HomeIndustriesProductsServicesPartnersAbout Us

Customer Login

Home arrow CFD-FASTRAN User Tips arrow Axisymmetric 2D Convergent-Divergent Boattail Nozzle Simulation Using CFD-FASTRAN
Axisymmetric 2D Convergent-Divergent Boattail Nozzle Simulation Using CFD-FASTRAN

The NASA D-1.22-L boattail nozzle configuration was obtained from the MADIC (Multidisciplinary and Design Industrial Consortium) program. The geometry definition and the flow conditions are documented in NASA TP 1766 [1]. This user tip presents a validation of numerical methods against experimental data.

Geometry Definition 

The model represents a cylinder with a sharp conical front end, tapered blunt aft end and a convergent-divergent internal nozzle. The geometry is discretized using a structured mesh. A 2D axisymmetrical model is build in order to reduce the CPU time requirements. The model contains 3 structured zones. On the solid walls, a boundary layer mesh has been used. The total number of the cells is around 16,600. The boattail nozzle mesh is presented in Figure 1.

Image

Figure 1.  2D Axisymmetric Model

Numerical Results

The initial and boundary conditions used for the NASA D-1.22-L model are the following :

  • adiabatic walls on the solid boundaries,
  • symmetry on the axis of symmetry,
  • fixed static pressure exit condition,
  • inflow/outflow condition on other outermost surface bounding the domain,
  • fixed total pressure and total temperature at the inlet at the nozzle.
Rem  306,000   
M (free stream) 0.8  
T_total (free stream) 592 deg. R
P_total (free stream) 14.71 psia
T_total (nozzle inlet) 592 deg. R
P_total (nozzle inlet) 14.47 psia
gamma 1.4  

The simulation is done using the structured solver of CFD-FASTRAN. The Roe’s upwinding differencing scheme with min mod limitor is used. Steady state solution is obtained using the Point Jacobi fully implicit scheme. For the spatial discretization, high order numerical scheme is used. The Menter SST k omega turbulence model has been used. On figure 2 we presented the Mach number.  

Image

Figure 2. Mach number for NASA D-1.22-L model

The plot below compares the computational results to the experimental data. The comparison is for pressure coefficient along the aft external wall of the boattail nozzle.

Image

  Figure 3. Comparison Experiment (red)/CFD-FASTRAN (black)

We can note a good agreement between numerical prediction and experimental measurements.

Regards,
Daniel Vinteler
CFD Support Manager - France

REFERENCES

[1] NASA TP 1766, Investigation of Convergent-Divergent Nozzles Applicable to Reduced-Power Supersonic Cruise Aircraft, Bobby L. Berrier and Richard J. Re, December 1980.

[2] www.grc.nasa.gov/WWW/wind/valid/madicnoz/madicnoz01/madicnoz01
 

Motion Model Dependencies in CFD-FASTRAN

Moving-body models available in CFD-FASTRAN are highly suited to simulate complex prescribed and six-degree-of-freedom (6DOF) motions of rigid bodies. In many engineering problems, this translates to multiple bodies moving relative to one another.

Low Mach Preconditioning and Dual Time Stepping in CFD-FASTRAN

Density-based schemes employing time-marching procedures available in CFD-FASTRAN provide excellent stability and convergence characteristics for high-speed compressible flows (typically M >0.5).

Improved Residual Plotter in CFD-FASTRAN-GUI

In V2006, the Residual Plotter in CFD-FASTRAN-GUI has been improved to plot all of the output files that CFD-FASTRAN creates. In addition to the global residual history (RSL), CFD-FASTRAN may produce output files with the extensions *.RZ, *.FORCE, *.VFORCE, *.AFS, *.DYNA, *.DYNB, *.KINA, *.KINB, depending on the problem.

Improvements to Parallel Simulation Setup in CFD-FASTRAN

Setting up and launching a parallel simulation has become much simpler and easier in FASTRAN. This note discusses some of these developments. Let’s start with things that have not changed. There are still two versions of CFD-FASTRAN solvers for parallel cases. The difference between the two versions is the underlying parallel communication mechanism, the choice of which is decided by the type of mesh used.

Chemical-kinetic Model for Mars Atmosphere Re-entry Applications

The shock layer flow over a blunt body entering a planetary atmosphere at a hypersonic speed will dissociate and partially ionize. A reliable prediction of the flow-field for such application requires a chemical-kinetic model. For Mars atmosphere, the five species Park'94 is considered [1]. The dissociation of CO2 is producing C, CO, CO2, O and O2.


© 2012 ESI Group CFD Portal