Page 24 - 3DMP Spring 2022

3D The AMNOW Challenge: Developing Digital Thread Data
engineering team and Principal Advisor, metallic additive and metals manufactur- ing with The Barnes Global Advisors. “At the end, the Army will also have extensive knowledge and data regarding the capa- bilities three of more suppliers for each combination when they need to procure these types of parts and materials to help them overcome supply-chain and spares issues.” The AMNOW Challenge builds upon this by looking at the art of the pos- sible when AM is combined with AI and data analytics.”
The three suppliers providing materials and process data for the AMNOW Chal- lenge are ATI Specialty Metals, Penn Unit- ed and Innovative 3D Manufacturing. Each already has completed several AM builds of fuel-system elbows currently being conventionally manufactured for use in U.S. Army rotorcraft. Researchers at Penn State University developed the original parameter sets for the build as well as test coupons. The project then transitioned over to these three suppliers to further develop production-represen- tative builds—more elbows and fewer test coupons. All of the builds have undergone extensive testing—tensile and fatigue tests, as well as composition, metallog- raphy and density testing.
The Project’s End Goal
“To tie the AMNOW Challenge to the overall program,” adds Ashley Totin, AMNOW Program Manager and senior project engineer for NCDMM, “we are creating a secure bidirectional commu- nication space to allow suppliers to down- load contracting documents, and TDPs (technical data packages) and upload manufacturing process information. That’s the end goal of the program, so that the Army can come in, create a job that goes out to the suppliers electroni- cally with the TDP, and the supplier builds the parts. However, the contract is not only for the hardware deliverable; there’s also a data deliverable, including digital data from the printing machines that becomes an operations report submitted electronically into the secure government cloud space called AMIP—advanced man-
ufacturing intelligence platform.”
AMIP data, as outlined in a project brief, include several mechanical prop- erties—tensile, fatigue strength, relative density, etc.—as well as radiographic data, thermal-processing history, feedstock cer- tification and much more. The entire RFQ, bid and contracting process will mimic the actual DoD contracting process, and suppliers will have access to a low-cost, open-standards-based data acquisition
and control system.
For the AMNOW Challenge, the five
semifinalist data analytics entities, select- ed in mid-February, will take the data sets provided by the three builders, correlate them, discover which builds created anomalies and what data correlates to that, and finally recommend process and parameter changes in order to improve their results. The five down-selected semi- finalists:
• University of Waterloo
• Applied Optimization/Penn State University Applied Research Laboratory
• Northeastern University
• Cognistx
• Addiguru.
Each will present, in late April, their
concepts for improving productivity and quality to an evaluation panel with rep- resentatives from across industry, the DoD and from NIST. The panel will then select two winners that will receive funding from the AMNOW program and work with the three builders to refine their processes.
“We’re not really looking to change the part design,” says Slattery, “since the goal is to develop a solution to address sup- ply-chain issues and manufacture spare parts. There’s really little or no design free- dom here, we just want good data corre- lation and process predictions.”
New AMS Specs
From the AMNOW program, Slattery and Totin expect to generate at least eight new SAE AMS (Aerospace Materials Sys- tems) specifications, “three from the 316L stainless-steel project alone,” says Slat- tery. Some are powder specs, others are material or part specs—acceptance cri- teria, thermal processing specs, and actu-
For the AMNOW Challenge, three metal-AM suppliers have employed laser powder-bed fusion to build these fuel-system elbows (and test coupons) currently conventionally manufactured for U.S. Army rotorcraft. Each supplier demonstrated compliance with a new contract requirement for in-situ data by securely uploading complex static and time-series data as part of production. Five R&D entities are provided access to that data and charged with seeking ways to further refine every aspect of the build process to improve productivity and quality.
al statistically based mechanical-property minimums.
“At the same time, we’re working to help develop the supply chain,” Totin adds, “showing not only we make the fuel elbows but that an Army design or pro- curement authority could qualify machines. So, we are also supporting an AMS machine-qualification specification using three simulated qualification builds.”
“Early in the AMNOW program,” Totin explains, “the team came to realize that there is no industry standard for report- ing the data coming from an AM machine. Enter AMNOW partner LECS Energy, which is providing its Learning Integrated Manufacturing System to col- lect process data during the builds.”
“LECS CEO Nat Frampton has been working with the machine builders to try to develop a standard for the data coming from the machines,” Totin says, “to lower the costs for the small manufacturer.”
3DMP
22 | 3D METAL PRINTING • SPRING 2022
3DMPmag.com