Multi-Scale Modeling of Failure of Continuous Carbon Fiber Composites for Virtual Allowables

Aerospace and automotive industries are increasingly using continuous carbon fiber composite to answer lightweighting challenges and meet high performance requirements. Such composite materials combine a high strength and stiffness with low density, making them competitive compared to classical metal based solutions. But this unique combination of material properties comes along with increased complexity across several scales, requiring extensive material characterization. Starting at the constituent level, matrix and fiber behavior drive the ply response, which typically shows anisotropy and non-linearity. Failure mechanisms are various (matrix cracking, fiber breakage, fiber-matrix debonding, delamination). Different ply stacking sequences must be investigated in the design process. And various laminate tests must be performed, such as unnotched tension/compression and open-hole tension/compression. Across all scales variability must be considered, such as constituent property variability or fiber volume fraction variability. This overall complexity results in extensive experimental coupon testing, expensive from both a time and cost point of view. Hence there is a strong need for a simulation solution which predicts composite coupon properties, replacing physical testing by virtual testing. This paper describes the technology behind the simulation software recently developed by e-Xstream engineering aiming at generating virtual allowables. The software offers micromechanics and continuum damage to describe the complex composite material behavior combined with a non-linear finite element analysis generator and solver to efficiently predict notched and unnotched coupon testing results and cover any desired test matrix. Variability can be applied to micro-level parameters such as constituent properties or fiber amount in order to generate A or B-basis allowables, thereby significantly reducing the amount of required physical tests and allowing a deeper understanding of the material parameters driving the composite strength.