บทถอดความฉบับเต็มของ 48: Jonah Myerberg Pt. 2 - Inside Desktop Metal's Blueprint for Manufacturing

it's very important that those particles be cleaned before they're densified. Because once they're densified, then it's sealed. That entire metal part, like you imagine, like a solid metal part that's not a sponge, that's what comes out of the sintering furnace. And so during that densification process, you don't want to trap contaminants inside of the three-dimensional object because that can turn into, you know, Like changing its chemical makeup, essentially. That's right. That's right. Or even having inclusions within the metal that would affect the properties. Yeah. Okay. So one thing that you did mention that I just have to ask, because now you've mentioned a couple times, the green body. Why do they call it that? It's a good question. Early on in the manufacturing process, it's weak. So, for instance, when we talked about green bodies that were created 100 years ago through pressing, which is still around today, this is a manufacturing process that is widely used in metals manufacturing. It's called powdered metallurgy. And you take the powder and you compress it into the shape that you want. That's your green body. And then you put it through a centering process to make it strong. That green body that's created through compression is weak. You could break it. You could snap it. It's brittle because it's just a bunch of particles just pressed together. Gotcha. And the same thing is true with metal injection molding. When you injection mold that mixture of wax and metal powder, that green body is also weak. And you can snap it and you can break it. But once you put it through the centering process, it becomes the solid. you know, fully dense piece of metal. And the same is true with binder jetting, where now we've evolved the manufacturing process of powdered metallurgy and metal injection molding to a new sensor-based process of binder jetting, where you create this green body from the same powders glued together. And that green body is relatively weak. You can snap it. You could break it. But then you put it into the sintering furnace and the sintering furnace densifies it, strengthens it, and gives it that final property of fully dense metal that you need. So desktop metal, what really set you guys apart like out of the gate was your proprietary, I don't know if that's the right word to use, printing process, this binder jetting that you talked about, which was one of your co-founders. discovered this process. Is that correct? Yes. So binder jetting of metal powder was developed early on in the 80s by one of my co-founders, Ellie Sachs. And Ellie was a professor at MIT, still is. Okay. And he developed the technique, the process of gluing together the powder using a printhead. And the technologies, the wide, page-wide... printers and print heads that we use today weren't available back then. And so what we did is we took the work that Ellie developed in the 80s and we accelerated it with two-dimensional printing technologies that had been

developed over the past 30, 40 years, really leveraging 2D printing as well as the powdered metallurgy that had evolved over the last 100 years. So putting this all together into a new technique of creating that green body and then putting it through the censoring process. Well, and like we said before, like your founding group and bringing in different professors in your background versus his, like it just seems like you guys kind of had this cohort of very different backgrounds that kind of made up this group to create something that was. you hadn't seen before. And that's just super interesting that you guys all have different backgrounds and different industries that you might have come from, but you created something that hadn't really been done before. That's right. That's right. So we created this multidisciplinary team from professors at MIT in metallurgy, in ceramics, and in mechanical engineering that allowed us to create a very kind of vertically integrated printing system. So everything from the materials through the machines and the process. And one of the things, one of the materials that binder jetting prints very well and is not possible to be printed any other way, not possible with lasers, is a ceramic material. Ceramic material, unlike metal material, doesn't like to be melted. So it really requires a sintering process to create complex geometries or even simple geometries. So in fact, ceramics are traditionally manufactured through this press and sinter method that we were talking about before to create very simple shapes that are then kind of machined into more complicated shapes. But you create these monolithic blocks by taking ceramic powder and compressing them and then sintering them into a full density. and then machining them. And ceramics are difficult to machine. They'll wear out your bits. And so creating complicated machined parts from ceramics is difficult. So since ceramics start as a powder anyway and have to go through a censoring process, they're a perfect fit for binder jetting. And so binder jet 3D printing, the desktop metal method, is a very efficient and effective way to make ceramic parts. So what is the application for some of these ceramic parts? Like where would you normally be seeing these? So right now, a lot of our applications are being driven by military and defense, as well as aerospace and space programs. And specific ceramic materials like silicon carbide are very, very attractive materials for defense applications, aerospace applications. These are materials that are very strong. very thermally conductive and very stable. So unlike metals that might