No abstract

In this work we present the numerical simulation of air-assisted liquid atomization at high pressure using the Smoothed Particle Hydrodynamics (SPH) method. Different post-processing tools are applied to facilitate the comparison with experimental observations. This allows to quantitatively validate the numerical method against the experiment, in terms of (i) frequency of the Kelvin-Helmholtz instability that develops on the jet surface, and (ii) statistical distribution of the jet intact length. The qualitative comparison also shows a good prediction of the jet global instability and of the fragmented liquid lumps, with regards to length and time scales. In addition, the post-processing tools also give access to the local parameters of the generated spray in the vicinity of the nozzle, which are not easily accessible in a real experiments. Using these tools, 1D profiles and 2D maps of the liquid phase properties such as the volume fraction, the droplet concentration, the Sauter Mean Diameter (SMD) and the droplet sphericity are presented. Because of the Lagrangian nature of the SPH method, it is also possible to monitor the whole atomization cascade as a causal tree, from the primary instabilities to the spray characteristics. This tree contains various information such as the fragmentation spectrum and the breakup activity, which are of great interest for researchers and engineers. Hence, the capability of the Smoothed Particle Hydrodynamics (SPH) method for simulating air-assisted atomization at high ambient pressure is demonstrated as well as its applicability to realistic configurations. This is a first step towards the development of a complete virtual spray test-rig.

In this paper, the complex two-phase flow during oil-jet impingement on a rotating spur gear is investigated using the meshless smoothed particle hydrodynamics (SPH) method. On the basis of a two-dimensional setup, a comparison of single-phase SPH to multiphase SPH simulations and the application of the volume of fluid method is drawn. The results of the different approaches are compared regarding the predicted flow phenomenology and computational effort. It is shown that the application of single-phase SPH is justified and that this approach is superior in computational time, enabling faster simulations. In the next step, a three-dimensional single-phase SPH setup is exploited to predict the flow phenomena during the impingement of an oil-jet on a spur gear for three different jet inclination angles. The oil’s flow phenomenology is described and the obtained resistance torque is presented. Thereby, a significant effect of the inclination angle on the oil spreading and splashing process as well as the resistance torque is identified.

In modern jet engines mostly air blast atomizers are used for the liquid fuel injection. The prediction of the spray generated by such atomizers was for a long time not feasible because of restricted computing resources. However, with modern super-computers the prediction of the atomization has come into reach. In the present paper a new approach for the numerical prediction of the primary atomization is presented. The methodology is based on the Smoothed Particle Hydrodynamics (SPH), which has originally been developed in the context of astrophysics. The numerical predictions to be presented were performed for a planar model atomizer, for which a vast amount of experimental data was collected by us previously. The major objectives of the numerical predictions are to elaborate the mechanism governing the effect of thickness of the trailing edge of the prefilmer on the size of the droplets and the temporal droplet formation rate.

In this paper the complex two-phase flow during oil-jet impingement on a rotating spur gear is investigated using the meshless Smoothed Particle Hydrodynamics (SPH) method. A comparison of single-phase SPH to multi-phase SPH simulation and the application of the Volume of Fluid method on the basis of a two-dimensional setup is drawn. The results of the different approaches are compared regarding the predicted flow phenomenology and computational effort. It is shown that the application of single-phase SPH is justified and that this approach is superior in computational time, enabling faster simulations. In a next step, a three-dimensional single-phase SPH setup is exploited to predict the flow phenomena during the impingement of an oil-jet on a spur gear for various jet inclination angles. Thereby, a significant effect of the inclination angle on the oil spreading and splashing process is revealed. Finally, a qualitative comparison to an experimental high-speed image shows good accordance.

At the “Institut für Thermische Strömungsmaschinen” (ITS) a numerical method based on the the meshfree “Smoothed Particle Hydrodynamics” (SPH) approach has been developed with the objective of computing primary breakup in the vicinity of fuel spray nozzles [1, 2]. In recent publications the successful application of the code to different flow problems is demonstrated [3, 4]. In this paper we present the first application of the method to investigate a simplified, but applied fuel spray nozzle geometry of the swirl cup design in 2D. The atomization process of Jet-A1 at ambient and at high pressure conditions is compared in terms of film flow development, mixing and spray characteristics. The influence of pressure is pointed out and quantified. The study demonstrates that the SPH method is a suitable toolbox for the analysis and the design of fuel spray nozzles. Unique analysis tools that are not available in grid-based CFD methods are presented and applied. Droplet distributions are extracted, which can be considered as possible input in subsequent Euler-Lagrange computations.

