Elsevier

Computers & Fluids

Volume 136, 10 September 2016, Pages 467-484
Computers & Fluids

Computational analysis of nozzle geometry variations for subsonic turbulent jets

https://doi.org/10.1016/j.compfluid.2016.05.033Get rights and content

Highlights

  • Effect of nozzle built-in components on the jet flow development is assessed.

  • Numerical method is successfully validated against references from the literature.

  • A Cartesian mesh method for three nozzle geometries with increasing complexity.

  • Highly resolved large-eddy simulation for turbulent hot jets.

  • The jet near field is dominated by flow structures induced by the geometry.

Abstract

Large-eddy simulations (LES) of turbulent hot jets emanating from realistic helicopter engine nozzle configurations at a Reynolds number of Re = 7.5 × 105 and a Mach number of M=0.341 are conducted. The numerical method is based on hierarchically refined Cartesian meshes. The nozzle wall boundaries are resolved by a conservative cut-cell method. Three nozzle geometries of increasing complexity are considered, i.e., the flow fields of a clean geometry without any built-in components, a nozzle with a centerbody, and a nozzle with a centerbody plus struts are computed. The numerical method is validated by solutions for a single, a coaxial, and a chevron nozzle jet problem. A grid convergence study shows that the essential flow characteristics due to the intricacy of the nozzle geometry are well resolved. The results evidence that the flow field in the region 35 nozzle radii downstream of the exit is dominated by flow structures induced by the geometry. Compared to the clean geometry, the other two configurations show enhanced turbulent mixing. The centerbody and centerbody-plus-strut nozzle configurations reveal a spectral peak in the near nozzle exit region at St=0.15 which is caused by the wake flow of the centerbody.

Introduction

The noise generated by engine jets is to a large extent influenced by the flow structures in the exit region of the nozzle. For this reason, any kind of aeroacoustic investigation should be based on a flow field that contains the essential flow structures. That is, it is not only necessary to at least approximately resolve the boundary layer on the casing of the nozzle that defines the transition of the turbulent structures from the wall-bounded to the free-shear layer but also the flow characteristics that are determined by the built-in components of the nozzle. Due to the geometric intricacy defined by those components there exist only a few studies in which the impact of the geometry variations of the nozzle interior has been investigated.

Xiong et al. [37], for instance, evaluated the aerodynamic efficiency of the Fan Flow Deflection (FFD) method in a supersonic turbofan jet, where the FFD method could control the flow development in the near field. Using vanes the turbulent kinetic energy downstream of the nozzle exit was redistributed such that the peak turbulence intensity on the downwards direction was reduced by 50% which caused a thrust penalty of approx. 0.1%. They showed the shape of the airfoil vanes to have an important impact on the aerodynamic performance of the FFD method. Brown et al. [11] developed the so-called offset stream technology concept, in which the jet is reshaped by using wedges, vanes, and S-ducts. In other words, the turbulent flow field was controlled at sideline and downstream regions of the jet, which allows to azimuthally redirect the jet flow development. Each component was analyzed and their efficiency was compared with a baseline configuration. The modified configurations by the offset stream technology revealed an earlier potential core collapse. Similarly, offset stream nozzles at different operating conditions were numerically investigated by Dippold et al. [14]. They concluded that the presence of the S-duct and the vanes reduces the turbulence intensity on the lower side of the jet plume with a less than 0.5% thrust penalty at take-off which results in an acoustic shielding effect to the location on the downwards direction, while an increase of the turbulence intensity was observed on the upper side of the jet plume. Additionally, the thrust reduction was found to be less than 0.1% at cruise condition. Moreover, Papamoschou [27] investigated the FFD method in experiments that the vanes can control the flow field such that the flow is deflected in the downward direction of the jet axis which shortens the primary potential core of the jet and enhances the turbulent mixing for a coaxial nozzle. A joint experimental and numerical study on the control of a jet plume with beveled nozzles was successfully carried out by Viswanathan et al. [34]. They investigated the deflection of the jet stream by varying the bevel angle and showed that a deflection can cause noise reduction in the aft angles. It was additionally shown that an increase on the bevel angles over 30 ° causes a remarkable alteration on the turbulent flow field and results in a large thrust reduction of the engine.

To analyze the turbulence generation by internal cylindrical spokes in a nozzle, Dahl and Nichols [12] performed LES and determined the vortex shedding frequency of the spokes. They concluded that adding such turbulence generators, fairly enhances the jet shear layer development. In real nozzle applications, there exist quite often internally mixed multi-shear-layer flows. In general, such multiple jet streams alter the turbulent mixing quite dramatically. The potential core length of a dual stream jet was remarkably altered by the change in the splitter area ratio in an experimental investigation by Bridges [9]. Another example is a twin converging-diverging nozzle whose flow and acoustic fields were analyzed numerically by Brés et al. [8]. The S-duct, Y-duct, and bypass-duct geometries upstream of the twin nozzle significantly change the flow field inside the nozzle. It is clear that such complex nozzle configurations and their built-in components possess particular influence on the jet flow development.

