Characterization of Deterministic Correlations for a Turbine Stage: Part 1 — Time-Averaged Flow Analysis

Author(s):  
Fabien Bardoux ◽  
Francis Leboeuf ◽  
Cédric Dano ◽  
Clément Toussaint

This paper analyses the flow in a transonic turbine stage, using time-dependent numerical results. Unsteady blade-row interaction has repercussions on the time-averaged flow, which are represented by the so-called “deterministic correlations”. These correlations appear in the system of equations governing the time-averaged flow; they can be divided into four types with different physical meanings. Time-dependent results enable direct computation of these correlations in both rotor and stator frames of reference. The computed deterministic correlations are analysed in the paper, in order to bind them to physical phenomena and to evaluate their influence on the time-averaged flow field. This analysis is also intended to help assess the shortcomings of simple mixing-plane methods and more complex approaches using deterministic correlation models. While the first part focuses on one particular type of deterministic correlation, the so-called “spatial correlation”, the second part attempts a more detailed analysis of time-dependent results and gives some clues to the orders of magnitude of the four types of deterministic correlation. The conclusions should be taken with caution; they may partly depend on the present turbine configuration with a specified structure of unsteadiness and on the present turbulence model.

Author(s):  
J. W. Barter ◽  
P. H. Vitt ◽  
J. P. Chen

A 3D, viscous, time-accurate code has been used to predict the time-dependent flowfield in a transonic turbine stage. Two analytical techniques are used to understand the unsteady physics. One technique takes into account interaction effects associated with reflected waves bouncing between blade rows while the other neglects them. Both techniques model the exact blade counts using phase-lag boundary conditions. The analytical techniques are validated by comparing to unsteady pressure measurements which have been made on the vane and blade surfaces at midspan. The analytical results are then used to understand the importance of interaction effects when the blade rows are close-coupled and when they are more widely spaced. The results show that interaction effects must be taken into account in order to accurately predict the unsteady loading on the upstream blade row. However, for the downstream blade row, interaction effects are second order and do not routinely need to be taken into account in the design process.


1985 ◽  
Vol 107 (4) ◽  
pp. 945-952 ◽  
Author(s):  
M. Koya ◽  
S. Kotake

Fully three-dimensional periodic flows through a turbine stage of stator and rotor are studied numerically by solving time-dependent three-dimensional Euler equations with the finite-volume method. The phase relation of stator and rotor flows and the related blade-row interaction are accounted for in the time-space domain. The established method of numerical calculation makes a practical contribution to predict actual turbine flows through a turbine stage of stator and rotor which have an arbitrary number of blades.


2015 ◽  
Vol 137 (12) ◽  
Author(s):  
L. He ◽  
J. Yi ◽  
P. Adami ◽  
L. Capone

For efficient and accurate unsteady flow analysis of blade row interactions, a space–time gradient (STG) method has been proposed. The development is aimed at maintaining as many modeling fidelities (the interface treatment in particular) of a direct unsteady time-domain method as possible while still having a significant speed-up. The basic modeling considerations, main method ingredients and some preliminary verification have been presented in Part I of the paper. Here in Part II, further case studies are presented to examine the capability and applicability of the method. Having tested a turbine stage in Part I, here we first consider the applicability and robustness of the method for a three-dimensional (3D) transonic compressor stage under a highly loaded condition with separating boundary layers. The results of the STG solution compare well with the direct unsteady solution while showing a speed up of 25 times. The method is also used to analyze rotor–rotor/stator–stator interferences in a two-stage turbine configuration. Remarkably, for stator–stator and rotor–rotor clocking analyses, the STG method demonstrates a significant further speed-up. Also interestingly, the two-stage case studies suggest clearly measurable clocking dependence of blade surface time-mean temperatures for both stator–stator clocking and rotor–rotor clocking, though only small efficiency variations are shown. Also validated and illustrated is the capacity of the STG method to efficiently evaluate unsteady blade forcing due to the rotor–rotor clocking. Considerable efforts are directed to extending the method to more complex situations with multiple disturbances. Several techniques are adopted to decouple the disturbances in the temporal terms. The developed capabilities have been examined for turbine stage configurations with inlet temperature distortions (hot streaks), and for three blade-row turbine configurations with nonequal blade counts. The results compare well with the corresponding direct unsteady solutions.


