This video has been sponsored by Opera. It was almost two years ago when I saw this video by Steve Mould, where he showed something that I'd never seen before, called the Atomic Trampoline. He also showed how it worked by just getting a regular ball bearing, and by dropping it from about a foot and a half, and honestly, it was pretty lame. I'm just kidding though, and that was exactly what he expected to happen, and this was because this cylinder was just made of some generic stainless steel. This on the other hand was the actual atomic trampoline, and for the most part, it was basically the same thing.
However, what made it a lot more interesting, was that it had this extra layer of special metal on top, and when he dropped the ball onto it, the difference was almost unbelievable. It just kept on bouncing, almost like it was never going to stop, And it just didn't make sense to me how this kind of generic looking piece of metal was able to make such a huge difference. Steve quickly cleared up a lot of this confusion though by explaining that it was made using a relatively new and unique type of material called amorphous metal or metallic glass, which was very different than normal metal. He then wanted to talk a lot more about it and why it was so bouncy, but the whole time the only thing that I could think of was that I really wanted to buy one of these Atomic Trampolines.
The only problem was that according to Steve, and hours of my own research, a setup like this was now impossible to buy, and Steve was only able to get this one by borrowing it from another channel called Grand Illusions. When I found this out, I was honestly pretty disappointed, and I still don't really know why nobody makes them anymore. What I do know though, is that not being able to buy one kind of made me want one even more, and at this point, it was pretty clear that there was only going to be one way for me to get an atomic trampoline, and I was going to have to try and make one myself. The only tricky part was that I still didn't entirely know what metallic glass was, and I definitely had absolutely no idea how to make it, but for some reason, I felt that I could do it, and I decided to just go for it.
So, I just jumped right into working on it, and I felt that the best and most important first step was to figure out how a disk of metallic glass is typically made. However, even after hours of searching on Google, and reading anything that seemed remotely relevant, I wasn't able to find much of anything. This is because, as far as I could tell, there was basically no information about specifically making a disk, and I was only able to find generic industrial methods that showed how to make things like ribbon. On the bright side though, all of this research still wasn't a complete waste of time, because I was able to find a lot of interesting information about metallic glass in general. For example, there are apparently an insane number of metallic glasses, where each of them have completely different compositions, and they all vary in their physical properties, like density, elasticity, and hardness.
This also meant that they all likely varied in their level of bounciness, and because of this, I almost definitely couldn't just make a disc of it, using the easiest recipe that I could find. In my mind, if I wanted it to bounce just as well as it did in the video, I was gonna have to make the exact same metallic glass. So I went back to Steve's video, and this time I listened to it much more carefully, And it was around the seven minute mark that he thankfully said exactly what it was made of, and that it contained zirconium, beryllium, titanium, copper, and nickel.
With a bit more research, I was then able to find that this composition in particular was called Vitreloy 1, and it was actually one of the first commercially viable, and mass-produced metallic glasses. Its exact composition was also conveniently listed on Wikipedia, and it was apparently 41.2% zirconium, 22.5% beryllium, 13.8% titanium, 12.5% copper, and 10% nickel. At this point, I now knew exactly what I had to make, which I felt was some good progress, but what was honestly kind of scaring me was the fact that it contained beryllium.
This is because in general, beryllium is known to be toxic and horrible, and when I looked it up, It's said that breathing in its dust or fumes could cause permanent scars to develop in the lungs. This in turn could then apparently lead to something called chronic beryllium disease or berylliosis, which is an incurable and potentially fatal lung disease. It really sounded like it was basically just the asbestos of metals, and what made it even worse was that apparently just one single exposure could potentially lead to berylliosis. Overall, this all honestly sounded pretty terrifying, and at first, it- kind of made me want to just abandon the entire project. However, after thinking about it for a while, and reading a lot of stuff about beryllium in general, I started thinking that it was actually possible for me to work with it safely, and I would just have to follow two very important rules.
Number one, make as little beryllium dust and vapor as possible, and number two, absolutely don't breathe in any of it. In general, rule two was actually relatively easy, and all I needed was some good ventilation and maybe a mask, but rule one was a bit trickier, and it would severely limit the ways that I could make the desk. This is because minimizing dust would mean that I wouldn't be able to shape the metallic glass like normal metal by grinding, cutting, or machining it, and instead, I was gonna have to somehow directly make pretty much the exact disc that I wanted.
In my mind, this left me with only one real way, which would be to melt all of the ingredients together and to pour them into some sort of mold, and at first, this seemed relatively simple. However, when I looked into things a bit more, I realized that there were a couple small details that were gonna make this process way more difficult. The first was that all of the ingredients were sensitive to oxygen, and this meant that I could only melt them with absolutely no air present.
