Abstract
Energy release rates obtained by geometrically linear and geometrically nonlinear finite element analyses of homogeneous, symmetrically delaminated single leg bending specimens are presented for a variety of materials, specimen geometries and fixture dimensions. It is shown that certain test geometries will exhibit strong nonlinear effects; thus, critical energy release rates obtained from tests of these geometries, using data reduction procedures that are based on linear theory, may contain significant errors. The nonlinear finite element results are used to develop empirical relationships between energy release rate as predicted by the nonlinear analyses and those predicted by linear analyses. These empirical relationships are shown to be valid over a wide range of specimen material properties, material property ratios (e.g., Young's modulus to shear modulus) and geometric properties of both the specimen and fixture, including fixture roller diameters. Thus, the empirical relationships may be used in a quantitative manner to design tests in order that significant nonlinear effects do not occur prior to fracture, and hence linear data reduction procedures remain valid. Alternatively, the empirical relationships may be used to interpret test results where nonlinear behavior occurs. Both uses are illustrated by example for typical laminated composite materials.