Metal 3D Printing Brings Big Improvements in Speed, Cost, and Quality
By Mark Shortt
If you feel the power of Desktop Metal’s claim that it’s reinventing how metal parts are manufactured, you’re not alone. Some of the biggest names in manufacturing and venture capital—BMW Group, GV (formerly Google Ventures), GE, Stratasys, and New Enterprise Associates—have bought into the company’s vision of making metal 3D printing more affordable, more accessible, and much, much faster. That’s largely because Desktop Metal has developed an innovative metal 3D printer, the Production System, which it calls the “fastest 3D printing system for mass production of high-resolution metal parts.”
Based on a new approach to metal 3D printing called Single Pass Jetting (SPJ), the DM Production System™ is said to build metal parts in a matter of minutes, instead of hours. It operates at speeds of up to 8,200 cubic centimeters per hour, which the company says is 100 times faster than laser-based systems. The technology also doesn’t require tooling, further reducing the time needed to make complex metal parts.
Desktop Metal (www.DesktopMetal.com) has garnered more than $200 million in funding from investors enamored with the prospect of mass producing strong, functional, and high-quality metal parts at unprecedented speeds. The Burlington, Massachusetts-based company is on a mission to make metal 3D printing more affordable and accessible, from prototyping through mass production, and has launched two systems—the DM Studio System™ and DM Production System™—to make those goals a reality.
The Studio System, designed to bring metal 3D printing to an engineer’s desk or the shop floor, is a complete platform that includes a printer, a debinder, and a sintering furnace. Desktop Metal began shipping the Studio System to early customers—including Google’s Advanced Technology and Products (ATAP) group—in December as part of its Pioneers Program roll-out. The Production System, designed for mass production of metal 3D printed parts, is scheduled to begin shipping in early 2019.
Other Pioneer customers are the U.S. Navy’s Naval Surface Warfare Center Dahlgren Division; Built-Rite Tool & Die; The Technology House; Medtronic; and Lumenium LLC.
Opening Doors to Opportunities
“This marks the first time our team will be able to use metal 3D printing for rapid prototyping of our hardware parts,” said David Beardsley, manager of Google’s ATAP, in a release announcing the rollout. “For prototyping, we have previously relied upon casting or using plastic 3D printing. Now with the Studio System, our team will experience shorter lead times, faster product development cycles and the benefits of functional prototypes in an array of metals on demand and in the lab. We look forward to exploring and developing potential applications for many of our projects.”
Jonah Myerberg, a co-founder of Desktop Metal and the company’s chief technology officer, told D2P that the company has essentially taken a technology that numerous industries have been watching very closely—metal 3D printing—and brought it closer to their engineers and manufacturers than it’s ever been before.
“We’ve started opening doors and opportunities for them,” said Myerberg in a phone interview. “We’re launching two different machines, two different systems, and what those two systems enable is going to be almost kind of unpredictable going forward. Every peripheral industry around manufacturing is going to be affected if and when this technology takes off.”
For engineers, accessibility means a lot of different things. With the Studio System, accessibility means “having the engineers’ connection to rapidly prototyping and evolving real parts,” Myerberg said. Rather than having a “looks like, feels like, fits like” model of a part that hasn’t yet gone into production, the engineer is able to produce a part that is usable, functional, and representative of the final materials and properties.
“That will immediately give the engineer something to test, to really stress and exercise, and then to evolve the design faster. Because let’s face it, the design process is an iterative process of failure. You design something, and it’s not perfect. It either doesn’t work, or you want to add something new, and then you have to redesign and redesign. So the faster you can do that, the more successful an engineer becomes as he or she develops the product.”
Metal 3D printing has actually been around for about 20 years, Myerberg said, and has been used for rapid prototyping of metal parts. But it hasn’t been easily accessible to engineers for various reasons, the first being that it’s always been expensive. The process requires big machines, complicated systems, and powdered metals that need to be controlled. “Essentially, these are factory level machines that cost a million dollars, plus whatever the facility’s modifications are, plus dedicated operators and expenses to run them,” Myerberg said. “They don’t sit in your office the way that a polymers printer does.”
Companies that could afford the machines have scaled factories by adding more and more machines to increase capacity. They might have hundreds, or even more than a thousand, million-dollar machines in a factory producing aerospace parts.