Then the second was that to form metallic glass, the molten metal had to be cooled as fast as possible, and this meant that the mold that I used had to have some sort of cooling system. Altogether, this definitely wasn't ideal, but I still felt that it was doable, and at this point, I was fully committed to the project. So, I immediately dove right into planning it, And the first thing that I did was reach out to Steve Mould to get the exact dimensions of his disc, and he was nice enough to help me out. I then spent the next few weeks trying to come up with a setup that I thought would work, and my first idea was to do something similar to when I made Purple Gold.
This is because to do that, all I needed was a simple furnace with a crucible, and with a small tube shooting in some inert argon gas, I was able to easily protect- all of the molten metal from air. I was also able to quickly pour it all out into a graphite mold, and even though it was extremely reactive, it only seemed to get slightly oxidized on the surface. In my mind, this basically solved pretty much my entire problem with air, and I figured that for the cooling part, the only thing that I would have to do differently was pour it into a perfectly sized little disk mold, and to quickly cool it, Using something like a computer heatsink.
What was also nice and extremely important Was that it was very easy for me to do all of this in one of my fume hoods Which would completely protect me from the scary beryllium. Overall, I really felt that this solved pretty much all of my issues and I was actually very close to testing it out Until I randomly stumbled onto this paper that I had somehow never seen before which explicitly talked about making discs of metallic glass. They also included a photo of something that they made, and it was pretty much exactly what I wanted, and what was especially nice was that they explained exactly how they did it. Specifically, they said that all of the ingredients were arc-melted together and then suction cast using water-cooled copper molds under a purified argon atmosphere. At first, a lot of these words honestly didn't mean very much to me, but with a bit more searching, I found that at least on a lab scale, this seemed to be a somewhat common way to make metallic glass.
It also seemed like the main part of the setup was this thing called an arc melter, and with even more searching on Google, I was able to find a bunch of photos. I then took a closer look at what appeared to be the simplest one, except what kind of concerned me. was that I honestly had no idea what I was looking at. After examining it a bit more though, I realized that it was actually a lot simpler than I thought, and it was basically just a standard welding setup. with some extra parts. This was especially highlighted by the fact that it was powered by a common welding power supply, and that the positive output was connected to a standard tungsten welding electrode.
There were of course some major differences though, and the first was that the other wire, which is normally attached to the thing that's getting welded, was instead attached to a water-cooled copper crucible. Then on top of this, All of it was enclosed in a glass or metal shell and filled with an inert gas like argon, which would completely protect it from air. With this setup, it was also apparently possible to liquefy almost any metal, and I found this really cool demo of one on YouTube.
The moment that it was turned on, it immediately released some blindingly bright light, and this is because a plasma was being generated between the electrode and the copper crucible. This plasma was then so hot that it was able to almost instantly liquefy any metal that it touched, and it honestly kind of blew my mind how fast it was able to melt it. With that being said though, after seeing this I almost immediately wanted one, mostly because it looked really cool, but also because it really did seem like an arc melter would be both the easiest and safest way to work with the metallic glass.
What I also think made it even more enticing was that with a slightly larger one like this, it was apparently possible to incorporate a custom mold into the water-cooled copper crucible. This would then allow me to directly suction cast the molten metal into the exact disc that I wanted, and at the same time it would also cool it extremely quickly, which was very important, and at this point, I was pretty much fully committed to getting an arc melter. The only problem was that just by looking at it, I knew that it was going to be expensive, and after talking to someone on Alibaba for over a week, the final quote that I got was $14,000. This was way more than the close to zero dollars that I would have had to spend on my other setup, and the number honestly kind of shocked me. It was clearly the much better and safer option though, and on top of this, They confirmed that it was no problem to make a disc mold with the exact dimensions that Steve had given me.
So, I decided to just go for it, and after a few months of patiently waiting, I eventually received this large wooden box. I then spent the next 15 minutes taking everything apart, and I got rid of all the packaging, and I was very happy to see that I now had everything that I needed. More specifically, I had the power supply, a water chiller, and a vacuum pump, and of course, the space-age looking arc melter.
Now, all I had to do was put it all together, by first plugging in the wires from the electrode and the crucible to the power supply, then by connecting the chiller to all of the blue water lines, and lastly, by attaching the vacuum pump to this little pipe sticking out on the side. I also went and got my argon tank, and I attached it to this part at the top, and when I was done, I felt that it was looking pretty good. I then opened it up to check out the copper crucible, and I was really happy to see that it was exactly what I had asked for.
In the middle, there was the custom mold, which I will talk more about later, and around it, there were four small rounded wells, which I could use to melt metal. There were also two custom ones with an exact- diameter of 44 millimeters, and I had made these just in case the mold somehow totally failed. This way I would at least have something to potentially fall back on, and like the mold, I'll talk more about them later on. Either way, with it all assembled, it was finally time to test it out, and I was pretty excited.
