Search

A particular turbine cascade design is presented with the goal of providing a basis for high quality investigations of endwall flow at high-speed flow conditions and unsteady inflow. The key feature of the design is an integrated two-part flat plate serving as a cascade endwall at part-span, which enables a variation of the inlet endwall boundary layer conditions. The new design is applied to the T106A low pressure turbine cascade for endwall flow investigations in the High-Speed Cascade Wind Tunnel of the Institute of Jet Propulsion at the Bundeswehr University Munich. Measurements are conducted at realistic flow conditions (M2th = 0.59, Re2th = 2·105) in three cases of different endwall boundary layer conditions with and without periodically incoming wakes. The endwall boundary layer is characterized by 1D-CTA measurements upstream of the blade passage. Secondary flow is evaluated by Five-hole-probe measurements in the turbine exit flow. A strong similarity is found between the time-averaged effects of unsteady inflow conditions and the effects of changing inlet endwall boundary layer conditions regarding the attenuation of secondary flow. Furthermore, the experimental investigations show, that all design goals for the improved T106A cascade are met.

Turbine disc cooling is required to protect vulnerable components from exposure to the high temperatures found in the mainstream gas path. Purge air, bled from the latter stages of the compressor, is introduced to the turbine wheelspace at low radius before exiting through the rim-seal at the periphery of the discs. The unsteady, complex flowfield that arises from the interaction between the purge and mainstream gases modifies the structure of secondary flows within the blade passage. A computational study was conducted using an unsteady Reynolds-averaged Navir–Stokes (RANS) solver, modeling an engine-representative turbine stage. Preliminary results were validated using experimental data from a test rig. The baseline secondary flowfield was described, in the absence of purge flow, demonstrating the classical rollup of the horseshoe vortex and subsequent convection of the two legs downstream. The unsteady behavior of the model was investigated and addressed, resulting in recommendations for modeling interaction phenomena in turbines. A superposed purge flow, resulting in egress through the upstream rim-seal, was shown to modify the secondary flowfield in the turbine annulus. The most notable effect of egress was the formation of a large plume forming near the pressure minima associated with the blade suction surface. The egress was turned by the mainstream flow, creating a vortical structure consistent in rotational direction to the pressure-side leg of the horseshoe vortex; the pressure-side leg was subsequently strengthened and showed an increased radial migration relative to the unpurged case. The egress plume was also shown to overwhelm the suction-side leg of the horseshoe vortex, reducing its strength.

To enhance solution accuracy with high-order methods, an implicit solver using the discontinuous Galerkin (DG) discretization on unstructured mesh has been developed. The DG scheme is favored chiefly due to its distinctive feature of achieving a higher-order accuracy by simple internal sub-divisions of a given mesh cell. It can thus relieve the burden of mesh generation for local refinement. The developed method has been assessed in several test cases. First, an examination on the mesh-dependency with different DG orders for discretization has demonstrated the expected mesh convergence order of accuracy correlation. The flow around a cylinder solved with up to the fifth-order accuracy demonstrates the need for a high-order geometrical representation corresponding to the high-order flow field discretization. A calculated turbulent flat plate boundary layer demonstrates the capability of the high-order DG in capturing the laminar sub-layer for a given coarse mesh (y + >20) without a wall function, in a clear contrast to a conventional second-order scheme. The focal test case of the present work is a high-pressure (HP) turbine cascade with the flow losses being strongly influenced by both the laminar separation induced transition on the blade suction surface and the secondary flow development in the endwall region. Fully turbulent RANS solutions consistently overpredict the losses, which can be significantly reduced by an empirically specified transition at a mid-chord point. Of particular interest is a contrasting behavior of pure laminar flow solutions. A steady laminar flow solution is difficult to converge, and even when converging, tends to give an unrealistically large separation. On the other hand when the laminar solution is run in a time-accurate unsteady mode, a qualitatively different flow pattern emerges. The separated shear layer is then shown to lead to coherently shed unsteady vortices with a net entrainment from the main stream to the near-wall region. The resultant time-averaged near-wall flow then effectively becomes a reattached boundary layer. Both overall and distributed losses computed from unsteady laminar solutions are shown to be consistently much better than the fully turbulent RANS solutions. It is remarkable and quite unexpected that for such a seemingly complex 3D flow problem, the straightforward unsteady laminar solutions with no empirical input seem to be comparable with (or slightly better than) the tripped RANS with empirical input.

In low-pressure turbines (LPTs), around 60–70% of losses are generated away from end-walls, while the remaining 30–40% is controlled by the interaction of the blade profile with the end-wall boundary layer. Experimental and numerical studies have shown how the strength and penetration of the secondary flow depends on the characteristics of the incoming end-wall boundary layer. Experimental techniques did shed light on the mechanism that controls the growth of the secondary vortices, and scale-resolving computational fluid dynamics (CFD) allowed to dive deep into the details of the vorticity generation. Along these lines, this paper discusses the end-wall flow characteristics of the T106 LPT profile at Re = 120 K and M = 0.59 by benchmarking with experiments and investigating the impact of the incoming boundary layer state. The simulations are carried out with proven Reynolds-averaged Navier–Stokes (RANS) and large-eddy simulation (LES) solvers to determine if Reynolds-averaged models can capture the relevant flow details with enough accuracy to drive the design of this flow region. Part I of the paper focuses on the critical grid needs to ensure accurate LES and on the analysis of the overall time-averaged flow field and comparison between RANS, LES, and measurements when available. In particular, the growth of secondary flow features, the trace and strength of the secondary vortex system, and its impact on the blade load variation along the span and end-wall flow visualizations are analyzed. The ability of LES and RANS to accurately predict the secondary flows is discussed together with the implications this has on design.

The experimental results reported in this contribution address the time-dependent impact of periodically unsteady wakes on the development of profile and end wall boundary layers and consequently on the secondary flow system. Experimental investigations are conducted on an annular 1.5 stage axial turbine rig at Ruhr-Universität Bochum’s Chair of Thermal Turbomachines and Aeroengines. The object under investigation is a modified T106 profile low-pressure turbine (LPT) stator row at a representative exit flow Reynolds number of 200,000. By making use of an annular geometry instead of a linear cascade, the influence of curvilinear end walls, nonuniform, increasing pitch across the span and radial flow migration can be represented. Incoming wakes are generated by a variable-speed driven rotor equipped with cylindrical bars. Special emphasis is put on the wake-induced recurrent formation, suppression, weakening, and displacement of individual vortices and separated flow regimes. For this, based on a comprehensive set of time-resolved measurement data, the interaction of impinging bar wakes and boundary layer flow and thus separation and its periodic manipulation along the passage end walls and on the blade suction surface are studied within the frequency domain.

