To start this assignment I started with my 2D drawings and had to make major changes to get everything to fit together. I began by going through all of my measurements and making sure that before I cut anything out, I would have a final product that would fit together and function. I removed extraneous pieces that were included in the original design such as the two bosses and ball that lock everything in place. I decided to use pins instead.

My initial designs looked like this:

As this images shows, I had not yet optimized my design for the final version. I wanted to first cut out all of the parts from cardboard in order to aid the design process by seeing a 3D version of what I was making. Before cutting I used calipers to measure the thickness of the cardboard so that I knew that the pieces would fit together (mainly the stanchions in the base). The cardboard version looked as follows:

I knew, based off of this version, that my basic concept for the design would work. I was able to cut down some of the pieces to fit around the dowel such as the propeller spindle and the pull ring in the center. I thought that I would go ahead and cut a version out of wood (after remeasuring and adjusting for material thickness) in order to further analyze my design.

I spent a long time analyzing this wooden version. I quickly realized that I could not make the propeller spindle and the pull ring work without modifying the design to account for the dowel (as shown by the fact that neither are present in this prototype). I had envisioned that I would be able to quickly cut the parts to fit but I learned that this would result in a subpar final product. I also learned from assembling this prototype that it was incredibly difficult to glue all eight of the disks with the center holes lined up. I originally had the angle of attack on the propellers as flexible…

…the propellers could rotate in the center hole of the spindle because the diameter of the hole was the same as the width of the propeller end. However, I wanted to create a more fixed propeller angle.

I spent about an hour going through each part of the design and writing down what I needed to change:

Some of the highlights of this process are as follows. I had to readjust for wood thickness again because the plywood I would use in the final version was 0.206″ thick while the plywood I used in the second prototype was 0.156″ thick. I wanted to shorten the base by 2.5 inches to reduce empty space. I needed to raise the height of the stanchions by 1/4 inch to give the flywheel more room. I had to make the middle pieces of the pull ring and propeller spindle with cut outs for the dowel. I needed to add small dowel holes to align the eight disks. I needed to determine an attack angle for the propellers and I needed to make a precisely sized hole. I also wanted to make the bottom of the two base plates a full sheet with no cut outs so I lowered the height of the tab in the stanchions by less than 1/10 of an inch so they wouldn’t protrude below the top plate.

Cutting out the final pieces only took one try but was difficult because of warping in the wood and the increase in thickness of the material from the prototype (the focus also changes when cutting such a large area). I did roughly 5 test cuts of different apertures before cutting the final in order to lock in the power and speed settings while ensuring that the holes for the propeller in the spindle and stanchions in the base were all a snug fit. I also made sure that the hole for the drive shaft in the spindle would allow the shaft to spin freely. This photo shows the pieces after cutting. You can see in the lower left the holes cut in the pull rings and the spindles for the shaft, the lack of holes in the bottom base plate, the dowel holes in the disks, and the addition of a nameplate (bottom right). I also cut around the big knots shown in the “b” side of the plywood.

Assembly was the hardest part. I first sanded down all leading faces and edges to remove any signs of burning from the laser. When gluing up my pieces I used both superglue and wood glue to bond the wood. I would spread an even layer of wood glue first and then augment with superglue. Both glues have strengths and weaknesses and I used both in order to maximize their strengths. The wood glue provides a much stronger bond than superglue but requires at least 15 minutes of clamp time to set and then as long as 24 hours to fully cure. However, after that time, the bond is actually stronger than the wood itself. For superglue, it fully sets in less than 30 seconds but does not have the holding power of the wood glue since the wood itself is very porous. Thus, the majority of the glue seeps into the grain and does not bond. The superglue bond can be pulled apart with not too much force (I tested it). So, to recap, when using superglue combined with wood glue you get the strength of the wood glue without having to clamp it (due to the quick bond time of the superglue). This method is not the best for every wood-related situation but is great for small, hard-to-clamp, parts such as many of these.

I continued assembling by using a rasp to remove a small bit of material from the pull ring and propeller spindle to fit the shaft, even though I had made holes for the shaft I still needed to fix the thickness difference between the 0.206″ middle layer with the shaft hole and the 0.337″ shaft. I had roughly 1/10 of and inch to sand and rasp away to ensure a tight fit. I chose this method over having two middle parts because then I would be left with too large of a gap and no great way to fill it.

When assembling the eight flywheel disks I used the small dowels mentioned above to ensure the lineup was perfect:

First I would place each disk on the four dowels and ensure it fit properly before adding glue and sliding it down the rest of the way. I glued the flywheel in two halves and sanded each using the drill press as a lathe (on low rpm) to do so:

After gluing the two halves together, I progressed to final assembly. This meant drilling holes in the main shaft for the string and for the two pins I integrated to ensure the shaft stayed centered. I used a small dowel for this and a 1/8″ drill bit on the drill press. Knowing that drill bits have a tendency to run (even on flat surfaces), I made sure to use a punch to mark the holes so that the bit would know where to go. This worked perfectly. I test fit the parts and realized that the shaft was sticking a little too much in the stanchions. I used a drill bit that was basically the same size as the shaft hole to deburr it. This worked perfectly. I fit everything together and glued everything together.

This is how close I was able to align all of the holes using the small dowels:

I sanded everything down with 240 grit sandpaper (getting rid of excess glue and smoothing out the surface). I then finished everything with Minwax clear gloss. I would have preferred to use an oil-based finish. I find that polyurethanes kill the warmth of the wood and are also toxic. An oil based finish would have also required less care around the moving parts of the project since it acts less like a glue. The one plus of the polyurethane is that it does add a more protective coating than the oil and for a piece that will be in use, this may be a better option. On that note, I did have to make sure that I wiped down all of the excess poly and made sure that none of it seeped into the wooden “bearings” and glued them together.

Even before finishing, the flywheel ran beautifully and smoothly. Luckily I did not let any finish get into the bearings and thus the only issue I had post-finishing was squeaking from the wood-on-wood contact. I added a little tallow with a toothpick and this fixed the squeaking.

Here are photos of the finished product:

Here is a video of it in motion: https://www.youtube.com/watch?v=8mQdqgYC1hk

(check out my other woodworking videos if you’d like)

I was able to get roughly 15 seconds of spin time. I think that is great, especially considering that it is a wooden bearing. I used roughly 1.5 feet of string but I assume the speed and spin time would increase with a longer string.

Now, for the elephant in the room: I messed up a small element of the assembly. When I read the assembly instructions I saw that it said “Glue everything together.”

I assumed that the instructions meant EVERYTHING when they said “everything”. In fact, they may have. However, gluing the pull ring in place, as I did, makes it impossible to pull a string wrapped inside the ring. I believe that the ring was meant to spin freely and thus keep the string from getting tangled in the wheel. Although it would be impossible for me to take off the now-glued pull ring, I think it does look good. As shown in the video, it not just still runs, but still runs great! All that needs to be done to use it is wrap the string tightly around an exposed area of shaft and pull! Since the string is not tied to the shaft, it cannot get tangled due to the fact that it simply gets pulled off.

I had fun with this project. I thought of it a lot like a reverse jigsaw puzzle. You start with a completed image and have to think about what pieces you need to put it together. I am very pleased with my final version even with its one hiccup.