Fresh Ideas around Additive Manufacturing with James DeMuth
Carpenter Additive's Vice President, Ben Ferrar, is joined by James DeMuth, Co-Founder & CEO at Seurat Technologies. Seurat's area printing process focuses 2 million points of laser light on a bed of metal powder, each point fully controllable to create fully melted net-shaped metal components. James describes how he and the company saw an opening in the industry for a new way of thinking and jumped into it.
You can read the transcript or listen to the full episode below.
Hello and welcome back to another episode of PowderHeads, a Carpenter Additive podcast. With each episode of PowderHeads, we bring you the minds of industry experts and delve into the topics that are defining how additive manufacturing is making an impact on our world. Today, we set our eyes and ears on Massachusetts-based and creatively named Seurat Technologies. Seurat's CEO, James Demuth sits down with Ben Ferrar vice president of Carpenter Additive Seurat's area printing process focuses 2 million points of laser light on a bed of metal powder, each point fully controllable to create fully melted net-shaped metal components like so many in the AM industry. James describes how he and the company saw an opening in the industry for a new way of thinking and jumped into it. It's a super interesting discussion on fresh ideas around manufacturing. Thanks for listening and enjoy the conversation.
Ben Ferrar (01:09):
So today on, uh, on PowderHeads, we have James Demuth from Seurat technologies. James, thanks very much for making the time to, to come and speak to us today on Powder Heads. I guess we should, we should kick off with you telling us a little bit about your story and, and giving us a bit of an intro.
James Demuth (01:28):
Sure. Terrific. Ben, thanks for having me. It's uh, terrific to be here. Uh, yeah, so, you know, uh, Seurat technologies, um, named after the French pointless painter, George Seurat, um, you know, similar to, uh, how he really changed the, that people perceived images and paintings. Um, he built up his images by the careful placement of millions of points of paint, which we perceive of as light. We thought there was some interesting similarity because we take what previously was a, uh, you know, a single point process of laser powder bed fusion. And we've essentially turned into a massively parallel, uh, uh, process where we're basically printing an image at a time. Um, and we're doing this as a way to massively scale up throughput rates, uh, control, quality, um, and lower the prices all at the same time. So, you know, uh, it's a little bit of a history, you know, what's in the name, right?
James Demuth (02:22):
Um, so, uh, you know, this was originally conceived at Lawrence Livermore, national laboratory AK lasers, lasers, nothing but lasers and, uh, a lot of experience manipulating, shaping, blending, you know, patterning laser light to do what you need to do. And originally this was developed as a solution for a problem that existed in fusion power, uh, to build a fusion power plant. And we just needed advanced designs and, uh, production rates that, you know, were massively higher than what exists today in laser powder bed. And this was a solution we came up with and decided to spin it out.
Ben Ferrar (03:01):
And, and so what was that key moment where when you realized actually, look, we can apply this technology in this area and it'll have such a significant impact..
James Demuth (03:12):
Yeah. You know, it was, uh, well, the, the key moment was the frustration started, right? Like, you know, oh my God, it's gonna take 200 years to just make one of these fusion chambers. We're looking to have fusion, plants deployed all around the world. And, uh, yeah, that's, that's not gonna work. So something needs to change in order to make this happen. And then it's like, all right, well, how do we do that? Well, what's a really cheap source of energy diaod lasers, right? That's a really refined yet also cost effective. It's semiconductor based. So very, very scalable. Um, and you know, a lot of people were actually interested in doing diode lasers for additive back then. Um, and we saw a lot of companies try and fail to do it sort of in parallel with us. Um, you know, I think we, we just happened to be lucky that we were in the right place at the right time. We had, you know, we're working with a lot of very smart people and we're able to, you know, surmount those, those issues that present themselves and, you know, uh, on how you actually do the process. So, and
Ben Ferrar (04:12):
So can you ex I mean, just for, for the people listening, can you explain some of those challenges? Are they wavelength? Are they related to wavelength? Are they the absorptivity of the powders? What, what are the challenges that you had to overcome?
