Prefilming Airblast Atomizer Research Articles

Two approaches for constructing multivariate injection models for prefilming airblast atomizers

More realistic droplet starting conditions for Euler–Lagrangian simulations enable e.g. more precise soot prediction in jet engines. Up to now, mainly the droplet size distribution of sprays is considered, but not the multivariate dependence structure of droplet size, starting position and initial velocity. A novel concept for extracting multivariate spray data efficiently from detailed simulations of the atomizing process into high fidelity Euler–Lagrangian simulations of spray combustion is presented in this paper. Therefore, simulations of a prefilming airblast atomizer using the Smoothed Particle Hydrodynamics method are considered. The multivariate dependence structure in the spray is identified using rank transformation. Two models of different nature are proposed which are able to reproduce the multivariate dependence structure. The first model follows a data-driven approach using vine copulas and marginal distributions. In contrast, the second model is based on human knowledge and assumptions enabling deeper insights into the atomization process. Both models demonstrated to reproduce the multivariate character of the spray data effectively. An assessment of their capabilities reveals that the first model might be more suitable for spray data of annular injectors.

Flame dynamics and oscillation characteristics based on statistical analysis in a combustor with prefilming and non-prefilming atomization

Suppressing thermoacoustic oscillations in low-emission combustors during their operation poses a major challenge. In this study, we report dynamic flame tests on combustors with prefilming (S1) and non-prefilming (S2) airblast atomizers. We used the image fast Fourier transform, proper orthogonal decomposition, dynamic mode decomposition, and continuous wavelet transform to obtain the spatial distribution of pulsations and spectral characteristics of the flames. The results reveal that the flames of S1 and S2 were significantly different. The combustor of S1 had a dome-attached flame confined to the primary combustion zone, while S2 had a lifted flame that filled the entire combustor. As the rate of airflow at the inlet increased, the flame of S1 exhibited oscillatory combustion, while the flame of S2 remained stable under all tested conditions, which was consistent with observations of its dynamic images. No characteristic peak was observed in the spectra of S2 under all operating conditions, and under rates of inlet airflow of 40 and 60 g/s for S1. However, pulsations in the chemiluminescence signals of the flame had a primary frequency of 116.4 Hz and secondary harmonic at 232.4 Hz at 80 g/s for S1. At 100 g/s, the S1 flame exhibited a primary frequency of 142.9 Hz, secondary harmonic at 285.4 Hz, and tertiary harmonic at 428.3 Hz. Minor adjustments to the geometry of the airblast atomizer can thus significantly alter the mode of spray–wall interactions and impact flame dynamics. Consequently, this study proposes a new control technique for instability suppression.

The Effect of Film Development on Primary Breakup in a Prefilming Airblast Atomizer

Abstract Liquid fueled gas turbines are likely to remain a dominant force in aviation propulsion for the foreseeable future, and therefore understanding the atomization process is key to meeting future emissions and performance legislation. To make experiments and simulations possible, simplified geometry and boundary conditions are often used, for example, simulations of primary atomization often use a fixed film height and velocity. This paper aims to quantify the effect of a fully developed unsteady film on the atomization process. A custom Coupled Level Set & Volume of Fluid (CLSVOF) solver with adaptive meshing built in OpenFOAM v9 is used. A simulation of the atomization process in the Karlsruhe Institute of Technology atomization experiment (Warncke et al., 2017, “Experimental and Numerical Investigation of the Primary Breakup of an Airblasted Liquid Sheet,” Int. J. Multiphase Flow, 91, pp. 208–224) is presented. A precursor simulation of the film development is used to provide accurate, temporally and spatially resolved inlet boundary conditions. These results are compared to previous CLSVOF simulations from Wetherell et al. (2020, “Coupled Level Set Volume of Fluid Simulations of Prefilming Airblast Atomization With Adaptive Meshing,” ASME Paper No. GT2020-14213)” using traditional boundary conditions. The unsteady film has doubled the modal ligament length and widened the distribution, and is now in better agreement with experimental measurements. A clear correlation in both time and space is observed between the film, atomization process, and spray. The sauter mean diameter (SMD) is significantly increased, again giving better agreement with the experiment. A discussion of extracting statistical descriptions of the spray is given, outlining the unfeasible computational cost required to converge droplet diameter distributions and other high order statistics for a case such as this.

