Accurate Material Modeling and Analysis of Fiber-Filled Thermoplastics to Enable Light Weighting in Automotive Applications

Usage of fiber-filled thermoplastics in automotive structural applications are increasing due to their inherent advantages over metal, which include lighter weight and simplification in assembly. However, accurately predicting the performance of a fiber-filled thermoplastic part can be challenging due to presence of non-linearity and anisotropy in the material behavior. This paper describes material characterization and modeling of fiber-filled thermoplastics for accurate prediction of part performance to enable rapid use of these lighter materials in automotive applications. The grade used for the study is a 30% glass filled PEI, SABIC’s ULTEM^TM 2300 Resin. Accuracy of the fiber orientation prediction is clearly demonstrated by the plaque level flow simulation validation with the CT-Scan data, followed by structural validations with specimen and part level tests. This study involves a representative cylinder-shaped part that accommodates different thicknesses and multiple manufacturing options. This part is tested for a variety of loading scenarios such as uniaxial tension, uniaxial compression and flexural loading to build confidence in the material data and modeling procedure. Results of simulations performed in widely used commercial software ABAQUS^TM and Digimat^TM compared against part tests show high level of accuracy in strength and stiffness predictions for 3mm thickness whereas other thickness specimens show deviations due to skin-core effect. The standard practice of qualifying a material model based on a single thickness specimen is also investigated in this work.