I was also kind of tempted to just immediately try making some metallic glass, but at the same time, I knew that it was an absolutely horrible idea to do my first attempt with beryllium. I mean, it definitely made a lot more sense to first actually learn how to use it before working with something dangerous. So instead, I decided to start out with some iron, which I felt would be similar to the metallic glass because it would also have a high melting point, and when it was molten, it would be sensitive to air. I also just happened to have all of this high purity iron, which was perfect for this, and I loaded up one of the small wells. I then lowered down the main chamber and I clamped it shut, and I turned on the vacuum pump to pull out all of the air.
To monitor this, I just looked at the pressure gauge, and when I eventually saw that it stopped going down, I closed the vacuum, and I opened the valve to the argon. This caused the argon gas to rush in and for the pressure to increase again, and when it eventually got back to around zero, I turned off the gas, and I flipped back on the vacuum. I then repeated this exact same process three more times, and when I was done, I closed both the vacuum and the argon valve.
All of these cyclings should have then left me with a chamber that had effectively no oxygen or moisture in it, and only inert argon, and I was nearly ready to start blasting the iron. The only thing left to do was to flip on the power supply, and to quickly do a horrible job setting up a camera to look inside, and at this point, I just had to click the pedal on the floor to make it go. So, I quickly lined up the electrode above the iron, and I was honestly a bit nervous, but I just went ahead and clicked it, and the whole thing immediately lit up.
Within a fraction of a second, the argon between the electrode and the crucible had been turned into a plasma, And I don't know why I was so surprised that it was working. It's working! The only problem was that it seemed to be struggling to melt the iron, and this was because the power supply was still at its lowest setting. I started it like this because I wanted to start safe, but clearly it needed more power to generate a hotter plasma, and I quickly ramped it up. As I did this, it immediately got a lot brighter, and not long after, There was thankfully some molten iron.
This was actually really impressive, considering that iron melts at around 1500 C, or 2800 Fahrenheit, and to do this in seconds was kind of insane. What I thought was even more impressive though, was that in less than a minute later, it had completely liquefied, and I had an absolutely ripping hot blob of iron. It also didn't seem to have any oxides forming, which told me that the inert atmosphere was working, and overall, I was very happy with my purchase.
I mean, it really was just as cool as I had imagined, and even if I completely failed to make the metallic glass, I had absolutely no regrets. That's really cool. But with that being said, the iron wasn't gonna get any more liquidy, so I stopped blasting it, and I let it cool down.
Then, several minutes later, when I felt that it was ready, I popped open the chamber, and I saw my beautiful little blob of iron metal. I was also really happy to see that it didn't look like it got oxidized at all, but it was surrounded by a bunch of black junk, which I think is just carbon. I then took a closer look at it, and when I flipped it over, I noticed that the bottom side wasn't properly melted. I really thought that I had avoided this by absolutely blasting it, but clearly the cold copper really did its job and the stuff at the bottom would probably always struggle to melt.
But moving on, the machine clearly worked and now the only other thing that I had to do before working with the scary beryllium was test the mold. Up until now, I also never explained how this mold even worked, but thankfully It was supposed to be pretty simple. The basic idea was that the mold itself was made from two separate pieces that were just sitting snugly in the crucible, and between these two pieces there was a cavity in the shape of a disc, which had a diameter of exactly 44 millimeters and a width of 3.5 millimeters. Then according to the manual to fill this with metal, all I had to do was melt some metal on top of the mold, which apparently wouldn't fall through the hole, until I opened the vacuum valve. This was then supposed to pull a vacuum through the hole in the bottom, which would cause the metal on the top to get quickly sucked in, and once it was full, the cold crucible would cause it to almost instantly freeze.
After that, I just had to pull the entire mold out of the crucible and open the two halves, and this would leave me with, at least theoretically, A near perfect disc of metal. In my mind this all seemed relatively simple and straightforward, and to get things going, I just loaded a bunch more iron onto the mold. I then sealed the chamber and cycled it like before, and I blasted it until it was completely molten.
Now, to turn this red hot metal into a hopefully beautiful disc, I apparently just had to quickly open this valve on the side and pull a vacuum. I then expected to see all of the metal to just quickly get sucked into the mold, except, well, absolutely nothing happened. It just sat there with seemingly no intention of going into the hole, even with the vacuum on full blast.
I then spent the next several minutes trying a bunch of different things to make it work, like increasing the power and blasting it directly in the center above the hole, but it didn't seem to make a difference. I also tried it again using a lot less iron, and I heated it for a lot longer, which I was hoping would help it melt at the bottom, and magically fix everything, except that still totally failed. So, using this mold apparently wasn't quite as easy as I assumed, and in a desperate attempt to figure out what was going on, I ended up spending the next few days trying a bunch of random things.
