Prediction of spring-back deformation associated with aeronautical composite parts after curing process

Abstract

The aerospace industry is increasing its utilization of composite-made structures for safety and efficiency reasons but what comes with that is a phenomenon called spring-back which is defined as the deviation of the finished part from the intended design. Thus, there is a critical need for a reliable method to prevent and correct the spring-back problem. The present work employs Finite Element Analysis (FEA) to predict the spring-back behavior of composite structures by modelling the stretching of the first ply of the laminate and integrating it with an interface component between the laminate part and tool. The research study is conducted in 3 phases with pre-impregnated laminates. The first phase is for flat laminates of various sizes and thicknesses. A good agreement was obtained between the FEA simulation results and the experimental data, particularly for the smaller and thinner samples. The predominant mechanism that contributed to the spring-back warpage is found to be the in-plane stress from the ply stretching. The second phase is for curved laminates with the critical parameters identified in the flat phase being maintained with an added mechanism referred to as the Corner Effect. It was discovered that the warpage is due to the in-plane stress contributed predominantly by the corner stretching. The remainder is due to the tool-laminate interface and it has been shown that its properties are independent of the laminate geometry. The third and final phase is to study the effect of angled plies on spring-back behavior. The experimental results showed that the inclusion of angled plies in the laminate significantly raises the severity of the spring-back warpage and that orientation of the deformation is largely aligned by the fiber orientation of the first ply which reinforces the hypothesis from the first phase and overall research. Finally, the scope of the current research study only concerns with predicting the spring-back behavior but as an extension for future study, the author recommends future works focusing more on correcting the spring-back deformation via tool modification by establishing yet again both an experimental and simulation base for validating the FEA model produced in the current study for all 3 phases.