Author(s):  
G. Persico ◽  
P. Gaetani ◽  
C. Osnaghi

An extensive experimental analysis on the subject of the unsteady periodic flow in a highly subsonic HP turbine stage has been carried out at the Laboratorio di Fluidodinamica delle Macchine (LFM) of the Politecnico di Milano (Italy). In this paper the blade row interaction is progressively enforced by increasing the stator and rotor blade loading and by reducing the stator-rotor axial gap from 100% (very large to smooth the rotor inlet unsteadiness) to 35% (design configuration) of the stator axial chord. The time-averaged three-dimensional flow field in the stator-rotor gap was investigated by means of a conventional five-hole probe for the nominal (0°) and an highly positive (+22°) stator incidences. The evolution of the viscous flow structures downstream of the stator is presented to characterize the rotor incoming flow. The blade row interaction was evaluated on the basis of unsteady aerodynamic measurements at the rotor exit, performed with a fast-response aerodynamic pressure probe. Results show a strong dependence of the time-averaged and phase-resolved flow field and of the stage performance on the stator incidence. The structure of the vortex-blade interaction changes significantly as the magnitude of the rotor inlet vortices increases, and very different residual traces of the stator secondary flows are found downstream of the rotor. On the contrary, the increase of rotor loading enhances the unsteadiness in the rotor secondary flows but has a little effect on the vortex-vortex interaction. For the large axial gap, a reduction of stator-related effects at the rotor exit is encountered when the stator incidence is increased as a result of the different mixing rate within the cascade gap.


Author(s):  
Can Ma ◽  
Xinrong Su ◽  
Xin Yuan

Unsteady blade row interactions considerably affect the performance of turbomachinery consisting of multiple blade rows. However, most aerodynamic optimizations of turbomachinery are based on mixing-plane steady flow simulations which cannot account for the unsteady effects of blade row interactions. In this work, the rotor of a two-dimensional transonic turbine stage is optimized using an in-house unsteady aerodynamic optimization system that allows for a more accurate modeling of the unsteady flow features occurring in multi-row turbomachinery configurations. The gradients of the objective function and constraint to the design variables are efficiently calculated with the discrete adjoint method. In the developed adjoint-based unsteady aerodynamic optimization system, the unsteady Reynolds-Averaged Navier-Stokes equations are solved using the harmonic balance method with an in-house code. The adjoint equations are derived by hand from the discrete form of the unsteady flow equations. The present results demonstrate the efficiency and capability of the unsteady aerodynamic optimization system for turbomachinery with multiple blade rows.


Author(s):  
Brian L. Venable ◽  
Robert A. Delaney ◽  
Judy A. Busby ◽  
Roger L. Davis ◽  
Daniel J. Dorney ◽  
...  

A comprehensive study has been performed to determine the influence of vane-blade spacing on transonic turbine stage aerodynamics. In Part I of this paper, an investigation of the effect of turbine vane-blade interaction on the time-mean airfoil surface pressures and overall stage performance parameters is presented. Experimental data for an instrumented turbine stage are compared to two- and three-dimensional results from four different time-accurate Navier-Stokes solvers. Unsteady pressure data were taken for three vane-blade row spacings in a short-duration shock tunnel using surface-mounted, high-response pressure sensors located along the midspan of the airfoils. Results indicate that while the magnitude of the surface pressure unsteadiness on the vane and blade changes significantly with spacing, the time-mean pressures and performance numbers are not greatly affected.