James Demuth (04:25):
Yeah, so we do things a bit differently. Um, laser powder, bed fusion, as we know it today, right? You scan around the laser, uh, you essentially make a weld pool and you drag it around, right. The process we're doing is more like an exposure. Think about like a, a camera exposing some film, right? So we are, uh, we apply a pattern to our laser, and then we project it down to the powder bed into a relatively large area, you know, um, our production machines looking at about 10 millimeters on the side with millions of pixels within it. So you have this really high resolution pattern. You're projecting it down to the powder bed and you're essentially flash exposing an area. And so you have, because you're doing everything in parallel, you know, all these pixels are happening in parallel. You can actually devote a lot more time to each one of them, um, using diode lasers alone. The problem that we had is as you illuminate it, diode lasers, just aren't intense enough to get the heat in fast enough to just get everything melted. They were melting the top layer of powder only, and not really getting the substrate below them hot really at all, relatively
Ben Ferrar (05:35):
Welding through to the layer below. Right?
James Demuth (05:37):
Yeah. So you'd end up getting the powder. It would all whip together. And then it would form Essentually a, a sheet, a barrier that would prevent you from getting any more light down to the substrate. And then any more heat you provided just would cause those powder particles to ball up and the base would actually cool further. And you'd sort of diverge from where you wanted to be. And so the insight we had was, oh gosh, we gotta deliver more energy, you know? Right. When you get, it's great to use the dilators of preheat, that's terrific. But like right, when you get close to melting, like hit it with a big burst of energy before the topology changes. Right. And so it's this, it's this really carefully tuned time dependent behavior that we found is actually very, very repeatable. And so it's, you know, you're, you're essentially doing exposure. We, we preheat the powder to its melting point with the diodes, and then we bring this big slug of energy in to sort of finish the process. And, and this is, you know, the, the shape, the pulse shape over time, overall, this very critical. Um, and so being able to manipulate that is, is crucial to getting the process to work. And so well, not everyone has those knobs that their disposal to be able to tune that in. And, uh, we did.
Ben Ferrar (06:50):
Oh, great. Yeah. And so is there, I mean, what's, what's one thing that you'd wish you'd known when you started?
James Demuth (06:59):
Well, we definitely, uh, I think we, we underestimated how many knobs that we really needed. We were doing something that's so very, very different. Um, we had some pulse control knobs when we started, but they weren't quite enough. Um, and had we had we done, you know, obviously when we started, if we had, um, if we had those knobs implemented from day one, uh, you know, you could argue, you know, we, we bought off the shelf equipment initially, if we had had to develop it ourselves, maybe we wouldn't have gotten anywhere. Right. Cause it's just too big of a hurdle. So, you know, it's hard to say it's really hard to say we had to go down the path that we went down, um, at some level and it, it got us to where we are today, but obviously knowing what we've done in the past, uh, you know, we have to spread powder differently than other people. Um, we've gotta illuminate the laser. We, you know, we illuminate different than others. The whole printing process is different and everything kind of folds in around that. You know, we can't just go get, uh, a used powder spreader from another manufacturer because we would need to frankly, just redesign it anyways. Um, so it's, it's been challenging.
Ben Ferrar (08:09):
So explain that then. So why, if fundamentally right, is still the same additive layer technology. Why is there a difference in the way you have to spread your powder?
James Demuth (08:20):
Yeah. Um, it's, it's all about how thick the powder layer is relative to the powder size distribution. And there's sort of a, a thickness you get, eh, two to three particles thick that gets you a nice spread. That looks beautiful. Right? Well, that turns out, you know, because we're not in the keyhole welding regime, we're not able to penetrate as deep. And so we actually need substantial amounts of shine through in order to really make the process shine, um, pun intended. So we, we end up having to be really close to the bitter edge on layer thickness, which is, is interesting. Um, it presents some, some interesting challenges for us now, not to say that's gonna be the case. Always. We see there's actually some unique opportunities to do some pulse shaping and potentially overcome that. But, you know, we've done all of our cost modeling and all of our scaling based off of how having thin layers and what's exciting is we're able to show how, even with that layer of thickness, we can crush it in terms of competing with machining casting and eventually one day even forging applications. So, uh, to the extent that we can go thicker, it's just gonna be gravy.
