“AM manufacturers don't really have an understanding of the operating needs & operating environment of the military.”

In the latest episode of our Innovators on Innovators series, the focus is on additive manufacturing in the military. 

Tali Rosman [TR], who spent a couple of years with the Israeli Air Force as part of her national service and is now the Elem Additive General Manager at Xerox, sits down with Michael Pecota [MP], a Contract Support subject matter expert for the Department of Defence (DoD), where he specialises in additive manufacturing. 

The pair convene just weeks after the installation of an ElemX liquid metal 3D printing system on a U.S. naval vessel and 18 months after the deployment of a machine at the Naval Postgraduate School in California. 

In the past couple of years, there have been several other announcements that point to an increasing application of 3D printing in the defence sector.

Across an hour of conversation, Rosman and Pecota explore the opportunities for AM in this space, touch on the considerations that are needed to implement AM in military organisations, and discuss the need for baseline quality standards in 3D printing. 

Below, we have the full transcript of the pair's exchange. 

MP: So your background, we got a brief intro last time, but you were in foreign foreign Air Force?

TR: Yeah, the Israeli Air Force.

MP: What did you do with them? Like, what was your main job day in day out?

TR: It was nothing exciting, I was in the computers unit, just writing code, yelling at other people writing code, that kind of stuff.

MP: That’s the similarities! The form of communication has got to be the same across all services, all branches, and all countries. Yelling and swearing is just standard.

TR: Yeah. Although that was the interesting thing, the American DoD is certainly different than the Israeli one.

MP: How so?

TR: You guys have a lot more respect for the titles. And I mean that in a good way. We call everybody by their first names, we openly express opinions and the interactions I had across the DoD are more people calling each other captain and by the rank, captain and major and this and the last name.

MP: Yes. And it's not for everybody. I did a lot of stuff on my last cruise I think with NATO. And we got to work with the German Navy, a lot of different navies. It was right after Russia invaded Crimea and so my group went out there as like response, we went to Latvia, Lithuania, Estonia, Poland, Germany, France, just saying, 'hey, we're still here, keeping up foreign policies.' There was a lot of fun, but getting to work with all the different navies, which I've never done before, you definitely saw a shift in dynamic. I think it was the German Navy, if I remember correctly, and they actually were encouraged to have spouses on the ship, so they were there as couples and they would have estate rooms and they had functioning relationships and I thought that was amazing.

Get your FREE print subscription to TCT Magazine.

Exhibit at the UK's definitive and most influential 3D printing and additive manufacturing event, TCT 3Sixty.

TR: Yeah, I think that's also one of the differences probably between the Israeli military and the US one is that for us, you go home almost every weekend  worst case maybe every other weekend, and in the US you could be deployed and not come about home for six/ nine months.

MP: [Laughs] So I left home in 2008 and I think I have been back now maybe six times in all these years. We do this thing, so if you're out at sea, and you have either a steak dinner or a lobster dinner, or crab, they're like 'hey guys, we got something special.' We're like 'oh shoot, how long are we extending for?' and then like halfway through the meal, you'll be eating and they'll come over the loudspeaker, 'hey, everybody, we got some unfortunate news, we're doing this mission and we're gonna stay out for another two months', like 'knew it! You don't whip out the steak for nothing.'

TR: Okay, well, you guys have steak and lobster, so you're in good shape.

MP: Oh, trust me. That's why you know it's something special because that is not the case. I usually was on nights, usually night check, and one of the dinners, they individually counted out six pieces of ravioli for everybody and they're like, 'this is it.' So that's the norm, I hope to never have breakfast rice again, where you had rice for lunch, and then the leftovers became dinner and they had just enough leftover where it was breakfast the next morning, I hope to never have breakfast rice again in my life.

TR: So, it’s consistent that the food is horrible across armies then.

MP: Oh God, it is. We would do these on reps where we're carrying all the food onto the ship, you'll do the big line off the dock and you'll be handing it, like conveyored on. And you'd have these boxes, and it would say, 'so in so meat, not fit for human consumption.' And you're just passing it along, you're like, 'I'm gonna eat this - Don't tell me that says that.'

TR: Maybe instead of talking about plastic and metal printers on ships, we should talk about food printers.

MP: [Laughs] something!

TR: That should be the market.

MP: Anything! It's a nice segue into something. I mean, you guys have been doing quite a bit with the Naval Postgraduate School in the lead up to putting one of the first, that I'm aware of, mind you, I've only been - only - but I've been in this realm since 2015. But you guys actually put a metal printer on a ship. Have you had any feedback or anything from that?

