0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2015;():V05CT00A001. doi:10.1115/GT2015-NS5C.
FREE TO VIEW

This online compilation of papers from the ASME Turbo Expo 2015: Turbine Technical Conference and Exposition (GT2015) 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to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.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

Internal Air Systems and Seals (With Turbomachinery)

2015;():V05CT15A001. doi:10.1115/GT2015-42029.

Conduction in thin discs can be modelled using the fin equation, and there are analytical solutions of this equation for a circular disc with a constant heat-transfer coefficient. However, convection (particularly free convection) in rotating-disc systems is a conjugate problem: the heat transfer in the fluid and the solid are coupled, and the relative effects of conduction and convection are related to the Biot number, Bi, which in turn is related to the Nusselt number. In principle, if the radial distribution of the disc temperature is known then Bi can be determined numerically. But the determination of heat flux from temperature measurements is an example of an inverse problem where small uncertainties in the temperatures can create large uncertainties in the computed heat flux. In this paper, Bayesian statistics are applied to the inverse solution of the circular fin equation to produce reliable estimates of Bi for rotating discs, and numerical experiments using simulated noisy temperature measurements are used to demonstrate the effectiveness of the Bayesian method. Using published experimental temperature measurements, the method is also applied to the conjugate problem of buoyancy-induced flow in the cavity between corotating compressor discs.

Topics: Rotating Disks
Commentary by Dr. Valentin Fuster
2015;():V05CT15A002. doi:10.1115/GT2015-42044.

This paper reports on various effects on the flow through rotating radial holes (centrifugal, centripetal) in conjunction with the geometries of hole and surrounding annuli. The aerodynamic behavior of radial rotating holes is different from the one of axial and stationary holes due to the presence of centrifugal and Coriolis forces acting in the main flow direction. Furthermore, the geometry of the inlet and outlet region is often influencing the separation behavior of the flow at the holes. To investigate the flow phenomena and the discharge behavior of these radial holes in detail, an existing test rig containing two independently rotating shafts (co- and counter rotating) was used. Experimental and numerical investigations have been performed for both flow directions through the radial holes (centripetal and centrifugal), for different hole geometries (oblong holes and round holes), inlet types (rounded and sharp), length to diameter ratios (variation of either length or diameter) and gap widths between inner and outer shaft. For each of these geometrical variations flow properties have been varied such as pressure ratio across the holes, incident Mach number and rotational speed of both shafts. To enable large parametric studies and grid independency studies an optimization model with completely automatic grid generation, CFD simulation and post-processing has been set up. As a main result of the current studies it was found, that the shaft to hole diameter is another parameter of interest for the flow behavior through shaft holes. For a centripetal flow through the shaft holes and a decreasing inner gap width, the discharge coefficient was observed to increase initially before it drops significantly. In addition, measurements of centripetal flow though oblong holes revealed higher discharge coefficient in comparison with round holes and equal length to diameter ratio.

Topics: Rotation , Annulus , Geometry
Commentary by Dr. Valentin Fuster
2015;():V05CT15A003. doi:10.1115/GT2015-42106.

The design of the newest aircraft propulsion systems is focused on environmental impact reduction. Extensive research is being carried out with the purpose of improving engine efficiency, enhancing crucial features, in order to decrease both fuel consumption and pollutant emissions. A lot of improvements to fulfill these objectives must be made, focusing on the optimization of the main engine parts through the utilization of new technologies. The leakage flow reduction in the turbo machinery rotor-stator interaction is one of the main topics to which numerous efforts are being devoted.

Labyrinth seals, widely employed in the aerospace field thanks to their simple assembly process and maintenance, can be the means to achieve these objectives.

This paper mainly focuses on the optimization of the labyrinth seal stator part, characterized, in modern Low Pressure Turbines (LPT), by a honeycomb cell pattern.

The first phase of this study deals with the implementation and validation of a Computational Fluid Dynamics (CFD) numerical model, by using the experimental data available in the literature. Discharge coefficients obtained by numerical simulations, performed at different clearances and pressure ratios on both smooth and honeycomb non-rotating labyrinth seals, are presented and compared to the literature data.

Then, for both convergent and divergent flow conditions, the effects on the discharge coefficient due to variations in several cell pattern parameters (i.e. cell diameter, depth and wall thickness) and fin tip thickness are shown. For these analyses the values of clearance and pressure ratio are set at a constant value.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A004. doi:10.1115/GT2015-42324.

The ingress of hot gas through the rim seal of a gas turbine depends on the pressure difference between the mainstream flow in the turbine annulus and that in the wheel-space radially inward of the seal. This paper describes experimental measurements which quantify the effect of ingress on both the stator and rotor discs in a wheel-space pressurised by sealing flow. Infrared sensors were developed and calibrated to accurately measure the temperature history of the rotating disc surface during a transient experiment, leading to an adiabatic effectiveness. The performance of four generic (though engine-representative) single- and double-clearance seals was assessed in terms of the variation of adiabatic effectiveness with sealing flow rate. The measurements identify a so-called thermal buffering effect, where the boundary layer on the rotor protects the disc from the effects of ingress. It was shown that the effectiveness on the rotor was significantly higher than the equivalent stator effectiveness for all rim seals tested. Although the ingress through the rim seal is a consequence of an unsteady, three-dimensional flow field, and the cause-effect relationship between pressure and the sealing effectiveness is complex, the time-averaged experimental data is shown to be successfully predicted by relatively simple semi-empirical models, which are described in a separate paper. Of particular interest to the designer, significant ingress can enter the wheel-space before its effect is sensed by the rotor.

Topics: Turbines , Disks
Commentary by Dr. Valentin Fuster
2015;():V05CT15A005. doi:10.1115/GT2015-42326.

Sealing air is used in gas turbines to reduce the amount of hot gas that is ingested through the rim seals into the wheel-space between the turbine disc and its adjacent stationary casing. The sealing air attaches itself to the rotor, creating a buffering effect that reduces the amount of ingested fluid that can reach the surface of the rotor. In this paper, a theoretical model is developed, and this shows that the maximum buffering effect occurs at a critical flow rate of sealing air, the value of which depends on the seal geometry. The model, which requires two empirical constants, is validated using experimental data, obtained from infra-red (IR) temperature measurements, which are presented in a separate paper. There is good agreement between the adiabatic effectiveness of the rotor estimated from the model and that obtained from the IR measurements. Of particular interest to designers is that significant ingress can enter the wheel-space before its effect is sensed by the rotor.

Topics: Turbines , Disks
Commentary by Dr. Valentin Fuster
2015;():V05CT15A006. doi:10.1115/GT2015-42327.

Rim seals are fitted in gas turbines at the periphery of the wheel-space formed between rotor discs and their adjacent casings. These seals, also called platform overlap seals, reduce the ingress of hot gases which can limit the life of highly-stressed components in the engine. This paper describes the development of a new, patented rim-seal concept showing improved performance relative to a reference engine design, using URANS computations of a turbine stage at engine conditions. The CFD study was limited to a small number of purge-flow rates due to computational time and cost, and the computations were validated experimentally at a lower rotational Reynolds number and in conditions under incompressible flow. The new rim seal features a stator-side angel wing and two buffer cavities between outer and inner seals: the angel-wing promotes a counter-rotating vortex to reduce the effect of the ingress on the stator; the two buffer cavities are shown to attenuate the circumferential pressure asymmetries of the fluid ingested from the mainstream annulus. Rotor disc pumping is exploited to reduce the sealing flow rate required to prevent ingress, with the rotor boundary layer also providing protective cooling. Measurements of gas concentration and swirl ratio, determined from static and total pressure, were used to assess the performance of the new seal concept relative to a bench-mark generic seal. The radial variation of concentration through the seal was measured in the experiments and these data captured the improvements due to the intermediate buffer cavities predicted by the CFD. This successful design approach is a potent combination of insight provided by computation, and the flexibility and expedience provided by experiment.

Topics: Design , Turbines
Commentary by Dr. Valentin Fuster
2015;():V05CT15A007. doi:10.1115/GT2015-42408.

In an attempt to manage CFD computations in aero engine heat exchanger design, this work presents the best strategies and the methodology used to develop a holistic porosity model, describing the heat transfer and pressure drop behavior of a complex profiled tubular heat exchanger for aero engine applications. Due to the complexity of the profile tube heat exchanger geometry and the very large number of tubes, detailed CFD computations require very high CPU and memory resources. For this reason the complex heat exchanger geometry is replaced in the CFD computations by a simpler porous medium geometry with predefined pressure loss and heat transfer.

The present work presents a strategy for developing a holistic porosity model adapted for heat exchangers, which is capable to describe their macroscopic heat transfer and pressure loss average performance. For the derivation of the appropriate pressure loss and heat transfer correlations, CFD computations and experimental measurements are combined. The developed porosity model is taking into consideration both streams of the heat exchanger (hot and cold side) in order to accurately calculate the inner and outer pressure losses, in relation to the achieved heat transfer and in conjunction with the selected heat exchanger geometry, weight and operational parameters. For the same heat exchanger, RAM and CPU requirement reductions were demonstrated for a characteristic flow passage of the heat exchanger, as the porosity model required more than 80 times less computational points than the detailed CFD model. The proposed porosity model can be adapted for recuperation systems with varying heat exchanger designs having different core arrangements and tubes sizes and configurations, providing an efficient tool for the optimization of the heat exchangers design and leading to an increase of the overall aero engine performance.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A008. doi:10.1115/GT2015-42430.

This paper presents results from an extensive experimental study on the rubbing behavior of labyrinth seal fins and a honeycomb liner. The objective of the present work is to improve the understanding of the rub behavior of labyrinth seals by quantifying the effects and interactions of sliding speed, incursion rate, seal geometry and seal fin rub position on the honeycomb liner. In order to reduce the complexity of the friction system studied, this work focuses on the contact between a single seal fin and a single metal foil. The metal foil is positioned in parallel to the seal fin to represent contact between the seal fin and the honeycomb double foil section. A special test rig was set up enabling the radial incursion of a metal foil into a rotating labyrinth seal fin at a defined incursion rate of up to 0.65 mm/s and friction velocities up to 165 m/s. Contact forces, friction temperatures and wear were measured during or after the rub event. In total, 88 rub tests including several repetitions of each rub scenario have been conducted to obtain a solid data base.