The turbulent flow in the jet near field can be roughly divided into two distinct flow categories [31]. The first category is defined by large-scale turbulent structures that determine the low frequencies of the acoustic field downstream of the nozzle exit, while the small-scale turbulent structures which are located in the shear layer near the nozzle exit define the high frequency near-field acoustics. These turbulent scales are affected by the nozzle geometry. Although some attention has been paid on the impact of the nozzle built-in components to the jet development, a comparison between a clean nozzle, i.e., a nozzle without any built-in components, and nozzles with a step-by-step increased complexity of the interior geometry such that the influence of the geometric variation can be investigated, is still missing. Since the built-in components inside the nozzle break the large scale turbulence into smaller scales, it is necessary to highly resolve the interior flow field to capture the turbulence generated by the nozzle and its built-in components. The accurate prediction of such a turbulent flow requires a scale resolving large-eddy simulation (LES) using a high mesh resolution to cover the essential part of the turbulence energy spectrum.

In this study, a Cartesian hierarchical mesh with local refinements is used to simulate the inner and outer flow of three nozzles with varying geometric complexity. Besides a clean nozzle, one nozzle configuration contains a centerbody and another configuration consists of a centerbody plus struts such that compared to the clean configuration without any built-in components additional shear layers and wakes are generated. These flow features will influence the velocity profile, turbulence intensity etc. at the nozzle exit. Inside the nozzle an isotropic mesh, i.e., a constant refinement level is used to resolve the fully turbulent flow. A mesh resolution study with a baseline and a fine mesh resolution is presented which allows to quantify the effect of the nozzle wall resolution on first and second moment profiles at the nozzle exit and to show the impact of the resolution on the turbulent mixing and the jet development. Unlike previous jet studies in which a laminar to turbulent transition was excited by a forcing of the jet, see e.g [6], [16], [21]., a fully turbulent flow is obtained at the nozzle exit such that no forcing of the shear layers is necessary. At the inlet boundary of the nozzle, inflow conditions defining the flow downstream of the last turbine stage are imposed [33].

This paper is structured as follows. First, the numerical method is presented. Then, the numerical method is validated by a single, a coaxial, and a chevron nozzle jet problem for which reference data are available from the literature. Then, the various jet flow configurations are defined. The discussion of the results starts with a grid convergence analysis before the flow fields of the three nozzle geometries, i.e., the clean nozzle, the centerbody nozzle, and the centerbody-plus-strut nozzle, are juxtaposed. Finally, some conclusions are drawn and an outlook is given.

Section snippets

Governing equations

The governing equations are the non-dimensional Navier–Stokes equations for unsteady, compressible flow VQtdV+AH¯·ndA=0,where n is the normal vector of the surface dA, Q=[ρ,ρu,ρE]T is the vector of conservative variables with the density ρ, velocity vector u, and the total specific energy E=e+u2/2 containing the specific energy e. The flux vector H¯ contains the inviscid H¯i and the viscous part H¯vH¯=H¯i+H¯v=(ρuρuu+pu(ρE+p))+1Re0(0τ¯τ¯u+q).The Reynolds number is defined by the fluid

Validation

To validate the solution method a single cold jet, a coaxial hot jet, and a cold jet exhausting from a chevron nozzle are computed and the results are compared with data from the literature [2], [3], [10], [21], [24], [35], [36], [38]. Computational details of the validation studies are summarized in Table 1.

Bogey and Bailly [3] studied the effect of the initial shear layer thickness and the forcing on the jet development. They observed that an increase of the shear layer thickness at the

Flow parameters

The flow parameters are alike for the various nozzle configurations. The Reynolds number based on the jet inlet diameter Di and the mean velocity Ui=116[m/s], where the subscript ‘i’ denotes the nozzle inlet condition with an air mass flux at a temperature Ti=862.64[K], is ReDi=ρiUiDiηi=7.5×105. The density is ρi=0.409[kg/m3] and the dynamic viscosity is ηi=3.79×105[kg/ms]. The Mach number at the inlet is M=Uia=0.341 based on the jet inlet velocity Ui and the speed of sound a. The main flow

Results

In the following, the results of the flow fields of the three nozzle configurations, i.e., the clean nozzle, the centerbody nozzle, and the centerbody-plus-strut nozzle, are discussed in detail. First, a grid convergence study is presented for the centerbody nozzle configuration to determine a mesh resolution which is sufficient to capture the major length and time scales. Then, instantaneous and mean flow fields of the various nozzle configurations are analyzed to assess the influence of the

