PowderHeads: Episode 8

The Future of Additive Technologies with Byron Kennedy

Listen in as William Herbert, Carpenter Technology's Director of Technology and R&D, sits down with Byron Kennedy, CEO and Founder at SPEE3D, for an interesting discussion that ranges from solar car racing and electric vehicles, to facing the challenge of making metal parts, and covering the exciting future of additive technologies.

You can read the transcript or listen to the full episode below. 

Full Transcript

Introduction (00:10):
Hi everyone. And welcome back to PowderHeads, a Carpenter Additive podcast. With each episode of PowderHeads, we bring you the minds of industry experts and delve into topics that are defining how additive manufacturing is making an impact on the world. Today, we're going down under. Will Herbert, Carpenter Technology's, vice president of technology and R&D, hops on the line with Byron Kennedy, CEO and founder of Melbourne, Australia-based SPEE3D, a company founded to address the basic challenge that making metal parts is hard. Byron describes a journey that begins with racing on the world, solar car challenge team, and then moves into a method of additive manufacturing that doesn't rely on laser technology. However, it is the software that Byron believes is the real differentiator. It's a super interesting talk. Enjoy the conversation...

Will Herbert (01:07):
Byron, welcome to PowderHeads. Uh, you're joining us from Melbourne Australia. So thanks very much for coming on the show. So you are, um, Byron your background, You're an engineer and you're an inventor. And I know you've started a couple of companies now, but what led you into additive and maybe just give people listening a bit of a background as to how we got to this point. You start in this company is SPEE3D?

Byron Kennedy (01:35):
Yeah, so, so our background, my background, um, goes back a few years now. Um, originally, uh, when I was at university, I was involved in solar car racing. So, uh, we built solar cars and yeah, it's quite interesting. The, the generation of electric vehicles, which is coming out now was really on the back of that solar car work. So, so research university, people will sometimes, you know, question why you do it. Well, you know, it may not be the outcome happening right there. And then, but you know, in 20 years you can see what then has happened with the electric vehicle industry. So, so we started with solar car racing, We developed some electric motor technology, myself and my co-founder Steve, uh, which is used in solar cars, Uh, we ended up commercializing that technology. We spun it off into a company, built that company and we sold it to a large U.S. manufacturer of electric motors.

Byron Kennedy (02:34):
Uh, we worked in manufacturing them for 10 years, tooling up that electric motor that was very successful generating 50 odd million in revenue by the time we left. And, um, and so in that 10 years, we learn a lot about manufacturing. We saw this technology, 3d printing coming, um, and, and when I'm talking 3D printing today, I'm talking just metals. We don't do in the plastic so they'll be around that metal space. Uh, so we saw this technology coming, but the reality in the production space in the production world was the existing laser-based technology. Although, you know, quite fascinating, the kind products you could produce was just too slow and too expensive for production. And our background was production, s o we thought to ourselves, could we actually take 3d printing into the production world? And thus, uh, when we finished up at the, um, motor company, we formed, um, Speed 3d and the vision from day one was always clear, uh, 3d printing for manufacturing.

Byron Kennedy (03:43):
So, um, so how we actually came across the technology, I can delve a little bit into that. Um, the company uses a deposition technology or printing technology called cold spray, in which we fire metal pellot powders at supersonic speeds about a thousand meters a second, when it hits a surface that then sticks and creates this full density part. And it was interesting, it was used by the us military for repair, um, but used in a different way. So when we saw that technology, we thought, could we adapt that, um, let's say a coating technology and adapt that to be able to build parts and that's the, the company was born.

Will Herbert (04:28):
So how much were you able to take off the shelf as it were, and just assemble into a machine and how much was really having to rethink the way this was done to make it into a 3d type of, uh, you know, deposition layout?

Byron Kennedy (04:43):
Yeah. So, um, as an engineer, you're always wanting to improve things, but you've also got to take a step back and say what's the best for the customer. Um, and in a production sense, having, um, verified commercial, um, components within a printing technology is ideal. So, you know, some of the challenges with the laser based systems is, you know, they're very finicky. They're, you know, they're, they're fantastic as a lab tool, uh, will they see their way into full scale production? Um, the jury's out on that one. Um, so when we came with this, we can have this as a different way saying, you know, what's industry proven and robust that we can use and what can we then where where's the value that we can add. So we looked at that for a while and that's, we've, we've come up with a combination of very robust proven technology.

