A novel dual-stage failure criterion based on forming limit curve for uncured GLARE

Abstract

Forming limit curve (FLC) is an essential criterion for evaluating the formability of fiber metal laminates (FMLs) in mass fabrication. However, conventional methodologies are unable to predict inner fiber failure, very sensitive to wrinkling defects, resulting in low measurement accuracy and stability. Hence, a novel dual-stage failure criterion considering the progressive failure of fiber and metal layers has been developed and validated by utilizing a notch-pattern sample and forming of typical thin-walled structure, with four typical categories of glass laminate aluminum reinforced epoxy (GLARE). The Nakajima tests of a series of notch-pattern and conventional pattern specimens with different widths were conducted in this study, demonstrating the superiority of wrinkling and delamination prevention of the notch-pattern by enhancing 22.84 % stress triaxiality and reducing 39.51 % compress strain in the edge area. Notably, punch reaction force and strain field analyses unveiled a mutation phenomenon indicative of premature fiber failure while the aluminum remained intact, which was also verified by numerical and interrupted methods. Furthermore, the acoustic emission in-situ method was adopted to monitor the progressive damage evolution of uncured GLARE in the Nakajima test, providing solid support for the occurrence of fiber failure at the mutation site. Leveraging these mutations, a list of dual-stage forming limit curves was devised to quantify the progressive damage behavior of uncured GLARE under biaxial tensile conditions. Finally, the practical application of this novel dual-stage forming limit curve methodology was demonstrated in the manufacturing process of a box-shaped part, serving as a proof of concept for its effectiveness. This study offers a foundational guideline for characterizing the intrinsic failure mechanisms of fiber metal laminates, particularly internal fiber failure under various strain conditions, thereby enhancing predictive accuracy.