Total Pressure Loss Reduction in Annular Diffusers

Abstract

Power output and efficiency of gas turbines depend strongly upon the achievable pressure rise in the subsequent diffuser. In combination with the requirement to keep diffuser length to a minimum, ever steeper opening angles are sought, while avoiding diffuser stall. In terms of diffuser pressure rise, the boundaries of what is achievable can be pushed further if the tip leakage vortices from the last stage are used to re-accelerate the diffuser boundary layer, thus delaying separation onset. Such measures have been shown to decrease total pressure losses as well. In this paper, we show that the benefit of total pressure loss reduction in vortex-stabilised diffusers becomes more pronounced for steeper opening angles by means of a numerically and experimentally validated approach. In extension, we provide evidence that the loss production in highly loaded vortex-stabilised diffusers, which would stall otherwise, can be brought down to the level of non-stalling diffusers. Furthermore, we present a detailed analysis of the different loss mechanisms and their response to vortex-stabilisation of the diffuser. NOMENCLATURE Symbols � cross-sectional area of the diffuser AR area ratio of the diffuser �, � flow velocity �� pressure recovery coefficient �� specific isobaric heat capacity �r reduced frequency h enthalpy (default: static) l chord length meridional coordinate number of blades � rotational speed in revolutions per minute � pressure (default: static) Euler radius � specific gas constant � temperature (default: static) � rotational velocity � generalised spatial coordinate � axial coordinate � flow angle, whirl angle curve diffuser half-opening angle Δ difference diffuser effectiveness total pressure loss coefficient circumferential coordinate Lamé constant Λ loss rectification number dynamic viscosity � kinetic energy coefficient � rectified total pressure loss coefficient � density � generalised spatial vector Σ stabilisation number , Ψ flow coefficient, loading coefficient Subscripts I, II rotor inlet/outlet plane eff effective corr correlated in, out diffuser inlet/outlet ref reference rel relative t turbulent quantity tot total quantity enthalpy-induced dilatational shearing-induced thermodynamic � vorticity-induced INTRODUCTION Without the use of exhaust diffusers, the expansion of hot gas achievable in turbines is bounded by the ambient pressure. Only the subsequent conversion of kinetic energy into static pressure, realised by an increase in cross-sectional area in the diffuser, allows for considerably higher expansion ratios in the turbine. As a consequence, the power output and—assuming constant heat input—efficiency increase. The resulting main aerodynamic design goal of exhaust gas diffusers is to convert as much kinetic energy as possible into static pressure, i.e., maximise the ratio of the static pressure rise over the diffuser to the kinetic energy at diffuser inlet. Diffuser designers tend to call this ratio pressure recovery and express it in terms of the non-dimensional pressure recovery coefficient �� = � − �i � ,i − �i . (1) International Journal of Gas Turbine, Propulsion and Power Systems April 2019, Volume 10, Number 2 Manuscript Received on October 2, 2018 Review Completed on April 10, 2019 Copyright © 2019 Gas Turbine Society of Japan