Effects of structured prefilmer lip on spray characteristics in planar prefilming airblast atomization

Effective liquid fuel atomization is an ultimate aim for the prefilming airblast atomizer in aero-engines, and the liquid fuel breakup behavior is the key indicator of atomization quality. Whereas, the prefilmer lip structure is unnoticed but of great influence on this breakup process. In this paper, the rectangular block structured prefilmer lip and the triangular prism structured prefilmer lip are delicately designed and adopted in planar prefilming airblast atomization. Accordingly, the spray characteristics, including the liquid film morphology, droplet distributions, and atomization quality are comprehensively investigated under structured prefilmer lips, using a coupled LB-LES model. The numerical results reveal that when the rectangular block structured prefilmer lip is adopted, the multihole morphology appears on the liquid film under the cutting effect of the rectangular sawtooth, thus contributing to the liquid breakup process. The droplet equivalent diameter in the atomization process distributes exponentially, and 70%-80% of the droplet sizes are less than 100 μm. The droplet velocity shows a normal distribution with its average value located from 1 m/s to 2 m/s in the streamwise direction when the air velocity is 20 m/s. Ultimately, according to droplet size, droplet velocity, and atomization angle, along with the atomization efficiency coefficient, the atomization quality under the rectangular block structured prefilmer lip is better than that with triangular prism structured prefilmer lip and unstructured prefilmer lip, and the highest atomization quality is obtained with rectangular block width of 0.3 mm.

Experimental investigations on dynamic wave structures of a prefilming atomizer

An experimental study was performed on a planar prefilming airblast atomizer to investigate the dynamic wave behavior on a prefilmer. Using a high-speed camera, backlighting visualization and reflection imaging were applied to record the breakup process of the liquid film and the temporary evolution of wave structures. Image post-processing methods were used to obtain the film thickness, surface wave frequency, breakup length, and ligament breakup frequency. Depending on non-dimensional aerodynamic numbers, surface waves on the prefilmer were classified into three mechanisms. In addition, correlations were derived from experimental data and used to predict atomization characteristics. The deviations of surface wave frequency and breakup length were limited to ±10%. These predictive correlations can assist in resolving engineering issues regarding prefilming atomizers.

A computational methodology to account for the liquid film thickness evolution in Direct Numerical Simulation of prefilming airblast atomization

Prefilming airblast atomization is becoming widely used in current aero engines. Fundamental studies on the actual annular configuration of airblast atomizers are difficult to realize. For this reason, researchers have also focused on planar configurations. In this regard, the Karlsruhe Institute of Technology (KIT) developed a test rig to conduct experimental activities, conforming a large database with results for different working fluids and operating conditions. Such data allow two-phase flow modelers to validate their calculations concerning primary atomization on these devices. The present investigation proposes a Direct Numerical Simulation (DNS) study on the KIT planar configuration through the Volume of Fluid (VOF) method within the PARIS Simulator code. The novelty compared to DNS works reported in the literature resides in the use of a boundary condition that allows accounting not only for the gas inflow turbulence but also for the spatio-temporal evolution of the liquid film thickness at the DNS inlet and its related effect on turbulence. The proposed methodology requires computing precursor single-phase and two-phase flow Large-Eddy Simulations on the prefilmer flow, with the assumption that the flow between computational domains is one-way coupled. Results are compared to DNS that only account for a constant (both timewise and spanwise) liquid film thickness at the domain inlet, validating the full methodology workflow. The proposed methodology is shown to improve the qualitative description of the atomization mechanism, as the different stages of breakup (liquid accumulation behind the prefilmer edge, bag formation, bag breakup, ligament formation and ligament breakup) coexist spanwise for a given temporal snapshot. This implies a more continuous atomization than the one predicted by the constant film thickness case, which showed the same breakup stage to be present along the prefilmer span for a given instant and led to a more discretized set of atomization events. The proposed workflow allows quantifying the influence of the liquid film flow evolution above the prefilmer surface on primary breakup frequency and relevant atomization features.

Numerical Investigation on Mechanism of Swirling Flow of the Prefilming Air-Blast Fuel Injector

Prefilming air-blast atomizers are widely used in modern gas turbine combustors. Due to insufficient awareness of the coupling mechanism of multi-stage swirling flow in gas turbines, there is a lack of effective methods for flow field optimization in combustor. In this study, the effect of some critical parameters on the flow field of a prefilming air-blast atomizer was analyzed with CFD. The parameters include the angle and number of the first swirler blades, the angle of the second swirler blades and the angle of sleeve. Furthermore, the coupling mechanism of two-stage swirling airflows of prefilming air-blast atomizer was discussed. Moreover, the influence of the interaction between two-stage counter swirling airflows on the characteristics of flow field was explained. The results show that with the increase in SNi, the axial length of the primary recirculation zone decreased, while the radial width increased. The starting position of primary recirculation zone (PRZ) moves forward with the increase in SNo. Reducing the sleeve angle β helps to form the primary recirculation zone. The results indicate that it is the transition of tangential velocity of airflow to radial velocity that promotes the formation of the PRZ. These results provide theoretical support for optimization of the flow field in swirl combustor.