I also ended up trying it with lower melting point metals like aluminum and tin to see if that would make a difference, and with those, I was actually able to get some to go into the mold, but the result was always trash. Many more repeated failures then eventually pushed me to the point where I started thinking that there was actually something wrong with the mold, I didn't know I'd just suck in. And I was honestly getting pretty close to just giving up.
Thinking that this is just not the way to do it. However, I then decided to try it one last time doing something different, and I'm really glad that I did, because to my surprise, it actually seemed to work. Almost all of the metal was sucked into the mold, and when I popped it out and opened it up, I was shocked to see a nearly perfect desk.
Oh my god, it worked! That's perfect! It's actually perfect! It's, it's, it's a nice little disc.
Wow. I then attempted it a couple more times, doing the exact same thing, to make sure that it wasn't just a fluke, and it worked every time. This of course told me that the mold wasn't broken or something, and it was kind of just poorly designed, and super finicky. What I mean by this is that for it to work, I found that I had to use a specific amount of metal, heat it for a certain amount of time, and open the vacuum valve at a particular speed. If I didn't do all three of these things properly, it would usually fail, where the metal would either refuse to go in, or only a very small amount would, and in the worst cases, most of it would just get sucked out the bottom of the mold.
But with that being said, my mold crisis was now apparently solved, And at this point, I was confident that I could actually do it with beryllium, and it was time to try making some metallic glass. So I went and got all of the metals that I would need for this, which as I mentioned earlier, was zirconium, beryllium, titanium, copper, and nickel. All of these were also the highest purity that I was able to find, and besides the beryllium, which was the lowest at 99.5% pure, all of the others were at least 99.9.
But moving on, with all of these metals in hand, what I had to do next was measure them out, and get the exact amounts that I would need for that Vitreloy 1 recipe. I also decided to start out with the beryllium, which was going to be a bit more difficult, because it could only be handled in the fume hood, and when I felt that I was ready, I cracked open the bag. I then made sure to handle it as carefully as possible, so that any dust that was let off would be kept at a minimum, and I quickly found a piece that I liked. I then gently placed it on the scale, and the goal here was to try and measure out exactly 1.34 grams. With such large chunks, this wasn't really going to be possible, but eventually I was able to get 1.348 grams, which was close enough, And I sealed it all in a vial.
I then spent the next 20 minutes measuring out five more vials of the beryllium, which would give me a total of six separate attempts at making the metallic glass, and I felt that this would be more than enough. This meant that at this point, I was done picking through all of the dusty beryllium, and what I had to do next was the extremely fun part, where I got to load the masses of these six samples, into a fresh excel sheet. From there, I was able to do a bunch of calculations, where the first one was to convert all of the molar percentages in the recipe to mass percentages, and when I was eventually done several minutes later, I had the final amounts of each metal.
What was also essential to confirm was that the total mass of each run added up to around 40 grams, which would give me a volume of about 6.6 milliliters. This is because, as I mentioned earlier, the amount of metal was very important, and at least based on the tin, I found that this was the ideal amount. But either way, with these numbers in hand, I ended up spending the next hour measuring out all of the other metals, which was honestly a lot more tedious than I thought.
This was mostly just because there was a lot of them, but also because I had to use bolt cutters to carefully shave off small amounts of the pellets to get to the exact measurements. Eventually though, I was able to finish measuring everything out, where this was what I would be using for run 1, and I thought that it was kind of interesting to see it laid out like this. It was also kind of satisfying to see the other 25 vials that I would be using for the rest of the runs, and at this point I was almost ready to make some metallic glass. The only other thing that I still had to do was move the entire arc melter to my fume hood, which was honestly kind of a huge pain, and to also add a HEPA filter to this part here, which would trap any beryllium dust or vapor that might get pulled out by the pump.
Also, while I was at it, I decided to add a two-way valve to the argon input at the top, which I should have done from the beginning. And I replaced the power supply with one that I'd found on Amazon, because the original one was kind of a piece of crap. With all of this in place though, I was finally ready to try making some bouncy metal, and after months of preparation and waiting, I was very excited to say the least. So I just immediately got started, and the first thing that I had to do was add all of the metals to the copper crucible, which you might have noticed. was different than the one before.
This is because I'd swapped out the one with the mold for this one that instead had a huge well in the center, and that would actually be able to hold all of the loose metal pellets. To a small well in the back, I also added a bunch of extra titanium pellets, which was very important, and I'll explain why in a bit. With everything loaded up though, I then closed the chamber and I started clearing it from air, just like I had done many times before, and while I was doing this, I kept the fume hood glass as low as possible. I had also hooked up the vacuum pump to another foot pedal, and I moved the argon valve really close to the opening of the fume hood, which allowed me to do all of this while being completely protected. Right now, this was probably a bit overkill, but it wouldn't be later on, and to prevent any possible screw-ups, I decided to just do it like this every time the chamber had to be cleared.
