Revenue reached $20.7 million for the quarter ended June 30, up 52.3% from $13.6 million a year earlier. The company also returned to a positive gross margin, reporting 21.5% compared with negative 11.7% in the second quarter of 2025.

The improvement was driven largely by Velo3D’s core printer and parts business. Revenue from 3D printers and parts rose 57% to $19 million from $12.1 million a year earlier. Velo3D attributed the increase to higher average selling prices, product mix, and higher revenue from its Rapid Production Solutions (RPS) parts production business.

The company expects system sales to remain its main source of revenue this year, but said RPS is expected to account for a larger share of the business under its new go-to-market strategy.

Gross margin also improved. Velo3D generated $4.4 million in gross profit during the quarter, compared with a $1.6 million gross loss a year earlier. The company said the improvement reflected higher selling prices, a more favorable product mix, increased RPS revenue and manufacturing efficiencies. The improvement also reflected a change in how certain labor and overhead costs were allocated between production costs and operating expenses.

Velo3D still reported a loss, but it was smaller than a year ago. GAAP net loss was $11.5 million, or 39 cents per share, compared with a $13.3 million loss, or 94 cents per share, in the same period last year. Adjusted net loss improved to $9 million from $11.4 million, while adjusted EBITDA was negative $8.1 million compared with negative $8.9 million a year earlier. At the same time, expenses increased as Velo3D continued investing in the business. Operating expenses reached $15.5 million, up from $10 million a year earlier.

More Cash, Less Debt

Velo3D ended June with much more cash than it had at the start of the year. In fact, cash and cash equivalents stood at $91.1 million as of June 30, compared with $39 million at the end of 2025. The increase came primarily from financing rather than operations. Velo3D generated about $99.5 million in net cash from financing activities during the first six months of the year, while using roughly $39.5 million in cash for operations.

In April, the company sold some 3.6 million shares in a registered direct offering, raising about $50 million in gross proceeds and $46.6 million after issuance costs. It raised another $59.4 million through its at-the-market stock offering program during the second quarter, or roughly $57.4 million after issuance costs. Velo3D raised much of that cash by issuing shares. Along with debt-to-equity conversions, the company reduced its total debt by more than 70% to $8.2 million at the end of June.

CFO Jim Suva pointed out that “With approximately $91 million in cash and cash equivalents at quarter end, Velo3D has greater financial flexibility to execute our growth strategy and support capacity expansion, technology development and customer programs while maintaining a disciplined approach to capital allocation. Combined with our significantly reduced debt, we believe our strengthened balance sheet supports our ability to execute our strategic initiatives, scale our operations and capitalize on the growing demand for advanced metal additive manufacturing solutions across the aerospace, defense, energy and space markets.”

Growing Orders

New orders reached $29 million in the second quarter, while Velo3D ended June with a $31 million backlog. The business is also expanding its manufacturing capacity. It recently announced a new production campus in Livermore, California, which it expects to become operational later this year. The site is expected to triple the company’s manufacturing capacity and become its primary production center. The expansion also supports Velo3D’s RPS business, which produces parts directly for customers instead of selling them 3D printers.

Partnerships also grew during the quarter. Mears Machine Corporation ordered its fifth Sapphire XC system, with options for two more, while a new partnership with Aurelia Technologies will focus on metal AM for next-generation gas turbine systems.

VELO3D Sapphire XC Metal 3D Printer. Image courtesy of VELO3D.

Velo3D raised its full-year revenue forecast to between $65 million and $75 million, up from its previous range of $60 million to $70 million. The rest of its 2026 guidance remains unchanged. Velo3D expects gross margin to exceed 30% in the second half of the year. Adjusted operating expenses are expected to be between $45 million and $55 million, while capital spending is expected to reach $40 million to $50 million, mainly to expand its RPS business. The spending will depend on available financing, explained management.

Velo3D is still targeting positive adjusted EBITDA in the second half of 2026. For the first six months of the year, revenue reached $34.5 million, up from $22.9 million a year earlier. Its first-half net loss narrowed to $18.5 million from $38.3 million. The company now enters the second half of the year with higher revenue guidance and plans to expand production capacity.

But, this does not necessarily mean that there has been a change in the goals of the project. Whereas a few years ago everyone bandied about the term “hypersonics” like it was some secret code word to the magical treehouse, now the in-crowd has gotten very tight-lipped about the technology. It’s much more of a “those who know” kind of a thing. We’re in the “Quiet Hypersonics” phase. Or the third wave of hypersonics.

The first wave of hypersonics was the uni demonstrators phase, when most of the efforts could have doubled as some of the world’s most high tech kayaks, hollow like Casanova’s late Friday night promises via text. The stuff that did work did so for 30 seconds; kind of a government-sponsored, off-season fireworks show of burning Nomex, glass, and composites. Then came the leap phase, in which hypersonics where everywhere…save for in the sky. Now we’ve got the Area 51 phase, with everyone pretending to be on a shuttle to Groom Lake, White Sands, or Yuma every week. But, for those jockeying for position, the Last Train to Yuma has already left the station. I swear that guys at MILAM will rub sand on their shoes to seem legit and take their kids to Cape Canaveral for the third time just to get the appropriate mosquito bites to fit in.

Design optimized propulsion system component made with DMP. Image courtesy of 3D Systems.

And 3D Systems has kept mum on this project so far, so that’s wonderful and keeping with what the government wants now. Hypersonics went from being a cool science experiment to the cutting edge of any future engagement with China. So good work on being tight-lipped; in fact, they were so tight-lipped, I thought this project had been cancelled. So this is very good news for 3D Systems, since, as we stated in our strategic overview article for the firm, they could use more exposure from the magical Gravy Train that is the US government.

