Production? Stacked parts? Supports? Costs? Speed? Patents? Distribution? Materials? Competition? Ever since Desktop Metal officially presented its MIM powder-based metal AM system, we have been wondering if this technology will truly be able to compete with traditional manufacturing on large batch part production. We sat down and spoke about this with Desktop Metal CTO Jonah Myerberg, also a Co-founder of the company, and we got a lot of answers to these and other big questions.
The first thing that comes to mind watching the slides from the official Desktop Metal presentation (presumably similar to the one that got the company investments from the likes of Google, BMW, GE and Stratasys) is that in Desktop Metal’s automated factory of tomorrow there are four furnaces to each 3D printer. This is an interesting change with respect to metal powder bed fusion automated factories, where there are several 3D printers to each finishing station and it gives a clear indication as to the speed-to-firing ratio of the Desktop Metal Production system. This double pass binder jetting process is up to 100 times faster than powder bed fusion processes (clearly without including the firing process).
“Our Single Pass Jetting process delivers speeds up to 8200 cm3/hr, With no tooling needed, it’s the fastest way to manufacture complex metal parts,” Mr. Myerberg says, “combined with low-cost MIM powder, high throughput, and simple post-processing deliver per-part costs that are competitive with traditional manufacturing processes—and up to 20x lower than today’s metal 3D printing systems.”
Materials are the next subject we discuss. The list of materials that can be used on the Desktop Metal production systems is huge because the system is able to process any powder used in traditional MIM processes. It already includes several core alloys commonly used in AM such as 316L and 17-4 PH Stainless steel, as well as H13 Tool Steel, Inconel 625 (a nickel-based superalloy) 4140 Copper, C11000 Chrome and Kovar F-15. Development alloys include several types of steel, as well as carbide, magnetics and titanium, even a new Catamold material in partnership with BASF.
Mr. Myerberg then goes on to discuss the state of Desktop Metal’s current business development. “We have launched in North America right — he says. “The next four years is going to be spent installing our first printers close by, here in the United States. We’ve entered into sales networks that are utilized by other 3D printers and CAD software providers. So, for instance, Stratasys is our partner and early investors and for that we also share a sales channel with them. That is the same sales channel that SolidWorks uses. And so, we end up reaching customers who know 3D printing and know it very well. Our sales channel knows how to support and maintain 3D printers very well. And so, it’s a very, very strong channel here in the United States. We’re going to take that channel into Europe next year and then into Asia and Japan after that.”
One of the biggest upcoming challenges in establishing Desktop Metal’s market presence may be HP, as the digital printing giant has been growing its network on polymer powder bed fusion and is now eyeing the metal AM market. Myerberg is not intimidated “We’re not convinced that HP is going to be addressing the same mass production market we are targeting” he explains. “Just by looking at what they’ve done for the polymer printers you can see that while they say that they’ve addressed mass production of plastic parts, at this time they have definitely taken a step in the right direction but by no means are they as fast or as efficient as injection molding for most types of components. If they take a similar approach with metal then they’ll be addressing the prototyping and low-volume production markets. I think there’s no indication that they’re going after the high-volume mass production market that we are targeting with our production systems. Either way, the more the merrier: there is plenty of space in the AM market for everyone to grow.”
Another possible area of growth is ceramics. Companies like XJet began by focusing on metal and are now moving toward technical ceramics. The possibilities for AM expansion in the ceramics components market are significant. “Ceramics are possible but we’ve decided to focus on metal initially. There’s no reason why you couldn’t do ceramics on our production systems because ceramics uses a very similar technology. It’s ceramic injection molding. Our CIM is very similar to metal injection molding”.