“The first step for us is to bring accessibility to the engineer,” Myerberg said. “Let him design parts, work in his office space, and let’s remove the barrier to entry for him to actually start playing around with metals. The second is making the metals more relevant to what you would see in production.”
That second step is important when an engineer wants to take a part into mass production. Although laser-based metal additive manufacturing technologies are capable of producing real metal parts out of steel, titanium, and aluminum, the microstructures of those parts are way different from those made by traditional production processes like casting, machining, or stamping. That’s because as the metal powder melts, it goes from a solid to a liquid, and then solidifies back into a solid.
“The microstructure of those parts is unlike anything that these engineers have ever seen before,” said Myerberg. “That’s why it’s taken companies like GE 10 years to qualify them for their applications. Because when you cross section and you look at the microstructure of these parts, you see a structure that’s not really made any other way.”
Because no other part in mass production looks like that, the only way to take that part into mass production is to scale the process. That means buying more machines to do more of the same laser welding techniques that produce the same properties in a part. It’s a route that most companies simply cannot afford.
A Less Expensive Way to Mass Produce Metal Parts
To avoid this problem, Desktop Metal designed its systems to use metal injection molding (MIM) materials with properties that are familiar to engineers. As the parts come off the engineers’ desks and are tested, their properties are well understood, making it much easier and less expensive to take those parts into mass production. Options include metal injection molding and hot isostatic pressing (HIP). But if you want to take a geometry that’s been designed for additive manufacturing into mass production, that’s where the Desktop Metal Production System comes in, Myerberg said.
“Now the engineer has been able to prototype, to design with the Studio System, and now the manufacturer can catch that design and put it into the same type of qualification plan, with the same type of materials and the same properties, on the Production System, and get the same properties out of the final part.”
Touted as the first metal 3D printing system for mass production, Desktop Metal’s Production System is said to provide the manufacturing speed, quality, and cost-per-part needed to compete with traditional manufacturing processes, many of which require tooling. Its combination of low-cost MIM powder, high throughput, and simple post-processing are said to result in per-part costs that are not only competitive with traditional manufacturing processes, but as much as 20 times lower than competing metal 3D printing systems.
—Jonah Myerberg, co-founder and CTO, Desktop Metal
Myerberg said that the Production System is also accessible because of its speed, and speed equals cost.
“If you’ve got a machine that builds at a couple of cubic inches per hour, like a laser based system does, then you need more machines to build those parts,” he explained. “But the way that Desktop Metal’s Production System is set up is like a printing press, like a 2D printing press, where it can produce 8,000 cubic centimeters per hour. Speed translates directly into cost now because you’re competing with traditional manufacturing processes like stamping and casting, where you would normally have to tool up a die, dedicate that die to a single geometry, a single part, and then move forward. The printer can now accept in any geometry; it doesn’t have to be a castable geometry or a stampable geometry, and then produce that without any tooling.
“The real key to the accessibility of the Production System is the fact that it competes at the cost target, and usually the cost target level of all other traditional manufacturing, and it produces parts that are well understood in their properties.”
The Production machine is by no means an inexpensive machine. Myerberg said that it costs 10 times more than the Studio machine, which lists at $49,900 for the printer, and $120,000 for the complete system, including debinder and furnace. But the Production System is 100 times faster than other machines, he added.
“The speed allows you to take that overhead—that capital cost—off the table. That becomes a fraction of the overall cost of the parts, and then the bulk of the part cost is the material. This is designed around using materials that are extremely low cost. There’s already a market for the materials that we’re using in powdered metal and metal injection molding, in which millions of tons of material are made each year. We’re tapping into that supply chain and getting low cost materials that we can use as is.
—Jonah Myerberg, co-founder and CTO, Desktop Metal
“We’ve introduced a new process to make final metal parts that have known properties of known metals,” he continued. “Everybody in the world knows 17-4 stainless steel, or 316 stainless steel, or 4140 Chromalloy. We’re using chemistries that are well known by engineers. They choose their material, they use this brand new printing process to create their parts, but at the end of the day, they end up with the properties of these known materials. It gives them something to directly measure against and validate. They know what they should be getting out of this new process because they’re using a known material. So we’re not trying to introduce multiple, serial inventions at the same time. We’ve introduced a new process using a known material, so the customer is able to quickly validate that.”