The results show that rub forces are mainly a function of the rub parameters incursion rate and friction velocity. Overall, the results demonstrate a strong interaction between contact forces, friction temperature and wear behavior of the rub system. The presented tests confirm basic qualitative observations regarding blade rubbing provided in literature.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A009. doi:10.1115/GT2015-42550.

As a type of contacting seal technology, brush seals provide superior sealing performance and flexible behavior. Brush seals have found increasing application in more challenging high-temperature locations in recent years. Thus, the frictional heat generation between the seal bristles and mating surfaces is becoming another major concern for stable operation of brush seals. This study presents detailed investigations on the conjugate heat transfer behavior of brush seals using Computational Fluid Dynamics (CFD) and Finite Element Method (FEM) approaches. A dual-energy equation was proposed to describe the conjugate heat transfer in the porous bristle pack region under local thermal non-equilibrium conditions. The heat transfer CFD model was established with consideration of anisotropic thermal conductivity and a radius-dependent porosity of the bristle pack. The frictional heat generation was calculated from the product of the bristle-rotor frictional force and sliding velocity. The bristle-rotor frictional force was obtained from the brush seal FEM model with consideration of internal friction and aerodynamic load on the bristles. The temperature distribution of the brush seal was predicted at various operational conditions using the iterative CFD and FEM brush seal model. The effects of pressure ratios and rotational speeds on the temperature distribution and bristle maximum temperature of the brush seal were investigated based on the developed numerical approach. The effect of frictional heat generation on brush seal leakage was also analyzed.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A010. doi:10.1115/GT2015-42580.

Within this paper a continuation in brush seal testing for flexible load regimes in a steam turbine is given. Besides the well-known main design parameters of brush seals, e.g. the bristle pack thickness, the bristle diameter or the lay angle of the bristle pack, this paper focuses on the axial inclination of the bristle pack and particularly the affinity of bristle pack oscillations at low inclined bristle packs and small pressure differences. As it was presented in GT2014-26330, the axial inclination of the bristle pack is an important design parameter for brush seals. Along with a clearly increased blow-down capability and a reduced stiffness the seals tend to exhibit an enhanced axial bristle pack width during pressurization. It was previously shown that a low axial inclination of the bristle pack results in a loose package and in bristle pack oscillations until pressure differences of 10 bar. Above pressure drops of 10 bar the resulting higher abrasive behavior stops and a well sealing brush seal with a loose bristle pack is given. Regarding the renewable energy sources for necessary changes in steam turbine operations, a flexible sealing system with an enhanced wide operating range is requested. To capture all positive behaviors of low inclined brush seals for pressure differences until 10 bar, a design to safely avoid bristle pack oscillations is required. With this background low inclined brush seals with a new back plate design were tested at the Institute’s cold air test facility in Braunschweig up to a pressure difference of 4 bar. The facility allows detailed sealing performance investigations including real time bristle pack observations. The present paper shows and discusses overall experimental results of brush seals with different axial inclinations mounted with an adjustable back plate to determine the influence of the back plate design on the bristle pack oscillations. Furthermore, these new results together with older measurements from 2012 were used to develop a theory regarding the changes that result from contact between the bristle pack and the adjusted back plate. Finally, the design for a pressure balanced back plate will be shown.

Topics: Pressure
Commentary by Dr. Valentin Fuster
2015;():V05CT15A011. doi:10.1115/GT2015-42609.

Flow and heat transfer in an aero-engine compressor disc cavity with radial inflow has been studied using computational fluid dynamics (CFD), large eddy simulation (LES) and coupled fluid/solid modelling. Standalone CFD investigations were conducted using a set of popular turbulence models along with 0.2° axisymmetric and a 22.5° discrete sector CFD models. The overall agreement between the CFD predictions is good, and solutions are comparable to an established integral method solution in the major part of the cavity. The LES simulation demonstrates that flow unsteadiness in the cavity due to the unstable thermal stratification is largely suppressed by the radial inflow. Steady flow CFD modelling using the axisymmetric sector model and the Spalart-Allmaras turbulence model was coupled with a finite element (FE) thermal model of the rotating cavity. Good agreement was obtained between the coupled solution and rig test data in terms of metal temperature. Analysis confirms that use of a small radial bleed flow in compressor cavities is effective in reducing thermal response times for the compressor discs and that this could be applied in management of compressor blade clearance.

Topics: Modeling , Cavities , Inflow
Commentary by Dr. Valentin Fuster
2015;():V05CT15A012. doi:10.1115/GT2015-42624.

This paper describes the design and testing of a fluidic “air-curtain” type seal application to reduce tip leakages on a small high-speed single stage axial turbine.

The initial experimental investigations were carried out to demonstrate the application of the fluidic type seal on a “frozen rotor” test of a turbine. The “frozen rotor” test is carried out using an actual turbine rotor in a static test facility without rotation. These preliminary tests provide the first experimental validation of fluidic “air-curtain” type seals working to reduce over-tip leakage in a turbine shroud on a actual turbine geometry.

Ultimately the final stage of the work will be to demonstrate the fluidic seal working in a full rotating facility but these results provide a logical and important step towards that ultimate goal.

Topics: Rotors , Turbines
Commentary by Dr. Valentin Fuster
2015;():V05CT15A013. doi:10.1115/GT2015-42852.

In order to ensure safety, predictable and acceptable life of gas turbine engines an important task is the design of the secondary air system (SAS).

To avoid hot gas path ingestion, one of the most critical concerns in a well-performing SAS is the understanding of the air motion within of the stator-rotor cavity systems. The state of the art of the fluid solver tools used to predict the flow behaviour inside cavities is generally based on codes with correlative approaches, in order to reduce calculation times and computational resources.

In order to improve the precision of correlations, by means of data with reduced associated uncertainty, experimentalists performed an experimental campaign selecting a simple test case composed by a rotating disc facing a flat stator. Imposing several working conditions, the rotational speed was varied in order to gain a rotational Reynolds number up to 1.2 × 106 which is one order of magnitude less than real seals working conditions. The analysis of the swirl rate as well as the frictional moment exerted by the rotor on the fluid was carried out based on different sealing mass flows. Pressure probes were used to obtain radial distribution of static and total pressure on the stator side. In order to deepen the analysis on the flow behaviour inside cavity, flow field measurements using a 2D PIV system was performed on two different planes: velocity contours were used to help the detection of the core region.

Finally, the wide database of the experimental results were used to improve a simple model able to predict the behaviour of rotor-stator cavities including the effect of the inlet geometry of the cavity.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A014. doi:10.1115/GT2015-42944.

The hot gas ingestion through rim seals of counter-rotating turbine disks with annulus flow is investigated by numerical simulation in this paper. Three models were calculated in this paper, counter-rotating turbine cavity with annulus and radial sealing inlet, the same geometric configuration but in rotor-stator system and the counter-rotating turbine cavity without annulus. The axial Reynolds number of annulus was set at 5.99×105 and rotating Reynolds number was set at 2.35×105, the non-dimensional sealing flow rate was changed from 279 to 1116. The results of simulation indicated that counter-rotating cavity with radial inlet was found to have a much higher sealing efficiency than the stator-rotor cavity at the equivalent condition. The flow path of ingestion gas was considered to be the cause for this. For the difference of flow structure of cavities, the annulus air is ingested the middle of cavity for counter-rotating cavity rather than flowed towards the static wall for rotor-stator cavity. Therefore, the sealing efficiency for counter-rotating cavity is high in both of the walls but low inside the cavity. In addition, preswirl of sealing air was found to affect the sealing efficiency by changing the flow inside cavity and making the early mixing of sealing air and ingestion gas.

Topics: Turbines , Annulus , Cavities
Commentary by Dr. Valentin Fuster
2015;():V05CT15A015. doi:10.1115/GT2015-42946.

The heat transfer characteristics of counter-rotating disks with annulus flow were investigated numerically in this paper, in which the effects of coolant inlet conditions (inlet geometry, turbulent parameter and inlet preswirl) of the cavity were emphasized. The axial Reynolds number of annulus Rew was set to be 5.99×105 and rotating Reynolds number Reϕ was set to be 2.35×105. Two kinds of cooling air inlet, radial inlet from the middle of cavity and axial inlet from one side of cavity, were adopted and investigated. The turbulent parameter and preswil ratio varied from 0.028 to 0.197 and −1.5 to 1.5, respectively. According to the calculation results:a vortex pair that generated by the radial inlet and non-preswirling cooling air is somehow unstable which rendered a complex flow field and wall heat transfer pattern in the cavity. For the influence of pumping effect of rotating disks, preswirling of cooling air led to uneven division of cooling air for two counter-rotating disks in radial cooling air inlet type. More cooling air flows towards the disk with same rotating direction of preswirl cooling air and leads to higher Nu in that wall which was enhanced by the increase of preswirl ratio. Axial cooling air inlet results in a much more stable flow than that with radial inlet. The Nu of the wall that cooling air flow towards was higher than that of the other one and this difference increased with the increase of turbulent parameter λT.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A016. doi:10.1115/GT2015-43117.

Previous studies on feed pump performance conducted on the ULB-ATM lubrication system test bench showed the dependence of cavitation and volumetric efficiency with inlet pressure, rotational speed and aeration. This paper presents a technique of aeration measurement applied on the test bench. After a description of the device and a theoretical review of the aeration, the paper shows that the method has been tested with success. The paper illustrates it with a series of tests showing the level of aeration obtained in different operating conditions and a comparison with a pump characteristic measured on the bench.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A017. doi:10.1115/GT2015-43184.