Ben Ferrar (09:29):
Yeah. Brilliant. So, uh, so give us, uh, give us an, uh, some insight into the vision then where, where do you see this technology could be in five years, 10 years, or do you believe that this, the value add that this technology and this process has over existing, uh, powder bed, fusion technologies?
James Demuth (09:48):
Yeah. So our big hairy audacious goal is by 2030. We are, we have a machine printing parts at around $25, a kilo, all in printed part costs that includes amortization machine service, post-processing feed stock, you name it. And you know, we're talking about an estimated market size when you get down to those, those price points at around 32 million metric tons a year. Um, that's a, a, a lot of metal. Um, and you know, we're talking about price points that are directly competitive with, you know, high volume casting, low volume forging, um, and obviously pretty much everything to do with machining, although machining's got precision aspects, but we see that we're gonna be able to address that in a, that's sort of like a, another vector that we can address as well, um, even today. So that's kinda like our, our goal for economics, you know, scalability wise, you know, we we're, we see this technology as the ability to scale to production rates that can meet and meet all the needs of conventional manufacturing today.
James Demuth (10:53):
And I mean, what do you need to break into the conventional manufacturing market? You need the,economics to be there, you need the production rates to be there and you need the quality to be there. And so we see that we have an opportunity to check all of those boxes, which says, Hey, our, our sites are squarely set on, you know, how do we bring the, uh, capabilities of additive, right? All these, you know, being able to consolidate multiple parts into one, make it true to the design that you want it to be, you know, driven by the needs of the product, not by the manufacturing process, right. Um, whether it's, you know, supply chain streamlining, reduce your inventories, produce parts on demand, like bring additive to conventional manufacturing. Um, that's our ultimate goal. And, you know, I think we're incredibly excited to see when you do that, you open up the ability to essentially just be massively creative. It's, it's gonna be a whole new world in terms of, um, manufacturing and what comes outta that
Ben Ferrar (11:45):
Well, and, and not just manufacturing, right. But the, the design side of things, people have to, we, we have to have a, almost a complete cultural change in the way we design and develop and the tools that we use today to be able to, to be able to do that. Right. And, um, I mean, that cultural challenge is, is clearly something that within, I feel, uh, you know, within the additive industry, right, we're, we're still a relatively small, uh, small group, right. By comparison to the rest of the, the world's manufacturing capability and design capability, what do you think are the key things that we have to do as an industry to, to really penetrate into those existing industries in those existing mindsets to, to change their, to change their perception of what's possible?
James Demuth (12:31):
Yeah. Well, you know, being able to do to design for additive, um, you know, whether or not it's required. Right. It's more, um, when I say design for additive, I don't mean like, um, you know, needing to learn to, to, to speak
Ben Ferrar (12:44):
At time, orientate your part, or,
James Demuth (12:47):
You know, it's, uh, it's more like, Hey, we want to, uh, leverage those benefits of that. It brings, right. You know, due topology optimization to, you know, make that part be lighter, stronger, better, um, for your application, it's a little bit of a chicken in the egg problem, right. And so the technology needs to be able to be there in order to justify using that, uh, those different applications for additive or for your place, conventional manufacturing, um, you know, we're looking at massive amounts of data. Um, I'm almost, you know, there's a huge barrier to overcome, which is gonna be, you know, if you have a build bed, that's multiple meters on a side and you've got six micron pixels on it, you've got an absurd, a amount of data per build. And, you know, we're, we're looking at high volume stuff, so we're repeating the same build over and over and over again, but just moving that amount of data around and parsing it, like we look at, you know, the, the different slicing programs that are out there today. And it can take a long time just to do simple builds. I mean, holy cow, when you're talking about stuff that's really big and really detailed, um, we're gonna need really a big step up in computational processing power just to be able to handle it. Right. I mean, we can dive into a thousand different areas of this. That's just one that popped to mind when you brought it up.
Ben Ferrar (14:11):
No, no, it's a, it is a good point. Right. And then it doesn't help, right. That we're in the middle of a, you know, a chip shortage generally. Right. Um, that there's, there's huge amounts of capacity that need to be increased just to sustain what we're, what we're consuming today from a, from a, a chip perspective. Um, and you know, you then take these types of technologies and multiply of that by orders of magnitude in terms of what will be required to process a component in the future. Right.
