Known as the Venice of the North, Amsterdam has a vast system of canals and bridges. Within this network are 165 canals and more than 1,500 bridges, some dating back as far as the 1600s. Over the years, many of these aging structures have undergone significant repair while others are still in need of maintenance.
Recently, a heavily trafficked bridge in Amsterdam spanning Oudezijds Achterburgwal, a canal in the city’s Red Light District, found itself next in line for a rehab. The rehab project, however, wasn’t just minor; the bridge had to be removed during its repair. Seeing an opportunity to showcase a new technology, MX3D, a tech company based in The Netherlands, approached the city with a proposal to produce a replacement bridge while the historic one got its facelift. Albeit temporary, that replacement bridge had to adhere to city codes to keep the city’s pedestrians safe crossing over it. It also had to be rigid enough to sustain a boat crashing into it.
After checking all of the necessary – and critical – boxes, MX3D won the opportunity to produce the bridge. The company also won the American Welding Society’s 2024 award for “Outstanding Development in Welded Fabrication.” Interestingly, traditional welding wasn’t involved in the project.
Instead, the secret sauce that garnered AWS’s attention was MX3D’s expertise in the development and application of large-format robotic wire arc additive manufacturing (WAAM), a type of additive manufacturing that relies on traditional welding wire as its feedstock. MX3D’s final design and delivery of the bridge resulted in a solution that met Amsterdam’s safety standards while wowing passersby, connecting them to a new technology they’d never been exposed to before.
Ambitious designs
Using welding wire along with computer control, the WAAM technology allows MX3D to – layer by layer – create the complex geometries the company needs for its end products. In addition to the bridge in Amsterdam, MX3D has gained recognition for applications in architecture, infrastructure, and even nuclear and automotive parts, proving how versatile the technology can be.
Celebrating its 10th year in business, the team at MX3D got their start working at Joris Laarman Lab, the studio that ultimately designed the bridge. With a focus oriented toward digital manufacturing, Fillipo Gilardi, the company’s R&D manager, says their early interest in new printing methodologies eventually led to the spinoff of MX3D.
“My background is in industrial design, so during school, I mainly studied the process of additive manufacturing in industrial scenarios,” Gilardi says. “We had a very prolific workshop for internal production, but based on our backgrounds in additive, we set out to develop a technology that would allow us to manufacture even more in-house. We had an old welding machine, so we attempted to use that to put welds on top of one another instead of next to each other. And it worked.
“After that initial success, we spent most of our time developing the core points of the technology,” he adds. “That included everything from the more practical side of printing to drafting the first sketches of what the software behind it would be in the future.”
During those early development years, the goal was to gain industry trust and manufacturing know-how. The team there accomplished that by engaging with welding experts; structural engineers; robotic and power source providers; and welding institutions, such as AWS and The Welding Institute in the U.K.
“These relationships were incredibly useful because, ultimately, they possess a lot of the know-how and IP that we needed to keep improving,” Gilardi says. “It was also critical to have their input to evaluate what we do.”
Today, the company’s approach to 3-D printing methods comes with a specific focus on printing big – producing large, functional and structurally sound objects that meet incredibly high standards. To produce its parts, MX3D leverages a range of ABB’s industrial robots equipped with Fronius GMAW/CMT welding machines. Interested parties can also invest in the company’s turnkey system, which is marketed as the M1 Metal AM System and comes with the company’s proprietary software, MetalXL. Fully hardware agnostic, the system can be integrated with any brand of robot or welding power source the customer prefers.
“For the bridge project and in general, we learned to appreciate the ABB 2600ID welding robots because of their size and reach, but also because of the hollow wrist for integrated dressing,” Gilardi explains. “They’re also lean enough to be easily reconfigured in different setups. We can join them to an external axis on top or on the side. We can put them on a rail or even mount them upside down. They’re very flexible.
“With the M1 system, we can print, maybe not the most complex things, but definitely the biggest,” he adds. “We just printed a nuclear application of several meters, weighing more than 5 tons. The work we’re doing for our nuclear customers or for other customers with critical components is helping to reduce their wait times. We print a lot of components for the energy industry, most of which are casted components. Casting or forging an impeller for a cooling pump in a power plant only takes about one day, but the demand at these facilities creates a sometimes year-long wait list.”
Speaking of wait times, Gilardi says the supply chain has a lot of adapting to do to catch up to the growing use of welding wire as an additive feedstock. While the use of traditional, highly available welding wire helped the burgeoning technology initially get off the ground, larger critical projects require feedstock from the same batch. When uncommon or exotic alloys are involved, the challenge to source the wire inevitably grows.