During the last decades a large effort has been made to continuously improve turbomachine efficiency. Besides the optimization of the primary flow path, also the secondary flow losses have been reduced considerably, due to the use of more efficient seals. Brush seals, as a compliant contacting filament seal, have become an attractive alternative to conventional labyrinth seals in the field of aircraft engines as well as in stationary gas and steam turbines. The aim of today’s research related to brush seals is to understand the characteristics and their connections, in order to be able to make performance predictions, and to ensure the reliability over a defined operating period. It is known that inevitable frictional contacts lead to an abrasive wear on the rotor side as well as on the bristle side. The wear situation is essentially influenced by the resulting contact force at the seal-to-rotor interface during the operating time. This contact force depends on the seal’s blow down capability, which is mainly determined by the geometrical design of the bristle pack, e.g. the axial inclination of the investigated seal design, in combination with the design and material of the surrounding parts, as well as the thermal boundary conditions. For realistic investigations with representative circumferential velocities the TU Braunschweig operates a specially developed steam test rig which enables live steam investigations under varying operating conditions up to 50 bar and 450 °C. Wear measurements and the determination of seal performance characteristics, such as blow down and bristle stiffness, were enabled by an additional test facility using pressurized cold air up to 8 bar as working fluid.

This paper presents the chronological wear development on both rotor and seal side, in a steam test lasting 25 days respectively 11 days. Interruptions after stationary and transient intervals were made in order to investigate the wear situation. Two different seal arrangements, a single tandem seal and a two-stage single seal arrangement, using different seal elements were considered. The results clearly show a continuous wear development and that the abrasive wear of the brush seal and rotor is mainly due to the transient test operation, particularly by enforced contacts during shaft excursions. Despite the increasing wear to the brushes, all seals have shown a functioning radial-adaptive behavior over the whole test duration with a sustained seal performance. Thereby, it could be shown that the two-stage arrangement displays a load shift during transients, leading to a balanced loading and unloading status for the two single brush seals. From load sharing and in comparison with the wear data of the tandem seal arrangement, it can be derived that the two-stage seal is less prone to wear. However, the tandem seal arrangement, bearing the higher pressure difference within one configuration, shows a superior sealing performance under constant load, i.e. under stationary conditions.

Topics: Wear , Steam
Commentary by Dr. Valentin Fuster
2015;():V05CT15A018. doi:10.1115/GT2015-43231.

The application of compliant filament seals to jet engine secondary air systems has been shown to yield significant improvements in specific fuel consumption and improved emissions. One such technology, the leaf seal, provides comparable leakage performance to the brush seal but offers higher axial rigidity, significantly reduced radial stiffness and improved compliance with the rotor. Investigations were carried out on the Engine Seal Test Facility at the University of Oxford into the behavior of a leaf seal prototype at high running speeds. The effects of pressure, speed and cover plate geometry on leakage and torque are quantified. Early publications on leaf seals showed that air-riding at the contact interface might be achieved. Results are presented which appear to confirm that air-riding is taking place. Consideration is given to a possible mechanism for torque reduction at high rotational speeds.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A019. doi:10.1115/GT2015-43238.

Industrial and aeroderivative gas turbines use exhaust systems for flow diffusion and pressure recovery. The downstream balance-of-plant systems such as heat recovery steam generators or selective catalytic systems require, in general, a steady, uniform flow out of the exhaust system. One detrimental effect of having these downstream systems is the increased back pressure. These combined-cycle systems increase the back pressure on the free power turbine which results in decreased power output and efficiency.

Aeroderivative gas turbines for mechanical drive application have a wide operational envelope. In general, at baseload, the exhaust back pressure ranges from 1.5 to 2.5 kPa above ambient pressure. Increased exhaust back pressure results in changes to power turbine secondary flows by changing the cavity flow dynamics, sealing flows, and rim seal ingestion. This impacts the thermal characteristics of turbine rotor discs and their lives.

The primary motivation for this research and development work was to develop solution for secondary air system and investigate the impact of high exhaust back pressure on power turbine disc thermals. At first, 1-D system-level power turbine secondary flow analyses were carried out with normal back pressure (3.0 kPa) and with high back pressure (11.37 kPa). In addition, 3-D computational fluid dynamic simulations were performed to understand the cavity flow dynamics and disc heat transfer coefficient variations. These results were used in a high-fidelity 2-D thermal modeling of the power turbine to study the impact of back pressure on turbine disc thermal characteristics and their lives.

The fluid and thermal predictions were validated using normal back pressure full-scale full-load test results. Cooling mass flow rate, static pressure, air temperature, and metal temperature predictions are compared with test results over a wide operating range. The numerical predictions are in good agreement with test results.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A020. doi:10.1115/GT2015-43293.

This paper presents an investigation to improve the design of a generic fin heat exchanger, using novel discrete adjoint solver tools in the computational fluid dynamics (CFD) software FLUENT. A baseline design is analyzed initially to evaluate flow resistance and heat transfer. Optimization is conducted by deploying the adjoint solver. The heat load and the drag force are combined into an objective function using a Reynolds analogy approach. Sensitivities of the objective function to geometric changes are predicted by the adjoint, and then the mesh is morphed, and the predictions are verified by the full CFD solutions. Predetermined, engineering driven, geometric changes are explored and compared, and the range of validity of the predictions is evaluated. An algorithm is then developed to implement steepest descent, constrained optimization based on the adjoint solution. The algorithm is applied iteratively on the fin heat exchanger, and a comparison is performed between the change in objective function predicted by the adjoint, and that calculated in full CFD solutions on morphed meshes. The insight gained on the directions of design changes and attending quantitative improvement of the design objective function is very useful to guide the optimization process. This is enabled by the adjoint solver’s capability to robustly evaluate the sensitivities of the objective function to all solution variables, and predict changes in observables.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A021. doi:10.1115/GT2015-43378.

This paper describes the development of laminated seals for stator-stator sealing in gas turbines. Cloth seals were introduced as stator-stator seals in the 1990’s and resulted in a significant improvement in sealing performance over the rigid seals then in service. Laminated seals presented here are proposed as an improvement to the existing cloth seals. They demonstrate improved leakage performance over cloth seals in aligned and offset conditions. Several versions of laminated seals were developed and tested before arriving at a seal geometry that satisfied the leakage, manufacturability and assembly requirements. These seals therefore provide an alternative to cloth seals that leak less and are durable, cost-effective and robust in assembly.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A022. doi:10.1115/GT2015-43406.

The paper presents a computational procedure for heterogeneous coupled analysis of 1D flow network models of air engine secondary air systems and 2D/3D solid thermo-mechanical finite element models of engine components. We solve an unsteady heat transfer problem over solid domain coupled to a sequence of structural static and steady flow problems using a quasi-steady state approximation. Strong coupling is achieved at each time step by a fixed-point iteration, based on the successive solution of the fluid and the solid sub-problems. The procedure is applied to a 2D axisymmetric finite element model of an intermediate pressure turbine assembly coupled to a flow network model of whole engine secondary air system simulated through a square cycle. The simulation results are compared to reference stand-alone predictions showing important non-negligible coupled effects and component interactions of a multidisciplinary multi-physical nature resolved in an efficient and automatic fashion.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A023. doi:10.1115/GT2015-43465.

Advanced contacting seals, such as leaf seals or brush seals, can offer reduced leakage during engine operation when compared to conventional labyrinth seals. The flexible elements of these seals provide better compliance with the rotor during flight manoeuvres. The functionality and performance retention attributes of an engine-scale prototype leaf seal have been investigated on a seal test facility at Rolls-Royce that achieves engine-representative pressures and speeds and allows dynamic control of the seal position relative to the rotor, both concentric and eccentric. In this paper the experimental setup and the test method are described in detail, including the quantification of the measurement uncertainty developed to ASME standard PTC 19.1. Experimental data are presented that show the variations in leakage and torque over typical variations of the test parameters. Insight is gained into the interactions between the operating pressure and speed and the concentric and eccentric movements imposed on the seal.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A024. doi:10.1115/GT2015-43496.

The bearing chamber of an aeroengine houses roller bearings and other structural parts. The spatial limitation, high operational speeds of the HP shaft and the proximity to the combustion chamber can make the operating conditions of the bearing chamber challenging. A roller bearing consists of an inner race, an outer race and a cage constraining a number of rolling elements. In the aeroengine application, oil is introduced into the bearing chamber via the inner race regions of the bearing into the rolling elements interstices. This provides lubrication for the roller bearings. The source of heat in the bearing chamber is mainly from rolling contact friction and the high temperature of combustion. Cooling results from the oil transport within the bearing chamber and thus an efficient transport of oil is critical to maintaining the integrity of the entire structure. The bearing chamber contains the oil which is eventually scavenged and recycled for recirculation. Experiments have been conducted over the years on bearing chamber flows but often simplified to create the best emulation of the real aeroengine. The complexity of the bearing chamber structure is also challenging for experimental measurements of the oil characteristic in the roller bearing elements and the bearing chamber compartment. Previous experiments have shown that the oil continuum breaks up in the bearing chamber compartment but it is not quantitatively clear how and what parameters affect these. Previous simulation attempt of bearing chamber, also, have been limited by the boundary conditions for the oil. This work presents a computational fluid dynamics (CFD) transient simulation of flow in the bearing sector in an attempt create boundary conditions for such models. The current results show that the oil emerges in the form of droplets into the bearing chamber compartment with speed of the order of 10% of the shaft rotation.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A025. doi:10.1115/GT2015-43501.