2009 ◽  
Vol 131 (3) ◽  
Author(s):  
G. Persico ◽  
P. Gaetani ◽  
C. Osnaghi

An extensive experimental analysis on the subject of the unsteady periodic flow in a high subsonic high pressure (HP) turbine stage has been carried out at the Laboratorio di Fluidodinamica delle Macchine of the Politecnico di Milano (Italy). In this paper the aerodynamic blade row interaction in HP turbines, enforced by increasing the stator and rotor blade loading and by reducing the stator-rotor axial gap, is studied in detail. The time-averaged three-dimensional flowfield in the stator-rotor gap was investigated by means of a conventional five-hole probe for the nominal (0 deg) and highly positive (+22 deg) stator incidences. The evolution of the viscous flow structures downstream of the stator is presented to characterize the rotor incoming flow. The blade row interaction was evaluated on the basis of unsteady aerodynamic measurements at the rotor exit, performed with a fast-response aerodynamic pressure probe. Results show a strong dependence of the time-averaged and phase-resolved flowfield and of the stage performance on the stator incidence. The structure of the vortex-blade interaction changes significantly as the magnitude of the rotor-inlet vortices increases, and very different residual traces of the stator secondary flows are found downstream of the rotor. On the contrary, the increase in rotor loading enhances the unsteadiness in the rotor secondary flows but has a little effect on the vortex-vortex interaction. For the large axial gap, a reduction of stator-related effects at the rotor exit is encountered when the stator incidence is increased as a result of the different mixing rate within the cascade gap.


Author(s):  
B. Paradiso ◽  
G. Persico ◽  
P. Gaetani ◽  
O. Schennach ◽  
R. Pecnik ◽  
...  

The unsteady and fully three-dimensional aerodynamics of HP turbines represent a relevant research branch for future aero-engine design. When stator-rotor interaction mechanisms and clocking effects are of concern, advanced measurement techniques as well as unsteady CFD codes are required. An extensive study on this topic was carried out in a one and a half stage transonic turbine operating at Graz University of Technology. Two steady and unsteady measurement techniques (Laser Doppler Velocimetry and a Fast Response Aerodynamic Pressure Probe) and an unsteady 3D CFD code were applied to the problem. In this paper, the 1st vane – rotor interaction is presented and discussed in detail to provide the basis for the analysis of the rotor – 2nd vane and the vane-vane interactions. The rotor-exit flowfield is mainly characterized by the wake, the hub passage vortex, the tip leakage vortex and the trailing edge shocks. All the flow structures except the tip leakage flow are strongly influenced by the first vane; in particular the main source of blade row interaction is the first vane trailing edge shock, that periodically alters the rotor trailing edge shock and the rotor hub passage vortex. The comparison with the CFD assesses the interpretation of the flow physics, and supports the identification of the first stator effects at the second stator inlet. A discussion on the stage performance is also provided.


1999 ◽  
Vol 121 (4) ◽  
pp. 663-672 ◽  
Author(s):  
B. L. Venable ◽  
R. A. Delaney ◽  
J. A. Busby ◽  
R. L. Davis ◽  
D. J. Dorney ◽  
...  

A comprehensive study has been performed to determine the influence of vane-blade spacing on transonic turbine stage aerodynamics. In Part I of this paper, an investigation of the effect of turbine vane–blade interaction on the time-mean airfoil surface pressures and overall stage performance parameters is presented. Experimental data for an instrumented turbine stage are compared to two- and three-dimensional results from four different time-accurate Navier–Stokes solvers. Unsteady pressure data were taken for three vane-blade row spacings in a short-duration shock tunnel using surface-mounted, high-response pressure sensors located along the midspan of the airfoils. Results indicate that while the magnitude of the surface pressure unsteadiness on the vane and blade changes significantly with spacing, the time-mean pressures and performance numbers are not greatly affected.


Author(s):  
Björn Laumert ◽  
Hans Mårtensson ◽  
Torsten H. Fransson

Based on the results of time dependent 3D viscous computations the aerodynamic mechanisms that cause the unsteady pressure fluctuations on the vane and rotor blade surface of a high-pressure transonic turbine are identified and separately classified in a phenomenological manner. In order to be able to describe separately the influence of wake, potential and shock distortions on the blade surface pressure at design operation conditions, the stator exit Mach number is increased as to enhance the shock distortions and lowered as to enhance potential and wake distortions. In a comprehensive approach the observations from the off-design conditions are utilized to classify every major perturbation observed in the perturbation space-time maps at design operation conditions. The spanwise variations caused by the inherent 3D nature of the flow field and promoted by the 3D shape of the rotor blade are addressed.


Sign in / Sign up

Export Citation Format

Share Document