For this final, we chose the knight as our piece, because I like knights and they are intetesting. Plus I thought that there could be a lot of interesting forms of a horse, like a unicorn. So the first class we spent looking on Thingiverse for a piece, and we found lot of neat parts. Some were really detailed and it seemed difficult to CNC a positive of them. In the end, be chose a cool design that we both agreed on, which is a seahorse looking, aquatic horse piece.

 

 

 

 

 

So once we we chose our design, we had to deal with it to make it a half horse. Also, apparently I have interesting file names. This was the hardest part of the final. So the next class after we picked our design, we had to learn how to use Fusion 360. It sucked. The video was confusing. So we spent that entire class struggling to deal with Fusion 360. Then I think the next class we went through a class tutorial, which was very helpful. We were somewhat productive that class. We first had to import our file into MeshMixer to lower the faces. But I believe that our file already had less than 10,000 faces. But we went through the process anyways. We changed the file so many times that I don’t even remember which one we started with. But, in the latest file we used, we edited the design in MeshMixer to lower the faces, and then exported it out. Then we imported that lower face file into Fusion 360.

 

 

 

 

 

Now, Fusion 360 is relatively simple now that I know how to use it! But at the beginning it was foreign. Fast forward a lot of struggle and unproductive class periods, we finally were able to use the file we were working with. The main struggle that we faces was the red circles in the 3D adaptive clearing that kept appearing. We tried to make them go away by expanding the bounding box, but it did not work. We also tried going through all of the steps we had done so far on another file that was simpler. But, it still came up with red circles, and the file was stupidly tilted. So, once we learned that they were okay to have in the file (after Thanksgiving break), then we started trying to make it on the Carvey with our original aquatic horse file. A helpful tip we got (from Noah and Pedro) was to change the Stepover on the Parallel smoothing to make it lower, so we did that and in Fusion 360 it made more passes, so that it would be smoother. Another trouble we encountered was exporting the Fusion 360 file into Easel. It is very inefficient. We had to go in to Easel, download some file, move the file into a location which took many steps to get to, including showing a hidden file, and then finally getting to export the file from Fusion 360 into the files. Then we had to find the file we just exported, which is in a different location, and import it into Easel. Like Fusion 360, once we learned how to do it, it was an easy process.

 

 

 

Working with the Carvey was frustrating. We have many failed cuts to where the Carvey would start cutting through the edge (top) of the wood, when in the file it was supposed to stay within the wood piece. So I tried moving the bounding box by making it smaller, and also trying to offset the piece more downward. We finally were able to get a successful cut. Then we edited the bounding box so the bottom of the box was intersecting with the base of the horse, that way we would have a way to pour in the mold. Once we did that, we still had issues with the Carvey, like how it would break through the top of the wood. But then finally we were able to get a good piece after “starting from scratch”. Starting from scratch, meaning starting over in the CAM component of Fusion 360. It was not hard to start over since we had learned how to use Fusion 360. During this part, I picked one of the earlier files, where the horse was more defined and had more faces, so it has more detail compared to the first successful CNC part we made. We did change a few things with the roughing settings. Before we had to change the Maximum Rough Stepdown from 5 mm to 2.5 mm, which made the cut take longer. But it did reduce the Carvey jerking a bit, so that was good.

 

 

 

 

With the new CAM component, we ran it and it still broke through the top of the wood.  And once the roughing was fine and stayed within the wood, but then the smoothing was offset lower than it should have been, so we ended up with an awkward looking horse. So then I made another file, and made it so we would use it with the X-Carve. We attempted to use the X-Carve, but were unable to get it started. So at that point I decided to just offset the piece very far towards the bottom, since the issue is that the Carvey would move up in the Y-axis and cut. We also noticed that it would cut more on the right side in the X-axis than it was predicted to do so. So once I checked and made sure that everything was in place and our cut would stay on the wood, we cut it again, and success! Finally. Once we were able to get our first one done, we went ahead and made a mold using the Smooth-Cast Mold Star. The cure time for it was about 30 minutes, but for the first one we left it for 2 hours, and it turned out well. Then after that was done, we had to CNC the other half of the horse. Since our piece is facing a direction, we had to go back into Fusion 360 to make another file. ugh.

 

 

 

 

Back in Fusion 360, we were much more productive than the beginning of the project since we knew how to use it. We found a Mirror function in the program, so we used that to make our other half horse facing the opposite direction. We went through all of the steps and finished that file relatively quick. But at the time the Carvey was being occupied (for many hours), so we tried to make a file with registration keys, in case casting failed. We tried making those files, and got one side of the horse with extruding circles, but we were unable to get the other side with holes in it. So we went ahead and called it a day on the files. Later that day we came back to make the cut, and it turned out well, so we went ahead and molded it and left it overnight.