In this work, we present the results of the numerical investigations of periodic wake–secondary flow interaction carried out on a low pressure turbine (LPT) equipped with modified T106-profile blades. The numerical predictions obtained by means of unsteady Reynolds-averaged Navier–Stokes (URANS) simulations using a k-ω-model have been compared with measurements conducted in the same configuration and showed a good agreement. Based on the verified numerical data, the Q-criterion has been employed to characterize the secondary flow structures and accurately identify their origin. An analysis of the fundamental wake kinematics and the unsteady vortex migration revealed dominant interaction mechanisms such as the circumferential fluctuation of the pressure side horseshoe vortex (HSV) and its direct interaction with the passage vortex (PV) and the concentrated shed vortex (CSV). Finally, a correlation with the total pressure loss coefficient is provided and a link to the incoming wake structures is given.

Experimental studies have been conducted on a modified T106 low pressure turbine (LPT) profile in an annular 1.5 stage axial turbine rig at the Chair of Thermal Turbomachines and Aeroengines, Ruhr-Universität Bochum. The rig setup allows the highly resolved measurement of unsteady wake–stator flow interaction in both space and time. Incoming wakes are generated by a variable-speed driven rotor equipped with cylindrical bars. In the present paper, an experimental approach to the investigation of unsteady phenomena is proposed. Time-averaged and instantaneous measurement data from 2D flow field traverses at the stator exit are provided for the analysis of the periodically unsteady vortex formation, displacement, and suppression. Additional time-accurate blade pressure data are used to study the relationship between the flow structures downstream of the stator row and the immediate intermittent wake impact on the blades. Bar wake kinematics is also discussed in relation to the observations.

Improvements in stage isentropic efficiency and reductions in total pressure loss are sought in a 1.5 stage axial turbine. This is representative of power generation equipment used in thermal power cycles, which delivers about 80% of the 20 × 1012 kWh world-wide electricity. Component-level improvements are therefore timely and important toward achieving carbon dioxide global emission targets. Secondary flow loss reduction is sought by applying a nonaxisymmetric endwall design to the turbine stator hub. A guide groove directs the pressure side branch of the horseshoe vortex away from the airfoil suction side, using a parametric endwall hub surface, which is defined as to obtain first-order smooth boundary connections to the remainder of the passage geometry. This delays the onset of the passage vortex and reduces its associated loss. The Automatic Process and Optimization Workbench (apow) generates a Kriging surrogate model from a set of Reynolds-averaged Navier–Stokes simulations, which is used to optimize the hub surface. The three-dimensional steady Reynolds-averaged Navier–Stokes model with an axisymmetric hub is validated against reference experimental measurements from the Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen. Comparative computational fluid dynamics (CFD) predictions with an optimized nonaxisymmetric hub show a decrease in the total pressure loss coefficient and an increase in the isentropic stage efficiency at and off design conditions.

The flow field in a ribbed triangular channel representing the trailing edge internal cooling passage of a gas turbine high-pressure turbine blade is investigated via magnetic resonance velocimetry (MRV) and large eddy simulation (LES). The results are compared to a baseline channel with no ribs. LES predictions of the mean velocity fields are validated by the MRV results. In the case of the baseline triangular channel with no ribs, the mean flow and turbulence level at the sharp corner are small, which would correspond to poor heat transfer in an actual trailing edge. For the staggered ribbed channel, turbulent mixing is enhanced, and flow velocity and turbulence intensity at the sharp edge increase. This is due to secondary flow induced by the ribs moving toward the sharp edge in the center of the channel. This effect is expected to enhance internal convective heat transfer for the turbine blade trailing edge.

This paper presents a new unconventional philosophy for high-pressure (HP) vane design. It is proposed that the most natural design starting point for admitting and accelerating flow with minimum loss and secondary flow is a trumpet-shaped flow-path which gradually turns to the desired angle. Multiple trumpet-shaped inlets are seamlessly blended into the (annular or partitioned) combustor walls resulting in a highly lofted flow-path, rather than a traditional flow-path defined by distinct airfoil and endwall surfaces. We call this trumped-shaped inlet the fully lofted oval vane (FLOvane). The purpose of this paper is to describe the FLOvane concept and to present back-to-back CFD analyses of two current industrial gas turbines with conventional and FLOvane-modified designs. The resulting designs diverge significantly from conventional designs in terms of both process and final geometric form. Computational fluid dynamic predictions for the FLOvane-modified designs show improved aerodynamic performance characteristics, reduced heat load, improved cooling performance, improved thermal–mechanical life, and improved stage/engine efficiency. The mechanisms for improved performance include reduction of secondary flows, reduced mixing of coolant flow with the mainstream flow, reduced skin friction, and improved coolant distribution. In the two current industrial gas turbine engines, the absolute (percentage point) improvement in stage isentropic efficiency when the FLOvane design was included was estimated at 0.33% points and 0.40% points without cooling flow reduction, and 1.5% points in one case and much more is expected for the other case when cooling flow reductions were accounted for.

An experimental study has been performed in a transonic annular sector cascade of nozzle guide vanes (NGVs) to investigate the aerodynamic performance and the interaction between hub film cooling and mainstream flow. The focus of the study is on the endwalls, specifically the interaction between the hub film cooling and the mainstream. Carbon dioxide (CO2) has been supplied to the coolant holes to serve as tracer gas. Measurements of CO2 concentration downstream of the vane trailing edge (TE) can be used to visualize the mixing of the coolant flow with the mainstream. Flow field measurements are performed in the downstream plane with a five-hole probe to characterize the aerodynamics in the vane. Results are presented for the fully cooled and partially cooled vane (only hub cooling) configurations. Data presented at the downstream plane include concentration contour, axial vorticity, velocity vectors, and yaw and pitch angles. From these investigations, secondary flow structures such as the horseshoe vortex, passage vortex, can be identified and show the cooling flow significantly impacts the secondary flow and downstream flow field. The results suggest that there is a region on the pressure side (PS) of the vane TE where the coolant concentrations are very low suggesting that the cooling air introduced at the platform upstream of the leading edge (LE) does not reach the PS endwall, potentially creating a local hotspot.

This paper considers the interaction of tip clearance flow with three-dimensional (3D) separations in the corner region of a compressor cascade. Three-dimensional numerical computations were carried out using ten levels of tip clearance, ranging from zero to 2.18% of blade chord. The 3D separations on the blade suction surface were largely removed by the clearance flow for clearance about 0.58% of chord. For this cascade, experimental results at zero and 1.7% chord tip clearance were used to assess the validity of the numerical predictions. The removal mechanism was associated with the suppression of the leading edge horseshoe vortex and the interaction of tip clearance flow with the endwall boundary layer, which develops into a secondary flow as it is drifted towards the blade suction surface. Such interaction leads to the formation of a new 3D separation line on the endwall. The separation line forms the base of a separated stream surface which rolls up into the clearance vortex.Copyright © 2006 by ASME