TR: Yeah, so I think as you were saying it's, as far as I'm aware, I'm familiar with plastic printers on ships, but I haven't been familiar with another metal printer on a ship, I'm always very cautious in saying, 'oh, it's the first time ever,' because you don't know, but to the best of my knowledge, it's the first time there's a metal printer on a ship. It's been there for three, four weeks now. So it's new and fresh, but so far, the feedback is good. They're printing parts and once a day when they have connectivity, they send us photos, 'oh, here's what we printed today.' So obviously, I haven't been able to touch and feel and analyse the parts that they printed on ship and make sure they're exactly consistent with what they've been printing in land. But at least in the photos and in the informal email feedback, it looks very positive.

MP: Yeah, I can't go into too much, but I've actually been involved with that the last week, at least in chains and things. So yeah, a lot of the feedback I've had is that their samples have been comparable to the printers at NPS, Naval Postgraduate School, which was what they wanted to do to my understanding and correct me if I'm wrong, it wasn't really to make parts at sea this first time. But to evaluate the printer and see how it does in sea state, how do the items compare to what was expected. And they did a lot of couponing. At least that's my initial understanding. Have you have you received other feedback?

TR: Yeah, so the goal was again, because it's novel to put a metal printer on the ship was really to try and print the exact same parts in land and at sea and then the next time the ship docks, take the parts off the ship and do a thorough analysis to compare them to the parts made in land and see if there were any deviations. With that said though, I understand they were very enthusiastic about the printers like ‘oh, well we're done printing those tests. Can we already start cranking out some parts’ and I saw they printed a couple wrenches and things like that that are kind of usable on the ship now so we're really waiting for it to dock, get those parts analysed and initial feedback is positive. So, we're excited about that.

MP: So like when I've seen other metal build prints, mostly I deal with like laser powder bed and things like that in our circles. Are you guys doing a lot of the standard couponing around the around the item to do that destructive and non-destructive testing?

TR: We're certainly printing a lot of test parts for customers as they're exploring the technology and getting up to speed and then there are customers that are actually already using it to make parts that have already purchased the printer, are comfortable with technology and are printing actual parts with it.

MP: So, with a lot of the laser powder bed but there is always a certain level of refinement you have to do afterwards. A lot of that's very granular and you do it down the... the prints coming off of the Xerox system I've seen are very, almost cold sprayish, where they exceed a little bit beyond the boundaries and don't have clear lines in the in the layers, what kind of post processing and refinement is critical to that system to bring it back down to the expected shape and geometry.

TR: So we don't have the shrinkage and warpage that you have with other technologies. So in fact, the part as it comes off from the printer is already usable. But that depends on the use case, because the surface finish, so there's two things in post processing that you may want to do. The first one is around the surface finish - as it comes out from the printer, it's sand casting, or better, we call it slightly better than sand casting quality. So definitely you might want to sandblast or machine or what have you for a smoother surface finish. And then the second thing is a T6 heat treat to improve the strength. So depending on the use case, you might be comfortable with the material properties as they come up from the printer, and you are not going to T6 heat treat it. But for most use cases, you probably would want to put it, for a few hours, in the oven, which part of the whole spiel of the liquid metal jetting technology and the mini factory in a box is that this is kind of a regular conventional oven, it's not proprietary, like some of the other technologies, and because the part pops right off from the build plate after the print, instead of putting the whole build plate in an oven, which would require to have a decent sized oven, here, you're just putting the part in. So, you can have a relatively small conventional oven for a few hours, improving the strength and you're good to go.

MP: So you've seen other metal builds where they put the whole like right out the printer on the build plate and everything into the oven for other technologies?

TR: I think there are some technologies where because the part is still on the build plate, you put the parts with the build plate in it.

MP: Interesting. I just personally haven't seen it, normally they remove the part and then put it in. So I'll have to look into that.

TR: Yeah, I mean, for us and I think every technology has its benefits and its strengths and weaknesses. So I don't think anybody's kind of replacing the other. I think it's all complementary technologies depending on what you want to make. And so I think our strength is really in the deployability. And the fact that it's not just about putting a printer in a box, in a Conex box, but putting the manufacturing, which by that I mean the ability to get a usable part in your hands, in a single Conex box. And that would include, again, putting an oven there if you need to, etc.

