Q&A: GE Additive on bringing its binder jet technology to market & designing 3D printers for new sectors

To RAPID + TCT in May, GE Additives wanted to talk binder jet – aware that it hadn’t always been forthcoming with updates along its now five-year journey with the technology.

Then, the main thrust of the conversation was that it wasn’t willing to launch the technology before it was ready, and to help move things along, it had brought on board a fourth Beta partner in Kenna metal.

“We want to make sure that whatever we bring actually does what we say it can do,” GE Additive Product Line Leader for Binder Jet Brian Birkmeyer [BB] customs duty TCT six months ago. By October, the company felt ready. Or at least, sure enough that over the next 12 months it would be. Shipments of its Series 3 binder jet system are now expected to commence in the second half of 2023, around six years after the company announced its intentions to develop a binder jet solution.

To Formnext, TCT caught up with Birkmeyer, as well as General Manager Engineering & Technology Chris Schuppe [CS] to understand what had changed since May, what had driven this endeavour, and what needs to happen next for GE Additive to meet its H2 2023 target.

TCT: Brian, when we spoke at RAPID + TCT in May, GE Additive wasn’t publicly stating a date for when the Series 3 metal binder jet machine would be commercially available. Then, last month, the company announced deliveries of the machine would commence in the second half of 2023. So, what progress has been made in the interim?

BB: So, we’ve continued to progress the platform through its validation testing. We’ve got some good results off the machine there, learned. As you know, we designed the machine and then we’ve been testing it, hence the validation testing. We’ve learned a couple of things, we should move a sensor here or there to get better readings, things like that. And so, right now, we’re going through and making some of those tweaks to the final spec. But otherwise, as we go forward, we’re going to be trying to test our machines to their absolute limits, try to break them where we can, make sure that our customers cannot. And then, as we go into next year, we’ll be starting into production.

TCT: You’re now at the Series 3 machine, with Alpha and Beta systems coming before it. What are the main differences between this to-be-commercialised product and those earlier iterations of the platform?

BB: The Alpha machine was the first iteration of the machine whereby a couple of years ago, we took our first crack at it, if you will, built it intentionally larger, such that we can take subsystems off, things like that, test it . And then our Beta machines were intended to be the end production spec, right? Everything that the Alpha then fed. And then those Beta machines will be fully reconciled, right? I mentioned a tweak of a central location or something like that, so we’ll go through and reconcile those Beta machines, run them through validation testing to make sure your production spec matches, and then we’ll be good.

The only thing that maybe didn’t get covered in what I just said versus what you maybe were after, which is a lot of machines that our Beta partners are currently on today, we call it the Series Two Beta. Everything I just described prior to this was the Beta machines of the Series 3 that we’ve been running, but the Series Two Betas are, generally speaking, more of a lab-scale piece of equipment, they’re not inert capable, so couldn’t be working [on] reactive alloy development on the machine at least during that time. So, most of that work has been in non-reactive alloys, things of that nature, with our customers. And really what that machine has done is it’s provided us a lot of great work experience with those customers to then feed into the Series 3 to say, ‘okay, look, the production grade machine that I bring to the market needs to be capable of this.’ That’s really the difference.

An hour of labor touch time in the aviation industry is a reasonable conversation on the part. In automotive, that’s a joke, right? They just think that’s funny.

TCT: Those developments have taken place over a five-year period, so a lot of time, effort and money has been invested into this project. What motivated GE Additive to enter the metal binder jet space?

BB: If you look across our other modalities today, laser and EBM have their market spaces where they play very nicely. But if you look at big opportunities for us, neither of them really play in automotive at appreciable scales. For us, binder jet was our attempt to go pursue automotive-level scaled volume, and really start to compete more head to head with castings and the type of price points that castings are at. And so, if you look across the three different technologies we offer, [we have], to me, a good three-way Venn diagram. Depending on the application that you have, one technology is going to be the best suited to print it. Depending on the thing, I might be able to print it in all three, but if I can do that, then maybe binder jet is a better option because it’s more cost effective. But if I have something like a LEAP Fuel Nozzle, I’m not going to be able to print that any other way besides laser. So, it just depends on the features and everything. But [binder jet] helps us tap into different market spaces that our other modalities otherwise don’t.

TCT: GE Additive enters the metal binder jet market alongside a number of high profile 3D printing OEMs – what does the company see as the differentiating elements of its offering?

BB: To me, there are three different core things that I think are big differentiators. One ice binds chemistry. So, we’ve been working in binder chemistry actually preceding GE Additive altogether through our Global Research Centre. We built upon that expertise and then developed the machine organically from there versus inorganic. So, binder chemistry is, I think, a big one that allows us to print and sinter large parts as well as small ones, but we can get into bigger parts as well. We have no known maximum wall thickness that I can’t get a good clean burn out of the binder. So, you’ve got good green strength, you’ve got good clean burnout as well. So, good material properties post-sinter where I don’t have residual carbides or oxides to worry about. Binder chemistry is extremely important.