Byron Kennedy (05:41):
And then we added a lot of software to it, interestingly, so the technology uses a proven six axis robots, eight off the shelf, as you know, um, we use, um, uh, pneumatic systems, which are all proven in the field switch, like, you know, bit very high quality seems PLCs, all of that sort of stuff where you'd expect in an industrial product, you want to be using net gear. Um, and then really the sophistication is in the software and that's really, that's what we bring to the table. Um, and we see ourselves as a commercial software company, um, industrial software company, um, and really, you know, we have the hardware platform, which we sell, which is, which is unique, but the key to it is the software. And if you actually compare that today with traditional metal 3D printing, so metal 3D printing has relatively simple software, but very sophisticated hardware. Um, and that's a challenge going forward for manufacturing because if the next version of printer comes out, you can't just upgrade. You know, you have to buy new hardware where with this system that the hardware remains fairly constant throughout and all of the updates really coming into the software. So that's where this proven model of production will then come in because people aren't afraid to invest in the hardware because they know all of this, the improvements are going to come through the software.

Will Herbert (07:12):
Okay. So you mean, you guys could do sort of push updates to the software and you can keep it current as it were in terms of the latest learnings on how to control it without having to necessarily switch out the whole unit, uh, for, you know, a great expense every time.

Byron Kennedy (07:29):
Yeah. A hundred percent. So, um, you know, we have our software teams in-house and they push out the, the updates at regular intervals, um, in terms of the hardware, as we said, it's, you know, we do have rocket nozzles and robotics and all those sorts of things, but the, um, you know, it's, it's proven technology. It's, it's robust, it's Hardy, you know, the, probably the biggest, um, Testament to that was, you know, we, we actually took one of our printers would put it on the back of a military vehicle. So I run my truck. This is in conjunction with the Australian army. We took it out into the middle of the Bush. Um, and this was at the top end of Australia in conditions of 40 degrees, which is what's that a hundred Fahrenheit, um, 90% humidity, um, put it in the middle of nowhere and then built parts in the middle of the Bush and, and the, the setup time, you know, they they'd put it on the back of trucks and drop it off the back of trucks. And there's some fantastic videos of them doing that by the way on that YouTube site. Um, and, um, we'd be up and running in about an hour and that's Testament to the, um, to the robustness and reliability of these systems that you can actually do it, you know, there's, there's no way you could put a laser machine out in the middle of the Bush, um, and, and have it operating with within an hour. Uh, and that's some of the real key markets for us as it is this, um, uh, parts on demand.

Will Herbert (09:00):
Yeah. Some of them are challenging enough in a nicely controlled, uh, lab or production environment. I agree with you. That's so, so that's interesting. So they're pretty rugged machines. And tell us a little bit about the sweet spot for the applications. I mean, what are you really trying to promote this towards in terms of the adopters and, and, and where are they taking these?

Byron Kennedy (09:21):
Yeah, yeah. So, so two distinct markets we're seeing, um, one of them is in, um, this, um, replacement of key components. Um, and, and you have to think of this a little differently. So when, you know, give you a prime example is working with the military. So we do a lot with the military, um, the military, uh, um, uses of expensive infrastructure. So when I say expensive infrastructure, I'm talking, um, you know, boats and ships and, and then, uh, vehicles and tanks, and on all of that hardware, um, that hardware is only as good as the weakest link. And, um, so, so the case studies we've done with the military are replacing parts, which are, um, which are not complex parts, interestingly, so 3d printing parts on about it having to be the complex parts and, and, and, you know, finding the case studies around the complex parts.

Byron Kennedy (10:20):
Um, that's not the reality in industries like the military and, um, the oil and gas industry and, and the mining or the rail industry. These are parts which have to be built then and there. So if you, you know, one of the examples which we use, and we got a few out there is, um, there's this camlock fitting. So it's essentially a hose fitting, like a relatively big hose fitting. Um, but it uses, it's used for transporting fresh water around a ship. And these things, corrode after time, as you could imagine. Um, and when they corrode, um, and, and then, then you can't have fresh water on the ship. So the ships are grounded. Um, and the challenge then is if they don't have the part, they then have to send a, a vessel over the toe, that one back, and then a third vessel then to protect.