MP: So, in comparison, if you were going back to what we were talking about earlier, the military application focused on here. What things do you think, in AM as a whole, like 3D printing as a whole, would be good in that military setting? And what things that you've seen today do you personally think would be difficult to implement due to the military nature and fundamental aspect of working in that environment?

TR: So, we were talking about the US military and having six-nine months deployment and remote locations where it's hard to get parts into, and also locations that you might struggle getting a tonne of inventory, or every part you may need. I think for those purposes of getting not even complex parts, just what the Navy's printing, wrenches and things like that, the ability to make them on demand at the point of consumption. I think that's huge. I think the extreme other end of the spectrum might be super specialised parts for aircraft and the likes. So I think for that we're not there yet with liquid metal jetting. But there could be other technologies that are better suited to fit for us. It's really about what I call the backup generator. So the ability to just get the parts that you need on demand, it might not be perfect. And if you can be on the grid, quote-unquote, you might prefer to be on the grid, but in the instances where you can, you're really happy that you have the liquid metal jetting options. And it's, I'll call it good enough, or better to be used if that makes sense.

MP: No, I feel you, I mean, that's definitely been my understanding as well in this past few years is good enough is the target area for AM right now because if you're, I'm in the engineering community backing up the foreign deployed and even daily use military operations, and there isn't a good understanding yet on, I would say, most AM technologies, there's a lot of standardisation out there for like FFF, for your big industrial systems like Stratasys, due to their work with the Air Force years ago. And then you've got a really good understanding of laser powder bed systems, but still, it's not like you can go into SolidWorks and CAD up a part, like you can with traditional manufacturing, say, 'Hey, I'm gonna make it out of a block of titanium.' And then you can do a lot of work to understand how it's going to behave, so there's a lot of mathematical equations behind that, so you understand what it will do. AM because of the layer bonding, you don't have that, there hasn't been enough research to create those algorithms so that you can have these computer programmes say 'yes, this will perform this way.' Like on the laser powder bed, we had an item that tonnes of information out there on it was for the V22 linkage assembly, I think on our side, it was like the first flight critical part ever made, I know I'm getting those phrases wrong, they always get on me like, 'oh, flight critical means this, safety is this.' So I might be using that wrong, but either way, it outperformed expectations, and even the OEM won, so like, they put it on there, they tested it, they pulled it, they did all the destructive and non destructive thinking it would operate in this bandwidth. And it did significantly more and the last I heard, it's been a few months, that part is still in service on that aircraft.

So, with AM it's just so new, that getting... you were talking about that forward deployed aspect, you put a printer, it's your backup generator, as you were saying, it's ready to go. But on this side, the engineering community needs to understand there needs to be standardisation. How are they trained? Are they operating the machine appropriately? Are we relying on their knowledge of the slicer? Or setting up that environment exactly the same way every time? Or is it automated? Is the machine so tuned in that every time you print something, it's going to behave the same way? That's what's really lacking and that's why I'm thankful for a lot of groups like I don't know if you have any familiarity with America Makes, all the ASTM standards, like standardisation of processes, qualifications, training, and material. Material's a big one, I think is critical to that everyone always talks about the distributed network, globalisation of AM, that capability, similar to what you have today with paper printers and faxes, it's the same wherever you go, they fall under the same criteria, you print a paper at Kinko's here, I'm in Maryland, or send it across the world, it's going to print the same way. AM definitely definitely, definitely is not there yet. But it takes all of that to do so. On my side, we've started to set up that network the last few years, so we've put a lot of effort into standardisation of process, quals, specs, so that things will be the same. But that takes so much work and fundamental knowledge and processes to make that happen, that it's quite daunting in the military aspect to be honest.

TR: Yeah, but I think the military offers a flexibility that you're not going to find in corporate. So I'll give you an example. We had one of the customers we were talking to, they were looking at it because their lead time to get certain parts that they're after is eight to nine months. And with ElemX they can literally make those parts in a day or two, it's a no brainer. Then they said 'Ah, but our internal qualification process for part takes about six months.' So the whole thing about a backup generator is that you need to be able to call it to action immediately. If you can get the part in the day, but then it takes you six months to qualify it internally, that doesn't make a lot of sense. And it doesn't save you a lot of time. I do think one of the things in the military is that sometimes commanders on the ground have the flexibility and the empowerment to say, 'this is good enough.' I need this part now. I don't have another way to make it. I'm going to use what's coming from the machine, which you're not gonna get in a corporate environment. And then to your point, I think one of the things is that, yes, it has to be much easier. The software workflows have to be much easier. And that's for everything from knowing the part is printable, how is it going to function? What will be the material properties and qualification of the part? Otherwise, you're killing the case for additive manufacturing, or you're making it exponentially more challenging?