James Demuth (14:42):
Right. No, I think you, I'm sure there's gonna be ways to get smart. Right. You know, you could imagine if you've got a single part and it's in a build volume and it's just repeated over and over and over again, like you could probably do some algorithms, so you only actually model one part and then that's just translated and repeated over and over and over. I I'm sure we can get smart about this. Right. And that's, what's gonna have to happen. People need to get efficient because that has downstream, you know, you just mentioned the chip shortage, right. Like, how do you overcome that? Well, you know, don't put so onerous requirements, right. Find a more efficient way to do things. Um, you know, speaking of efficiency, right. You know, we, um, we, you know, additive manufacturing, it's got pretty low loss, right. I mean, you've got support structure, but in the actual, like printing of the process, you generally, the main losses are from spatter and partially conglomerated powder particles sort of, you know, be changing their particle distribution.
James Demuth (15:37):
And then they get sived out later, um, is like, what, 1% to 5% loss it's not big, um, as opposed to, let's say casting where you got SP screws and other things that sort of add up, um, you know, we, we have a process of pulse shaping that completely eliminates spatter and injecta, it just doesn't happen. Um, we have, it looks like a brief flame, and that means there's a, a big step forward. We see it in terms of repeatability and reproducibility, because you don't have these little errors or defect sites Being distributed throughout your entire print. Um, so, you know, we always try and list sort of like, or think about what are all the ways that we can be better for people better for the planet. Um, being more efficient is one. Um, but when we're, we actually look into direct ways, uh, we think a lot about, you know, we're an electrically driven process.
James Demuth (16:29):
We're potentially just basing casting, which is predominantly driven by, uh, you know, Coke or natural gas. And those have a lot of CO2 emissions. It's about almost, uh, half a ton of CO2 omitted per ton of metal produced. Like you look at all the metal amount of metal produced, you know, casting's 20% of the overall manufacturing industry. Um, so you have to sort of, you know, dial that in, but casting, you know, if you could displace casting and, you know, at that $25 per kilo level, the market size would be potentially around two and a half giga, tons of CO2 reduced per year, um, emitted from manufacturing. So, you know, as a planet, we gotta get to 50 by 2050. All right. Um, 50 giga tons a year by 2050, I think, was the target, right? So two and a half out of 50 by just, you know, one, everything helps. Right. So, um, how can, how can we do that? And, you know, additive in general is electrically driven. It just can't scale. So how do you scale? And I think that's the big question that we're, that we're answering.
Ben Ferrar (17:30):
And so, and, and so what applications of you, are you excited about the, for this technology versus the traditional technologies?
James Demuth (17:40):
Yeah. You know, I think that that dovetails really strongly with the industry that we're talking about. Um, you know, we see strong interest from automotive, from consumer electronics, right. You know, you see, uh, the Bugatti break Caliper out there, but you don't see the Honda civic break Caliper out there. Right. So how do we get, uh, you know, how do we get that level of, you know, commodity level pricing yet still maintain quality and hit those really high volume production rates? Um, you know, there's always gonna the ability be the ability for additive to sort of reach down the cost curve. Like you could have two parts, but because the added part is cheaper or sorry, it's lighter weight, it combines multiple parts in the one, you know, you have supply chain benefits, you can effectively compete with something that was initially lower priced. Um, but it's combine make it even better and better and better. So, um, you know, I think this just the goal here is win-win for everyone involved. Um, how do we, how do we help, you know, make manufacturing, uh, easier and more accessible to everyone?
Ben Ferrar (18:43):
Yeah, no, I, I agree. I mean, let, as if we, if we just talk about you for a second, right. Clearly, uh, origins of your career and origins of this idea was Lawrence oliver, national labs, and very much on the technical side. And about 12 months ago, you took over the sort of CEO position within the organization. Has that impacted your ability to spend time on the, on the technical development side? What parts have you enjoyed? What parts haven't you enjoyed of, of that sort of transition?
