ASME Conference Presenter Attendance Policy and Archival Proceedings

2013;():V03CT00A001. doi:10.1115/GT2013-NS3C.

This online compilation of papers from the ASME Turbo Expo 2013: Turbine Technical Conference and Exposition (GT2013) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

General Interest

2013;():V03CT14A001. doi:10.1115/GT2013-94242.

This paper compares predictions from a 3-D Reynolds-Averaged Navier-Stokes code and a statistical representation of measurements from a cooled 1-1/2 stage high-pressure transonic turbine to quantify predictive process sensitivity. A multivariable regression technique was applied to both the inlet temperature measurements obtained at the inlet rake, and the wall temperature and heat transfer measurements obtained via heat-flux gauges on the blade airfoil surfaces. By using the statistically-modeled temperature profiles to generate the inlet boundary conditions for the Computational Fluid Dynamics (CFD) analysis, the sensitivity of blade heat transfer predictions due to the variation in the inlet temperature profile and uncertainty in wall temperature measurements and surface roughness is calculated. All predictions are performed with and without cooling. Heat transfer predictions match reasonably well with the statistical representation of the data, both with and without cooling. Predictive precision for this study is driven primarily by inlet profile uncertainty followed by surface roughness and gauge position uncertainty.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A002. doi:10.1115/GT2013-94277.

Up to the present time, virtually all production gas turbine engines utilize high-pressure turbine components that are cooled by various modes of bulk, or macro, internal flow passages and networks, many of which are then linked to external film cooling by holes through the walls. Typical examples of macro cooling include serpentine passages, swirl chambers, pin banks, and impingement jets. All cooled turbine airfoils in commercial operation today utilize cooling channels and film holes that are considered to be macroscopic in physical magnitude. The conventional delimiters between macro and micro cooling are (1) whether the feature (internal passages and holes) can be repeatedly manufactured via investment casting methods, and (2) whether the flow passage (film hole, impingement orifice, dust hole) may become plugged by particles in the cooling fluid. The acceptable sizes are relative to the turbine size and the operating environment. This study examines the limitations and weaknesses inherent in macro cooling and the reasoning that demands gas turbine cooling change to micro cooling methods. A brief history of the developments in micro cooling for turbine airfoils is presented, including the reasons why it has not yet become commercially viable. A simple cooling study is used to demonstrate the very significant performance gains that can be obtained with micro cooling, such as cooling flow reductions approaching 40%, and thermal stress (gradient) reductions of 50%. The key to realizing these gains in full lies in cost effective manufacturing and durability in operation.

Topics: Cooling , Gas turbines
Commentary by Dr. Valentin Fuster
2013;():V03CT14A003. doi:10.1115/GT2013-94291.

Developments of High Pressure Turbine (HPT) blades and vanes are particularly affected by the accuracy and consistency in heat transfer prediction techniques. The conventional wisdom in heat transfer analysis considers that aerodynamics fully determines Heat Transfer Coefficient (HTC) distribution along the blade surface, so that wall temperature should have no influence on HTC. The effect of the wall temperature on the heat transfer coefficient has been rarely studied, although there have been some rather scattered correlations regarding the influence of the wall-inflow temperature ratio on HTC, largely based on simple geometry configurations. There seems to be a need to answer two related questions:

a) to what extent the conventional wisdom is valid, in particular for a transonic HPT blading?

b) what is a consistent way by which the HTC should be worked out in CFD (as well as in experiment)?

In this paper, computational analyses are carried out on the effects of wall thermal boundary condition on external HTC for a 2D Nozzle Guide Vane (NGV) profile. The study is performed by using Fluent® commercial Reynolds-Averaged Navier Stokes Equations solver. Further computations are also made using a coupled solid-fluid Conjugate Heat Transfer (CHT) method with an internally cooled blade. The present results show a strong dependence of HTC on wall temperature, far above that predicted by using the existing correlations, highlighting the importance of upstream boundary layer history on the HTC distribution. Influence of the wall temperature on the trailing edge shock position has been also observed. Based on the results found, a new method is proposed, allowing to model the HTC dependence on wall temperature with much improved accuracy as demonstrated.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A004. doi:10.1115/GT2013-94335.

Flow field near the turbine blade platform is very complex due to the secondary flow motions such as horseshoe vortices, passage vortices and endwall cross flows. It is therefore extremely difficult to predict the platform heat transfer distribution. As the secondary flows are largely affected by platform profile/shape, a number of investigators have studied different platform profiles to minimize aerodynamic loss and heat load. Understanding of the platform heat transfer has become especially critical in recent years, because of firing temperature increase and low NOx combustion requirement, as it is directly related to turbine durability. Three different axisymmetric platform profiles were designed and experimentally studied: flat profile, dolphin nose profile and shark nose profile. All of them were based on the existing engine hardware designs. The measurements were conducted in a high-speed linear cascade, which consisted of five blades and six flow passages. The test platforms were made of FR4 material and painted with Thermo-chromic Liquid Crystal (TLC). The test article was kept in the plenum located under the cascade at the pre-test condition. At the start of each test, the test blade/article was inserted into the cascade rapidly and then two CCD cameras recorded the color changes of the TLC on the platform surface. Engine representative Reynolds numbers were studied from 300,000 to 600,000 and the corresponding inlet Mach numbers were ranged from 0.12 to 0.24. The upstream section of the flat profile platform showed a typical flat plate heat transfer pattern with boundary layer development. The shark-nose and dolphin-nose platforms resulted in lower heat transfer coefficients on the upstream region compared to that for the flat profile, and the peak values moved slightly downstream from the leading edge due to possibly different secondary flow patterns. The heat transfer rate increased with increased Reynolds number for all three platform shapes, while the flat profile showed a higher increase rate. Zone averaged heat transfer distributions in addition to local values were also presented to show the effect of platform profile. In general, the flat profile platform resulted in a higher overall heat transfer rate than that for the other two profile platforms, which suggested that a good design of contoured profile platform could reduce the heat load and aerodynamic loss in gas turbine blade.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A005. doi:10.1115/GT2013-94345.

This study focuses on local heat transfer characteristics on the tip and near-tip regions of a turbine blade with a flat tip, tested under transonic conditions in a stationary, 2-D linear cascade with high freestream turbulence. The experiments were conducted at the Virginia Tech transonic blow-down wind tunnel facility. The effects of tip clearance and exit Mach number on heat transfer distribution were investigated on the tip surface using a transient infrared thermography technique. In addition, thin film gages were used to study similar effects in heat transfer on the near-tip regions at 94% height based on engine blade span of the pressure and suction sides. Surface oil flow visualizations on the blade tip region were carried-out to shed some light on the leakage flow structure. Experiments were performed at three exit Mach numbers of 0.7, 0.85, and 1.05 for two different tip clearances of 0.9% and 1.8% based on turbine blade span. The exit Mach numbers tested correspond to exit Reynolds numbers of 7.6 × 105, 9.0 × 105, and 1.1 × 106 based on blade true chord. The tests were performed with a high freestream turbulence intensity of 12% at the cascade inlet.

Results at 0.85 exit Mach showed that an increase in the tip gap clearance from 0.9% to 1.8% translates into a 3% increase in the average heat transfer coefficients on the blade tip surface. At 0.9% tip clearance, an increase in exit Mach number from 0.85 to 1.05 led to a 39% increase in average heat transfer on the tip. High heat transfer was observed on the blade tip surface near the leading edge, and an increase in the tip clearance gap and exit Mach number augmented this near-leading edge tip heat transfer. At 94% of engine blade height on the suction side near the tip, a peak in heat transfer was observed in all test cases at s/C = 0.66, due to the onset of a downstream leakage vortex, originating from the pressure side. An increase in both the tip gap and exit Mach number resulted in an increase, followed by a decrease in the near-tip suction side heat transfer. On the near-tip pressure side, a slight increase in heat transfer was observed with increased tip gap and exit Mach number. In general, the suction side heat transfer is greater than the pressure side heat transfer, as a result of the suction side leakage vortices.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A006. doi:10.1115/GT2013-94356.

Steam turbine start-up has a significant impact on the cyclic fatigue life. Modern steam turbines are operated at high temperatures for optimal efficiency, which results in high temperature differences relative to the condition before start-up. To achieve the fastest possible start-up time without reducing the lifetime of the turbine components due to excessive thermal stress, the start-up procedure of cyclic turbines is optimized to follow the specific material low cycle fatigue limit. For such optimization and to ensure reliable operation, it is essential to fully understand the thermal behavior of the components during start-up. This is especially challenging in low flow conditions, i.e. during pre-warming and early loading phase. A two-dimensional numerical procedure is described for the assessment of the thermal regime during start-up. The calculation procedure includes the rotor, casings, valves and main pipes. The concept of the start-up calculation is to replace the convective effect of the steam in the turbine cavity by an equivalent fluid over-conductivity that gives the same thermal effect on metallic parts. This approach allows simulating accurately the effect of steam ingestion during pre-warming phase. The fluid equivalent over-conductivity is calibrated with experimental data. At the end of the paper the impact of ingested steam temperature and mass-flow on the rotor cyclic lifetime is demonstrated. This paper is a continuation of papers [1] and [2].

Topics: Steam turbines
Commentary by Dr. Valentin Fuster
2013;():V03CT14A007. doi:10.1115/GT2013-94583.

The Inverse Flux Solver for Arbitrary Waveforms (IFSAW) algorithm is a transient, simultaneous solution of time resolved adiabatic effectiveness, η(t), and heat transfer coefficient, h(t). Numerical simulations showed IFSAW maintained its high accuracy despite two experimental sources of error typically found when using a transient heat transfer method. The traditional transient method involves exposing a film cooled wind tunnel model at uniform temperature to a step change in freestream temperature. The experimental design results in nearly one-dimensional heat transfer and allows the surface to be modeled as semi-infinite. Typically, the surface temperature history is correlated to an analytical solution to the governing heat transfer equation (yielding η and h), but the required temperature step change is impossible to achieve in a laboratory. This paper first analyzed the error introduced by imperfect step changes and evaluated an alternative methodology, IFSAW, requiring only an arbitrary change in freestream temperature occurring at any rate. Secondly, severe error in h (found in locations where η is near unity because the surface temperature changes little from the initial temperature) was shown to be mitigated using IFSAW combined with a gradual change in coolant temperature at any point during measurement. With both complications, IFSAW maintains its ability to determine periodic η(t) and h(t) waveforms. In these ways, IFSAW is shown to be superior to the legacy transient method.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A008. doi:10.1115/GT2013-94685.

The energy requirements associated with recovering greenhouse gases from Integrated Gasification Combined Cycle (IGCC) or Natural Gas Combined Cycle (NGCC) power plants are significant. The subsequent reductions in overall plant efficiency also result in a higher cost of electricity. In order to meet the future demand for cleaner energy production, this research is focused on improving gas turbine efficiency through advancements in gas turbine cooling capabilities.

For this study, an experimental approach was developed to quantify overall effectiveness and net heat flux reduction for a film-cooled test article at high temperature and pressure conditions. A major part of this study focused on validating an advanced optical thermography technique capable of distinguishing between emitted and reflected radiation from film-cooled test articles exposed to exhaust gases in excess of 1000°C and 5 bar. The optical thermography method was used to acquire temperature maps of both external and internal wall temperatures on a test article with fan-shaped film cooling holes. The overall effectiveness and heat flux were quantified with one experiment. The optical temperature measurement technique was capable of measuring wall temperatures to within ±7.2°C. Uncertainty estimates showed that the methods developed for this study were capable of quantifying improvements in overall effectiveness necessary to meet DOE program goals. Results showed that overall effectiveness increased with an increase in blowing ratio and a decrease in mainstream gas pressure while heat flux contours indicated consistent trends.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A009. doi:10.1115/GT2013-94715.

Typical turbomachinery flows are too complex to be predicted by analytical solutions alone. Therefore numerous correlations and test data are used in conjunction with numerical tools in order to design thermally critical components. This approach can be problematic because these correlations and data are not fully independent of the boundary conditions applied. The heat transfer coefficients obtained are not only dependent on the aerodynamics of the flow but also on the thermal boundary layer created along the surface. The adiabatic heat transfer coefficient is the only one which is independent of the thermal boundary conditions, as long as the energy equation can be considered linear with respect to the temperature. However, a proper prediction of the surface temperature cannot be obtained with the adiabatic heat transfer coefficient alone.

This paper first reviews the concept of adiabatic heat transfer coefficient and its application to turbomachinery flows. Later, a concept is introduced to allow interchanging between different definitions of heat transfer coefficient and boundary conditions, i.e. constant heat flux or constant wall temperature. Finally, a typical configuration for measuring the adiabatic heat transfer coefficient on a turbine blade and the conversion to other definitions of heat transfer coefficient is presented and evaluated. It is shown that with the technique presented here even small deficiencies of some experiments can be compensated for.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A010. doi:10.1115/GT2013-94741.

The Air Force Research Laboratory Turbine Research Facility is a transient blowdown facility that allows simultaneous measurement of unsteady heat transfer and aerodynamics on full scale engine hardware. It is unique for its size and consequent blowdown duration. Compared to engine validation testing, full scale, short duration turbine-rig testing is able to provide very large amounts of rotating turbine flowfield information using far less energy, orders of magnitude lower cost, and greater instrumentation selection.

This paper provides an updated review of the facility’s history, operation, and enhancements since its initial construction over two decades ago. Historical connections to pioneering work in short-duration turbine heat transfer testing are highlighted, and an overview of past developments, features, and capabilities is given. More recent experimental and computational integration is described using a suite of in-house developed CFD design and analysis tools. Example test programs include a non-proprietary 1+ 1/2 stage research turbine rig, which is the most heavily instrumented high pressure turbine tested in the facility to date.. Recent data illustrates the character of unsteady airfoil shock interactions that may lead to large levels of resonant stress or turbine high cycle fatigue. The paper ends with a brief discussion of future work.

Topics: Turbines
Commentary by Dr. Valentin Fuster
2013;():V03CT14A011. doi:10.1115/GT2013-94754.

In high-speed unshrouded turbines tip leakage flows generate large aerodynamic losses and intense unsteady thermal loads over the rotor blade tip and casing. The stage loading and rotational speeds are steadily increased to achieve higher turbine efficiency, and hence the overtip leakage flow may exceed the transonic regime. However, conventional blade tip geometries are not designed to cope with supersonic tip flow velocities. A great potential lays in the modification and optimization of the blade tip shape as a means to control the tip leakage flow aerodynamics, limit the entropy production in the overtip gap, manage the heat load distribution over the blade tip and improve the turbine efficiency at high stage loading coefficients.

The present paper develops an optimization strategy to produce a set of blade tip profiles with enhanced aerothermal performance for a number of tip gap flow conditions. The tip clearance flow was numerically simulated through two-dimensional compressible Reynolds-Averaged Navier-Stokes (RANS) calculations that reproduce an idealized overtip flow along streamlines. A multi-objective optimization tool, based on differential evolution combined with surrogate models (artificial neural networks), was used to obtain optimized 2D tip profiles with reduced aerodynamic losses and minimum heat transfer variations and mean levels over the blade tip and casing. Optimized tip shapes were obtained for relevant tip gap flow conditions in terms of blade thickness to tip gap height ratios (between 5 and 25), and blade pressure loads (from subsonic to supersonic tip leakage flow regimes) imposing fixed inlet conditions. We demonstrated that tip geometries which perform superior in subsonic conditions are not optimal for supersonic tip gap flows. Prime tip profiles exist depending on the tip flow conditions. The numerical study yielded a deeper insight on the physics of tip leakage flows of unshrouded rotors with arbitrary tip shapes, providing the necessary knowledge to guide the design and optimization strategy of a full blade tip surface in a real 3D turbine environment.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A012. doi:10.1115/GT2013-94921.