When I was eventually done though, I was ready to start melting stuff, and I flipped on the new power supply. I then started blasting the titanium at the back, and this wasn't something that I had done before, but with the metallic glass, this was very important. This is because, even though the argon that I was using was 99.999% pure, It probably still contained a very small amount of moisture, carbon dioxide, and oxygen, which could all react with my metallic glass and potentially contaminate it. So, I was of course gonna have to do something to try and get rid of them, and thankfully, this was supposed to be easy to do by just melting some titanium, except unfortunately, I was gonna have to do this every single time. With that being said though, I was now finally ready to try combining all of the metals, and to melt the scary beryllium, and I would be lying if I said that I wasn't at least a bit nervous.
I was also kind of assuming that this whole process would be relatively clean and nice, especially after melting the titanium, but I was almost immediately proven wrong. Right away, there was a bunch of smoke that was being let off, which I thought was interesting. Considering that the metals were supposed to be super pure.
However, I assumed that it was probably just from some minor impurity, and I was hoping that it wouldn't get too much worse. Except it apparently did. I mean, it really looked like the ball of beryllium had just exploded and released a whole bunch of toxic dust and smoke. And this was definitely something that I did not like.
This wasn't too surprising though, considering that the beryllium was the least pure metal out of them all, and I should say that despite all of this nasty smoke, I still felt that things were going okay. But anyway, what I had now was a really hot and kind of crusty looking blob of potentially metallic glass, and in an ideal world, this melting step would have now been done. Unfortunately though, the world that we live in is far from ideal, and in this case, there was probably still a bunch of unmelted metal at the bottom, which I was gonna have to deal with, using this convenient tool that was built into the machine.
At least according to the manual, it was supposed to allow me to easily flip over the metal. However, I quickly realized that this was gonna be extremely frustrating. After five minutes though, I was eventually able to flip it, and on the bottom, it was possible to see the outline of all the pellets that clearly didn't melt, and I just immediately started blasting it. It then all quickly started melting again, and getting much more evenly mixed, and what was nice, was that it didn't seem to let off nearly as much smoke. What was less nice though, was that according to every source that I had read, To get a good uniform mixture of the metals, I was gonna have to flip it at least three more times.
This definitely wasn't ideal, but I also really didn't have a choice, and when I was eventually done over 20 minutes later, I had a nice blob of red hot metal. Now all I had to do was just let it sit there and wait for it to cool down, and for the cold crucible to quickly pull away its heat. At the same time though, I felt that it was a good idea to watch it very closely, and to see exactly how long it took to solidify.
This is because in general, when making metallic glass, getting it to freeze as fast as possible was extremely important, and with the Vitreloy-1, this was going to have to happen in less than a few minutes. This way, there wouldn't be enough time for any of the metal atoms to crystallize, or to form any ordered structure, before solidifying. Instead, at least in theory, they should all get locked in a random and amorphous state, and this is supposed to form a metal with very different properties that's commonly referred to as a metallic glass.
With that being said though, at this point, it looked like it had completely solidified, and considering that it had only taken about a minute, I was relatively confident that this was metallic glass, and that I now had a beautiful blob of bouncy metal. Before drawing any real conclusions though, I was first gonna have to get it out of this beryllium riddled chamber, and with this setup, it was thankfully pretty easy. All I had to do was lower the glass as low as possible, and to pull out all of the dust by turning on the vacuum pump, and trapping it in that little filter.
I then refilled it with argon, and I repeated this cycle a few times, just to make sure that there was absolutely nothing left. After that, I flipped the two-way valve, which opened it to air and equalized the pressure, and now I was able to safely open it. So, I carefully flipped up both of the clamps, and I slowly lifted the chamber, to reveal my beautiful chunk of hopefully metallic glass.
I then removed it from the crucible using some tongs, and I wiped it down with some wet paper towel, to remove any toxic residues, and I took a closer look. What immediately stood out to me though, and what honestly kind of surprised me, was that at least in this form, it didn't seem like it was anything super special. It really didn't feel any different than normal metal, and I mean, I know that I was holding it with tongs, but still.
With a density somewhat similar to pure iron, it kind of just felt like one of those iron blobs. that I'd made earlier. I of course couldn't just judge it by how it felt though, and I was gonna have to be a bit more scientific, and I was gonna have to do an actual test. So I went and got one of my ball bearings, which are normally used to just shoot from a slingshot, and when I dropped it, I was pleasantly surprised by what I saw.
I mean, it obviously wasn't even close to as good as what Steve Mould had, but it- definitely seemed like it was abnormally bouncy, and I thought that it made a really interesting sound. There was something about it that just sounded springy to me, and I felt that this was a really good sign. To be absolutely sure that this was anything special though, I of course had to compare it to some regular metal, and I decided to just use one of the blobs of iron.