change their shape as temperature changes. Just very reactive. Yeah, exactly. So if you increase the temperature, a metal could grow and expand. And the opposite, if you go down to very, very cold temperatures that are in space, material like a metal will shrink. It's called the coefficient of thermal expansion or the CT. these ceramics can be designed to have very, very low or even zero or negative CTs. And that is very attractive when you start manufacturing these components on Earth and then putting them up into space or putting them into very high temperature applications. So binder jetting has been used now to create optics, mirrors, and... from telescopes to directed energy weaponry, so focused lasers that require these mirrors for beam shaping. And this is taking products, assemblies like telescopes that might be the size of this room and shrinking them down to the size of a bread box, something much smaller. And they still have like all the same... they would have when they were these colossal things. Yes, and that is one of the major advantages of additive manufacturing, where you can take assemblies that are very large and complicated. But because they're really only large and complicated because you have so many different pieces that you have to manufacture independently and then piece together. What additive manufacturing allows you to do is consolidate assemblies. So if you take up an assembly that might be 30 pieces that have to be manufactured independently, you can now print those 30 pieces together as a single piece. It allows you to bring everything down in size and weight and complexity. And one thing that I've just been thinking about based on other people that have been on the podcast, we have a lot of different people that have been on that are fabricators, manufacturers, whoever it may be. And when we go to different trade shows like Fabtech and you see all these insane machines, whatever it may be doing, we see water jetting and CNC machining. One thing I'm thinking about is how if people needed to make metal parts. What would be the difference in doing metal 3D printing versus just doing CNC machining? Yeah, yeah. It's a great question, and it's often a misrepresentation of what 3D printing really is. What we try to do is set the expectation of 3D printing just like you would expect any other manufacturing process to be part of the overall, the total process. So sometimes the misconception is that All you need to do is throw a digital file at a 3D printer and you get your part out the other side. And it's done and it's ready to go. But that's not the case out of, you know, for really any manufacturing process, right? There's no replicator like, you know, in Star Trek where you push the button and you get what you want out and it's finished. 3D printing, just like any other manufacturing process, is just one piece of the puzzle. Just like in casting. When a part comes out of the mold of casting, it's a bit of a mess, and it needs to be cleaned up. And often it's unpredictable in its geometry, and so high-precision surfaces need to be machined. High-precision holes need to be drilled. So casting works in combination with CNC machining, just like 3D printing works in combination with CNC manufacturing. So you have these technologies working together. that make an

overall more efficient process. An example of CNC that you bring up is a good one because traditionally CNC machining will buy raw materials off the spot market that are your commodity materials, like a cylinder of steel or a large beam of steel. And that inside of that prism of steel will be the part that you want at the end of the day. So you have to remove all of that steel. to get to your final part. Which to me on the outside sounds somewhat wasteful. It does. That's right. And what 3D printing brings is this ability to, instead of starting with this giant prism that you need to remove 90% of the material from, you print a part that is much closer to the final part that you want. It's called near net shape. And it's very close to what you want, but not exactly. And you've printed extra material. in areas where you're going to come back and machine and remove that extra material. That's interesting. So you're just moving from two different directions to end up at the same point. That's right, because when you look at these parts that were machined, in order to remove the material and reduce the weight as much as possible, you have to touch every surface within that part with the machine tool, which can be long and expensive to do. But a lot of the surfaces that you've touched are not super critical surfaces. They can be left as cast and not machined. That's what 3D printing does is it allows you to print everything as a near net surface and then come back and precision machine them. So 3D printing is just a new tool in a manufacturer's tool belt or a designer's tool belt that is part of the manufacturing process. Okay. So in this 3D printing, what kind of materials are you guys usually printing with? Yeah. It's interesting because almost every different metal alloy and ceramic alloy could be 3D printable one way or another. So there are, as we talked about before, there were a number of 3D printing processes that have emerged over the past 20 years. And each one has its sets of alloys that it's best at printing. different metal alloy is somewhat 3D printable or is 3D printable in general. And so what we focus on at Desktop Metal with the Sinterpaste materials are stainless steels, super alloys, like nickel super alloys, as well as titanium and copper. But our bread and butter are stainless steels and nickel super alloys. Super alloy sounds crazy. What exactly is