~ Casting and molding.
~ post date: 2017.4.24, recent update: 2017.4.26

Background

Molding is the process of fabricating by shaping pliable or liquid materials in a rigid formwork. Casting is the process of pouring a liquid material into a mold to achieve a desired shape when the liquid cures. In this post I will focus on making a silicone mold from milled wax and then casting other materials, such as liquid plastic into the mold.

Contents
  • Star puzzle : Child's play?
  • ...Oomoo...
  • Liquid plastic : Smooth-Cast 300 Series
  • Bioelectrochemical testing pod
  • Jump : Index
  • Project files
  • Download starPuzzle mold form STL
  • Star puzzle : Child's play?

    For my niece's birthday, I three-d printed a star puzzle. The toy however could never be dismantled after I muscled it together due to poor printer calibration on my part (a running theme in this blog). Further, duplicating the toy many times over comes with the time factor of three-d printing. Hint : slow. So, I thought it would be fun and good proving grounds to revisit this project in this exercise. And, the other day she asked if I had made the better version of her toy yet and gave me no leniancy for time spent building my first robot.

    These photos set the scene. I downloaded the STL file from a previous student of the How to make Almost Anything course at MIT.

    In a sea of hongbao (red envelopes stuffed with money) this little broken puzzle was a champion.

    I remodeled the toy into a mold measuring about 37 x 79 x 17 mm. There are two inside angles the mill will not be able to complete. I will try to hit those with some hand tools before making the mold.

    When I thought more about casting, I realized getting a perfect cast is going to be tricky. Using this method, I would only need to mill one form, cast it twice and then when I cast the final object I will have some excess material inside my air / casting channels. Looking at this now, I might edit that channel so it bottlenecks briefly before the finish surface. That could help guide a blade and make it easier to snap off the extra material.

    I thought if I oriented the final piece so the top surface was level, there might be less trimming involved. However, I would need to either mill two forms, or drill into one mold after it is made.

    Inspecting the final casting phase, I am concerned there is a chance air bubbles could get stuck in the square corners.

    After consulting the regional review group, we decided to pursue my second option by cutting one form and adding a second material to the form to make the air and casting channels for one mold.

    I used Modela Player 4 to generate the tool paths. You can do that too. First export your file to STL format then open it in Modela Player. I formatted the model to only include cutting surfaces before exporting. In Modela, check Set > Model > Origin and ensure the model origin is where you assume it to be. Then you can start making processes.

    It would be wise to start with a rough cut.

    Cut the "top" side of the surface.

    I chose a 4mm square bit. Different bits, different advantages. Remember, one 4mm square cut bit can be very different from another. For instance, these are two 3mm square endmills, one is capable of more than 2x the cutting depth than the other. This is an important consideration for walls perpendicular to the XY plane.

    Double check the cutting area is correct visually and by the numbers.

    After simulating the different cutting patterns, I found countour lines preferable. It offered a nice balance of speed and precision. For the cutting parameters I kept the defaults for modeling wax, except I changed the cutting in amount to 3mm.

    Next, I setup a finish cut process using a 3mm square bit. I might have liked to try a 2mm square bit but there were none in stock in the lab today. We did not think a round bit, another option, would make a big difference on the sloped sides of the model.

    For the finishing path, I chose scan lines in the X direction to see how the model looked after. It was a quick job for the machine; if I wanted a better finish I could cut more after (for instance another path in the perpendicular direction).

    In the cutting parameters I changed the path interval from 0.1mm to 0.3mm, a 10% offset.

    When the jobs are satisfactory, select Create Toolpath and there are some options to preview the job in 3D and get time estimates, which in my experience were hit and miss, big time miss.

    When finished, select Cut and output the toolpaths to individual files.

    Roland VPanel for MDX-40A works much in the same way as the application for the SRM-20. Set the origin and kick it. The Z origin does not need to be as precise as is the case when milling PCBs. In fact, the first cut of the rough cut is set to 0.2mm Z.

    Cut the files individually, sequentially, and make tool changes when necessary before cutting.

    While cutting, some of the removed material will accumulate in the cutting zone. Periodically remove this material so it does not hurt the form. It can be collected, cast into a form and reused later! On the Roland Modela MDX-40A press the View button to pause the job. Long press the View button with the cover closed to resume after clearing the debris.

    The rough cut is on the right and the first finish cut on the left. The rough cut took about 10 minutes and the finish cut 30. I decided I would like to make the model smoother so I did another finish cut with a 1.5mm bit offset at .1mm which needed nearly two hours.

    I had to make a new bit as this tool was not in the application's inventory. Options > Add/ Remove Tool

    The cutting depth of the 1.5mm bit was not sufficient for the outer walls of the form which is fine because I am not seeking higher resolution there. I reduced the work area by 3mm on all sides to add the necessary space for the endmill to work.

    Nice looking form. Now I can get busy with making molds!

    Download project files

    ...Oomoo...

    "OOMOO® 25 & 30 are easy to use silicone rubber compounds that feature convenient one-to-one by volume mix ratios (no scale necessary). Both have low viscosities for easy mixing and pouring . . . vacuum degassing is not necessary. Both products cure at room temperature with negligible shrinkage." These silicones also have great tear strength.

    Oomoo is a two part substance which requires thorough mixing before use. I am using Oomoo 30, which has a ration of part A : part B, 100 : 130. If you trust the buckets, you can do that. You might want to check that datasheet though.

    I filled the form with water to get an idea of how much oomoo I would have to mix. That word will never get old.

