Finite element simulation of axial crushing of flat and C-channel composite coupons using a modified mesomodel

Abstract

The present article investigates the axial crushing behavior of IM7/8552 material system. Two types of specimens are considered-(a) flat coupons with saw-tooth trigger and (b) C-channels with 45° chamfer as a trigger. All the crushing test data (load, displacement and velocity history curves) is shared in the Crashworthiness working group (CWG) consortium of the Composite material handbook (CMH17.org). Flat crush coupons are tested at the University of Utah and the C-channels are tested at the John H. Glenn Research Center of the National aeronautics and space administration (NASA). Three types of layups are considered for flat coupons: QI1-[90/±45/0]2S, HL2-[902/02/±45/02] and HL3-[90/45/02/90/-45/02]. Two types of layups are considered for C-channels: HL2-[902/02/±45/02] and HL3-[90/45/02/90/-45/02]. Both the specimens are subjected to crush loading with an initial velocity and lumped mass attached to the impactor. All the ply and interface material properties come from the current authors' testing campaign except for strength and stiffness in the fiber direction. The crushing test cases are simulated in LS-Dyna software. Ply and interface behavior is described via user-defined material subroutines of LS-Dyna software. Each ply is modeled using 3D solid elements and the interfaces are modeled explicitly using cohesive elements. Intraply behavior is described using a novel modified meso model proposed by the present authors. Intraply model accounts for nonlinear damage, plasticity, ply fracture toughness and finite element mesh size. Interply response is modeled using bi-linear law. Predictions from present FE models showed a good correlation with crush test data. The effect of finite element model parameters e.g., filtering frequency, mesh size, interface properties, ply fracture toughness, element erosion criterion, contact stiffness and friction coefficient on crushing predictions is demonstrated by conducting parametric studies.