The main objective of the present work was to develop a porous media based axi-symmetric bolt protrusion drag and heat transfer model for fast and efficient aero-thermal coupling of a rotor-stator cavity with rotor mounted bolts. Traditionally, detailed non axisymmetric features like bolts, holes etc. are either approximated through windage correlations or ignored which could result in significant difference in disc temperature prediction. Protrusion drag and windage work terms are introduced into the bolt porous zone in an axisymmetric model to simulate non axisymmetric effects of protrusions. The drag, tangential velocity and adiabatic disc surface temperature results from the simplified axisymmetric model are compared with 3D CFD model predictions and experimental data for a range of rotational and throughflow Reynolds numbers. The simplified porous media based models are found to predict the drag and windage heat transfer with reasonable accuracy compared to 3D sector CFD results. However, 3D sector CFD under-predicts the high core flow swirl and mixing of the hot fluid withinthe rotor-stator cavity and also under-predicts the adiabatic disc surface temperature inboard of the bolt, compared to experimental data, particularly for the rotationally dominated flow case. The comparison of disc temperature with measurements for the through flow dominated case of λT=0.2 and Cw = 105 is satisfactory.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A026. doi:10.1115/GT2015-43503.

This paper explores the role of surface tension, grid resolution, inertia term representation and temporal discretisation scheme in the numerical simulation of shear-driven thin-film rimming flows. An ideal film formulation and solution strategy, suitable for the simulation of smooth, shock and pool solutions is presented. Shock and pool solution stability is shown to be dependent on the provision of sufficient grid refinement to resolve key flow features present in the solution such as steep fronts and small wavelength capillary waves. A minimum grid refinement criterion is proposed based on the findings from a parametric study. A previously established dependence of solution stability on surface tension is shown to be linked to the sensitivity of the wavelengths of disturbances in the capillary zone on the surface tension coefficient. Solution strategies utilising un-physically high surface tension values to guarantee stability, are explored and shown to enhance stability by modifying the solution in the capillary zone to one that is resolvable on the available grid. The role of inertia in solution stability is also investigated and simplified inertia representations are shown to primarily affect accuracy but not stability.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A027. doi:10.1115/GT2015-43506.

An isothermal thin-film flow over a rotating plate has been simulated using the depth-averaged Eulerian Thin-Film modelling (ETFM) approach. The model setup is based on published experimental and numerical Volume of Fluid (VOF) CFD studies of the same problem to allow for model validation. A range of controlled film inlet heights and mass flow rates are explored together with varied plate rotational speeds ranging from a stationary plate (50rpm) to 200 rpm. While the VOF model has previously been shown to accurately reproduce film thickness, the Eulerian thin-film model is shown to provide predictions of comparable accuracy at a much lower computational cost. The model is also shown to be able to reproduce the film solution’s sensitivity to variations in fluid properties due to changes in inlet temperature. A full 3D domain has been used in this study and the ETFM model is also shown to be able to reproduce azimuthal film thickness variations and surface features similar to those previously observed in experiments.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A028. doi:10.1115/GT2015-43509.

Oil-air separation is a key function in aero engines with closed-loop oil systems. Typically, aero engine air/oil separators employ the use of a porous medium such as open cell metal foams, as a secondary separation mechanism. Assessing its impact on overall separation is important since non-captured oil is released overboard. Computational fluid dynamics offers a possibility to evaluate the metal foam separation effectiveness.

A pore scale numerical modelling methodology is applied to determine the transport properties of fluid flow through open cell metal foams. Microcomputer tomography scans were used to generate a 3D digital representation of commercial open cell metal foams of different grades. Foam structural properties such as porosity, specific surface, pore size distribution and the minimum size of a representative elementary volume are directly extracted from the CT scans. Subsequently, conventional finite volume simulations are carried out on the realistic tomography-based foam samples. Simulations were performed for a wide range of Reynolds numbers. The feasibility of using standard Reynolds-averaged Navier-Stokes (RANS) turbulence models is investigated here. As part of the method validation, samples with varying lengths were simulated. Pressure drop values were compared on a length-normalized basis against in-house experimental data.

The oil phase was modelled using a Lagrangian particle tracking approach. Boundary conditions for the oil phase were extracted from a previous CFD simulation of a full breather device in the ground idle regime (worst separation effectiveness). Steady state particle tracking simulations were run for droplet diameters ranging from 0.5–15 μm, and for flow inlet velocities ranging from 10–60 m/s. Stochastic tracking was taken into account in order to model the effects of turbulence on the particle trajectories. Simulations were run on different types of foam and the results are compared qualitatively. The procedure has shown that pore scale modelling is a valid tool to capture the flow field and model oil separation inside open cell metal foams. However, at the moment there is no experimental data available for validation of the oil phase modelling.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A029. doi:10.1115/GT2015-43517.

Rim seals in the turbine section of gas turbine engines aim to reduce the amount of purge air required to prevent the ingress of hot mainstream gas into the under-platform space. A stationary, linear cascade was designed, built, and benchmarked to study the effect of the interaction between the pressure fields from an upstream vane row and downstream blade row on hot gas ingress for engine-realistic rim seal geometries. The pressure field of the downstream blade row was modeled using a bluff body designed to produce the pressure distortion of a moving blade. Sealing effectiveness data for the baseline seal indicated that there was little to no ingress with a purge rate greater than 1% of the main gas path flow. Adiabatic endwall effectiveness data downstream in the trench between the vane and blade showed a high degree of mixing. Extending the seal feature associated with the vane endwall indicated better sealing than the baseline design. Steady computational predictions were found to overpredict the sealing effectiveness due to underpredicted mixing in the trench.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A030. doi:10.1115/GT2015-43541.

The efficiency of power transmission systems is increasingly targeted with a view to reducing parasitic losses and improving specific fuel consumption (SFC). One of the effects associated with such parasitic losses is gear windage power loss and this mechanism can be a significant contributor to overall heat-to-oil within large civil aeroengines. The University of Nottingham Technology Centre in Gas Turbine Transmission Systems has been conducting experimental and computational research into spiral bevel gear windage applicable to an aeroengine internal gearbox (IGB).

The two-phase flows related to gear lubrication, shrouding and scavenging are complex. Good understanding of such flows can be used to balance lubrication needs with need to minimise oil volumes and parasitic losses. Previous computational investigations have primarily employed discrete phase modelling (DPM) to predict oil behaviour under the shroud [1, 2]. In this paper modelling capability has been investigated and extended through application of FLUENT’s Eulerian multiphase model. In addition, DPM modelling linked to FLUENT’s Lagrangian film model has been conducted. A control volume with periodic symmetry comprising a single tooth passage of the bevel gear has been modelled to keep the computational cost down.The results from both models are compared to each other and to available experimental visual data.

Both models are found to perform acceptably with the Eulerian multiphase model yielding results closer to those observed experimentally. The use of DPM with a Eulerian film model is suggested for future work and extension to a full 360° model is recommended.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A031. doi:10.1115/GT2015-43549.

In many aeroengines the accessory power offtake is achieved using a spiral bevel gear set running off one of the main shafts. The crown and bevel gears are housed in an internal gearbox. Over the past few years the Nottingham University Technology Centre (UTC) in Gas Turbine Transmission Systems has researched flow near spiral bevel gears both computationally and experimentally using a purpose-built test rig. In the current investigation the rig was configured with a Trent crown gear and slightly modified shroud covering the full 360° of the gear. No external containment chamber was fitted and all testing was conducted single-phase (air only) at 5,000 rpm. Laser Doppler Anemometry (LDA) was used to obtain the three components of flow velocity at a shroud exit slot and at shroud inlet. A 2D system was utilised and thus two measurements were required at each point to give the 3 velocity components.

The LDA technique enabled detailed mapping of flow features over the chosen regions, which included areas very near the shroud surfaces. Data was obtained over two measurement regions:

1) a volume mapping the air “jet” exiting the shroud exit slot at top dead centre (TDC) and

2) an area capturing the flow structures local to the shroud inlet.

Combined the results form an excellent set of high quality, detailed, 3-component flow data for direct use in validating CFD models and/or to define CFD boundary conditions.

At the shroud exit slot the maximum velocity measured was 46.2 m/s with the jet velocity dispersing over the measurement volume such that by 26 mm from slot plane the maximum velocity was less than 20 m/s. The jet angle was found to be only 16° off perpendicular azimuthally and 22° down from perpendicular. Data from the top 5 slots shows good similarity indicating the detailed data for the TDC slot is probably applicable to all slots.

Air entering the shroud comes down the shroud face and up the rotating end face of the gear shaft. The azimuthal velocity component at shroud inlet was around 20 m/s; this is of the order of 50% of the maximum linear shaft surface speed. Within 3 mm of the rotating gear face the azimuthal velocity is less than 1 m/s. Detailed measurements were obtained only at one angular location but sufficient additional measurements were obtained to determine that for the purposes of CFD validation the results can be considered representative.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A032. doi:10.1115/GT2015-43561.

This paper presents a computational investigation into the effects of inlet swirl on the fluid flows computed for a spiral bevel gear and shroud pairing. This is with a specific focus on aeroengine internal gearboxes where the power offtake gear is located in a chamber where a highly swirled environment exists.

Previous work for a single rotating gear and stationary shroud [1] has been performed in isolation and with idealised boundary conditions at both the inlet and outlet of the system. This effectively decouples the shrouded gear subsystem from the rest of the gearbox. In the present study a parametric investigation has been conducted varying the amount of inlet swirl and the effect this has on the mass flowrate of air “pumped” through the gear. The paper presents data showing that compared to no inlet swirl, a higher mass flowrate is induced when there is swirl, with up to 20% higher mass flowrate occurring when the tangential velocity at shroud inlet is 50% of the axial velocity component. For swirl above 50% the mass flowrate drops back somewhat remaining higher than for the no-swirl case for values investigated. Gear windage power loss is a function of mass flowrate and consequently higher windage losses occur for higher mass flowrates.

In an aeroengine flow through the gear exits through slots in the shroud and this paper shows that the swirl velocity component at the shroud exit holes is relatively insensitive to the velocity components entering the gear/shroud system for a given geometry.

Further to this, the effect of the shroud outlet geometry on the flow leaving the back of the gear has been investigated and quantified. It is shown that accurate geometric representation is required as the outlet geometry has a significant effect on the computed mass flow through the gear-shroud system and consequently on the computed windage power loss.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A033. doi:10.1115/GT2015-43601.

