Aerothermal Study of Intermediate Turbine Ducts

Doctoral thesis, 2016

An intermediate turbine duct guides the flow from the smaller diameter high pressure turbine to the downstream larger diameter intermediate or low pressure turbine. It is typically equipped with a structural vane transferring loads for the core to the external parts, and supplies the core with necessary services such as oil and air. Flow in an intermediate turbine duct is highly complex, influenced by the upstream turbine stage flow structures, which include tip leakage flows and non-uniformities originating from the upstream high pressure turbine vane and rotor. The complexity of the flow structures makes predicting them using numerical methods difficult, hence there exists a need for experimental validation. This thesis includes results from experiments conducted in two different facilities, the Chalmers Large-Scale Low-Speed Turbine Facility and the Oxford Turbine Research Facility. At Chalmers, the emphasis was at aerodynamic measurements including static pressure, total pressure, velocities and flow angles in the intermediate turbine duct. In Oxford, the experimental campaign included both aerodynamic and heat transfer measurements with the purpose of describing the aerothermal flow through the intermediate turbine duct. A new intermediate turbine duct with a medium turning vane was designed, manufactured, instrumented and installed in the Oxford Turbine Research Facility. Three high pressure turbine stage inlet conditions were studied; a uniform inlet flow and two low-NOx swirl profiles. Three adjacent intermediate turbine duct vanes were instrumented in an identical manner, to study the vane-to-vane clocking effect. Instrumentation includes static pressure tappings and thin film heat flux gauges installed on the intermediate turbine duct vane and endwalls at discrete locations. A three-hole pressure probe equipped with thermocouples above and below was traversed at the intermediate turbine duct exit. Steady and unsteady CFD predictions were performed of the intermediate turbine duct on its own and the full 1.5 high pressure turbine stage. The results from the test campaign in Oxford show the clocking effect to have a larger effect on the surface aerodynamics and heat transfer of the intermediate turbine duct vane than the introduction of a low-NOx swirl profile at the high pressure turbine stage inlet. The clocking effect is for example seen to change the incidence to the intermediate turbine duct vane. The change in inlet condition is seen to have local effects in the intermediate turbine duct. The CFD results are found to in general match the experimental data fairly well, both in aerodynamics and heat transfer. A comparison of steady and averaged time averaged unsteady CFD is noted to provide similar results.