Turbomachinery turbine blade vibratory stress prediction

The objective of this thesis was to develop a methodology to predict the vibratory stresses of a turbomachinery turbine blade. The aerodynamic excitation phenomenon was not studied in this research. Furthermore, the turbine blades studied in this research had the characteristics of being unshrouded, uncooled and used mainly in small to medium sized turbomachineries. This thesis consists of four main subjects. The first subject, composed of an analysis of experimental results, was done to ex tract damping values of the turbine blades as well as resonances. The damping values extracted from the experimental data were used to determine analytical vibratory stresses with FLARES. Furthermore, the resonances were identified during the data reduction and therefore, the experimental vibratory stresses were extracted. These values were later used to correlate the analytical vibratory stresses predicted using FLARES. The second subject elaborated on an analytical method with finite element analysis using contact elements to determine natural frequencies and mode shapes of the turbine blades. The new analysis with contact elements surpassed all expectations with respect to the current analysis being performed at Pratt & Whitney Canada. The natural frequencies were compared with experimental data, and were found to be in good agreement. Furthermore, the mode shapes were compared with the current analysis results, and were found to be identical. The third subject describes an experimental method to test the blades in a controlled environment to extract natural frequencies, damping and mode shapes. The experimental testing was not performed with great success. The main deficiency was the excitable frequency range created by the high-frequency speaker. Finally, the fourth subject compares the vibratory stress experimental values and the prediction of vibratory stresses through an analytical tool. Using the FLARES tool, the modal amplification factor was found for every resonance of the PWC Engine 1 HPT Blade, PWC Engine 2 HPT Blade and the PWC Engine 3 CT Blade. Therefore, it can be concluded that the FLARES analytical tool can predict accurate vibratory stress levels due to a resonance for an unshrouded, uncooled turbomachinery turbine blade fairly well. More work needs to be done on the CFD part of the solution to predict more accurate unsteady pressure levels at the higher engine rotating speeds.