Primary breakup of liquid fuel in the vicinity of fuel spray nozzles as utilized in aero-engine combustors is numerically investigated. As grid based methods exhibit a variety of disadvantages when it comes to the prediction of multiphase flows, the "Smoothed Particle Hydrodynamics" (SPH)-method is employed. The eligibility of the method to analyze breakup of fuel has been demonstrated in recent publications by Braun et al, Dauch et al and Koch et al [1,2,3,4]. In the current paper a methodology for the investigation of the two-phase flow in the vicinity of fuel spray nozzles at typical operating conditions is proposed. Due to lower costs in terms of computing time, 2D predictions are desired. However, atomization of fluids is inherently three dimensional. Hence, differences between 2D and 3D predictions are to be expected. In course of this study, predictions in 2D and based on a 3D sector are presented. Differences in terms of gaseous flow, ligament shape and mixing are assessed. Keywords Multiphase Flows -Fuel Atomization -Smoothed Particle Hydrodynamics -Aero-Engine -Combustor IntroductionAiming at a reduction of pollutant emissions of air-traffic, academia and industry both invest in research to investigate processes causing the formation of pollutants of aero-engine combustors. One aspect influencing the formation of pollutants is the quality of the injected fuel spray and its placement within the gaseous flow field of the combustion chamber. Because of limited optical access and challenging thermodynamic conditions, experimental studies are costly and cannot provide detailed information about breakup of the fuel in the close vicinity of fuel injectors. Hence, numerical investigations analyzing the local two-phase flow are desired. At the "Institut für Thermische Strömungsmaschinen" (ITS) a numerical code based on the Lagrangian "Smoothed Particle Hydrodynamics" (SPH)-method has been developed by Koch, Hoefler and Braun [5,4]. The objective is to predict the liquid fuel breakup in the close vicinity of the fuel spray nozzle. Conventional grid based methods exhibit a variety of inherent shortcomings, which can be overcome by a fully Lagrangian approach. In recent publications the potential of the code in terms of two-phase flow predictions has been demonstrated successfully by Braun et al, Dauch et al and Keller et al [1,3,6]. Current state-of-the art methods for combustor design do not take into account the details of primary breakup. Correlations based on empirical studies are employed to impose droplet initial conditions for subsequent EulerLagrangian CFD predictions. These correlations must be tuned to each individual setup and need to be calibrated. Most of the correlations are valid only for low pressures and temperatures. A detailed simulation of primary breakup can overcome these shortcomings. It might provide transient droplet initial conditions, which can be used as input for subsequent Euler-Lagrangian predictions. Even if SPH computations are too costly for every day design studies, the...

Predictions of the primary breakup of fuel in realistic fuel spray nozzles for aero-enginecombustors by means of the SPH method are presented. Based on simulations in 2D, novel insightsinto the fundamental effects of primary breakup are established by analyzing the dynamics ofLagrangian-coherent structures (LCSs). An in-house visualization and data exploration platformis used in order to retrieve fields of the finite-time Lyapunov exponent (FTLE) derived from theSPH predictions aiming at the identification of time resolved LCSs. The main focus of this paperis demonstrating the suitability of FTLE fields to capture and visualize the interaction between thegas and the fuel flow leading to liquid disintegration. Aiming for a convenient illustration at a highspatial resolution, the analysis is presented based on 2D datasets. However, the method and theconclusions can analoguosly be transferred to 3D. The FTLE fields of modified nozzle geometriesare compared in order to highlight the influence of the nozzle geometry on primary breakup, whichis a novel and unique approach for this industrial application. Modifications of the geometry areproposed which are capable of suppressing the formation of certain LCSs, leading to less fluctuationof the fuel flow emerging from the spray nozzle.