MP: Very and to your comment about the military and forward commanders having a little more leeway. In my experience, I think that's, sometimes, I come from an aviation background, I was repairing helicopters for 11 years. And in the aviation world, at least, everything is documented, it's so rigid. If you went in there, and all I had to do was like flip a switch, on/off switch, you come back inside and say, 'Hey, I turn this switch off in accordance with this publication at this time, and it needs to reference up, and then you have quality assurance and quality assurance on top of that, it's always a minimum of a two man job, somebody doing the action, somebody witnessing it, that's qualified to witness it. And then behind the scenes, there's somebody all the way back in the QA office, that will come through and do periodic inspections to make sure those processes are being adhered to, like the bigger ones, like hey, you said you did it in accordance with this publication, show me where you got that publication. We are so documented and thorough that from an aviation side, it's very, very difficult to do new, to try things, to experiment, because there's a risk. If you're driving a Jeep, and that's your job everyday is to drive a Jeep, and all you need is a door handle part, right? And it breaks, you can 3D print that door handle, if it falls off, there is no consequence of failure. If it breaks 100% of the time, worst case is, you're gonna use the other door, but on an aircraft, if you're flying and a door handle breaks, that door handle can be sucked into the engine, lots of bad things can happen, that could be your point of egress, your single point of egress.

TR: That's the thing. What I'm saying is, 80/20 rule, right? Okay, so let's not try, because I think a lot of people try with additive manufacturing, from day one, to aim it at the most complex, the most challenging parts. I'm saying the opposite. I'm saying no, no, no 80/20 rule, let's knock out the 80% that are not a primary part of an aircraft that if it breaks, the whole plane could go down, let's focus on that 80% that we know we can do. Get confidence this way, qualify everything that way. And by the way, even if you don't get to that 20%, you're still fine, because if we're all that 80%, you no longer have to carry inventory, you can really focus your inventory on the most critical things that additive manufacturing can do. And you're clearing way in your inventory for that, because there's 80% of the parts that arguably you can do today already.

It's really difficult to advocate and say, 'hey, if you spend millions of dollars for the next couple of years supporting these printers, we're going to make super low critical things that nobody cared about anyway, but we're gonna do it really well.'

MP: No, I'm totally with you, trust me, you're ringing true to my personal beliefs. I'm a big fan of putting a very cheap desktop 3D printer everywhere and anywhere and saying, 'Hey, make low critical stuff, make knobs, covers buttons, stuff for ground equipment, do that all the time.' Even if you're not in military, I wish we had more 3D printers just out in the community and people were doing things that they can. From my perspective, I've seen a lot of hardships with implementing something like a metal system or an industrial system in the same viewpoint and saying, 'Hey, let's tackle low critical stuff' because of the price, the threshold for operating it, for having qualified people, not having them go on rotation, be out of the command, all of the logistical stuff that not just operating a printer comes with to make it as successful. That's incredibly difficult, time consuming, and it costs a lot of money.

And when you're talking in this environment, looking for a resource sponsor, who's going to fund this large process so that we can have engineers even evaluating those parts and make sure they're doing the low critical parts. It's hard. It's really difficult to advocate and say, 'Hey, if you spend millions of dollars for the next couple of years supporting these printers, we're going to make super low critical things that nobody cared about anyway, but we're gonna do it really well.' Resource sponsors they want those big wins, they want to advocate and say 'yes, we save this much money, we had a return on investment, we got these aircraft or equipment up, we solve these problems,' it's hard when you're focusing on lower criticality items and saying 'this is where we start, we're going to get there. But we have to do this many years of work, we need this much money to get there.' It's difficult. And that's definitely what I have seen for everybody in the AM community across all the services is really getting those resource sponsors aligned with a big picture idea that's going to take it past even their time in office, like, hey, we have a ten year plan. Ten years gets us to those critical items, but we need the funding, the manpower to do it. And people really don't want to do that. They're like, where's my quick wins? I think sometimes personally, it's where are those wins that are going to take effect while I'm in this seat, so that I can promote, so that I can go on to bigger and better things, I don't have time for a ten year strategy, let me do anything within two year, one to two years, so that I can say, 'I did this.'