Myerberg said that customers will come to Desktop Metal with parts to print that really weren’t designed for additive manufacturing, but were instead designed to be stamped or cast, or somehow manufactured a different way.
“They ask us ‘Can you print this?’ and our answer to them is, of course, ‘Yes, sure, we can print these as a starting point, but for you to really take full advantage of everything that additive manufacturing has to offer, you’re going to want to take another hard look at your designs and start redesigning your parts, and redesigning parts to be more optimized for mass, optimized for printing, optimized for anything. And what the customer says is ‘Yeah, yeah, yeah, we’ll get to that,’ but it’s a lot of work to redesign, and especially a lot of work to go back and reteach their engineers the rules of additive manufacturing and, essentially, how to model parts that will print well.
“So we kind of feel a responsibility to our customer to help them in that process also. The additive manufacturing process is a full loop, from concept all the way to finished part. And we are responsible to participate in that and make it as painless for the customer as possible. It’s one thing to tell the customer, ‘Hey, design this for additive manufacturing,’ but it’s another for them to go back and really do it and understand it.
“And I think that will come in time. Universities are already starting to teach this, but when I went to school, I certainly didn’t learn Design for Additive Manufacturing. Design for Additive is new, and most universities still do not teach Design for Additive. So we’re kind of the bridge between the physical world and the digital world—the digital world of CAD and the physical world of parts. We’re connecting the CAD, through a 3D printer, into real parts,” Myerberg said.
Engineers can now take CAD from SolidWorks and grow it into parts that are both functional and printable, using the Live Parts software. Customers are seeing this, he said, and saying “Wow, I can quickly generate complicated geometry that is functional and printable.”
Myerberg believes that the 3D printing industry, or community, needs to continue to show technologies like what Desktop Metal is developing.
“We need to show that these are applicable to mass production, not just prototyping anymore, not just aerospace or medical, but that it’s achievable and it’s reliable enough to apply to any metals application,” he said. “That takes some time to evolve and to adapt, but it’s going to be a lot faster than what happened in aerospace, where it took 10 years to qualify for the first part to be flown.
“I think next year, there will be more parts in consumer electronics; there will be more parts in automotive—all of these industries that have been watching medical, watching aerospace, and wishing that they had accessibility to metal 3D printing. As we bring the cost down, and we bring the speed up and we bring the complexity down, and we start to bring metal 3D printing into the office, companies are going to start to use it more, embrace it more.
“This next generation of engineers—the first thing that they think of is printing—like, ‘OK, I’m going to design a part, and I’m going to print it.’ We want to encourage that. We’re going to encourage that in the right way, teach that in the right way. And as that starts to happen, we’re going to start to stress the world’s supply of powder, and so we need our powder manufacturers to get on board with expansion to their capacity, and they all are ready. All of them are really excited about this. None of them are nervous, and saying ‘Oh, gosh, this is going to drown me.’ No, they’re all very excited about adding more atomizers, becoming higher quality producers, competing with each other.
“We’re a small enough business right now, and the opportunity is so large. I say ‘small enough business’ as a whole: Metal 3D printing is very relatively small compared to the mass production market of metal parts in the world. So the demand is huge. There’s opportunity for everyone, and this common, rising tide is going to float all of our boats.”
The Production System was created by the inventors of ground-breaking technologies in 3D printing and 2D printing—binder jetting (by Ely Sachs) and single pass inkjet (by Paul Hoisington). Desktop Metal’s co-founders include Ric Fulop, CEO; Jonah Myerberg, chief technology officer; Rick Chin, vice president, software development; and Ely Sachs, who, along with colleagues at MIT, invented the Binder Jetting process in 1988. Also among them are Christopher A. Schuh and A. John Hart, both of whom have Ph.D.’s, and Yet-Ming Chiang, Sc.D.
Intelligent Layering Process Yields Low-Cost Metal Additive Parts
In the world of metal additive manufacturing, cost is, by far, the number one barrier to entry. But the cost of the machine is just the beginning of a long, costly journey to being able to produce parts in volume, to the specifications required. Printing a prototype part is one thing, but to be able to move into production requires a whole ecosystem of expertise that companies need to have.
