Goal
This week is going to be a little bit busy. Because this week we have to design a mold, which is not easy at all. Then we must machine it, and it is going to take several hours. Then make a mold which takes some time to dry and finally cast parts form it. I'm pretty sure I'm not going to make it this week.
Last week MIT lecture review
Last week I was late and I missed most of the assignments showcase. The script did not chose me either. 6 weeks in a row now. The probability of being picked by the script next week has raised to P=1-(0.75^6) = 82.2%. We are approaching what the statisticians name the certain event (P=1).
This week we are doing quite an interesting topic: Molding and casting. I like it because I've never done a mold. And I only have cast tons of concrete. Neil told us two reason for using this technique instead of 3D printing. One, the surface finish is much better and smoother. And two, you can cast thousands of pieces with a single mold much faster than 3D printing. It is an interesting tool to make repetitive and modular things: from ice cubes, chocolate tablets, stamps, and construction pieces. We'll be using machinable wax (much denser than normal wax) to make the positive of the mold, silicone or urethane rubber for the mold itself and we have quite a lot of materials to cast, each more toxic than the other, as I could see. Which lead us to security, again. Our instructor Tomas showed us a picture of him with the allergic reaction he had because he used Crystal Clear without gloves and mask. His body has developed allergy to the product for the rest of his life.
Designing the mold. First attempt
This time I followed Neil's advice about spiral development. He told us to make something small going through all steps before doing the big job. This led me to the first failure of the week. Shall we start from the end. These two images below should look the same
What failed was not a single factor but a combination of three:
- I wanted to finish it quickly so I designed a small piece.
- I wanted to finish it quickly so I used a 1/8 bit
- I wanted to finish it quickly so I made really small details
You should not go that fast. None of these factors is a problem in itself. You can do 1 and 3, or 1 and 2. But definitely, you cannot mill something that small with a 1/8 bit. Also I later noticed that if you want to make a mold of something 2.5D it is better to use the laser cutter and acrylic. It will be faster and you will obtain better results.
Designing the mold. Second attempt
For the second attempt I wanted to design a 3D object. I liked the idea of making something modular that could be used for construction. At the beginning I was thinking of a spatial structure node like you see in the image below, since these items are very expensive.
But I left this assignment unfinished and at the end of the semester, while doing the machine assignment we felt the need to make a timing belt so I designed one in kokopelli. The best thing about kokopelli is that it allows you to design the same way your brain works, adding and removing volumes. So I created a .cad file for the timing belt that allows me to display five outputs: the belt, the plastic generator, the silicone mold, the wax and the gear that matches the belt. Is not possible to visualize this with any other piece of software.
Milling the mold
I milled the wax with the Roland Modela, finding it unexpectedly fast (15 to 20 minutes for the rough pass using a 1/8 flat end bit and one hour for each finish pass XZ and YZ using a 1/16 ball nose bit). The quality of the finish is unbelievable, even the most little detail is reproduced.
Casting the Silicone mold
The next step was casting the silicone. I used OOMOO 25 silicone rubber that someone left in the shelf, but when I opened the pots I knew that something was wrong. The content of the yellow pot was too dense, and the content of the blue pot was very liquid at the top with a huge solid ball (maybe already hardened silicone) at the bottom. Anyway I mixed them in a 1:1 proportion (eyeballing) as stated in the instructions. I tried my best to not including bubbles into the mix. I poured the mix in the mold, it was so dense that it was difficult. I don't remember my mates pouring such a dense mix.
A couple hours later (75 minutes should be enough, but it was still sticky) when I removed the silicone mold my concerns were confirmed. Lots of huge bubbles were trapped while mixing and/or while pouring and they could not find their way out. Also the product could not get into the female tabs of the wax mold.
I finally obtained a new unopened OOMOO 25 product and I casted again the silicone mold. There were still some bubbles but the finish was much, much better. At the top, the old silicone mold, at the bottom the new mold with fresh product.
Casting the plastic sections
I used Smooth-Cast 310 (matt white plastic) to cast three sections of plastic. It is possible to cast more or less sections according to the desired length of the timing belt. This time mixing products A and B is much easier because they are both liquid and you can mix them easily. At the beginning the mix is transparent and it becomes white when it's cured.
Casting the timing belt (at last)
I used urethane rubber as the material for the timing belt. Urethane rubber is elastic and timing belts are not supposed to be elastic so I reinforced the belt against tension with sewing thread. You can see the result below.
The result
Producing a reliable timing belt was a failure. It is not only way too elastic, but it is also supposed to be a high precision part. But the finish is really like home made and uneven. If you make a big mold you will probably success in your goal, but I think this technique has some limitations for small parts due to the viscosity of the products used for making the molds. Luciano already told me that I would fail, but I was hoping that the reinforcement could make it work.
What I learned
The hardest part of this assignment is mixing products A and B. Sounds easy, but it is not. There will appear undesired bubbles, and you will find it very difficult, if not impossible, to get rid of them without a vacuum device. I also recommend you to analytically calculate the volume of mix, and using graduated devices to carefully prepare 10 to 15% more than the exact quantity. I say so because I only had detachable coffee cups at hand for mixing and pouring and I wasted more product than needed. Molding and casting is an interesting technique for making repetitive objects, but take into account that for some small parts it can be difficult to achieve the desired result.
Download files
You can download all the files related to this week here.