this super alloy? Yeah, so I think super alloys got their name based upon pushing the performance of that alloy well above what a normal steel alloy could handle. And so when you look at the strength, that is required at temperature for a lot of these applications. Stainless steels just, they don't cut it. A lot of stainless steels, you know, are perfectly fine at two, three, 400 degrees Celsius and operating at that temperature. They don't lose too much of their strength. But that's the key is how high of a temperature can you operate? that metal under load before it loses its strength and becomes really too weak and too soft. And so at high temperatures, traditional stainless steels will become weaker. Their yield strengths will go down. They'll end up creeping more. And if you creep under pressure, under applications, that's how you fail. So super alloys were developed to resist that. creep at higher and higher temperatures. So in other words, to maintain their yield strengths as the temperature of the application was pushed higher and higher. So some examples of these applications are within turbine engines. These are very, very high compression chambers, high pressure, high temperature compression chambers within a turbine engine that has to run at a very high temperature. and is spinning at 100,000 RPMs or greater. So your centrifugal forces on these components can be extreme. And if the metal doesn't maintain its strength, it will creep and will fail. That's how accidents happen, yeah. That's how accidents happen. So the industry, for the turbine industry, the energy generation industry, jet propulsion have developed. new alloys that perform better and better and better at higher and higher and higher temperatures within these applications. And these are the alloys that are being used today in the defense and aerospace applications as we're seeing jet engines become miniaturized. So these large jet engines that are the size of this room are starting to become these tiny little jet engines that... that power drones, right? Instead of large aircraft, these engines power drones. And because of that, they have to be much more efficient. They have to be much more powerful on a, you know, on a thrust per weight basis. And so these applications are extremely demanding for the highest performing super alloys. And is a super alloy like, a mixture of different metals and like kind of taking all their best components out of those or is it something totally different yeah so just like any uh alloy you you have um you know a variety of different metals that come together to create these special properties um and a lot of these super alloys are their base metal is a nickel which has a very high temperature uh performance um and then you blend in other components that uh that give you the properties that you need to. And what happens with these super alloys is they become very high performers at high temperatures. But they often also become problematic with welding. So you can't weld a lot of these super alloys. Because they're supposed to be withstanding the heat. So welding seems like it would work against it a little bit. I mean, it's not exactly that reason. the microstructure and the crystalline structure and the grain boundaries and everything that happens within these super alloys as they're heat treated to the performance levels that they need to work at, they become more sensitive to heat affected zones or gradients within the metal, temperature gradients that form, especially as you're trying to precipitate different.

uh grain structures throughout the metal so that the long story short is that they become very difficult to weld so in other words you don't want molten metal sitting next to solid metal um because that doesn't it tends to not react well gotcha and that's exactly the way that laser-based printing is performed because you have solid metal next to liquid metal and it's just kind of like fuses together in layers. That's right. And so if you have a very weldable metal, that's no problem. You can do that. But when these super alloys, the metals often become less weldable. And so you don't want to put molten metal next to solid metal. So that's when sinter-based additive, the type of process that I'm describing with desktop metal, become very efficient because you're you're taking all of the alloy at once, bringing it up to the same temperature, which is below the melting temperature, allowing sintering to take place, and then cooling it down. So it's a much more homogenous and even thermal process. And therefore, we're able to print these very, very high-performance super alloys, like Inconels. and MARM247 and CM247. These are all advanced super alloys that were developed for the energy generation and jet propulsion industry. We're able to 3D print them where laser based processes are not. Would not be able to do that. Right. So tell me a little bit more about the I mean, you've been with Desktop Metals since the beginning. So I want to know more about the evolution of your product of your. printers, because I know you said you guys are very vertically integrated where you're there from like getting the client or the customer set up with their printer, helping them learn how to use it. And you even said like on the raw material end of it as well. So you're kind of through that entire process. But tell me, how did it start? with the very first printer you made to what are you guys making right now? Yeah, so it's a great question. So over the past 10 years, we've discovered that we have two basic categories of customers that we work with. The first customer is really getting to know... printing, especially metal 3D printing, and wants to do research or explore it on a kind of an educational level. And those customers are universities or national labs where they want to explore different types of materials in 3D printing, or they just want to learn about the 3D printing process, teach their students, which is becoming really important right now. Because when I grew up in university, we didn't learn about 3D printing in our classes. We took manufacturing classes and we were taught about all the different types of manufacturing processes that were out there. But 3D printing was not one of them. And that's changed now. 3D printing is becoming a mainstream topic and a mainstream manufacturing process that is taught in schools. So universities want to give their students hands-on experience. So 3D printers that are compatible with universities, research, that

