Browse Publications Technical Papers 05-13-03-0020

Modeling Dynamic Failure of Woven Carbon Fiber Thermoplastic Composites Using Empirical- and Multiscale-Based Material Cards in LS-DYNA MAT054 05-13-03-0020

This also appears in SAE International Journal of Materials and Manufacturing-V129-5EJ

This article focuses on the axial crush performance of woven carbon fiber/polyetherimide (CF/PEI) thermoplastic composites using MAT054 in LS-DYNA. The CF/PEI laminates are fully characterized empirically using mechanical test methods, and the experimental data is used to develop a MAT054 material card for LS-DYNA. A second MAT054 material card is developed separately using virtual test data gathered from a multiscale representative volume element (RVE) of the same woven CF/PEI composite. The two MAT054 material cards are implemented in LS-DYNA simulations of single hat-shaped tubular structures under impact. Using only constituent properties from material datasheets as inputs to the multiscale model, the homogenized tensile modulus of the multiscale RVE was within 4% of the experimental composite laminates. However, the multiscale methods overpredicted the tensile and compressive strain to failure and homogenized material strength of the composite laminate. Despite these differences in the two material cards, during an axial crushing simulation, the simulated peak impact load differed by less than 2%. The results showed that the initial elastic response of the composite structure was very similar for each of the two material cards. After failure began to occur, however, the higher strength of the multiscale-based material card resulted in higher predicted specific energy absorption (SEA) compared to the empirical-based material card for 5 mm and 10 mm thick laminates. For thicker laminates, the overprediction of strength by the multiscale-based material card resulted in nonphysical wrinkling in the composite structure due to incomplete failure through all layers of the composite structure.


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 19% off list price.
Login to see discount.