This project aimed to establish an additive manufacturing (AM) enabled investment casting process using ferrous scrap from forward operating bases (FOB). Ferrous waste materials were first characterized, and a waste material database was created to store critical information about scrap for sorting and recycling. A blending model was created to select scrap for remelting and making target steel alloys in the field.

In the investment casting process, stereolithography (SLA) printed patterns were used to replace the traditional wax patterns, which significantly increased the versatility and flexibility of the manufacturing process. Two types of SLA printers were studied and compared. Both printers were proven feasible in the established procedure. Zirconium silicate-based primary slurry and Zircon sand were used in the ceramic shell-making process. A dipping-stucco process was established to fabricate ceramic shells, and a burnout process was designed and optimized for SLA pattern removal. MagmaSoft was utilized to assist with the casting tree design. Three demonstrative components were successfully made using a similar compact casting system in the foundry at Worcester Polytechnic Institute.

To satisfy the need of the FOB, a three-container-based mobile foundry was designed which enabled rapid manufacturing of urgently needed parts in the fields. Casting facilities and inspection equipment were configured into three standard-size shipping containers. Thermal simulations were conducted in ANSYS software to ensure a safe operating environment. A list of commercially available microgrid systems and instruments needed in the mobile foundry was compiled with supplier information.

Data-driven tools have been established and studied for post-treatment after casting. An artificial neural network model was created and applied to study the relationship between a wide range of steel chemical compositions and multiple mechanical properties. The model used alloying element concentrations in steel and basic heat treatment information as inputs. The model outputs included mechanical properties, including yield strength, ultimate tensile strength, Brinell hardness number, and elongation.

Overall, this project successfully established an AM-enabled investment casting process that utilizes ferrous scrap from FOB, designed a mobile foundry system to meet urgent manufacturing needs, and created data-driven tools for post-treatment after casting. The findings of this project provide a solid foundation for future work in this area.