NPARC Alliance Validation Archive
Validation Home   |   Archive   >   Mach 4.5 Turbulent Flow over a Flat Plate

Mach 4.5 Turbulent Flow over a Flat Plate

Introduction

In this case, a turbulent boundary layer is developed from a Mach 4.5 flow over a flow plate. The flow conditions are given in Table 1. The Wind-US simulations used several grids and turbulence models. The Wind-US simulations were compared with experimental data from Coles1-3. The skin friction and boundary layer data are available in Table 2. are compared to computational silumations from Wind-US. Typically, the boundary layer along a smooth, flat plate will transition from laminar to turbulent. It is well known that most CFD codes are unable to predict this transition accurately. The Wind-US solutions are fully turbulent flow, e.g. the boundary layer is solely turbulent and does not transition from laminar to turbulent. The available experimental data includes results for several methods of tripping the boundary layer to turbulent earlier than what would naturally occur.

Table 1. Freestream conditions. 
Mach Pressure (psia) Temperature (deg R) Angle-of-Attack (deg) Angle-of-Sideslip (deg)
4.512 0.0974246 108.841 0.0 0.0

Table 2. Experimental data from Coles1-3.
Skin Friction Data FP4p5_ExpCf.dat
Boundary Layer Data FP4p5_ExpBL.dat

Download All Validation Case Files

A TAR file containing all the files (grids, solutions, results) for cases presented here is available for download. To unpack the files, use the command:

tar -xzvf FP4p5.tz

Table 3. TAR file of all case files.
Wind-US 3.162
FP4p5.tz

Grid

A 2-D, structured grid was created; it is pictured in Figure 1. The grid consisted of 46 points in the streamwise (x-) direction and 81 points in the transverse (y-) direction. In the streamwise direction, the grid was packed at x=0, the beginning of the flat plate. In the transverse direction, the grid was packed to a nominal spacing of y+=1.0 along the lower boundary, to ensure that the boundary layer was sufficiently resolved. The viscous wall portion of the grid was 36 inches long.

2-D, structured Mach 4.5 flat plate grid.
Figure 1. 2-D, structured Mach 4.5 flat plate grid.

The 3-D, structured grid was created by extruding the 2-D, structured grid one cell, 12 inches wide, in the normal (z-) direction. The 3-D, structured grid is shown in Figure 2. The 3-D structured grid was saved in the unstructured format, to be run with the Wind-US unstructured solver.

3-D, structured Mach 4.5 flat plate grid.
Figure 2. 3-D, structured Mach 4.5 flat plate grid.

A 3-D, unstructured grid was created using the same boundary spacing as the 2-D and 3-D, structured grids. The unstructured grid kept the quadrilateral prism boundary layer from the structured grid. Triangular prisms were used to fill in the remainder of the flow domain. The grid was clustered at x=0 and x=21.5 (the location of the boundary layer probe). The unstructured grid had a total of 12,808 cells.

Along the lower boundary, the 14 points upstream of x=0 were inviscid and the 31 points downstream of x=0 were viscous.

3-D, unstructured Mach 4.5 flat plate grid.
Figure 3. 3-D, unstructured Mach 4.5 flat plate grid.


3-D, unstructured Mach 4.5 flat plate grid; zoom of leading edge.
Figure 4. 3-D, unstructured Mach 4.5 flat plate grid; zoom of leading edge.

After creating the unstructured grids, CFPART was used to create lines in the grid to enable the most efficient use of the unstructured line solver. The grid files and CFPART input files are included in Table 4.

Table 4. Files for Grid Creation
Description Gridgen File GMAN Script CFPART Script CGD Grid File
2-D, Structured Grid, Structured Solver FP4p5_StrStr.gg N/A N/A FP4p5_StrStr.cgd
3-D, Structured Grid, Unstructured Solver FP4p5_StrUns.gg N/A FP4p5_StrUns.cfpart.inp FP4p5_StrUns.cgd
3-D, Unstructured Grid, Unstructured Solver FP4p5_UnsUns.gg FP4p5_UnsUns.gman.com FP4p5_UnsUns.cfpart.inp FP4p5_UnsUns.cgd

Boundary Conditions

Boundary conditions for this flat plate case are as shown in Figure 5. Note that the first 14 points along the bottom are set to inviscid, such that the leading edge of the plate is at grid point 15 in the streamwise direction along the bottom of the grid. Although only boundary conditions for the 2-D, structured grid are shown, the boundary conditions are the same for the other grids. For the 3-D grids, the inviscid wall boundary condition is applied to the sidewalls (i.e., z=constant).

Schematic of boundary conditions applied to flate plate grid.
Figure 5. Boundary conditions, shown for the 2-D, structured grid.

Computational Strategy

Convergence was determined by monitoring the residuals and two flow field quantities key to this case: the wall skin friction coefficient and the u-component of velocity at x=21.5 inches. The algorithm settings used in the DAT file for these unstructured solver cases are listed in Table 5. Note that settings for the structured solver (i.e., choice of RHS method) were different.

