Clearances between gas turbine casings and rotating blades is of quite importance on turbo machines since a significant loss of efficiency can occur if the clearances are not predicted accordingly. The radial thermal growths of the blade may be over or under predicted if poor assumptions are made on calculating the metal temperatures of the surfaces exposed to the fluid. The external surface of the blade is exposed to hot gas temperatures and it is internally cooled with air coming from the compressor. This cold air enters the radial channels at the root of the blade and then exists at the tip. To obtain close to realistic metal temperatures on the blade, the Conjugate Heat Transfer (CHT) approach would be utilized in this research. The radial thermal growth of the blade would be then compared to the initial guess. This work focuses on the interaction between the external boundary conditions obtained from the commercial Computational Fluid Dynamics software package CFX, the internal boundary conditions along the channels from a 1D flow solver proprietary to Siemens Energy, and the 3D metal temperatures and deformation of the blade predicted using the commercial Solid Mechanics software package ANSYS. An iterative technique to solve CHT problems is demonstrated and discussed. The results of this work help to highlight the importance of CHT in predicting metal temperatures and the implications it has in other aspect of the gas turbine design such as the tip clearances.

Flutter is an aeroelastic instability phenomenon that can result either in serious damage or complete destruction of a gas turbine blade structure due to high cycle fatigue. Although 90% of potential high cycle fatigue occurrences are uncovered during engine development, the remaining 10% stand for one third of the total engine development costs. Field experience has shown that during the last decades as much as 46% of fighter aircrafts were not mission-capable in certain periods due to high cycle fatigue related mishaps. To assure a reliable and safe operation, potential for blade flutter must be eliminated from the turbomachinery stages. However, even the most computationally intensive higher order models of today are not able to predict flutter accurately. Moreover, there are uncertainties in the operational environment, and gas turbine parts degrade over time due to fouling, erosion and corrosion resulting in parametric uncertainties. Therefore, it is essential to design engines that are robust with respect to the possible uncertainties. In this thesis, the robustness of an axial compressor blade design is studied with respect to parametric uncertainties through the Mu analysis. The nominal flutter model is adopted from [9]. This model was derived by matching a two dimensional incompressible flow field across the flexible rotor and the rigid stator. The aerodynamic load on the blade is derived via the control volume analysis. For use in the Mu analysis, first the model originally described by a set of partial differential equations is reduced to ordinary differential equations by the Fourier series based collocation method. After that, the nominal model is obtained by linearizing the achieved non-linear ordinary differential equations. The uncertainties coming from the modeling assumptions and imperfectly known parameters and coefficients are all modeled as parametric uncertainties through the Monte Carlo simulation. As iv compared with other robustness analysis tools, such as Hinf, the Mu analysis is less conservative and can handle both structured and unstructured perturbations. Finally, Genetic Algorithm is used as an optimization tool to find ideal parameters that will ensure best performance in terms of damping out flutter. Simulation results show that the procedure described in this thesis can be effective in studying the flutter stability margin and can be used to guide the gas turbine blade design.

This thesis documents an experimental investigation into the effect of trailing edge Injection on the aerodynamic performance of turbine blades conducted at Virginia Polytechnic Institute and State University (VPl&SU). A brief description of the arrangement, instrumentation and data acquisition system of the VPl&SU Transonic Cascade Wind Tunnel is given. Testing was conducted under a number of test conditions. Baseline data was obtained for the blades with no trailing edge injection. The blades were then tested for two different blowing rates to test the effect of blowing rate on the total pressure loss coefficient, L. Tests were conducted at a variety of save cascade exit Mach numbers ranging from 0.79 to 1.36. Measurements were taken at three locations downstream of the cascade blade trailing edges. The algorithm used to calculate the L from the measured data is discussed. Results of the testing indicate that trailing edge injection has a negligible effect on the total pressure loss coefficient. Correlations of cascade exit Mach number to L are given. The development of L downstream of the blade trailing edge is discussed. / Master of Science