is one category and that's our kind of entry-level printer that we offer and we enable universities to teach with or national labs to research with. On the other hand, we have a number of customers who now want to scale 3D printing into manufacturing. Like making many, many parts. That's right. They want to use it and they want to use it in production. And for that, for those applications, we develop a different... class of printers, printers that will go into a production environment, produce at high volume, fast, inexpensive, and allow our customers to return on their investment. There has to be an ROI for those printers. You know, if you're going to invest hundreds of thousands or a million dollars in a printer. They got to be able to get some parts out of it. That's right. You need to be able to produce. So, yeah, so in both of those two, Those two categories are very different printers. So as you mentioned, we stay very vertically integrated. We provide raw materials. We provide printers. And then we provide the processes that are used to run the printers. And then we steer our customers towards post-processing applications that can help them reach their final. their final part, you know, geometry, whether it's a machining operation like CNC or it's a heat treating operation or a polishing operation. All of the metal parts that come out of our 3D printers are just like any other metal part that's produced, can be machined, heat treated and polished the same exact way. Wow. So, and tell me a little bit, so I know you are the co-founder and you are the CTO as well. And while I would imagine that that, entails a lot of different things that you probably have to do. You probably oversee a ton of different things for the company. But what is your role within the company right now? Are you actually designing these systems? Are you looking for the right customer? Are you kind of doing all of the above? Yes. So because we're so vertically integrated, we are evolving every aspect of the three-dimensional printing process. New printers, absolutely. We've been working with our customers for 10 plus years and we take direction from the input from our customers. So we constantly are listening to the voice of the customer and then reacting to their needs for the next product. For printers, that means we're making the printers faster, larger. able to mass produce less expensively so that the customers can reach the next level in performance of 3D printing. We're also introducing new materials. So we're working on the next generation of metal alloys, of ceramics, and listening to the customers of what type of materials they need to solve the problems that they're faced with. Absolutely. That is... so interesting. And when I think we were talking a little bit earlier, just with the evolution of our industry and 3D printing and manufacturing and kind of all of the above was this topic of AI. And obviously, everybody is talking about AI, like the normal everyday person, companies are talking about how they can implement it. And I think it is a little bit interesting to get just because you guys are such a, you know, tech focused, very progressive, kind of like forward thinking company. What are your thoughts and opinions about AI? And are you guys using it within your company? Yeah. Yeah. So, I mean, AI is something that I think everyone has to begin to learn about, right? If you're never going to use it, you need to at least know about it. Yeah. And the same is true with 3D printing. If you are not... about 3D printing, then you're already kind of behind. Really? And this is why I was excited to come on your podcast and talk about 3D printing with you because it's easy to look at the industry from the inside and think how big we are and how many different opportunities there are. But the minute you step outside of the 3D printing industry, you realize how small it still is. It's still very early and it's... its history. And it's not well known. And you're asking all these questions that are great ones. And we realize that the minority of manufacturers really understand 3D printing. So just like AI, you need to at least be aware of where 3D printing is and where it's going, even if you're not going to immediately use it. You need to know about it. AI is the same way. And I think that AI is not going to steal our jobs, right? It's the people who are using AI, the best, who are going to be the ones that end up stealing our jobs. So we need to learn about these technologies and understand them. And within Desktop Metal, one area where we're using AI is in material discovery. So as a mechanical engineer, we design parts. as much for their application as for their manufacturing process. So when you look around the world and you see a part that's designed for a certain application, you'll also find many features within that part that is designed for