This is actually my 1st post, but I've been building models for many years- usually in the course of product design/engineering, but I don't have a forum quite like this in my repertoire so I this will be a nice addition!
Anyway- so what prompted this research is a particularly tricky mold that I have a few questions on. It's tricky for two major reasons:
1. It is a shape sometimes described as a "hyperhemisphere" which is a hemisphere that continues past the centerline creating a big ole' undercut. Think of a fishbowl, or lightbulb. Usually something like this would be made in blown glass. It's basically a hollow sphere with a hole in it (in my case around 30-50% of the diameter).
2. Both the inside and outside of the sphere need to be optically clear surfaces- this is why blown glass doesn't work, too much surface variance and too much distortion. I'm looking to improve on that.
My solution so far has been to try to mold two rigid patterns in soft materials. The patterns are accurate optical surfaces and I have been trying to avoid using any release on them for fear of impacting the clarity of the surface. So far so good in that regard. You just have wrestle and have some patience to remove the patterns and stretch the molds. Each sphere pattern also contains a little flange which exists primarily to align them when they are combined. This way you can locate the smaller (positive) sphere mold concentrically with the larger sphere (female) sphere mold and when you pour the transparent rigid urethane into the void- voila you have a perfect part! (in theory).
The larger sphere is no problem- Just pour platinum silicone around my "lolipop" pattern and pop it out. The interior sphere is harder, because it has to end up being a very soft and compliant material in order to pull out of the undercut, and the source pattern is rigid. To accomplish this, I tried using a soft (40A) urethane for a mold- removed the pattern and then poured dragonskin (smooth-on) into it. Unfortunately it turns out these aren't compatible evidently, and the part inhibited.
Ok- so I will instead make a new mold with conventional platinum around my pattern, and this time pour dragonskin into it.
But now I'm going to need to pour the rigid clear urethane into a mold who's outer surface is platinum rubber and inner surface is dragonskin.
Is that going to work?
More specifically- I think the dragonskin will be ok with the undercut,
(BTW, it's kind of neither here nor there, but I installed a 3rd, less precise wax lolipop core as well so that I could melt/break it out and have a more collapsible interior mold, or apply vacuum to self-deflate)
but here are the things I'm specifically wondering about:
1. Will dragonskin be effective in preserving an optical surface?
2. Is it mandatory to use release in these processes and is my worry justified about using a release (re: optics)?
3. Will I be able to peel the dragonskin away from the rigid clear urethane? I'm not trying to make an overmold surface, but this is probably how I would normally go about it if I were.
4. What, if any worries about inhibition or other issues should I have with this new scenario?
5. What other materials might be a better choice? eg. I could maybe make that interior core in solid wax also, but I rarely have had motivation to make a wax part. I wonder if that process would be as effective or what tricks would be involved- I'd like to leave that interior surface as unscathed as possible and I worry about smears with wax. What are my other material options?
If you've read this far, sorry for the long first post, but I hope I've made this clear (pun intended!) and I hope someone can offer some advice. Thanks so much for reading!
Here is a diagram: http://i.imgur.com/7er6QK8.jpg