By no means have I forgotten about this project, so I wanted to provide a little update. As is my wont, before moving onto producing actual final pieces, I've been devoting an OCD amount of time thinking about and experimenting with perfecting my "process." (Not that I don't do this when it's entirely out of proportion in many cases, but in this situation I have the plausible excuse that I want to make sure I've got everything absolutely right before I start encapsulating hard-to-find vintage electronics components permanently in resin.)
I had for some time been vexed by the fact that I wasn't able to get the back of the first test casting from frosted all the way back to optically clear, which I mentioned in my last post. Due to the meniscus formed when doing one-part molds, I know I'm going to have to sand flat all the castings at last at the base, so I wanted to know I could get it fully optically clear.
My second test casting (again using random found electronics components for encasulation) came out better in some respects than the first and worse in others. Because I fully vacuum degassed this time (until no bubbles were coming out of the resin at all), there were absolutely zero bubbles in the final cured part.
However, I did get some weird waviness in the surface (which I somehow neglected to photograph) that didn't happen last time. This required me to sand flat all the surfaces and not just the base, deepening my concern about my ability to get a sanded surface back to perfectly clear. Smooth On recommends heating up your mold before casting to avoid "suck back" with Crystal Clear, which may be the problem I was encountering. I'm going to order a toaster oven and try this for my next test. Incidentally, I'll also be able to use the oven post-cure the resin, which should speed up the long wait during which Crystal Clear remains tacky at the surface, preventing demolding for several days after pulling out of the pressure chamber.
Going in successive grits from 80 up to 3000 in sand paper and then moving to Novus polishes, I got a decent but not excellent result.
You can see on the upper slanted face in particular some fine scratches, and in the second photo the whole thing just looks generally pretty cloudy.
After much research, I decided to try the foam polishing pads and machine polishing compounds in the 3M automotive clear-coat polishing system.
Due to their use in the car body shop industry, they're pretty expensive. I believe the full set of polishes and pads cost me about $150. But using it with my pneumatic DA sander/polisher did yield a visibly better result. (This actually is an apples-to-apples comparison, because I went back through the sandpaper grits for this second test.)
So this was like 90% there. But under very close inspection with bright light, I could still see some machine swirl marks, which was driving me crazy. So I took a deep dive researching into the world of automotive finishing, in particular in the area of clear coats on black paint jobs, to which this project is most analogous.
After much reading and experimentation I finally worked out that I wasn't getting a good result buffing because my pneumatic polisher was losing power when buffing a high speeds, so my machine polishes weren't being used to their full effect. I don't have the world's most powerful compressor, nor does it have a very large reserve tank, so this isn't surprising, but I don't know why it took me so long to hit upon this as the culprit.
I decided to order both a pro auto refinish DA polisher and a rotary polisher, both in electric versions, to see which would suit my purpose better. I chose the Rupes Big Foot Mini and the Flex PE8. These are both "detail finishers" in that they accommodate relatively small pads. I use mostly 3" 3M Trizact pads, which again are crazy expensive but very good for wet sanding.
Conventional wisdom has it that DA polishers, because they are newer and offer an "extra feature" (an additional axis of movement), are somehow always superior. I didn't actually find that to be the case. I found the DA polisher really difficult to control on a small item like a prop, and that even at full speed it took a lot longer to get much work done with it. Unlike my pneumatic one, however, it did actually improve the finish and didn't leave little mini-swirls. The Flex rotary polisher, however, is really easy to control and runs incredibly smoothly. It also can cut and buff through relatively deep scratches really quickly, so I found that I strongly prefer it. Some people are worried about overheating the clear coat surface when using rotary polishers, which is a valid concern, but I found that lubricating the foam with some water in addition the compound entirely mitigated that problem for me. To avoid adding swirl marks and holograms, rotary polishers take a bit more practice to get the feel of than DA polishers, but I didn't find the learning curve particularly steep.
Another nice feature of the Flex is it can do all the way down to 1" pads for really tight corners, which on props is really important. Most polishers are for large auto body work and thus 6" or larger.
I also found that for the sanding, I fared better with hand polishing using 3M wetordry papers rather than the Trizact discs, waiting to switch to machine work until I was up into the polishing/buffing stage. An extremely useful tool in this regard was this 3M pad:
It makes it much easier to sand a smooth surface than just one's fingers.
Here is the final result I was able to get after some practice with the rotary polisher. At last, I'm quite happy, and it's basically optically clear.