Fluidic devices are of interest with turbomachinery internal air systems for modulation of cooling air and other applications. Generally, the flow states within a fluidic device are switched by control flow or flows. For most fluidic devices the switching procedure is almost instantaneous and hence it is difficult to characterize the performance of a device experimentally. The objective of this research is to numerically investigate the dynamic characteristics of a control flow operated fluidic device. In this study the dynamic characteristics of a nozzle during switching is considered. The simulations considered the unsteady interaction of the control flow with the nozzle jet for two different switching scenarios namely, switching of high to low flow state and vice versa. The magnitude of static pressure applied at the control port was identified as a controlling parameter and had to be below a critical value to achieve stable switching. The CFD solutions show that this is related to the flow physics and critical momentum flux ratios for switching are calculated for the present device.

Topics: Fluidic devices
Commentary by Dr. Valentin Fuster
2015;():V05CT15A034. doi:10.1115/GT2015-43607.

Conventional labyrinth seal applications in turbomachinery encounter a permanent teeth tip damage and wear during transitional operations. This is the dominant issue that causes unpredictable seal leakage performance degradation. Since the gap between the rotor and the stator changes depending on engine transitional operations, labyrinth teeth located on the rotor/stator wear against the stator/rotor. This wear is observed mostly in the form of the labyrinth teeth becoming a mushroom shape. It is known that as a result of this tooth tip wear, leakage performance permanently decreases, which negatively affects the engine’s overall efficiency. However, very limited information about leakage performance degradation caused by mushroom wear is available in open literature.

This paper presents a study that numerically quantifies leakage values for various radii of mushroom shaped labyrinth teeth by changing tooth-surface clearance, pressure ratio, number of teeth, and rotor speed. Analyzed parameters and their ranges are mushroom radius (R=0–0.508mm), clearance (cr=0.254–2.032mm), pressure ratio (Rp=1.5–3.5), number of teeth (nt=1–12), and rotor speed (n=0–80krpm).

CFD analyses were carried out by employing compressible turbulent flow in 2-D axi-symmetrical coordinate system. CFD leakage results were also compared with well-known labyrinth seal semi-empirical correlations.

Given a constant clearance, leakage increases with the size of the mushroom radius that forms on the tooth. This behavior is caused by less flow separation and flow disturbance and the vena contracta effect for flow over the smoothly shaped mushroom tooth tip compared to the sharp-edged tooth tip. This leakage increase is higher when the tooth tip wear is considered as an addition to the unworn physical clearance, since the clearance dominates the leakage.

The leakage affected by the number of teeth was also quantified with respect to the mushroom radius. The rotational effect was also studied as a secondary parameter.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A035. doi:10.1115/GT2015-43638.

Brush seals require custom design and tailoring due to their behavior driven by flow dynamic, which has many interacting design parameters, as well as their location in challenging regions of turbomachinery. Therefore, brush seal technology has not reached a conventional level across the board standard. However, brush seal geometry generally has a somewhat consistent form. Since this consistent form does exist, knowledge of the leakage performance of brush seals depending on specific geometric dimensions and operating conditions is critical and predictable information in the design phase. However, even though there are common facts for some geometric dimensions available to designers, open literature has inadequate quantified information about the effect of brush seal geometric dimensions on leakage.

This paper presents a detailed CFD investigation quantifying the leakage values for some geometric variables of common brush seal forms functioning in some operating conditions. Analyzed parameters are grouped as follows; axial dimensions, radial dimensions and operating conditions.

The axial dimensions and their ranges are front plate thickness (z1=0.040–0.150in.), distance between front plate and bristle pack (z2=0.010–0.050in.), bristle pack thickness (z3=0.020–0.100in.), and backing plate thickness (z4=0.040–0.150in.).

The radial dimensions are backing plate fence height (r1=0.020–0.100in.), front plate fence height (r2=0.060–0.400in.), and bristle free height (r3=0.300–0.500in.).

The operating conditions are chosen as clearance (r0=0.000–0.020in.), pressure ratio (Rp=1.5–3.5), and rotor speed (n=0–40krpm).

CFD analysis was carried out by employing compressible turbulent flow in 2-D axi-symmetric coordinate system. The bristle pack was treated as a porous medium for which flow resistance coefficients were calibrated by using literature based test data.

Selected dimensional and operational parameters for a common brush seal form were investigated, and their effects on leakage performance were quantified. CFD results show that, in terms of leakage, the dominant geometric dimensions were found to be the bristle pack thickness and the backing plate fence height. It is also clear that physical clearance dominates leakage performance, when compared to the effects of other geometric dimensions. The effects of other parameters on brush seal leakage were also analyzed in a comparative manner.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A036. doi:10.1115/GT2015-43895.

A modification of a conventional straight four-tooth labyrinth seal with inclined teeth including single and double stepped notches is proposed. The variants are to be numerically modeled and evaluated for potential leakage reduction through the seal as a result of the developing flow field. The CFD methodology for numerical investigation is first validated by comparison with literature data from static labyrinth seal experiments. The objective is to numerically analyze the proposed “Notched Inclined Teeth” with solid and honeycomb lands and verify the leakages. Another objective is to compare these leakages with the “Stepped Double Notched Straight Tooth” variant discussed by the present authors in a previous paper.

Results indicate that the proposed modifications — single and double stepped notched inclined teeth, systematically reduce the seal leakage compared to the baseline straight and forward inclined teeth due to higher turbulence, higher blockages by introducing vortex in leakage flow through step and cavities, and higher flow resistance as compared to baseline model.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A037. doi:10.1115/GT2015-44067.

Bristle tip contact forces and resulting stress levels under engine conditions are critical to optimizing brush seal performance as well as to achieving operational safety. Literature survey reveals the lack of test data and analysis methods for evaluating seal stiffness and stress levels under operating conditions. In an attempt to meet this need, a custom test rig design and methodology has been developed to perform stiffness tests under pressure and rotor speed of 3000 rpm. Finite element simulations have been performed for brush seals and results have been correlated with the test data of this study. Considering the critical importance of contact loads on brush seal overall performance and system health, and due to the complicated structure of brush seals, where bristles are contacting with each other as well as with the backing plate and the rotor, CAE analyses with high fidelity is required to simulate the test and turbine operating conditions. For this purpose, FE methodology has been developed for structural analyses of brush seals. 3D finite element models of brush seals have been constructed and simulations have been performed for pressurized rotor-rub conditions. CAE model of brush seals includes rotor-bristle, bristle pack-backing plate and inter-bristle contacts with friction. Simulations with non-rotating rotor and transient analyses with rotating rotor have been conducted, and the extracted bristle tip force levels are correlated with the test results. Inertial effects during dynamic tests have also been simulated through transient analyses and results show good agreement with the dynamic test data. Displacement and stress profiles obtained from correlated FE models give better understanding of brush seal behavior under turbine operating conditions.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A038. doi:10.1115/GT2015-44068.

Brush seals are complex structures having variety of design parameters, all of which affect the seal behavior under turbine operating conditions. The complicated nature of the seal pack and frictional interactions of rotor, backing plate and bristles result in nonlinear response of the brush seal to variances of design parameters. This study presents CAE based characterization of brush seals, which aims to investigate the main effects of several brush seal design parameters on brush seal stiffness and stress levels. Characterization work of this study includes free-state rotor rub (unpressurized seal), steady state (pressure load without rotor interference) and pressurized-rotor interference conditions.

Commentary by Dr. Valentin Fuster
2015;():V05CT15A039. doi:10.1115/GT2015-44069.

While the efficiency of a brush seal is measured by its leakage rate, the overall performance of the seal is mostly affected by wear rate and durability. Seal stiffness and hysteresis behavior play important roles in determining the leakage performance and rotor stability due to the fact that they directly affect wear rates and pressure load capacity of the seal. The complicated nature of the bristle, rotor and backing plate interactions at typical operating conditions makes it difficult to determine the stiffness and durability of brush seals. In this study, test and computer aided engineering (CAE) methodologies have been developed to simulate brush seal stiffness and stress levels at unpressurized conditions. Unpressurized stiffness tests have been conducted by using two different test rigs, one of which uses a simple metallic pad and the other one uses a full-sized rotor for seal interference measurements. Test results for the two different rigs have been compared and the drawbacks of the simple stiffness test rig have been detailed in this study. CAE analyses at unpressurized conditions have been conducted by using 3D finite element (FE) models, and analyses have been correlated with the stiffness tests. The influence of rotor rotation has also been analyzed at unpressurized seal conditions. Transient simulation results also demonstrated good agreement with the dynamic stiffness tests of the brush seals.

Topics: Inspection , Stiffness
Commentary by Dr. Valentin Fuster

Transitional Flows (With Turbomachinery)

2015;():V05CT16A001. doi:10.1115/GT2015-42504.

The influence of low to moderate Reynolds number and low to moderate turbulence level on aerodynamic losses is investigated in an incidence tolerant turbine blade cascade for a variable speed power turbine. This work complements midspan heat transfer and blade loading measurements which are acquired in the same cascade at the same conditions. The aerodynamic loss measurements are acquired to quantify the influence of Reynolds number and turbulence level on blade loss buckets over the wide range of incidence angles for the variable speed turbine. Eight discrete incidence angles are investigated ranging from +5.8° to −51.2°. Noting that the design inlet angle of the blade is 34.2° these incidence angles correspond to inlet angles ranging from +40° to −17°. Exit loss surveys, presented in terms of local total pressure loss and secondary velocities have been acquired at four exit chord Reynolds numbers ranging from 50,000 to 568,000. These measurements were acquired at both low (∼0.4%) and moderate (∼4.0%) inlet turbulence intensities. The total pressure losses are also presented in terms of cross passage averaged loss and turning angle. The resulting loss buckets for passage averaged losses are plotted at varied Reynolds numbers and turbulence condition. The exit loss data quantify the impact of Reynolds number and incidence angle on aerodynamic losses. Generally, these data document the substantial deterioration of performance with decreasing Reynolds number.

