I read this thread earlier and was trying to work out how I would approach it.
I’m not sure any of the techniques or materials I have used myself would suit this product (HDPE is never pourable, so in order to get fine detail, it is necessary to either injection-mould (difficult for amateur/small scale), or press-forge it using a two-piece rigid mould - and I can’t think of a path to getting a rigid and truly heatproof forging mould in less than four processes, which is a pain, and potentially loses surface detail fidelity.
If you changed the design to resemble a disc-shaped medallion, it would be fairly simple to carve the thing out of many different resistant materials on an entry-level CNC router/mill (the existing design could also be CNC carved, but would require more than a simple 3 axis machine).
But Im assuming you don’t want to change the design at all, so starting from basics, and given the durability and safety stipulations in the brief, I would:
[ul]
[li]3D print the two pieces of the negative casting mould (is it possible to 3D print with sufficient detail to make two mating mould halves?)[/li][li]Cast the hall pass into the 3D printed mould using 2 part polyurethane rubber, or some other flexible but tough casting resin.[/li][/ul]
If you use a clear resin, you could mix in cool stuff such as foil flakes, or you could mix up small dyed batches of resin and pour them sequentially for stripes/banding, or simultaneously for swirls.
The printers themselves use something called g code, but that’s not relevant, because every 3D printer will come with software to generate that for you. In practice, the format you want for your models is STL. Anything at all that can do 3D design can export an STL file. I use TinkerCad, a free website. As for how hard it is, my introduction to it came in the form of “You’re going to be teaching a bunch of middle-schoolers how to do this, starting in about five minutes”. Of course there are different levels of proficiency, and I’d have had a hard time making this thing with that level of proficiency, and couldn’t have done most of my other projects at all, but that’s all just a matter of playing around and seeing how the different bits and pieces work. At the highest levels, if you want to create your own shape generators, you’ll need some programming know-how, as well as algebra and possibly some trig, but you can do a heck of a lot without going that far (as it is, I’ve only found the need to create one shape generator myself).
I intend to find out. My hunch is that, for the sorts of shapes suitable for rigid molds, it should be possible: 3D printed pieces seem to have the greatest loss of fidelity where there’s overhangs, and the finished parts of the piece droop a little as the new layers are added. But a rigid mold can’t have any of those, anyway.
Well, I’ve finished it. I have some pictures; let me know if you can view them.
I printed the mold out of PLA. The nonplanar shape of my seam was enough for proper registration between the mold halves, so I didn’t need to add any dimples. I then weighed out some low-temperature gluesticks, mixed them in an old can with some dross from the original pass, and heated it in an oven to 325º F (just a little hotter than the melting point of PLA). Meanwhile, I saturated the mold halves with water and froze them, and used cooking spray for a mold release.
The original plan was to pour a little in the halves to capture the detail, then to put the halves together for a fill pour, but I found that once I’d poured enough for the detail, I’d poured almost all of my filling anyway. So I proceeded to just press the halves together. I also put a bent bit of paper clip wire in to reinforce the eye-hole, which was the first part to fail on the original (I don’t have any concrete plans for that eye-hole, but it seems like it might someday be handy to put it on a lanyard).
I had also originally planned to add pigment, but Mom assured me that she had some tempura powder (which I’m guessing would have worked), and so I didn’t bother to buy any. It turns out that she does, but not in anything close to the color I wanted.
Results were mixed. It seems that hot glue, even at a hundred degrees above its melt temperature, is still too thick and/or quick-cooling to fill a mold well: I had a lot of trouble with voids, that I had to fill as well as I could with a second pass. If I had a different mold material, I might have been able to address this by putting the whole mold into the oven and letting it take its time to ooze into all of the cracks and crevices, but I think that would have been pushing my materials too far. I’m also not too wild about the appearance: The translucency makes it hard to make out even the thick-embossed lettering, much less the thinner fractal figure. And it doesn’t slide in and out of the holster nearly as easily as the first one.
The good news, though, is that I still have the mold, so I can still experiment with other materials and techniques.