    After some maths, I had poured two paper cups with two parts. Paper cups are great for mixing small amounts. They are flexible and have hard corners and straight sides making them easy to scrape. Stir the two together thoroughly. Try to get the air bubbles out, scrape all the sides and watch the color change.

    I primed the mold by painting some of the oomoo onto the critical surfaces. Hopefully this helps me to avoid air bubbles in these areas.

    After slowly filling the form with my oomoo mixture, I shaked and pounded the hell out of it for several minutes to entice air bubbles to the surface.

    After an hour, you may see the edges start to peal from the wax and when touching the oomoo, it is no longer sticky. You can even slowly peal the edges back with your fingers to get an idea of the state of curing of the oomoo. Curing of the oomoo: one of my favorite D&D spells. When the time is right, whip it out.

    I released the oomoo after 2 hours. That may have been rushing it becasue the datasheet recommends six hours before pulling the mold and another 4 hours of post curing.

    Looks good. Rinse and repeat. I will set it and pull the other half of the mold in the morning.

    The second mold is better than the first. I used more oomoo accidentally when pouring for mixing, and the thickness seems to help. Also, this mold had proper time to cure inside the form. The first mold is ever so slightly curved and the second not. I cast a third immediately after pulling this one.

    The fit is great.

    Using a small box cutter, I removed a channel for casting. The first time trying to use this channel, I had problems getting liquid plastic to pour through the bottom side of the hole so I doubled the size of it. For the liquid plastic, it might be a little too big now.

    Download project files

    I will post links to resources I have found helpful here.

  • Oomoo Datasheet : Oomoo 25, Oomoo 30.
  • Liquid plastic : Smooth-Cast 300 Series

    The first substance I attempted to cast was Smooth-Cast 300 Series, a liquid plastic with a curing time of just 10 minutes. Here is an overview from the datasheet:

    "The Smooth-Cast® 300 Series of liquid plastics are ultra-low viscosity casting resins that yield castings that are bright white and virtually bubble free. Vacuum degassing is not necessary. They o er the convenience of a 1A:1B by volume or 100A:90B by weight mix ratio. The di erences between them are pot life and demold time.
    These resins readily accept llers (such as URE-FIL® 3 from Smooth-On) and can be colored with SO-Strong®, UVO® or Ignite® color tints (Smooth-Cast® 325 series accepts pigments better than the Smooth Cast® 300 series). Fully cured castings are tough, durable, machinable and paintable. They resist moisture and mild solvents. Applications for Smooth-Cast® 300 Series Liquid Plastics include reproducing small to medium size sculptures, making prototype models, special e ect props and decorative jewelry."

    Start by analyzing the datasheet. I am using Smooth-Cast 300. 3 minutes of pot life and ten minutes for curing. Rapido. That means I need to mix the two liquids, 1:1, stir, add color and cast inside of three minutes.

    Gather the materials. Smooth-Cast 300 mixes well with S-Strong colors. I take a dab of color on my stick and mix it with the A side of the plastic before adding the B side. Then, spray some of the release agent into the mold and seal it together. I used cardboard to make the stresses even across the mold.

    This time, I checked the volume in my original Rhinoceros file.

    Command: Volume
    Volume = 10924.7095 (+/- 1e-05) cubic millimeters

    !:1 mixing volumes so I need minimum 5.5g each. I used 7g to be safe. Mix the two sides, and pour it into the mold. As I am pouring, I take time to move the mold around to try to get all the air bubbles out and pre-soak all the interior surfaces before filling.

    Wait ten minutes from the time you first mix the materials together and then open and take a look. My first cast was very nice. I think the channel for casting is a little too big. In the next mold I am making I will try something smaller.

    Humbling myself after an initial success is good. From the time of mixing the plastic to casting, I wasted too many precious seconds. While casting, the material stopped going into the mold so I thought it was finished. Turns out, it was solidifying inside the mold before filling.

    Here the scar from cutting away the extra material can be clearly seen. Unfortunate.

    Otherwise, looks nice. I did not notice the milling toolpaths along the inside surfaces of the piece. When the puzzle is locked together, these surfaces are invisible. Regardless, I will hit those with some sand paper in the wax mold before casting any more oomoo.

    And after making a bunch...

    Download project files

    I will post links to resources I have found helpful here.

  • Smooth-Cast 300 Datasheet : Inform yourself.
  • Bioelectrochemical testing pod

    I would like to begin prototyping my final project. The first challenge is setting up a plant microbial fuel cell system for generating electricity with moss.

    This work will heavily rely on the scientific research from Paolo Bombelli, et al. I thought I could cast this pot...

    ...and thinking how to do that with ooloo is tough. I think slip casting or vacuum forming would be the easiest route. I have experience vacuum forming and more, and would have some concern with the finishing that would need to be done after pulling a form for this object. If I slip cast a plaster mold with clay, the plaster has a long drying time and then each cast requires its own extended time plus glazing the inside to make it water resistant is extra effort.

    I came across a project by Nicolas Kisic Aguirre. He used four forms to make an sleeve. Maybe I can do something similar with two forms. First, grasshopper.

    Next, a positive and a negative form which slot togther so they may float in space. Looking at this now, I think I can improve the slotting on the bottom, which is shown as the top in the second image. The blue section could wrap up the wall of the red a little to make a tighter seal.

    Jump : Index

    J.travis Russett © 2017
    Creative Commons License All the work contained within is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License
    You may remix, tweak, and build upon my work non-commercially, as long as you credit me and license your new creations under the identical terms.