And then you just have general machine design. The whole system has been designed truly to reduce touch time in between all the different process steps as well. If you’re trying to operate in an automotive scale marketplace, any touch time that I have is money spent. So, the system is designed to automate everywhere where it makes sense to, transporting material from MHS powder to machine all automated, and then be automatable where someone like, name your automotive OEM, already has their AGV or AMR systems that they want to operate with. We want to make sure they can interface with those, but we don’t push a solution that they might otherwise be married to. Machine design is extremely important for high throughput environments.

CS: On machine design, one of our goals is certainly minimize operator touch time but also minimize exposure to powder. If you look at that, from the moment the pattern shows up in the factory to the moment that part leaves the factory, there’s very limited or almost no exposure to the raw powder, which today, in industry, is generally a risk that we all face .

BB: And then also our Amp software is, I think, a big piece of the puzzle here. With binder jet, obviously, you have all the distortion and shrinkage as you go through sintering. What Amp allows us to do is pre-distort and pre-compensate those parts. That way, as I print and then sinter based off a sinter profile that we recommend, you get down to the part and feature tolerances that you’re after. Those three things together help bring a build robust solution to the market.

TCT: On the Amp software, what is the big challenge that this platform will address for users of your binder jet technology?

BB: Iteration in the development cycle. So, once you have your build geometry that you know comes out well post sinter, you just rinse and repeat, right? We talked throughput, this machine technology wants to make millions of parts a year of the same part – great. But also in the development cycle, if I have to iterate 10, 15 times to pre distort, if I build my nominal CAD, it’s going to shrink and distort, so then I got to try to adjust and try again. What Amp allows you to do is reduce those iterations down to one or two versus 20. And ideally, you hit it right the first time, but depending on the tolerance that you have to hit, you may say, ‘okay, I need to add just a little more here.’ But it gets you 99% of the way there such that you don’t have to keep iterating over and over and over again.

CS: It’s very similar to if you think about the parallel to the casting industry. When you talk to people in the casting industry, there’s simulation of castings, right? And they talk about shrink, they talk about draft angles, they talk about cooling rates. Similar conversation, just casting has been doing it for centuries, we’ve been doing binder jet for a few years. So, different points in the learning curve, but trying to really speed that ramp into production as quickly as possible.

BB: And of course, the benefit of binder jet over casting is casting, once I make my tooling, I’m stuck with it. With binder jet, if I print it, and I scan it, and they say, ‘I’m a little bit out here,’ I just try it one more time and tweak that thing. And you should be good.

Read more: Formnext 2022: 3D printing goes big while software shines

TCT: And on the machine design aspect, is it a challenge to develop a machine that caters for everyone’s needs and pain points?

CS: It’s a little bit challenging, but I think maybe a good example, if you look at our laser technology, we try to make the laser technology suitable to make aviation-grade quality hardware. You look at our booth there and you go through some of the sample parts, we take parts that we’ve printed with aviation-grade quality parameters, and we may speed up those parameters for industries that need lesser material capability to lower the cost. So, we have some ability to tune with the software to get from maybe a very high barrier to entry market down into a market that doesn’t care as much about the material capability or long term material capability. So, we have some knobs, but to say it’s perfect to go across all industries, it’s probably an overstatement.

Part of the reason we have beta partners on this program was we wanted to go into the automotive space. [But] we didn’t have a lot of experience in that space. So, getting those beta partners upfront, talking through the design of the machine and what’s important to them versus maybe what’s historically important to us when you think about touch times in the aviation industry. An hour of labor touch time in the aviation industry is a reasonable conversation on the part. In automotive, that’s a joke, right? They just think that’s funny.

TCT: Deliveries of the Series 3 machine are slated for the second half of next year – what happens between now and then?

BB: For us, really, it’s scaling that supply chain. So, we’re in the process of standing up our facility to actually touch all the production parts and make them at any appreciable scale. There are two things: one is continuing through that validation cycle to make sure [we’re] getting a lot of touch time under our belts to make sure I can get through 100,000 cycles, things like that. Also validating my preventive maintenance schedules. ‘Okay, yeah, this thing’s failing when I expect it to,’ these sorts of things. But then also, like I said, just scaling up our supply chain and our service teams as well. [Our service teams], quite frankly, today, are primarily staffed by our engineering teams, but in the future, of course, we need to be able to be well tied in with our services organization writ large. So, today most of our operations are US based, so it’s also translating some of that knowledge to Europe and other markets as well. So, there’s a few different pieces of the puzzle, but it’s all in the plan.

Want to discuss? Join the conversation on the Additive Manufacturing Global Community Discord.

Get your FREE print subscription to TCT Magazine.