Sander van Vliet

@Fablab Amsterdam

Molding and Casting

Week 12 (20 april - 27 april)


This lecture was about the various types of molding and the materials and additives which are being used for casting. Also the process of mixing, pouring, curing and demolding the material and the safety precautions one should take when working with these materials. See this weeks' class lecture file to get an overview of the subjects of the lecture.


A white plastic wall mounted panel

I wanted to spend the time for this weeks assignment on something that I could use for my final project. I was not 100% sure it would be possible, but I would try to make a wall mountable control panel which will include a CO2 sensor and an LCD, visible on the front of the panel. It will also contain a main board, a wifi module and possibly a temperature sensor.

Designing it
I measured the dimensions of the CO2 sensor and the LCD and calculated the dimensions the panel needed to have. Then I started drawing in Rhino on friday night and finished the rough design of the panel.
After a weekend where I didn't have time to work on it, I stayed up sunday night trying to create two molds from the model. I was spending a lot of time trying to understand the shapes and the counter shapes a.k.a. the positives and the negatives of: 1- the mold of the mold 2- the mold of the final product and 3- the final product itself. We are making these three objects for this weeks assignment, but in fact the two last ones are formed by the design of the first one. This is a mold of a mold which we fabricate by milling blocks of machinable hard wax with the shopbot. I wanted to design and use only one mold part and use only one block of wax, but this was not possible for my design because the panel I am making is a hollow structure. This would need two nesting mold parts, so I have to make a mold consisting of two parts. Drawing the molds was a true mindf#@! and I kept finding myself trying to answer the same questions over and over and ending up almost desperate, trying to find a different approach to get my brains around it. Using play doh clay creating the parts and counterparts helped me visualize and finally I got a more clear vision of how it needed to be modelled. I still didn't finish drawing that night though and went to bed -mentally exhausted- at 2:30.
Monday morning, after 4,5hrs of sleep I woke up with the clear vision still on my mind and continued drawing.

Design issues
A few things I realized were not possible/correct during drawing:

  • I created a chamfered edge on the inside/rearside of the panel, but this could not be created by milling because this is an undercut situation. It would need the wax mold to be milled from the other side as well. I had drawn the edge chamfered to make the CO2 sensor go in more easily, but it wasn't a problem to remove this detail from the design.

  • I had added a chamfer to what would become the bottom of the key hole (1) instead of what would become the top of the keys (2). As many times before during this part of the process, I didn't take into account that this mold is a negative of the product it will form. I corrected it and added the chamfered edge on the part that would form the keys.

  • And Monday around noon I finally had finished the two models! panel wax mold of rubber mold two parts source file - rhino

    Checking the design
    I wanted to mill the wax blocks and also wanted to pour the rubber on this day still, so I rushed to de Waag and discussed my design with our local instructor Mickael. He noticed the following issues:

  • I could have added keys to the two parts that form the sensor and LCD openings (1) to make the mold more stable although we both assumed it would not be a problem without them.
  • The inner corners of the rectangular LCD opening (2) will be round with the diameter as big as the milling bit's diameter intead of a sharp 90 degrees corner like in the drawing. This was not generating an issue for my design.
  • I would need to use a ball end milling bit for the Shopbot to be able to form the u shaped cable opening (3) in the bottom. Because I needed a flat end milling bit for other features I did not want to loose in the design I accepted the fact that this opening would not be formed exactly the way I had drawn it.
  • I needed to check what the deepest point of the mold was to be able to decide which milling bit I could use for this job. This was 28mm (4) and adding a few mm's to stay clear from the collet I now knew that the milling bit I would use needed to stick out of the collet at least 30mm. I could have used the 5mm, but I chose to use the 4mm flat end milling bit just to be 100% sure it will go through all the narrow parts of the design.