This study compares surface pressure measurements and predictions for a high pressure turbine first-stage nozzle vane. The surface pressure measurements were taken in a 3D annular cascade, consisting of four airfoils and five passages. The cascade was uncooled, axisymmetric at both inner and outer endwalls, and reproduced the design intent Reynolds and Mach numbers of the real engine component. Static pressure measurements were taken along the airfoil profile at 15, 50, and 85% span, with duplicate midspan measurements across the four airfoils for assessing the tangential periodicity of the flow. Static pressure measurements were also taken on the inner and outer endwall surfaces of the center airfoil passage, with 40 measurement points uniformly distributed over each endwall. Three methods of surface pressure prediction were compared with the data: (1) a 2D inviscid CFD solution of a single airfoil passage at fixed spanwise locations, (2) a 3D RANS CFD solution of a single airfoil passage, and (3) a 3D RANS CFD solution of the full five-passage cascade domain. Both of the single-passage solutions assumed flowfield periodicity in the tangential direction and compared favorably to the center passage airfoil data. This finding suggested that the cascade center passage was sufficiently representative of the full-annulus turbomachine environment and validated the cascade for further experimental studies. The adjacent airfoil pressure measurements quantified the passage-to-passage variation in the cascade flowfield, and the 3D full-cascade CFD compared favorably with the peripheral airfoil data. The full-cascade CFD also compared favorably with the data on both endwalls: with an average and maximum deviation of 0.5 and 2%, respectively. These findings provide confidence in the 3D CFD methods for use in determining local flow rates from cooling/leakage geometry, and serve as an important first step toward validating the methods for real-engine blockage effects like coolant and endwall contouring.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A013. doi:10.1115/GT2013-94926.

Short-duration facilities have been used for the past thirty-five years to obtain measurements of heat transfer, aerodynamic loading, vibratory response, film-cooling influence, purge flow migration, and aeroperformance for full-stage high-pressure turbines operating at design corrected conditions of flow function, corrected speed, and stage pressure ratio. This paper traces the development of experimental techniques now in use at The Ohio State University (OSU) Gas Turbine Laboratory (GTL) from initial work in this area at the Cornell Aeronautical Laboratory (CAL, later to become Calspan) in 1975 through to the present. It is intended to summarize the wide range of research that can be performed with a short-duration facility and highlight the types of measurements that are possible.

Beginning with heat-flux measurements for the vane and blade of a Garrett TFE 731-2 HP turbine stage with vane pressure-surface slot cooling, the challenge of each experimental program has been to provide data to aid turbine designers in understanding the relevant flow physics and help drive the advancement of predictive techniques. Through many different programs, this has involved collaborators at a variety of companies and experiments performed with turbine stages from Garrett, Allison, Teledyne, Pratt and Whitney, General Electric Aviation, Rocketdyne, Westinghouse, and Honeywell. The Vane/Blade Interaction measurement and CFD program, which ran from the early eighties until 2000, provided a particularly good example of what can be achieved when experimentalists and computational specialists collaborate closely. Before conclusion of this program in 2000, the heat-flux and pressure measurements made for this transonic turbine operated with and without vane trailing edge cooling flow were analyzed and compared to predictive codes in conjunction with engineers at Allison, United Technologies Research Center, Pratt and Whitney, and GE Aviation in jointly published papers.

When the group moved to OSU in 1995 along with the facility used at Calspan, refined techniques were needed to meet new research challenges such as investigating blade damping and forced response, measuring aeroperformance for different configurations, and preparing for advanced cooling experiments that introduced complicating features of an actual engine to further challenge computational predictions. This required conversion of the test-gas heating method from a shock-tunnel approach to a blowdown approach using a combustor emulator to also create inlet temperature profiles, the development of instrumentation techniques to work with a thin-walled airfoil with backside cooling, and the adoption of experimental techniques that could be used to successfully operate fully cooled turbine stages (vane row cooled, blade row cooled, and proper cavity purge flow provided). Further, it was necessary to develop techniques for measuring the aeroperformance of these fully cooled machines.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A014. doi:10.1115/GT2013-94927.

There is a growing trend toward the use of conjugate CFD for use in prediction of turbine cooling performance. While many studies have evaluated the performance of RANS simulations relative to experimental measurements of the momentum boundary layer, no studies have evaluated their performance in prediction of the accompanying thermal boundary layer. This is largely due to the fact that, until recently, no appropriate experimental data existed to validate these models. This study compares several popular RANS models — including the realizable k-ε and k-ω SST models — with a four equation k-ω model (“Transition SST”) and experimental measurements at selected positions on the pressure and suction sides of a model C3X vane. Comparisons were made using mean velocity and temperature in the boundary layer without film cooling under conditions of high and low mainstream turbulence. The best performing model was evaluated using modification of the turbulent Prandtl number to attempt to better match the data for the high turbulence case. Overall, the models did not perform well for the low turbulence case; they greatly over-predicted the thermal boundary layer thickness. For the high turbulence case, their performance was better. The Transition SST model performed the best with an average thermal boundary layer thickness within 15% of the experimentally measured values. Prandtl number variation proved to be an inadequate means of improving the thermal boundary layer predictions.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A015. doi:10.1115/GT2013-94942.

Aerodynamic loss and endwall heat transfer for a turbine blade are influenced by complex vortical flows that are generated at the airfoil-endwall junction. In an engine, those flows interact with clearance gaps between stationary and rotating components, as well as with leakage flow that is designed to exhaust through the gaps. This paper describes experimental measurements of endwall heat transfer for a high-pressure turbine blade with an endwall overlap geometry, as well as an upstream leakage feature that supplied swirled or unswirled leakage relative to the blade. For unswirled leakage, increasing its mass flow increased the magnitude and pitchwise uniformity of the heat transfer coefficient upstream of the blades although heat transfer further into the passage was unchanged. Leakage flow with swirl shifted the horseshoe vortex in the direction of swirl and increased heat transfer on the upstream blade endwall, as compared to unswirled leakage. For a nominal leakage mass flow ratio of 0.75%, swirled leakage did not increase area-averaged heat transfer relative to unswirled leakage. At a mass flow ratio of 1.0%, however, swirled leakage increased overall heat transfer by 4% due to an increase in the strength of the vortical flows.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A016. doi:10.1115/GT2013-94945.

Porous metals have long been considered as an ideal material in which to manufacture turbine components given the inherent large convective surface area. One consideration, however, in using porous metals is the increase in pressure drop that accompanies these materials. To characterize increases in pressure drop for porous materials, flow measurements were made on numerous porous metal coupons. The porosity of the coupons investigated had a range of four in terms of density. A technique for determining the effective internal flow area from pressure drop measurements was developed to provide an effective diameter. The pressure drop measurements were compared to an ideal isentropic compressible-flow nozzle and to a smooth, straight-walled tube. The comparisons show that the porous channels have a similar, but much larger pressure drop than the smooth walls. The experiments performed demonstrated that these porous geometries can be scaled to provide generalized pressure drop characteristics for all geometries.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A017. doi:10.1115/GT2013-95008.

The endwall heat transfer characteristics of forced flow past outlet guide vanes (OGVs) in a linear cascade have been investigated by using a liquid crystal thermography (LCT) method. The Reynolds number is kept at 250,000 and both on- and off-design conditions are tested. For the on-design condition where the incidence angle of OGVs is 30°, no obvious flow separation phenomenon was observed; on the contrary, for the off-design conditions where the incidence angle of OGVs is 0° and −31°, respectively, remarkable flow separation was noticed. The results indicate that the incidence angle of OGVs has a significant effect on the endwall heat transfer. In general, the endwall heat transfer coefficients for the off-design conditions are higher than the corresponding on-design condition. In addition, a preliminary CFD analysis was performed and presented. Basically, the results are consistent with the experiments but further investigations are needed in the future work.

Topics: Heat transfer , Design
Commentary by Dr. Valentin Fuster
2013;():V03CT14A018. doi:10.1115/GT2013-95009.

According to current trends in the energy market, heavy duty gas turbines are increasingly being used to fill gaps in the power energy supply and are less frequently operated in pure steady-state base load conditions. This tendency implies more rapid load ramps and is confirmed by utilities’ requirements for more operational flexibility in order to increase their net revenues. In order to assess the effects of such load variations on temperature gradients withstood by the various components, a series of simple correlations are derived that take in account key operating parameters of gas turbines. To this end, each blade and vane has been schematized as a compound of different portions to which specific values of cooling efficiency and gas temperature were assigned. This results in a simplified model of the engine allowing for the prediction of the temperature gradients on the base material of the critical zones of blades and vanes as a function of different cooling schemes. Method results can subsequently be exploited both to improve thermal design of hot gas path components, as well as to set up material testing campaigns targeting any specific duty cycle.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A019. doi:10.1115/GT2013-95048.

Engine companies typically emphasize research which has been conducted at conditions as close to engine conditions as possible. This focus on engine relevant conditions often causes difficulties in University research laboratories. One particularly difficult testing regime is high speed but low Reynolds number flows. High speed low Reynolds number flows can occur in both low pressure turbines under a normal range of engine operating conditions and in high pressure turbines run at very high altitudes. This paper documents a new steady state closed loop wind tunnel facility which has been developed to study high speed cascade flows at low Reynolds numbers. The initial test configuration has been representative of a first stage vane configuration for a UAV turbofan which flies at a very high altitude. The initial test section was configured in a three full passage four-vane linear cascade arrangement with upper and lower bleed flows. Both heat transfer and aerodynamics loss measurements were acquired and are presented in this paper. Heat transfer measurements were taken at a Reynolds number of 720,000 based on true chord and exit conditions at Mach numbers of 0.7, 0.8, and 0.9. Exit survey measurements were conducted at a chord exit Reynolds number of 720,000 over a similar range in Mach numbers. However, this facility has the capability to run at chord Reynolds numbers of 90,000 or below in the present configuration which uses an approximately three times scale test vane.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A020. doi:10.1115/GT2013-95055.

Relatively small scale turbulence is known to intensify in the presence of a stagnation region due to the elongation of these eddies by the mean strain field of the approach flow. Experimental evidence also demonstrates that the large scale eddies are blocked as they approach presence of the stagnation surface. Recent heat transfer measurements suggest that very high intensity turbulence or turbulence in the presence of very large scale leading edge regions may not be as strongly influenced by the stagnation region strain field. Understanding the physics of turbulence is critical to the improvement of turbulence models which are used to predict the surface heat load in gas turbine hot sections.

This paper documents the response of high intensity turbulence in the approach flow of two large cylindrical leading edge regions. Measurements of turbulence intensity, scale, spectra, and dissipation have been acquired for five elevated levels of turbulence in the approach flow of two large diameter (0.1016 m and 0.4064 m) leading edge regions. Generally, three influences were observed. Initially, in the presence of the largest cylinder the smaller scale higher intensity turbulence showed increased decay due to longer effective convection times. Secondly, dissipation levels, as estimated from the inertial subrange of the one-dimensional spectra, initially decreased then increased as the strain field intensified in the presence of the stagnation regions. Finally, the measurements indicated that the energy in the low wave number spectra was increasingly blocked in the near wall region of the leading edge.

Topics: Turbulence
Commentary by Dr. Valentin Fuster
2013;():V03CT14A021. doi:10.1115/GT2013-95061.

As a result of continuous enhancement of the capability of it’s gas turbines, proposed uprated versions the SGT-100 engine were predicted have pre-mature oxidation of the honey-comb seal material above the CT1 blade. This would lead to an increase in over tip leakage and degradation of engine performance. A project to increase the service life of the honeycomb was started and various options considered. Initially a CFD analysis of the region was carried out in order to understand the flow field in the cavity surrounding the HP blade shroud, and hence the source of the high temperatures. Further CFD studies were carried out to assess the effect of applying cooling air into the cavity, and the optimal position to introduce such flow so it had maximum effect on the honeycomb temperatures. Modifications were then carried out to engine components to replicate cooling proposed by the CFD analysis and a series of engine tests with, and without the cooling were carried out. The Shroud temperatures were measured during engine operation by infra-red Pyrometry, which confirmed the effectiveness of the cooling modification.

Topics: Cooling , oxidation
Commentary by Dr. Valentin Fuster
2013;():V03CT14A022. doi:10.1115/GT2013-95108.

The effect of hot streaks on deposition in a high pressure turbine vane passage was studied both experimentally and computationally. Modifications to Ohio State’s Turbine Reaction Flow Rig allowed for the creation of simulated hot streaks in a four-vane annular cascade operating at temperatures up to 1093°C. Total temperature surveys were made at the inlet plane of the vane passage, showing the variation caused by cold dilution jets. Deposition was generated by introducing sub-bituminous ash particles with a median diameter of 11.6 μm far upstream of the vane passage. Results indicate a strong correlation between surface deposits and the hot streak trajectory. A computational model was developed in Fluent to simulate both the flow and deposition. The flow solution was first obtained without particulates, and individual ash particles were subsequently introduced and tracked using a Lagrangian tracking model. The critical viscosity model was used to determine particle sticking upon impact with vane surfaces. Computational simulations confirm the migration of the hot streak and locations susceptible to enhanced deposition. Results show that the deposition model is overly sensitive to temperature and can severely overpredict deposition. Model constants can be tuned to better match experimental results, but must be calibrated for each application.

Topics: Turbines
Commentary by Dr. Valentin Fuster
2013;():V03CT14A023. doi:10.1115/GT2013-95164.

HPT operate at high pressure and temperatures. One of the most important loss sources is the tip leakage flow on the rotor tip region. The flow that leaks in this region does not participate in the energy transfer process between the hot gas and rotor blade row. Hence, the main flow suffers a penalty to maintain the energy conservation. To try decreasing this mass flow leakage some techniques can be applied. The most common are the winglet and squealer rotor tip configuration. These techniques improve the turbine performance, but some attention should be taken into account because the temperature distribution changes on this region for different tip configurations. In this work, the winglet and squealer tip geometries are compared with the common flat tip configuration. The analysis was performed for design and off-design conditions. The HPT developed in the E3 program was used as baseline turbine to explore the differences of the flowfield on the rotor tip region. The results are compared and discussed in detail.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A024. doi:10.1115/GT2013-95236.

Variations from manufacturing can influence the overall pressure drop and subsequent flow rates through supply holes in such applications as film-cooling, transpiration cooling, and impingement cooling that are supplied by micro-channels, pipe-flow systems, or secondary air systems. The inability to accurately predict flow rates has profound effects on engine operations. The objective of this study was to investigate the influence of several relevant manufacturing features that might occur for a cooling supply hole being fed by a range of channel configurations. The manufacturing variances included the ratio of hole diameter to channel width, the number of channel feeds (segments), the effect of hole overlap with respect to the channel sidewalls, and channel Reynolds number. Results showed that the friction factors for the typically long channels in this study, were independent of the inlet and exit hole configurations tested. Results also showed that the non-dimensional pressure loss coefficients for the flow passing through the channel inlet holes and through the channel exit holes were found to be independent of the channel flow Reynolds number over the range tested. The geometric scaling ratio of the hole cross-sectional area to the channel cross-sectional area collapsed the pressure loss coefficients the best for both one and two flow segments for both the channel inlet and channel exit hole.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A025. doi:10.1115/GT2013-95281.

Supersonic wind tunnels provide controlled test environments for aerodynamic research on scaled models. During the experiment, the stagnation pressure in the test section is required to remain constant. Due to the nonlinearity and distributed characteristics of the controlled system, a robust controller with effective flow control algorithms is required, which is then capable of properly working under different operating conditions. In this paper, an Extended Kalman Filter (EKF) based flow control strategy is proposed and implemented in the controller. The control strategy is designed based on the state estimation of a real blowdown wind tunnel, which is carried out under an EKF structure. One of the distinctive advantages of the proposed approach is its adaptability to a wide range of operating conditions for blowdown wind tunnels. Furthermore, it provides a systematic approach to tune the controller parameters to ensure the stability of the controlled air flow. Experiments with different initial conditions and control targets have been conducted to test the applicability and performance of the designed controller. The results demonstrate that the controller and its strategies can effectively control the stagnation pressure in the test section and maintain the target pressure during the stable stage of the blowdown process.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A026. doi:10.1115/GT2013-95687.

Gas turbine engine efficiency and reliability is generally improved through better understanding and improvements to the design of individual components. The life limiting component of the modern gas turbine is the high pressure (HP) turbine stage due to the arduous environment. Over the last 50 years significant research effort has been focused on advancing blade cooling designs and materials.

Due to practical limitations little fundamental research on the turbine system is performed in the operating gas turbine engine. Consequently different types of experimental approaches have been developed over the last 4 decades to study the flow and in particular the heat transfer and cooling in turbines.

In general the facilities can be divided into continuous running or short duration and cascade or rotating. Over the last 30 years short duration facilities have dominated the research in the study of turbine heat transfer and cooling.

The Oxford Turbine Research Facility (formerly known as the QinetiQ Turbine Test Facility, The Isentropic Light Piston Facility and The Isentropic Light Piston Cascade) is a short duration facility developed and built in the late 1970s and early 1980s for turbine heat transfer and cooling studies.

This paper presents the developments and measurements taken on the facility over the last 35 years, including the type of research that has been conducted and, the current capability of the facility.

Topics: Turbines
Commentary by Dr. Valentin Fuster
2013;():V03CT14A027. doi:10.1115/GT2013-95823.

This paper presents an industrial perspective on the potential use of multiple-airfoil row, unsteady CFD calculations in high-pressure turbine design cycles. A sliding-mesh unsteady CFD simulation is performed for a high-pressure turbine section of a modern aviation engine at conditions representative of engine take-off. The turbine consists of two stages plus a center-frame strut upstream of the low-pressure turbine. The airfoil counts per row are such that a half-annulus model domain must be simulated for periodicity. The total model domain size is 170MM computational grid points, and the solution requires approximately 9 days of clock time on 6,288 processing cores of a Cray XE6 supercomputer. Airfoil and endwall cooling flows are modeled via source term additions to the flow. The endwall flowpath cavities and their purge/leakage flows are resolved in the computational meshes to an extent. The time-averaged temperature profile solution is compared with static rake data taken in engine tests. The unsteady solution shows a considerable improvement in agreement with the rake data, compared with a steady-state solution using circumferential mixing planes. Passage-to-passage variations in gas temperature prediction are present in the 2nd stage, due to non-periodic alignment between the nozzle vanes and rotor blades. These passage-to-passage differences are quantified and contrasted.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A028. doi:10.1115/GT2013-95841.

Heat transfer in a high-pressure turbine configuration (from an experiment documented in [1–2]) has been analyzed by means of large-eddy simulation. Blair’s large-scale rotating rig consists of a first stator, a rotor and an exit stator. Flow and heat transfer in the first stator are assessed for two configurations — with and without the presence of turbulence generating grid. A particular challenge here is that turbulence grid generates fairly high levels of inlet turbulence with turbulence intensity (TU) of about 10% just upstream of leading edge; this in turn moves the transition location upstream in a dramatic fashion. As far as the rotor blade is concerned, the flow and heat transfer is also analyzed experimentally for a range of incidence angles assessing the pressure side heat transfer increase at negative incidence angles. Several challenging aspects relevant to flow in the rotor are also considered — the three-dimensionality of pressure side flow separation at negative incidence, the impact of upstream stator wakes, as well as the role of surface roughness.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A029. doi:10.1115/GT2013-95886.

The aim of this study was to provide a characterization of non-symmetrical operation in two counter-flow primary surface exhaust gas recuperators installed in parallel flow loops. The hybrid system emulator test rig and facility designed and operated by the Department of Energy, National Energy Technology Laboratory located at the West Virginia (USA) campus was used for the study. Various tests from the past years often resulted in non-symmetrical operation, indicated by significantly variant temperature measurements at the outlets of the recuperators.

Some specific tests have been carried out in order to identify the possible cause of this flow unbalance. The isolated effects of bleed air, cold air and hot air valve on the heat exchangers flow unbalance have been studied. Also, the impact of load bank changes on flow distribution has been considered in this study. Each test has been carried out in close loop fuel valve speed control. The influence of each independent variable in the study on parallel recuperator flow distribution has been quantitatively characterized using temperatures and a heat balance. Both the bleed and the cold air compressor bypass valves showed an appreciable impact on the heat exchangers flow unbalance, while hot air valve and load bank changes had minimal effect.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A030. doi:10.1115/GT2013-95909.

The quality of information, which is necessary to help designers further improve turbine engine performance, requires sophisticated analytics working hand-in-hand with well-developed experimental methods. Historically, the test instrumentation used in harsh, real engine conditions was short-lived, invasive, and not very accurate. This was accepted as the practical reality and the data obtained in an engine test cell had been used as little more than a sanity check or a trend indicator during the design process. Today expectations are much higher and the challenge is to develop experimental tools that can deliver the accuracy required to verify analytical predictions, calibrate computer models and to provide ground for critical design decisions in a way which was not possible before.

Successful introduction of UCTS (Uniform Crystal Temperature Sensor) Technology to the leading engine manufacturers demonstrated that it has the potential to overcome the typical issues of testing in a real engine environment. It is robust, non-intrusive and capable of high accuracy temperature measurement. It is based on the mechanism of heat transfer conduction, of which the fundamental theory is rigorous and simple. Our experience has shown that in order for a UCTS-based system to realize its promise, all potential sources of error must be tightly managed. LG Tech-Link identified important factors of influence that could complicate measurement and increase its uncertainty. Among them are variations in the part’s geometry, TBC thickness, boundary conditions, and installation methodology. These have been selected as the focus of this study.

The authors of this paper are using 3D Finite Element Analysis (FEA) methodology to investigate the possible pitfalls in the process of UCTS application that could cause loss of accuracy. It is the authors’ intention to probe the sensitivity of temperature at the location of the sensor to the major technological factors. The findings emphasize the value of collaboration between instrumentation, test and analytical engineers when planning engine tests and interpreting their results. Practicing engineers will be able to use the presented recommendations, methodologies and case studies to ensure the application of UCTS in their projects is accurate, compatible with test objectives and cost effective.

Commentary by Dr. Valentin Fuster
2013;():V03CT14A031. doi:10.1115/GT2013-95984.

Advances in turbine-based engine efficiency and reliability are achieved through better knowledge of the mechanical interaction with the flow. The life-limiting component of a modern gas turbine engine is the high-pressure (HP) turbine stage due to the arduous environment. For the same reason, real gas turbine engine operation prevents fundamental research. Various types of experimental approaches have been developed to study the flow and in particular the heat transfer, cooling, materials, aero-elastic issues and forced response in turbines. Over the last 30 years short duration facilities have dominated the research in the study of turbine heat transfer and cooling.

Two decades after the development of the von Karman Institute compression tube facility (built in the 90s), one could reconsider the design choices in view of the modern technology in compression, heating, control and electronics. The present paper provides first the history of the development and then how the wind tunnel is operated. Additionally the paper disseminates the experience and best practices in specifically designed measurement techniques to both experimentalists and experts in data processing. The final section overviews the turbine research capabilities, providing details on the required upgrades to the test section.

Topics: Turbines , Compression
Commentary by Dr. Valentin Fuster

Combustors (With Combustion, Fuels and Emissions)

2013;():V03CT17A001. doi:10.1115/GT2013-94209.

Gas turbine engine uses diffuser system to decelerate the compressor exit flow velocity before it enters combustor, it is important to design the compact structure and high performance of the diffuser for gas turbine engine. The diffuser and combustor dome configurations are critical flow path parameters in the design of a low-pressure-loss, high-performance combustion system. With rising of the inlet Mach number of the combustor, dramatically increasing of the diffuser total pressure loss and flow separation. So a new distributor diffuser was designed. In this paper preliminary results from an experimental investigation into the aerodynamic performance on a rectangle combustor-diffuser system with seven distributor plates were presented. Measurements were taken in the diffuser section to assess the diffuser performance characteristics under various conditions, the appropriate outlet flow field can be attained by changing the plate area ratio and form. Tests were carried out to investigate the influence of distributor diffuser plate geometry. During these measurements for each parametric configuration, data were obtained at 24 different flow rates through the distributor diffuser, it gave the conclusion that the distributor diffuser area ratio could be more than traditional diffusers with shorter construction and higher pressure recovery performance, while the flow loss through it was not beyond the traditional limit. Overall static pressure recovery improves and overall total pressure loss reduces with increasing distributor diffuser area ratio, and the increased flow rates through the distributor diffuser gave rise to a higher total pressure loss. The total pressure loss fraction was less than 2.5% when Mach number changed from 0.3 to 0.38; if the area ratio was more than 2.1, the diffuser loss coefficient remained less than 0.3, pressure recovery coefficient more than 0.5 and area ratio up to 2.45. There exists an area ratio in 1.6∼2.0 which makes diffuser outlet flow field distribution more uniform; Baffle structure can adjust the flow field distribution of outlet diffuser. As a result, the distributor diffuser can be potentially satisfied with demands for high performance combustor.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A002. doi:10.1115/GT2013-94474.

Reverse-flow combustors have been used in heavy, land-based gas turbines for many decades. A sheath is typically installed over the external walls of the combustor and transition piece to provide enhanced cooling through hundreds of small jet impingement cooling, followed by a strong forced convention channel flow. However, this cooling is at the expense of large pressure loss. With the modern advancement in metallurgy and thermal-barrier coating technologies, it may become possible to remove this sheath to recover the pressure loss without causing thermal damage to the combustor chamber and the transition piece walls. However, without the sheath, the flow inside the dump diffuser may exert nonuniformly reduced cooling on the combustion chamber and transition piece walls. The objective of this paper is to investigate the difference in flow pattern, pressure drop, and heat transfer distribution in the dump diffuser and over the outer surface of the combustor with and without a sheath. Both experimental and computational studies are performed and presented in Part 1 and Part 2, respectively. The experiments are conducted under low pressure and temperature laboratory conditions to provide a database to validate the computation model, which is then used to simulate the thermal-flow field surrounding the combustor and transition piece under elevated gas turbine operating conditions.

The experimental results show that the pressure loss between the dump diffuser inlet and exit is 1.15% of the total inlet pressure for the non-sheathed case and 1.9% for the sheathed case. This gives a 0.75 percentage point (or 40%) reduction in pressure losses. When the sheath is removed in the laboratory, the maximum increase of surface temperature is about 35%, with an average increase of 13%–22% based on the temperature scale of 23 K, which is the temperature difference of bulk inlet and outlet temperature.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A003. doi:10.1115/GT2013-94475.

The objective of Part 2 is to employ a computational scheme to investigate the difference in flow pattern, pressure drop, and heat transfer in a gas turbine’s dump diffuser and over the outer surface of the combustor with and without a sheath. Both experimental and computational studies are performed. In Part 1, the experiments are conducted under low pressure and temperature laboratory conditions to provide a database to validate the computation model, which is then used to simulate the thermal-flow field surrounding the combustor and transition piece under elevated gas turbine operating conditions.

For laboratory conditions, the CFD results show that (a) the predicted local static pressure values are higher than the experimental data but the prediction of the global total pressure losses matches the experimental data very well; (b) the total pressure losses are 1.19% for the no-sheath case and 1.89% for the sheathed case, which are within 3% of the experimental values; and (c) the temperature difference between the sheathed and non-sheathed cases is in the range of 5∼10K or 16%–32% based on the temperature scale between the highest and lowest temperature in the computational domain.

In summary, removing the sheath can harvest a significant pressure recovery of approximately 3% of the total pressure, but it will be subject to a wall temperature increase of about 500K (900°F or a 36% increase) on the outer radial part of the transition piece, where the flow is slow due to diffusion and recirculation in the large dump diffuser cavity near the turbine end. If modern advanced materials or coatings could sustain a wall temperature of about 200K higher than those currently available, the shield could be removed with the condition that a special cooling scheme (such as a water spray system) must be applied locally in this region. Otherwise, removal of the shield is not recommended.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A004. doi:10.1115/GT2013-94667.

State-of-the-art liner cooling technology for modern combustors is represented by effusion cooling (or full-coverage film cooling). Effusion is a very efficient cooling strategy based on the use of multi-perforated liners, where metal temperature is lowered by the combined protective effect of coolant film and heat removal through forced convection inside each hole. The aim of this experimental campaign is the evaluation of the thermal performance of multi-perforated liners with geometrical and fluid-dynamic parameters ranging among typical combustor engine values. Results were obtained as adiabatic film effectiveness following the mass transfer analogy by the use of Pressure Sensitive Paint, while local values of overall effectiveness were obtained by eight thermocouples housed in as many dead holes about 2 mm below the investigated surface. Concerning the tested geometries, different porosity levels were considered: such values were obtained both increasing the hole diameter and pattern spacing. Then the effect of hole inclination and aspect ratio pattern shape were tested to assess the impact of typical cooling system features. Seven multi perforated planar plates, reproducing the effusion arrays of real combustor liners, were tested imposing 6 blowing ratios in the range 0.5–5. Test samples were made of stainless steel (AISI304) in order to achieve Biot number similitude for overall effectiveness tests.

To extend the validity of the survey a correlative analysis was performed to point out, in an indirect way, the augmentation of hot side heat transfer coefficient due to effusion jets. Finally, to address the thermal behaviour of the different geometries in presence of gas side radiation, additional simulations were performed considering different levels of radiative heat flux.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A005. doi:10.1115/GT2013-94794.

Modern aviation combustors run at high fuel-air ratios to achieve high turbine inlet temperatures and higher turbine efficiencies. To maximize turbine durability in such extreme temperatures, the blades are fitted with film cooling schemes to form a layer of cool air between the blade and the hot core flow. Two terms that are utilized to evaluate a cooling scheme are the heat transfer coefficient (h) and the local driving temperature, namely, the adiabatic wall temperature (Taw). The literature presents a method for calculating these two parameters by assuming the heat flux (q) is proportional to the difference in freestream and wall temperatures (TTw). Several researchers have shown the viability of this approach by altering the wall temperature over a finite range in low temperature environment. A linear trend ensues where the slope is h and the q = 0 intercept is adiabatic wall temperature. This technique has proven valuable since constant h is known to be a valid assumption for constant property flow.

The current study explores the validity of this assumption by analytically predicting and experimentally measuring the h and q at high T and low Tw characteristic of a modern combustor. Both a reference temperature method and temperature ratio method were applied to model the effects of variable properties within the boundary layer. To explore the linearity of the heat transfer with driving temperature, the analysis determined the apparent h and Taw which would be measured over small ranges of Tw by the linear method discussed in the literature. This study shows that, over large Tw ranges, property variations play a significant role. It is also shown that the linear trend technique is valid even at high temperature conditions but only when used in small temperature ranges. Finally, this investigation shows that the apparent Taw used in the linear convective heat transfer assumption is a valid driving temperature over small ranges of Tw but cannot always be interpreted literally as the temperature where q(Taw) = 0.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A006. doi:10.1115/GT2013-94874.

State-of-the-art liner cooling technology for modern combustion chambers is represented by effusion cooling (or full-coverage film cooling). Effusion is a very efficient cooling strategy typically based on the use of several inclined small diameter cylindrical holes, where liner temperature is controlled by the combined protective effect of coolant film and heat removal through forced convection inside each hole. A CFD-based thermal analysis of such components implies a significant computational cost if the cooling holes are included in the simulations, therefore many efforts have been made to develop lower order approaches aiming at reducing the number of mesh elements. The simplest approach models the set of holes as a uniform coolant injection, but it does not allow an accurate assessment of the interaction between hot gas and coolant. Therefore higher order models have been developed, such as those based on localized mass sources in the region of hole discharge.

The model presented in this paper replaces the effusion hole with a mass sink on the cold side of the plate, a mass source on the hot side, whereas convective cooling within the perforation is accounted for with a heat sink. The innovative aspect of the work is represented by the automatic calculation of the mass flow through each hole, obtained by a run time estimation of isentropic mass flow with probe points, while the discharge coefficients are calculated at run time through an in-house developed correlation. In the same manner the heat sink is calculated from a Nusselt number correlation available in literature for short length holes. The methodology has been applied to experimental test cases of effusion cooling plates and compared to numerical results obtained through a CFD analysis including the cooling holes, showing a very good agreement. A comparison between numerical results and experimental data was performed on an actual combustor as well, in order to prove the feasibility of the procedure.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A007. doi:10.1115/GT2013-95039.

Effusion cooling technology has been assessed in past years as one of the most efficient methods to maintain allowable working temperature of combustor liners. Despite many efforts reported in literature to characterize the cooling performances of those devices, detailed analysis of the mixing process between coolant and hot gas are difficult to perform especially in case superposition and density ratio effects become important. Furthermore, recent investigations on the acoustic properties of these perforations pointed out the challenge to maintain optimal cooling performance also with orthogonal holes which showed higher sound absorption.

This paper performs a CFD analysis of the flow and thermal field associated with adiabatic wall conditions to compute the cooling effectiveness. The geometry consists of an effusion cooling plate drilled with 18 holes and fed separately with a cold and hot gas flow. Two types of perforations equivalent in porosity and pitches are investigated to assess the influence of the drilling angle between 30 and 90 deg. The reference conditions considered in this work comprehend an effective blowing ratio ranging between 1 and 3 at isothermal conditions (reaching a maximum hole Reynolds number of 10000) and high inlet turbulence intensity (17%). This set of conditions was exploited to perform a validation of the numerical procedure against detailed experimental data presented in another paper. Inlet turbulence effects highlighted by measurements for the slanted perforation were also investigated simulating a low turbulence condition corresponding to 1.6% of intensity. Furthermore the nominal DR = 1.0 was increased up to 1.7 to expand the available data set towards typical working conditions for aero-engines.

Steady state RANS calculations were performed with the commercial code ANSYS® CFX, modeling turbulence by means of the k — ω SST. In order to include anisotropic diffusion effects due to turbulence damping in the near wall region, the turbulence model is corrected considering a tensorial definition of the eddy viscosity with an algebraic correction to dope its stream-span components. Computational grids were finely clustered close to the main plate and inside the holes to obtain y+ < 1, to maximize solver accuracy according to previous similar analysis.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A008. doi:10.1115/GT2013-95235.

Steady-state Reynolds Averaged Navier-Stokes (RANS) equations are solved in the present numerical investigation to simulate the reactive two-phase flow in a model aero-engine combustor, and the reactive flow field with NOx emissions is analyzed. The gaseous phase is modeled by the modified SST turbulence model, and the liquid phase is modeled by Lagrangian tracking method considering the droplet breakup, collision and evaporation. Turbulence-combustion interaction is modeled by the extended coherent flame model, and NOx emissions are modeled by solving the species transport equation based on the assumption of frozen temperature. The fuel system of the present simulated combustor is radially staged, with a main stage employing the principle of lean prevaporized and premixed (LPP) concept to reduce pollutant emission, and a pilot stage burning a diffusion flame for flame stability. For the exit temperature quality improvement, dilution air is assigned with little amount of airflow. Detailed numerical results including exit temperature distribution, dominant burning performances and species distributions are evaluated for the combustion with and without dilution air. The influence of upstream burning characteristics to downstream temperature distribution is assessed. Numerical prediction of NOx emission demonstrates its capability of a reasonable reduction, and the exit temperature pattern with the dilution air is also able to fulfill its design target.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A009. doi:10.1115/GT2013-95495.

To accommodate the customer’s expectations for operational flexibility and low power generation costs, a gas turbine has to be robust, flexible and cost effective. Since its introduction in 1993 and with its more than 7.5 million operating hours and over 54’000 starts, the GT13E2 gas turbine has already demonstrated to be a most flexible and reliable engine. It is being used in connection with many different applications, and meets a very broad range of environment and operation conditions. The GT13E2 upgrade 2012 described in this paper further improves these capabilities.

The next generation of GT13E2 combustors is improved for increased lifetime, reduced total life cycle cost and implementation of a low emission dual fuel AEV burner system. The basic design philosophy for the lifetime improvement is adapted from the well-proven GT24 and GT26 annular combustors. The liner segments represent Alstom’s proven technology of sealed TBC coated metallic combustor liners that can expand in their fixations. The application of a thermal barrier coating onto the segments is simple and cost-effective. The design is robust so that the liners have to be checked only at major inspections and are not subject to reconditioning/replacement at hot gas part inspections. The closed-loop cooling arrangement is used for the backside cooling of the hot gas liner segments and to maintain the large structural components at a constant temperature. This combustion segment improvement is combined with the AEV (Advanced EnVironmental) burner. All the mentioned features result in a marked improvement of the operating and cyclic lifetime of the GT13E2 combustor.

This paper describes the development and validation process for the implementation of the combustion liner segment technology of the GT13E2. The various design phases from concept development to validation including the generic tests and final engine implementation are described and substantiated.

Commentary by Dr. Valentin Fuster
2013;():V03CT17A010. doi:10.1115/GT2013-95499.

Nonuniform combustor outlet flows have been demonstrated to have significant impact on the first and second stage turbine aerothermal performance. Rich-burn combustors, which generally have pronounced temperature profiles and weak swirl profiles, primarily affect the heat load in the vane but both the heat load and aerodynamics of the rotor. Lean burn combustors, in contrast, generally have a strong swirl profile which has an additional significant impact on the vane aerodynamics which should be accounted for in the design process. There has been a move towards lean burn combustor designs to reduce NOx emissions. There is also increasing interest in fully integrated design processes which consider the impact of the combustor flow on the design of the HP vane and rotor aerodynamics and cooling. There are a number of current large research projects in scaled (low temperature and pressure) turbine facilities which aim to provided validation data and physical understanding to support this design philosophy. There is a small body of literature devoted to rich burn combustor simulator design but no open literature on the topic of lean burn simulator design. The particular problem is that in non-reacting, highly swirling and diffusing flows, vortex instability in the form of a precessing vortex core or vortex breakdown is unlikely to be well matched to the reacting case. In reacting combustors the flow is stabilised by heat release, but in low temperature simulators other methods for stabilising the flow must be employed. Unsteady Reynolds-averaged Navier-Stokes and Large eddy simulation have shown promise in modelling swirling flows with unstable features. These design issues form the subject of this paper.

Commentary by Dr. Valentin Fuster

Structures (With Structures and Dynamics)

2013;():V03CT18A001. doi:10.1115/GT2013-94293.

To achieve high efficiency, modern gas turbines operate at temperatures that exceed melting points of metal alloys used in turbine hot gas path parts. Parts exposed to hot gas are actively cooled with a portion of the compressor discharge air (e.g., through film cooling) to keep the metal temperature at levels needed to meet durability requirements. However, to preserve efficiency, it is important to optimize the cooling system to use the least amount of cooling flow. In this study, film cooling optimization is achieved by varying cooling hole diameters, hole to hole spacing, and film row placements so that the specified targets for maximum metal temperature are met while preserving (or saving) cooling flow. The computational cost of the high-fidelity physics models, the large number of design variables, the large number and nonlinearity of responses impose severe challenges to numerical optimization. Design of experiments and cheap-to-evaluate approximations (radial basis functions) are used to alleviate the computational burden. Then, the goal attainment method is used for optimizing of film cooling configuration. The results for a turbine blade design show significant improvements in temperature distribution while maintaining/reducing the amount of used cooling flow.

Commentary by Dr. Valentin Fuster
2013;():V03CT18A002. doi:10.1115/GT2013-94306.

In the last five years Uncertainty Quantification (UQ) techniques became popular to predict gas turbine performances. Taking into account the uncertainties in the input parameters it is possible to evaluate the impact of random variations and to overcome the limitations of deterministic studies. These methods, that only recently have been widely used in computational fluid dynamics, have some limitations that must be considered.

One of the most important limitations is that these models cannot predict a “Black Swan” (BS) event. In probability a Black Swan is an event rare, possible and with serious consequences. A reliable design requires a correct evaluation of the probabilities of occurrence of the Black Swan that could strongly affect the life of the turbine. Black Swans are generated by the variability of the input parameters in the “tail” of the statistical distributions. Being far from the mean value design geometry/condition, these events have a low probability of occurrence. In this paper is shown that the use of the Gaussian distribution for the input parameters could strongly underestimate the probability of occurrence of a Black Swan event. Despite that most of the models used in UQ for aerodesign are neglecting the problem.

As an example of Black Swan, the hot gas ingestion across a stator is analysed. The gaps have been assumed to be affected by uncertainty with a variation of +/-50% of the nominal value. By using a Monte Carlo simulation with 108 realizations and a Gauss distribution as input, the configuration is initially considered reliable. The six sigma criterion is also satisfied and the probability to have a failure is only 2.54 10−4%. However if a “fat tail” for the input distribution is used instead, the probability to have hot gas ingestion becomes 2.33%, 104 times higher.

Most of the methods used in literature aim to have an accurate reproduction of the PDF moments such as mean, standard deviation, skew and kurtosis. However the “tail” of the distribution affects the gas turbine life and must be considered. In particular “fat tails”, the mathematical origin of Black Swans events, can have serious consequences, but in modern stochastic models used for computational fluid dynamics they are not accounted for.

Commentary by Dr. Valentin Fuster
2013;():V03CT18A003. doi:10.1115/GT2013-94845.

At preliminary design stages of the turbine discs design process, reducing uncertainty in the thermal prediction of critical parts models is decisive to bid a competitive technology in the aerospace industry. This paper describes a novel approach to develop adaptive thermal modeling methods for non-gaspath turbine components. The proposed techniques allow automated scaling of disc cavities during preliminary design assessment of turbine architectures.

The research undertaken in this work begins with an overview of the past investigations on the flow field in cavities of the air system surrounding the turbine discs. A theoretical approach is followed to identify the impact of the design geometry and operation parameters of a simplistic rotor-stator cavity, with special focus on swirl and windage effects. Then, a parametric CFD process is set up to conduct sensitivity analysis of the flow field properties. The CFD sensitivity analysis confirmed the parameter influences concluded from the theoretical study.

The findings from the CFD automated studies are used to enhance the boundary conditions of a thermal FE-model of an actual high pressure turbine. The new set of thermal boundary conditions adapts the flow field to changes in the cavity parameters. It was found that the deviation to experimental data of the traditional preliminary modeling technique is about 4 times higher as the deviation of the CFD-enhanced technique. When running the FE-model through a transient cycle, the results from the CFD-enhanced method are significantly closer to the test data than those from the traditional method, which suggests there is high potential for using these adaptive thermal techniques during turbine preliminary design stages.

Commentary by Dr. Valentin Fuster
2013;():V03CT18A004. doi:10.1115/GT2013-94972.

Gas turbine discs are classified as Critical Parts since, in the event of their primary failure, high energy debris can be released potentially resulting in hazardous consequences to the aircraft. Critical Parts are monitored during the life cycle of a gas turbine engine to ensure that integrity is established and maintained. The predicted safe cyclic life for an engine disc must be calculated as part of this monitoring process. For calculating the life of a turbine disc, a thorough understanding of material properties, operating conditions, metal temperatures and the resultant stress field is required. These inputs are obtained variously by component or whole engine testing or by predictive methods. These methods evolve over time, and materials may need to be changed, so for legacy engine designs, it’s important that the monitoring process recognises this and reacts appropriately. This paper describes the application of probabilistic methods to determine the uncertainty of turbine disc cyclic life for a two shaft low by-pass ratio gas turbine engine designed originally in the 1950s but predicted to be in service to beyond 2030. For the subject gas turbine the original material used to manufacture the turbine discs was declared technically and commercially obsolete. A new material was selected, requiring a new cyclic life to be determined. Rather than run an engine test to measure temperatures of the new discs, an analytical approach was adopted involving air system and thermal modelling and robust design techniques. These included Monte-Carlo analyses and the linking of thermal modelling and cyclic lifing codes using optimisation tools. It is shown how a probabilistic approach to air system and thermal modelling has enabled: (i) a quantitative judgment on the value of an air system survey (ii) the uncertainty of thermal predictions and the resultant variation in life to be quantified. These methods and results have then been used to release a safe cyclic life of a turbine disc for operation in an aircraft without the use of a dedicated thermocouple test.

Topics: Turbines , Disks , Uncertainty
Commentary by Dr. Valentin Fuster
2013;():V03CT18A005. doi:10.1115/GT2013-95708.

A computational methodology is proposed to predict the running clearance of a six-tooth straight-through rotating labyrinth seal numerically by taking into account both the centrifugal and thermal growths. Four different angular velocities ranging from 0 to 3000 rad/s are chosen to study the influence of rotation on the leakage flow rate. The detailed leakage flow fields and the structural deformations are presented. Further, different pressure ratios in the range of 1.1 to 2.5 have been investigated for a wide range of initial clearances. The methodology is validated against the available data in the literature. It is found out that there is a significant reduction in leakage flow rate by incorporating the radial growth for a particular operating condition. However, for a given initial clearance, the rotation has negligible effect on the reduction in the leakage flow rate, except at pressure ratios lower than 1.7. Further; the rotation has more prominent effect for smaller clearance values.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In