So, after a month of attempts, I tried a different approabsp; Instead of moving the heat source, I iried l the vat of liquid fluorite. That seemed to initially have the same issues as before, but I persisted and tried tinkering with the process. I found that if I lowered it too quickly, it'd form a gss, and as I slowed dowe that the vat was lowered, I started getting ion of crystals. The problem was that I was getting multiple crystals all over the vat, and they remained fairly small, eventually getting encased in gss themselves.
I ged out the vat for a vat made of polished steel, and ged up my entire apparatus. Instead of manually l the vat, which seemed to always result in it getting lowered too quickly, I ged it to be gear driven, with a gravity driven escapement meism to very slowly lower the vat away from the heat. The first few tests were slightly more successful than the first, but I was still having the issue of too many ion points.
After another month of testing in total, I ged the vat design again, this time having the bottom of the vat shaped the same as a singur fluorite crystal. My hope being that the lowest point, which would be furthest from the heat source, would also provide the starting for ion, and that crystal would be where all the new fluorite deposits.
This method had mixed, but improvis. Intermediate results would still have too muucleation , with new crystals growing off the sides of the vat, or even on the face of another crystal. I decided then that I still had a few too many variables to trol to properly form this process. For one, I was still attempting to heat the vat using charcoal, so the temperature was likely fluctuating a lot he top of the vat. A sed issue was likely that the whole process was still exposed to air.
So, I took a month and a half attempting to alleviate both of those issues. While I'd love to use a fixed rate hydrogen fire to gee the heat for this process, we don't really gee enough heat for the vat I've been using for testing. We , however, improve the sistency of the heat in a different way. If we used standardized charcoal pellets over raw charcoal, the general temperature fluctuations should be smaller with the more sistent fuel sourbsp; Of course, produg charcoal pellets was easier said than done.
First the charcoal o be pulverized into a powder, which was fairly easy with iven teology. However, the process of pressing and cuttis was a little tricky to rediscover, and I got fairly lucky. I desighe pellet mae to press powder through small die, like how we make wire. However, since I want regur pellets, I wahe cutting process and the pression process automated and tied together meically, so they'll always work at the same pabsp;
Ultimately, I lucked out in that I just so happeo time the passage of the bde whiocks the pellets off with when the pressive roller would finish f powder through the die. If I hadn't timed it like that, the mae would have output a lot of failed pellets in addition to a few good ones, and I'd probably have wasted a day or two figuring out what was wrong. After hooking the pellet press up to a small windmill, I then tasked some goblins with making charcoal pellets from now on. Besides being potentially useful to me, the stable burn of pellets will probably also be useful in some preetalw which the bcksmiths might want to do.
I also did some design work on how the heating element is fueled, to keep the heat source banced. Thanks to having already worked on the pelletizer, I already had a good idea of how to make sure our fuel source was evenly distributed, and I designed a rotati sourd scoop to stantly rotate above the fire, p new pellets on slowly, while also scooping excess pellets off that might have piled too high.
The vat itself was fitted with a stone lid, with a stone shaping vacuum valve, so I could attempt to remove any potential impurities that might occur from that end. After finishing this new design, I've realized that I only have a little uwo months left before I o start training goblins enics to get that facility running more permaly. Which basically means if this new desig work then I'll have to put artificial fluorite on hold again for a few seasons.
Results seemed promising enough in the first real trial that I tinued my attempts. I say first real trial because there were multiple meical hiccups during the first few days of tests. Seven days in, I had my first real taste of success, albeit a bit ckluster. The crystal seed at the bottom did grow to be the rgest crystal i, but there were still numerous other crystals f throughout the vat's walls. That trial growth ran over 12 hours in total. The rgest crystal still had off-growths on it as well, so I slowed the growth process down further by iira gears into the escapement meism.
After multiple trials of increasing duration spanning 25 days in total and some upgrades to the escapement meism, I finally succeeded as much as I think I'll be able to with our current teology. There were still some sedary crystals growing o walls, but the main fluorite crystal in the ter of the vat was the bulk of the mass. The main crystal ractically fwless, and because it was grown on the surface, it was still extremely hot to the touch.
The first artificial fluorite crystal that showed success ractically colorless, but had a very faint e tinge, a remnant of what little of the natural impurity was left after melt cyg it so many times. The crystal also produced the fai amount of heat still. The rgest natural fluorite that I had pulled from the limited amount I mined was about 4 inches along it's diagonal. By parison, this crystal was about twice that height at just over 8 inches.
While an 8-inch octahedron is nice, I'd like to grer ones iure, now that I know that it's possible to grow them artificially. It's also important that I hahe crystals as little as possible as we melt them down to reform them, as to attempt to keep as much of the natural inclusion material as possible. For research purposes, we actually want to do the opposite, and heavily process and filter until we have naturally pure fluorite, which we then attempt to add inclusions into.
Irospect, I should have probably been wary of the transparent fluorite, in case it too produced any unknown magical effects. Though by the time I realized that, I'd been handling this new crystal for a few days while experieng no symptoms. With just over a month left, I want to grow a colored fluorite crystal this size, a some ptes cut from that crystal to see how they perform.
I ended up having to mine more fluorite to make a good batolten fluorite material. That being the case, I decided to mine enough for a few batches of crystal growth, just in case I had any other sudden ideas while I worked. While I did successfully grow an 8-inch tinted fluorite crystal, it's color was still slightly faihaural crystals I had mio make it. Since I grew it on the surface, with ambient mana about, it was actually quite hot to the touch even after it should have teically been cooled down.
Thankfully, fluorite is pretty soft, and I was able to easily cut an inch thick pte from the ter, leavih a squarish pte measuring just over 5.5 inches per side. Then, I was left with two square pyramids each 3.5 iall. While I could titing the pyramids down, they could each be useful on their own as well, and the rger the crystal, the more effitly it turns mana into its final product, so breaking it down further might be a waste.
After testing the pte, it still seems like crystals, or at least fluorite, rely pretty mutirely on volume fical effectiveness. The pte, while only 1 inch thick, and 5.5 io a side, has nearly three times the volume of a 4-inch fluorite crystal, and seems to produ appropriate amount of heat that I'd expect for that volume. The difference, however, is that it has a much shorter distao it's ter from the surface, meaning it dissipate heat geed across it's volume faster, making it much more useful for industrial purposes.
Although I don't have proof of it yet, I also suspect it probably intake mana more effectively, given it has a higher surface area to volume ratio. While a small crystal would have an even rger surface to volume ratio thae would, small crystals suffer from being uo produce much of a magical effect, losing any advahey might have had.
After only nine days of work though, I recalled that I should probably go do some levelling again before the new year begins so I went and did thirteen days worth. , I'll o train the new goblins on how to operate the cryogenics facility. After those goblins are trained, and I'm fident in their abilities, I pn to design a much rger fluorite crystal growing vat. Sihe 8-inch crystal was already quite hot on the surface, I'll actually need a detly sized underground facility to do this work in.