Impact of Swirl Flow on the Penetration Behaviour and Cooling Performance of a Starter Cooling Film in Modern Lean Operating Combustion Chambers

Author(s):  
B. Wurm ◽  
A. Schulz ◽  
H.-J. Bauer ◽  
M. Gerendas

An experimental and numerical study is presented that deals with the impact of the swirled hot gas main flow on the penetration behaviour and cooling performance of a starter cooling film. Within modern combustion chambers designed for lean combustion the whole fuel/air mixing process is done by the fuel injectors without any additional mixing ports. Typically swirl stabilization is used within this kind of combustion chambers. The swirl flow interacts in a particular way with near wall cooling flows like starter cooling films which assure a proper wall cooling near the fuel injector. Experiments without combustion show the impact of the swirled main flow on the stability of the starter cooling film. Thermal analyses reveal a reduced cooling performance of the starter film near the stagnation area of the swirl flow. Laser optical measurement techniques reveal a significant reduced penetration of the starter cooling film close to the stagnation area. Numerical simulations show the reason for the reduced starter film performance in areas which cannot be accessed by optical measurement techniques. Based on experimental and numerical data different adaptive hole geometries where tested in combination with heat shield ribs in order to improve the starter film cooling performance. Results show that the combined application of heat shield ribs and adaptive cooling holes stabilize the starter cooling film and lead to a homogenous cooling performance.

Author(s):  
B. Wurm ◽  
A. Schulz ◽  
H.-J. Bauer ◽  
M. Gerendas

An experimental study on combustor liner cooling of modern direct lean injection combustion chambers using coolant ejection from both effusion cooling holes and a starter film has been conducted. The experimental setup consists of a generic scaled three sector planar rig in an open loop hot gas wind tunnel, which has been described earlier in Wurm et al. (2009, “A New Test Facility for Investigating the Interactions Between Swirl Flow and Wall Cooling Films in Combustors, Investigating the Interactions Between Swirl Flow and Wall Cooling Films in Combustors,” ASME Paper No. GT2009-59961). Experiments are performed without combustion. Realistic engine conditions are achieved by applying engine-realistic Reynolds numbers, Mach numbers, and density ratios. A particle image velocimetry (PIV) measurement technique is employed, which has been adjusted to allow for high resolution near wall velocity measurements with and without coolant ejection. As the main focus of the present study is a deeper understanding of the interaction of swirl flows and near wall cooling flows, wall pressure measurements are performed for the definition of local blowing ratios and to identify the impact on the local cooling performance. For thermal investigations an infrared thermography measurement technique is employed that allows high resolution thermal studies on the effusion cooled liner surface. The effects of different heat shield geometry on the flow field and performance of the cooling films are investigated in terms of near wall velocity distributions and film cooling effectiveness. Two different heat shield configurations are investigated which differ in shape and inclination angle of the so called heat shield lip. Operating conditions for the hot gas main flow are kept constant. The pressure drop across the effusion cooled liner is varied between 1% and 3% of the total pressure. Results show the impact of the swirled main flow on the stability of the starter film and on the effusion cooling performance. Stagnation areas which could be identified by wall pressure measurements are confirmed by PIV measurements. Thermal investigations reveal reduced cooling performance in the respective stagnation areas.


Author(s):  
B. Wurm ◽  
A. Schulz ◽  
H.-J. Bauer ◽  
M. Gerendas

An experimental study on combustor liner cooling of modern direct lean injection (DLI) combustion chambers using coolant ejection from both effusion cooling holes and a starter film has been conducted. The experimental setup consists of a generic scaled three sector planar rig in an open loop hot gas wind tunnel, which has been described earlier in Wurm et al. [1]. Experiments are performed without combustion. Realistic engine conditions are achieved by applying engine-realistic Reynolds numbers, Mach numbers, and density ratios. A Particle Image Velocimetry (PIV) measurement technique is employed, which has been adjusted to allow for high resolution near wall velocity measurements with and without coolant ejection. As the main focus of the present study is a deeper understanding of the interaction of swirl flows and near wall cooling flows, wall pressure measurements are performed for the definition of local blowing ratios and to identify the impact on the local cooling performance. For thermal investigations an infrared thermography measurement technique is employed that allows high resolution thermal studies on the effusion cooled liner surface. The effects of different heat shield geometry on the flow field and performance of the cooling films are investigated in terms of near wall velocity distributions and film cooling effectiveness. Two different heat shield configurations are investigated which differ in shape and inclination angle of the so called heat shield lip. Operating conditions for the hot gas main flow are kept constant. The pressure drop across the effusion cooled liner is varied between 1% and 3% of the total pressure. Results show the impact of the swirled main flow on the stability of the starter film and on the effusion cooling performance. Stagnation areas which could be identified by wall pressure measurements are confirmed by PIV measurements. Thermal investigations reveal reduced cooling performance in the respective stagnation areas.


