To begin this project, I created a drawing in Adobe Illustrator that is an exact, 2D trace of the blade of one of my kayak paddles, called a Werner Shogun. This is pictured above. The process followed for this aspect is similar to that of the keychain and CNC plaque I made previously, view those for more details.
Once I had made this illustration, I transferred it into Fusion 360. In Fusion 360 I began creating sketches based on the initial drawing. These were aligned to a plane perpendicular to the illustration. This allowed me to create a 3D model. This model was made by taking the first plane, sketching the cross-sectional view of the paddle at this point, and then making another plane offset but on the same axis as prior. These offset planes were duplicated and sketches were created for each aligning point on the original paddle. These sketches were then lofted together, which creates a volume by joining the space between two planes. After this, the right-side model paddle blade looked very similar to its real-life counterpart. I created a mirror plane at the base of the paddle, where it would attach to the shaft, creating the left blade, which also aligned to its real-life counterpart. At this point, I created models for both paddle blades in the modeling software. The 3D model is visible below, but note this image is from much farther along in the process with additional modifications.
As you work through this project collect photos, screenshots, videos etc. in order to add them to your website submissions.
Your website should walk through each step you took to complete your project.
Upon the creation of these models, we decided to CNC them out of insulation foam before making a finalized version out of wood. The CNC is able to cut the foam much faster than the wood. We make a stock, or body that encased the blades and met the dimensions of the foam to begin with, and toolpaths for the CNC to follow to mill out the blades. Initially, the toolpaths of the milling operation used took hours to create, so we switched to using parallel passes for the X & Y axes. This exponentially sped up the process.
The first physical iteration of the blades was done in foam. We used a sort of jig that was already on the table, which was just a piece of wood with a 90 degree bend that was aligned to the table. When we performed the flip mill, the wood was not properly aligned the the top and bottom faces of the paddles did not line up. See below (Middle two blades in the second picture are the first iteration)
To fix this problem, we added another operation to the first setup, in which the CNC would drill a hole all the way through the stock and into the table. This allowed us to run pegs into the table and stock when it was flipped, giving us exact alignment. This was tested on the second iteration of the paddle, and worked very well. With this success, we decided to go ahead and make a wood version. This was made of poplar, and the stock matched the dimensions of the second foam one. The flip mill with pegs went very well, the halfs were aligned. These are visible above as the bottom two blades.
Problems arose upon further inspection. The paddle was extremely thin, to the point the wood was translucent, at one point on each blade. We do not know why this happened, and hopefully wouldn't be too much of a hinderance, as the thin spot was not near the edge of the blade. Further problems were uncovered when we began drilling out the end of each blade, where the shaft would attach. The model of the blade was only marginally bigger than the shaft would be, which gave us a very small error margin when drilling them out. We attached the blade to the table of the drill press (turned 90 degrees vertical) with clamps. It was very hard to get the paddle straight and centered under the bit. Once we got them aligned and began drilling, we would have to constantly readjust to keep them conformed. We broke through the thin walls several times between the two blades, rendering them almost useless. We also realized the table was not truly locked at the 90 degree vertical angle we set it at, and would move as we were drilling. Paddle prior to drilling is visible below, as well as when it was in- progress on the CNC.
Answer the following reflection questions at the end of your site.
What went well during this project?
Creating the initial 3D model was easier than I expected, and learning to use the loft tool went well.
What challenges did you encounter?
Aligning the flip mill proved to be quite difficult, as did setting up the toolpaths in the way we wanted.
If you were to do this project again, what would you change?
See Subpage Paddle V2. We widened the shaft attachment point, fixed several of the lofts, especially around the ends. We also changed how we drilled out the ends.