Endwall Loss Reduction of High Lift Low Pressure Turbine Airfoils Using Profile Contouring: Part I — Airfoil Design

Abstract

This paper presents the reasoning for and the design process of contouring a high lift front-loaded low pressure turbine (LPT) airfoil near the endwall to reduce the endwall loss. The test airfoil, L2F, was designed to the approximate gas angles with 38% larger pitchwise spacing than the widely studied Pack B airfoil. Being more front-loaded with a higher stagger angle, L2F is shown to produce more endwall losses than Pack B. It is suggested that the high endwall loss of L2F is due to the high stagger angle, not front-loading as usually suggested in the literature. A procedure is presented to approximate the front-loading and stall resistance of L2F and obtain a low stagger version of that airfoil, designated L2F-LS. A contoured airfoil is then designed by transitioning L2F into L2F-LS at the endwall to obtain a benefit from the reduced stagger angle at the endwall. The contouring process generates a fillet, so the contoured airfoil is referred to as L2F-EF (“Endwall Fillet”). Predictions in this paper suggest endwall loss reductions between 17% and 24% at Re = 100,000. Linear cascade experiments in Part II [1] of this paper indicate that L2F-EF reduces endwall losses more than 20% compared to L2F. The overall conclusion is that the stagger angle has a significant effect on endwall loss and should be considered for designing high lift LPT airfoils at the endwall.

Investigation of Low-Pressure Turbine Endwall Flows: Simulations and Experiments

“…In addition, the L2F-EF geometry (L2F with endwall fillet) was considered which blends from a L2F airfoil at mid-span to a L2F-LS at the endwall. 24 The blending extends over 9.5% of the total span (=0.3325C x ) at both endwalls. The simulations were set up according to the AFRL experiments where upstream wakes and surface roughness were not considered.…”

“…The Reynolds number based on axial chord, C X , for the simulations was Re=100,000 and thus the same as in the experiments and earlier RANS calculations. 24,25…”

“…In the RANS calculations by Lyall et al 24 a turbulent boundary layer with a thickness of δ 99 =0.025H=0.0875C x was prescribed at the inflow boundary which was located at x=-1.4C x . The following approximations hold for a turbulent flat plate boundary layer:…”

“…x . From the inflow velocity profiles of the Lyall et al calculations 24 the displacement and momentum thickness were computed as δ * =0.012543C x and θ=0.0092353C…”

“…The AFRL decided to investigate profile contouring to reduce the endwall losses (e.g., Lyall et al 24 and Sangston et al 25 ). A new geometry was designed that blends from a L2F profile inside the passage to a low-stagger L2F-LS airfoil near the endwall.…”