Within a European research project, the tip endwall region of low pressure turbine guide vanes with leakage ejection was investigated at DLR in Göttingen. For this purpose a new cascade wind tunnel with three large profiles in the test section and a contoured endwall was designed and built, representing 50% height of a real low pressure turbine stator and simulating the casing flow field of shrouded vanes. The effect of tip leakage flow was simulated by blowing air through a small leakage gap in the endwall just upstream of the vane leading edges. Engine relevant turbulence intensities were adjusted by an active turbulence generator mounted in the test section inlet plane. The experiments were performed with tangential and perpendicular leakage ejection and varying leakage mass flow rates up to 2%. Aerodynamic and thermodynamic measurement techniques were employed. Pressure distribution measurements provided information about the endwall and vane surface pressure field and its variation with leakage flow. Additionally streamline patterns (local shear stress directions) on the walls were detected by oil flow visualization. Downstream traverses with five-hole pyramid type probes allow a survey of the secondary flow behavior in the cascade exit plane. The flow field in the near endwall area downstream of the leakage gap and around the vane leading edges was investigated using a 2D particle image velocimetry system. In order to determine endwall heat transfer distributions, the wall temperatures were measured by an infrared camera system, while heat fluxes at the surfaces were generated with electric operating heating foils. It turned out from the experiments that distinct changes in the secondary flow behavior and endwall heat transfer occur mainly when the leakage mass flow rate is increased from 1% to 2%. Leakage ejection perpendicular to the main flow direction amplifies the secondary flow, in particular the horseshoe vortex, whereas tangential leakage ejection causes a significant reduction of this vortex system. For high leakage mass flow rates the boundary layer flow at the endwall is strongly affected and seems to be highly turbulent, resulting in entirely different heat transfer distributions.

This paper presents an experimental study of the flow mechanisms of tip leakage across a blade of an unshrouded turbine rotor. It shows the design of a new one-and-1∕2-stage, unshrouded turbine configuration, which has been developed within the Turbomachinery Laboratory of ETH Zurich. This test case is a model of a high work (Δh∕u2=2.36) axial turbine. The experimental investigation comprises data from unsteady and steady probe measurements, which has been acquired around all the bladerows of the one-and-1∕2-stage, unshrouded turbine. A newly developed 2-sensor Fast Response Aerodynamic Probe (FRAP) technique has been used in the current measurement campaign. The paper contains a detailed analysis of the unsteady interaction between rotor and stator blade rows, with particular attention paid on the flow in the blade tip region. It has been found that the interaction of the rotor and the downstream stator has an influence on the development of the tip leakage vortex of the rotor. The vortex is modulated by the stator profiles and shows variation in size and relative position to the rotor trailing edge when it stretches around the stator leading edge. Thereby a deflection of the tip leakage vortex has been observed, which expresses in a varying circumferential distance between two neighboring vortices of ±20% of a rotor pitch. Furthermore, a significant influence of quasi-stationary secondary flow features of the upstream stator row on the secondary flow of the rotor has been detected. The geometry and flow field data of the one-and-1∕2-stage turbine will be available to the turbomachinery community for validation and improvement of numerical tools.

A feature-based jet model has been proposed for use in three-dimensional (3D) computational fluid dynamics (CFD) prediction of turbine blade film cooling. The goal of the model is to be able to perform computationally efficient flow prediction and optimization of film-cooled turbine blades. The model reproduces in the near-hole region the macroflow features of a coolant jet within a Reynolds-averaged Navier-Stokes framework. Numerical predictions of the 3D flow through a linear transonic film-cooled turbine cascade are carried out with the model, with a low computational overhead. Different cooling holes arrangements are computed, and the prediction accuracy is evaluated versus experimental data. It is shown that the present model provides a reasonably good prediction of the adiabatic film-cooling effectiveness and Nusselt number around the blade. A numerical analysis of the interaction of coolant jets issuing from different rows of holes on the blade pressure side is carried out. It is shown that the upward radial migration of the flow due to the passage secondary flow structure has an impact on the spreading of the coolant and the film-cooling effectiveness on the blade surface. Based on this result, a new arrangement of the cooling holes for the present case is proposed that leads to a better spanwise covering of the coolant on the blade pressure side surface.

Prediction of flow field and heat transfer of high rotation numbers and density ratio flow in a square internal cooling channels of turbine blades with U-turn as tested by Wagner et al. (ASME J. Turbomach., 113, pp. 42–51, 1991) is the main focus of this study. Rotation, buoyancy, and strong curvature affect the flow within these channels. Due to the fact that RSM turbulence model can respond to the effects of rotation, streamline curvature and anisotropy without the need for explicit modeling, it is employed for this study as it showed improved prediction compared to isotropic two-equation models. The near wall region was modeled using enhanced wall treatment approach. The Reynolds Stress Model (RSM) was validated against available experimental data (which are primarily at low rotation and buoyancy numbers). The model was then used for cases with high rotation numbers (as much as 1.29) and high-density ratios (up to 0.4). Particular attention is given to how secondary flow, velocity and temperature profiles, turbulence intensity, and Nusselt number area affected by Coriolis and buoyancy/centrifugal forces caused by high levels of rotation and buoyancy in the immediate vicinity of the bend. The results showed that four-side-average Nu, similar to low Ro cases, increases linearly by increasing rotation number and, unlike low Ro cases, decreases slightly by increasing density ratio.

Flow characteristics in stationary two-pass channels with a sharp 180-deg turn have been measured using LDV (laser Doppler velocimeter), directing special attention to the influence of the size of the turn clearance on the flow structure. The main features of the flow, flow separation and recirculation, secondary flow, turbulence intensities, measured with Re=3.5×104 for three turn clearances, are presented. A close comparison of the velocity data with local Sherwood number distributions on the channel walls reveals that the wall-normal velocity mainly dominates the heat transfer in the channel, but the wall-parallel component also contributes locally to heat transfer after the turn.

This paper presents the results of time-resolved flow field measurements of a multistage shrouded axial turbine. The unsteady interaction mechanism between the rotor’s secondary flow vortices, the rotor’s wake and the adjacent blading at the exit plane of the first turbine stage is of prime interest and analysed in detail. Three key phases are identified for one blade passing event. The first phase shows a quasi undisturbed convection of the rotor’s secondary flow field into the downstream stator. The second phase shows a migration of high loss fluid from the wake layer into the passage and horse-shoe vortices at the rotor hub section. The relative motion between the rotor and stator blades brings the two vortices closer to the wake layer and lets the flow features interact with each other. The third phase focuses on the rotor indigenous hub vortices that are bent and stretched around the stator’s leading edge. The signal analysis of the time-resolved flow field indicates a high level of unsteadiness at the stator’s pressure side. The associated unsteadiness within the flow field is evaluated and quantified on the basis of pitchwise averaged space-time diagrams. The obtained results are finally discussed and explained using two flow schematics within and at the end of the paper.Copyright © 2004 by ASME