Experimental study on effects of prefilming and non-prefilming atomization on the combustion characteristics in a realistic full annular combustor

The interaction between liquid fuel and the walls poses a major challenge to using the airblast atomizer in the combustor of an advanced aero-engine because the mechanism of this interaction is not fully understood. In this study, the authors experimentally examine the influence of the interaction between liquid particles and the walls on the performance of a full annular combustor in terms of fuel atomization, the characteristics of the flow field, ignition and lean blowout (LBO), the combustion efficiency, and the distributions of temperature at the outlet corresponding to prefilming and non-prefilming methods of atomization. The results show that the Sauter mean diameter (SMD) of the prefilming airblast atomizer was smaller than that of the non-prefilming airblast atomizer when the fuel pressure difference was small. As the fuel pressure difference was continuously increased, the SMD of the prefilming airblast atomizer gradually became larger than that of the non-prefilming airblast atomizer. The morphology of the flow field downstream of the airblast atomizer could be changed by adjusting between the prefilming and non-prefilming modes, i.e., the flow field downstream of the prefilming mode was a typical swirling recirculation zone while that downstream of the non-prefilming mode had a recirculation zone on both sides of the forward jet at the center. When the inlet air flow is small, the ignition and LBO performance of the non-prefilming scheme were better than those of the prefilming scheme. As the inlet air flow was continuously increased, however, the ignition and LBO performance of the prefilming scheme became better than those of the non-prefilming scheme, and it had a higher combustion efficiency under the ground idle condition for different fuel–air ratios. The quality of temperature distribution at the outlet corresponding to the non-prefilming scheme was better than that of the prefilming scheme as well.

Effects of the structured lips on spray characteristics in prefilming airblast atomization

Nowadays, prefilming atomizers have been widely applied in aero-engine, industrial fields, and agricultural irrigation. In this study, atomization characteristics of the prefilming atomizers with structured lips have been investigated. The influence of eight structured lips on the spray breakup process has been measured using high-speed planner laser introduced fluorescence and particle/droplet image analysis in the planner prefilming air-blast atomizer. Synchronously, multi-directional visualization was used to measure hydrodynamic behavior. Rec-3 with a depth of 1 mm, a width of 1 mm, and a pitch of 0.5 mm has the smallest droplet size although it has medium breakup length and ligament velocity. For all studied structured prefilmers, rectangle and triangle structured lips performed good atomization effect than no structured lips except tri-1.

Investigation of Water Films Shed From an Airfoil in a High-Speed Flow

Abstract The breakup of liquid films in high-speed flows is found in many applications. These include prefilming air blast atomization found in fuel injectors and shedding of droplet off of airfoils. In this work, a test rig has been developed and applied to study the breakup of water films from the trailing edge of an NACA 0012 airfoil placed in a high-speed air flow. The results provided reflect a new detailed dataset focused on air velocities above 75 m/s. Computational fluid dynamic (CFD) simulations were used to guide the development of the rig and the associated boundary and inlet conditions. In this work, air velocities up to 175 m/s were used with water films between 0.4 cm2/s and 3.6 cm2/s. The water film was introduced onto one surface through a series of 0.5 mm holes separated by 1 mm at a location 35 mm downstream of the leading edge of the vane. High speed video was used to document the sheet breakup behavior from the edge of the airfoil. Image analysis was used to identify the breakup length of ligaments formed. Laser diffraction was used 50 mm downstream of the trailing edge of the vane to determine the droplet size distribution and associated representative average diameters generated. A series of test conditions were run between 50 m/s and 175 m/s using multilevel factorial designed experiments. The results obtained were subjected to analysis of variance to generate correlations for breakup length and droplet representative diameters as a function of the conditions studied. The analysis was conducted on dimensionless versions of the variables studied to help connect the results to the physics of the situation. For example, rather than using air velocity to correlate the behavior, the Weber number (based on gas phase density and velocity) was used owing to its historical ability to describe liquid breakup processes. The results obtained are compared with other studies in the literature, and good agreement was found for the conditions at which a comparison could be made. It is observed that, for the conditions studied, liquid film thickness has little effect on the resulting droplet sizes—especially for the higher air velocity cases studied. The results also illustrate the sensitivity to choice of the characteristic length, which is generally consisted of trailing edge geometry or boundary layer thickness. In contrast, liquid film does influence the ligament breakup length, although the time to break up is less affected. Overall, the correlations developed provide engineering tools to help estimate ligament and drop size behavior for water films shed from airfoils in a high-speed flow.