the manufacturing process. And over the course of the history of other manufacturing processes, like let's say CNC machining or die casting, Materials have been developed and evolved for those manufacturing processes. So when you want to cast a steel or an aluminum, you're using casting alloys or alloys that were developed specifically for casting processes. Same with machining. There can be alloys of aluminum and alloys of steel that were developed for machining because they machine better than others. And the same is true for additive manufacturing. And this is where it gets really exciting right now because we're on this growth curve of additive manufacturing and we haven't yet... the alloys for additive manufacturing yet. So we're using AI now to discover and develop brand new alloys that are specifically designed for additive manufacturing and give you even better properties than what's currently available today. So that's like, is this just something you're utilizing kind of in the background to be doing A lot of research you guys might not have the

time or patience. You guys got a lot of things to do probably to be doing. And have you seen like a lot of success using AI specifically for that topic? Yes, yes. So material discovery can be extremely complicated. You've got, you know, tens of different alloying elements that can come together to form very complicated, you know, grain structures. It can be very unpredictable. And so there have been simulation tools developed to help to discover this. But for the most part, it's been manually driven by metallurgists and material scientists who have, through experimentation, just tried different combinations. But now AI brings in a whole other level of being able to... and understand what's happening at the atomic level and crystalline level of alloy design. And so just like casting alloys were developed in aluminum to flow better through a melt flow process and fill a mold better than other aluminums, we're going to see the same thing happening now with sensor-based additive where alloys of super alloys are developed that have even better performance than the current super alloys and have better printability than the current alloys as well. Wow. So just with your experience with using it, what do you see for the future of AI and all the industries that might be using it? Where do you see it going? It's probably a big question. It's a huge question. I'm very excited about the, you know, just the possibilities that it brings. Again, it's just another tool. And if leveraged in the right way, it will help, you know, 10x, 100x the performance of what we're doing. Absolutely. And we're already seeing that, you know, and the industry is seeing that. It's still super early in the history of AI, you know, to make any vast predictions. But, you know, these next 10 years are going to be very interesting. It's going to be very, very interesting. That is for sure. And one thing just with the evolution of your company and your printers and even with, you know, pairing with tools like AI, where do you see the evolution of your company going? And do you see maybe like, because I was telling you earlier that I'm used to the 3D printers that you can even get to have inside your house and making like little. plastic parts and fun little things like that. Do you see the metal 3D printing side of that becoming like commercialized in a sense of like anyone just being able to have one like on their desk at home and like things like that? Or is this probably going to stay more of like manufacturing industry is going to be having this tool? Oh, no. I think the ultimate goal is to open up. the accessibility of metal 3D printing exactly the way that you're experiencing it with plastic 3D printing. How that evolves or how that happens, I have no idea. But it's very early in the history of 3D printing. And it's not every day that a new manufacturing process is born. So we have a handful of metals manufacturing processes out there. And you mentioned a couple of CNC machining

and die casting. sand casting and metal injection molding. And it's not every day that a brand new way to make a metal part is formed. So we're 20, 25 years into the history of metal 3D printing. And that is typically with manufacturing processes where things begin to accelerate. When we reach maturity and we begin to we begin to have more mass adoption and there's more focus. And I do think that the goal is to ultimately have a metal printer inside of your house. I think if you rewind 20 years from now, it would have been hard to imagine a metal 3D printer inside of a university or a school. Absolutely. But there's high schools now that have... are metal 3D printers inside of them to teach students in vocation. Which even back when I was in high school, which I mean, it wasn't that long ago, but like that, that would have been very odd to me. It would have been like, we are living in the future right now. That's right. That's right. So yeah, I think ultimately that's the goal. It's beginning to happen now with plastic 3D printing. We've really hit an inflection point on the accessibility of 3D printers in the home. Yeah. You know, I'm a true believer that, you know, contractors will all have a 3D printer on their truck just like they have a hammer and a chisel and a screwdriver. That if, you know, that 3D printing will become a just a normal tool of the trades. And right now that's happening. It's not, you know, it's not a speculation. That's happening right now with. plastic 3D printing. And I know it will happen with metals at some point. Yeah. Do you think with speeding along that timeline of more people... Even seeing that that can be a part of their