The company mentions that the GEN-II DMP-1000 program is dealing with high-speed printing of high-temperature materials, and seems to be centered around a 1000 x 1000 x 1000mm build volume DMP system.

Dr. Michael Shepard, Vice President, Aerospace & Defense at 3D Systems, said,

“This award increases total project funding to $27.4 million and reflects the U.S. Air Force’s continued confidence in our technology, its development path, and our teams. It has been exciting to see the system come together, and we look forward to working with additional partners on new applications as the platform matures.”

We are creeping towards hypersonics, and the victory could be an important one. For the polymer skins and internal metal components, additive is sure to play a part. Making everything compact is extra important with a vehicle that needs to survive itself and where every gram of weight matters. Having said that, the budgetary picture is hazy. Without much of an explanation, budgetary requests are down from $6.9 billion to $3.9 billion in a year. There is no clear end point for hypersonics, no clear goal apart from total dominance, but no defined path to get there. From a recent report, “The United States has neither deployed an operational hypersonic boost-glide system nor provided Congress with a clear funding trajectory for doing so, while Russia and China have moved past the development phase into deployment.” I would, by the way, disagree that Russia is deploying hypersonics. It’s using rockets that can go at hypersonic speeds in their terminal phases, but not have the varied flight patterns that China and the US are working on which will make them so difficult to predict. But, the budgetary tea leaving does point to a lack of focus.

Industrial Gas Turbine component built with Certified HX parameters, manufactured on the DMP Factory 500 (Image courtesy of GF Casting Solutions).

There does not seem to be the political will at the moment to really make America win at hypersonics. So the technical folks are making do and trying to get as far ahead as they can. Meanwhile, money is being diverted to exigent missile production for current US needs. And the “take all your money” Golden Dome plot is being readied to become a vehicle to make us all safe through a careful appropriation to those specifically close to the administration. Now I’m not saying that a great big Maginot line in the sky isn’t a bad idea, it’s just that the US will need to win in hypersonics to sufficiently counter Chinese advances and deter hypersonic launches. Even if the Golden Dome could protect the continental US, what of US forces overseas? Without an offensive and defensive hypersonics capability for them, the US will not be able to really conduct any overseas operations well.

So 3D Systems is making progress on a real need. Possibly the long term growth in hypersonics will give the machine they’re working on a lot of custom. But, in getting the Air Force to help them develop a new larger system ideal for rockets and defense, the company is getting a real chance to grow in aerospace and defense more broadly.

Meanwhile, AM also played a role in multiple other projects during the two exercises, warranting some follow-up discussion on both. First, there was Manufacturing Attritable Systems at Scale (MASS), an effort revolving around 3D printed vessels, both manned and unmanned. This was part of a showcase called ‘Print on the Move,’ the point of which is exactly what it sounds like: expeditionary manufacturing.

Marines, aided by participants from Naval Surface Warfare Center Carderock (NSWCC), used a large-format polymer 3D printer and a spray foam and coating system to produce multiple versions of a seven-meter raiding craft. The NSWCC team leveraged their experience making a similar five-meter hull, which initially took 16 hours, using $11,000 of materials, making the process both cheaper and faster than conventional methods.

According to Marissa Stecko, an engineer at NSWCC, “We have reason to believe that, once fully scaled, we could manufacture a 5-meter platform every four hours with a team of 12, and a 7-meter platform every six hours. We’re still in the prototyping phase, but there’s a tremendous amount of potential.”

They saved the most ambitious phase of MASS for last: the team started a print en route to the White Beach Naval Facility in Japan, then transferred the printer to the 3D printed vessel, where production of the component continued while in motion.

As I noted in a post from a few months ago, the US military is quite intent on using 3D printing to showcase expeditionary manufacturing capabilities specifically in the Indo-Pacific, with the subtext there being self-explanatory. One symptom of the military’s prioritization of this objective is accelerated adoption of containerized manufacturing systems, and Firestorm Labs is front-and-center in that context.

Firestorm’s xCell being lifted onto the USS Essex. Image courtesy of Firestorm.

Firestorm demonstrated its unique value proposition yet again on the way to RIMPAC, where representatives from the company used the xCell containerized manufacturing unit to train service-members from the Army, Navy, and Marine Corps how to print parts on the move aboard the USS Essex. Despite this being the first time that the xCell was demonstrated at sea, they managed to print more than 1,000 parts over the course of two weeks, including a dozen of Firestorm’s Squall FPV drones. Once the ship arrived in Hawaii, Marines flew the Squalls as part of a counter-UAS exercise, providing striking validation of what Firestorm’s ecosystem can do.

Firestorm’s CEO Dan Magy, said, “The Indo-Pacific logistics problem is real, and most of the answers on offer live in a pitch deck. We printed flight-ready aircraft and the parts a crew actually needed, in the conditions they’d actually face. That’s what manufacturing at the edge looks like when it works.”

This doesn’t even cover all of the 3D printing activity at both events, not to mention other events the US military has engaged in over the summer, but these examples seem to be most representative of (1) the things the Pentagon is currently trying to acquire more of and (2) how it’s using AM to speed up the process of getting those things. The US military services and the companies they’ve been working with to build up organic 3D printing capacity — mostly small startups — have shown consistent progress at these sorts of events over the last several years, but the activity this summer implies that all those efforts have started to hit critical mass.