  • I did not create a pouring- and a vent opening, because I thought that I did not need it. I wanted to keep the rearside of the 'inside' mold part (on the right on the image above) open by making the walls which I would make around the wax mold the same height as the wax mold itself. I only realized at this point that this wasn't possible because then the inner and outer parts of it would not have been connected that way. They would have been able to move relative to each other which would have made the mold very unstable. Mickael had a few potential ideas how to work around this.
  • One way is obvious; change the design file and add a pouring and a vent opening to the models. For time reasons I didn't want to do this.
  • Another way is filling the mold with rubber without having a pouring opening by filling one part of the mold and pressing the other part on/it it, pushing the excess rubber out. It is possible but the risk of having trapped air bubbles inside is higher that way. There also is a chance that the top mold part will not be fully filled with rubber.
  • The third option is placing two pieces of hard non absorbent material like acryllic in the wax mold before pouring the rubber. The rubber will not be where the two acryllic pieces are and so this will create two openings in the rubber mold.

  • I chose this third option so I did not have to change the design file. Now I was ready to setup the shopbot to mill the two wax mold parts.

    Milling the wax mold parts with the Shopbot
    After fixing the block of wax to the Shopbot bed with doublesided tape I created a roughing toolpath and a finishing tool path for each of the mold parts milling jobs and ran them, see below.

    These are the settings I used for the roughing jobs and the finishing jobs:

    Creating a rubber mold from the wax mold
    After milling the wax mold parts I did not have time anymore to pour the rubber at the Fablab because it was closing time. I decided to do this at home so I took the materials I needed home and fixed the walls to the mold. I used glue and tape to close the seams so the rubber would not leak out. I read the instructions and setup everything in the bathroom for this job. It has a ventilation system and we would not need to be in there so the rubber could cure there all night.

    Then I added an even amount of parts A and B of the urethane rubber product Vytaflex 30 to come to 650ml total. I had estimated this amount by measuring the volume of the mould roughly with a ruler.
    After stirring for three minutes I could pour it in the mold, see below.

    Creating the plastic panel from the rubber mold
    After I removed the walls and had taken apart the wax mold parts the rubber mold I created looked nice! The rubber feels very strong and the edges were very detailed and consistently formed. I did notice an issue though... I had forgotten to place two pieces of material before I poured the rubber so the mold did not have a pouring- and a air vent opening! Too bad, because after I did not draw the openings, Mickael had given me this option to solve this and now I had totally forgot to do it.
    Only one option now which was making two holes in the rubber mold. I tried drilling with a sharp drill, but after a few attempts in my leftover rubber piece I decided I will cut the rubber with a knife. After having done this I sprayed mold release agent onto the two mold parts and let it sit for a while. Then I put the parts together and fixed them using some materials which I found in the Fablab.

    I had opened a model of my final product (the panel) in Netfabb to determine how much material I needed (70,63cm3) and placed the mold setup under fume hood. I placed one side on a piece of wood so the mold would be standing under the angle I had determined before. This was necessary for the bubbles to move up through the liquid plastic to the air vent opening (left opening). I assumed I needed some extra material for the bit that would stay behind in the cup and which I would maybe spill so I mixed 50ml of each parts A and B of Smoothcast 305 white plastic and poured it carefully in the pouring hole until the mold seemed to be filled up.

    Then I slammed on the table to vibrate the mold to help the air bubbles inside the plastic to come up and leave the mold through the air vent opening. I could see bubbles come up and the plastic level go down so it worked. I poured a little bit of plastic to fill the mold again and repeated this process for a few minutes.

    On the train home I opened the rubber mold and there it was, the hard white plastic panel!
    It had some flashing at the parting line, and ofcourse you see material where the pouring- and air vent openings were, but it looked fine although I noticed something had gone wrong! Where there should have been openings for the LCD and the CO2 sensor was a thin bit of plastic which I found very strange. Apparently the mold parts didn't touch eachother well enough and the liquid plastic had flown in between them. I will manually remove this to have the openings and I will finalize the panel by removing the flashing.

    You can read about the finishing process and see the end result of the panel here