FAILURE OF A COMPOSITE CRUCIFORM SPECIMEN UNDER BI-DIRECTIONAL LOADING

Som R. Soni, Mohan Balan, Charles Cross

ABSTRACT

Glass/VCAP-75 and Glass/AMB-21 woven fabric, polymer matrix composites (PMCs), are potential high temperature materials being considered for aircraft engine applications. Both of these material systems have been subjected to a considerable amount of mechanical testing, including bi-directional combined tension and bending as well as bi-directional in plane loading. For both material systems, the 12-ply laminate orientation is (0/45)3S. All the effective properties (compressive, tensile and shear) of these materials have been measured by AdTech (1), and these properties are utilized for predicting the laminate behavior.

This paper deals with the stress and strength analysis of cruciform specimen shown in Figures 1 and 2. Both analytical and experimental studies are conducted. A general purpose FEA software (SDRC I-DEAS) is used to develop a finite element model shown in Figures 7 and 8 to conduct the stress analysis of a composite specimen under bi-directional loading conditions. The loading is in the x and y directions of the specimen. At the intersection of the cruciform legs, two types of curvatures are considered to create different stress concentrations. One increases the stress concentration while the second decreases the stress concentration at the intersection. The representative case finite element results for maximum stress components for both kinds of specimens are given in bar chart form. These results give an indication of the possible mode and location of failure. The experimental results are compared with the FEA predictions. Failure loads for specific loading is calculated by using the computer code (2) called, “Automated System for Composite Analysis”, (ASCA). The newly developed unique multi-axial testing system in the Turbine Engine Fatigue Facility of the Propulsion Laboratory, Air Force Research Laboratory (AFRL/PRTC) at Wright Patterson Air Force Base is used. There exists a good agreement between experimental and predicted results.