If you compare this to RIMPAC 2022, for instance, that was the first time that the US Navy demonstrated 3D printing at sea aboard an operational vessel, also the USS Essex. Evolving from there to printing over 1,000 functional parts over a span two weeks, on a completely different system, just four years later, makes a statement.

Similarly, four years ago, maritime drones were barely on anyone’s radar. Already though, US service members aren’t merely printing different iterations of watercraft that can be modified between serving as manned and unmanned systems. They’re actually printing components in those vessels while they’re operational.

This establishes a pretty high bar for success. The next challenge is to achieve all of the same capabilities, at scale.

Featured image courtesy of DVIDS

With regard to the filament market, the beautifully made slide above seems amazing. But I think that this would be an incorrect conclusion to draw. One major area of focus is the whole idea that this is a “razor and blades” model, a common trope in the business world. Now, I don’t think that is where the real power of Bambu and the others lies. I think that a software-experience-driven company that gathers more accurate usage data and, in effect, digitizes the Material Extrusion process while making it easier and more reliable to print at low cost by using sensors, images, and LIDAR, is more than a razor-and-blade play.

To me, ease of use, reliability, and low cost were the real reasons to adopt well-working Bambu units. The company then uses its access to the entire toolchain of data to get better printing results, not only for its filaments but for all filaments. It then becomes the first billion-dollar revenue company in 3D printing within five years. That’s around one-twentieth of the entire rest of the market, which took billions in VC and other investor capital and 40 years to grow to that point. It used sensors and software to solve a problem that everyone else was trying to solve via mechanical engineering. And whereas you had printers that worked for $5,000, they had one that works just as well for $1,000, then $500, and now $200. The firm then rolled out all of its innovations across its many printers, experimented with models, and pushed out hardware, sensor, and software updates to all models.

To me, this is obviously a commoditization play. Bambu wants to be the creation gateway between the digital and physical worlds. It wants to own the idea to the product pipeline for many or anyone. By building an app, desktop software, various machines, file sharing, AI creation, customization, filament, firmware, sensor, and data platform, it wants to, in effect, be a Microsoft Office for making stuff. Or, if Google is the operating system for the digital world, Bambu wants to be the OS for the physical world.

To me, the razor blade stuff is too simple, too stupid, too short-sighted. And it won’t really work when you’re targeting a community of engineers and inventors. Imagine if you wanted to be the richest person in the world. Is making a Nespresso for stuff going to cut it?

Imagine if you had to compete with Meituan, a multi-billion-dollar revenue delivery service that invested in Snapmaker, and cars to vacuum cleaner firm Xiaomi, which has its own 3D printer and also invested in Snapmaker, as well as DJI that invested in Elegoo, and on the side, you kind of have to compete with your frenemy at Tencent who invested in Creality. This is a completely insane situation to be in. Tencent has revenues of $107 billion, Xiaomi $64 billion, Meituan $51 billion, and DJI only has $11 billion. And do you really think that those guys care about a $20 billion revenue industry? Or that these firms are super excited about “it’s like Nespresso but for stuff”? The play is much more fundamental. Filament revenue is a way to fund the machines (as is revenue from the machines), and the machines provide data to the software, which gets better, which lets people make more things better, which drives filament sales and more files, and that’s how the autocatalytic cycle works. This could potentially be a gateway architecture like the phone or the PC was. So, to me, the pathway is to sustain growth in order to power functionality and make the entire platform more frictionless.

Manual Slicer Config

Now, putting in slicer configuration and dialing in filament did suck, don’t get me wrong. But I didn’t crawl under my bed to hide from it. It was annoying and made you less adventurous with choosing different filaments. Now I can pop pretty much any filament in a Bambu, and it works well. If there’s no RFID I can just select if from a menu. A few times I’ve not found it, but the Bambu does a great job of printing it anyway. The generic settings on Bambu systems are killer. That all doesn’t sound very razor and blades-like to me. If anything, it’s a convenience play for new users that keeps them loyal. Nespresso does not allow you to put other people’s coffee in their machines. Emphatically here, I think that razor and blades is completely not what is going on in 3D printing. This is a completely wrong conclusion by our AI tools. To me, the bigger, more correct, picture is in the gateway strategy.

And it’s a little more frictionless to put Bambu filament into the AMS, but not that much really. And I don’t even see it as a real pain now. So to me, that’s not the real driver here. I also don’t agree that pure mechanical performance is what’s happening here. Bambu is software-driven. And it’s the ease and reliability of the overall system that is driving sales into new markets. All the new folks aren’t buying this for 400mm/s or whatever; they don’t even know what that means. They buy it because this is a 3D printer that they can use, and they believe it because they’ve seen enough other people talk about it or use it.

Model T

If you want to annoy people at car companies, ask them why, if they’re so focused on their product, don’t they outsource everything except making the tires? After all, that clearly it is their core business. A car company will typically sell four tires for every car it sells. They’re clearly tire companies masquerading as car companies. And I could totally imagine an AI analyzing Ford and concluding that the company was clearly entering into a razor-and-blade play. It would make these super-inexpensive automobiles to sell tires. And indeed, initially, Ford did make its own tires. So it may have been a very obvious thing to think about. Now, surrounded as we are by disposable products and examples of razors and blades, we have a selective perception bias.

But, if you want to transform the world, I think you’re going to have to build bigger moats and do better in this day and age. And if you looked under the hood at Ford, you might come to another conclusion and think that the mechanization, rote work, and assembly line were the real technology. Or you may think that the car’s serviceability and simplicity drove its adoption. It may be due to the fact that it was affordable and very tough. This point is overlooked; it was a car for places without roads or without good roads, not just another car for Manhattan or London. The Model T also quickly became a core for delivery trucks, ice trucks, and vans. So perhaps it’s the vehicle’s versatility that led to its success. Lighter steel and better shifting may also have helped. You may think of the Model T as a volume play or a commoditization.