“It is a significant investment, and the million dollar machine is just the start,” said Matt Sand, president of 3DEO (www.3deo.co), a technology and manufacturing company in Gardena, California. “And for binder jetting systems, the maintenance alone on the inkjet heads, the spray heads, is upwards of $75,000 per year.”
Sand’s technical partners, Matt Petros and Payman Torabi, were intent on finding a way around this formidable cost barrier. They zeroed in on inkjet, a technology originally designed for 2D printing, rather than 3D binder jetting. The question on their minds was “How can we invent a system that takes a lot of the cost out of it?”
For Petros and Torabi, the answer was a design that doesn’t use an inkjet to spray the binder. Instead, it uses a low cost spray head that lacks the complexity of an inkjet and leverages established technologies like CNC milling.
“We have a very low cost spray head, and we use that to bind the entire layer. Then we come back and we CNC the layer with an end mill—with a micro end mill that’s as small as 125 microns,” said Sand. “It’s the smallest drill bit you’ve ever seen. And so it’s really interesting: Old school manufacturing, CNC machining meets the new school of additive. We’re building the parts layer by layer, so we get all the advantages and complexities of additive manufacturing, while at the same time, drastically reducing the machine costs.”
3DEO’s technology, Intelligent Layering®, is said to unlock the potential of additive manufacturing by reducing final part cost by as much as 80 percent while meeting the MPIF Standard 35. Sand cites three main contributors to the low part cost—extremely low machine cost, the use of commodity materials, and creative software design.
“It started with kind of a crazy idea; it was iterated through so many different technologies and approaches, to finally get to where we are today. I think it’s an elegantly simple solution, but it’s simplicity on the far side of complexity. It took a lot of R&D by the guys on the technical side to get to where we are.”
Sand emphasized that 3DEO is both a technology company and a manufacturing company.
“We view ourselves first and foremost as a technology company. A lot of manufacturing companies—and there’s nothing wrong with this at all—will just buy off-the-shelf machines to be able to produce. We’re a technology company because we’ve actually invented this new, kind of breakthrough, low cost technology that we use for ourselves.”
According to 3DEO, the Intelligent Layering process begins by spreading a thin layer of metal powder over the build area. A binder is then applied to the entire layer being built. A cutter then shapes the perimeter of the part, layer by layer, before the next layer of powder is spread. Next, the completed part is put into a high-throughput furnace for sintering. Finally, a finishing process may be applied, depending on the application.
“The way we think about our process is ‘MIM (metal injection molding) without the molds,’” said Lance Kallman, 3DEO’s vice president of business development, in an interview at Singularity University’s Exponential Manufacturing Summit in Boston. “The founders actually created the technology with low cost in mind because there’s obviously a very high end market for metal additive right now. So they created a process that ties to the metal injection molding standards MPIF 35 (Metal Powder Industries Federation’s Materials Standards for Metal Injection Molded Parts).
“We’re using MIM powders, so once the layering is complete, the part goes into a MIM furnace. At the end, when our parts come out, they’re just like MIM parts, but we created them without all the tooling costs required to create a metal injection mold. The CNC machine that we use is very low power, which means very low cost because all it’s doing is cutting powder and glue; it’s not cutting a finished molded part.”
Building a Metal Additive Manufacturing Supply Chain in House
Sintavia, a metal additive manufacturing supplier in Davie, Florida, has centered its business model on growing an ecosystem of additive manufacturing expertise inside the walls of its vertically integrated manufacturing facility. The company is unusual in that sense. Rather than outsourcing or subcontracting work to additive manufacturing suppliers, Sintavia has taken pains to develop and nurture what amounts to a complete 3D metal manufacturing supply chain—mainly for the aerospace and defense industry— within its own facility.
“As a fully integrated supply chain, Sintavia is poised to meet all the quality, validation, and post processing needs, all in one facility,” said Sintavia President and Chief Technical Officer Doug Hedges, in an interview. “What Sintavia has done is blaze a trail into an unknown territory in 3D printing. The reason we are able to undertake this process is because of the resources we have in house to test and control production from beginning to end.”
