Secondary Flows and Losses in a Highly Loaded Low Aspect Ratio Transonic Axial Flow Turbine Stage

Author(s):  
Ananthakrishnan Kaliyaperumal ◽  
M. Govardhan
Author(s):  
K. L. Lewis

Selected experimental results, obtained from a detailed investigation into the flowfields within two low speed multistage turbines, are presented. A repeating stage condition occurred typically after two stages, with the secondary flows an important factor in the low aspect ratio geometry. A tracer gas technique was employed to identify the dominant mechanisms of spanwise transport and their relative significance. In the first stages of both machines, tracer transport was more intense near the endwalls than at mid-span whilst in the multistage environment the transport was approximately constant across the whole span. The convective influence of classical secondary flow, shroud leakage and wake passage through a downstream blade was identified and shown to be as significant as turbulent diffusion in effecting cross-passage and spanwise transport. The data show that spanwise transport should be included within any throughflow model and are used to calibrate two scaling models. These models are presented in Part 2 where the influence of incorporating spanwise transport into a throughflow model is investigated.


1994 ◽  
Vol 116 (2) ◽  
pp. 179-186 ◽  
Author(s):  
K. L. Lewis

Selected experimental results, obtained from a detailed investigation into the flow fields within two low-speed multistage turbines, are presented. A repeating stage condition occurred typically after two stages, with the secondary flows an important factor in the low aspect ratio geometry. A tracer gas technique was employed to identify the dominant mechanisms of spanwise transport and their relative significance. In the first stages of both machines, tracer transport was more intense near the endwalls than at midspan, while in the multistage environment the transport was approximately constant across the whole span. The convective influence of classical secondary flow, shroud leakage, and wake passage through a downstream blade was identified and shown to be as significant as turbulent diffusion in effecting cross-passage and spanwise transport. The data show that spanwise transport should be included within any throughflow model and are used to calibrate two scaling models. These models are presented in Part 2, where the influence of incorporating spanwise transport into a throughflow model is investigated.


Author(s):  
Christopher Clark ◽  
Graham Pullan ◽  
Eric Curtis ◽  
Frederic Goenaga

Low aspect ratio vanes, often the result of overall engine architecture constraints, create strong secondary flows and high endwall loss. In this paper, a splitter concept is demonstrated that reduces secondary flow strength and improves stage performance. An analytic conceptual study, corroborated by inviscid computations, shows that the total secondary kinetic energy of the secondary flow vortices is reduced when the number of passages is increased and, for a given number of vanes, when the inlet endwall boundary layer is evenly distributed between the passages. Viscous computations show that, for this to be achieved in a splitter configuration, the pressure-side leg of the low aspect ratio vane horseshoe vortex, must enter the adjacent passage (and not “jump” in front of the splitter leading edge). For a target turbine application, four vane designs were produced using a multi-objective optimization approach. These designs represent: current practice for a low aspect ratio vane; a design exempt from thickness constraints; and two designs incorporating splitter vanes. Each geometry is tested experimentally, as a sector, within a low-speed turbine stage. The vane designs with splitters geometries were found to reduce the measured secondary kinetic energy, by up to 85%, to a value similar to the design exempt from thickness constraints. The resulting flowfield was also more uniform in both the circumferential and radial directions. One splitter design was selected for a full annulus test where a mixed-out loss reduction, compared to the current practice design, of 15.3% was measured and the stage efficiency increased by 0.88%.


Author(s):  
K. L. Lewis

In Part 1 of this paper, a repeating stage condition was shown to occur in two low aspect ratio turbines, after typically two stages. Both turbulent diffusion and convective mechanisms were responsible for spanwise transport. In this part, two scaling expressions are determined that account for the influence of these mechanisms in effecting spanwise transport. These are incorporated into a throughflow model using a diffusive term. The inclusion of spanwise transport allows the use of more realistic loss distributions by the designer as input to the throughflow model and therefore focuses attention on areas where losses are generated. In addition, modelling of spanwise transport is shown to be crucial in predicting the attenuation of a temperature profile through a turbine.