To me, Ford’s transformational success was the result of all of those things, expressed most clearly in its prices, which were $850 in 1908 and $260 by 1925. For the same car! Ford’s $5-a-day pay increased salaries so his workers could afford the car with three months’ wages. To me, the Model T is a value innovation play that transformed the world. There were many cars before Ford, and many car companies, too. Now there was a car that, through the sum total of its innovations, made motorized transport practical and affordable. And it did so everywhere. Affordable personal transport was just an idea before the Model T and became a market after it. To profoundly reimagine the world as it could be, and to then imagine the product that could make it so, is already a magical thing. But, to then build a business that actually delivers on this while making the product better and cheaper all the time, that’s a revolution. And I think that this is what is happening to the filament market.

Notes on AI Research

On these pages you can see some completely entertaining images. These illustrate my points, and the only thing I had to do was feed my article into Notebook. I didn’t tweak anything or mess about with it; it ran in the background. So this is pretty powerful stuff. On the whole however, for now it’s very much a scary tool to use AI for research. I’ve been testing this extensively for a number of months now. And even with capabilities increasing, a lot of danger still remains. I’m no AI naysayer at this point, but I’m also not a cheerleader. I’m trying to figure out where it makes sense to use it. I’m more of a “It sure is a nice circular saw, it cuts well, but should we be using it to cut hair and mow the lawn?” kind of guy.

Chainsaw Moments

These are some of the biggest AI errors I’ve found so far over the past half year:

• AI crawlers and analysis tools routinely miss big trends, large datasets, and major data sources.
• They don’t know they miss them, so they infer a lot from their not being there, and this leads to huge errors.
• These tools infer a lot from very little. Often, a lot of grounded opinion or a basis of knowledge is built on three opinions or one sentiment repeated.
• AI is super good at categorizing and classifying things, but it’s not too good at super large datasets.
• A lot of data is nonetheless classified or interpreted incorrectly due to incorrect technical understanding, which surprised me.
• It sometimes seems as if it “latches on” to a key idea and then won’t let go, even if it’s incorrect.
• AI tools are often sycophantic and reconfirm any and all biases that you may have.
• Initial queries, data, or classes can really screw up later information.
• Old data is often passed off as new.
• Often, LLMs conflate or mix together two completely foreign concepts into one.
• AI tools seem to jump to conclusions a lot.
• When an authoritative source is wrong, and their lie is retold, it’s hard for the AI to find it or root it out.
• AI tools often try to make things slick and spiffy, which causes an enormous number of errors.
• There appears to be bias in the information entered initially or towards the top of the datasets.
• AI´s images or presentation can be very good, fast, and entertaining, which often hides faults.
• Eg data is not meant for study or interpretation but to convince.
• Much of the imagery, tables, and graphs can appear convincing, but obscure the truth because it seems holistic or logical.
• Generally, it presents itself as more knowledgeable and presents its conclusions more forcefully than the data often allows.

So these are some of the issues that I encountered. I would totally recommend using AI tools for research in areas that you understand, for general office work, and for creating images, graphs, and other materials. For brainstorming, making new connections, and rechecking things, it’s great. But it’s not suited for learning new things, new subjects, or understanding them well. Yes, there will be very powerful business tools created with AI. But I really do think that some people are going to completely gut their companies and blow their careers by staring blithely into these things and steering blindly by them. Already, we can see research at competitors degrade as they use more AI without sufficient safeguards. There’s such a font of information there, and it appears so quickly that you build up trust with every query. Rechecking data, cleaning up your datasets, eliminating bad data, and checking key assumptions becomes more important the less familiar you are with a subject. The mistakes are sometimes subtle, and sometimes embarrassingly obvious. So, to re-collate, present, and see connections in that which you know, it can be a great tool. But, for now at least, I’d stay away from learning anything new with it, so as not to make you learn the wrong things.

The team consisted of Jin-Hyeong Lee, Kyeong Pyo Kim, Lijie Ding, Michael Li, Min Chan Kim, Kyu Hyun, Ji Hoon Kim, Jan-Michael Y. Carrillo, and Suk-kyun Ahn, all from Pusan’s School of Chemical Engineering and School of Mechanical Engineering. On the ORNL team, researchers joined from the Neutron Scattering Division & Center for Nanophase Materials Sciences. So here we’re seeing a multinational team, from very different disciplines, working together to make a lot of candle power shine brighter. It’s at this kind of interdisciplinary frontier that we expect to find major advances in our industry.

One of the key ingredients here is smectic ink; this is a Liquid Crystal Elastomeric ink that, through differences in print speeds and temperature, can switch the alignment of its molecules. Usually in liquid crystal inks, the smectic phase occurs when molecules are neatly aligned in direction and layers. This is in contrast to the nematic phase, where molecules are directionally aligned by interspersed, and the cholesteric phase, where they’re kind of fanned out haphazardly. To remember this: smectic is ladyfingers in a box, nematic is ladyfingers assembled into tiramisu, and cholesteric is I threw the tiramisu. So what they’ve done here is made an ink that allows you to repeatedly switch between these states by changing process variables such as print temperature and speed. That then could let you make a part where portions of it can elongate or contract.