TR: So that's interesting. So you think the problem is they want the amazing, shiny parts that they can allude to instead of, actually, I've solved the problem. It's just not a quote-unquote, sexy problem.

MP: Oh yeah, I don't just think that, I know that [laughs]. I have been in very many meetings with lots of people on the same topic, this is striking relevant to things that are happening right now, to be honest, I am seeing this day in and day out that yeah, it's those super sexy parts, those ones like, hey, we did this in metal, look, there was an OEM lead time of three years, they wanted millions of dollars to stand up a new production line, they have us, we don't have the technical data to do it. Those are the ones because these days, everything in my field is related to data and metrics. There's all of these things saying, 'hey, for this aircraft, and this platform, here's what we call the leading degraders, here are my head hurters, these are the ones that I need answers for today, because I know they're problems.' And the higher up the list they go is the more money they're worth, the more money they're gonna have to throw at the problem to solve it, thinking traditional manufacturing, traditional contracting even, it's those big figures that get a lot of attention. But on the other side, I think that AM manufacturers, the OEMs for the machines themselves, don't really have an understanding of the operating needs and the operating environment of the military. And so they're starting to create products, whether it's software products, whether it's the machines themselves, the materials, and they're trying to sell it to these military individuals and forward thinking people and saying we can do this. And in reality, they can't, their machines can't or nor will never be fully cyber compliant. Their materials and the material handling requirements might not meet the specification for deployment, like no, you can put that printer there, but you can't get the material through the supply chain to get there any quicker than you could get that OEM part. Or even just things like having a requirement to touch the cloud or that stuff to do cloud based slicing, and not localised computing.

All of these things are non starters for a lot of the equipment that we're starting to see come out into the field. And I've seen this a couple of dozen times now, where a very high military leader says I want to be forward thinking I want this printer in this location. We want to try it, we want to see and the printer gets there and then all the logistics stuff comes in the programme management stuff. Hey, it's here, but we ran out of material, how do we get it, how how do we pay for it? We're a local forward deployed unit, I have systems for ordering equipment, and this isn't built into that. No one thought of this or, 'Hey, I'm a printer operator, I've got the machine, I'm qualified, but I haven't been through a year more of CAD, I can't design anything' or 'hey our licence expired three months in, how do we get a new licence?' All of these things I have personally, in my active duty time, seen, tried to find solutions for and that's one of the phrases I often say is nobody wants another printer. Nobody in the services wants a new printer in a new location, they want the support, they want the the full infrastructure for the stuff they have, they just want the traditional support, and to be able to use it and trained.

TR: But to your point, it's so funny, it's easier to get a million dollars to buy a new printer than it is to get $1,000 for a kilogramme of material.

MP: Bingo! You're 100% right. And that doesn't even go for AM. The thing to avoid, from my personal experience and time in, hazmat. Anything that is qualified as hazmat, it has entirely different procurement cycles, even if you're home, like when I was stationed here at Pax River, we had, it was one of the welding shops, and they went down for a simple gas canister, they needed this inert gas, they've been using it for 20 years. And for whatever reason, that particular manufacturer of the bottle they were using last month was gone. Maybe they didn't meet military qualifications. Maybe they lost their certs, I don't know. And so it's like, 'okay, we know we need it but there's no supplier.' So we had to reach out all the way to the base, because the base here controls the hazmat, so now you're dealing with the base hazmat folks who are removed from daily operations anyway, and say, 'Hey, I need you to do the research, find me a subsequent alternative.' By the way, this isn't just my location, this is everybody in the fleet. So now we're talking about a much larger procurement cycle or a larger contract. And it can be months and/or years until there's a replacement in the supply system to fix that. And that's just unfortunate. And I personally think that's why we've seen the big push at least in the Department of Defence on the US side for FFF, for material extrusion systems that don't require any specific or specialised hazmat. Yeah, every few months, you need like lithium grease or alcohol to clean it up, those are already in the supply system, but the material itself, the thing that you need to operate the printer, the filament, is a solid, it is not considered a hazmat. So, it goes through purchasing and procurement as no problem at all.