Commentary by Dr. Valentin Fuster
2015;():V05CT16A002. doi:10.1115/GT2015-43326.

Roughness below a certain, admissible level will not increase an aerofoil’s skin friction and thus will not impact on engine performance. In this paper a simple model is developed that demonstrates how this admissible roughness height changes through the engine. The model is determined by combining existing analytical models and is backed up by computational validation where necessary.

It is shown that, given a fixed inlet/exit stagnation temperature and pressure to a blade row, the admissible roughness height is only a weak function of chord Reynolds number and Mach number for typical gas turbine blades. The aerofoil geometry/duty is also shown to have little impact. This allows the model to give a general picture of where roughness matters, irrespective of the exact details of the flow conditions/blade geometries.

The model shows that the admissible roughness height decreases as the stagnation pressure increases. The lowest admissible roughness level occurs at the high pressure end of the compressor at sea-level; the admissible roughness increases as the stagnation pressure drops towards the front of the compressor. Turbines are also most sensitive to roughness close to the combustor, but even there the admissible roughness will be around three times greater than at the rear of the compressor.

Commentary by Dr. Valentin Fuster
2015;():V05CT16A003. doi:10.1115/GT2015-43329.

Current criteria used to determine whether rough surfaces affect skin friction typically rely on a single amplitude parameter to characterize the roughness. The most commonly used criteria relate the centreline averaged roughness, Ra, to an equivalent sandgrain roughness size, ks. This paper shows that such criteria are oversimplified and that Ra/ks is dependent on the roughness topography, namely the roughness slope defined as the roughness amplitude normalized by the distance between roughness peaks, Ra.

To demonstrate the relationship, wake traverses were undertaken downstream of an aerofoil with various polished surfaces. The admissible roughness Reynolds number (ρ1u1Ra1) at which the drag rose above the smooth blade case, was determined. The results were used to demonstrate a 400% variation in Ra/ks over the roughness topographies tested.

The relationship found held for all cases tested, except those where the roughness first initiated premature transition at the leading edge. In these cases, where the roughness was more typical of eroded aerofoils, the drag was found to rise earlier.

Commentary by Dr. Valentin Fuster

Combustors (With Combustion, Fuels and Emissions)

2015;():V05CT17A001. doi:10.1115/GT2015-42217.

In order to deepen the knowledge of the interaction between modern lean burn combustors and high pressure turbines, a real scale annular three sector combustor simulator has been assembled at University of Florence, with the goal of investigating and characterizing the generated aerothermal field and the hot streaks transport between combustor exit and the high pressure vanes location. To generate hot streaks and simulate lean burn combustors behavior, the rig is equipped with axial swirlers, fed by main air flow that is heated up to 531 K, and liners with effusion cooling holes that are fed by air at ambient temperature. The three sector configuration is used to reproduce the periodicity on the central sector and to allow to perform measurements inside the chamber, through the lateral walls.

Ducts of different length have been mounted on the swirlers, preserving the hot mainflow from the interaction with coolant. Such configurations, together with the one without ducts, have been tested, using different measurement techniques, in order to highlight the differences in the resulting flow fields.

First of all, isothermal PIV measurements have been performed on the combustion chamber symmetry plane, to highlight the mixing phenomena between the mainflow and cooling flows. Then a detailed investigation of the mean aerothermal field at combustor exit has been carried out, for nominal operating conditions, by means of a five hole pressure probe provided with a thermocouple, installed on an automatic traverse system. With the aim of analyzing the hot streaks transport and the flow field modification towards the vanes location, such measurements have been performed on two different planes: one located in correspondence of the combustor exit and the further one placed downstream, in the virtual location of the vanes leading edges.

Therefore, an experimental database, describing the evolution of the flow field in a combustor simulator with typical traits of modern lean burn chambers, for different injector geometries, has been set up.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A002. doi:10.1115/GT2015-42218.

In order to deepen the knowledge of the interaction between modern lean burn combustors and high pressure turbines, a real scale annular three sector combustor simulator has been assembled at University of Florence, with the goal of investigating and characterizing the generated flow field. To generate hot streaks and simulate lean burn combustors behavior, the rig is equipped with axial swirlers, fed by main air flow that is heated up to 531 K, and liners with effusion cooling holes that are fed by air at ambient temperature. The three sector configuration is used to reproduce the periodicity on the central sector.

Ducts of different lengths have been mounted on the swirlers to reduce the interaction of the mainstream with the coolant. Such configurations have been tested, using different measurement techniques, in order to highlight the differences in the resulting flow fields.

The work presented in this paper shows the experimental campaign carried out to investigate the flow turbulence at combustor exit, in isothermal conditions, by means of hot wire anemometry. The goal has been achieved by investigating each test point twice, using an automatic traverse system equipped, in turn, with two split-fiber probes, that allow to measure the velocity components on two planes orthogonal to each other. A method for the time correlation of the signals obtained by the two different tests has been used.

In order to analyse the turbulence decay towards the vanes location, such measurements have been performed on two different planes: one located in correspondence of the combustor exit and the further one placed downstream, in the virtual location of the vanes leading edges.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A003. doi:10.1115/GT2015-42278.

To deepen the knowledge of the interaction between modern lean burn combustors and high pressure turbines, a non-reactive real scale annular trisector Combustor Simulator (CS) has been assembled at University of Florence, with the goal of investigating and characterizing the combustor aerothermal field as well as the hot streak transport towards the high pressure vanes. To generate hot streaks and simulate lean burn combustor behaviors, the rig is equipped with axial swirlers fed by a main air flow stream that is heated up to 531 K, while liners with effusion cooling holes are fed by air at ambient temperature. Detailed experimental investigations are then performed with the aim of characterizing the turbulence quantities at the exit of the combustion module, and specifically evaluating an integral scale of turbulence. To do so, an automatic traverse system is mounted at the exit of the CS and equipped to perform Hot Wire Anemometry (HWA) measurements. In this paper, two-point correlations are computed from the time signal of the axial velocity giving access to an evaluation of the turbulence timescales at each measurement point. For assessment of the advanced numerical method that is Large Eddy Simulation (LES), the same methodology is applied to a LES prediction of the CS. Although comparisons seem relevant and easily accessible, both approaches and contexts have fundamental differences: mostly in terms of duration of the signals acquired experimentally and numerically but also with potentially different acquisition frequencies. In the exercise that aims at comparing high-order statistics and diagnostics, the specificity of comparing experimental and numerical results is comprehensively discussed. Attention is given to the importance of the acquisition frequency, intrinsic bias of having a short duration signal and influence of the investigating windows. For an adequate evaluation of the turbulent time scales, it is found that comparing experiments and numerics for high Reynolds number flows inferring small-scale phenomena requires to obey a set of rules, otherwise important errors can be made. If adequately processed, LES and HWA are found to agree well indicating the potential of LES for such problems.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A004. doi:10.1115/GT2015-42402.

Turbine entry conditions are characterized by unsteady and strongly non-uniform velocity and temperature and pressure fields. The uncertainty and the lack of confidence associated to these conditions require the application of wide safety margins during the design of the turbine cooling systems, which are detrimental for the efficiency of the engine. These issues have been further complicated by the adoption of lean-burn technology in modern aeroengines, identified by many manufacturers as the most promising solution for a significant reduction of NOx emission. Such devices are in fact characterized by a very compact design, whereas the strong swirl component generated by the injector is maintained up to the end of the flametube due to the absence of dilution holes, which in conventional combustors provides the required pattern factor.

Bearing in mind complexity and costs associated to the experimental investigation of combustor-turbine interaction, CFD has become a key and complementary tool to understand the physical phenomena involved. Due to the well-known limitations of the RANS approach and the increase in computational resources, hybrid RANS-LES models, such as Scale Adaptive Simulation (SAS), are proving to be a viable approach to resolve the main structures of the flow field.

This paper reports the main findings of the numerical investigation of a hot streak generator for the study of combustor-turbine interaction. The results were compared to experimental data obtained from a test rig representative of a lean-burn, effusion cooled, annular combustor, developed in the context of the EU project FACTOR. Steady RANS and unsteady SAS runs were carried out in order to assess the improvements related to hybrid models. Additional simulations were performed to investigate the effect of the periodicity assumption and the impact of liner cooling modelling on the exit conditions.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A005. doi:10.1115/GT2015-42403.

This paper reports the main findings of a numerical investigation aimed at characterizing the flow field and the wall heat transfer resulting from the interaction of a swirling flow provided by lean burn injectors and a slot cooling system, which generates film cooling in the first part of the combustor liner. In order to overcome some well-known limitations of RANS approach, e.g. the underestimation of mixing, the simulations were performed with hybrid RANS-LES models, namely SAS-SST and DES-SST, which are proving to be a viable approach to resolve the main structures of the flow field. The numerical results were compared to experimental data obtained on a non-reactive three sector planar rig developed in the context of the EU project LEMCOTEC.

The analysis of the flow field has highlighted a generally good agreement against PIV measurements, especially for the SAS-SST model, whereas DES-SST returns some discrepancies in the opening angle of the swirling flow, altering the location of the corner vortex. Also the assessment in terms of Nu/Nu0 distribution confirms the overall accuracy of SAS-SST, where a constant over-prediction in the magnitude of the heat transfer is shown by DES-SST, even though potential improvements with mesh refinement are pointed out.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A006. doi:10.1115/GT2015-42478.