Byron Kennedy (11:15):
So it's not one ship, it's three ships. So this $50 part, you know, it could end up costing, uh, the, the Navy, you know, $2 million in, in, in running costs. So they're the kind of parts which we're finding is very, very important in the real world. Um, and, and, you know, this is not about geometric optimization and over-engineering, it's about getting a part in, in an hour or two, and actually getting that vehicle and vessel running it. So it's not the cost of the part, it's the cost of the underlying infrastructure, which is key.

Will Herbert (11:54):
Yes, The downtime. Yeah. Sort of avoided avoiding that downtime. But it's curious, I was looking at that case study bar, I mean that you put $66 and then assume is that us dollars Australian? I don't know what the exchange rate is, but that's, that's still not a really expensive component. And it was about the size of that say grapefruit, is that right?

Byron Kennedy (12:16):
Okay. So, so in terms of costing, we, um, uh, our processes have better similar price point to a Casper. He had the same cost part, so, you know, uh, rough numbers here. So we, you know, we come from the manufacturing world and in the manufacturing world, you talk about dollars per kilogram. So 3d printing talks about dollars per CC, which I've never understood because what's the CC. Um, and, and you know, how, how much has that partway? So, um, the, um, it's all about dollars per kilos. So in the manufacturing world, so, uh, a, you know, steel, steel rod, Dollar a Kilo, you never going to compete with that. Um, uh, die-cast parts, you know, around the five to ten a kilo, and then you get sand cast parts, you know, from, from twenty to a hundred dollars a kilo. Um, and, and interestingly, if you look at that, it does then apply in the plastics world.

Byron Kennedy (13:10):
So there's, you know, low-cost plastics meeting cost by and high cost plastics. So really in manufacturing, you have to be in that sub one hundred dollars a kilo. Um, and that's really where this technology sits. So, you know, uh, we'll be producing parts for somewhere between 20 to $50, a kilo, um, aluminium, a bit more expensive, interestingly, because of the powder price being higher today, um, up towards that 50 to $100 kilos, and that's including full Amortization, the machines, labor, overhead, everything in there. So it's very easy then for us to go to a customer and say, right, aluminium top end a hundred dollars a kilo, how much does your partway that's that's then the cost of the part. So for that part, it would have been 660 grand part, I suspect, um, that's the $66 a kilo. So very, very simple numbers that you can say do actually have a business case for this part. Um, the way to do it is to white, and then you can work it out from there. So the challenge of course, with the laser technology it's around that thousand or $5,000 kilos, um, and, um, and really you have to be sub hundred dollars for any industry, any, any material, interestingly, you need to be sub a hundred dollars a kilo to hit production.

Will Herbert (14:26):
Yeah. It makes a lot of sense. I wasn't, I'm going to switch tack a little bit here. Um, Byron, and really what I'm interested in is seems like in Australia, there's some really unique types of technologies like yours coming out, um, some more game changing type stuff. Whereas we see in Europe and in the U S um, and parts of Asia, it's more sort of following the key themes of this laser powder bed and so on. I mean, what the scene down in Australia, what's the, the group of engineers and metallurgical people coming together to make this happen?

Byron Kennedy (15:00):
Yeah, it quite interesting. Um, but discussion with we've had colleagues, um, overseas as well about, um, how do we actually get, um, you know, new technologies into the market. It may be that in Australia, because we're a bit remote that we're willing to try a few things early on. Um, and, and we do see that. So even some of the European technologies come down to Australia and, and get them adopted in Australia first, before taking him back to mainland Europe. So, um, is that a cultural thing? Is that a geographical thing? I suspect it's, it's a little bit of both. It's the ability to try things down here and, you know, if it does not work in Australia, it's not a huge market, I suppose, um, to lose. Um, but, um, there is, uh, a, a, um, certainly a philosophy of trying things, uh, and, and getting them onto the market.

Byron Kennedy (15:54):
Um, but I want to extend that also to other regions like South America, um, Africa, um, you know, we, we've just got a pretty going into South America now, and that's all about supplying the South American market because, you know, the vast majority of parts are manufactured offshore. Um, that's, that's into El Salvador. So, so they kind of get a lot of parts. And, um, and so the market there is to being out to build those parts that, um, you know, they just kind of import from the us or Europe for whatever reason. So it's, it's a bit of this. Um, you needed that in necessity, um, rather than going through and ticking all the boxes. You know, you've got to tick all the boxes in many, many sectors still, but, um, you know, it's having the ability to say, yes, we'll give that a shot.

