not least, I added a name plate with some basic information about the project that is spotted in the bottom lefthand corner of the machine.

And here is the final product (from the front)!

And from the back, side, and top.

And here is a video of it working! I have realized that with enough vaseline, anything will move, especially gears, dowels, and metal bars. But honestly, all things considered, my machine runs way smoother than I was expecting based off how the last couple days of work went.

Overall, this project took me an insane amount of hours between CAD, manufacturing, and assembly. I learned a lot, and definitely feel more proficient in CAD, laser cutting, and water jetting which is a net positive. However, this project caused me lots of frustration as well.

Because of this, there are a handful of things that I would do different in the future.

  • First, I would consider assembly while I was designing my machine. I would think about how I could design it to ensure I could take it apart without breaking anything in the process. I would ass some additional supports to ensure that the walls and everything stay parallel.
  • Next, I would like to have had more time to stain my wood, or paint it or something to give the entire machine a little more character.
  • In the future, I think I would also plan on choosing a simpler design. I didn’t really realize how complicated it had gotten until it was too late and I had progressed too far to turn around.
  • I would also build extra supports that would ensure the walls and gears stay parallel with one another since that would improve the smoothness and effectiveness of my gear system.

Overall, I am very happy with how this project turned out! I successfully made a machine where leaves fall from a tree which is what I set out to do from the beginning. Even though my design and plan changed a lot throughout the process, I am proud of myself for sticking with it through lots of frustration and stress, and I definitely learned a lot through it.

Now for a quick cost breakdown:

  • Labor: 40 hr * $15 / hr (my wage from this summer) = $600
    • ~ 12 hr on CAD
    • ~ 4 hr on illustrator adjusting and readjusting files to cut
    • ~ 6 hr manufacturing parts
    • ~ 18 hr assembling, trouble shooting, fixing problems etc.
  • Vinyl cutter usage: 0.5 minutes * $10 / hr = $5 (estimated based off laser cutter costs because I couldn’t find a rent-a-vinyl-cutter website)
  • Laser cutter usage: 4.5 hours *  $15 / hr = $67.5 (FACTORYeNOVA hourly rate)
  • Water jet usage: 1 hour * $20 / hr = $20 (from WardJet estimated cost)
  • Adobe Illustrator usage: $21 (one month of Adobe Illustrator)
  • OnShape usage: FREE

Total: $850.86

Total w/o labor: $250.86

Total w/o labor and machine time: $137.36

Anyway, that’s all I got for now. Time for me to go catch up on sleep and my other classes and commitments!

Bryn

Print Friendly, PDF & Email