Table 5. Algorithm settings.
Field Str-Str SST Str-Str SST, w/Comp Corr, Press Dilat Off Str-Str S-A Str-Uns SST Str-Uns SST, w/Comp Corr, Press Dilat Off Str-Uns S-A Uns-Uns SST Uns-Uns S-A
Version Wind-US 3.162 Wind-US 3.162 Wind-US 3.162 Wind-US 3.162 Wind-US 3.162 Wind-US 3.162 Wind-US 3.162 Wind-US 3.162
Grid 2-D, Structured 2-D, Structured 2-D, Structured 3-D, Structured 3-D, Structured 3-D, Structured 3-D, Unstructured 3-D, Unstructured
Solver Structured Structured Structured Unstructured Unstructured Unstructured Unstructured Unstructured
Iterations 5,000 5,000 5,000 5,000 5,000 5,000 5,000 5,000
Convergence Order 13 13 13 13 13 13 13 13
Method Default Default Default IMPLICIT UGAUSS LINE EXACT_LHS VISCOUS JACOBIAN FULL CONVERGE FREQUENCY 11 SUBITERATIONS 6 IMPLICIT UGAUSS LINE EXACT_LHS VISCOUS JACOBIAN FULL CONVERGE FREQUENCY 11 SUBITERATIONS 6 IMPLICIT UGAUSS LINE EXACT_LHS VISCOUS JACOBIAN FULL CONVERGE FREQUENCY 11 SUBITERATIONS 6 IMPLICIT UGAUSS LINE EXACT_LHS VISCOUS JACOBIAN FULL CONVERGE FREQUENCY 11 SUBITERATIONS 6 IMPLICIT UGAUSS LINE EXACT_LHS VISCOUS JACOBIAN FULL CONVERGE FREQUENCY 11 SUBITERATIONS 6
CFL 1.0 1.0 1.0 AUTO DECREASE 2 CFLMAX 200 AUTO DECREASE 2 CFLMAX 200 AUTO DECREASE 2 CFLMAX 200 AUTO DECREASE 2 CFLMAX 200 AUTO DECREASE 2 CFLMAX 200
Limiters Default Default Default DQ LIMITER ON RELAX 0.5 DQ LIMITER ON RELAX 0.5 DQ LIMITER ON RELAX 0.5 DQ LIMITER ON RELAX 0.5 DQ LIMITER ON RELAX 0.5
Dissipation Default Default Default TVD BARTH 3.0 TVD BARTH 3.0 TVD BARTH 3.0 TVD BARTH 3.0 TVD BARTH 3.0
Boundaries IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON IMPLICIT BOUNDARY ON
RHS Default Default Default HLLE SECOND, VISCOUS FULL HLLE SECOND, VISCOUS FULL HLLE SECOND, VISCOUS FULL HLLE SECOND, VISCOUS FULL HLLE SECOND, VISCOUS FULL
Gradients Default Default Default LEAST_SQUARES LEAST_SQUARES LEAST_SQUARES LEAST_SQUARES LEAST_SQUARES
Turbulence Model SST SST SPALART SST SST SPALART SST SPALART
Turbulence Corrections none COMPRESSIBLE, PRESSURE DILATATION OFF none none COMPRESSIBLE, PRESSURE DILATATION OFF none none none

Input Files

The options from Table 5 were incorporated into the DAT input files. The input files for the standard SST turbulence model cases listed above are found in Table 6. The other turbulence model options (i.e., Spalart-Allmaras, SST with compressibility correction) are included, but are commented out.

Table 6. Input files.
Case DAT File
Str-Str SST FP4p5_StrStr.dat
Str-Uns SST FP4p5_StrUns.dat
Uns-Uns SST FP4p5_UnsUns.dat

Post-Processing

Several flow quantities were computed for each case: streamwise (u-) and transverse (v-) velocity; wall skin friction; boundary layer thickness, momentum thickness, and displacement thickness; and Reynolds number. Post-processing scripts were created to compute each of these flow quantities. Table 7 includes the CFPOST scripts used to extract and compute some of these flow quantities.

Table 7. Post-processing scripts for SST turbulence model solutions.
Case CFPOST Script Full Post-Processing Script
Str-Str SST FP4p5_StrStr.cfpost.com FP4p5_StrStr.pp.csh
Str-Uns SST FP4p5_StrUns.cfpost.com FP4p5_StrUns.pp.csh
Uns-Uns SST FP4p5_UnsUns.cfpost.com FP4p5_UnsUns.pp.csh

Results

Plots of streamwise (u-) velocity and transverse (v-) velocity through the boundary layer.
Figure 6. Plots of streamwise (u-) velocity and transverse (v-) velocity through the boundary layer.


Plot of streamwise (u-) velocity compared to Coles experimental data.
Figure 7. Plot of streamwise (u-) velocity compared to Coles experimental data1-3.


Plots of skin friction and integrated boundary layer thickness.
Figure 8. Plots of skin friction and integrated boundary layer thickness.


Plots of skin friction compared to Coles experimental data.
Figure 9. Plots of skin friction compared to Coles experimental data1-3.


Plot of L2 residuals for determining convergence.
Figure 10. Plot of L2 residuals for determining convergence.

References

  1. Coles, D., Measurements in the Boundary Layer on a Smooth Flat Plate in Supersonic Flow: I. The Problem of the Turbulent Boundary Layer, Jet Propulsion Laboratory Report No. 20-69, June 1953.
  2. Coles, D., Measurements in the Boundary Layer on a Smooth Flat Plate in Supersonic Flow: II. Instrumentation and Experimental Techniques at the Jet Propulsion Laboratory, Jet Propulsion Laboratory Report No. 20-70, June 1953.
  3. Coles, D., Measurements in the Boundary Layer on a Smooth Flat Plate in Supersonic Flow: III. Measurements in a Flat-Plate Boundary Layer at the Jet Propulsion Laboratory, Jet Propulsion Laboratory Report No. 20-71, June 1953.

Contact Information

This validation test case was performed by Vance Dippold and Keven Lenahan. Contact: Vance Dippold, MS 5-12, NASA Glenn Research Center, 21000 Brook Park Road, Cleveland, Ohio, 44135.


Last Updated: Friday, 27-Jan-2012 10:32:21 EST