This paper presents the evaluation of the turbulence radiation interaction (TRI), which is usually neglected in Reynolds averaged simulations (RAS) of computational fluid dynamics (CFD) problems. As a result of neglecting this interaction, it is obtained that the radiative heat loss in the flame and the radiative thermal load on the combustor is too low. The TRI is evaluated using the optically thin fluctuation approximation (OTFA) in coupled (radiation and CFD) calculations in a highly turbulent lean premixed methane flame. The validity of the OTFA is investigated in terms of the resulting change in temperature and reaction progress. Two common absorption-emission models are used in Monte-Carlo radiation calculations: the weighted-sum-of-grey-gases (WSGG) and the statistical narrow band correlated-absorption coefficient (SNB-CK). Both models show that the radiative thermal load on walls is increased up to 27 % due to TRI. In contrast, the flame shape is not affected since local temperatures changes are less than 25 K.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A007. doi:10.1115/GT2015-42484.

The exit temperature distribution had a great effect on reliability and security in a gas turbine. In this paper, the exit temperature distribution of a small engine reverse-flow combustor with three injectors test module was experimentally obtained to qualitatively analyze the influence of the primary zone operating condition by changing the fuel air ratio at the ambient pressure and temperature condition. Under the nearly identical air condition, there was no obvious difference on the mixing performance with different fuel flow rate. The hot zones occurred at the same position of the combustor exit section, and the temperature declined in the radial direction from the center. It could be seen that the radial temperature profiles in FAR of 0.022–0.03 were almost same. Malvern experimental results showed that the air fuel ratio of swirler cup ranges from 5 to 40 and the droplet distribution index n could not be increased or decreased by the ratio at different air pressure drop. The air fuel ratio of combustor swirl cup had reached more than 5 which fuel particle had been nearly stable and not got some variation by changing the fuel mass rate. As a result, the increase of fuel air ratio had no impact on fuel atomization uniformity in combustor dome. The fuel had been completely atomized when the combustor fuel air ratio ranged from 0.022 to 0.03, and its impact on the droplet size and uniformity of fuel could be neglected. With the uniform fuel spray, a numerical study of the whole combustor had been made to analyze the strong relation between swirl flow and jets of primary holes and dilution holes. The dilution jets had a strong effect on quenching flame and temperature dilution. Along the combustor flow direction, the temperature difference became less and less obvious, the addition of fuel would enhance the combustion intensity mainly in combustion zone, but with an effect of dilution jet, the temperature distributions had little deviation when increasing the fuel air ratio. And it showed a same phenomenon that different fuel air ratio would make the same exit temperature distribution which was found to be in line with the experimental results. In a word, for the primary zone operating condition in the combustor, it almost had no effect on the temperature distribution at the exit of the combustor by changing the fuel air ratio from 0.022 to 0.030 in primary zone at normal pressure and temperature condition.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
2015;():V05CT17A008. doi:10.1115/GT2015-42584.

Over the last ten years there have been significant technological advances toward the reduction of NOx emissions from civil aircraft engines, strongly aimed at meeting stricter and stricter legislation requirements. Nowadays, the most prominent way to meet the target of reducing NOx emissions in modern combustors is represented by lean burn swirl stabilized technology. The high amount of air admitted through a lean-burn injection system is characterized by very complex flow structures such as recirculations, vortex breakdown and precessing vortex core, that may deeply interact in the near wall region of the combustor liner. This interaction makes challenging the estimation of film cooling distribution, commonly generated by slot and effusion systems.

The main purpose of the present work is the characterization of the flow field and the adiabatic effectiveness due to the interaction of swirling flow, generated by real geometry injectors, and a liner cooling scheme made up of a slot injection and an effusion array. The experimental apparatus has been developed within EU project LEMCOTEC and consists of a non-reactive three sectors planar rig; the test model is characterized by a complete cooling system and three swirlers, replicating the geometry of a GE Avio PERM (Partially Evaporated and Rapid Mixing) injector technology.

Flow field measurements have been performed by means of a standard 2D PIV (Particle Image Velocimetry) technique, while adiabatic effectiveness maps have been obtained using PSP (Pressure Sensitive Paint) technique. PIV results show the effect of coolant injection in the corner vortex region, while the PSP measurements highlight the impact of swirled flow on the liner film protection separating the contribution of slot and effusion flows. Furthermore an additional analysis, exploiting experimental results in terms of heat transfer coefficient, has been performed to estimate the net heat flux reduction (NHFR) on the cooled test plate.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A009. doi:10.1115/GT2015-42587.

International standards regarding polluting emissions from civil aircraft engines are becoming gradually even more stringent. Nowadays, the most prominent way to meet the target of reducing NOx emissions in modern aero-engine combustors is represented by lean burn technology. Swirl injectors are usually employed to provide the dominant flame stabilization mechanism coupled to high efficiency fuel atomization solutions. These systems generate very complex flow structures such as recirculations, vortex breakdown and processing vortex core, that affect the distribution and therefore the estimation of heat loads on the gas side of the liner as well as the interaction with the cooling system flows.

The main purpose of the present work is to provide detailed measurements of Heat Transfer Coefficient (HTC) on the gas side of a scaled combustor liner highlighting the impact of the cooling flows injected through a slot system and an effusion array. Furthermore, for a deeper understanding of the interaction phenomena between gas and cooling flows, a standard 2D PIV (Particle Image Velocimetry) technique has been employed to characterize the combustor flow field.

The experimental arrangement has been developed within EU project LEMCOTEC and consists of a non-reactive three sectors planar rig installed in an open loop wind tunnel. Three swirlers, replicating the real geometry of a GE Avio PERM (Partially Evaporated and Rapid Mixing) injector technology, are used to achieve representative swirled flow conditions in the test section. The effusion geometry is composed by a staggered array of 1236 circular holes with an inclination of 30deg, while the slot exit has a constant height of 5mm. The experimental campaign has been carried out using a TLC (Thermochromic Liquid Crystals) steady state technique with a thin Inconel heating foil and imposing several cooling flow conditions in terms of slot coolant consumption and effusion pressure drop. A data reduction procedure has been developed to take into account the non-uniform heat generation and the heat loss across the liner plate.

Results, in terms of 2D maps and averaged distributions of HTC have been supported by flow field measurements with 2D PIV technique focussed on the corner recirculation region.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A010. doi:10.1115/GT2015-42868.

Increasing pressure to reduce pollutant emissions such as NOx and CO, while simultaneously increasing the efficiency of gas turbines, has led to the development of modern gas turbine combustors operating at lean equivalence ratios and high compression ratios. These modern combustors use a large portion of the compressor air in the combustion process and hence efficient use of cooling air is critical. Backside impingement cooling is one alternative for advanced cooling in gas turbine combustors. The dome of the combustor is a primary example where backside impingement cooling is extensively used. The dome directly interacts with the flame and hence represents a limiting factor for combustor durability. The present paper studies two aspects of dome cooling: the impingement heat transfer on the dome heat shield of an annular combustor and the effect of the outflow from the spent air on the liner heat transfer. A transient measurement technique using Thermochromic Liquid Crystals (TLCs) was used to characterize the convective heat transfer coefficient on the backside of an industrial heat shield design provided by Solar Turbines, Inc. for Reynolds numbers (with respect to the hole diameter) of ∼ 1500 and ∼ 2500. Reynolds-Averaged Navier Stokes (RANS) calculations using the k-ω SST turbulence model were found to be in good agreement with the experiment. A standard heat transfer correlation for impingement hole arrays overestimated the mean heat transfer coefficient compared to the experiment and computations, however this could be explained by low biases in the results.

Steady state IR measurements were performed to study the effects that the spent air from the heat shield impingement cooling had on the liner convective heat transfer. Measurements were taken for three Reynolds numbers (with respect to the hydraulic diameter of the combustor annulus) including 50000, 90000, and 130000. A downstream shift in the flow features was observed due to the secondary flow introduced by the outflow, as well as a significant increase in the convective heat transfer close to the dome wall.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A011. doi:10.1115/GT2015-43035.

Numerical computations are performed on three configurations of a model gas turbine combustor geometry for cold flow conditions. The purpose of this study is to understand the effect of changes to combustor passage section on the location of peak convective heat transfer along the combustor liner. A Reynolds Averaged Navier-Stokes equations based turbulence model is used for all the numerical computations. Simulations are performed on a 3D sector geometry. The first geometry is a straight cylindrical combustor section. The second model has an upstream diverging section before the cylindrical section. Third one has a converging section following the upstream cylindrical section. The inlet air flow has a Reynolds number of 50000 and a swirl number of 0.7. The combustor liner is subjected to a constant heat flux. Finally, liner heat transfer characteristics for the three geometries are compared. It is found that the peak liner heat transfer occurs far downstream of the combustor for full cylinder and downstream convergent cases compared to that in the upstream divergent case. This behavior may be attributed to the resultant pressure distribution due to the combustor passage area changes. Also the magnitude of peak liner heat transfer is reduced for the former two cases since the high turbulent kinetic energy regions within the combustor are oriented axially instead of expanding radially outward. As a consequence, the thermal load on the liner is found to reduce.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A012. doi:10.1115/GT2015-43135.

Combustor liner of present gas turbine engines is subjected to high thermal loads as it surrounds high temperature combustion reactants and is hence facing the related radiative load. This generally produces high thermal stress levels on the liner, strongly limiting its life expectations and making it one of the most critical components of the entire engine. The reliable prediction of such thermal loads is hence a crucial aspect to increase the flame tube life span and to ensure safe operations.

The present study aims at investigating the aero-thermal behavior of a GE DLN1 (Dry Low NOx) class flame tube and in particular at evaluating working metal temperatures of the liner in relation to the flow and heat transfer state inside and outside the combustion chamber. Three different operating conditions have been accounted for (i.e. Lean-Lean partial load, Premixed full load and Primary load) to determine the amount of heat transfer from the gas to the liner by means of CFD. The numerical predictions have been compared to experimental measurements of metal temperature showing a good agreement between CFD and experiments.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A013. doi:10.1115/GT2015-43819.