This thesis documents the experimental research conducted on a transonic turbine cascade. The cascade was a two-dimensional model of a jet-engine turbine with an, approximately, 1.2 design, exit Mach number, and was tested in a blow-down type wind-tunnel. The primary goal of the research was to examine the effect of trailing edge thickness on aerodynamic losses. The original cascade was tested and, then, the blades were cut-back at the trailing edge to make the trailing edge thicker. The ratios of the trailing edge thickness to axial chord length for the two cascades were 1.27 and 2.00 percent; therefore, the ratio of the two trailing edge thicknesses was 1.57. To simulate the blade cooling method that involves trailing edge coolant ejection, and to examine the effect of that on aerodynamic losses, CO₂ was ejected from slots near the trailing edge in the direction of the flow. Two different blowing rates were used, in addition to tests without CO₂. A coefficient, L̅, was used to quantify aerodynamic losses, and this was the mass-averaged total pressure drop, normalized by dividing with the total pressure upstream of the cascade. The traversing, downstream total pressure probe was stationed at one of three different locations, in order to investigate the loss development downstream of the cascade. The two cascades were tested for an exit Mach number ranging from 0.60 to 1.36. The research suggested that the main influence of the trailing edge thickness on losses is through affecting the strength of the trailing edge shock system, since L̅ was almost the same for the two cascades in the subsonic Mach number region. The losses mainly differed (larger for the cut-back cascade) in the Mach number region of 1.0 to 1.2. In this region, the difference in loss maximized, showing a loss for the cut-back cascade 20 to 30 percent more than the original cascade. The CO₂ was found to have no significant effect for high Mach numbers; for low Mach numbers, the high blowing rate slightly decreased the loss. Finally, the loss, nearly, stopped to increase after one axial chord length downstream of the cascade. / Master of Science

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1979. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND AERONAUTICS. / Vita. / Includes bibliographical references. / Ph.D.

A 2-D CFD / FEM model to simulate thermal stresses in a turbine blade has been set up using the software FLUENT and FIDAP. The model was validated against the data of Bohn et. al. (1995) and was used to simulate 5 test cases. The numerical model was set up for a single Mark II nozzle guide vane (NGV) and utilised the appropriate boundary conditions for the surrounding flow field. A commercially available software code, FLUENT, was used to resolve the flow field, and heat transfer to the blade. The resulting surface temperature profile was then plotted and used as the boundary conditions in FIDAP (a commercial FEM code) to resolve the temperature and stress profile in the blade. An additional solver within FLUENT essentially superimposes an additional flow field as a result of the NGV vibration in the flow field. The pressure, temperature and heat transfer coefficient distribution, from FLUENT, were compared to those from Bohn et. al. (1995). The model predicted the distributions trends correctly, with an average over-prediction for temperature, of 10 % on the suction side and 6 % on the pressure side. This was restricted to the region from leading edge to 40 % chord on both sides of the blade. The blade temperature and equivalent stress contour trends were also correctly predicted by FIDAP. The blade temperature was over-predicted by and average of 1.7 %, while the equivalent stress magnitude was under-predicted by a worst case of 43 %, but the locations of maximum stress were correctly predicted. The reason for the differences between the stresses predicted by FLUENT / FIDAP and the data given in Bohn et. al. (1995), is believed to be the results of the temperature dependence of the material properties for the blade (ASTM 310 stainless steel), used in the two studies, not being identical. The reasoning behind this argument is because the distribution trends and contour variation, predicted by the model, compared favourably with the data of Bohn et. aI., and only the equivalent stress magnitude differed significantly. This completed the validation of the FLUENT / FIDAP model. The model was used to simulate test cases where temperature (i.e. turbine inlet temperature or TIT), at the model inlet (Le. the pressure inlet boundary in FLUENT), was set up to be time varying. Four simplified cases, viz single shock, multiple shocks, simplified cycle and multiple cycles, and a complex cycle (a mission profile) were simulated. The mission profile represented typical gas turbine operational data. The simulation results showed that stress was proportional to TIT. Changes in TIT were seen at a later time in the stress curve, due to conduction through the blade. Steep TIT changes, such as the shock loads, affected stress later than gentler TIT changes - the simplified and multiple cycles. These trends were consistently seen in the complex cycle. The maximum equivalent stress was plotted against TIT to try and develop a loose law that gives maximum equivalent stress as a function of TIT. A 4th order polynomial was fitted through the maxima and minima of the maximum equivalent stress plot, which gave the maximum and minimum stress as a function of TIT. This function was used calculate the maximum and minimum and mean equivalent stress using the TIT data for the mission profile. Thus, the FLUENT I FIDAP model was successfully validated, used to simulated the test cases and a law relating the equivalent stress as a function of TIT was developed. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2002.