Even though Sintavia can be characterized as an additive manufacturer, Hedges said, that only begins to tell the story of the company’s depth of expertise around that designation. Quality is a prominent issue in aerospace manufacturing, as it is for all critical industries, and Sintavia prides itself on quality control.
“It takes a wide variety of skills to make this all work properly,” said Hedges. “We have expertise in designing for additive manufacturing, so the engineers know how to design these parts and help the customers get to their end product. We have the expertise in post-processing—that’s hot isostatic pressing (HIP), heat treatment, CNC machining, et cetera. We have the expertise in metallurgy, and it’s complemented by our mechanical testing. So that’s really what Sintavia’s about, and when I speak of that, it really is all about quality control. All the resources that we have to do this job ultimately come down to quality control.”
For more on Sintavia, see The Metal Additive Manufacturing Supply Chain Is in the House.
A New Era of 3D Printing
Adaptive Corporation, Inc. strives to enable innovation by applying technology to streamline business processes, reduce costs, and improve efficiencies throughout the product development lifecycle. Adaptive is a reseller of Markforged 3D Printers, like the Onyx Series and Metal X, which are used to make carbon fiber composite and metal printed parts, respectively.
Frank Thomas, a metrology and additive manufacturing specialist for Adaptive Corporation, has worked with a variety of manufacturing companies in the areas of engineering, metrology, and additive manufacturing, as both an implementation consultant and product specialist. Over the past 10 years, he has focused on connecting engineering and manufacturing, specifically around quality, and now additive manufacturing. His goal is to help companies better connect the “virtual” to the “physical,” thereby improving their time to market and reducing cost.
Thomas said that until fairly recently, additive manufacturing was used most often as a tool to create parts that you could hand to somebody so that they could see it, touch it, and provide some input as to what might need to be changed or modified. But that’s changed in recent years as new materials have been developed that enable printers to make stronger, more durable parts.
“Metal printing has always been there, but that has an economic value proposition that’s a bit challenging for it. The ABS and nylon and other plastic 3D printers, up until the last couple of years, weren’t necessarily dimensionally accurate, and then they had challenges creating a part that’s functional. That’s what I think is different about the market today, compared to just, really, a couple of years ago.”
Thomas said that Adaptive markets 3D printers that feature dimensional accuracy and the ability to yield a part that is functional, depending on the application. He’s also seeing a lot of interest in metal 3D printing.
“Where metal 3D printing comes from is the argon laser based systems. The companies that have had applications or use cases for them have made the investments, and they’ve been huge investments. They probably start at half a million dollars and go up, and that doesn’t even count the facility that’s required to be able to certify and implement something that’s an argon laser based system. It’s very, very costly.”
Thomas said that Adaptive is in the process of bringing to market an additive metal machine that’s very different from the argon laser based systems. It’s based around the metal injection molding (MIM) process, so it doesn’t employ any laser technology. He sees it as a totally new kind of metal 3D printer.
“It does not employ argon gas compressed in a 3D printing environment,” he said. “We’re basically using a filament material that is a powdered metal with two binders. We have the ability with software to scale the part, so that after we’ve heat treated the part, the net shape of the part is the nominal design shape. We’re able to predict all of that throughout the process. It’s very exciting, and we expect it to be revolutionary.”
The ability of the machine to print metal powder that’s bound in a plastic matrix eliminates the safety risks associated with traditional metal 3D printing machines, which incorporate a high power argon laser that emits argon gas.
“If you’ve seen traditional 3D metal printers, typically, the print volume is relatively small, but if you look in the front window of the printer, it looks like a bomb chamber. I mean, it’s so well insulated and it’s so well protected. And why is that? Well, you’re working with a laser; it’s emitting argon gas, which needs to be controlled. If there’s a spark or some kind of problem with that environment, the possibility of fire or explosion is very, very high.
“That doesn’t exist in our environment. We’re not using a laser, and we’re not emitting argon gas as part of our 3D printing process. The other side of that includes all of the environmental things that are required in using the traditional 3D printing machines with the high powered laser. If you go into those environments, you’ll see the workers that are usually dressed in spaceman’s outfits, you know, to protect themselves, even in the environment when the machine has been shut down for some period of time. So it’s a very complicated solution.”
