Advances In Super Plastic Forming Simulations

Advances in superplastic forming simulation with advanced finite element analysis have enabled timely qualitative and quantitative assessments of part and tool design with a decreased reliance on manufacturing trial and error. Implementing a model based engineering (MBE) and full digital thread approach to superplastic forming manufacturing processes allows for optimization of the design, including to maximize part quality and repeatability at a significant cost and time savings.

Finite element models (FEMs) are developed using the commercial software LS-DYNA. Simulations of the Titanium Superplastic Forming Diffusion Bonding (Ti-SPFDB) forming process are performed. This advanced analysis provides key insights into the complex interactions between the die (tooling surfaces) and pack (4-sheet Titanium part). The first phase of analysis includes assessment of die closure, leading to pivotal understanding of the mechanisms for wrinkle formation. After die closure, the FEM is utilized to analyze the pack behavior under pressurization. Complex forming of the core sheet cells will begin to form according to the pack and die design. Results provide key insight into how a particular die and pack design are likely to form in industrial settings. These predictions lead to early insights of potential risks in sheet wrinkling and regions of excessive thinning before the die and pack are manufactured. Understanding these risks a priori aids in panel design, allowing multiple design iterations and forming sensitivity on subscale trials studies before an actual part is formed.

This is a significant improvement over an approach that relies on trial-and-error, providing the opportunity to realize significant cost savings within the overall design and manufacturing process of Ti-SPFDB parts. Through improved simulation capabilities with LS-DYNA, these costs can be alleviated through an iterative MBE approach, in which simulation results advise necessary changes to tooling design and pack configuration, all of which occurs prior to real-world forming.