A facility for the investigation of stall flutter in aircraft engine compressors and fans was designed. Stall flutter was achieved in the test fan and verified through sonic and photographic methods. The frequency components of the sonic output during flutter were determined using a real-time analyzer. This frequency analysis indicated a dominant peak within 7 percent of the theoretical torsional natural frequency of the blades. Photographs taken during stall flutter indicated the presence of an interblade phase angle. The effect of blade stagger angle, flow incidence angle and solidity on flutter speed was determined. / Master of Science

Gas turbine blade tip heat transfer and tip gap flow phenomena has been explored experimentally in a stationary cascade for blade exit Mach numbers = 1.2 to 1.4. Experimental results were found to agree well with qualitative predictions performed at GE Aircraft Engines. The pressure distribution in the blade tip cavity of a grooved tip blade was found to vary little with either Mach number or tip gap height. The tip cavity pressure was, however, a strong function of location. The tip cavity pressure distribution coupled with the pressure side distribution near the tip was speculated to drive the leakage flow across the blade tip from mid-chord aft based on surface flow visualization studies using an oil/dye mixture. Heat flux on the tip cavity floor was successfully measured using a thin-film Heat Flux Microsensor. Results of these measurements are consistent with previous studies in the subsonic regime. The convection coefficients on the tip cavity floor were found to be three times those found on the suction side airfoil surface near the trailing edge. Convection coefficients were found not to vary with either tip gap height or Mach number. The fluctuating component of heat flux was found to be at least 25% of the total heat flux. / Master of Science

The aerodynamic and thermodynamic performance of an aircraft gas turbine directly affects the fuel consumption of the engine and the life of the turbine components. Hence, it is important to be able to understand and predict the fluid flow and heat transfer in turbine blades to enable the modifications and improvements in the design process. The use of numerical experiments for the above purposes is becoming increasingly common. The present thesis is involved with the development of a flow solver for turbine flow and heat transfer computations. A 3-D Navier-Stokes code, the Moore Elliptic Flow Program (MEFP) is used to calculate steady flow and heat transfer in turbine rotor cascades. Successful calculations were performed on two different rotor profiles using a one-equation q-L transitional turbulence model. A series of programs was developed for the post-processing of the output from the flow solver. The calculations revealed details of the flow including boundary layer development, trailing edge shocks, flow transition and stagnation and peak heat transfer rates. The calculated pressure distributions, losses, transition ranges, boundary layer parameters and peak heat transfer rates to the blade are compared with the available experimental data. The comparisons indicate that the q-L transitional turbulence model is successful in predicting flows in transonic turbine blade rows. The results also indicate that the calculated loss levels are independent of the gridding used while the heat transfer rate predictions improve with finer grids. / Master of Science

The behavior of 8-harness satin woven Celion 3000/PMR-15 graphite-polyimide was experimentally investigated. Unnotched and center-notched specimens from (0)₁₅, (0)₂₂, and (0,45,0, - 45,0,0, - 45,0,45,0)₂ laminates were tested. Material properties were measured and damage development documented under monotonic tension, sustained incremental tension, and tension-tension fatigue loading. Damage evaluation techniques included stiffness monitoring, penetrant-enhanced X-ray radiography, laminate deply, and residual strength measurement. Material properties of the woven graphite-polyimide were comparable to those of woven graphite-epoxy. Damage development in woven graphite-polyimide was quite different than in non-woven graphite-epoxy. Matrix cracking was denser and delamination less extensive in the graphite-polyimide material system, and as a result, increases in notched residual tensile strength were much lower. A ply level failure theory was used to successfully predict the notched tensile strength of the (0,45,0, - 45,0,0, - 45,0,45,0)₂ laminate based on experimental data from the (0)₂₂ laminate. A simple method was used to simulate fatigue damage in a (0)₂₂ notched specimen to predict residual strength as a function of fatigue life. The advantages and disadvantages of the ply level failure theory used in this study are discussed. / Master of Science