Turbine Outlet Guide Vane Flows

Doctoral thesis, 2007

New design goals on modern jet engines often leads to more highly loaded turbines with fewer stages in order to save weight and cost. This gives higher swirl angles into the outlet guide vane (OGV), implying a more difficult aerodynamic design. Structural requirements and weight reduction goals in recent designs of turbine rear frames (TRF) also often requires non-cylindrical shrouds with a three-dimensional polygonal shape and sunken engine-mount bumps. This has sparked a renewed interest in design methods and validation cases for LPT/OGV flows with complex end-wall geometries and engine mount bumps. Very little work has been conducted in the field of low pressure turbine outlet guide vanes (LPT/OGVs). This thesis concerns the aerodynamics around a highly loaded OGV, both with and without an engine mount bump. In the first part of the thesis a linear cascade was designed using new techniques including both CFD and classical analytical approaches. The test-facility proved to work as expected, and properties like load distributions, losses, outlet flow angles and the evolution of the secondary flowfield was extensively monitored. To get a better view of the downstream secondary flow development, traverse measurements for half the span and one pitch around the middle vane were performed for several downstream positions in the cascade. Since the experimental mesh was very resolved the streamwise vorticity could be analyzed for a better understanding of the secondary flow progress. For all these measurements numerical models as implemented in FLUENT were validated. The turbulence models tested in this thesis are the realizable k-ε model, the k-ω SST model, the Spalart-Allmaras model and the Reynolds Stress Model (RSM). The author would recommend the use of the realizable k-ε model in the design process for a new OGV design. The realizable k-ε model does not always resolve the flow perfectly well, but it is robust and converges well with good results. It is also relatively fast, which is important when many design iterations are needed. For thorough investigations of the flowfield the k-ω SST model is recommended. This model is superior to the other models in predicting the progress of the secondary flow. It was often used by the author to increase the understanding of the experimental results. Its results for the development of the flowfield around the engine mount bumps were very satisfying.