The heat soak-back occurring in engines under post-shutdown conditions is a well-known phenomenon. This behavior is caused by the transmission of accumulated heat from hot parts or cavities during idle operation (such as turbines, contention rings, etc.) to colder ones (combustion chamber, injectors, etc.) when the airflow inside the engine approaches nullity. Then stagnant fluids in components such as the injectors (mainly TDE) and bearings become exposed to this heat which spreads by conduction, radiation and natural convection through the engine, and potentially leading fuel or oil to decompose and to form a build-up of carbon through a phenomenon called “coking”. Heat soak-back to engine components on shutdown, due to the thermal inertia of heated turbine parts, has the potential to cause deposits to build up in fuel injectors which can over time block the injectors. Blocked or partially blocked injectors must then be removed from the engine, inspected and sent for cleaning. Both soak-back and coking phenomenon have already been investigated by some motorists through experimental and structural (FEA) studies [1]. To the author’s knowledge however, no CFD model considering the airflow has yet been discussed, mainly because of the computing resources and the time it requires to simulate this unsteady phenomenon. As part of the present study and in order to fill in the gap on the availability of numerical data in the open literature for the heat soak-back occurring in a gas turbine combustor, the following investigation implies CFD simulation to predict the thermal behavior and magnitude of such a soak-back and its potential consequence on the fuel passages. A previous CFD simulation done by the authors showed that the use of a radiation model was required to provide some very reasonable results. As a follow up, the work to be presented in this paper will provide a more complete numerical soak-back procedure that can be used to predict the thermal behavior inside the combustor of a just shutdown gas turbine engine. Prior to the heat soak-back analysis, a non-premixed combustion model is run to simulate idle condition. Then these more realistic results for idle are used as initial conditions for the analysis of the transient heat dissipation occurring after shutdown. The following work includes a quick description of the experimental setup, and an introduction to the operational conditions for a simplified test rig. The full numerical procedure is then described. An analysis highlights the improved ability of the numerical model in predicting when the coking temperatures are reached using the adopted modeling techniques. It is observed that results obtained by the present model compare well with the experimental data to validate the simulation of this not so obvious natural convection phenomenon for a better understanding of this transient problem.

Commentary by Dr. Valentin Fuster
2015;():V05CT17A014. doi:10.1115/GT2015-43998.

We examine experimentally the transition from a single flame stabilized along the inner shear layer (ISL) to a double flame stabilized along both the inner and the outer shear layers (OSL) and spreading over the outside recirculation zone (ORZ) in a fully premixed swirl-stabilized combustor. This work is mainly driven by previous studies demonstrating the link between this transition in the flame macrostructure and the onset of thermo-acoustic instabilities [1, 2]. Here, we examine the transition mechanism under thermo-acoustically stable conditions as well as the dominant flow and flame dynamics associated with it. In addition, we explore the role of changing the thermal boundary conditions around the ORZ and its effect on the presence or absence of the flame there. We start by analyzing the two flames bounding the transition, namely the single conical flame stabilized along the ISL (flame III) and the double conical flames with reactions taking place in the ORZ (flame IV). A dual chemiluminescence approach — using two cameras with a narrow field of view focused on the ORZ — is undertaken to track the progression of the flame as it reaches the ORZ. During the transition, the flame front, initially stabilized along the ISL, is entrained by OSL vortices close to where the turbulent jet impinges on the wall, leading to the ignition of the reactants in the ORZ and the ultimately the stabilization of the flame along the outer shear layer (OSL). This ORZ flame is also subject to extinction when the equivalence ratio (ϕ) is between values corresponding to flames III and IV. For ϕ lower than the critical transitional value, the flame kernel originating from the ISL-stabilized flame is shown to reach the ORZ but fails to grow and quickly disappears. For ϕ higher than the critical value, the flame kernel expands as it is advected by the ORZ flow and ultimately ignites the reactants recirculating in the ORZ. Sudden and extreme peak-to-peak values of the overall heat release rate are found to be concomitant with the ignition and extinction of the ORZ reactants. Finally, Different thermal boundary conditions are tested by modifying the heat flux through the combustion chamber boundary, particularly around the ORZ. We find that the transition is affected in different ways: while the transition from flame III to IV (i.e. as ϕ increases) is insensitive to these changes; flame IV persists at lower ϕ as its value is reduced when heat losses through the boundaries are diminished.

Commentary by Dr. Valentin Fuster

Structures (With Structures and Dynamics)

2015;():V05CT18A001. doi:10.1115/GT2015-42098.

In order to achieve high working efficiency, modern gas turbines operate at high temperature which is close to the melting points of metal alloys. However, the support of turbine end suffers the thermal deformation. And the journal center position is also changed due to the effects of high temperature and shaft gravity. Tangential or radial supporting structures, which are composed of supporting struts, diffuser cones, hot and cooling fluid channel, are widely used in gas turbine hot end. Cooling technology is usually used to keep the bearing temperature in a reasonable range to meet requirements of strength and deformation of the supporting struts. In this paper, three major assumptions are proposed: (a) radiation is not considered, (b) cooling flow system is only partially modeled and analysis assumes significantly higher cooling flow that is not typical for current engines, and (c) only steady state heat transfer is considered. And a 3D fluid-solid coupled model based on finite-element method (FEM) is built to analyze the performances of both the tangential and the radial support. The temperature distribution, thermal deformation and stress of supports are obtained from CFD and strength analysis. The results show that either the tangential or radial support is used in a 270MW gas turbine; the thermal stress is about 90.3% of total stress which is produced by both thermal effects and shaft gravity. Comparing to the results from radial supports, it can be seen that the struts stress and position variation of journal center of tangential support are smaller. Due to a rotational effect of the bearing housing caused by the deformation of the tangential struts, the thermal stress in these tangential struts can be relieved to some extent. When both thermal effect and shaft gravity are considered, the stress of each tangential supporting strut is almost uniformly distributed, which is beneficial to the stability of rotor system in the gas turbine.

Topics: Fluids , Gas turbines
Commentary by Dr. Valentin Fuster
2015;():V05CT18A002. doi:10.1115/GT2015-42655.

Gas turbine engines must withstand severe thermo-mechanical conditions during present-day load operation, characterized by cyclic transients and long dwell times. Indeed engine components are subject to thermal transient conditions, thermo-mechanical strain and stress fields; those are not easily measurable during operation, making calculations hardly confirmable. All these operational factors can lead a turbine component life reduction, finally increasing lifetime costs.

The developed approach has been based on several calculations, such as thermal and FEM stress evaluation on the rotor components, tuned or validated by different field measurements carried out by thermocouples in the rotor core and the pre-tightening load variation of tie-rod. Transient disks and tie-rod temperatures (calculated by an in-house Secondary Air System code) have been tuned on experimental data. Thus, for rotor thermo-mechanical analysis more reliable boundary conditions have been provided. Rotor FEM analysis has been finally checked comparing the variation of the tie-bolt tension (calculated by FEM analysis) with the experimental behaviour observed during different operating conditions.

Topics: Engines , Rotors
Commentary by Dr. Valentin Fuster
2015;():V05CT18A003. doi:10.1115/GT2015-42940.

This study builds upon previous work by the authors, using a combination of 3D conjugate heat transfer (CHT) computational fluid dynamics (CFD) and finite element analysis (FEA) to characterise the thermal bow behaviour of a simple compressor shaft and case model under natural cooling. As with previous studies by the authors, body temperatures obtained from 3D CHT CFD solutions at set time intervals are transferred to FEA, where the physical distortion associated with the asymmetric thermal load is measured. The current study examines the influence of a range of shaft design parameters on the severity and duration of the shaft deformation. The parameters of interest include shaft length, annulus geometry, degree of shaft ventilation, and shaft internal cavity geometry. Each time the baseline model is modified to analyse the contribution of a parameter, the model is allowed to cool down from representative operational temperatures for a period of 180 minutes, over which time the shaft thermal bow, and shaft-to-case clearance, are measured. The results of this study indicate that the shaft’s thermal bow response and shaft-to-case clearance over time are highly sensitive to changes to its geometry, whereas the change in 180-degree out-of-phase shaft-to-case clearance is more sensitive to the case geometry, rather than the shaft. These results indicate that increasing the length of the shaft, reducing its wall thickness, or introducing a rising-line annulus, will increase the severity of the shaft thermal bow phenomenon; whereas introducing disc geometry inside the shaft will reduce the severity of the bow.

Commentary by Dr. Valentin Fuster
2015;():V05CT18A004. doi:10.1115/GT2015-43497.

In the present study an application for efficient cooling of turbine liner segments employing pulsating impinging jets was investigated. A combined numerical and experimental study was conducted to evaluate the design of a case cavity device which utilizes the periodically unsteady pressure distribution caused by the rotor blades to excite a pulsating impinging jet. Through an opening between the main annulus and a case cavity, pressure pulses from the rotor blades propagated into this cavity and caused a strong pressure oscillation inside. The unsteady CFD results were in good qualititative agreement with the measurement data obtained using high frequency pressure transducers and hot wire anemometry. Furthermore, the numerical study revealed the formation of distinct toroidal vortex structures at the nozzle outlet as a result of the jet pulsation. Within the scope of the measurements the influence of the operating point on the pressure propagation inside the cavity was investigated. The dependence of shape and amplitude of the pressure oscillation on engine speed and stage pressure ratio was found to be in accordance with an analytical consideration.

Topics: Cooling , Jets , Turbines , Blades
Commentary by Dr. Valentin Fuster
2015;():V05CT18A005. doi:10.1115/GT2015-44031.

Tighten force has much influence on tie-bolt fastened rotor dynamics. Temperature distribution in tie-bolt fastened rotor results in thermal expansion of rotor and rods. The difference of thermal expansion between rotor and rods causes the variation of bolt load. With considering the thermal contact conductance, the thermal model of tie-bolt fastened rotor was established by finite element method and the axial temperature distribution was obtained. The influences of surface roughness, nominal contact pressure and axial position of contact on axial temperature distribution were analysed. Based on temperature distribution in the tie-bolt fastened rotor, the variation of tighten force was investigated. Results show that nominal contact pressure, surface roughness and axial contact arrange have different influences on the variation of tighten force with temperature.

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