Supersonic Base Flow

Felipe Lenzi Rocha

Queen’s University Belfast
06/05/2014
 Overview of the presentation
• Introduction;
• Numerical Methods;
• Discussion;
• Conclusion;
• Bibliography;
 Introduction
• What is the supersonic base flow?
It is the interaction between a fluid with a mach number greater
than one and a fluid with a mach number lower than one;

• Why is it important to study?

The understanding of this fenomena is important when a new
missile, projectile or launcher is been developed;

• Use of numerical methods to simulate the supersonic base flow;

 Introduction
Problems related with the supersonic base flow:
• Dramatic decrease of the pressure at the base;
• Increasing base drag;
• Instability;
• Shock waves;
• Poor knowledge of unsteady features of base flows;
• Difficulty to have reliable experimental data;
• Unsteadiness of the recirculation zone;
Figure of ”Effects of blunt trainlind edge flow discharge in supersonic regime” from B.H. Saracoglu, G. Paniagua, J. Snachez
and P. Rambaud.
 Numerical Methods
Reynolds-averaged Navier-Stokes (RANS)
• Advantages:
 Lower computational cost;
 Predict attached flows very well with a low computational
cost;

• Disadvantages:
 Great dependence toward turbulence modeling;
 Do not predict the mixing layer expansion satisfactory;
 Have been derived from the imcompressible form;
 Numerical Methods
Large-Eddy Simulation:
• Advantages:
 Well suited for prediction of free shear flows;
 Low computational cost compared with dynamic
models;
 Can predict separated flows more accurately;
• Disadvantages:
 Higher computational cost compared with RANS;
 Numerical Methods
Zonal Detached Eddy Simulation (ZDES)
• Advantages:
 Has given encouraging results for a wide range of
flow configurations exhibiting massive separations;
 Grid refinement occurs only in the regions of interest
such as the separated area in the case of base flows;
• Disadvantages:
 Higher computational cost;
 Discussion
• Simon, Deck and Guillen developed an
experiment to investigate the supersonic base
flow;
• The body used in their experiment was a
cylinder;
• Unstady simulations have been performed with
ZEDS, RANS and LES;
• Thus, the behavior of such methods can be
investigated and compared;
 Discussion
Data of the experiment:
• M = 2.46;
• U = 593.8 m/s;
• Re = 45x10^(6);
• P = 31415 Pa;
• T = 145 K;
• R = 31.75 mm
• Two different meshs [ Nxyz 5x10^6 (M1) and Nxyz
13.5x10^6 nodes (M2);
 Discussion

Figure from Simon, Deck and Guillen

Characteristic features of supersonic base flows can be evidenced and

points out the unsteady nature of such flows with the presence of
numerous turbulent scales. The separation point is fixed by the
geometry at the corner;
 Discussion

Figure from [2]

 Discussion
• RANS results poorly agree with the experiment;
• RANS results failed to reproduce the shear layer expansion
behind the base partly due to excessive turbulent eddy
viscosity;
• In the LES calculation, the incoming boundary layer is
responsible for the under prediction of the mixing layer
growth rate;
• As the compression region is approached, the ZDES closely
matches the experimental data. At reattachment, good
agreement is observed except at the maximum value where
a more rounded profile is observed in the experiment;
 Discussion

From Simon et al [2]

 Discussion
What can effect our solution?
• The Incoming Turbulent Boundary Layer;
• The Subgrid Scale Model;
• The Numerical Scheme;
 Conclusion
• ZDES and others RANS/LES simulations of
supersonic base flows have been performed on
two high density grids;
• By comparing the LES and RANS/LES results, the
incoming boundary layer thickness needs to be
properly matched to predict the proper amount
of turbulence production in the near wake;
• Various factors can influence our solution;
 Bibliography
[1] “Effects of blunt trailing edge flow discharge
in supersonic regime” - B.H. Saracoglua, G.
Paniaguaa, J. Sanchez b, P. Rambauda;
[2] “Reynolds-Averaged Navier–Stokes/Large-
Eddy Simulations of Supersonic Base Flow” –
Simon, Deck and Guillen;