Comparing the effect of geometry and stiffness on the effective load paths in non-symmetric laminates

© 2019, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. In aerospace composite material design, it is common to encounter load bearing components that vary in thickness across their length. In plate design, ply drops, tow-steering, and the addition of stiffeners, all act to change both the section geometry and the effective stiffness of the part. Often, due to aerodynamic design constraints, the geometric profile must transition non-symmetrically, i.e. thickness is built up from a reference surface, meaning the mid-surface of the plate does not remain on a constant plane. These localised changes in geometry, and associated change of position of the mid-surface, lead to inherently three-dimensional states of stress. As a consequence, and especially for composite structures, there is the potential for significant through-thickness stresses and/or stress concentrations, leading to failure—for example debonding or delamination. By investigating the effects of geometric and effective stiffness changes, we are able to gain physical insight into structural behaviour in the regions of geometric transition. This is achieved through a parametric study, whereby we compare the behaviour as predicted by Classical Laminate Theory—a commonly utilised two-dimensional approach—with a finite element analysis based on the Unified Formulation by Carrera and co-workers. Based on these investigations, we are able to illustrate how rates of profile change and/or stiffness variation are linked to variance in the predicted location of the neutral plane of the two approaches which acts as a proxy measure for predicting through-thickness behaviour. Finally, we discuss the potential opportunity to utilise laminates that possess non-standard layups to tailor the load path through geometric transitions, thus offering a potential route for elastic tailoring that minimises undesirable through-thickness stresses.