TR: So, it's spot on exactly what we're doing with the liquid metal jetting, right, it's completely safe, we don't have all the environmental health and safety and the spool of wire, we always say it looks very analogous to an FDM/ FFF technology because you literally have a spool of wire without any special storage requirements or anything of that sort. For that reason, but to your point, yeah, it's speaking of going after the sexy shiny objects, instead of the meat and potatoes of what a lot of units need, it is often easier to get a million dollar for a printer than 1000 bucks for materials. And then the other thing you were saying, which I relate to, and I heard that in an AMUG panel a few months ago, is that this needs to be easy enough for a 20 year old soldier to operate. It can't be that it's aimed at the expert that has two years of the learning curve cetera, especially not if you're going to do it in a distributed fashion, if you're just going to have it in one centralised facility. That's a different story. You can have a set of experts that are there. But if you want to put it on ships, you want to put it in remote deployments of army or marines, you want to put it in various bases all over the world, you're not gonna have these experts everywhere. So you have to make it much easier. And I think to make it much easier. Well, that's not just the printer manufacturer. That's a few companies coming together. And for all the talk on consortiums and putting software and the hardware and the material and the user together in the same room to solve the problems, are you seeing that actually happening? Or is it still siloed?

MP: 100% siloed. It is so siloed, it hurts.

TR: [Laughs] You didn’t even think.

MP: Talking about it earlier, America Makes I think is great. I've seen really good things and, I'm going to be honest with you, I've also seen some really bad things, I've seen projects go in with great intention that I personally was waiting like,' Okay, if this comes through, what comes out of it will benefit us all.' And sometimes it's been, 'it just hasn't worked.' Or, you know, sometimes contracting doesn't work, there's stuff in there. But as a whole, I think they're heading the right direction. The idea there that we're going to centralise knowledge bring folks in, I love the intent of American Makes, what it represents and what it could be. And there's plenty of things like that where different programmes SBAs, SDRs, and things like STTR, once those go through or while... the government owns a lot of the data, so there's been a couple projects that I've seen where like, hey, we want to work with this system and/or this software, 'oh, hey, that was started with an SBIR years ago, and the government has a dev token, you can access that, that happened to me last year, in fact, so I called up the company, found out who owned the dev token, and we were able to get the data we needed because we owned it.

TR: So the burden on putting all of this together and making sure you have truly an end to end workflow from software and materials, who do you expect to lead it? Should it be you guys i.e., military? Should it be the printer OEM? Should it be entities like America Makes?

MP: That’s a good question. I don't think there really is anybody set up to do it other than the OEMs, and possibly a group like America Makes to drive it in working with the OEMs. The military is so big, so vast, so many different cogs and we said stovepipes earlier, data gets lost, ownership, who owns that? How do I get access to it? It's very difficult. So I think if the OEMs started to do that better, it would be a lot better for all of us. I have an interesting comparison. And that's I have a desktop operated, not desktop, it's rather big, but a CNC in my garage. And it's open source, it's wonderful. Every time I make something or carve it on my CNC, you type in like three or four different basic settings. It's feeds and speeds. But at the end of each carve, it asks you, 'was it good? Was it bad?' If it's good, it'll it say perfect. It puts all of those presets into a larger network, back to the OEM, and if it's bad, it asks you a few follow on questions. 'What was it?' 'Did you see chip out?' Was it braking?' So what you end up with is your users are creating a database for you of what's working, what settings aren't. And so then it gives you recommendations. So if you go to carve something and you're using this bit, it'll say 'hey, we recommend this, this or this.' I have loved that for my CNC and that's something I would love to see fed back into the AM community is that, 'hey, all of these settings, even a desktop printer operating Cura has hundreds of individual settings, if you get into the advanced.' Did it work? Yes. If it worked, then hey, let's push that back. And eventually we will have this database of user entries that, 'hey, I selected negative two infill?' Oh, hey, flags red, we based on our information, we know that 100% of the time, this does not work.

TR: Yeah, but you’re doing that on your CNC machine at home. I guess part of the problem is, if I were to put it on a 3D printer in the military, I wouldn’t be able to connect it to a network that would feed me back this information.

MP: But you could have localised data, even if you don't have the benefit of the bigger network. You could go into it with an existing database like, hey, the OEM provided this database based on all their testing, we won't drill into it, but then you forward deploy, where you are in that environment. And even if you're not forward deployed, you're at home, you bring it into a military installation, you work with that initial database, and then you collect your own data on top of that, like for the last year you've been operating it and every time you use this tip you failed, it can build upon that initial database for localised information. And that's it. That's a nice medium strategy. It could still work and maybe every few years or every few months, you bring in or export entire databases. Believe it or not, that's how.. when you're working on aircraft, you have a programme that documents maintenance. And that's how that system kind of works. You go out with your database and log files, you operate forward deployed based on what you have. And then when you come back, you import all of that into the 'mothership,' back into your home base, and you pick up, so there's a way to marry that between those two different schools of forward deployed and isolated and relying on larger networks, I think.

