Failure Modeling and Simulation of Composites Subjected to Bypass and Bearing Loads

Abstract

An analytical model to predict the bypass and bearing strength of composites is developed. The model includes a contact algorithm, finite deformation theory, nonlinear material behavior, and the progressive damage of composites. The orientation and density of matrix-cracks and fiber damage are represented through a damage tensor, an internal state tensor that is thermodynamically consistent within the framework of continuum damage mechanics. The damage tensor evolves as damage associated with different failure modes accumulates. Matrix-cracks are predicted using a three-dimensional failure criteria, which is based on the stresses acting on a potential fracture plane. For use with the failure criteria, a new anisotropic shear constitutive law is postulated governing the transverse and the in-plane shear nonlinear mechanical behavior. For the prediction of initiation and progression of multiple delaminations through the thickness of the composite, a cohesive-decohesive constitutive law is adopted. The analytical model is implemented in a finite element commercial code via user subroutines. The problem of a notched composite loaded in tension is examined. The general characteristics of the predicted failure modes are in good agreement with the experimental observations obtained from the open literature. Numerical simulations are also conducted for the filled hole tension, double sided bypass and bearing load, and single sided bypass and bearing test configurations. The predicted strength and the strain behavior is in good correlation with in-house experimental data. The predicted failure modes are consistent with observations made in the open literature. A common challenge in the development of aircraft and spacecraft structures is maintaining structural integrity in the presence of mechanically fastened joints. This challenge is amplified when the structures include composite laminates, which have shortcomings from a microscopical and macroscopical standpoint. In the micro-scale, stress concentrations develop between stiff fibers and relatively compliant matrix material, and in the macro-scale stress concentrations develop near discontinuities such as a drilled hole. These stress concentrations lead to failure modes that can initiate at loads below the ultimate strength of the composite material. The focus of this paper is to quantify joint strength via an improved simulation of failure modes in composite structures. The design and analysis of bolted joints in composite structures is complex and uncertain because failure loads depend on a combination of factors such as material selection, stacking sequence, bolt clamping force, loading vector, geometric configuration, and manufacturing defects. The contact between the bolt and the bolt hole may induce large strains and high stress concentrations in the vicinity of the bolt hole boundary. Eventually an accumulation of localized failures ‐ such as fiber failure, delamination, and matrix-cracks ‐ that propagate from the edge of the bolt hole leads to the ultimate failure of the mechanically fastened joint. This complexity requires analytical procedures that can account for these variables in order to reasonably predict the response of new structures. Analytical work needs to be developed based on lessons learned from the extensive experimental work characterizing failure mechanisms that occur in mechanically fastened composite joints. 1‐6 Failure modeling of composite joints is challenging due to the multi-scale damage mechanisms that occur in composite laminates: micro-cracks in the micro-scale, and delaminations and through-the-thickness cracks in the macroscale. Significant work has been conducted in modeling and simulating the failure of composites using progressive