At last year’s Additive Manufacturing Strategies (AMS), we spent an entire morning of a three-day show on DED, and specifically pressure vessels. To many, pressure vessels are a rather obscure 3D printing application. But to us, they’re absolutely essential for applications in oil & gas, nuclear energy, rockets, hydrogen, industrial power, and underwater structures. A lot of any new energy infrastructure, whichever path it takes, will end up using a lot of pressure vessels. And with DED, we can make huge ones at relatively low cost. We can also test them to specifications and they can pass, meaning a more surefooted adoption path for engineers and management. That pass and fail is something we’re sorely missing with other parts, such as components under fatigue load.
So it’s obvious that the Department of Energy (DOE)’s Oak Ridge National Lab and Idaho National Laboratory (INL) would take an interest in pressure vessels. Indeed, ORNL’s earlier work with Lincoln Electric has led to that firm’s service in making pressure vessels. In working with ARC, ORNL led to a path forward in digitizing energy infrastructure, and its work with GKN has led to that firm making aerospace DED parts. In fact, the first steps towards DED were made by Westinghouse engineers in 1920 and a patent was awarded in 1925, the Baker Method of Making Decorative Articles.
As well as an early example of DED, the invention was also a remarkably accurate depiction of the overall look and feel of DED parts. And DED indeed still uses “ornamental arc welding, more especially to utilizing an electric are, such as is ordinarily employed for electric welding, for the formation of deposits to produce receptacles or containers of ornamental and useful shapes.” Now we’ve moved a bit beyond the ornamental alone, but very prescient. And the first motion stage in DED was a guy called Ralph. Later on, Sandia, the only US national lab named after a melon, would develop LENS, meaning that since the very beginning, DED has been intertwined with the US energy establishment.
ORNL and INL want to continue work on wire arc DED (WAAM) for pressure vessels. The aim is to improve quality and make even larger things. As we heard from Lincoln Electric, they can already make huge components that weigh thousands of pounds, so the Manufacturing Demonstration Facility (MDF) must be working on some pretty huge parts. The two labs will use AI to help make the pressure vessels.
Shannon Bragg-Sitton, Idaho National Laboratory associate laboratory director for Energy and Environment Science and Technology, said,
“INL and its industry partners will accelerate development of new reactor designs, components and manufacturing methods by applying AI tools, such as those developed under Prometheus, to support the commercial adoption and deployment of nuclear energy and meet the growing needs of American communities, manufacturers and AI data centers.”
Robert Wagner, ORNL associate laboratory director for Energy Science and Technology, stated,
“ORNL’s unique strength is our ability to connect world-class science with the Manufacturing Demonstration Facility’s capabilities to move innovation from research to real-world impact. By working with industry to demonstrate, validate and qualify advanced manufacturing technologies, we can reduce risk, accelerate deployment and strengthen the domestic supply chains essential to America’s energy future.”
Initially the team made a 3 foot by 5 foot nuclear part out of steel using the multi WAAM platform MedUSA with three print heads. They also made neutron sensor brackets for Antares Nuclear‘s R1 Mark-0 micro reactor.
ORNL lead researcher Patxi Fernandez-Zelaia stated that,
“We would like to achieve born-qualified pressure vessel components using data gathered during printing to confidently assess their worthiness for extreme environments,”
While Jorgen Rufner, INL group lead for advanced manufacturing, says,
¨With this project, we’re bringing together expertise from both labs to integrate AI and data science with advanced 3D printing so we can evaluate a part’s performance in real time, while it’s being printed, instead of waiting for post-production testing,”
Now it is worth a little unpacking of the “Oakridgese” here because ORNL is working on something completely astounding. What they want to do is use sensors, AI, simulation, and some pretty amazing computers to make new articles that take into account any environmental and design constraints, with the machine, material, and physics itself to such a level that you could make a qualified product on the first print. So all that is going on changes from black box to completely knowable. This will accelerate DED adoption significantly because it will expand the speed of getting parts made and massively decrease development time and errors.
ORNL and INL are working on some pretty amazing technology here. And the challenges must be very daunting. But, if you’re active in DED, I’d make a beeline for the MDF because this could be some truly competitive market-changing technology if it comes to market.