Then, all I had to do was drop the ball bearing, and I was honestly a bit worried that it would end up bouncing just as well, and that what I had made was actually trash, except it was almost laughably bad. It wasn't even close to anything that I would say even resembled a bounce, and this told me that I was definitely on the right track. At the same time, I also felt that turning what I had into a proper flat disc, and that giving it a solid base to sit on, would make it significantly bouncier, and that was of course what I had to try next. Before moving on though, I was first gonna have to finish melting the five other runs that I had prepared, and I did my best to do this as fast as I could. However, despite this, it was still a very long and tedious process, and what made it even worse was that runs three and six apparently had beryllium that was even more explodey, and it looked like the beryllium was sprayed everywhere.
It also filled the entire chamber with so much smoke that it was almost impossible to see anything, and this was quite unfortunate to say the least. On the bright side though, the other runs seemed to be totally normal, and not even make that much smoke, and I also ended up seeing something that I thought was really interesting. More specifically, this was during run 4, when I had accidentally touched it before it had fully solidified, and this caused a small amount of it to get pulled away as a little string of goo, and this wasn't like any other metal I had ever seen. In my opinion, it was a lot more like melted plastic, or molten glass, And I felt that this was another really good sign that I had actually made metallic glass.
But anyway, when I was eventually done four hours later, I had six beautiful blobs of metal, and I was feeling pretty hopeful. I also did a quick ball bearing test to confirm that they were all bouncy, and at least from this, they all seemed to be about the same. The only sad part was that when I weighed them all, The ones that were extra explodey had lost way too much mass, and this told me that most of the beryllium had probably just been sprayed around the chamber, and I wasn't going to be able to use them.
However, I didn't think that this was a huge issue, because even without them, I would still have four attempts to make the discs, and I felt that this was more than enough. With all that being said though, at this point, it was getting really late, and I was completely exhausted, and I decided to come back to it the next day. When I returned in the morning, I was really motivated to get started, and the first thing that I did was swap out the crucible for the one with the mold.
I then dropped in some fresh titanium, and onto the mold, I added the metallic glass that I had made in the very first run. After that, I closed everything up, quickly did all of the argon cyclings, and melted the titanium, and I was ready to go. So I just started blasting it using the same technique that I'd successfully used on the tin, and when I felt that it was hot enough, I carefully opened the vacuum to the mold. This then caused almost all of the metal to get quickly sucked in, only leaving behind a small bead at the top, and I was genuinely shocked that it had actually worked on my first try. So apparently, all of my practice had paid off, And now, inside the mold, there should have been a nice, and hopefully perfect disk of metallic glass.
So, the only thing left to do was to get it out, and to do that, I quickly cycled it with the gas, and I flipped open the chamber. Then to release the mold, I added some screws, and with a makeshift piece of metal, I was able to carefully pull out the entire thing. With this in hand, the only thing left to do was open it up, And inside of it, I was expecting to see a nice and beautiful disc, but I was immediately disappointed. This is because what I got instead was an ugly piece of crap that somehow didn't even fill the entire mold, and what bothered me even more was that it seemed like it was pulled in very weirdly, which caused this gap to form in the middle.
On top of this, there was a lot of metal that was definitely getting sucked out the bottom, as evidenced by the tail, and what was also really bad, was that it looked like even if the button at the top had gone into the mold, It would have only barely been able to fill it. Overall, this was a little sad to say the least, but at the same time, I wasn't too worried because I of course still had a few more attempts, and I felt that they would be more than enough to succeed at least once. However, apparently, they were not, and while the next attempt was technically better, the one after that was an absolute disaster, where almost all of the metal had gone right through the mold.
This left me with only one last try, and at this point, I felt that it was pretty clear that I was gonna have to do something different, so I decided to cut off some of the extra stuff from all of the failed attempts, which really did feel like trying to cut glass. I then combined all of it together, and I was hoping that with all of the extra metal, it would actually fill the mold, except it was the complete opposite, and I was never able to get any to go in. All that I ended up with was this huge, and kind of useless chunk that I couldn't cut or break apart because of the beryllium, and at this point, I couldn't help but feel pretty frustrated with how things were going. I mean, it was probably a bit naive to think that it was all just gonna magically work, especially because the vitreloy had completely different properties than the tin, which probably required a different technique, and I had never even worked with it, but still. After all of these failed attempts, I couldn't help but kind of lose any faith that the mold was actually gonna work or be remotely reliable, and I really did not have the motivation to spend another entire day measuring out all of the metals, and then doing all of the melting and flipping, only for it to just probably fail again.
