High-Fidelity Progressive Damage Predictions of Composite T-Joints With Seeded Defects in CB 2 ATA

View at publisher
Authors: Anand Karuppiah, Rui Li, Vijay Goyal, Jim Lua
Cited by: 0
References: 16

Abstract

High-fidelity progressive failure analyses of composite T-joints, both without and with seeded defects, are performed under pull-off tensile, side-bend, and mixed-shear loading using the Composite Bonded and Bolted Toolkit for Abaqus (CB2ATA). The CB2ATA model employs an advanced continuum damage formulation that incorporates physics-based LaRC04 damage criteria to predict matrix and fiber damage progression in T-joint specimens, along with a mixed-mode cohesive element method to model the initiation and propagation of interface delamination cracks. Given the relatively thick adhesive layer and its nonlinear material behavior, a nonlinear shear constitutive model is explicitly included in the simulation, along with inserted cohesive interfaces between the adhesive layer and the adjacent composite substrates. Feature-driven mesh refinement, accompanied by mesh convergence studies, is used to accurately capture stress concentrations at edges, the noodle region, and locations of seeded defects. A smearing approach based on lumped ply modeling is applied to non-critical regions of the composite skin to further improve computational efficiency. Model verification and validation are performed using Tee-stiffeners without and with Teflon inserts at the throat and tip locations. The tool demonstration is conducted through blind predictions of the load–deflection response, failure load, and damage mechanisms and sequences in T-joints, with the predictions compared against experimental data.

Fields of Study & Subjects

Mechanical Behavior of Composites Material Properties and Applications Fatigue and fracture mechanics

Publication details

Journal / source
AIAA SciTech Forum
Publication date
8 January 2026