Influence of the wettability on the liquid breakup in planar prefilming airblast atomization using a coupled lattice Boltzmann–large eddy simulation model

In this paper, an efficient coupled lattice Boltzmann–large eddy simulation model [X. An et al., “Coupled lattice Boltzmann-large eddy simulation model for three-dimensional multiphase flows at large density ratio and high Reynolds number,” Phys. Rev. E 104, 045305 (2021)] based on the Allen–Cahn phase-field theory is introduced for simulating the liquid breakup in planar prefilming airblast atomization. This is the first time that the lattice Boltzmann method is used in the three-dimensional numerical investigation of prefilming airblast atomization to the best of our knowledge. The present model utilizes two evolution equations: one is used to capture the fluid interface, and another is adopted to solve hydrodynamic properties. An advanced multiple-relaxation-time scheme is also applied for the collision operator to enhance the numerical stability. To investigate the influence of the wettability on the liquid breakup accurately, a simple and efficient wetting boundary scheme is delicately designed and strictly validated. Additionally, to evaluate the atomization quality intuitively, an atomization efficiency coefficient is proposed for characterizing the liquid breakup process. The numerical results reveal that the influence of the wettability lies in the liquid accumulation phenomenon at the edge of the prefilmer and the droplet movement in the vertical direction. The atomization quality adopting a non-wetting prefilmer is better than other cases, according to the atomization efficiency coefficients, the mean droplet equivalent diameters with their size distribution proportions, and the atomization angles in the vertical direction. In addition, it is also found that the droplet proportion above the prefilmer increases as the contact angle increases, and the proportions on both sides of the prefilmer account for 50% at the contact angle of 90°.

A hybrid moment of fluid–level set framework for simulating primary atomization

This paper presents a hybrid moment of fluid–level set (HyMOFLS) method of liquid/gas interface reconstruction for the application to simulate primary atomization of liquid fuel. This method combines the moment of fluid (MOF) method and the level set framework in the coupled level set volume of fluid (CLSVOF) method. In this hybrid method, the MOF and CLSVOF methods are used to reconstruct the interface in the under-resolved and the resolved regions of the flow, respectively. An interface surface resolution metric called interface resolution quality (IRQ) is introduced to identify and classify these two flow regions in the computational domain. Such a strategy classifies/tags each computational cell with MOF or CLSVOF method based on a threshold value for the IRQ. The MOF method uses liquid volume fraction as well as centroids of liquid and gas phases for the interface reconstruction. The CLSVOF method uses the level set for describing the interface and liquid volume fraction for conserving the mass. The phase centroids in the HyMOFLS method are computed and transported on-the-fly during the cell tagging process. The transport of the liquid volume fraction, level set, and the phase centroids are performed using a directionally split algorithm. This algorithm is coupled with the Navier-Stokes equations solver that uses ghost fluid method as well as consistent mass and momentum flux computation for the momentum equation. Various numerical tests that exhaustively assess the capabilities, accuracy, and computational time consumption of the HyMOFLS method under multiple flow conditions and configurations are presented. The results from these tests suggest that this hybrid framework is capable of well capturing the liquid/gas interface belonging to thin and under-resolved structures that are often encountered in simulations of turbulent atomization of liquids. Following these tests, a detailed parametric study on the threshold value of IRQ is presented to test its effect on the accuracy of interface reconstruction. Finally, this hybrid framework is employed to simulate turbulent jet injection and pre-filming planar Airblast atomization cases of engineering applications. The proposed HyMOFLS method is found to achieve a good balance between the accuracy and the computational cost of reconstructing the liquid/gas interface for complex and turbulent atomization configurations.

