week 8: molding and casting

Prior Experience: 0/5

I have no prior experience with this, except for CAM from make something big week.

This week is molding and casting week. The assignment is:

group assignment:
- review the safety data sheets for each of your molding and casting materials,
- then make and compare test casts with each of them extra credit: try other molding and casting processes
individual assignment:
- design a mold around the stock and tooling that you'll be using,
- mill it (rough cut + three-axis finish cut), and use it to cast parts
- extra credit: use more then two mold parts

Here are Anthony's thoughts on molding and casting/tips for visualization: https://gitlab.cba.mit.edu/classes/863.22/site/-/blob/master/doc/MoldingAndCasting/readme.md

The steps this week are to CAD a design, mill it out of the wax (this is the positive), pour the oomoo into the wax (this is the negative), and then cast it in hydrostone.

Group Assignment

I went in to the archshops with a group of several other people, and Jen introduced us to the week's assignment. She told us about the safety data sheets, showed us previous examples of the machined wax as well as the oomoo molds (a house and a gear). Then we went to the chemicals room and mixed some oomoo/sorta-clear (which is food safe) ourselves, and poured them into the wax mold.

Machined wax:

Mixing the sorta-clear with Diego:

Pouring the sorta-clear:

The final mold:

Individual Assignment

Here is the archshop tutorial for molding and casting week: https://archshops.mit.edu/moldcast.php. It has some good information about the specifics of this assignment.

CAD Modeling: Mobius Strip

I modeled a Mobius strip following this tutorial: https://www.youtube.com/watch?v=IQzEjLYfBoI.

The wax block we have is 3x6x1.5 inches, so for a two-sided mold, the footprint should fit within a 3in x 3in section. The width of the smallest endmill is .125". Also, the walls of the mold should be around .25".

Step 1: Sketch a circle

Step 2: Add a perpendicular line with its midpoint on the circle

Step 3: Go to the Surface tab and sweep the line 0.5 distance along the circle, turning 90 degrees. (A mobius strip turns 180 degrees total around the whole circle.)

A cylinder is not what we want.

Half a mobius strip is what we want:

Step 4: Sweep the other half of the strip.

Step 5: Thicken the surface.

This makes the 2D surface into a 3D body.

I did two thicken operations, one on each sweep.

Then I duplicated the mobius strip and moved it.

It turns out that the mobius strip actually has 180 degree rotational symmetry, so I could just machine one mold and pour the oomoo twice, but I decided to do two anyway so the process would not take as long.

Unfortunately, after cutting it in half using Split in Fusion, I realized that this would require undercuts.

I decided to 3D print the model for the mold, and come up with another design to machine out of the wax, since I could not figure out how to make the mobius strip so that it didn't have undercuts. I probably should have thought harder about my model before I modeled it but I couldn't visualize it very well in my brain.

I split the model in half, and then rotated one of them (which turns out to be the same as the other one, which I didn't realize).

I arranged them in a box:

Then I added the locking cylinders for the two molds to fit together:

Here's the final model for the mold:

3D print: Mobius Strip

I exported my Fusion design as an .stl file and sent it to 3D print. I uploaded to Dropbox, opened the file in 3DWOX on the shop computers, then sliced the layers and added supports.

Then I exported the gcode and uploaded it to a USB drive.

Then I sent it to the 3D printer and started printing it. Here is the result:

Oomoo Time

I mixed up some oomoo (the ratio is around 1:1) and poured it into my mold.

I actually ran out of oomoo and there wasn't enough so after the oomoo dried I mixed up some sorta-clear (also approximately 1:1 ratio) and poured it over the oomoo. They actually didn't stick together so the mold is not very ideal, but it actually worked fine for the next part; I just put the two halves on top of each other.

I came back around 5 hours later to pour the hydrostone.

Hydrostone Time

I added some water to the hydrostone until it reached a honey-like consistency, and then poured it into the mold.

I ended up with two halves (I was going to cut a channel so I could put the two halves together and pour the whole thing at once, but the mold was already kind of janky and I didn't want it to break), and for each half I just poured some more hydrostone and stuck it on top again. I ended up with two mobius strips.

You can kind of see the division line but honestly after a little sanding it looks completely fine.

CAD Design: Dinosaur Pencil Holder

Obviously I did not get to machine the wax for the Mobius strips, so I ended up designing something else to cut out of the wax.

I want to make a horizontal dino holder, based on someone else's previous dinosaur project (http://fab.cba.mit.edu/classes/863.21/Architecture/people/HannahGazdus/week8.html).

Here is the inspiration:

I am copying the stegosaurus.

Here's my drawing:

I changed my document settings to inches and then uploaded my dinosaur as a decal.

I wanted to sculpt the ball, but I gave up because it seemed unnecessarily complicated and I also wanted to make sure my model would actually be machinable. I will just make a thick dinosaur.

I traced it with splines and lines.

Then, I extruded everything.

Then, I filleted the body and mirrored the dinosaur.

I added some more fillets and we are done :D

Here's a render of the final dinosaur:

Now, it's time to put it inside the box so that we can make the CAM file.

I made a box and mirrored it.

I spent a while editing the file so that it would work in the wax mold (I made sure the .125" endmill could fit everywhere I wanted to be cut, changed the teeth, changed the sizes of the walls to be .25", which is what was specified in the archshop tutorial, added the channel to pour hydrostone into the tail, added some fillets, and so on.)

I tried to export as .3mf but I didn't work for the CAM software. The .step file worked but was invisible in Rhino/the wrong color, but Jen fixed it. Now, it's time for CAM and machining the wax block.

CAM

Jen had the CAM file ready and went over it with me. For the toolpaths, because of the way the tools are designed, it's better to step down a little bit at a time to clear material away, because the tools prefer to be cutting sideways than drilling down.

The Dropbox template is a good place to start from (the .emcam file: https://www.dropbox.com/sh/2ibzht24sr2myp2/AACniCgHutA1RI6WwfR15eoja/molding%20and%20casting?dl=0&subfolder_nav_tracking=1), but for each file you will have to delete some paths and add stuff as well.

The blue lines are cut lines and the yellow lines are movement lines. To display the toolpath, click on it and click the wavy lines to show it. The lengths and diameters of the tools are set (in the tool library made by Jen), but when the actual cutting is done, I have to manually change the tools in and out (there are only three tools I'm using -- the 1/4" and 1/2" flat, and the 1/4" ball tip.)

The spindle speed is how fast the tool spins, and the feed rate is how quickly you plunge it. These are set manually by Jen, and they can be found using a chip load calculator. We want these feeds and speeds to be proportional, and you have to figure out through practice what is realistic.
After the rough cut, leave some stock to cut into on the drive, so that the finer toolpaths can cut it away. The stepover is how far apart the lines are.

After the CAM file with the roughing toolpaths, the contour toolpaths cleaning up the outlines, and the pockets for the holes, double check everything and export the gcode, in groups based on the tool. Then we load the code onto the machine and run the gcode by tool, switching out the tools.

Here is the machine: