Turbine Platform Cooling and Blade Suction Surface Phantom Cooling From Simulated Swirl Purge Flow

The paper presents the swirl purge flow on platform and a modeled land-based turbine rotor blade suction surface. Pressure sensitive paint (PSP) mass transfer technique provides detailed film cooling effectiveness distribution on platform and phantom cooling effectiveness on blade suction surface. Experiments have completed in a low speed wind tunnel facility with a five blade linear cascade. The inlet Reynolds number based on the chord length is 250,000. Swirl purge flow is simulated by coolant injection through fifty inclined cylindrical holes ahead of the blade leading edge row. Coolant injections from cylindrical holes go through nozzle endwall and a dolphin nose axisymmetric contour before reach platform and blade suction surface. Different “coolant injection angles” and “coolant injection velocity to cascade inlet velocity” results in various swirl ratios to simulate real engine conditions. Simulated swirl purge flow uses coolant injection angles of 30, 45, and 60 degrees to produce swirl ratios of 0.4, 0.6, and 0.8, respectively. Traditional purge flow has coolant injection angle of 90 degree to generate swirl ratio of 1. Coolant to mainstream mass flow rate ratio (MFR) is 0.5%, 1.0% and 1.5% for all swirl ratios. Coolant to mainstream density ratio maintains at 1.5 to match engine conditions. Most of the swirl purge and purge coolant approaches platform, but small amount of the coolant migrates to blade suction surface. Swirl ratio of 0.4 has highest relative motion between rotor and coolant and severely decreases film cooling and phantom cooling effectiveness. Higher MFR of 1% and 1.5% cases suffer from apparent decrement of the effectiveness while increasing relative motion.

Development of High Order LES Solver for Heat Transfer Applications Based on the Open Source OpenFOAM Framework

The computation of heat transfer phenomena in gas turbines plays a key role in the continuous quest to increase performance and life of both component and machine. In order to assess different cooling approaches computational fluid dynamics (CFD) is a fundamental tool. Until now the task has often been carried out with RANS simulations, mainly due to the relatively short computational time. The clear drawback of this approach is in terms of accuracy, especially in those situations where averaged turbulence-structures are not able to capture the flow physics, thus under or overestimating the local heat transfer. The present work shows the development of a new explicit high-order incompressible solver for time-dependent flows based on the open source C++ Toolbox OpenFOAM framework. As such, the solver is enabled to compute the spatially filtered Navier-Stokes equations applied in large eddy simulations for incompressible flows. An overview of the development methods is provided, presenting numerical and algorithmic details. The solver is verified using the method of manufactured solutions, and a series of numerical experiments is performed to show third-order accuracy in time and low temporal error levels. Typical cooling devices in turbomachinery applications are then investigated, such as the flow over a turbulator geometry involving heated walls and a film cooling application. The performance of various sub-grid-scale models are tested, such as static Smagorinsky, dynamic Lagrangian, dynamic one-equation turbulence models, dynamic Smagorinsky, WALE and sigma-model. Good results were obtained in all cases with variations among the individual models.

Full Coverage Shaped Hole Film Cooling in an Accelerating Boundary Layer With High Free-Stream Turbulence

Full coverage shaped-hole film cooling and downstream heat transfer measurements have been acquired in the accelerating flows over a large cylindrical leading edge test surface. The shaped holes had an 8° lateral expansion angled at 30° to the surface with spanwise and streamwise spacings of 3 diameters. Measurements were conducted at four blowing ratios, two Reynolds numbers and six well documented turbulence conditions. Film cooling measurements were acquired over a four to one range in blowing ratio at the lower Reynolds number and at the two lower blowing ratios for the higher Reynolds number. The film cooling measurements were acquired at a coolant to free-stream density ratio of approximately 1.04. The flows were subjected to a low turbulence condition (Tu = 0.7%), two levels of turbulence for a smaller sized grid (Tu = 3.5%, and 7.9%), one turbulence level for a larger grid (8.1%), and two levels of turbulence generated using a mock aero-combustor (Tu = 9.3% and 13.7%). Turbulence level is shown to have a significant influence in mixing away film cooling coverage progressively as the flow develops in the streamwise direction. Effectiveness levels for the aero-combustor turbulence condition are reduced to as low as 20% of low turbulence values by the furthest downstream region. The film cooling discharge is located close to the leading edge with very thin and accelerating upstream boundary layers. Film cooling data at the lower Reynolds number, show that transitional flows have significantly improved effectiveness levels compared with turbulent flows. Downstream effectiveness levels are very similar to slot film cooling data taken at the same coolant flow rates over the same cylindrical test surface. However, slots perform significantly better in the near discharge region. These data are expected to be very useful in grounding computational predictions of full coverage shaped hole film cooling with elevated turbulence levels and acceleration. IR measurements were performed for the two lowest turbulence levels to document the spanwise variation in film cooling effectiveness and heat transfer.