Detailed numerical simulations of primary atomization of airblasted liquid sheet

This paper investigates the primary atomization of airblasted liquid sheet using detailed numerical simulations. The atomization of liquid sheet under airblasting conditions involve complex mechanisms and a thorough understanding is necessary. A planar pre-filming airblast atomization configuration have been chosen to study the breakup of liquid sheet/film injected on a solid flat plate. We have investigated an operating point directly relevant for high altitude relight condition of the aircraft. This configuration has been chosen based on the experimental investigation of Gepperth et al. [S. Gepperth, A. Müller, R. Koch, H.-J. Bauer, Ligament and droplet characteristics in pre-filming airblast atomization, Proceedings of the ICLASS, 12th Triennial International Conference on Liquid Atomization and Spray Systems, September 2-6, Heidelberg, Germany, 2012] for the airblast atomization. The numerical simulations have been performed using in-house Navier–Stokes solver that uses consistent mass and momentum flux computation technique. The purpose of this work is to provide a comprehensive database and analyses of the airblast atomization of liquid sheet. This include studies on the effect of velocity profile on the atomization characteristics, occurrence of secondary atomization and drop coalescence, and extraction of near-field atomization characteristics. The qualitative analyses of the results from the simulations showed that there are two major atomization mechanisms of liquid film breakup — sheet/bag breakup and ligament breakup. The drop diameter and velocity distributions computed from the simulations was found to be of the same order of magnitude although under-predicting the experimental data. Based on the atomized drop data, both the secondary atomization and drop coalescence have been observed to occur in the simulations. The quantitative analyses of the near-field liquid ligaments results revealed the lengths of these ligaments are of the same order of magnitude as the experimental data while an under-prediction in the ligament velocity has been observed. Finally, an excellent agreement between simulations and experimental data has been found for the Sauter Mean Diameter (SMD) of the atomized droplets.

Spray Drop Size Distribution and Velocity Distribution issued from a Prefilming Airblast Atomizer

Spray Drop Size Distribution and Velocity Distribution issued from a Prefilming Airblast Atomizer

A novel inlet boundary condition for DNS-VOF simulations of the primary breakup of prefilming airblast atomizers

Research leading to these results has received funding from the Clean Sky 2 Joint Undertaking under European Union?s Horizon 2020 research and innovation programme through the ESTiMatE project, grant agreement 821418. We acknowledge PRACE for awarding us access to computational resources on JOLIOT CURIE-AMD at GENCI@CEA, France (proposal 2019204944). Support given to Carlos Moreno by Universitat Politecnica de Valencia through the ?FPI-Subprograma 2? grant within the ?Programa de Apoyo para la Investigacion y Desarrollo (PAID-01-19)? is acknowledged. We must thank Marco Crialesi and Wojciech Aniszewski for their technical advice on custom modifications on PARIS, and Johan Sundin for his implementation of the contact angle model. We also thank Achille Schmitter, Hugo Martinez and Lucas Gonzalez for help processing the 1-phase LES, 2-phase LES and DNS, respectively.

Best Architecture of an Artificial Neural Network to Model Prefilming Airblast Atomization: Not So Deep Learning

Abstract A fully connected artificial neural network (ANN) is used to predict the spray characteristics of prefilming airblast atomization. The model is trained from the planar prefilmer experiment from the Ph.D. thesis of Gepperth [Experimentelle Untersuchung des Primärzerfalls an generischen luftgestützten Zerstäubern unter Hochdruckbedingungen, Vol. 75. Logos Verlag Berlin GmbH], in which shadowgraphy images of the liquid breakup at the atomizing edge capture the characteristics of the primary droplets and the ligaments. The quantities extracted from the images are the Sauter Mean Diameter, the mean droplet axial velocity, the mean ligament length, and the mean ligament deformation velocity. These are the prescribed output of the ANN model. In total, the training database contains 322 different operating points at which different prefilmers, liquid types, ambient pressures, film loadings, and gas velocities were investigated. Two types of model input quantities are investigated. First, nine dimensional parameters related to the geometry, the operating conditions, and the properties of the liquid are used as inputs for the model. Second, nine nondimensional groups commonly used for liquid atomization are derived from the first set of inputs. These two types of inputs are compared. The architecture providing the best fitting is determined after testing over 10,000 randomly drawn ANN architectures, with up to 10 layers and up to 128 neurons per layer. The striking results is that for both types of model, the best architectures consist of a shallow net with the hidden layers in the form of a diabolo: three layers with a large number of neurons (≥24) in the first and the last layers, and very few neurons (≈12) in middle layer. This shape recalls the shape of an auto-encoder, where the middle layer would be the feature space of reduced dimensionality. The trend highlighted by our results, to have a limited number of layers, is in contrast with recent observations in deep learning applied to computer vision and speech recognition. It was found that the model with dimensional input quantities always shows a lower test and validation errors than the one with nondimensional input quantities. The best architectures for both types of inputs (dimensional and nondimensional input) were tested versus the experiments. Both provide comparable accuracy, which is better than typical correlations of Sauter mean diameter (SMD) and droplet velocity. As the models take more input parameters into account compared to the correlations, they can predict the experimental data more accurately. Finally, the extrapolation capability of the models was assessed by training them on a confined domain of parameters and testing them outside this domain. It was found that the models can extrapolate at larger gas velocity. With a larger ambient pressure or a lower trailing edge thickness, the accuracy decreases drastically.