The present paper investigates the effects of a fan-shaped hole endwall cooling geometry on the aero-thermal performance of a nozzle vane cascade. Two endwall cooling geometries with four rows of holes were tested, for different mass flow rate ratios: the first configuration is made of cylindrical holes, whereas the second one features conical expanded exits and a reduced number of holes. The experimental analysis is mainly focused on the variations of secondary flow phenomena related to different injection rates, as they have a strong relationship with the film cooling effectiveness. Secondary flow assessment was performed through downstream 3D aerodynamic measurements, by means of a miniaturized 5-hole probe. The results show that at high injection rates, the passage vortex and the 3D effects tend to become weaker, leading to a strong reduction of the endwall cross flow and to a more uniform flow in spanwise direction. This is of course obtained at the expense of a significant increase of losses. The thermal behavior was then investigated through the analysis of adiabatic effectiveness distributions on the two endwall configurations. The wide-banded thermochromic liquid crystals (TLC) technique was used to determine the adiabatic wall temperature. Using the measured distributions of film-cooling adiabatic effectiveness, the interaction between the secondary flow vortices and the cooling jets can be followed in good detail all over the endwall surface. Fan-shaped holes have been shown to perform better than cylindrical ones: at low injection rates, the cooling performance is increased only in the front part of the vane passage. A larger improvement of cooling coverage all over the endwall is attained with a larger mass flow rate, about 1.5% of core flow, without a substantial increase of the aerodynamic losses.

It is the aim of this publication to attract the designers attention to the end wall flow interactions of shrouded high pressure turbines. One of the key issues for designing better turbines is the understanding of the flow interactions set up by the presence of labyrinth seals. Those interaction flows are carefully examined in this publication using the control volume analysis and the radial equilibrium of forces acting on streamlines. The consequences on secondary flow development and mixing losses are discussed and quantified. Out of this insight, design recommendations are derived, which attempt to make use of the nature of the labyrinth interaction flow. The open labyrinth cavities are classified in a systematic way. The aim of this approach is to work out the characteristic differences between hub and tip cavities and those having a leakage jet or sucking main flow fluid into the labyrinth. The influence on the main flow is discussed in terms of the incidence flow angle of downstream blade rows and the associated loss production mechanisms. The design strategies presented in this paper follow two paths: (a) Optimization of the mixing losses of the leakage jets at hub and tip is estimated to result in an efficiency increase of up to 0.2%. (b) The nonaxisymmetric shaping of the labyrinth interaction flow path aims at the secondary flow control in downstream blade rows. This approach might contribute in the same magnitude of order as reduction in the mixing losses.

Laser-Doppler velocimetry and pressure measurements are presented of the local velocity and wall pressure distributions in a rotating two-pass square duct with staggered ribs placed on the leading and trailing walls at an angle of 45deg to the main stream. The ribs were square in cross section with the radii of rounds and fillets to rib height ratios of 0.33. The rib-height/duct-height ratio and the pitch/rib-height ratio were 0.136 and 10, respectively. The duct Reynolds number was 1×104 and rotation number Ro ranged from 0 to 0.2. Results are documented in terms of the evolutions of both main flow and cross-stream secondary flow, the distributions of the pressure coefficient, and the variation of friction factor with Ro. For CFD reference, the periodic fully developed flow condition is absent for the present length of the rotating passage roughened with staggered 45deg ribs. In addition, the relationships between the regional averaged Nusselt number, transverse and convective mean velocity component, and turbulent kinetic energy are addressed. Using these relationships the general superiority of heat transfer enhancement of the staggered 45deg ribs arrangement over the in-line one can be reasonably illustrated. Simple expressions are obtained to correlate the friction factor with Ro, which are lacking in the published literature for passages ribbed with staggered 45deg ribs. The staggered 45deg ribs are found to reduce the friction loss to about 88%±1% of the in-line 45deg ribs for the rotating passage under the same operating conditions. The respective contributions of the angled ribs and passage rotation on the passage friction loss are identified.

Measurements are presented of the local velocity and wall static-pressure distributions by using laser-Doppler velocimeter and pressure transducers, respectively, in a rotating two-pass square duct with ribs placed on the leading and trailing walls at an angle of 45 deg to the main stream. The ribs were square in cross section and in a parallel mode of arrangement. The rib-height/duct-height ratio and the pitch/rib-height ratio were 0.136 and 10, respectively. The duct Reynolds number was 1×104 and rotation number Ro ranged from 0 to 0.2. Results are addressed in terms of the evolutions of both main flow and cross-stream secondary flow, the distributions of the pressure coefficient, and the variation of friction factor with Ro. In addition, the relationships between the regional averaged Nusselt number, transverse and convective mean velocity component, and turbulent kinetic energy are documented. Simple expressions are obtained to correlate friction factor with Ro, which are lacking in the published literature for ducts ribbed with 45- deg ribs. The 45-deg ribs are found to reduce the friction loss to 60% of the 90-deg ribs for rotating duct under the same operating conditions. For CFD reference, the fully developed flow condition is absent for the rotating ribbed duct investigated. The measured evolution of complex secondary flow vortices is believed to be a challenge to numerical simulations.

Laser-Doppler anemometry has been applied to approximately two-dimensional turbulent air flow in rotating two pass channel with turbulator of rectangular cross section (AR=3:1). The axis of rotation is normal to the axis of the duct, and the flow is radially outward/inward. The duct is of finite length and the walls are isothermal. Two sided oppositely ribbed channel including one sided ribbed U bend of p/e=8 at e/DH=0.27 are experimentally conducted with ReD=5000 and 10,000. The main features of the flow, reattachment length, recirculation zone, and mean velocity as well as turbulent intensity and shear stress distributions are presented in ribbed ducts. The measured flow field is found to be quite complex, consisting of secondary cross-stream flows due to the Coriolis effects and centrifugal forces with rib-roughened surfaces.

Shear layers shed by aircraft wings roll up into vortices. A similar, though far less common, phenomenon can occur in the wake of a turbomachine blade. This paper presents experimental data from a new single-stage turbine that has been commissioned at the Whittle Laboratory. Two low-aspect ratio stators have been tested with the same rotor row. Surface flow visualization illustrates the extremely strong secondary flows present in both NGV designs. These secondary flows lead to conventional passage vortices, but also to an intense vortex sheet which is shed from the trailing edge of the blades. Pneumatic probe traverses show how this sheet rolls up into a concentrated vortex in the second stator design, but not in the first. A simple numerical experiment is used to model the shear layer instability and the effects of trailing edge shape and exit yaw angle distribution are investigated. It is found that the latter has a strong influence on shear layer rollup: inhibiting the formation of a vortex downstream of NGV 1 but encouraging it behind NGV 2.