 

 

 

Now with our two half horse molds, we were ready to cast. We decided to use 300 and 300Q in order to mold because we have used it in the past and it has a very quick curing time. Since we did not have registration keys, we tried various techniques in order to clamp the two pieces together. For our first try we had a rubber band around the two pieces, not too tight to deform the figure, and then had a piece of wood on each side of the mold and clamped it together. It kept it together, but the issue was that the cast did not reach the top part of the horse which is thin and detailed.

 

 

Then we tried to do the same set up, but with 325, which is new for me. I do not like 325. It takes a long time to set compared to the 300 and 300Q. Plus it left everything sticky in the end, so I had to wash out our molds and dry them so we could cast again.

 

 

Then we got the advice (from Wettergreen) to first pour into one half of the mold, and then sandwich the other mold and keep them together. So we tried that, and it worked, but it was messy and we had to move fast. The issue was trying to get the molds to line up perfectly so the horse would not be super tilted or awkward looking. So we tried doing this a few times. We also had to move pretty fast since it has such a quick set time. In one of the casts, the 300 and 300Q started setting as I was pouring it into the mold, so half of the horse’s face is missing.

 

 

Then I tried a different technique, where I rubber banded the pieces together, not too tight to where it deformed the pieces, and poured. I poured in a little, then separated the bottom (top of horse) of the mold, so liquid would reach the top, and then close it. Then I would pour the rest of the casting material into the mold. It was messy, but it worked. So yay! But I did clean up the table after the attempts so it was how I had found it earlier. I kept making them until I had two that were lined up decently well.

 

 

Then I started to make the black ones. I used the black paint that is mixed in with one of the parts of the casting material. At this point I had run out of Popsicle sticks, so I started to use little stir straws. The first one I used only a little bit of the paint, and I mixed it in with Part B before I added in Part A. After mixing for a little, it instantly turned it black. It was cool. Then adding in Part A, it was hard to tell when it was fully mixed, so I just had to tell by the temperature of the stuff. Then I poured it into the mold, using the same technique as before, and waited for it to set. It took a bit longer to set, possibly because I did not mix it well enough, or because of the black dye. But when it set, it turned to a grey color instead of black. When I took it out it looked a bit weird, which is okay since it was the first time trying. It could also be due to the fact I was eyeballing the amount I poured, so it could be why the face of the horse looked a bit gooey. Also it could be because I did not mix it enough, since there were obvious streaks of black in the model.

 

 

Then I tried it again, and I added a little more dye to it, and stirred the dye longer to make sure it was evenly distributed. It does not take a lot of dye for it to work. We were running low on the black dye, so I did not want to use too much. I went through the same process, and was more precise this time with my measurements as well. So then I got a nice black horse.

 

 

After I finished casting, I cleaned up some more, and picked which horses I wanted to go into the final elimination round *dun dun dun*. I picked two white horse and the grey and black one. I decided if I were to try anything, I would try it on the grey one first since it has a melted face, so he was my lab rat. To finish the knight (get it?), I went ahead and used an exacto-knife to cut off some extra parts on the pieces. I made it so that it was less obvious that it came from a two-part mold by cutting off the excess. One of the white horses is offset in one way and the other white horse is offset in another way. The black horse is pretty straight and the grey horse is eh. 

 

 

Then the next day, we sanded the horses down some more. We sanded the areas where the two molds were put together and we also sanded down the base so it would stand straight and flat. I accidentally sanded away too much of the bottom on the black horse to where it has no base now. Originally in the design it has a base which adds height and depth to the piece, but the piece slipped and it got sanded off on the belt sander. So I went ahead and sanded off the entire base so it would not look lopsided and awkward. So now it is just short. Then we sanded the white horse and it ended up with no base as well. Then we sanded the other white horse, and it does have a base. So we will turn in the white and black horse that have no base, so they can be equally short, and have the white horse with the base as an example of how they should have been if I had not slipped up.

 

 

 

Then we also cut off the excess wood from the CNC cuts we made. They look cute. The bounding box is almost a square, which looks very nice.

 

 

I think this project was easier than the midterm project was. This was more of a partner project than the midterm one was. And I really like molding and casting, so casting all of the pieces was fun for me. I think that going through Fusion 360 as a class was very helpful, because we were lost the first day. Then it just took time to become accustomed to Fusion 360 and how it works. Once we were able to do that, everything was fine. We also spent a lot of time trying to get rid of “errors” (red swirly down circles) when we did not need to. I think we spent about a week or so stuck dealing with that. But once we were able to get through that and CNCing, then things picked up pretty quickly. In the end, I think it was a good final and it was very fun. I was also tempted to make horses with all of the colors of dye. Then we could have rainbow knights! That would be cool! Just randomly place them around the OEDK and have the freshman do a scavenger hunt to find all of the knights. That would be an interest thing to do.