A COMPARATIVE STUDY OF DIRECT NUMERICAL SIMULATION AND EXPERIMENTAL RESULTS ON A PREFILMING AIRBLAST ATOMIZATION CONFIGURATION

A COMPARATIVE STUDY OF DIRECT NUMERICAL SIMULATION AND EXPERIMENTAL RESULTS ON A PREFILMING AIRBLAST ATOMIZATION CONFIGURATION

Experimental Investigation of Spray Characteristics of Pre-Filming Air-Blast Atomizers II – Influence of Liquid Properties

One of the most important advantages of twin-fluid atomizers over other atomizers is the opportunity of atomizing liquids in a wide range of liquid properties efficiently. Thus, the fundamentals of spray dynamics, such as spray pattern, liquid breakup length, and spatial droplet size evolution using different liquid properties need to be investigated in some detail. The purpose of this study is therefore to examine and describe the influence of the liquid viscosity and surface tension on the spray performance of prefilming air-blast atomizers. A high-speed background shadowgraphy technique associated with particle tracking and Phase Doppler Anemometry (PDA) as a non-intrusive high-resolution local measurement technique were utilized to study the atomization process regarding the influence of the liquid properties on the spray quality. Generally, the break up mechanism is considerably influenced by the liquid viscosity. The breakup length ( ) is dependent on liquid viscosity with for Pa = 0.25 and 0.6 bar air pressure, respectively. Droplet size distribution profiles of different liquids in axial and radial directions are studied. No significant influences of the surface tension and viscosity were observed on the mean droplet velocity downstream of the spray. A minor increase of the droplet mean diameter was observed with liquid viscosity variations from 22 to 111 mPas, however, the liquid surface tension was verified to play an important role in the atomization process at prefilming airblast atomizers as well. According to the available experimental data, it has subsequently been possible to develop an original and descriptive model based on a dimensional approach at different air pressure ranges.

Influence of the ambient pressure on the liquid accumulation and on the primary spray in prefilming airblast atomization

The influence of the ambient pressure on the breakup process is investigated by means of PIV and shadowgraphy in the configuration of a planar prefilming airblast atomizer. The ambient pressure is varied from 1 to 8 bar. Other investigated parameters are the gas velocity and the film loading. From single-phase PIV measurements, it is found that the gas velocity in the vicinity of the prefilmer partly matches the analytical profile from the near-wake theory. As done in previous publications, the characteristics of the liquid accumulation are extracted from the shadowgraphy images of the liquid phase directly downstream of the prefilmer. Two different characteristic lengths, as well as the ligament velocity and a breakup frequency are determined. In addition, the droplets generated directly downstream of the liquid accumulation are captured. Hence, the spray Sauter Mean Diameter (SMD) and the mean droplet velocity are given for each operating point. The novelty of this study is that a scaling law of these quantities with regard to ambient pressure is derived. A correlation is observed between the characteristic length of the accumulation and the SMD, thus reinforcing the idea that the liquid accumulation determines the primary spray characteristics. In this paper, a threshold to distinguish the zones between primary and secondary breakup is proposed based on an objective criterion. It is also shown that taking non-spherical droplets into account significantly modifies the shape of the dropsize distribution, thus stressing the need to use shadowgraphy when investigating primary breakup. Additionally, the ambient pressure and the velocity are varied accordingly to keep the aerodynamic stress ρgUg2 constant. This leads to almost identical liquid accumulation and spray characteristics. Hence, it is confirmed that the aerodynamic stress is a more appropriate parameter than the gas velocity or the ambient pressure to characterize prefilming airblast breakup. Finally, SMD correlations from the literature are compared to the present experiment. Most of the correlations calibrated with LDA/LDT measurement underestimate the SMD. This highlights the need to use shadrowgraphy for calibrating primary breakup models.

Investigation of the Secondary Atomization in Prefilming Air-Blast Atomizers

Investigation of the Secondary Atomization in Prefilming Air-Blast Atomizers