James Demuth (19:15):
Yeah. You know, it's interesting. Um, I definitely don't get to do as much technical work as I, as I used to. Um, but you know, part of growing in an organization is at some level replacing yourself and every way you can, along the way, right, as a founder, you kind of need to be, uh, a little bit of the Jack of all trades in the sense that, you know, if there's a, if there's a role that's not being filled, you need to pinch hit, right. Until it is filled in the beginning, you're pinch hitting everywhere. And then slowly, slowly, slowly, you can fill in those different roles. So, you know, I think we've got a very effective team that we've built up, you know, um, you know, initialy it was small. And then as you grow, you, you essentially get various lieutenants in each one of their different areas of expertise and they fill out their teams underneath them.
James Demuth (19:59):
And then all of a sudden, you know, you've got just a, really, a much stronger, robust organization there, you know, kudos to our VP of engineering, um, Curash Comsatt from making that happen and really flushing out the team and, you know, empowering all of our, our technical folks and, um, and non-technical folks in their various areas to make that happen. Um, but you know, it's, we've been trying very hard from day one to hire for scaling and to build our organization to scale such that when the time is needed, we're ready to ready to rock. And, you know, I think this is a, a Testament to it in that, you know, when I made the transition, um, to the CEO role here, it was, it was quite seamless from the technology point of view. Right. We have gotten, you know, we have all of our pieces in place, um, is my heart still in it? Of course. Um, you know, you just, you'll never, you know, that's serve your background. You'll never, uh, divorce yourself from that completely, but, uh, it's, uh, at some level we gotta bring in the cash to make this vision come true. And, uh, that's what I've really, you know, starting in August of 20, uh, 20, 20, uh, that's what I switched gears to, to make happen. So, and since then we've raised two rounds, uh, of funding and, uh, yeah, we're, we're very excited to, to make our, turn our vision into a reality
Ben Ferrar (21:21):
And just like being the British person living in America, just explain pinch, hitting to me.
James Demuth (21:29):
So it's liking cricket. No, I, I, I don't know how cricket works. Um, sorry. Um, you know, like it's kind of a point of view, right? There's a, a position that needs to be played. And, uh, until that the, the, you know, until the, uh, we bring in someone to handle it, you know, I, I would in, in that role, right. Um, for, for the, for the meantime, so, uh, it's, uh, it's just filling the gaps until we, we hired the right person to, to take it off on, on their own.
Ben Ferrar (21:59):
I, I, uh, I did see a, I went to, uh, my first baseball game and, uh, I was actually, I, yeah, like, I, I don't know, six months ago in last summer something I know. Yeah. And, uh, and, uh, and I actually managed to see like a, a grand slam walk off finish, and the people that I went to the game with were like, you'll never see that again, that, you know, it was, it was incredible. It's like, okay, well, uh, don't bother going back now. Oh
James Demuth (22:32):
Man. Oh man.
Ben Ferrar (22:35):
I had, it had to be explained what it was. Right. Because as you say, right, it's a bit different from the bit different from the falls and sixes in cricket that I'm used to, but it was certainly
James Demuth (22:46):
Environment in the stadium sort of carried you, uh, gave you the impression it was a big deal.
Ben Ferrar (22:51):
Yeah. Yeah. It was unbelievable. Um, so I, I guess, you know, there's, there's a fair amount that you guys have still, uh, still got to do around the strategic side of what you're, what you're trying to do, but where, where, what is the direction you're going to go in? Is it it machine sales? Is it part manufacturing? Is it partnerships, joint ventures? How do, how do you see your, your market, uh, shared growing over the next and in, in which areas?
James Demuth (23:22):
Yeah, so we are looking, you know, know what you want to be when you grow up. When we grow up, we're looking to deploy print facilities, print Depot, as we go, all them, you know, initially, you know, near us, you know, nationally, and then internationally, essentially wherever large OEMs need parts manufactured. Right. And you know, it either can be a surround facility. It could be a joint venture with that. OEM could be a joint venture with, you know, an electrical producing company, um, a powder manufacturer, maybe even, I mean, you know, we're talking, you know, our, that 20, 30 machine, um, is, is a beast, right. It's like two, uh, 2000 kilos an hour. Um, for one of those machines, I mean, this is like, whoa, like you, you're talking big. Right. We, we get price breaks from our system growing big. That's how we scale.
