My first attempt at the carrier was a fun educational adventure, but with disappointing final results.
I initially had the idea that it would be cool to try to make the red carrier body as if it were a large injection-molded plastic part. The traditional means of accomplishing this in the prototyping world is vacuum casting. As much as I would love to, I obviously can't afford a $40,000 Renishaw Vacuum Casting system used by professional prototyping shops, so I've been lately trying to figure out a poor man's approach to replicating the same system and results.
The premise of industrial vacuum is pretty straightforward. You mix a polyurethane resin under vacuum and then use a robotic apparatus to pour the resin into the mold, also under vacuum. Because there is no air in the mold at all, air entrapment is not a concern, and mixing of the resin under vacuum also eliminates most pin-hole bubbles by de-gassing the resin before the pour.
Just de-gassing the resin before the pour and then pouring the resin into the mold at atmospheric pressure doesn't work for most molds because even with the best venting, air entrapment is still frequently an issue on many geometries. This is fine and well if you want to cast something and paint it, because you can always just do a little bit of body filler before painting if there are any voids. However, if you want to do something like I did with the sample carrier body and have the pigment actually in the polymer rather than as a thin layer of paint on top (thus more accurately simulating a mass-produced thermoplastic part), you need a truly void-free part. Extra positive material problem (such a vents and sprues) is no problem, because that can always be sanded and polished away.
So in order to actually degas the resin in mold to pull air out of entrapment, I've been building my molds with big reservoirs, pre-degassing the resin, pouring into the mold, and then degassing again. Here is what that looks like on the carrier.
I started with a CNC milled master model.
I then built up a mold. The copper wire is venting and the big off-white ABS block functions as a resin reservoir. The tape creates a parting line.
In order to vacuum such a large model and mold, I had to get a bigger vacuum chamber than I previously had.
And a much higher CFM pump
I was concerned about being able to pull bubbles out of the deep holes, so I pre-poured silicone into the negative spaces first.
I then fit the model sideways into a mold box and poured the rest of the silicone.
In order to degass that much silicone before pouring, I had to build an enormous box for it. Silicone expands many times its original volume on a first degassing.
After pouring the silicone into the mold box, I then did a final silicone degassing of the silicone in the mold box itself (this photo is a reflection viewed from below of the ceiling of my vacuum chamber).
I let it cure overnight.
Trimmed the mold and cut to the parting line in a zigzag pattern with a scalpel
Here is what the mold looks like with the reservoir, ready for casting. It's enormous and quite heavy.
And here after casting with sprues trimmed before demolding.
So here is the thing. The above described process works perfectly on some molds, but on others I tend to get foaming of the resin as the vacuum starts to approach -1 bar. That was sadly the case on this mold. This phenomenon actually introduces bubbles and leads to terrible results.
I was able to mitigate the foaming issues somewhat by stopping the vacuum as soon as I saw evidence of resin foaming in the mold, drilling in additional venting into the mold, and curing at around 400 kPa pressure, but results were still not perfect.
I still had some issues with air entrapment voids, presumably because I wasn't able to pull and hold a full vacuum on the mold with the resin in it due to the foaming issues.
In all cases, the top of the carrier (which faces downward during molding) came out fine.
But the voids on the bottom made the parts unusable, since if I want a consistent color throughout, I can't putty the carriers.
I think I need to try a new strategy. I need to figure out how to pull a full vacuum, degassing the resin and fully evacuating the mold of air, and only then introduce the resin into the mold. As it is, trying to evacuate the mold with resin in it, I suspect that the foaming is being caused by small amounts of air being left in the mold trying to escape around the resin as the vacuum approaches -1 bar. I figure if I can pull a full vacuum first and then pour the resin into the mold (like the true vacuum casting chambers do), this problem can be avoided. So my plan is to try to rig up a small robotic apparatus to hold the resin in a reservoir, let me degas the chamber fully, and then remotely release the resin into the mold. This will take some design and experimentation, but it should be fun. Until then, I don't expect to be able to cast a bubble- and void-free carrier. This this was a bit of a failed experiment, but I learned a ton and have some obvious new avenues to explore. Unfortunately, I'm currently in the process of packing up my workshop to move it across the country, so it'll be a few months before I can revisit these issues, but I'm looking forward to it.
I learned a few other things in making the mold. I did incorporate slight draft angles in the negative cylinders at the top to make demolding easier, which make that half of the mold very easy to pull apart from the model and castings. However, the bottom proved very difficult to de-mold, even with shots of pressurized air down the vents. Next time, I would design the bottom like this:
This would have made is so much easier to release from the mold, and also to paint the original master model without paint dripping down into inaccessible corners.