Despite all of this though, there was still some hope, and I hadn't completely given up yet, because there was still one last thing that I had to try, which was that special 44 millimeter well that I had mentioned earlier. I had made these just in case this exact scenario happened, where the mold completely let me down, and in my mind they were so simple that there was no way they wouldn't work. However, long story short, it again ended up being a complete disappointment, and this is because it didn't at all spread out into a thin layer like I was hoping it would, and instead it just sat there as a ridiculously thick blob. I then just kinda said screw it, and I decided to keep adding metal until I got a full disc, which ended up taking all of the metal that I had, and when I was eventually done, I was left with this horrifically ugly monstrosity. I mean, on the bright side, unlike all of the other attempts, it was actually a full disc, and there was part of me that wanted to just accept this as my final result and call it a day, but I knew deep down that it was complete trash.
There was also no way for me to fix it without a ridiculous amount of grinding or machining, and it was at this point that I basically just gave up. Over the next few weeks though, I couldn't stop thinking about it, even while I was working on other projects, and inevitably, I ended up with a bunch of ideas that I really wanted to try. For some reason, I was also feeling really confident that I could actually get the mold to work properly and consistently, and I decided to just jump right back into the project, and I started weighing out the metals. This time though, I was very careful to only choose the beryllium balls, and none of the shards which seemed to be more explodey, and I also weighed out 30% more of everything.
This is because I felt that 30% more metal would be the perfect amount, and that it would be just enough to compensate for all the metal that was getting shot out the bottom, while simultaneously not being so much that it would refuse to go into the mold. I'm not sure why, but I was also feeling weirdly optimistic, so I decided to only give myself four attempts. And with all of the metals ready, I did all of the same melting and flipping steps. When I was eventually done, I had these four beautiful blobs of metallic glass, which pretty much looked exactly the same as the ones from earlier, except they were 30% bigger, and I loaded the first one onto the mold. I then did all of the typical gas cyclings and titanium melting, and after hours of tedious work, it was finally time to see if any of my new ideas were actually good, or if it would all end in another complete failure, and I just started blasting the metal.
As usual, this caused it to all quickly melt, and now, the first major thing that I was going to do differently, was to just keep blasting it for a ridiculous amount of time. This is because I found out that unlike normal metal, metallic glass is significantly worse at conducting heat, and it was very likely, that it just took a lot longer for the entire thing to fully liquefy. So I decided to heat it for no less than a minute at near max power, And when I felt that it looked good, I moved the electrode to the middle, and I continued blasting it there for another 10 seconds. This would just really make sure that it was fully liquid above the hole, and then after that, the only thing left to do was open the vacuum valve. I also had to do this slowly and very carefully, which I felt would help prevent the metal from just blasting through the mold, and almost right away, It all started getting sucked in.
Then not long after, it started looking very similar to my other attempts, where there was just a small bead left at the top. The only difference this time was that the bead looked like it was slightly bigger, which kind of implied that it didn't all go in, but at the same time, I did start with a lot more metal, which made it hard to tell. Either way, I then let it cool, cycled the gas, and popped it out.
And part of me was mentally preparing for what was probably another miserable failure. However, that absolutely was not the case, and inside of the mold was what looked like a near-perfect disc. Oh my god, it actually worked!
I mean, it still had some issues, but it was way nicer than any other attempt, and I really couldn't have asked for anything better. Yes! Yessssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss I was actually able to make three more discs. Like the first one though, none of them were quite perfect, and they all had some slight defects, and on the surface of them, they all looked like they had some weird ripples. However, from what I could tell, this was extremely minor, and it only seemed to be visible because they were so shiny, and overall, I still felt that they were more than good enough.
But moving on. Now that I actually had it in disc form, what I of course had to do next was test out my theory from earlier, and to see if it was bouncier. So, I went and got my stainless steel base, which had been made by my friend Justin, who runs the channel The Thought Emporium, and he was able to machine it, to the exact same dimensions as Steve's.
I then carefully placed my disc on top of it, and I went and got my ball bearing, And for what felt like the millionth time, I was overcome with disappointment. This is because somehow it seemed to be only barely bouncy, and it was apparently even worse than that huge piece of trash that I had made earlier. At first, this kind of surprised me, and I started thinking that I probably just screwed up measuring or mixing the metals, but at the same time, I knew that something was off, Because of the sound that it was making.
Every time the ball hit it, it was making a clinking sound instead of the bouncy one that I had heard earlier, and in my opinion, this had to be happening because the disc wasn't perfectly flat. This was probably causing it to slightly pivot and to get slammed into the base every time it was hit, and it was very possible that this was absorbing enough energy to completely ruin the bounce. It also thankfully seemed like it was a relatively easy thing to fix, and the obvious answer was to just glue it down.