Endwall losses significantly contribute to the overall losses in modern turbomachinery, especially when aerodynamic airfoil load and pressure ratios are increased. In turbines with shrouded airfoils a large portion of these losses are generated by the leakage flow across the shroud clearance. Generally the related losses can be grouped into losses of the leakage flow itself and losses caused by the interaction with the main flow in subsequent airfoil rows. In order to reduce the impact of the leakage flow and shroud design related losses a thorough understanding of the leakage losses and especially of the losses connected to enhancing secondary flows and other main flow interactions has to be understood. Therefore, a three stage LP turbine typical for jet engines is being investigated. For the three-stage test turbine 3D Navier-Stokes computations are performed simulating the turbine including the entire shroud cavity geometry in comparison with computations in the ideal flow path. Numerical results compare favourably against measurements carried out at the high altitude test facility at Stuttgart University. The differences of the simulations with and without shroud cavities are analysed for several points of operation and a very detailed quantitative loss breakdown is presented.

This paper describes the time-varying aerodynamic interaction mechanisms that have been observed within a transonic high-pressure turbine stage; these are inferred from the time-resolved behavior of the rotor exit flow field. It contains results both from an experimental program in a turbine test facility and from numerical predictions. Experimental data was acquired using a fast-response aerodynamic probe capable of measuring Mach number, whirl angle, pitch angle, total pressure, and static pressure. A 3-D time-unsteady viscous Navier-Stokes solver was used for the numerical predictions. The unsteady rotor exit flowfield is formed from a combination of four flow phenomena: the rotor wake, the rotor trailing edge recompression shock, the tip-leakage flow, and the hub secondary flow. This paper describes the time-resolved behavior of each phenomenon and discusses the interaction mechanisms from which each originates. Two significant vane periodic changes (equivalent to a time-varying flow in the frame of reference of the rotor) in the rotor exit flowfield are identified. The first is a severe vane periodic fluctuation in flow conditions close to the hub wall and the second is a smaller vane periodic fluctuation occurring at equal strength over the entire blade span. These two regions of periodically varying flow are shown to be caused by two groups of interaction mechanisms. The first is thought to be caused by the interaction between the wake and secondary flow of the vane with the downstream rotor; and the second is thought to be caused by a combination of the interaction of the vane trailing edge recompression shock with the rotor, and the interaction between the vane and rotor potential fields.

Computations, based on the ensemble-averaged compressible Navier-Stokes equations closed by the shear-stress transport (SST) turbulence model, were performed to investigate the effects of leading-edge airfoil fillet and inlet-swirl angle on the flow and heat transfer in a turbine-nozzle guide vane. Three fillet configurations were simulated: no fillet (baseline), a fillet whose thickness fades on the airfoil, and a fillet whose thickness fades on the endwall. For both fillets, the maximum height above the endwall is positioned along the stagnation zone/line on the airfoil under the condition of no swirl. For each configuration, three inlet swirls were investigated: no swirl (baseline) and two linearly varying swirl angle from one endwall to the other (+30° to −30° and −30° to +30°). Results obtained show that both leading-edge fillet and inlet swirl can reduce aerodynamic loss and surface heat transfer. For the conditions of this study, the difference in stagnation pressure from the nozzle’s inlet to its exit were reduced by more than 40% with swirl or with fillet without swirl. Surface heat transfer was reduced by more than 10% on the airfoil and by more than 30% on the endwalls. When there is swirl, leading-edge fillets became less effective in reducing aerodynamic loss and surface heat transfer, because the fillets were not optimized for swirl angles imposed. Since the intensity and size of the cross flow were found to increase instead of decrease by inlet swirl and by the type of fillet geometries investigated, the results of this study indicate that the mechanisms responsible for aerodynamic loss and surface heat transfer are more complex than just the intensity and the magnitude of the secondary flows. This study shows their location and interaction with the main flow to be more important, and this could be exploited for positive results.Copyright © 2002 by ASME

Laser-doppler velocimetry and transient thermochromic liquid crystal measurements are presented to understand local fluid flow and surface heat transfer distributions in a rotating ribbed duct with a 180 deg sharp turn. The in-line 90-deg ribs were arranged on the leading and trailing walls with rib height-to-hydraulic diameter ratio and pitch-to-height ratio of 0.136 and 10, respectively. The Reynolds number, based on duct hydraulic diameter and bulk mean velocity, was fixed at 1.0×104 whereas the rotational number varied from 0 to 0.2. Results are compared with those of the rotating smooth duct flow in terms of maximum streamwise mean velocities Umax/Ub and turbulence intensities u′max/Ub, skewness of mean velocity profiles, secondary flow pattern, turn-induced separation bubble, and turbulence anisotropy. Nusselt number ratio mappings are also provided on the leading and trailing walls. The relationships between the fluid flow and local heat transfer enhancement are also documented. It is found that the rotating ribbed duct flow provides higher Umax/Ub,u′max/Ub, and stronger total averaged secondary flow and, hence heat transfer is enhanced. Comparisons with heat transfer data published by other research groups are also made. Furthermore, simple linear correlations between regional averaged Nusselt number ratio and rotation number are developed.

Within the present paper, a detailed experimental investigation is presented. The influence of blade loading on the development and interaction of secondary flow structures within an annular compressor stator cascade (CSC) is examined. Experimental results at 3% chord hub clearance were obtained at four different blade loadings. Included are blade and endwall flow visualization, time resolved measurements of the static pressure on the endwall and radial–circumferential hot-wire traverse measurements within the passage as well as five-hole probe traverse measurements at the inlet and the outlet of the passage. The experimentally obtained results give detailed insight on the effect of the incidence on the development and interaction of the clearance vortex, horseshoe vortex, and the passage vortex. Furthermore, it will be shown that a vortex breakdown of the clearance vortex occurs at higher loadings.

Efforts to reduce blade count and avoid boundary layer separation have led to lowpressure turbine airfoils with significant increases in loading as well as front-loaded pressure distributions. These features have been independently shown to increase losses within the secondary flow field at the end wall. Compound angle blowing from discrete jets on the blade suction surface near the end wall has been shown to be effective in reducing these increased losses and enabling the efficient use of highly loaded blade designs. In this study, experiments are performed on the front loaded L2F low-pressure turbine airfoil in a linear cascade. The required mass flow is reduced by decreasing the hole count from previous configurations and from the introduction of unsteady blowing. The effects of pulsing frequency and duty cycle are investigated using phase-locked stereo particle image velocimetry to demonstrate the large scale movement and hysteresis behavior of the passage vortex interacting with the pulsed jets. Total pressure loss contours at the cascade outlet demonstrate that the efficiency benefit is maintained with the use of unsteady forcing. [DOI: 10.1115/1.4026127]

The current design of low-pressure turbines (LPTs) with steady-blowing vortex generating jets (VGJ) uses steady computational fluid dynamics (CFD). The present work aims to support this design approach by proposing a new semi-empirical transition model for injection-induced laminar-turbulent boundary layer transition. It is based on the detection of cross-flow vortices in the boundary layer which cause inflectional cross-flow velocity profiles. The model is implemented in the CFD code TRACE within the framework of the γ-Reθ transition model and is a reformulated, re-calibrated, and extended version of a previously presented model. It is extensively validated by means of VGJ as well as non-VGJ test cases capturing the local transition process in a physically reasonable way. Quantitative aerodynamic design parameters of several VGJ configurations including steady and periodic-unsteady inflow conditions are predicted in good accordance with experimental values. Furthermore, the quantitative prediction of end-wall flows of LPTs is improved by detecting typical secondary flow structures. For the first time, the newly derived model allows the quantitative design and optimization of LPTs with VGJs.Copyright © 2013 by ASME