TR: But again, you’re putting the burden on the printer OEMs, so it looks like our hands are gonna be full.

MP: Well, a lot of that burden in other industries - AM is very new - but the burden is already there. When I make my coffee in the morning from my coffee pot, I don't question that it's going to work 100% of the time, I don't have to think about my bean count, I don't have to think about too much.

TR: [Laughs].

MP: Even if I put in the wrong amount of water. They've done that work and proven it out. And if you read the bottom of your coffeemaker, there's all these standards that it met, it met this standard, and this, it's an electrical appliance, and you get your 3D printer and maybe it met a standard depending on what printer you have, but the industry isn't there yet. And that's why it's going to take the OEMs working with groups like American Makes, ASTM to create those baseline standards for everybody so that we can have better equipment in the future. And your coffee maker, I don't think, got their back in the day by relying on the military for their source data. I like to think that that was the OEM.

TR: Yeah, no, absolutely. I think for as long as I've been in this industry, everybody's been talking about the click to print, you go, press two buttons, it starts to print, and you're done. And we all know that's not the case today.

Read more on metals in the military:

MP: I think we’re getting closer.

TR: That's my point, I think we're getting closer, I would have expected by now by the year 2022 for us to have already been there as an industry. But it's taking longer than I would have imagined ten years ago. But I think we're starting to get there, to a true ease of use. Certainly, I think with the low end, plastic printers are seeing that. And I think it's not long before we're going to be seeing it in production grade equipment as well. I think some technologies are just not going to lend themselves as well to it, but others are well positioned for it.

MP: I think a lot of it has to do with sharing data. Those low end printers, I think the big push was soon as those patents expired, everything became open source, people started to realise like, hey, I can collect and share data and I can leverage what other people have done, I can go look. My first printer that got me started, this whole thing was made of wood and zip ties. And I built things, I collected my own data, I did test coupons, I would make a tweak to a setting, and then I would post it, I would publish it, I would put it into forums and be like, hey, I was able to do this to my printer, tighten these belts down and increase the setting and still get these kinds of prints. That that big collection of data, understanding of what other people are doing, being able to see and not having to reinvent the wheel constantly, I think really pushed that. And I'll just say like the desktop printer side, for VAT photopolymerisation, some of the printer settings now are two or three things: you level the build plate, it's light on, its light off, its bed up and bed down. They're really becoming much more simpler to use. So I think that collectively, even for the large OEMs with more advanced technologies where there isn't data yet, sharing data, giving your users the ability to capture data appropriately and feed it back, but having a infrastructure set up for the collection of data so that you have comparable data, that you can do something with it later on is what's going to advance us as a group, as a technology but keeping data, hoarding data or not collecting it at all, is really what's going to hold back the industry from moving as fast as I we all want it to.

TR: I agree with that. And as the printer OEM side of this conversation, we are going to collaborate as much as needed on that and make all the efforts required. Because I think in the year 2022, there's no reason for us not to get to a place where it's seamless and easy and fast to adopt AM into the operations.

MP: No, absolutely. And although some people might think their printer is special, we've seen it these last ten years is that somebody puts out a technology, puts out something new, they give it a name, it'll be an acronym, this is the VSL4 style of printer. Well, you know what, it's going to be out  commercially, it's going to be out for a year or two, and somebody's going to make a tweak to it, they're going to put a curing light in a different area, they're going to do something in a different way and then they're going to release the VLBot6. It's like, okay, I get it, you have a product, you need to target it, but somebody else is going to do it or do it better soon. So if you, from the jump, say, 'Hey, I'm releasing this technology, I'm sharing everything about it, we are going to be agile, we are going to make improvements as we see it, I think as an OEM would do a lot better than what we've seen with some of the other technologies when they start new, and try to do something new.

TR: Yeah, absolutely. And I think it requires this agility, startup-like movement around the improvements, while also I call it having the corporate backing, because I think a lot of companies in the space underestimate the level of investment that's required into the hardware. And that actually, a lot of the investment is not in r&d, and having a prototype in the garage, but actually, the bulk of the investment comes in the engineering, and hardening and productising it, et cetera, which is why you're seeing hundreds of hardware startups. And yet, when you look at how many actually have more than 10 machines out there, the numbers just drop. And how many startups were super promising five, six years ago, and have not delivered on that promise. Because it requires such a rare combo of the startup mindset, but with corporate pockets, because it's hardware.

