Liz McCormick

HOW TO MAKE [ALMOST] ANYTHING_MAS.863

Week 07 - Molding & Casting

 


Project 02b

This weeks assignment was to design a 3d mold, machine it, and cast parts from it. Sounds simple, right? Here are my struggles.

Project Planning & Prep

 

Project 02a
When I made my Environment Box prototype for Week 05 I noticed that the acrylic wasn't particularly air tight. Though I'm not trying to make a vacuum, I did want to make something that had a predictable airtightness level. Since the acrylic was sitting right in the foam, the joints loosened up as I continued to work with the box. So I decided to make some custom silicon gaskets. The silicon will mostly be burried in the foam, but where it is exposed at the corners, I hope to wrap it in aluminum or some sort of sheet metal. (This is the red layer on joints A and B above).
Project 02c
Since the molding process was still a little foreign to me, I wanted to make sure I had the right design before I went through the effort of milling the wax, making the mold, then finally casting it. So I 3d printed the three different corner conditions. Above is joint A which occurs at an acrylic to acrylic corner condition. I ended up redesigning this piece to fix the interior condition. The small square part in the center wouldn't have anything to hold on to. The image on the left is the final design.
Project 02b
Joint B (left) is where the acrylic meets acrylic at a flat condition (this allows me to use smaller sheets of acrylic to keep costs down and to stay within the confines of the laser cutter bed). Joint C (right) is where the acrylic meets the foam and is a fairly straightforward joint. This is the joint that I used to move forward this week.

Machining

 

Project 02b
Since this was 3d milling, I had to import surfaces into mastercam, instead of contours (lines). The above image shows the geometry file that I imported into Mastercam.
Holy crap, did I just learn Mastercam? I think I did. This was my shining achievement of the week. HUGE huge thanks to Jen and Justin for sitting with me for so many hours until I finally figured it out.
Project 02c
toolpaths, surface parameters, collision detections? Nailed it. I didn't spend much time on the spindle speed since I can override that manually on the intellitek.
Retract depth: 1.75, feedplane: 1.6. Always make sure the values are on absolute. Plunge control: make sure "allow multiple plunges along cut" is selected and check the boxes for "Allow negative Z motion along surface" and Allow positive Z motion along surface"
Mastercam is not particularly user-friendly or intuitive, but the simulation tool is a really great feature. If you don't know what tool or what toolpath to use, just pick one and visualize it. It's fairly easy to troubleshoot once you know what the problem is. Though I'm not sure how much this software will affect my life in the long run, I'm pretty proud of myself this week... even though I printed half of my mold backwards.
Project 02c
For wax, max stepdown can be .5". The default is .1 and it was taking everyone forever. It easily tripled the milling time and mine was a fairly simple job.
Project 02c
Here's the tutorial Justin sent out about using the intellitek. It was super helpful. Just remember, if you accidentally hit the machine extents (too far left or right, etc), you have to re-home the machine before moving on.
Project 02c
I started my job with a .1 stepdown and it was taking forever, so I stopped it and resent the file with a .5 stepdown. Things moved much faster after that. Above is the image of the completed rough cut using the 3/8" bit - I couldn't see anything while it was milling because of all the wax scraps, so it was a fun reveal when I vacuumed all that stuff up.
Project 02c
Here it is with the 1/8" bit doing the final cuts.

Casting the Mold

 

Project 02c
Oomoo was super easy to work with. I measured by eye and didn't have any trouble with this process. After I finished mixing, I spent several minutes tapping the cup on the table to get the air bubbles out (bottom right picture). Thank God Jiamin was there, though, because I almost poured it in without putting any mold release on. I poured a thin bead from high up so that there wouldn't be any air trapped in the goop. After the mold was filled, I tapped it on the table for a few minutes and left it overnight to cure.
Project 02c
When I came back the next day, I thought, oh shit, how am I going to get the mold out of the wax? No fear though, it came right out. It helped that I overpoured the mix a little bit and had little "tabs" to get it started. Everything looked good and clean. No air bubbles.
Project 02c
Here's a dumb mistake - when I was working in mastercam I moved all the parts down so that they would sit into the wax a little more. However, I reimported some geometry from rhino to cut the registration marks (the cylinders) and didn't think to move that down as well. As you can see from the top right image above, these cylinders were too long and the mold would not close all the way without squishing them together. (I actually did pour it like this, so it really wasn't that big of a deal since the material is so squishy). After the fact, I cut them down with scissors to make them shorter. Now the mold closes on its own.

Casting

 

Project 02c
It was at this moment that I realized I cut the wax wrong. No turning back now.
I used wood scraps from the recycle bin to clamp the mold together. This was a delicate dance - because the material is so squisy, I couldn't clamp it too tight otherwise it would deform the mold. I needed to clamp it only tight enough to prevent leaking.
Project 02c
The hydrostone took a bit more effort to mix than the oomoo. I only started with a little bit of water, which was good, because it took a ton of powder to get the right consistency. The directions tell you to measure by weight. Feeling impatient, I did not do this. I started with water (as recommended) and kept adding powder in small amounts until I got the texture I wanted. Calvin (TA) described it as somewhere between milk and melted ice cream. I went a little thicker to nearly waffle batter consistency.
Project 02c
Again, I poured a small bead from up high to eliminate the air bubbles. This mixture was significantly more runny than the oomoo, so I did splash it around quite a bit while mixing and pouring it.
Project 02c
And here it is coming out of the mold. I let it cure for about 2-3 hours and took it out of the mold. It looks like it still had some curing to do, but it was solid enough to work with.
Project 02c
Not exactly the shape I had intended.
Project 02c
Here it is next to the 3d printed piece.

F is short for many things.

 

As I'm finishing up my page for the week, feeling pretty good about myself, I receive an email from Justin, our shop manager, addressed to whole section. I'm just going to leave it here without comment for now.
Subject: general observation of the week
"Nearly everyone I have worked with this week came into the shop not having familiarized themselves at all with the machine, the process of 3-axis machining, the constraints in the tooling, or anything else that could help inform the design. I did not see a single design that had been drawn with the machining process in mind. Many people wanted magical end mills that could be as long as needed, with the smallest diameters imaginable, and could machine with the rapidity and grace of a cheetah. Many people paid no attention to the simulation, which showed them rampant collisions between the toolholder, shaft and shank of the endmill, and their material, and ran the jobs anyhow with predicable results. Some even managed to break short, sturdy, 1/8" HSS endmills in wax, something I have never seen done before on that machine with those tools. In the end, most people wanted us to bend over backwards and do other elaborate gymnastics moves to make their jobs more machineable, rather than understanding the constraints of the tools and the process, and modifying their designs to meet those constraints.
The material, tool, and process must inform your design, not the other way around.
On the whole, I am giving this week a F for our section. Although many of you have machined something, it was done with such scant attention to the key lessons as to be almost valueless. However, I think this failure reflects more strongly on the staff, and the support (or lack thereof) that we offered this week, and I would like to make sure that improves. It is critical to double-back on this assignment, and absorb all lessons learned, before anyone embarks of further machining operations, and we will welcome you to continue meeting with us to discuss this assignment.
If you have any suggestions at all please send them, either to the group, or confidentially directly to me, as appropriate. I can also be available to meet in person to receive any feedback that you anyone would prefer not to put in written form. We need to do better."