Conclusions

The internal and external flow fields of three helicopter engines were computed by large-eddy simulations. The solution method was validated for a single cold jet, a coaxial hot jet, and a chevron nozzle jet at a Mach number of M = 0.9 and a Reynolds number of Re = 400,000. The overall comparison of the results showed a satisfactory agreement reference data from the literature. The three nozzle geometries consisted of a clean reference nozzle, a configuration with a centerbody, and a geometry

Acknowledgments

The research was funded from the European Community’s Seventh Framework Programme (FP7, 2007–2013) PEOPLE program under the grant agreement No. FP7-290042 (COPAGT project). The authors gratefully thank the Gauss Centre for Supercomputing (GCS) for providing computing time for a GCS Large-Scale Project on the GCS share of the supercomputer JUQUEEN [20] at the Jülich Supercomputing Centre (JSC) and High Perfomance Computing Center Stuttgart (HLRS). GCS is the alliance of the three national

References (38)

  • C.K.W. Tam

    Computational aeroacoustics-issues and methods

    AIAA J

    (1995)
  • A. Uzun et al.

    Some issues in large-eddy simulations for chevron nozzle jet flows

    J Propuls Power

    (2012)
  • XiaH. et al.

    Large-eddy simulations of chevron jet flows with noise predictions

    Int J Heat Fluid Flow

    (2009)
  • XiongJ. et al.

    Aerodynamic performance of fan-flow deflectors for jet-noise reduction

    J Propuls Power

    (2012)
  • V.H. Arakeri et al.

    On the use of microjets to suppress turbulence in a Mach 0.9 axisymmetric jet

    J Fluid Mech

    (2003)
  • C. Bogey et al.

    Effects of inflow conditions and forcing on subsonic jet flows and noise.

    AIAA J

    (2005)
  • C. Bogey et al.

    Influence of nozzle-exit boundary-layer conditions on the flow and acoustic fields of initially laminar jets

    J Fluid Mech

    (2010)
  • C. Bogey et al.

    Noise investigation of a high subsonic, moderate Reynolds number jet using a compressible large eddy simulation

    Theor Comput Fluid Dyn

    (2003)
  • G.A. Brés et al.

    Unstructured large eddy simulations for nozzle interior flow modeling and jet noise predictions

    AIAA Paper

    (2014)
  • Cited by (14)

    • Optimum structure of a laval nozzle for an abrasive air jet based on nozzle pressure ratio

      2020, Powder Technology
      Citation Excerpt :

      It is an attractive technology due to its distinct advantages including small thermal effect, small cutting forces, high machining versatility, and high flexibility [1–3]. Additionally, an abrasive air jet is widely used in areas such as material surface finishing, surface treatment, drilling, and grooving [4,5]. On the other hand, in the exploitation of coalbed methane, because coal is a multi-media of pores and fractures, the drilling fluid is likely to enter and have physical and chemical interaction with coal by conventional mining methods, thereby changing the strength of the rock, destroying the stability of the wellbore, and affecting the mining efficiency.

    • Influence of grid resolution on the spectral characteristics of noise radiated from turbulent jets: Sound pressure fields and their decomposition

      2020, Computers and Fluids
      Citation Excerpt :

      Experiments [23–28] have shown that serration modification to the round nozzle can bring as much as 6 dB reduction in peak noise during take-off with less than 0.5% thrust loss during cruise. Large-eddy simulations [29–33] have revealed the fundamentally different mechanisms of shear layer growth, with increased mixing caused by the nozzle geometry. Chevrons were also found to reduce the jet and wing/flap interactions [34].

    • Numerical analysis of the impact of exit conditions on low Mach number turbulent jets

      2017, International Journal of Heat and Fluid Flow
      Citation Excerpt :

      That is, the flow field was redirected by introducing built-in components inside the nozzle where they created multiple-shear-layer flow and modified the turbulent flow field which resulted in an earlier potential core collapse. Cetin et al. (2016) numerically analyzed the impact of the nozzle built-in components on the jet flow development. Their analyses for the various nozzle configurations, i.e., a clean nozzle, a centerbody nozzle, and a centerbody-plus-strut nozzle showed a similar flow development and spectral distributions in the outer free-shear layer.

    • Hazel Hen – leading HPC technology and its impact on science in Germany and Europe

      2017, Parallel Computing
      Citation Excerpt :

      The acoustic field is subsequently predicted in a second computational aeroacoustics step by solving the acoustic perturbation equations using unsteady source term data from the unsteady turbulent flow field. More details of the computational methods can be found in [6,7]. Fan industry increasingly demands for quieter and more efficient axial fans in a wide range of applications.

    View all citing articles on Scopus
    View full text