Endwall surface film cooling effectiveness was measured on a turbine vane endwall surface using the pressure-sensitive paint (PSP) technique. A double staggered row of holes and a single row of discrete slots were used to supply film cooling in front of the nozzle cascade leading edges. Nitrogen gas was used to simulate film cooling flow as well as a tracer gas to indicate oxygen concentration such that film effectiveness by the mass transfer analogy could be obtained. Cooling mass flow was controlled to be 0.5 to 3.0 percent of the mainstream mass flow. The free-stream Reynolds number was about 283,000 and Mach number was about 0.11. The free-stream turbulence intensity was kept at 6.0 percent for all the tests, measured by a thermal anemometer. The PSP was calibrated at various temperatures and pressures to obtain better accuracy before being applied to the endwall surface. Film effectiveness distributions were measured on a flat endwall surface for five different mass flow rates. The film effectiveness increased nonlinearly with mass flow rate, indicating a strong interference between the cooling jets and the endwall secondary flows. At lower mass flow ratios, the secondary flow dominated the near wall flow field, resulting in a low film effectiveness. At higher mass flow ratios, the cooling jet momentum dominated the near wall flow field, resulting in a higher film effectiveness. The comparison between hole injection and slot injection was also made.

The current paper deals with the aerodynamic measurements in the rotational frame of reference of the Axial Flow Turbine Research Facility (AFTRF) at the Pennsylvania State University. Stationary frame measurements of ‘‘Mainstream Aerodynamic Effects Due to Wheelspace Coolant Injection in a High Pressure Turbine Stage’’ were presented in Part I of this paper. The relative aerodynamic effects associated with rotor ‐nozzle guide vane (NGV) gap coolant injections were investigated in the rotating frame. Three-dimensional velocity vectors including exit flow angles were measured at the rotor exit. This study quantifies the secondary effects of the coolant injection on the aerodynamic and performance character of the stage main stream flow for root injection, radial cooling, and impingement cooling. Current measurements show that even a small quantity (1 percent) of cooling air can have significant effects on the performance and exit conditions of the high-pressure turbine stage. Parameters such as the total pressure coefficient, wake width, and three-dimensional velocity field show significant local changes. It is clear that the cooling air disturbs the inlet end-wall boundary layer to the rotor and modifies secondary flow development thereby resulting in large changes in turbine exit conditions. Effects are the strongest from the hub to midspan. Negligible effect of the cooling flow can be seen in the tip region. @DOI: 10.1115/1.1397303#

This paper reports endwall film-cooling investigations with single and multiple rows of fan-shaped film holes using temperature-sensitive paint (TSP). The experiments are carried out in a six-bladed linear cascade based on the geometry of a highly loaded gas turbine first vane. The film effectiveness performance of the cooling rows is investigated under the influence of enhanced near-wall secondary flow. Tests are conducted at three different loading conditions changing the profile incidence. Film-cooling injection is established at elevated coolant density ratios of 1.4 using heated carbon dioxide. Due to the finite thermal conductivity of the wall material, the heat conduction effects observed in the measured temperature fields are assessed by a newly developed data analysis based on a finite element thermal analysis and tracking algorithms along CFD-computed near-wall surface streamlines. The results showed that the coolant trajectories are visibly influenced revealing the intense interaction between the film jets and the near-wall flow field. These effects are certainly enhanced with higher incidence leading to increased streamwise coolant consumption and reduced wall coverage. At the cascade inlet, the film-cooling injection is significantly affected by the near-wall flow field showing distinct over- and undercooled regions. Due to the enhanced deflection and mixing of the film jets injected from a single row, area-averaged film effectiveness and wall coverage decreases about 9 and 11%, respectively. With adding more cooling holes to this endwall area, the influence of the enhanced secondary flow becomes more pronounced. Hence, larger reduction in film effectiveness of 23% and wall coverage with 28% is observed. For single row injection at the airfoil pressure side, the stronger secondary flow motion with intensified streamwise mixing leads to a visibly decreased endwall coverage ratio of about 38% and maximum flow path reduction of about 41%. In this case, film effectiveness is found to be reduced up to 47% due to the small amount of coolant injected through this row. This effect is significantly smaller when more cooling rows are added showing an almost constant cooling performance for all incidence cases.

In a cooperative project between the Institute of Aircraft Propulsion Systems and MTU Aero Engines GmbH, a two-stage low pressure turbine with integrated 3D airfoil and endwall contouring is tested. The experimental data taken in the altitude test-facility study the effect of high incidence in off-design operation. Steady measurements are covering a wide range of Reynolds numbers between 40,000 and 180,000. The results are compared with steady multistage CFD predictions with a focus on the stator rows. A first unsteady simulation is taken into account as well. The CFD simulations include leakage flow paths with disk cavities modeled. Compared to design operation the extreme off-design high-incidence conditions lead to a different flow-field Reynolds number sensitivity. Airfoil lift data reveals changing incidence with Reynolds number of the second stage. Increased leading edge loading of the second vane indicates a strong cross channel pressure gradient in the second stage leading to larger secondary flow regions and a more three-dimensional flow-field. Global characteristics and area traverse data of the second vane are discussed. The unsteady CFD approach indicates improvement in the numerical prediction of the predominating flow-field.

A special naphthalene sublimation technique is used to study the film cooling performance downstream of one row of holes of 35° inclination angle and 45° compound angle with 3 diameter hole spacing and relatively small hole length to diameter ratio (6.3). Both film cooling effectiveness and mass/heat transfer coefficients are determined for blowing rates from 0.5 to 2.0 with density ratio of unity. The mass transfer coefficient is measured using pure air film injection, while the film cooling effectiveness is derived from comparison of mass transfer coefficients obtained following injection of naphthalene-vapor-saturated air with that of pure air injection. This technique enables one to obtain detailed local information on film cooling performance. General agreement is found in local film cooling effectiveness when compared with previous experiments. The laterally-averaged effectiveness with compound angle injection is higher than that with inclined holes immediately downstream of injection at a blowing rate of 0.5 and is higher at all locations downstream of injection at larger blowing rates. A large variation of mass transfer coefficients in the lateral direction is observed in the present study. At low blowing rates of 0.5 and 1.0, the laterally-averaged mass transfer coefficient is close to that of injection without compound angle. At the highest blowing rate used (2.0), the asymmetrical vortex motion under the jets increases the mass transfer coefficient drastically ten diameters downstream of injection.Copyright © 2000 by ASME

