Predicted Turbine Heat Transfer For A Range Of Test Conditions

Author : Suzanne A. Southworth
Publisher :
ISBN 13 :
Total Pages : 200 pages
Book Rating : 4.:/5 (838 download)

DOWNLOAD NOW!

Download or read book Comparison of Steady and Time-accurate Predictions with Experiment for the Aerodynamics of a Fully Cooled Single-stage High-pressure Turbine written by Suzanne A. Southworth and published by . This book was released on 2006 with total page 200 pages. Available in PDF, EPUB and Kindle. Book excerpt: Abstract: The aerodynamics of a fully cooled single stage high-pressure turbine operating at design corrected conditions has been the subject of a thorough study involving experimental and computational work. The experimental configuration included a fully cooled, full-stage high-pressure turbine stage operating at design corrected conditions of corrected speed, flow function, and stage pressure ratio. The data reported in this thesis were obtained for a relatively low vane inlet Reynolds number condition in order to test the limits of the experimental system. The vanes and blades of the turbine and the stationary shroud immediately above the blade were instrumented. The vanes were instrumented with heat-flux gauges and pressure transducers at one span wise external location and two internal locations. The blades were instrumented with heat-flux gauges and pressure transducers at three different span wise external locations to provide external heat-flux and surface-pressure distributions, and internally with pressure transducers and miniature thermocouples to provide pressure and temperature histories within the airfoil cooling cavities. The stationary shroud was instrumented with heat-flux gauges and pressure transducers at several axial chord locations. Kulite pressure transducers were used to obtain all of the pressure data and double-sided Kapton heatflux gages were used for the heat transfer measurements on the vanes and blades while button-type gauges were used in the stationary shroud. Aerodynamic predictions were obtained using the computational fluid dynamics (CFD) codes Numeca's FINE/Turbo and Mississippi State University (MSU) Turbo, but for both calculations the addition of the film cooling gas was ignored. Further calculations are in progress that do include the film cooling, but results from those calculations are not reported herein. Both of these codes are 3D viscous codes, but FINE/Turbo was used to obtain both steady and time-accurate results while MSUTURBO was used to obtain only time-accurate results. Both FINE/Turbo and MSU Turbo utilize phase lagged boundary conditions to simplify the model and significantly reduce computing time and resources. The unsteady loadings, as predicted and measured, are compared for the blade, vane, and shroud as time-averaged, time series, and power series data. The steady CFD prediction was also obtained so as to provide the initial boundary conditions for the unsteady prediction. Therefore, comparisons of the steady CFD predictions and the time- averaged data were also made. The blade included a recessed tip geometry, which was included in the CFD model, and the CFD analysis also investigated different tip/shroud clearances to investigate the influence of the tip cap height on the downstream flow field. The analysis shows that both the steady state and time-accurate pressure predictions compare quite well with the experimental results. The steady and timeaveraged vane predictions are closer to the data than the steady prediction for the blade, but that is to be expected with the unsteady nature of the flow over the blade. The timeaccurate prediction for the blade is in very good agreement with the experimental results. The FINE/Turbo heat transfer predictions (MSU Turbo doesn't do heat-transfer predictions at the present time) are not showing similar trends as the data but this is likely due to the fact that the film cooling was ignored and due to the grid density utilized for these calculations. Since this work is meant to be a true prediction without any model optimization, it is interesting to see where the CFD produces the best results and where it has the most difficulty. It is also important to keep in mind that this work compares an uncooled CFD prediction to data obtained for a fully cooled turbine stage. This is the first time such data has been available and the uncooled CFD predictions will lend insight into the importance of cooling modeling, which has yet to become developed to a point that it is widely used in CFD predictions. Overall, the comparisons made here demonstrate the ability of two different CFD codes to successfully capture the unsteady flow physics on the blade surface and in the blade tip/stationary shroud region.