1984 ◽  
Author(s):  
Francis Leboeuf

A computational method for secondary flows in a compressor has been extended to treat stalled flows. An integral equation is used which simulates the inviscid flow at the wall, under the viscous flow influence. We present comparisons with experimental results for a 2D stalled boundary layer, and for the secondary flow in a highly loaded stator of an axial flow compressor.


2004 ◽  
Vol 128 (3) ◽  
pp. 492-499 ◽  
Author(s):  
Graham Pullan ◽  
John Denton ◽  
Eric Curtis

Experimental data and numerical simulations are presented from a research turbine with low aspect ratio nozzle guide vanes (NGVs). The combined effects of mechanical and aerodynamic constraints on the NGV create very strong secondary flows. This paper describes three designs of NGV that have been tested in the turbine, using the same rotor row in each case. NGV 2 used three-dimensional design techniques in an attempt to improve the performance of the datum NGV 1 blade, but succeeded only in creating an intense vortex shed from the trailing edge (as previously reported) and lowering the measured stage efficiency by 1.1% points. NGV 3 was produced to avoid the “shed vortex” while adopting a highly aft-loaded surface pressure distribution to reduce the influence of the secondary flows. The stage with NGV 3 had an efficiency 0.5% points greater than that with NGV 1. Detailed comparisons between experiment and computations, including predicted entropy generation rates, are used to highlight the areas where the loss reduction has occurred and hence to quantify the effects of employing highly aft-loaded NGVs.


Author(s):  
Ja´nos Vad ◽  
Ali R. A. Kwedikha ◽  
Helmut Jaberg

Experimental and computational studies were carried out in order to survey the energetic aspects of forward and backward sweep in axial flow rotors of low aspect ratio blading for incompressible flow. It has been pointed out that negative sweep tends to increase the lift, the flow rate and the ideal total pressure rise in the vicinity of the endwalls. Just the opposite tendency was experienced for positive sweep. The local losses were found to develop according to combined effects of sweep near the endwalls, endwall and tip clearance losses, and profile drag influenced by re-arrangement of the axial velocity profile. The forward-swept bladed rotor showed reduced total efficiency compared to the unswept and swept-back bladed rotors. This behavior has been explained on the basis of analysis of flow details. It has been found that the swept bladings of low aspect ratio tend to retain the performance of the unswept datum rotor even in absence of sweep correction.


2012 ◽  
Vol 134 (6) ◽  
Author(s):  
G. Persico ◽  
A. Mora ◽  
P. Gaetani ◽  
M. Savini

In this paper the three-dimensional unsteady aerodynamics of a low aspect ratio, high pressure turbine stage are studied. In particular, the results of fully unsteady three-dimensional numerical simulations, performed with ANSYS-CFX, are critically evaluated against experimental data. Measurements were carried out with a novel three-dimensional fast-response pressure probe in the closed-loop test rig of the Laboratorio di Fluidodinamica delle Macchine of the Politecnico di Milano. An analysis is first reported about the strategy to limit the CPU and memory requirements while performing three-dimensional simulations of blade row interaction when the rotor and stator blade numbers are prime to each other. What emerges as the best choice is to simulate the unsteady behavior of the rotor alone by applying the stator outlet flow field as a rotating inlet boundary condition (scaled on the rotor blade pitch). Thanks to the reliability of the numerical model, a detailed analysis of the physical mechanisms acting inside the rotor channel is performed. Two operating conditions at different vane incidence are considered, in a configuration where the effects of the vortex-blade interaction are highlighted. Different vane incidence angles lead to different size, position, and strength of secondary vortices coming out from the stator, thus promoting different interaction processes in the subsequent rotor channel. However some general trends can be recognized in the vortex-blade interaction: the sense of rotation and the spanwise position of the incoming vortices play a crucial role on the dynamics of the rotor vortices, determining both the time-mean and the time-resolved characteristics of the secondary field at the exit of the stage.


Sign in / Sign up

Export Citation Format

Share Document