During the last few decades extremely powerful Quasi-three-dimensional (3D) codes and fully 3D Navier-Stokes solvers have been developed and successfully utilized in the design process and optimization of multistage axial-flow compressors. However, most of these methods proved to be difficult in handling and extremely time consuming. Due to these disadvantages, the primary stage design and stage matching as well as the off-design analysis is nowadays still based on fast 2D methods incorporating loss-, deviation- and end wall modeling. Only the detailed 3D optimization is normally performed by means of advanced 3D methods. In this paper a fast and efficient 2D calculation method is presented, which already in the initial design phase of multistage axial flow compressors, considers the influence of hub leakage flows, tip clearance effects, and other end wall flow phenomena. The method is generally based on the fundamental approach by Howard and Gallimore (1992). In order to allow a more accurate prediction of skewed and nondeveloped boundary layers in turbomachines, an improved theoretical approach was implemented. Particularly the splitting of the boundary layers into an axial and tangential component proved to be necessary in order to account for the change between rotating and stationary end walls. Additionally, a new approach is used for the prediction of the viscous end wall zones including hub leakage effects and strongly skewed boundary layers. As a result, empirical correlations for secondary flow effects are no longer required. The results of the improved method are compared with conventional 2D results including 3D loss- and deviation-models, with experimental data of a three-stage research compressor of the Institute for Jet Propulsion and Turbomachinery of the Technical University of Aachen and with 3D Navier-Stokes solutions of the V84.3A compressor and of a multistage Siemens research compressor. The results obtained using the new method show a remarkable improvement in comparison with conventional 2D methods. Due to the high quality and the extremely short computation time, the new method allows an overall viscous design of multistage compressors for heavy duty gas turbines and aeroengine applications.

The present study is conducted to investigate the local heat/mass transfer characteristics for flow through perforated plates. A naphthalene sublimation method is employed to determine the local heat/mass transfer coefficients on the effusion plate. Two parallel perforated plates are arranged in two different configurations: staggered and shifted in one direction. The experiments are conducted for hole pitch-to-diameter ratios of 6.0, for gap distance between the perforated plates of 0.33 to 10 hole diameters, and for Reynolds numbers of 5,000 to 12,000. The result shows that the high transfer region is formed at stagnation region and at the mid-line of the adjacent impinging jets due to secondary vortices and flow acceleration to the effusion hole. For flows through the perforated plates, the mass transfer rates on the surface of the effusion plate are about six to ten times higher than for effusion cooling alone (single perforated plate). In general, higher heat/mass transfer is obtained with smaller gap distance between two perforated plates.Copyright © 2000 by ASME

A pressure-correction based, 3D Navier-Stokes CFD code was used to simulate the effects of turbine parameters on the tip leakage flow and vortex in a linear turbine cascade to understand the detailed flow physics. A baseline case simulation of a cascade was first conducted in order to validate the numerical procedure with experimental measurements. The effects of realistic tip clearance spacing, inlet conditions, and relative endwall motion were then sequentially simulated, while maintaining previously modified parameters. With each additional simulation, a detailed comparison of the leakage flow’s direction, pressure gradient, and mass flow, as well as the leakage vortex and its roll-up, size, losses, location, and interaction with other flow features, was conducted. Part II of this two-part paper series focuses on the effect of relative motion of the outer casing on the leakage flow and vortex development. Casing relative motion results in less mass flow through the gap and a smaller leakage vortex. The structure of the aerothermal losses in the passage change dramatically when the outer casing motion was incorporated, but the total losses in the passage remained very similar. Additional secondary flows that are seen near the casing are also discussed.

Experimental results are presented which show the influence on the secondary flow and its losses by a profile modification of the leading edge very close to the endwall. The investigation was carried out with a well-known turbine profile that originally was developed for highly loaded low pressure turbines. The tests were done in a low speed cascade wind tunnel. The geometrical modification was achieved by a local thickness increase; a leading edge endwall bulb. It was expected that this would intensify the suction side branch of the horse-shoe (hs-) vortex with a desirable weakening effect on the passage vortex. The investigated configuration shows a reduction of secondary losses by 2.1% points that represents approximately 50% of these losses compared to the reference profile. Detailed measurements of the total pressure field behind the cascade are presented for both the reference and the modified profile. The influence of the modified hs-vortex on the overall passage vortex can be clearly seen. The results of a numerical analysis are compared with the experimental findings. A numerical analysis shows that the important details of the experimental findings can be reproduced. Quantitative values are locally different. The theoretical approach taken cannot yet be used for an exact prediction of the loss reduction. However the analysis of the interaction and the resulting tendencies are considered to be valid. Hence theoretical investigations as a guideline for the design of a leading edge bulb at the endwall are a valuable tool.

Mass injection upstream of the tip of a high-speed axial compressor rotor is a stability enhancement approach known to be effective in suppressing stall in tip-critical rotors. This process is examined in a transonic axial compressor rotor through experiments and time-averaged Navier-Stokes CFD simulations. Measurements and simulations for discrete injection are presented for a range of injection rates and distributions of injectors around the annulus. The simulations indicate that tip injection increases stability by unloading the rotor tip and that increasing injection velocity improves the effectiveness of tip injection. For the tested rotor, experimental results demonstrate that at 70 percent speed the stalling flow coefficient can be reduced by 30 percent using an injected massflow equivalent to 1 percent of the annulus flow. At design speed, the stalling flow coefficient was reduced by 6 percent using an injected massflow equivalent to 2 percent of the annulus flow. The experiments show that stability enhancement is related to the mass-averaged axial velocity at the tip. For a given injected massflow, the mass-averaged axial velocity at the tip is increased by injecting flow over discrete portions of the circumference as opposed to full-annular injection. The implications of these results on the design of recirculating casing treatments and other methods to enhance stability will be discussed.

A comprehensive experimental investigation was undertaken to explore the flow field in the tip clearance region of a turbine rotor to understand the physics of tip leakage flow. Specifically the paper looks at its origin, nature, development, interaction with the secondary flow, and its effects on performance. The experimental study was based on data obtained using a rotating five-hole probe, Laser Doppler Velocimeter, high-response pressure probes on the casing, and static pressure taps on the rotor blade surfaces. The first part of the paper deals with the pressure field and losses. Part II presents and interprets the vorticity, velocity, and turbulence fields at several axial locations. The data provided here indicates that the tip leakage vortex originates in the last half chord. The leakage vortex is confined close to the suction surface corner near the blade tip by the relative motion of the blade and the casing, and by the secondary flow in the tip region. The tip leakage flow clings to the blade suction surface until midchord then lifts off of the suction surface to form a vortex in the last 20 percent of the blade chord. The relative motion between blades and casing leads to the development of a scraping vortex that, along with the secondary flow, reduces the propagation of the tip leakage flow into the mainflow. The rotational effects and coriolis forces modify the turbulence structure in the tip leakage flow and secondary flow as compared to cascades.

