Loading the ElemX 3D printer aboard the USS Essex © Ace Rheaume, US Navy, July 2022
Interview with Tali Rosman, entrepreneurial consultant in Miami (Florida) and ” Entrepreneur-In-Residence ” at Metropolitan University of Toronto (Canada)
–> Interview by Murielle Delaporte
A well-known expert in the field of digital fabrication and 3D printing, Tali Rosman is now a start-up consultant at Miami-based RHH Advisory and entrepreneur-in-residence at Metropolitan University of Toronto. She has lived in the USA since 2014, but previously studied and worked in Israel, France (where she graduated from INSEAD), the UK, Switzerland and Singapore. Among other things, she oversaw mergers and acquisitions at Stratasys, then the leading player in the additive manufacturing market, and in 2020 became CEO of Elem Additive at Xerox.
She describes this experience as her “greatest achievement”, as she was responsible for creating a new company, Elem Additive, from scratch, and launching a brand new product, ElemX. In fact, it is well known in the military field for having worked with the Naval Post Graduate School to test the very first liquid metal projection 3D printer, installed in July 2022 on a US Navy ship, the USS ESSEX, while deployed at sea.
In this interview, she describes the state of the art in metal 3D printing, its advantages and disadvantages, and the prospects she sees for this revolutionary field.
What are the features and main advantages of the technology you’ve developed with the ElemX printer (based on liquid metal spraying)?
The ability to be deployed is probably ElemX’s greatest asset, since we delivered the printer in a container, enabling it to withstand marine conditions such as vibration and corrosion.When I was hired in 2020 to create Elem Additive with a small team of researchers, who were behind the liquid metal spray technology, the market was dominated by plastic 3D printers and powder-based metal printers. My task was to move from the latter technology to safer, easier-to-deploy metal printers using liquid metal technology fed by a spool of wire. The challenge was therefore manifold, as we had to move away from powder-based technologies, which are toxic for the environment and highly explosive, making them unsafe and difficult to deploy.We have succeeded in developing a completely new technology, used today by the US Navy, the US Department of Energy and Siemens, based on metal wires (currently focused on aluminum) rather than powders. Wire is safe, non-toxic and non-explosive. Although it’s not “plug and play” yet, it’s not far off, and responds to imperatives of time and space (“when you need it”; “where you need it”).
Speed is of course another key advantage, with the ability to manufacture parts on demand, with a turnaround time of one day. Speed refers to the time it takes to produce a part, and is linked not only to printing speed, but above all to the reduction in post-processing time, enabling true on-demand, on-site production. Powder-based metal manufacturing requires post-processing equipment often outsourced to external suppliers, which increases delivery times and limits the ability to produce parts on demand in remote locations: post-processing for metal printing is cumbersome and costly. What users want are parts manufactured “on demand”. So it’s one thing to print in a few hours, but if post-processing takes weeks, it defeats the purpose, because it’s not “on-demand” manufacturing.
This ability to deliver a new, usable part on the same day that a part has broken or been lost is a game-changer for logistics and inventories in both the civilian and military sectors. All you need is space to install the printer. Although the ElemX is quite large, new market entrants such as the Belgian company Valcun now offer similar products with a smaller footprint and reduced power consumption. In addition to the printer, all that’s required are spools of wire (currently, ElemX and ValCUN solutions focus on aluminum) that can be purchased off-the-shelf and stored without any special processing, as well as post-processing equipment, which is minimal compared to powder-based technologies. Of course, this depends on the part, but most of the time, the part comes out of the printer in good condition, and post-processing will mainly involve milling to finish the surface. It depends on the part, of course, but most of the time, the part comes out of the printer in good condition, and post-processing will mainly involve milling to finish the surface to make it perfectly smooth, or simple heat treatment in an ordinary oven to strengthen the part. There’s no need for a proprietary oven, powder disposal equipment, HIP, etc.
Sovereignty is also strengthened, as 3D printing enables manufacturing to be relocated, creating promising new uncharted territory in terms of workforce training and raw material use. As the process has been greatly simplified, liquid metal spray 3D printing has another advantage: it requires less manpower, and training digital manufacturing engineers and mechanics is much easier and quicker. What’s more, the impact on the supply chain in general could quickly become perceptible for two reasons: firstly, there are none of the material complexities posed by powder-bed metal technologies, all you have to do is buy spools of wire that are readily available on the market. It’s worth noting that all emerging liquid-metal printers, from ElemX to ValCUN to GROB, initially focus on aluminum, which accounts for 25% of the global market for metal parts (steel accounts for 25% and the rest is divided between a few other materials). Secondly, you don’t need a large quantity of material, since you simply use what you need to produce the part, unlike subtractive manufacturing, such as machining, where you subtract material from a block of material. By definition, subtraction means tons of waste. Finally, less need for materials means fewer inventory problems: you no longer need to stock multiple parts, but only the materials you need to produce them.
What are the main obstacles to be overcome before 3D printing becomes a “form of default manufacturing” in some cases, as you describe it?
The relationship between 3D printing, or additive manufacturing, and the global supply chain can be compared to the relationship between a backup generator and the power grid: if the usual system is working, it’s probably best to use it, but if the supply chain is disrupted, a backup generator is needed, and that’s what 3D printing can offer and do. For some parts, additive manufacturing will always be used as a back-up generator, but for others, it could become the form of default manufacturing. This is currently the case on the front line in Ukraine, where many parts cannot be supplied and stocks are not always available.
At present, two major obstacles prevent 3D printing from being used more generally as a “default manufacturing mode”:
– The first is the technical difficulty of mass production. ElemX is used for a diversified, low-volume product mix on a deployed vessel or forward base, not for mass production in depots. However, this situation could change with the emergence of non-powder solutions. Israeli company Tritone, for example, uses a new process that fills a wax mold with metal paste, enabling it to produce thousands or even tens of thousands of parts. Although it’s not made for millions of parts, it’s already a different market from that of Xerox – or ADDiTEC today, since Elem X was bought by that company last summer – and Vulcan.
– The second major change concerns the certification and qualification process, which is particularly cumbersome in the military sector. Numerous debates are underway on this particular issue, which is slowing down the expansion of the use of 3D printing in the armed forces. One solution is to certify parts in advance; the other is to consider the level of quality required for non-critical parts or parts with a non-primary structure. The question is: “Do you really need this part to be certified?” For example, the handle on washing machines aboard military vessels can, if broken, become a source of nightmares on board. Does it have to be of impeccable quality? This brings us back to the argument of “good enough to work” quality, as opposed to “superior quality” criteria, which require a rigorous certification process to ensure that the printed part corresponds exactly to the initial part.
A final word on the prospects for additive manufacturing from your point of view?
A rigorous certification process or in situ quality control could very soon be greatly facilitated by the current exponential use of artificial intelligence (AI). A few months ago, I would have told you that AI was just a pipe dream, but today we’re all feeling the “ChatGPT” effect on multiple levels: whether in terms of quality assurance, or in terms of the degree of trust people now place in these new tools.
AI can in fact contribute to predictive maintenance by increasingly integrating quality control and end-user input, while speeding up the whole process. I’ve seen a few new startups beginning to master generative design. Nvidia recently announced the launch of an AI model for 3D design.
In addition, the question of security, both in terms of cybersecurity and intellectual property protection, has now been resolved by major technological advances in this crucial area. Some start-ups, such as Israel-based Assembrix, are already working on secure digital file transfer, to the extent that printing staff would not be able to access protected information. The difficulties slowing down the spread of additive manufacturing on a larger scale are therefore beginning to be resolved one by one in the field.