manufacturing process, is that just getting out and educating people? Just like you are right now with me today, I've learned a lot. Is that just kind of getting out there and educating more people on the metal 3D printing process? Like you said, more schools and universities are teaching it now, which I feel like is a pretty big step in the right direction. Yeah, yep. So for sure, when we step outside of the 3D printing world and we see how many... just aren't aware of metal 3D printing. That's really encouraging because there's so many opportunities to take metal 3D printing into areas where it's not right now. But that is a slow, systematic, and needs to be a controlled process so that we don't overstate expectations and we don't under-deliver in areas. So we're approaching it from that perspective. But yes, awareness is super important. reaching your listeners, starting to talk about this is the first step. In fact, you had a company on your podcast called Ryerson previously. And Ryerson is a great example of a huge steel company who has recently invested in 3D printing. They purchased a group in Pennsylvania that focuses on 3D printing. And I imagine the reason they did that is because they see 3D printing as the future, that they need to begin to learn about it. They need to be able to offer it and develop it. And they don't want to be behind. They want to be a leader in the space. And they're not the only one. There are a number of companies across. the world, big manufacturing companies that see 3D printing as a tool that they need to learn about. Just like AI. Just like AI. That if you don't learn about it now, then you'll be behind. And even if you don't use it this year or next year, it's still something that you need to watch. The awareness and the education. Because I mean, that's something that our team has talked about so much, just being in the metals industry, is that it's equally... like a very antiquated industry but then you go to shows like fab tech and you're like these are the most advanced machines and technologies i've ever seen in my entire life but then at the same time sometimes you talk to people and you're like this is still kind of antiquated because we we deal with that all the time being in the procurement side of metal products your raw commodity steel materials that we are online metals marketplace for those buyers and sellers to transact on and you would think that that wouldn't be super you're like we do we are on marketplaces all the time but in this industry sometimes people are like no i i don't want to i don't want to i just want to call people or i just want to send an email so i think we're definitely at an inflection point especially with things like ai and 3d metal printing that i just think People are starting to see you either got to get on the train or you will be very, very left behind. Yeah. And I Ð antiquated is Ð I wouldn't use that word because a lot of these Ð traditional manufacturing processes are just mature, right? Which is exactly where we want them to be. They've hit maturity. They've hit mass adoption. They've gotten to the point where they are reliable. And the materials feeding them become commodities. And so like your marketplace, you can shop for materials that can be applied to, you know, machines from, you know, a variety of different brands. And metal 3D printing is not there yet. The powders that we talked about using are still very specialized powders that help to improve the performance of the machine. But we're going to get to the point where metal powders are going to be traded like... The raw commodity stuff. That's right, like the other steels. Yeah. And when we get to that point, it's going to represent maturity and mass adoption. And so we need to be talking about it now. Yeah. Absolutely, which is exactly why we had you on this podcast, which I mean, I've really learned a lot today. It's been really interesting. Even your background with Power Tools, Formula One, you seem to just have a very lengthy career. And I'm excited to see where your company goes moving forward. And I mean, I've learned a ton about 3D metal printing that I genuinely never knew before. So I'm hoping a lot of our listeners got to learn that as well. Well, it's a really exciting time for advanced manufacturing in general, especially in combination with AI. But it's going to be a very exciting next few years. So, yeah, I can't wait to see where this goes also. And we're going to be driving the next generation in 3D printing technology. Absolutely. And for any of our listeners, where can they find more information about you, your company, if they're interested in potentially learning more and working with you? Yeah, so as I mentioned earlier, Desktop Metal was recently purchased by a company called Arc Impact Corporation. Okay. So you can find us online now by searching either one of those, either desktopmetal.com or Arc Impact Corporation, and you'll find information on our printers, on our raw materials, our capabilities, as well as case studies on all of our applications. And we're always looking for new ways that we can help. solve our customers problems um and we'll work with you from the very beginning to the very end so um you know look us up and reach out Yeah. And I think I found you on LinkedIn as well, which I know, I mean, at least in B2B, LinkedIn is a great way to connect and just like learn more about people's backgrounds. And so I think you're also available on LinkedIn. So that's always a good place to connect with people. Yep, absolutely. Yeah. Well, thank you so much for joining us today. I learned a ton and I'm excited to see what you do in the future. It's my pleasure. Thank you.