Flowfield and Heat Transfer Numerical Analysis of a Stator Vane Endwall With High Freestream Turbulence

Among the different parts subjected to hot gas flow, endwall heat transfer evaluation is particularly challenging because the flow is strongly affected by secondary effects. Large three-dimensional flow structures introduce remarkable spatial variation of heat transfer, both along streamwise and spanwise directions, making the use of simplified modelling approaches questionable in terms of reliability, and at the same time increasing the challenge for high fidelity computational methods. The aim of the present contribution is to describe the work done in the assessment of computational methods for the estimate of high pressure vane endwall heat transfer for industrial applications. Efforts were first devoted to the development and validation of an accurate computational procedure against a large set of aerodynamic and heat transfer data, available from literature, for both airfoil and endwall of a low-pressure linear cascade with low and high inlet turbulence levels. The analysis, focused on steady state computations, is principally devoted to the turbulence modelling assessment, including non-linear turbulence closure as well as transition modelling. Obtained results showed that the aerodynamics of both passage and endwall are well captured independently of the turbulence modelling while a large impact on both pattern and averaged value is verified for the heat transfer.

Film Cooling Effectiveness of Transonic Turbine Vane Suction Side With Compound-Angle Shaped Hole Configuration

This paper studies the effect of transonic flow velocity on local film cooling effectiveness distribution of turbine vane suction side, experimentally. A conduction-free Pressure Sensitive Paint (PSP) method is used to determine the local film cooling effectiveness. Tests were performed in a five-vane annular cascade at Texas A&M Turbomachinery laboratory blow-down flow loop facility. The exit Mach numbers are controlled to be 0.7, 0.9, and 1.1, from subsonic to transonic flow conditions. Three foreign gases N2, CO2 and Argon/SF6 mixture are selected to study the effects of three coolant-to-mainstream density ratios, 1.0, 1.5, and 2.0 on film cooling. Four averaged coolant blowing ratios in the range, 0.7, 1.0, 1.3 and 1.6 are investigated. The test vane features 3 rows of radial-angle cylindrical holes around the leading edge, and 2 rows of compound-angle shaped holes on the suction side. Results suggest that the PSP technique is capable of producing clear and detailed film cooling effectiveness contours at transonic condition. The effects of coolant to mainstream blowing ratio, density ratio, and exit Mach number on the vane suction-surface film cooling distribution are obtained, and the consequence results are presented and explained in this investigation.

Measurement and Prediction of Adiabatic Film Effectiveness of Combustor Representative Effusion Arrays

Thermal protection of gas turbine combustors relies heavily upon the delivery of a carefully managed film of coolant air to the hot-side of the combustor liner. Furthermore, improvements in engine sfc and the trend to ever more aggressive engine cycles means greater emphasis is being placed upon more efficient use of the proportion of combustion system air made available for cooling. As a result, there is a requirement to better understand the development of cooling films deposited onto the hot-side of the liner through complex effusion arrays. This study, therefore, is concerned with the prediction and measurement of adiabatic film effectiveness of a number of engine-representative designs. A RANS based CFD approach is used to predict film effectiveness in which computational cost is minimised by solving first for a single coolant passage to provide high fidelity, near-exit boundary conditions to the effusion arrays. Equivalent measurements are made for each test case using a Pressure Sensitive Paint (PSP) technique in which the oxygen-quenched fluorescence properties of the paint are employed together with a Nitrogen gas cooling simulant to determine adiabatic film effectiveness. This study demonstrates that whist the model under-predicts the mixing of the coolant with the main-stream flow, and hence the film development over the surface, the approach works well at quantifying the relative performance of each design.

Stagnation Region Heat Transfer Augmentation at Very High Turbulence Levels

A database for stagnation region heat transfer has been extended to include heat transfer measurements acquired downstream from a new high intensity turbulence generator. This work was motivated by gas turbine industry heat transfer designers who deal with heat transfer environments with increasing Reynolds numbers and very high turbulence levels. The new mock aero-combustor turbulence generator produces turbulence levels which average 17.4%, which is 37% higher than the older turbulence generator. The increased level of turbulence is caused by the reduced contraction ratio from the liner to the exit. Heat transfer measurements were acquired on two large cylindrical leading edge test surfaces having a four to one range in leading edge diameter (40.64 cm and 10.16 cm). Gandvarapu and Ames [1] previously acquired heat transfer measurements for six turbulence conditions including three grid conditions, two lower turbulence aero-combustor conditions, and a low turbulence condition. The data are documented and tabulated for an eight to one range in Reynolds numbers for each test surface with Reynolds numbers ranging from 62,500 to 500,000 for the large leading edge and 15,625 to 125,000 for the smaller leading edge. The data show augmentation levels of up to 136% in the stagnation region for the large leading edge. This heat transfer rate is an increase over the previous aero-combustor turbulence generator which had augmentation levels up to 110%. Note, the rate of increase in heat transfer augmentation decreases for the large cylindrical leading edge inferring only a limited level of turbulence intensification in the stagnation region. The smaller cylindrical leading edge shows more consistency with earlier stagnation region heat transfer results correlated on the TRL (Turbulence, Reynolds number, Length scale) parameter. The downstream regions of both test surfaces continue to accelerate the flow but at a much lower rate than the leading edge. Bypass transition occurs in these regions providing a useful set of data to ground the prediction of transition onset and length over a wide range of Reynolds numbers and turbulence intensity and scales.

