Additive will provide a simpler, more responsive supply chain for high-value parts, according to Velo3D CEO Benny Buller.
KIP HANSON, Smart Manufacturing: Benny, what prompted you to start Velo3D in 2014?
Benny Buller: Before 2014, I was an investor for two years, and I vowed never to invest in 3D printing. When everyone told me, “3D printing is this great technology. You could do whatever you need. The only thing that needs to improve is the cost.” I had seen that when you are in a race to reduce cost, it’s a race to the bottom. No one makes money in this race.
What changed my mind was a rocket company we’d invested in that decided to design their engines around 3D printing as a manufacturing technology. What they found is, it took them three-plus months to get any parts made because they had to iterate on those parts and change the design and kind of incorporate more and more intrusive DfAM [designing for additive manufacturing] to make them manufacturable, compromising part performance, until they got something that was both manufacturable and acceptable from performance level.
That surprised me. Then I started to talk with other people who were using the same technology, and I saw that this was universal. They were designing the parts and changing them in a way that would be manufacturable, but it wasn’t what they really needed.
So, I looked at that and I wondered, “What if we could solve that? What if we could eliminate these design constraints or these manufacturing constraints? That will open the design space.”
And the answer from others was, “This would be the most disruptive thing that happened to 3D printing—but don’t worry about it. It’s completely impossible.” And I thought, “Okay, I don’t know how to do that yet, but I’ll commit the next decade of my life to it.”
HANSON: Tell us about the growth of Velo3D over the last couple of years and about your expansion into Europe recently. What that might mean to your competitors?
Buller: Fundamentally, we are going after manufacturing customers that do not buy one system, they buy tens of systems. But they always start with one, right? Then they will go to two or three. Then five and six, and gradually they’ll build their fleet. So this is a land-and-expand strategy. So, the first layer of our commercial model is expanding the customer base.
The second layer is the layer of systems that we are shipping. We ship systems to new customers every year, but each one of our existing customers is continuing to buy systems. We have already observed that, on average, each of our existing customers is buying between one and one-and-a-half systems a year. As we go forward, you can expect our existing customer base to keep adding machines at this rate, and the new customers will keep adding machines in future years as well.
The third aspect is that each machine that is operating in the field in the first year of shipment, we recognize revenue from the sale of this machine, and then there is recurring revenue that is coming from this machine as it’s operating in future years. All of that is a revenue expansion that is very organic.
So you asked the question about Europe. The vast majority of our sales so far have been in the United States. We just sold our first system in Europe, and in December of 2020 we hired our first salesperson in Europe. We are now creating a headquarters for our European operation. Basically, we are going to replicate in Europe the sales effort that we have in the United States. It’s a direct sales effort with a very strong technical sales team and a very strong field service and application support team for the customers that will be there. We estimate the European market to be about the same size as the U.S. market.
And you asked, what does it mean to our competitors, in particular our European competitors? I think the answer is, it doesn’t mean much to them. The applications we are focusing on are those with requirements that our customers have found can’t be met with our competitors’ systems. We are pricing our systems intentionally to be premium-priced compared to our competitors. And we are actively discouraging our people from going after opportunities where our competitors can provide a solution. Our competitors, between them, have more than 5,000 systems that they’ve deployed in the field. If someone can do what they need to do with a competitor’s systems, they will do that in the end, and I encourage them very actively to do that and not try to work with us. We are focusing on the things that other people cannot do.
HANSON: Let’s talk about open versus closed systems. Some manufacturers continue to keep the operating parameters under lock and key. They only offer proprietary materials and so on. What are your thoughts on that?
Buller: When you look at powder-bed fusion, you can get closed parameters from different suppliers, but all of the suppliers today would allow you to buy a software package that will allow you to open the parameters. Whether it’s EOS or GE or SLM Solutions or Trumpf or Renishaw, you can develop parameters for any of those systems.
When people talk about closed and open systems, Velo3D is as closed as it can get, right? One thing that is really important to understand is that if you are printing with Inconel 718, as an example of a material, when someone takes a CAD file and creates from that a print file, this print file is completely independent on the [Velo3D] Sapphire machine that it’s going to be running on. It’s a universal print file for this specific part with this specific alloy, that now can be run on any Sapphire machine in the world, getting exactly the same outcome.
And so from that perspective, yes, it’s a closed system, but it’s a closed system that has been built to generate a very specific outcome. And the outcome is that when you have a build file, this build file will reproduce in any place in the world, on any machine, at any time, to exactly the same outcome. And this is an end-to-end solution provided to customers that want manufacturing, not people that want to play and develop parameters.
HANSON: Do you think that the industry is closer to the Holy Grail of serial production of metal parts? Do you see that happening?