MP: You know, what's interesting here is I've seen a lot of people in additive go from being a small company, like you said a startup where they can be agile, you want to make a change to your printer, you do it. You want to make a change to even how your company handles a certain process, like a spec or document something training, how you train people to use a printer. It's agile, okay, we saw a problem, we fixed it, we saw a problem, we fixed it. Once an OEM makes a jump and becomes larger - not a real company, that's terrible - but a big company. And I'm sure you understand this from being at Xerox, I'm part of a larger entity. Being agile becomes hard. I've been working with an OEM this last year and helping, like, 'hey, we have this problem. We think there needs to be this kind of solution' and seeing a corporate entity try to make change hurts. Like, it hurts me as a person. I'm like, 'wow, you operate just like the government. You operate just like we do here where change is hard.' And yeah, I hope that as different groups, and mainly printers, start to get bigger and OEMs, they build in the processes to their company's operations that they can make changes and adapt because personally, even in this new space additive companies aren't that old, I'm seeing that, I'm seeing that they are focused on becoming bigger, on growth on going back to the board and showing how much money they've made, that they start to become stagnant. They don't make change. This is their product, they can't. And in order to make a change, it's going to take a year two, three, and by then it's already irrelevant. People have come up with weird little solutions, or they've ditched that printer or that company altogether for another one.

TR: Yeah, I think that to me was the biggest thing about being in this industry for so long and in Stratasys specifically, I was in corporate developments, I got to see all the different startups, and I would revisit the pitch decks three years later, five years later, and it would be exactly the same. They haven't been able, even as startups, they haven't been able to progress fast enough, pivot, make the required changes, let alone in the larger companies.

MP: Going back to the theme of this podcast, the military application, I think that a lot of printer manufacturers do really well, they'll come up with a product, it works. And as they try to get bigger and better, their knowledge and their understanding of how things work is from the corporate side, or from industry, sometimes academia, and it's not a military background, so they don't understand that 'hey, in certain environments, you can't have a USB card, a method of data transfer that isn't hardened.' And so they create these products that are advancing technology, that are just incredible out in industry, but you go to do DoD applications, like 'nope, hey, that has a camera on it.' So there's risk involved there, that camera could be recording things, maybe it can see outside of the build plate, we've got to account for that from a cybersecurity standpoint. Maybe it requires on a method of data transfer over USB, or similar and 'whoops can't do that.' Maybe it requires connection to the internet, like we said, a hazmat maybe it's it's a huge printer, and you have to be able to access it from the back, the side, the bottom to do maintenance. And yeah, in field, we have limited space, and we can access it this way. So I think for a lot of people in industry to start to understand military applications, it's very simple. More companies just need to hire veterans.

TR: [Laughs] I think that’s the mic drop, more companies hire more veterans and it’s good.

MP: It’s good for everybody. So many veterans, maybe somebody's job in the service, they were a, I don't know, ordinanceman. So every day, they handle ordinance. When they go to exit the service, they have a hard time finding jobs, you don't have ordinance men in every city in every company. But this is what they do know, they have been operating possibly 20 or more years in that environment, just because they were an ordnance meant that doesn't mean they weren't in charge of hazmat at some point, they might have been in charge of data, you have all of these, what we call, collateral duties, other things that you've had to do in your line of work, that wasn't your primary tasking, those collaterals, I believe, mean more to this emerging technology in companies to understanding the fundamental nature of how the military works, than that individual's primary job, so  hire somebody that was an ordinanceman, a para rigger or something like that, and put them in your r&d into your logistics and pick their brain and be like, 'hey, what do you think?' Because they're gonna be like, 'No, I couldn't use this in my shop. Everybody knows that, it has a USB, you don't know that?' And you will be surprised the knowledge that every veteran has on daily operations that we think is common sense coming out to industry and you're like, no, your regular person does their specific tasks. They wrote these manuals, but they didn't know anything about hazmat. They didn't know anything about cyber security, cyber ops, all these other things that is common sense for us, isn't so common when you get out and I think more people need to recognise that and not just look at a veteran's primary job and what they did via their rating or MOS, but what are their other external experiences and how can they contribute?

TR: Absolutely.