Process and workflow automation — including post-processing and firing — represents yet another key advantage in the Desktop Metal’s strategy, implementing cloud-based manufacturing execution systems. “Each machine knows what’s going on with the other machine and the furnace knows what’s going on in the printer. The material cartridge is really easy to just pop in and out. “As we move from prototyping into low volume manufacturing and certainly into high volume manufacturing the value and advantages that automation can offer become very significant,” Myerberg explains. “We all want to get to the point where we have lights-out manufacturing. Just run the machines as needed and have low maintenance requirements. Right now, we are working with our first clients and partners on the production system. We want to fully understand how their factories work. Because as you get into high-volume manufacturing every customer approaches manufacturing differently. We’re asking them what kind of automation requirements they prefer, whether they would like the powder to be fed in through pipes and the into each machine or through kegs that we lift up and introduce into the machine. We are looking at what types of robots they want to utilize. Our systems right now are being designed in a way where customers can inject their own “DNA” into the automation process.”
The Desktop Metal Production system has a 350 x 350 x 350 mm build volume, which enables the production of relatively large nested parts. The system is fast enough to enable cost-effective mass production of relatively large size, stacked parts. “We realize that customers will want to be able to print larger and larger,” Myerberg admits. “Eventually, we want to be able to print an entire car chassis. And in order to do that we need to evolve both the machines and the processes. Going bigger and bigger can have an exponential impact on cost and complexity but that is certainly something that we are looking at. Ultimately AM will be used to make giant parts, even parts for trains and automobiles. At that point, you start to really open up adoption for assembly consolidation, where all these parts that maybe are pieced together with a hundred sheet metal parts can be reduced to one or two printed parts.”
Desktop Metal has made no secret that one of its primary targets is the automotive industry but several other industrial segments could benefit from this production-ready technology. “There are a number of verticals that we are pursuing,” Myerberg continues. “In general we can look at any engineering company that wants to rapidly design, redesign and then massively customize and optimize their functional components. Examples include what Caterpillar is doing with on-demand printing of spare parts. or what BMW is doing with design optimization. Other areas include small tools and a wide range of medical applications and even any metal component that you might find in mobile phones. These are all potential areas where metal components are being designed and produced at very very high volumes, in the order of hundreds of thousands or even millions of components. And these companies want to rapidly evolve them. They don’t necessarily want to offer the same tool for the same phone over the course of the year. They want to be able to modify and change as they discover problems without having to reinvest in tooling. So, this is a great opportunity for them and any engineering and manufacturing firm to use additive manufacturing to both cut costs and improve time efficiency.”
One delicate issue that has come up recently concerns ongoing litigation between MarkForged and Desktop Metal relative to both companies’ studio-based metal 3D printing process. Since both companies are located near Boston — along with Formlabs and others — the area has really become the center of 3D printing in the US.
“MarkForged has been around for a while and we’re all friends, we’re all in the same boat,” says Myerberg. “We are putting our own spin on the way that metal 3D printing should you know be done. And it’s really going to come down to the details. There are certain aspects of metal 3D printing that you cannot protect because they’re open and available and there are some aspects that you can predict. The Devil is in the proverbial details here. It’s about how do you really go about doing this in the right way? Where you get the right geometry and the right metallurgy for the finished part? And at Desktop Metal we focus very strongly on the metallurgy and the geometry of the part. We hired a number of metallurgists that come from the labs of our founders at MIT, who run the material science department there, and we’re very heavily invested in getting the metallurgy just right.”
“Right from the start — Myerberg continues — we’ve also invested in ways that you can create the geometry and maintain it through the sintering process like this interface layer, and the rafts that we put under our parts: these are the little details that it takes to really create. You know, complex geometry, not just two and a half D — parts that are very easy to build and stay together at first — but real 3D printed parts that have features and complexity that can’t be manufactured any other way. And then putting it all together within a system that starts with the feedstock that we control from the very, very beginning. We have one of the guys who invented MIM process, and he’s been working in MIM for 40 years. He’s focused on getting the recipe for the feedstock right from the very beginning. And that then goes into a printer through the Desktop Metal environment software that we’ve designed from the ground up through all of our software team, that comes from the CAD industry from SolidWorks, from a Petroleum engineer. And then connecting that with the binders and furnaces that all live on the Desktop Metal Network, all designed from the ground up here as a sub-metal to work with each other and with our process. And that’s that system right there. We feel it’s going to be a very very critical component of doing this. Doing this the right way and giving the users the right experience they need.”