Will Herbert (16:40):
And does the, does the Australian government help with grants? And does it underwrite some of this early stage stuff that, that, uh, promotes people to take a few more risks? Is that part of it?

Byron Kennedy (16:52):
Yeah, absolutely. Like, we've been very, we've been supportive, um, tremendously by the Australian government. So, um, so there's lots of good programs in, in Australia for doing that. Um, you know, people I'd say, see, you know, some people complain about the government support, but it's, it's always there if you need it for us, at least, um, you know, if we needed it, we'd, um, go and look for, for the appropriate program and, and, um, you know, you've got to put the time and effort in to get the money, um, but it's usually available. So yeah, that's, that's been a big, um, uh, help for us.

Will Herbert (17:25):
And what other things are you seeing that, uh, are interesting? I mean, obviously we haven't been out at the trade shows and things where usually we get an early look at some of the, the new trends and new technologies coming out, and then you're probably like me desperate to get back to some of those. But, uh, what have you been taking note of what's been picking your interest?

Byron Kennedy (17:47):
Yeah. Um, we've been, um, we've, we've had an excellent year, interestingly, um, in 2020 on the back of the work we did with defense. And we're really building off that at the moment into the, uh, other defense sectors globally now, uh, in terms of technology where, where, you know, we got a few interesting things coming on board. We're doing a lot of work on scanning at present. So we've been working on 3d scanning for, uh, about two years now with, with one of the universities and that's to integrate a scanner directly into the printer. Um, our printer is a bit upside down in that. Um, we actually have a nozzle which is attached to the fore. We spray the material upwards, and then the robot, um, is above the spray head and the parts then build downwards, but what then enables us to do so the parts built on the end of the robot arm.

Byron Kennedy (18:42):
Um, what that enables us to do is we can then Mount a scanner separately inside the build chamber. You build a, you put the part on the end of the robot, and then we can automatically scan and stitch all of the, um, the resulting parts together in an automated fashion. It created what type mesh to what we can do is have a rip full, full, reverse engineering technology within this printer. And that's what we've got coming out. And we're doing some case studies around that at the moment where we've got some, um, obsolete parts, which, which, uh, customers come to us and said, did the drawings. And so what we do is we've mounted them, the robot arm we scan it and then we can rebuild that part. So, um, so then, then there's a lot of opportunities then in these heavy industries.

Byron Kennedy (19:35):
So, you know, another one to choose like rail where, you know, [inaudible], and the, um, the carts, you know, may run for 30 but 50 years and they go drawings of some of some of those parts. So instead we can then scan it and then build those replacement parts. So I think scanning's coming, um, you know, it's not all that it's cut out to be you, can't just, um, it's not a plug and play technology, so that's why we've had to work at it for two years, but, but that's certainly next on the horizon for us. And then increasingly build size of the printers as well as coming.

Will Herbert (20:12):
I see an increase in the build volume as well. Super cool. Um, is it, is it limited the softer metals and your technology? Can you do steels and, uh, let's say harder, higher temp, melting temperature metals as well.

Byron Kennedy (20:28):
What spray process, um, has been around for, for 10, 20 years, um, and in cold spray, they spray in the, in the academic arena, they spray alloy materials. They spray titanium. they spray Inconel, steel, stainless steel, um, you know, many, many materials you, we focused on the high volume applications, um, which is, um, aluminium and the coppers and stainless steels to date. Um, because, um, because we can push the price right down, we can actually compete on price with, with the traditional materials. So there's no, no real limitations. It's, it's more just engineering time and development that goes into that. Um, and, and for, yeah, new materials development is always a very large job. So when you take on a new material, um, you know, you've got to invest the time and effort into it. So, um, so we're selective in what materials we go after.

Will Herbert (21:28):
And what, on that note about development projects? I mean, what is a typical one look like? Because, you know, with these, these new technologies that, uh, at least new to a lot of the end customers who you're looking to to promote it too, I mean, they've never used it and it oftentimes takes a lot of work upfront. So what what, what, what are the typical steps that you have to take them through?

