Fiber-reinforced ceramic-matrix composites (CMCs) depend upon inelastic mechanisms to diffuse stress concentrations associated with holes, notches, and cracks. These mechanisms consist of fiber debonding and pullout, multiple matrix cracking, and shear band formation. In order to understand these effects, experiments have been conducted on several double-edge-notched CMCs that exhibit different stress redistribution mechanisms. Stresses have been measured and mechanisms identified by using a combination of methods including X-ray imaging, edge replication, and thermoelastic analysis. Multiple matrix cracking was found to be the most effective stress redistribution mechanism.

Summary Thermoelastic stress analysis was used to document the effect of composite damage on the stress distribution in three ceramic matrix composites. Composite damage was found to significantly alter the thermoelastic response of each material, with the greatest effect noted in SiC/CAS. Thermoelastic imaging of these materials affords a more complete picture of how the various damage mechanisms affect the stress distribution. In particular, a stress concentration factor computed from thermoelastic images, serves as an indicator of stress redistribution. The stress concentration factors were computed by comparing notch root to far field temperatures, and monitored after the introduction of various amounts of damage. In each material, the stress concentration factor diminished as the damaging load approached the ultimate stress. Reduction in the stress concentration is associated with local changes in modulus, mechanistically arising from combinations of fiber, matrix and interface fracture. Stress redistribution occurs as a consequence of modulus changes, leading to lower notch sensitivity in each of the tested composites.

Summary The SPATE system provided a means for experimental ther‐moelastic analysis of high frequency modes of vibration of a complicated turbine impeller which did not facilitate analysis with conventional methods due to low stress response and high modal density. Testing also demonstrated a unique experimental stress analysis technique which was applied to a blade and vane airfoil and exhibited excellent stress distribution correlation with an established experimental technique (brittle lacquer) and with analytical stress distributions predicted by NASTRAN. Full field dynamic stress analysis of gas turbine compressor airfoils accomplished using thermoelastic techniques provided an efficient method of determining optimum strain gage locations for monitoring rig and engine vibratory stress. Significant cost savings could be realized using this technique from: (1) reduction in the amount of instrumentation required to characterize airfoil dynamic behavior and (2) increased vibratory response mode coverage thereby preventing the re‐running of engine test programs to measure vibratory stress response levels of unexpected resonances encountered during initial engine testing. Care is required when using SPATE results for vibratory modes resulting in high bi‐axial stress distributions since the system output indicated stress magnitudes based on the sum of the principal stresses which could be significantly higher than either of the individual principal stresses.