Author(s):  
Yongbin Ji ◽  
Bing Ge ◽  
Shusheng Zang ◽  
Jiangpeng Yu ◽  
Ji Zhang

Gas turbine combustors design nowadays is aimed at achieving extremely lower NOx emissions through involving more air into the combustor to perform lean combustion, which results in the reduction of cooling air ratio for the liner walls. In this context, effusion cooling, one of the most effective cooling strategies, is adopted on the liner for its advantages of providing well cooling protection with limited amount of air. The swirl flow structure generated by the injector to stabilize flame in most modern lean-burn combustor is very complex with recirculation and vortex breakdown. So the interaction between three dimensional main flow and jets issued from the effusion holes is significant when assessing effusion cooling performance on the liner. In the present work, detailed effusion cooling feature on both inner and outer liners of a scaled annular combustor equipped with three axial swirlers has been provided under non-reactive and reactive conditions. The main flow is electrically heated for the non-reactive condition, while premixed combustion is realized after methane is fueled into the injectors and mixed with the air in the surrounding passage for the reactive condition. Temperature distribution on the target bended plate with 7 rows of discrete cooling holes in an in-line layout is captured by infrared thermography, and the cooling effectiveness is then analyzed. Effects of coolant to mainstream flow rate ratio and equivalence ratio are evaluated respectively. Results show that the macro rotational flow generated by the swirl flows interacts with cooling film and leads to non-symmetric cooling protection circumferentially on both liners. Additionally, averaged cooling effectiveness is found to increase with the flow rate ratio. At reactive conditions, stagnation of the high temperature swirl flow impinging on the liner wall locates at X/D range of 0.4–0.5, which has not been observed at non-reactive conditions. Also cooling effectiveness results indicate that outer liner obtains better cooling protection than inner liner when reaction is activated. Finally, the effect of most interested parameter for combustion process equivalence ratio is surveyed at Φ=0.7, 0.8 and 0.9. With experimental results, the importance of the combustion is highlighted in weighing the effusion cooling performance on the real annular combustor liners, which can’t be predicted comprehensively by non-reactive investigations. To obtain more knowledge of this issue, future work concerned with the flow field and flame visualization needs to be done through experimental techniques and numerical methods.


Author(s):  
B. Wurm ◽  
A. Schulz ◽  
H.-J. Bauer ◽  
M. Gerendas

Based on experimental results on a liner of a modern direct lean injection combustion chamber using coolant ejection from both effusion cooling holes and a starter film, a method is presented that allows the assessment of the cooling performance of the liner. As the main focus of the present study is a deeper understanding of the interaction of swirl flow and near wall cooling flow, wall pressure measurements are performed for the calculation of local blowing ratios and local coolant mass fluxes. Thermal investigations allow the calculation of adiabatic film cooling effectiveness and heat transfer coefficients. The pressure drop across the effusion cooled liner is varied between 1% and 3% of the total pressure of the main flow. As experiments are performed without combustion and at low temperature, the influence of radiation is neglected. Results show the impact of the swirled main flow on the stability of the starter film and on the effusion cooling performance. Stagnation areas which could be identified by wall pressure measurements are confirmed by detailed PIV measurements. Thermal investigations reveal reduced cooling performance in the respective stagnation areas. For the definition of the non dimensional cooling efficiency the measurement area is sub divided into rhombic sections, which are located around each effusion cooling hole. Based on the measurement results presented, heat fluxes per unit area can then be calculated and put together to the cooling efficiency.


Author(s):  
Peter Katzy ◽  
Josef Hasslberger ◽  
Lorenz R. Boeck ◽  
Thomas Sattelmayer

The presented work aims to improve computational fluid dynamics (CFD) explosion modeling for lean hydrogen–air mixtures on under-resolved grids. Validation data are obtained from an entirely closed laboratory-scale explosion channel (GraVent facility). Investigated hydrogen–air concentrations range from 6 to 19 vol %. Initial conditions are p = 0.1 MPa and T = 293 K. Two highly time-resolved optical measurement techniques are applied simultaneously: (1) 10 kHz shadowgraphy captures line-of-sight integrated macroscopic flame propagation and (2) 20 kHz planar laser-induced fluorescence of the OH radical (OH-PLIF) resolves microscopic flame topology without line-of-sight integration. This paper presents the experiment, measurement techniques, data evaluation methods, and simulation results. The evaluation methods encompass the determination of flame tip velocity over distance and a detailed time-resolved quantification of the flame topology based on OH-PLIF images. One parameter is the length of wrinkled flame fronts in the OH-PLIF plane obtained through automated postprocessing. It reveals the expected enlargement of flame surface area by instabilities on a microscopic level. A strong effect of mixture composition is observed. Simulations based on the new model formulation, incorporating the microscopic enlargement of the flame front, show a promising behavior, where the impact of the augmented flame front on the observed flame front velocities can be detected.