So more so than a paper, they’ve shown a technology that can be used by many other researchers to make all sorts of interesting shapes and properties. Indeed, it’s easy to see how this can lead to the development of machines, microfluidics, and optoelectronic components. This is partially because Liquid Crystal Polymers are super weird. They’re kind of like glass in the sense that it’s difficult to tell if it’s like a very bad liquid or like a terrible solid, depending on how you look at it. But glass, of course, is super usable, changeable, and useful because of its changes of state and its weirdness. Liquid Crystal Polymers are kind of like that in a sense, and have very orderly crystalline structures when melting, different states, and low melt viscosity. This means that you could turn them into many things, including thin and complex structures. At the same time, these parts are inherently flame-resistant, have high strength, and good continuous service temperatures of around 240°C. That’s why we’ve been so enthusiastic about NematX, which made a dream of a printer to print these materials. The reason for all the marble and precise control on NematX’s system is precisely the crux of these materials; while the properties are inspiring, printing these materials day to day is not fun.

So it’s easy to see why making small machines out of these materials would be immensely valuable. Think of the sensor applications, the missile parts, the optoelectronics, the miniaturized aircraft components, the connectors, the drone parts, the circuit boards, and the antenna. It’s almost as if a whole imagined future of making integrated robots out of one material runs through this material, or this material coupled with copper or similar. Just looking at electronics, making lightweight, dimensionally accurate, thin-walled antennas for phones could be a huge application, and there’s much more besides. Implants, wearable sensors, implanted sensors, and defense could all be impacted by this, as could a new wave of optoelectronic devices. By changing molecular alignment during the print, lots of complex properties can be made. What’s more, this direct ink writing technique could be reproducible and work at scale. To do it, the team worked with X-ray scattering, as well as simulation and rheology tools to determine when and how they could make the material switch and switch back.

Professor Ahn stated,

“Our work provides the first demonstration of switching molecular alignment between two orthogonal directions using a single 3D-printable smectic LCE ink, simply by tuning the printing speed and temperature. Potential real-life applications include soft robotic actuators and artificial muscles, reconfigurable surfaces for haptic displays, and adaptive textures that regulate aerodynamic  drag. Over the next 5–10 years, this work could help 3D-printed objects go beyond just holding a fixed shape. Instead, they could actively change shape and carry out specific functions.”

The team has demonstrated not only repeatable switching, but also has done so with specific lattice and conformal shapes. It’s easy to see how, if this works, they could develop a technology to make complex assemblies that can move, change, and work at a very small scale while having high performance. This could be a very interesting technology to watch, as it could enable more researchers to do more.

While that presents a massive economic challenge, it also represents a big opportunity for the additive manufacturing (AM) industry to demonstrate that AM can help improve efficiency in production workflows that depend on copper. That’s true, moreover, for areas of the AM value chain utilizing the lowest-cost desktop machines, not only for users of industrial systems targeting serial production. For instance, Indiana-based Kupros, Inc. makes an all-metal, conductive filament called Cu29, which can be printed on consumer desktop FFF/FDM machines. Kupros also provides services supporting users’ adoption of the material.

Aiming to lower the barrier-to-entry for electronics 3D printing, Kupros just launched Cu29 V2, the second-generation of the product, with an initial batch of 90 kilograms that already looks to be sold out. Meanwhile, the company is also developing a patent-pending workflow for expanding use of Cu29 beyond printing conductive traces. Kupros is going after even more complex applications, including embedded electronics, conformal antennas, sensors, and really any use case where AM seems like a good fit for electronics. According to Kupros, it has sent “Cu29-enabled sensor test articles” to Oak Ridge National Laboratory (ORNL), which will evaluate the results.

If you’re attending Fed Supernova 2026 (Austin, Texas, August 18-20) or J-DAMMIT 2026 (Harrisburg, Pennsylvania, August 25-27), you can learn more about Kupros and Cu29 in person. You can also read an interview I did last year with the company’s founder, Ian Ramsdell, a veteran Navy officer.

In a press release about Kupros, Inc.’s release of Cu29 V2, as well as its ongoing work on a patent-pending workflow for 3D printed electronics on desktop FFF and FDM printers, Ramsdell, who also serves as the company’s CEO, said, “V1 proved that this was possible,” Cu29 V2 is about making it easier for engineers to actually deploy. We spent the last year printing, breaking things, rebuilding them, listening to our early adopters, and understanding where the workflow created friction. V2 represents everything we learned from putting the first generation into the hands of real users.

“[The new workflow] is where the technology becomes much bigger than filament. Our objective has always been to change how electronics are manufactured. If you can introduce conductive pathways and electronic components directly during a standard FDM/FFF workflow, the printer stops being only a mechanical manufacturing system. It begins becoming an electronics manufacturing platform.” 

I recently wrote about how the US Army DEVCOM Armaments Center was testing remote 3D printing for drone electronics repair at a recent exercise in the Indo-Pacific. One piece of feedback from the Marines participating in the drone repair challenge was that they want an automated solution for all-in-one electronics maintenance diagnostics and 3D printed electronics components, which can be used in rugged operating conditions.

Sciperio, the sister company to nScrypt, was the private sector partner working on the drone repair project with DEVCOM and the Marines. Since nScrypt has worked with Cu29, and since FFF/FDM machines are highly adaptable to the sorts of use cases the military is demanding, Kupros would seem like a good candidate to contribute to a diagnostic/repair tool. In any case, the exercise illustrates the rise of demand signals for low-cost, 3D printed embedded electronics.

From a longer term perspective, I think the education market could be another perfect match for Cu29. If reshoring is going to succeed in the US — and that appears to be a gigantic “if” — one of the biggest factors in determining that success involves maximizing the number of college-age and K-12 students who go on to work in manufacturing. That requires getting those future members of the workforce to become genuinely interested in manufacturing.

