The ever increasing use of composite materials in today’s society has created a drastic demand for better modeling of their behavior. The difficulty arises in that many modern composite structures are unique in shape and are exposed to a variety of loading situations. More specifically, loading scenarios which cause out-of-plane shear (Mode III) or mixed mode (Mode I + Mode II + Mode III) failure are of greatest challenge to model. This study investigates the capabilities of Simulation Composite Analysis (SCA), a composites software by Autodesk, in modeling failure in notched carbon-fiber composite panels loaded in Mode III. SCA was used with the finite element modeling software Abaqus/Standard (Dassault Systèmes) to model six different laminate stacking sequences. Three of the layups featured 40 plies through the thickness and the other three had 20 plies, with each containing either 10, 30, or 50 percent zero degree plies. The modeled panels were displaced as to create for a Mode III loading condition and the resulting maximum loads, load-displacement plots, and damage propagation outputs were compared to experimental results. It was found that SCA can determine the maximum failure load of the panels with an average of 11.6 percent deviation from experimental values. For one laminate stacking sequence in particular, the software determined maximum loads that deviated less than 1 percent from the experimental data. The load-displacement plots showed good correlations with experimental data in the linear region; however, the load-displacement behavior after damage was well modeled for only certain layups. The damage propagation paths for all the panel models were similar to the experimental panels in general, though self-similar damage propagation was not captured by the FEA models. Overall, Mode III failure in the notched carbon fiber panels was satisfactorily modeled for maximum load, but continued development is needed for predicting damage propagation paths. Modeling Mode III failure in composites is a difficult task; therefore, determining accurate methods in which to model such failure will be a substantial benefit to the composites engineering community. If low cost computer models can be established which accurately capture material damage and failure, the need for expensive and time-intensive experiments may be greatly reduced.