If the demand for 3D printed metal parts is going to grow significantly, especially for critical use cases, OEMs will have to be able to count on high-quality parts. Thomas believes the additive metal industry is up to the challenge because he’s already seen major improvements in quality in recent years.
“At the end of the day, this is really a materials game. If the materials that we’re able to bring to the market provide the end use quality that people are looking for, that’s critical.”
Thomas sells both desktop and industrial versions of Markforged 3D printers. One of the advantages of the desktop versions, he said, is the ability to actually walk through the door of a building with a 3D printer to show engineers. The opportunity to have an audience with a client helps remove many of the preconceived notions that people might have about 3D printing, especially for people familiar with older versions.
“We try to get in front of people and try and show them the actual solution and describe to them the process,” he said. “The desktop printers are really very portable, with the ability to put that into a car and drive out to somebody’s site, and, within 10, 15 minutes, you’ve got a 3D printer set up, running, and3D printing a part. It’s very visual for people, and it helps demystify the process.”
In earlier days, 3D printers were much bigger, more expensive, and way more complicated. Users printed onto a heated plate, Thomas said, and some of them had wires going into the heated plate. The overall complexity of the machines, to some people, was a bit scary, he said.
“Now, when you put one of our solutions on a desktop, in front of an engineer, or a manufacturing engineer, and you run the thing for them, there’s nothing scary about it. There’s nothing intimidating about it, and they just gravitate to it very well.”
Thomas emphasized that it’s important to educate clients and expose them to the possibilities that additive manufacturing creates. Simplicity, and the amount of time it takes to realize value, are key.
“The solutions that people buy in this economy need to deliver value measured in weeks and a month or two—companies don’t have the luxury of implementing solutions that are measured in months and years. Additive manufacturing is something that you can deliver to a client, and they can be using it tomorrow and getting some value off of it and generating valuable results for the organization.”
Desktop Metal Production System Is Built for Speed
Single Pass Jetting is reported to work 100 times faster than laser-based systems
Desktop Metal touts its Production System as being the first metal 3D printer for mass production. Powered by a technology called Single Pass Jetting, the Production System is reported to work up to 100 times faster than laser based additive manufacturing systems.
“Metal 3D printing could change much of the world around us if it was fast enough and cheap enough for mass production,” Desktop Metal says in a video on its website that outlines how the Production System works. “To date, metal 3D printing has been too expensive and too slow to change the world around us. At up to 100 times faster than existing technologies, the Production System unlocks the cost per part needed for mass production. For the first time, it’s possible to go to market with metal 3D printing.”
Here’s how it works, as told in the video:
The system combines two powder spreaders and one print unit into a single pass system to both spread metal powder and print. Unlike existing 3D printing, there is no wasted motion with Single Pass Jetting: A single pass starts in the powder spreader, where a metering system deposits metal powder, and a compacting system forms a layer as thin as a human hair. The print bar follows, jetting droplets of a binding agent. Millions are jetted per second, binding metal powder to form high resolution layers.
Anti-sintering agents are then deposited, making it possible for supports to fall off after sintering, saving hours of post processing. Once the layer is dried, the process repeats itself.
“The system combines all the necessary steps for printing into a single pass, so whenever there is movement, there is printing. This makes it possible to print parts in minutes instead of hours.”
Single pass jetting is bi-directional. The system combines all the necessary steps for printing into a single pass, so whenever there is movement, there is printing. This makes it possible to print parts in minutes instead of hours, according to Desktop Metal.
Once printed, the brown parts are densified in a micro-wave enhanced furnace that combines conventional heating with microwaves to speed up sintering. A closed loop thermal control system regulates temperatures in real time, as parts are heated to just below their melting point. Binder is removed, and metal particles are fused to form a dense solid.
The Production System is cloud connected. Sophisticated software manages the entire workflow, with profiles that are tuned to every build and material, from the printer to the furnace, delivering dense metal parts.
The result is sheer throughput. In the time that it takes laser based processes to produce just 12 impellers, Desktop Metal’s Single Pass Jetting technology would have produced more than 500, the company says.
Source: Desktop Metal (www.desktopmetal.com/products/production)