Experimental study of flow through a transonic linear cascade

Authors

Idalia Jagodzińska - Warsaw University of Technology, Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Nowowiejska 24, Warsaw, 00-665, Poland
Bartosz Olszański - Warsaw University of Technology, Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Nowowiejska 24, Warsaw, 00-665, Poland
Konrad Gumowski - Warsaw University of Technology, Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Nowowiejska 24, Warsaw, 00-665, Poland
Slawomir Kubacki - Warsaw University of Technology, Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Nowowiejska 24, Warsaw, 00-665, Poland

Abstract

The flow in turbomachinery blading in transonic regime often encourages the fundamental phenomena like laminar-to-turbulent transition, shock wave-boundary layer interaction and shock-induced separation. The CFD techniques, in most industrial applications, often neglect the existence of the laminar boundary layers and the laminar-to-turbulent transition. The present work aims at providing the database for improvement of the RANS-based techniques. For that purpose, the pressure measurements on the blade surface were performed as well as a PIV method was applied to study the flow characteristics in the wake region behind the blade. The experiments have been done in the open circuit intermittent, in-draft (suction) type wind tunnel. The 2D model of the low-pressure turbine (LPT) linear cascade, consisting of 8 blades in VKI LS-59 arrangement, was placed in the wind tunnel test section. The Reynolds number (based on the inlet velocity and the blade chord) varied between 2.59 x 10^5 and 2.73 x 10^5. The outlet Mach numbers were equal to Ma_2is = 0.777 (subsonic flow) and Ma_2is = 0.975 (transonic flow). The inlet and outlet conditions, in terms of the static pressure, static temperature, inlet turbulence level and air humidity are provided. The pressure measurements on the blade surface were done for the pressure tappings in the mid-span plane of the blade. The PIV was performed using the Nd:YAG 200mJ Litron laser and LaVision DaVis software. The laser sheet covers the downstream region of the cascade including rear part of the blades suction side. The PIV measurements allowed for a detailed analysis of the flow field downstream the cascade. The vortex shedding was reported behind the blade in both subsonic and transonic cases. In transonic case, the interaction of the shock-wave with the separated boundary layer on the blade suction side was reported. The resulting flow unsteadiness caused a perturbation of the wake turbulence further away from the blade trailing edge. In addition, the interaction of the shock-wave with the wake turbulence was visible near to the blade trailing edge. The measurements were supplemented by numerical simulations using an algebraic intermittency model. The simulations were realized using both the RANS and URANS techniques. Good agreement was obtained between measured and predicted the Mach number distribution on the blade surfaces. Differences were reported between measurement and simulation downstream the cascade. The differences between the experiment and predictions are thoroughly discussed.