Most turbine cascade studies in the literature have been performed in straight-endwall, high-aspect-ratio, linear cascades. As a result, there has been little appreciation for the role of, and added complexity imposed by, reduced aspect ratios. There also has been little documentation of endwall profiling with these reduced spans. To examine the role of these factors on cascade hydrodynamics, a large-scale nozzle guide vane simulator was constructed at the Heat Transfer Laboratory of the University of Minnesota. This cascade is comprised of three airfoils between one contoured and one flat endwall. The geometries of the airfoils and endwalls, as well as the experimental conditions in the simulator, are representative of those in commercial operation. Measurements with hot-wire anemometry were taken to characterize the flow approaching the cascade. These measurements show that the flow field in this cascade is highly elliptic and influenced by pressure gradients that are established within the cascade. Exit flow field measurements with triple-sensor anemometry and pressure measurements within the cascade indicate that the acceleration imposed by endwall contouring and airfoil turning is able to suppress the size and strength of key secondary flow features. In addition, the flow field near the contoured endwall differs significantly from that adjacent to the straight endwall. [S0889-504X(00)01104-1]

A flow modification technique is introduced in an attempt to allow increased turbine inlet temperatures. A large-scale two half-blade cascade simulator is used to model the secondary flow between two adjacent turbine blades. Various flow visualization techniques and measurements are used to verify that the test section replicates the flow of an actual turbine engine. Two techniques are employed to modify the endwall secondary flow, specifically the path of the passage vortex. Six endwall jets are installed at a location downstream of the saddle point near the leading edge of the pressure side blade. These wall jets are found to be ineffective in diverting the path of the passage vortex. The second technique utilizes a row of 12 endwall jets whose positions along the centerline of the passage are based on results from an optimized boundary layer fence. The row of jets successfully diverts the path of the passage vortex and decreases its effect on the suction side blade. This can be expected to increase the effectiveness of film cooling in that area. The row of jets increases the aerodynamic losses in the passage, however. Secondary flow measurements are presented showing the development of the endwall flow, both with and without modification. [S0889-504X(00)01004-7]

The purpose of this paper is to investigate tip-clearance and secondary flows numerically in a transonic compressor rotor. The computational method used is based on the numerical integration of the Favre-Reynolds-averaged three-dimensional compressible Navier–Stokes equations, using the Launder–Sharma near-wall k–ε turbulence closure. In order to describe the flowfield through the tip and its interaction with the main flow accurately, a fine O-grid is used to discretize the tip-clearance gap. A patched O-grid is used to discretize locally the mixing-layer region created between the jetlike flow through the gap and the main flow. An H–O–H grid is used for the computation of the main flow. In order to substantiate the validity of the results, comparisons with experimental measurements are presented for the NASA_37 rotor near peak efficiency using three grids (of 106, 2 X 106, and 3 X 106 points, with 21, 31, and 41 radial stations within the gap, respectively). The Launder–Sharma k–ε model underestimates the hub corner stall present in this configuration. The computational results are then used to analyze the interblade-passage secondary flows, the flow within the tip-clearance gap, and the mixing downstream of the rotor. The computational results indicate the presence of an important leakage-interaction region where the leakage-vortex after crossing the passage shock-wave mixes with the pressure-side secondary flows. A second trailing-edge tip vortex is also clearly visible.

For a single-stage transonic compressor rig at the TU Darmstadt, three-dimensional viscous simulations are compared to L2F measurements and data from the EGV leading edge instrumentation to demonstrate the predictive capability of the Navier–Stokes code TRACE_S. In a second step the separated regions at the blade tip are investigated in detail to gain insight into the mechanisms of tip leakage vortex-shock interaction at operating points close to stall, peak efficiency, and choke. At the casing the simulations reveal a region with axially reversed flow, leading to a rotationally asymmetric displacement of the outermost stream surface and a localized additional pitch-averaged blockage of approximately 2 percent. Loss mechanisms and streamline patterns deduced from the simulation are also discussed. Although the flow is essentially three-dimensional, a simple model for local blockage from tip leakage is demonstrated to significantly improve two-dimensional simulations on S1-surfaces.

A highly loaded turbine cascade has been redesigned with the objective to reduce the secondary flow by applying endwall contouring and three-dimensional airfoil design in the endwall regions. The overall loading and the axial area ratio of the cascade have been kept constant. With the tools of a three-dimensional design environment, a systematic study has been carried out regarding several features of the endwall pressure distribution and their influence on the secondary flow. Two optimized configurations have been investigated in a high-speed cascade wind tunnel. The flow field traverses showed improvements concerning the radial extent of the secondary flow and a decrease in secondary loss of 26 percent. Unfortunately this reduction was counterbalanced by increased profile losses and higher inlet losses due to increased blockage. The striking feature of the cascade with endwall contouring and three-dimensional airfoil design was a significant reduction of the exit flow angle deviations connected with the secondary flow. The predictions obtained by the three-dimensional Navier–Stokes solver TRACE_S showed a remarkable agreement with the experimental results.

A linear design system, already in use for the forward and inverse design of three-dimensional turbine aerofoils, has been extended for the design of their end walls. This paper shows how this method has been applied to the design of a nonaxisymmetric end wall for a turbine rotor blade in linear cascade. The calculations show that nonaxisymmetric end wall profiling is a powerful tool for reducing secondary flows, in particular the secondary kinetic energy and exit angle deviations. Simple end wall profiling is shown to be at least as beneficial aerodynamically as the now standard techniques of differentially skewing aerofoil sections up the span, and (compound) leaning of the aerofoil. A design is presented that combines a number of end wall features aimed at reducing secondary loss and flow deviation. The experimental study of this geometry, aimed at validating the design method, is the subject of the second part of this paper. The effects of end wall perturbations on the flow field are calculated using a three-dimensional pressure correction based Reynolds-averaged Navier–Stokes CFD code. These calculations are normally performed overnight on a cluster of work stations. The design system then calculates the relationships between perturbations in the end wall and resulting changes in the flow field. With these available, linear superposition theory is used to enable the designer to investigate quickly the effect on the flow field of many combinations of end wall shapes (a matter of minutes for each shape). [S0889-504X(00)00902-8]

A first-stage stator vane experiences high heat transfer rates, particularly near the endwall, where strong secondary flows occur. In order to improve numerical predictions of the complex endwall flow at low-speed conditions, benchmark quality experimental data are required. This study documents the flowfield in the endwall region of a stator vane that has been scaled up by a factor of nine while matching an engine exit Reynolds number of Reex=1.2×106. Laser Doppler velocimeter (LDV) measurements of all three components of the mean and fluctuating velocities are presented for several flow planes normal to the turbine vane. Measurements indicate that downstream of the minimum static pressure location on the suction surface of the vane, an attenuated suction side leg of the horseshoe vortex still exists. At this location, the peak turbulent kinetic energy coincides with the center of the passage vortex location. These flowfield measurements were also related to previously reported convective heat transfer coefficients on the endwall showing that high Stanton numbers occur where the passage vortex brings mainstream fluid toward the vane surface. [S0889-504X(00)00803-5]