2020 ◽  
Author(s):  
Zbigniew Stępień

The undesirable deposits forming on the surfaces of various internal parts of reciprocating internal combustion engines and the systems operating in conjunction with them worsen during the operation of the engines and threaten their proper functioning. The deposits form as a normal result of the processes of fuel injection and creating and combusting the fuel–air mixture in engines. It was not investigated until the beginning of the 21st century, when extensive multi-directional research began not only to identify the causes of these deposits, the mechanisms behind their formation, and the factors leading to deposit growth, but also to determine the chemical composition of various groups of deposits. Such research became necessary because engines must comply with gradually tightening regulations on environmental protection, necessitating the introduction of increasingly complex engine designs and strategies for controlling the processes of precise and divided fuel injection into the combustion chambers and advanced algorithms for controlling the combustion processes according to the combustion system and the purpose of the engine. However, it became apparent that the co-functioning of the increasingly complex engine technologies and solutions, particularly of fuel injection systems, may be significantly disturbed by the deposits forming inside them. More and more complicated engine designs with tighter and tighter tolerances of the working parts necessitate the multi-directional testing of harmful deposits. An increasing number of factors affecting deposit formation are being identified, which leads to the development of increasingly complex classifications and subdivisions of deposits according to their type, composition, and form. At the same time, the search for lower emissions and greater engine efficiency is driving further mechanical changes in engines and vehicles. The higher temperatures and pressures connected with these changes are likely to impact the fuel being handled within the fuel and combustion systems. Such effects will inevitably cause the deposit chemistry and morphology to change. The size of the coke deposits produced may disturb the processes of fuel atomization, of filling the engine combustion chambers and swirling the charge, and in consequence may affect the efficiency of filling and the quality of the fuel–air mixture. These problems led to the development of a number of standardized and unstandardized methods for assessing the size of deposits. It was found that in the case of SI engines, the deposits that most endanger correct engine operation are those which are formed in the combustion chambers, on the inlet valves, inlet ducts, and fuel injector tips. The most common sign of deterioration caused by deposits is the loss over time of the performance, usability, and operational value which were originally declared by the manufacturer. In the case of CI engines, the most dangerous are coke (carbon) deposits formed on the external surfaces of the fuel injector nozzle tips and inside the injector nozzle orifices. In Europe, mandatory procedures for assessing the size of different coke deposits formed on different components in both SI and CI engines are being developed by the Coordinating European Council for the Development of Performance Tests for Transportation Fuels, Lubricants, and Other Fluids (CEC). The theoretical part of this publication reports the problems of the deposits produced in reciprocating internal combustion engines and their fuel systems. It discusses standard and non-standard engine test methods for both quantitative and qualitative assessment of deposits and presents the significance of the assessment methods which are currently used for the classification of deposits. The publication also presents the scope of application and the usefulness of methods for determining the threats posed to the functioning of an engine by various types of deposits and methods for identifying the causes of deposit formation, in particular those related to the composition of the fuels and lubricating oils used. The effects which fuel composition and the engine’s construction and operating parameters have on various engine deposits, the possible causes of deposit formation, and the importance of modern deposit control additives and high-technology solutions in counteracting this detrimental phenomenon are also all discussed. The experimental part presents the results of research carried out at the Oil and Gas Institute – National Research Institute concerning: • the incomparability of measurements of fuel performance obtained from various engine tests, • studies on the influence of various deposit control additives on the formation of harmful engine deposits during engine tests, • the influence of fuel treatments on the deposit formation processes in internal combustion engines (described qualitatively or quantitatively), • determination of the impact which various chemical compounds, serving as contaminants within the fuels, have on deposit formation in internal combustion engines and fuel injection systems, • determination of the impact that various chemical structures of the compounds within the fuels and biofuel blends have on deposit formation in internal combustion engines and fuel injection systems, • studies on the influence of bio-components contained in both petrol and diesel fuels on tendency for deposits to form in internal combustion engines, and • multidirectional studies on the impact of FAME degradation processes in biodiesel fuel blends on the formation of harmful engine deposits.


2007 ◽  
Vol 7-8 ◽  
pp. 265-270 ◽  
Author(s):  
Thorsten Siebert ◽  
Thomas Becker ◽  
Karsten Spiltthof ◽  
Isabell Neumann ◽  
Rene Krupka

The reliability for each measurement technique depends on the knowledge of it’s uncertainty and the sources of errors of the results. Among the different techniques for optical measurement techniques for full field analysis of displacements and strains, digital image correlation (DIC) has been proven to be very flexible, robust and easy to use, covering a wide range of different applications. Nevertheless the measurement results are influenced by statistical and systematical errors. We discuss a 3D digital image correlation system which provides online error information and the propagation of errors through the calculation chain to the resulting contours, displacement and strains. Performance tests for studying the impact of calibration errors on the resulting data are shown for static and dynamic applications.


