THE EFFECT OF MACH NUMBER ON LP TURBINE WAKE-BLADE INTERACTION

M. Vera, H. P. Hodson

Whittle Laboratory

University of Cambridge, UK

R. Vazquez

TTP, Industria de Turbo Propul sores

Madrid, SPAIN

Abstract The techniques employed in high speed linear cascade testing to simulate the effect of unsteadiness are presented and compared with low speed counterparts. Results are obtained from a high speed cascade and a low speed cascade. Both are models of an existing (conventional) low pressure turbine blade. They are compared under steady and unsteady fbw conditions. The results show that the same quantitative values of losses are obtained, proving the validity of the low speed approach for profiles with an exit Mach number of the order of 0.64. The range of validity of the conclusions is extended by reference to a profile designed using current low pressure turbine design practice. Wake traverses using pneumatic probes reveal that the unsteadiness reduced the profile losses up to Mach numbers of 0.9

1. Introduction

The development of the low pressure turbine (LPT) has reached a stage where rises in efficiency are difficult to obtain. Furthermore, the LPT could represent one third of the total engine weight. Therefore, one of the current trends that designers have adopted is to improve the overall performance of the LPT by reducing its weight. This new philosophy leads to fewer blades each of which carries a greater aerodynamic load.

The LPT operates at the lowest Reynolds number in the whole engine. This means that the development of the boundary layers will be determined by the transition from laminar to turbulent fbw. Increasing the aerodynamic load in­creases the diffusion on the rear part of the blade and thus the risk of separation. In fact, the resulting blade usually features a large separation bubble in steady ft>w conditions. This can be partially or totally suppressed by transition to tur­bulent ft>w being caused by the wakes coming from an upstream row of blades. Due to the ability of the wakes shed by an upstream row to promote transition

203

K. C. Hall et al. (eds.),

Unsteady Aerodynamics, Aeroacoustics and Aeroelasticity of Turbomachines, 203-216. © 2006 Springer. Printed in the Netherlands.

in the neighborhood of separation, the study of wake-boundary interactions is of primary interest in the LPT environment. The large aspect ratio of the LPT blades (typically between 3 and 7) makes them appropriate for linear cascade testing.

Previous studies on wake induced transition phenomena for high speed fbws have been carried out by others researchers (Brunner et. al, 2000 and Coton et. al, 2002) but still, very little is known about the topic. This fact together with the current limitations of the CFD tools (Vilmin et at, 2003) suggests that experimental studies of blade-wake interaction phenomena in high speed fbws are needed. Even though the experimental techniques used in high speed testing are conceptually the same as those used in low speed, the problems encountered tend to be magnified at high speed. In addition, new challenges arise. Therefore, we should question when it is worth doing high speed cascade testing.

This paper aims to answer this question. It examines the extent to which the results from the low speed approach are meaningful. This paper follows pre­vious comparisons between high speed and low speed testing, (Wisler, 1984), (Hodson and Dominy, 1993), by presenting a comparison between high speed and low speed testing of LPT linear cascades. In the main part of this paper, two profiles are compared. These profiles are not identical but they both are intended to model an existing LPT blade in the cascade environment at low speed or at high speed. Facilities and instrumentation will also be compared. Finally, results from a modern LPT blade will be presented to demonstrate the extent to which the conclusions are valid. To conclude, the behavior of this blade is presented up to exit Mach numbers of 0.9.