Design, Characterization, and Verification of a Closed Loop Wind Tunnel With Linear Cascade and Upstream Wake Generator

A new cascade wind tunnel has been designed and constructed at the LSU Wind Tunnel Laboratory. The objective was to develop a versatile test facility, suitable for a wide range of experimental studies and measurements on turbine airfoils, especially with regards to film-cooling incorporating realistic unsteady effects due to passing wakes. The test section consists of a four passage linear cascade composed of three full blades and two shaped wall blades. The 2D blade shape profile of the cascade is a high-lift, low-pressure turbine L1A profile provided by the US Air Force Research Laboratories (AFRL), with a 152-mm axial chord. The Reynolds number based on the axial chord length at the nominal freestream velocity of 50 ms−1 is 500,000. A conveyor-based system was designed and fabricated to simulate the passing wakes of the upstream vanes (or blades) on the test blades (or vanes) depending on which airfoil types are put on the stationary frame and the moving frame of the conveyor. The original implementation uses blade profiles on the stationary frame and thick plate wake generators on the translating frame. Results are presented from hot-wire surveys conducted to characterize and qualify the velocity and turbulence intensity distributions and associated spectral characteristics at the cascade test section inlet, in the wake of the vanes and in the wake of the test blade. A blade instrumented with 123 pressure taps was used to acquire static pressure profiles of the cascade central blade, which were compared to the ones from the nominal airfoil design as well as to those obtained from a CFD simulation of the cascade flow. Incoming velocity and temperature profiles were found to be uniform to within a few percentage points, and the pressure coefficient distribution was found to be in good agreement with design values. The passage periodicity of the conveyor-belt-driven, flat-plates was verified and their wake was characterized. These results verified that the cascade wind tunnel operates according to design, thus proving to be a reliable test-bed for film cooling studies with and without unsteady wake effects. The design also incorporates an in-house-designed, miniature periscopic and adjustable laser sheet generating system integrated within the “dummy” blades to enable Particle Image Velocimetry measurements in the intra-blade domain.

Analysis of Turbulent Scalar Flux Models for a Discrete Hole Film Cooling Flow

Algebraic closures for the turbulent scalar fluxes were evaluated for a discrete hole film cooling geometry using the results from the high-fidelity Large Eddy Simulation (LES) of Bodart et al. [1]. Several models for the turbulent scalar fluxes exist, including the widely used Gradient Diffusion Hypothesis, the Generalized Gradient Diffusion Hypothesis [2], and the Higher Order Generalized Gradient Diffusion Hypothesis [3]. By analyzing the results from the LES, it was possible to isolate the error due to these turbulent mixing models. Distributions of the turbulent diffusivity, turbulent viscosity, and turbulent Prandtl number were extracted from the LES results. It was shown that the turbulent Prandtl number varies significantly spatially, undermining the applicability of the Reynolds analogy for this flow. The LES velocity field and Reynolds stresses were fed into a RANS solver to calculate the fluid temperature distribution. This analysis revealed in which regions of the flow various modeling assumptions were invalid and what effect those assumptions had on the predicted temperature distribution.

Experimental Study of Leakage Flow on Flow Field and Film Cooling of High Pressure Turbine Blade Platform

An experiment has been conducted to study stator/rotor disc cavity leakage flow on the platform of a highly loaded stationary linear blade cascade. The linear cascade consists of a scaled-up model of the high-pressure turbine blades of an E3 (Energy efficient engine) and leakage slot models installed under the platform. Experiments have been conducted to investigate the effect of the slot injection angle, leakage flow rates, distance between the leading edge of the blade and the slot, and spacing of the blades. The film-cooling effectiveness was measured by pressure sensitive paint (PSP), and the temperature fields and flow fields were investigated using laser-induced fluorescence (LIF) and particle image velocimetry (PIV), respectively. It was observed from the experiments that the leakage flow covered the surface of the blade platform when the distance between the leading edge and the slot was zero; however, with increasing distance, the horseshoe vortex dominates near the junction of the blade leading edge, and the leakage flow could not cover the region. It was also found that the leakage flow has an effect that promotes the formation of the horseshoe vortex for some experimental conditions.