Author(s):  
Robin Schmidt ◽  
Matthias Voigt ◽  
Konrad Vogeler ◽  
Marcus Meyer

This paper will compare two approaches of sensitivity analysis, namely (i) the adjoint method which is used to obtain an initial estimate of the geometric sensitivity of the gas-washed surfaces to aerodynamic quantities of interest and (ii) a Monte Carlo-type simulation with an efficient sampling strategy. For both approaches the geometry is parameterized using a modified NACA parameterization. First the sensitivity of those parameters is calculated using the linear (first order) adjoint model. Since the effort of the adjoint CFD solution is comparable to that of the initial flow CFD solution and the sensitivity calculation is simply a postprocessing step, this approach yields fast results. However, it relies on a linear model which may not be adequate to describe the relationship between relevant aerodynamic quantities and actual geometric shape variations for the derived amplitudes of shape variations. In order to better capture nonlinear and interaction effects, secondly a Monte Carlo-type simulation with an efficient sampling strategy is used to carry out the sensitivity analysis. The sensitivities are expressed by means of the Coefficient of Importance, which is calculated based on modified polynomial regression and therefore able to describe relationships of higher order. The methods are applied to a typical high pressure compressor stage. The impact of a variable rotor geometry is calculated by 3D CFD simulations using a steady RANS model. The geometric variability of the rotor is based on the analysis of a set of 400 blades which have been measured using high-precision 3D optical measurement techniques.


Author(s):  
Lars Högner ◽  
Alkin Nasuf ◽  
Paul Voigt ◽  
Matthias Voigt ◽  
Konrad Vogeler ◽  
...  

Geometric variations caused by manufacturing scatter can influence the aerodynamic performance of turbomachinery components. In case of nozzle guide vanes (NGVs), the capacity is of particular importance due to its influence on the entire engine behaviour, since often the narrowest cross section of the turbine, which limits the capacity, is found in the first NGV stage. Within this scope, the present paper illustrates different methods in order to quantify the impact of geometric variations of high pressure turbine (HPT) NGVs with respect to capacity change during the development process. At first, in the design phase, a parametric CAD model of the NGV can be used to perform an initial assessment of the effect caused by different geometric variations onto capacity. The results of this study can for example be used to set the tolerances for the subsequent manufacturing process. As soon as the first real hardware components become available, their geometry can nowadays be accurately captured using optical measurement techniques. Consequently, reverse engineering (RE) methods can be used to enable numerical assessment of geometric variability since manufacturing scatter is determined and incorporated into the subsequent CFD analysis. The process to perform this assessment is described in the second part of the paper and its results are compared to the initial CAD-based study. The investigation is conducted using an example of a state-of-the-art NGV stage provided by Rolls-Royce.


2017 ◽  
Vol 139 (11) ◽  
Author(s):  
Robin Schmidt ◽  
Matthias Voigt ◽  
Konrad Vogeler ◽  
Marcus Meyer

This paper will compare two approaches of sensitivity analysis, namely (i) the adjoint method which is used to obtain an initial estimate of the geometric sensitivity of the gas-washed surfaces to aerodynamic quantities of interest and (ii) a Monte Carlo type simulation with an efficient sampling strategy. For both approaches, the geometry is parameterized using a modified NACA parameterization. First, the sensitivity of those parameters is calculated using the linear (first-order) adjoint model. Since the effort of the adjoint computational fluid dynamics (CFD) solution is comparable to that of the initial flow CFD solution and the sensitivity calculation is simply a postprocessing step, this approach yields fast results. However, it relies on a linear model which may not be adequate to describe the relationship between relevant aerodynamic quantities and actual geometric shape variations for the derived amplitudes of shape variations. Second, in order to better capture nonlinear and interaction effects, a Monte Carlo type simulation with an efficient sampling strategy is used to carry out the sensitivity analysis. The sensitivities are expressed by means of the coefficient of importance (CoI), which is calculated based on modified polynomial regression and therefore able to describe relationships of higher order. The methods are applied to a typical high-pressure compressor (HPC) stage. The impact of a variable rotor geometry is calculated by three-dimensional (3D) CFD simulations using a steady Reynolds-averaged Navier–Stokes model. The geometric variability of the rotor is based on the analysis of a set of 400 blades which have been measured using high-precision 3D optical measurement techniques.


Sign in / Sign up

Export Citation Format

Share Document