You know what American kids care about? Electronics, and very little else. While that has of course led to some very bad outcomes, most notably addictions to social media, the potential flip side is that an interest in electronics could be channeled into more constructive pathways, like education in valuable lifelong skills. Electronics aren’t going anywhere, but instead of an entire generation hoping to someday be paid by people who watch them play video games, maybe the future workforce can be encouraged to leverage their interest towards acquiring useful technical capabilities.

Images courtesy of Kupros, via LinkedIn

When Katie Doran was a teenager, she spent years in braces, wore headgear, and went through a series of retainers. Like many patients, she thought her orthodontic treatment was behind her. It wasn’t. Years later, her teeth had shifted again. She tried clear aligners during the COVID-19 pandemic, but they didn’t work. When she eventually saw an orthodontist, she was told she would likely need surgery before starting another round of braces, followed by about 18 months of treatment.

“I kind of retired the idea of getting braces,” Doran, Vice President of Physical Operations at LightForce, told 3DPrint.com. “I thought my smile was just what it was.”

Instead, she became the first patient to receive LightForce’s new fully customized 3D printed metal brackets. Seven and a half months later, her treatment was finished, just in time for her wedding.

For Doran, the experience was different in more ways than one. As an executive at LightForce, she already understood the engineering behind the company’s customized brackets. But becoming the first patient to wear them gave her a completely different perspective. Designed from a digital treatment plan and 3D printed specifically for each patient, the brackets are the latest addition to LightForce’s personalized orthodontic platform.

Just six visits

The treatment begins with a digital scan at the orthodontist’s office. LightForce then creates a digital treatment plan showing how each tooth should move. Based on that plan, the company designs and 3D prints a unique bracket for each tooth. The metal brackets are manufactured in-house at LightForce’s facility in Wilmington, Massachusetts. Unlike some dental manufacturing processes that rely on printed molds or casting patterns, LightForce prints the metal brackets directly, with every patient receiving a completely customized set.

LightForce’s customized 3D printed orthodontic brackets are designed for each patient’s teeth using a digital treatment plan. Image courtesy of LightForce.

For Doran, that manufacturing process translated into a very different treatment experience. She remembers spending years in braces as a teenager. This time, the experience was completely different. Her entire treatment took seven and a half months and required only six appointments, including having the brackets placed and removed. Each follow-up visit lasted about 10 to 15 minutes.

“We would pull up my digital treatment plan, swap the arch wire, take a progress photo, and I’d leave,” she said. “As a working professional, spending less time in the orthodontist’s chair was one of the biggest advantages. My treatment also corrected a Class III bite without surgery, something I hadn’t expected.”

LightForce Orthodontics releases LightBracket Metal. Image courtesy of LightForce Orthodontics.

Dr. Bryan Lockhart, an orthodontist in private practice who has worked with LightForce since the company’s early days, says the answer starts long before the patient sits in the chair. With traditional braces, every patient receives the same brackets. Orthodontists then spend months making adjustments by repositioning brackets and bending wires to fine-tune the result.

Customized brackets change that process. Instead of starting with a standard bracket, orthodontists begin with a digital treatment plan designed for each patient. Each bracket is then manufactured specifically for the tooth on which it will be placed.

While clear aligners have become increasingly popular, especially among adults looking for a less noticeable option, braces continue to play an important role in orthodontics. Unlike aligners, which patients can remove and need to wear for most of the day, braces stay in place throughout treatment and give orthodontists more control over how teeth move, especially in more complex cases.

Today, orthodontic patients have more treatment options than ever before. Some choose clear aligners, while others prefer braces. Even within braces, patients can now choose between traditional and customized options, ceramic or metal, depending on their clinical needs and personal preferences.

Braces have also become more popular again, especially among younger patients. Social media and changing trends have made metal braces more accepted than they were a few years ago. In the case of LightForce, the company combines the control of traditional braces with brackets that are customized and 3D printed for each patient.

“The details have already been built into the treatment plan,” Lockhart said. “As treatment progresses, there is less need for many of the adjustments that have traditionally happened near the end of a case.”

Faster Appointments

The brackets themselves can also be placed more quickly. Instead of bonding each bracket one at a time, orthodontists use a custom tray to position all the brackets at once. According to Lockhart, that cuts down on the time patients spend in the chair during the first appointment.

For Doran, another benefit was comfort. Because the customized brackets could be placed lower on her teeth, she avoided using bite turbos, small pieces of material sometimes added to keep patients from biting directly onto their brackets.

Dr. Bryan Lockhart. Image courtesy of LightForce.

As Vice President of Physical Operations, Doran spends her days overseeing LightForce’s manufacturing of customized brackets. But going through treatment herself changed how she viewed the technology.

“I cared far more about how it felt to have this product in my body than about the manufacturing process,” she said. “Most patients feel the same way. The technology matters because it leads to a better experience, not because patients want to know how the brackets are made. The manufacturing process is incredibly cool. But the benefit is that it allows us to do something you simply couldn’t achieve with traditional manufacturing.”

Software for 3D printed braces. Image courtesy of LightForce Orthodontics

Lockhart believes customized braces are the next step in the evolution of orthodontics. Years ago, orthodontists bent nearly every wire by hand. Later, brackets with built-in prescriptions simplified treatment. Personalized brackets, he says, are simply the next stage.

Doran tells me orthodontics has changed very little over the past century, apart from the introduction of clear aligners, leaving plenty of room for innovation. For patients, however, those changes may be measured in much simpler ways, with fewer appointments, less time in the chair. And in Doran’s case, getting her braces off in time for her wedding.

