ETH Zurich researchers have come up with a rotary LPBF system, the RAPTURE, that can process multiple metal materials simultaneously. In their published paper, they explain that the system may be very well suited to making ring-like parts. Impellers, turbine and compressor components, pistons, pipes, valves, pressure vessels, engine components, bearings, connectors, and more could be good candidates for this technology. I hope it works in Inconel.

In this system, the deposition system and the soot extraction system are spinning synchronously with the laser, while the build platform is stationary. Gas is added locally only, and the system incorporates an extractor for soot and a localized inlet for gas. A small, angled, gravity-fed hopper feeds powder, in time with its own rotation, which is further spread by a carbon fiber brush. For hopper and extraction design, the team used ANSYS Fluent software and the like to optimize components. In a spinning printer such as this one, powder accumulation on the build zone is problematic, but the researchers seem to have solved this through component optimization and testing. But, they seem not to have solved how to modulate speeds depending on geometry. The build area is 200mm x 200mm, though a central 30mm ring is needed at the center, where parts can’t be built. Initial testing was done at 30 µm layer thickness. The system uses a SCANLAB hurrySCAN III, IPG 200W laser, and motion control from Hiwin and Aerotech.

Test prints were reported at 5 m/s, getting a Ra=89.8 μm, which the team hopes to improve. The researchers seem to have used 316L powder in all their testing. They expect to use a lot less powder than with rectangular 3D printers. They also said that there was a “16.6% build time reduction for the rocket nozzle and 10.3% for the gyroid ring compared to a linear system.”

“Savings exceeding 52% were predicted for thin-walled ring-shaped parts, highlighting a geometric dependency favoring parts with small cross-sections and wall thicknesses relative to the build area,” they continued.

The researchers think that perhaps walls could be put up to further increase speeds, and want to complete further work on beam shaping, hatching, and more.

In contrast to the soccer team-like papers we see often, this paper was written by just Markus Bambach and Michael Tucker, and the work was completed over 9 months. The printer was specifically developed to make rocket nozzles for bipropellant engines. The hope therefore is to make the system suitable for producing parts with multiple powders in future iterations.

Tucker said that “this process is ideally suited to rocket nozzles, rotating engines and many other components in the aerospace industry. They typically have a large diameter but very thin walls.” He also said, “At first we underestimated the extent to which the gas flow mechanism affects product quality. Now we know it’s crucial.” That in and of itself could be an interesting area ripe for optimization in many conventional machines.

Architecturally, we’re moving away from the rectangular recoating machines of yore that spend more time recoating than printing. If this system works well, then we could see a continuous 3D printing system where recoating time is either non-existent or minimized. That architecture could also perhaps be much faster overall when married to specific improvements in light engines.

I love the idea of having an LPBF machine—either for oil and gas as well as new space—that’s just for rings. I also love having a specific rocket nozzle system. Generally, it seems to make more sense to have systems optimized for specific parts or geometries. This kind of system may not suit everyone, but it may be the best for impellers or rocket nozzles, and this could be a significant market. If the main driver of LPBF adoption currently is New Space, it should follow that we see more optimized architectures for their specific needs. In this case, the RAPTURE may be, for now, very specific, but in so doing may be very viable indeed.