Process-to-Performance Simulation of 3D Woven Composite T-Joints

Abstract

Translate Article

View Video Presentation: https://doi.org/10.2514/6.2021-0313.vid The realization of practical, large integrated, bonded composite structures have long been a goal of the composite aircraft design community. One enabler for this goal is the ability to robustly bond composite skins to stiffeners. Typically, this is accomplished with T-joints manufactured using unidirectional prepreg material formed into a T shape by placing L-bend shapes back to back. These interface structures suffer from problematic delamination and manufacturing issues associated with the tight radii bends. One solution is the manufacture of T-joints using composites constructed with 3D woven textiles that are near-netshape in their architectures. These materials forgo the clear delamination paths by introducing fiber tows that are integrally woven through the preform, binding the out-of-plane directions together. Many complex 3D textile weave architectures are available that can potentially accomplish these desires. However, the design and analysis of such composite joints is a challenge due to the extreme complexity of the architecture and potential for manufacturing-induced defects. The work described here focuses on the process modeling of a planar portion of a T-joint, culminating in a stress analysis of the as-processed 3D preform accomplished at the geometric level of individual tows, or the meso-level. Specifically, the compaction process of a 3D ply interlock architecture was simulated and the resulting geometries ported to a stress analysis package. Additionally, it is desired to embed this tow-level stress analysis model within a global, homogenized model to inject global loads at the local level. The ultimate goal of this research effort is to combine textile preform compaction simulation with cure-process simulations to produce a process-to-performance workflow for these complex composite structures, while allowing for potential deviations from the optimum process.