One additional thing we often don’t think about, but should, is that we need several of these advantages stacked one onto another for the advantages to cumulate sufficiently for it to make sense to move to additive. Something we think of even less is that if something makes sense for additive, then adding more additive advantages to it is comparatively easy. In 3D printed helmets, we’re seeing that cooling and sweat wicking properties, as well as better moisture and heat management, generally are a major factor. This is helping to drive adoption of military and sports helmets. Similarly, heat management is driving broader adoption of wheelchair cushions as well. We also know that heat exchangers are a huge application in electronics, performance automotive, aerospace, and defense. Fluid handling and fluid handling systems, including soft robotics and things like hydraulics, are also growing. We’re also seeing more and more end-use parts in elastomeric components, something which we initially sucked at. Combining these disparate developments leads to something blindingly obvious and new at the same time.

On my walk yesterday, it was 20:00 and 32°C. The heat wave here in Europe has been brutal and unforgiving. A young man passed me, a lean construction worker wearing a yellow high visibility hat. The immensely wide brimmed hat hat had a cape behind him. For someone working outdoors all day in the scorching sun, this is a chillingly good hat. It’s ridiculous that we need something like this now, but I’m glad for him that he has it. Can’t we do a bit better, though?

 

View this post on Instagram

 

A post shared by nathaniel (@nathanielfong)

Body armor cooling is a big issue. The US military, for one, has tried again and again to develop body armor cooling systems. Japanese construction workers wear these puffy air jackets which are better implementations of what the military has come up with most of the time. A much less energy-intense system, the Ice Plate is amazing and essentially is an adaptation of putting a Camelbak (which a lot of soldiers already carry) in the freezer before putting it in between you and your plate carrier. The Body Armor vent uses the person’s breathing to vent an evaporative vest.

Tacvents are yet another alternative here: a corrugated rubber panel creates space and airflow, reducing temperature. Another option is the Maxxdri vest, which has over 700,000 customers. The company’s innovation was to create a 3D texturized space in a vest to create airflow and reduce heat buildup.

Now there’s no way we could make something like that, is there? Or add textures to it to better wick sweat? What about doing the same for shooting gloves, biking gloves, safety gloves, and other construction equipment? What about for firefighting helmets and gear? For all of these components, wearing them is going to be much more of a pain as heat persists. And if people stop wearing safety gear due to discomfort or heat, we will see more injuries, accidents, or deaths.

So the solution is easy. Let’s take the body armor example. We create corrugated elastomeric panels that have textures to wick sweat and reduce surface temperature. Perhaps they have Scilla-like structures to create more space and airflow. These panels are textured to remove air as the wearer walks and moves their arms. At the same time, a compression on a 3D printed bladder pushes air out periodically across the panel. There is room for ice inserts or adding Camelbaks to the structure if you’d like. The structure has porosity to promote airflow even further. And what’s more, it is contoured and the performance has been maximized according to the body.

Additionally, there will be an internal heat exchanger system whereby cold liquid could be routed around the body near the skin through channels. A pump can be used to move the liquid along manually to get it to circulate. The vest itself will be a flexible integrated heat exchanger. We could do this for helmets, gloves, and other safety gear as well.

Liquid Cooling Garments are already used in space gear and some professional settings. Usually here, PU tubes circulate water close to the body. This is circulated to a vent or cooling unit. Commercial versions of these have Peltier plates in areas such as the neck to aid cooling. These are electronic cooling plates that use the temperature gradient to cool. We could perhaps make some housing components for this.

What is also possible is to make some kind of ridiculously inefficient elastomeric Sterling engine to generate heat into fan movement to blow underneath the vest. Shape memory materials could also make heat exchangers for use close to the body, although work on these seems to be more industrial now. Generally however, I think we could make a better cooling vest for plate carriers, but also develop better cooling solutions for safety helmets, gloves, hearing protection, and more.

This is potentially a high-value application that could initially be funded through high-end implementations for soldiers, pilots, and the like. Later, the technology could scale to the millions of construction workers, policemen, and other people that need to be outdoors when it’s hot.

Now, a company founded in May 2024, Suzhou Dewu Technology Co., Ltd. (Dewu Technology), targeting the same market for ultra-high-resolution, tiny metal parts, has announced the closure of a pre-Series A round, in an exclusive from Chinese tech media site 36Kr. The numbers aren’t very specific, with the site simply noting that Dewu Technology has been funded thus far to the tune of “tens of millions of yuan,” which sounds like anywhere in the $5-15 million range. The leaders of the round were two Chinese investment firms, Beijing-headquartered GSR United Capital, and Jintou Zhiyuan, based in Jiangsu.

Dewu Technology claims its printers are capable of achieving wall thicknesses down to 30µm, with finishes down to 1µm. These are bold claims, especially for such a young startup, and there’s not much of a presence in (Western) online sources for the company. It seems somewhat relevant, however, that the 36Kr interview mentions that one of the institutions where “core team members” at Dewu were educated was Shanghai’s Jiao Tong University, which is one of the members of the aforementioned consortium involving BLT.

The interview is also filled with insightful observations from the company’s founder, Wang Zhou, related to demand trends in the global AM market. These are especially relevant to demand for metal parts used in consumer electronics applications:

Wang frames the company’s business model by noting that the most relevant question for customers “…is not whether it can be printed, but ‘how many procedures are still needed after printing.’ The post-processing cost of parts printed by traditional SLM accounts for 70%, and we have reduced this proportion to 10%-20%. …When the surface finish reaches Ra ≤ 1.5μm, many medical and consumer electronic parts can be used directly without post-processing.

