Week 8 - Molding and Casting

CAD

I thought about molding and casting the rest of the KBOS buildings to use it for my final project. However the buildings would simply end up extremely tiny (at most 1 cm in height), so I chose to mold and cast the air traffic control tower that I 3D printed in Week 4 to compare and contrast the two.

Anthony recommends building the part and then building the walls around it until you have a CAD of what the wax would look like. I split the building into two and at first tried to fit both halves onto the wax, so I could use one oomoo mold to cast. (In retrospect, this wouldn't have worked anyway, since I would either have had to glue the two casts together or made a second mold.) The building was too big to fit entirely so I stuck with one half. I forgot at first to ensure that the building was centered, so the two molds I would cast could fit together perfectly, and I had to spend some time fixing that.

Initial CAD of one part mold (left). Sketch of wax box (center). CAD to be milled (right).

If a two part mold, register the top and bottom halves.

CAD after adding indentations for the mold to fit together

The CAD file can be found here.

CAM

This week helped validate my choice of Fusion for CAD, as Fusion has an excellent interface for CAM. I referenced Anthony's recitation documentation for setup settings, then Anthony helped me generate tool paths. To rough out and finish the shape of the tower, we used adaptive clearing, parallel finishing, and contour finishing.

Adaptive clearing (left), parallel finishing (center), and contour finishing (right).

It took Anthony and me a couple of tries to find tool paths that would mill the window indentations. Pencil finishing turned out to the best choice.

Milling the windows using scallop (left) and pencil (right) finishing.

And here are all the CAM tool paths! Estimated machine time is about 9.5 minutes, though Anthony recommends just doubling the suggested mill times for a more realistic estimate.

Compilation of all machine tool paths.

Milling

We're using machinable wax, which can be sanded afterwards and even smoothed out with a heat gun. We hot glued the wax to the machine bed to keep it stationary. Calibrating the axes of the mill involves using a small metal plate, rotating the threads of the tool to match intended axis, and running an electric current through the tool head.

Original wax block (left). Calibration of X and Y axis (center). Milling (right).

Unfortunately, the machine's router seemed to encounter a bug, causing the tool head to "hiccup" every so often, so we stopped my milling job partway so Anthony could figure out the problem. Fortunately, my mold had been sufficiently milled out such that I'd be able to use it to mold and cast, even though it was a bit rough, and one of the legs had gotten damaged from the bug.

It turned out that the rack and pinion had gotten loose from machine vibration from all of the Week 6 OSB, causing it to lose position and skip a gear tooth or two. The machine then had to stop and accelerate to catch up, resulting in the "hiccuping" we'd seen. Anthony tightened the spring that sets how closely the rack and pinion are engaged and mentioned that it should have a lock washer or nut to reduce the effect of high frequency noise.

Machine debugging was needed, but the show must go on. Final mold on the machine bed surrounded by the wax flakes that had been machined away (left) and successfully removed from the machine (right).

Removing the wax mold from the machine bed turned out to be the most difficult step in the process. I spent a number of minutes trying to cut through the hot glue with a box cutter and attempting to pry out the wax with a spackle knife and for whatever reason somehow alarming the rest of the section.

Molding

Some fun facts about Oomoo! It's actually spelled in all caps. The shelf life of unopened Oomoo is a year. It has a 25 min potlife (the amount of time you can leave it in the cup for and still use it). Molding with Oomoo means that internal bubbles do not matter. Otherwise the usual process, particularly for thin parts, is to reduce bubbles is to blowtorch them to break the surface tension. Oomoo is good until 500 degrees F. (Algenate is another mold-making material, for objects that can't get hot.)

How To OOMOO:

Use gloves and clothes that can get dirty. (I got lucky this time by being extremely careful, and Dave lent us an apron.)
Determine the volume of oomoo needed by pouring water into the wax mold, then into your cups. Split the water evenly between two cups and mark the water line with a Sharpie (or in my case, aggressively with a pen); Oomoo ratio is 1:1 by volume.
Dry off the wax and cups with an air gun.
Mix the bottles of the oomoo silicone rubber compounds first with a wooden stick before pouring the designated amount into your cup. Overshoot your lines a bit.
Pour the more viscous liquid into the less viscous. Shape the cup's opening into an ellipse, tilt the mold if you want (if it's a one part mold), and lift the cup up as you pour to form a thin stream.
Mix until the color is uniform (about 2-3 min). Mix slowly and not as if you were beating eggs.
Tap the sides and bottom of the wax on the table to get rid of bubbles.
Let sit for 2-3 hours to cure.

Oomoo is messy (left). Completed mold 1 with mold 2 setting (right).

Once the Oomoo molds were ready for use, I cleaned the edges up with an exacto knife to compensate a bit for the CNC mill's bugs.

Both molds complete!

And now I'm ready to cast my tower!

Some pre-cleanup closeups.

Casting

Some fun facts about casting:

Add powder to water not water to powder. Powder to water ratios by weight are 100:22 for hydrostone and 100:20 for drystone. These aren't exact, as you can adjust based on viscosity/texture.
Use rubber bands or a clamp to hold the molds together. The pressures at the bottom of the mold requires more clamping force, and the density of your cast material makes a difference.
Use cardboard to keep the Oomoo from squishing up on itself.
Pouring excess material into the mold provides more pressure and material to fill in bubbles.
Hydrostone and drystone both require 30 minutes to demold.
Drystone can be dyed (Oomoo can be too), a fact that Anthony offered to demonstrate for us and instantly regretted when we chose flesh-colored dye. I sadly forgot to add this dye to my cast.

Given that my mold wasn't fully milled, the registers weren't functional, so I took extra care to line the molds up. I tried drystone at first, but my mixture was too viscous to be easily poured into my mold, so I switched to using hydrostone.

Hole to pour in the hydrostone (left). An ice cream sandwich (center). Rubber-banded and ready for casting (right).

After 30 minutes, the tower was ready to demold, though multiple days are needed for it to fully cure. The excess hydrostone at the top of the mold (that filled in the hole above the tower) broke off on its own.

Post-casting results! It stands!

After some filing, here's the final product!

Final KBOS ATC Tower cast.

Thanks to Anthony for his excellent recitation and CAM help, Dave for his apron, and Ben for making challenge coins for all of us.