Week 9: Molding and Casting

Project Plan

For molding and casting, I thought it would be interesting to try and cast one of the cycloids in the cycloidal drive. I figured that I could likley cast them out of materials that had much better structural integrity than a simple 3D printed part. I am working on the next iteration of the drive itself and so I figured it would be fine to mold and cast a scaled down version of the previous design to test if it would work. For reference, this is the cycloid that I am talking about.

I knew that I was going to be short on time this week becasue I was leaving Saturday morning and would get back mid day Monday which did not leave me very much time to work on it.

The Training

I went to the training that Gert hosted on Thursday demonstrating how to pour and cast. There wasn't anything particularly interesting, he just heavily emphasized the need to remove as many bubbles as possible. I am a mentor in The Deep, a makerspace on campus and they have a vaccum chamber which I figured could be useful for this.

Execution

To create the double negative mold, I figured I would start with the existing model of the cycloid that I wanted as the result, and then subtract it from a body to get the first negative. I already had the 3D model for what I was trying to cast, and so I was able to jump right in.

With that, I then subtracted the cycloid from a box to get the first negative. I also used the revolve tool to create some hemispheres to be used in the alignment of the molds. Neil was very particular about not being able to see the parting line of the cast part and so I figured I would try and hide it by making the seam right at the top surface of the cycloid.

What I now had modeled was the silicon molds that I would use to cast the final part. The next thing that I had to do was design the form to yield these molds. This was pretty simple, again I just made a box and subtracted the bodies from them to get the double negative that I would need.

With them molds printed, I went ahead and 3D printed them. Originally, the plan was to print them on one of the Formlabs Form 4 printers that are in the makerspace that I mentioned above. Unfortunatley, due to my trip over the weekend, I was short on time and needed to move forward so I just printed it on my Bambu P1S. I knew that this would introduce problems regarding the layer lines of the parts and I would have to attempt to remove them from the mold but I was willing to accept that.

To try and get rid of the layer lines, I wanted to try and use chapstick... Anthony talked about using honeycomb and rubbing it over the mold to fill in gaps or the space between the layers and I was curious if Burts Bee's could work. I experimented a little bit with an old print on my desk, a benchy, and it actually appeared to be working. I then tried my best to coat the mold in the chapstick, using a popsicle stick to try and reach in all the areas and then went ahead and poured the mold. I used Oomoo 25.

It says that it has a cure time of 75 minutes but when I checked it was still not cured. I then checked in again around 3 hours after and it had solidified but was still very much tacky and so I decided to just let it cure overnight. I am quite pleased with the result! I was nervous about bubbles in the mold but did my best to eliminate them and it worked! There were a couple things that I tried to do to minimize them. First, I tried to of course not introduce them. Then I found that if you hold the cup at the angle and spin it, the exposed bubbles often pop due to surface tension. I also tried pouring it between cups with a super small stream to eliminate bubbles which worked too.

The chapstick did not work as well as I had wanted it to so most of the layer lines are still visible. Now that I had the mold, it was time to try and cast the part itself. By the time that I was ready to cast, the arcihtecture department had run out of rockite which was my original planned material. It is my fault for being so behind and thus not being able to use that material but I figured I'd go ahead and try DAP's Plaster of Paris.

The instructions on the back say that it takes about 30 mins to cure. It also suggested a ratio of 2 parts plaster to 1 part water. I tried that initially and determined that it was too viscous and I wouldn't be able to pour it into my mold. I slowly added water until I felt that it was worth a try.

Throughout all of my attempts, I started by filling most of the bottom mold and then added the top mold after. I did this to try and fill the mold as much as possible and guarantee that it would spread evenly and also to minimize trapped air in the mold. This is what one of the pours looked like once I put the top on.

My first attempt didn't yield fantastic results, there were lots of air bubbles in the part but perhaps more importantly, the part broke as I pulled it out of the mold.

My second attempt, I tried adding a little bit more water to the plaster so it would flow even easier. I also tried to stur it more slowly as to introduce less air bubbles. I started the pour in the same way, with the bottom mold but when I placed the top mold on, it didn't seat completlety (I thought it had). When I pulled it out of the mold, one of the sides was completely solid and the holes did not go all the way through. There were a bunch of bubbles still and it also broke when I tried to pull it out of the mold.

I'm starting to think that plaster is too brittle of a material to use with my mold and that despite best efforts they will continue to break but we'll see. For my third attempt, I decided to try and give the vacuum chamber I try to see if I could go ahead and get rid of these bubbles. I was able to remove most of the internal bubbles but when I removed the top mold, it revealed much larger airbubbles that had been created becasue the plaster didn't fully spread out. It also broke when I tried to pull it out of the mold.

Thoughts and Reflection

I definitley fell behind this week. I poured my mold on Tuesday and started casting attempts early Wednesday morning. I'm quite furstrated with myself, but at the end of the day it was a learning experience. I probably should have printed and made my initial mold before I left for the weekend and then I would have been fine when I came back. Of course, that was the goal but I just couldn't execute on that.

All that is in the past and doesn't matter anymore. Regarding the actualy project for the week and the experience of molding and casting something, I was quite pleased. I'm really happy with how my mold came out and I think it would be great to cast the final design of the cycloidal drive out of resin. I think it would have much better structural integrity when compared to a 3D printed counterpart and it would also just be really satisfying to see clean, professional looking parts in my final project.

I'd like to revisit molding and casting once I have a final design of the drive system figured out and I look forward to refining my skills as I continue. I feel like I barely scratched the surface.