…Our technical route is naturally suitable for consumer electronics, because the cost saved in post-processing is real. This is also why overseas leading consumer electronics customers have come to us intensively since the second half of last year. Their design end has moved towards ultra-high-precision structures, and traditional 3D printing cannot meet the requirements…as consumer electronics enters mass production, multi-laser large-format equipment will further reduce the printing cost, and this advantage will continue to expand. Customers do not ultimately buy a piece of equipment, but a solution that can be delivered in mass production.”

Again, one can certainly question how consistent the company’s process is at delivering parts with such thin walls and such high-resolution finishes, and how close it is to scaling its business model. But it is noteworthy in itself that the company’s value proposition is tiny metal parts, and that this is, largely, specifically in response to growing demand from consumer electronics firms, particularly in light of the BLT and Farsoon announcements from earlier this year, as well as the regular influx of 3D printed electronics news over the course of this year.

It’s also worth mentioning that, in the last several years, seemingly every time non-Chinese (especially Western) observers express skepticism about Chinese technological progress — and above all, such progress in the context of consumer electronics — they are inevitably proven wrong. The most attention-grabbing example was probably the Huawei release of the Mate 60 phone, which demonstrated that China was capable of making chips with 7 nm capabilities, despite years of various Western sanctions.

The same trend has continued with Chinese AI firms. People scoffed at DeepSeek, and while that company may have been overhyped, initially, it wasn’t just hype, and at any rate, there are a number of other viable Chinese LLMs. Meanwhile, a Chinese company just announced that it succeeded in producing a DUV machine that is expected to begin shipping this year, enabling additional workarounds to Western chip sanctions.

All of that aside, isn’t it deeply arrogant for people in, say, the US, which at this point has been ceding ground in manufacturing for almost as long as it was the world’s dominant manufacturing power, to assume superiority over the country where more manufacturing has taken place in the 21st century, by far, than in any other country? It used to be that one could deflect that fact by saying all they make in China is goods that cost too little to be worth anyone else’s trouble. But that has long since ceased to be a valid comeback.

Images courtesy of Dewu Technology

Keyport validated its 17-4PH process on the EOS M290, one of the most widely used PBF systems in the world. The Navy has been working on this project for seven years, launching it as a Naval Innovative Science and Engineering (NISE) project and then transitioning it to a formal qualification project supported by the Submarine Industrial Base (SIB) program.

The Keyport Warfare Center went a step beyond validating the technique: it also tested what might happen to the print process during a power outage. The Navy researchers achieved that simply by stopping the build for 12 hours and then restarting it to gauge the results. According to the research team, who subjected the build to destructive testing following its completion, the stop-and-start nature of the print didn’t affect the final outcome.

The US Navy has effectively underwritten a significant chunk of American metal AM adoption growth thus far this decade. This step essentially represents a transition from “planting season” for the Navy’s AM buildup to the harvesting of experiential gains. Keyport notes that it’s already working towards qualification of 316L stainless. And, now that Keyport has qualified 17-4PH, the benefits won’t go to the Warfare Center alone, but are accessible to the broader Navy supply chain.

In a press release about NUWC Keyport’s receipt of approval for its 17-4PH process on the EOS M290, Bryce Weber, platform readiness and endurance technology lead in the command’s Undersea Systems and Sustainment Engineering Department, said, “This milestone is a key enabler for the NAVSEA organic industrial base [OIB] to begin upscaling metal AM technologies in a meaningful way. Keyport’s processes to achieve this qualification can be transferred to other commands and duplicated. This provides an easier “instruction set” for OIB sites, unlocking the number of qualified machines allowed to print critical shipboard parts during maintenance availabilities.

“These qualification credentials allow Keyport to develop and validate part-specific manufacturing processes that can be digitally shared, creating outsized impact. The Keyport team is known for its agile manufacturing engineering, prototyping and production readiness capabilities to solve parts availability issues.”

Image courtesy of EOS

Austal USA has already stood up the Digital SEA platform, which allows Navy suppliers to access the technical data necessary to provide the service with metal AM parts. The Barnes Global Advisors (TBGA), meanwhile, has just landed a BPMI contract for setting up a qualifications database for Inconel 625 and 316L stainless for US Navy suppliers, which I suggested will be incorporated into Digital SEA, since the latter is a centralized repository.

So the Navy will soon have qualifications data for at least three metals up-and-running on a digital platform. This should accelerate adoption, not only for these three metals, but, as soon as the operational capabilities of the centralized database model itself is validated, should also accelerate the qualification (and, thereby, adoption) of even more materials, likely at an even faster pace.

This is what progress looks like! Regardless of the actual goal that’s achieved, it’s perhaps more important at this stage in a tech buildup (the final steps from low-level production to full commercialization) to prove that an organization can set multiyear objectives for itself and accomplish them in relatively timely fashion. But the Navy has now done that and, the actual goal that it has achieved is arguably the most important one in terms of setting itself up for future success. You can’t produce parts at scale without qualifying the production process, and you can’t maximize the potential value unlocked by AM parts qualification without a robust digital ecosystem.

One other detail that the Navy mentioned in the Keyport press release is that it’s working on a project to develop in-situ monitoring (ISM) hardware. Phase3D already works with the US Navy and, in fact, has its Fringe Inspection installed at the US Navy AM CoE in Danville, VA — the physical home base for Digital SEA — where it is attached to an EOS M290. If this is what the Navy is referring to, that adds yet another catalyst which should speed up the momentum behind US metal AM adoption.

Featured image courtesy of Frank Kaminski, NUWC Keyport