Buller: When you talk serial production, you automatically think about the automotive market or commercial products, and the like. A lot of people, when they think serial production, they think BMW, VW, right? And my answer on that is, absolutely not. They may be doing some completely negligible things that have no business impact before that, but I will be dead before BMW and VW will be in significantly high-impact serial production with additive manufacturing.
And the reason for that is, when you look at products that are produced in volumes of tens of thousands a month, the designers design them in such a way that you could use very low value add manufacturing technology to make those parts. Those parts are made in such a way that their cost is mostly represented by the raw material cost that goes into them. And they use very low-cost raw materials, right? So these are very, very low-value parts.
I think the opportunity for 3D printing is in the market of high-value parts, and you find those high-value parts in industrial equipment, in aerospace, in energy, in power generation, in medical devices and medical equipment. You don’t find them in consumer products.
And those systems are manufacturing in volumes of tens, rather than tens of thousands, a month. Such high-value parts represent a $100 billion market today. It could become an almost $200 billion market by the end of this decade. And this is where metal additive manufacturing is going to play a very big role in serial production. We see this starting to happen today, and I think that in this coming decade we’ll see a tremendous acceleration of the growth in this segment.
HANSON: Other than laser powder-bed, what other forms of metal 3D printing do you think are promising?
Buller: I think that laser powder-bed is going to cement its position as the dominant additive manufacturing technology for high value parts. I am looking at the area of DED [directed energy deposition], and I haven’t made up my mind about it yet because it’s possible that near-net shape manufacturing with DED has a bright future as well. I’m not sure about that. And the reason why I say that is because you have a lot of capabilities that have been out there for a really long time, and yet we saw very little adoption of that in the real world. And so I need to figure out if there is a fundamental reason for that. The verdict is open.
HANSON: I believe you once stated that contract manufacturers are better suited than OEMs to advance metal additive. Why is that?
Buller: If you look at the last 50 years, what we have seen in every aspect of our economy is a trend of specialization. Every organization is doing what they are expert in, and they are outsourcing to other organizations the other parts. This means more and more complex supply chains that are more and more specialized and skilled. I don’t see any reason why this would be any different for additive. I think this trend is irrefutable and unstoppable. And at a micro level, what you can see is that contract manufacturers are much more nimble than OEMs. They can build the capacity much faster, and they have much less red tape. So, if an OEM wants to make parts by additive, it’s much less hassle for them to go and get parts, than to get all aspects of the organization aligned on buying a multimillion-dollar machine and so on.
And very often, when an OEM buys a machine, it initially gets only 20, 30 percent utilization. It’s a very expensive offering for the industry. When you build this capacity in contract manufacturing, it’s much better utilized.
But there is another corollary to that. If the capacity is highly utilized, these contract manufacturers get to build skills much faster because they use their technology much more. So, they are able to deliver much better products, much more quickly, much more efficiently than an OEM doing it by itself.
And the last thing is data. When we are engaging with customers, we see a ratio of about 10 to one between OEMs that want to become vertically integrated and OEMs that want to buy parts. For every OEM that wants to buy a system, there are 10 OEMs that just want parts.
HANSON: Can you say more about how you expect additive manufacturers to disrupt the industry?
Buller: I think what we are seeing now is an emergence of a new contract manufacturing supply chain. This supply chain will create a foundation on which companies that today are relying on a very archaic and nonresponsive supply chain, one that has lead times measured in quarters or in years, is replaced by a supply chain that can provide parts in basically batches of one if needed, within days or very few weeks. That changes the whole way most companies operate. That fundamentally changes how you plan, how you provide inventory and how you provide service to your customers. It fundamentally changes the product lifecycle.
On top of that, we are seeing how this new capability allows companies to develop new products that were not possible before. A lot of our customers take advantage of this technology to really design disruptive products, whether it’s in space exploration, flight, power generation or elsewhere.
HANSON: What other impacts do you expect AM to have, globally?
Buller: We are going to see much more localization. We are already talking with customers about the need for a contract manufacturing supply chain to be present in specific locations, whether it is in Southeast Asia or the Middle East or Alaska. I think that we are going to see a lot of local manufacturing in emerging economies that would be able to leapfrog the old manufacturing infrastructure. Very much like how emerging economies skipped the land line and went directly to mobile [phones]. We will see much more diversification. And we will see a much more agile supply chain.
But I don’t think that because of additive manufacturing, people will elect to become their own suppliers. Additive is still a highly skilled, high-stakes manufacturing method that needs to be done well, and you want to rely on people that are doing it at scale and at quality and with culpability. So, I think the result is diversity and localization, coupled with specialization.