Before I could do that though, I was first gonna have to clean it up a bit by cutting off all of the extra stuff with some bolt cutters, and by shaving down the nubs with some wet sandpaper. I also had to very lightly sand down the side that I was gonna glue so that it would stick better, and I did make a bit of dust doing this. but it was also completely unavoidable. When I eventually felt that it was looking good though, I shot a bunch of super glue onto the steel base, And I quickly squished on the desk. I also slid it around to make sure that everything was covered with glue, and I honestly did a pretty horrible job, and the glue is dripping everywhere.
However, regardless of that, I still felt that things were going pretty well, and after waiting a day for it to fully cure, I quickly cleaned it up. I then carefully slid some acrylic tubing over it, which I was able to find on eBay, and now... It was time to see if the glue had magically fixed everything.
So again, I went and got a ball bearing, and I nervously hovered it above the tube, and well, it did somehow magically fix everything. I mean, it definitely seemed like it was bouncing just like in Steve's video, and thankfully, it was no longer making that clinking sound. The new sound that it was making was definitely more bouncy, and it was much louder than I expected, And at this point, I was still genuinely surprised that it was actually working.
It actually works. That is really cool. It's actually amazing.
But with that being said, I was now super happy and relieved that the glue had worked, and at this point there was only one thing still left to do. And that was to get all of the other discs onto a steel base. So I quickly cleaned them up just like the first one, and I glued each of them to a fresh steel cylinder that I'd made using a service called Proto Labs Network.
This gave me a total of four nice looking little setups, and with all of the extra stuff that I had to cut off, I was actually able to make a fifth one. I then gave all of them their own tube, and now I was finally able to do what I dreamed of when I started this project, and I tested them all at the same time. This was honestly mostly just for fun, but I was also really curious to see if there would be any difference, and I wasn't surprised to see that there was.
I mean, I think the metallic glass itself was mostly fine, and I think the main difference just came down to the quality of the disc surface. For example, the third one had by far the crappiest surface, followed by the fourth, where both of them had little bumps and divots that I was hoping wouldn't be a problem, but they consistently performed the worst. In theory though, I think that this could all be easily fixed by just grinding and repolishing the surface, but as of right now, I don't really have a way to do that safely. At the same time, I'm also more than happy with numbers 1, 2, and 5, which seem to perform almost just as well as Steve's, and that's more than enough for me.
This is because it allows me to keep one for myself, give one to Steve, which I promised I would do in a comment on his video, and to also give one away to you guys. If you want a chance to get it, there are also no real requirements, and you don't need to subscribe or comment, Or do anything like that. All you need to do is send a nice email to giveaway at nylered.ca and after a week, I'll have the winner randomly picked.
And just as a point of safety, because I don't think I said it in the video, the metallic glass in this disc form is totally safe, as long as you don't break it apart or try to sand it or grind it. But with that being said, in the end, This project turned out to be a lot trickier than I thought, especially with the whole mold setup, and I'm glad that it actually worked out. What's also nice is that it gave me a huge interest in metallic glass in general, and I plan to do some more videos on it in the future.
For now though, I think I'm done working with nasty and toxic beryllium, and I'm gonna try focusing on some even more dangerous projects, like making nuclear-powered glow toys. Alright, so this idea is something that I've been wanting to do for years, ever since I made regular glow toys, and I think I'm pretty close to being able to do it. There are still just a few things to work out though, to make sure that it's all done safely, and it may or may not end up being really expensive. So, this is why, as usual, I'm extremely thankful for all of my supporters on Patreon, and for my YouTube members, for making projects like this possible.
And of course, for the sponsor of this video, Opera. I've been using Opera on desktop for a while now, and in my opinion, it's still my favorite browser. This is because it's both really fast and intuitive, and it's loaded with a bunch of extra features, and it also just got new updates to its native browser AI called ARIA.
One of the biggest updates is that you can now access ARIA without even using the sidebar, and it's possible to open it by just hitting control slash or command slash. Another nice update is that if I then hit tab, I can enter page context mode, which is a neat tool that uses ARIA to extract information and to summarize text on any page. Also, something that I want to add is that it's totally optional whether or not you use ARIA, and it would only be activated if you choose to turn it on. Besides ARIA though, they have loads of other features, like their tab islands, which has really helped me get all of my tabs under control, especially when I'm doing research for a project. What's also super convenient is that they even have a free built-in VPN service, an ad blocker, as well as this sidebar, which I think is really useful.
This is because it lets me easily access all of the social media that I often use, like Messenger, WhatsApp, Instagram, and TikTok, and it even has an integrated music app, where I can listen to things like Spotify or Apple Music. On top of all of this, they also have something called Lucid Mode, which I think is pretty cool, because with just a click of a button, it's possible to immediately enhance any video. With that being said though, I've been having a really nice time just using Opera in general, and I still think that it's worth giving it a try for just the sidebar alone.
It's also of course completely free to use, and you can download it right now by clicking the link in the description.