James Demuth (24:15):
And it makes it very much so that, you know, we're, that drives you really towards part production and part sales. Um, and you know, one of these facilities, uh, that we're talking about for 2030, it could be a third of a Gigot power production. So you can see why we'd be interested in partnering with a, you know, an electrical generation company, um, you know, and the amount of metal it's making. I mean, we're talking about, you know, 92,000 metric tons a year for one of these facilities. So it might make sense to just make how it is there.
Ben Ferrar (24:50):
And yeah, well, certainly right. I mean, um, you know, then it really does become about optimizing the supply chain, um, to make sure you've got it on, you know, as you need it. Right. It's like any lean manufacturing. I always relate it back, you know, bottle manufacturers. I think, I think shipping powder around the world's very similar to shipping plastic bottles, right? You're, you're shipping a whole lot of air taking up a lot of volume right. In powder, cuz it, cuz of the density of the material, it's like shipping plastic bottles. And that's why now if you look at, you know, people who make cosmetics or people who make, uh, drinks, right. They, you know, there's a factory that makes the product and there's a factory right. Next door making the bottles. Right. Uh, and, and I, I see a similar vision within the additive industry around powders because otherwise you're shipping 40% by volume of air right. To your, to your, around the world. Right. Which doesn't make sense. So yeah. I totally agree with that perspective.
James Demuth (25:54):
Yeah. I mean, we see that they're, they're not gonna be, you know, everywhere. Right. You do get cost savings by doing batch shipments. Right. So the bigger your order, the more it makes sense to have your distribution centers more spaced out. Um, but I think there's gonna be a nice blend, uh, of all of those. Right. And I guess the, kind of the long term goals that we'd have a portal online where even, you know, anyone could say, Hey, I wanna have a part made, um, same day or next day. Right. And it gets slipped in with sort of a, you know, more of a generic print. That's not, you know, maybe a, a high, you know, obviously not separate from a high volume printing going on, but if you can enable that now all of a sudden why bother having your own printer at home, if you can get that level of service. I mean, at the end of the day, people want, you know, our customers want parts, uh, that meet their quality requirements for the right price. Right. And uh, if you can fulfill that need, then you can alleviate a lot of headaches and frustrations from dealing with your own printer. Um, and just get to the good stuff, which is the part that you wanted. Right. Yeah.
Ben Ferrar (26:59):
And, and get the economic benefits of sharing the platform and, and sharing the, the print farm. Right. A again, right. It comes back to these technologies being really tuned to even digitizing the sales process, digitizing the, the order process. And, and I think, uh, I think we have the mindset within the industry where people were, will feel comfortable doing that because of how digital the technology is. Right. It, it sort of comes with the mindset that, uh, that means additive will be successful.
James Demuth (27:33):
Right. Yeah, absolutely. Absolutely.
Ben Ferrar (27:36):
Right. It's uh, it's been a pleasure having, having a discussion with you today. I've certainly, uh, certainly learned a lot, uh, about the technology. Um, I, uh, I thank you very much for joining us. Is there anything you want to say as a, as a final parting comment?
James Demuth (27:52):
Yeah. You know, I think, uh, you know, we, we've covered a lot here today, you know, really appreciate you, uh, having me on and, um, you know, if, uh, if anyone's interested in, in changing the world of manufacturing, uh, should come check out Suerats, uh, as a place to be. So
Thanks very much to James DeMuth for joining us on PowderHeads. We're looking forward to seeing how Seurat Technologies further develops their business and drives industry change. If you have questions or comments about what we discussed in this podcast PowderHeads, send them to firstname.lastname@example.org or visit our podcast page at www.carpenteradditive.com/powderheads. We continue to build an archive of all of our interviews there as well as additional material that provides insight and perspective on modern day additive manufacturing. PowderHeads is managed by Carpenter Additive and its parent company Carpenter Technology, a global leader in specialty alloys for over 130 years. Our goal is to help solve their most challenging material process problems. Learn more at CarpenterTechnology.com. Thanks again for listening and keep building!