“I’m not saying it’s hard to innovate – the industrial need for welding is very present,” Gilardi says, “but with a completely new use of a technology, improvements are always required. At the beginning, we were questioned a lot. Is it going to hold? What material are you printing and what are their properties? How can you validate? All of those things are required.”
Building standards
Additive manufacturing and the WAAM process aren’t new, of course. But, as these processes mature and their adoption grows, the development of standards become more and more critical. In fact, AWS published the D20.1 Standard for Fabrication of Metal Components using Additive Manufacturing in 2019. Because these comprehensive documents don’t materialize overnight, its development began years prior when the burgeoning technology was beginning to gain steam.
The development of standards like the D20.1 code book can take years as they involve a structured, collaborative process to ensure that the final standard is scientifically sound and practical for industry applications. The goal is to create a code that ensures safety, quality and uniformity in manufacturing practices across different sectors. Bill Mohr, the D20.1 committee chair, is leading that charge.
As a principal engineer within the structural integrity group at Edison Welding Institute, assessing fitness-for-service, design and fatigue of welded structures is part of Mohr’s day job. His experience, therefore, is incredibly valuable to the committee and its current task – creating a separate document, D20.2, which will be dedicated to the directed energy deposition (DED) technology of using wire as the feedstock, as leveraged by MX3D, versus powder.
When AWS developed the original 2019 document, the committee recognized that other standards bodies were focused on specifics for the multiple types of metal additive manufacturing available at the time. Therefore, AWS’s approach was to draft a more general document. Currently, that document, the D20.1, covers powder bed fusion machines and the DED process, an umbrella term for the WAAM process used by MX3D. When relying solely on the D20.1, engineers must provide the additional specs they need for their specific projects.
Understanding the growing use of DED with wire as the feedstock, the committee saw value in creating a separate document. Looking to the future, the plan is to have two documents, D20.1 and D20.2. When powder is the feedstock, D20.1 will be the document to use.
“Both of these are currently draft documents that the committee is working on,” Mohr says, “but both of them have objections that we have to work through to see whether we can come to a full consensus on the document.”
In the case of a standards document, objections aren’t necessarily a bad thing, though. Mohr stresses how important it is to garner as many participants in the process as possible to ensure that it is as comprehensive and watertight as possible.
“We’re getting more comments, which should eventually lead us to a document that’s as good as it can be,” he says. “The effort of attracting more people helps us uncover the different ways people will be working with the technology.”
AWS committees include researchers from universities, people from job shops, inspectors, representatives from organizations that are designing the equipment and feedstock, and other trade professionals that will use the product. Some of the commenters could be from organizations producing simple products while others could be making critical aircraft parts. This wide range of commenters also ensures that any potential materials will be covered, be it steel, stainless steel, titanium or a myriad of other useful alloys.
“We will have input from very different industries and applications, which is an important part of attempting to produce a very general document,” Mohr explains. “That’s where we see the future of this.”
Command and control
The DED process using wire as the feedstock – wire arc DED – can also be considered an umbrella term. As an example, the WAAM process used by MX3D refers specifically to the process of building a 3-D object using a welding arc to melt the wire feedstock while wire arc DED is the broader category within DED where a wire is used as the feedstock and melted with an arc as the heat source. Either way, the processes have a lot in common, including their use in small batch, large-format projects exemplified by MX3D’s bridge as well as the work it’s providing to the nuclear sector.
“I see these processes tending to diverge, but not quickly,” Mohr explains. “They will still have some kind of interface between them, but there will be parts and projects that are just a little too big to be made from powder. There are some really enormous opportunities for both, but the biggest things are not going to be made with powder.”
Currently, there are a number of competitors in the large-scale wire arc DED area, including MX3D as well as Big Metal Additive, Cranfield WAAM 3D and Lincoln Electric. Concurrently, they are all coming up with systems, designing processes and acceptable materials, while also developing the computer motion control to achieve the shapes they need. In that regard, “you might say additive manufacturing in general is about command and control,” Mohr adds.
When it comes down to it, wire arc DED and WAAM are technically welding processes; the difference is that, compared to manual welding, they can be precisely controlled. With that computer control, “you know beforehand what you’re going to get on each build,” he says.
Additive has been on the fringes of the metals fabrication world for a while, but now that it’s becoming more aligned with the welding world through the advancement of wire arc DED, it’s exciting to think about what the future holds.