Byron Kennedy (21:50):
So, so there's, um, um, it depends on the industry. It depends on the customer to, to, to a degree, but the, you know, the general questions are how good is this material? So, you know, if you're not making good metal you might as well use plastic. Um, so first of all, you know, you need to prove that it's good metal, uh, which we do invest in the normal process, you know, tinsel bars and the normal test procedures that, that, um, everyone would go through. So that's always the first step and then its proving out the geometry and actually then building business cases. Um, as I said, if, if, um, if people come to us and say, I want 'em to replace my die-cast part, and, you know, I want you to be $5 a kilogram, we'll say, you know, go and die cuts because that's what you want. Um, but, um, if they're in that 50 to $100 range, then, then there is an opportunity for, for us to compete. So, yeah, there's two parts of it. There's the technical validation, which is, you know, relatively straightforward, it's the same for most customers. Plus then the business case.

Will Herbert (22:58):
And do you have a team who will do that sort of applications development work and help customers make the first steps, uh, before investing in a machine themselves? How does the business model that work?

Byron Kennedy (23:11):
And we, we the, uh, work with customers directly, or, um, we have service bureaus and contract manufacturers around the world that has their printer, and, um, they also work with us. So, so our business model at Speed 3d is actually the delivery of the software and the hardware, um, where, you know, we, we do work with our customers in delivering the, um, material data and tensil arsenal and all those sorts of things. But ultimately if people want parts, um, it's working with one of the contract manufacturers that has our printer technology.

Will Herbert (23:48):
I see. So you can hand it off to them. Interesting. So maybe we can zoom out, um, again, just look at that broader time horizon, because, uh, as we started off you, you telling us about the automotive and the axial flux motors that you were working on. And we're really seeing that now becoming a reality in terms of the EV market has just gone crazy this year. Um, give us your perspective on the next 10 years here in autotech, what, what are we going to see? What are you excited about?

Byron Kennedy (24:20):
The, the key to additive is to get into true manufacturing and true, true production. Um, so what is true production? True production is, um, products, um, you know, a single product at 10 to 20 ton per year. That's the kind of numbers, you know, we need to get to, uh, to be able to really call this production technology, like, uh, you know, w we w we're toying around the edges on editing at the moment we haven't got the volumes to be able to call it production. It's a niche production at best present. Yeah. There are some good examples of outside that with the hip cups and the teeth and those sorts of things. Um, but our vision is to take it into the real volume side of things. And that's, that's where we, you know, we're getting up into bed 10, 20 hundred tons a year of product.

Byron Kennedy (25:12):
So, so that's really the key. Um, that's different from our defense market, of course, because the defense guys are after specific replacement parts for quiet value. And, you know, they're never going to get to a hundred ton of a replacement parts. We, you wouldn't expect them to, um, but, um, that would ease the vision then, and that one takes longer. So, so that's where for our company, we see it as a two-step process project process in that, um, you know, we're working with many manufacturers, the Loma had, who had the same vision to get up into this, you know, 10, 20, 50, a hundred ton a year product. Um, and then, um, but, um, to get there, you know, you need to take baby steps and the obsolete parts of the ones we're doing to start with as we work on those business cases. And just to give you a, a comparative benchmark, um, one of our printers can print 'em in a normal two shift operation about 25 ton of material per year. Um, so, you know, to get to a hundred times it's only four printers, um, yeah, that that's never the case in the laser technology. You know, I think they're about the 250 kilos, mark, per year of material actually printed. Um, so, you know, even the hit, um, twenty-five ton, I think that's a hundred printers, so, um, and the economics weren't work that sort of volume.

Will Herbert (26:45):
Fantastic. Well, Byron, uh, thank you very much for joining us. Uh, unfortunately we're out of time, we could talk about these topics all day, um, but best of luck with your new technologies. It's really interesting.

OUTRO (27:03):
Thanks very much to Byron Kennedy for sitting in on a PowderHeads episode, getting his perspective on AM and hearing about the user centric approach that his company is taking is notable. He's clearly bullish and optimistic about his business and where the industry's going. And that in our definition is a true powder head. If you have questions or comments about what we discussed in this podcast to PowderHeads, send them to powderheads@carpenteradditive.com or visit our podcast page at www.carpenteradditive.com/powderheads. We'll be building an archive of all 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 customers solve their most challenging material and process problems. Learn more at CarpenterTechology.com. Thanks again for listening and keep building.