Turbulence Modeling Resource

The Boeing Bump geometry consists of a surface whose height is a Gaussian function in the streamwise direction, x, and is tapered in the spanwise direction, z, by an error function to give the bump its "shoulders."

3.2.1. Bump Geometry

The following files provide the definition for the bump alone at the test scale (L=0.9144m). The two files represent the test model which was fabricated with one half wider that the other so that the joining seam was off centerline. Therefore, the left-hand side and right-hand side definitions are not exact halves of the configuration. Files are in IGS format.

The as-manufactured geometry was verified with a coordinate measuring machine (CMM). The measured contours were confirmed to be within +-0.125 mm of the CAD model for each of the plate sections. Reports containing the CMM measurements can be downloaded here:

The test article including the splitter plate and bump sections were fabricated with a surface roughness RMS approx 305 micrometer. The steps between plates were measured to be < 0.063 mm. A more detailed report of the panel step and surface finish measurements can be downloaded here:

The wind tunnel experiments were conducted in air at ambient conditions. Although cooling via the turning vanes was utilized during the experiments its purpose was to hold air temperature to within a 1 degree C band within the wind tunnel. Details on the reference conditions within the test section during a specific measurement are documented in the header section of the corresponding data file.

Two aerodynamic parameters were varied in the experiments to understand their influence on smooth body separation. The first was the flow Mach number or Reynolds number which was varied by changing the air speed via the fan RPM. For most measurements the emphasis was on M=0.1 and M=0.2, however, in several cases data were obtained across a larger set of conditions. The following table lists all Mach numbers tested along with the corresponding free steam velocity and Reynolds number (based on the test section width).

The second aerodynamic parameter varied was the boundary layer thickness upstream of the bump. This was varied independently of freestream velocity by placing the bump at different locations on the splitter plate so that the boundary layer had more or less distance to develop. Accordingly, two bump configurations were tested in this experimental study: configuration A where the bump apex was postitioned one test section width downstream from the splitter plate leading edge and configuration B where the bump apex was two test section widths downstream from the leading edge. An emphasis was placed on obtaining data with the bump in configuration A.

All experimental data are provided in human-readable ASCII DAT file format (.dat). The beginning of these files includes a commented block of text which specifies the bump geometry, test configuration, references conditions, derived quantities and variable name descriptions, with units. The data table consists of a single uncommented line with the parameter names followed by the data itself. These DAT files are readable by Microsoft Excel, MATLAB, Python and Tecplot. Additionally, the PIV and Kulite data are also provided in binary MATLAB MAT file format (.mat). The MAT files have smaller file sizes than the DAT files and for the PIV data the MAT file format preserves the 2D array structure whereas the DAT files store the 2D data as 1D arrays which have to reshaped after import. The MAT files can be read by MATLAB and Python.

Two coordinate systems are utilized in the data files. First, a right-handed global coordinate system with respect to the test section is denoted using capital X, Y, and Z to represent streamwise, vertical, and spanwise distances from its origin, respectively. The origin of the global coordinate system is located at the inlet of the test section, on the bottom wall, and at the center span between the left and right side walls. Second, a right-handed bump-based coordinate system is implemented when distances with respect to the bump are referenced, regardless of the global position of the bump apex. Here, lower case symbols x, y, and z represent streamwise, vertical, and spanwise directions, respectively. The origin of the bump-centric coordinate system is in the same streamwise plane as the bump apex, and located vertically at the top surface of the flat plate, and at the center span of the tunnel between the left and right side walls. For data files where the filename includes the position of the measurement the global coordinate system (X,Y,Z) is used. Within the data files the data coordinates are provided in both the global (X,Y,Z) and bump-centric (x,y,z) coordinate systems.
 

5.2.1. Importing .dat files into MATLAB

Generally, the .dat files can be imported into MATLAB by simply entering data=readtable('filename.dat'); in the command window. For some files MATLAB does not correctly handle the commented header section and throws an error. Therefore, a more robust way to import the data is to use the following commands in an m-file.

  infile = 'EmptyTunnel_TotalPressureBL_X486mmProfileB_M0p10.dat';
  comment= '%';
  
  S=readlines(infile);
  ch = S{1};, line=1;
  while ch(1) == comment
  	line = line + 1;
  	ch = S{line};
  end
  clear S
  
  data=readtable(infile,'NumHeaderLines',line-1);

Data are then accessed from the structure using data.variablename, e.g., data.U.

PIV data which are obtained as a 2D array will import from DAT files as 1D vectors. The following is an example of how to reshape and plot the SPIV data:

  data = readtable('BumpConfigA_SPIVcrossplanes_X1143mm_M0p200.dat');
  ydim = length(unique(data.y_L)); zdim = length(unique(data.z_L));
  zL = reshape(data.z_L,ydim,zdim);
  yL = reshape(data.y_L,ydim,zdim);
  U = reshape(data.U,ydim,zdim);
  figure
  contourf(zL,yL,U,256,'EdgeColor','none')
  c = colorbar;
  title(c,'$U$ [m/s]','interpreter','latex');
  axis equal
  xlabel('$z/L$','interpreter','latex')
  ylabel('$y/L$','interpreter','latex')

5.2.2. Importing .mat files into MATLAB

MAT files are imported into MATLAB by entering load('filename.mat') in the command window. Data are then accessed from the structure using data.variablename, e.g., data.U.
 

5.2.3. Importing .dat files into Python

The following commands will import the data in the .dat file and preserve parameter names while ignoring commented lines in the header.

  import numpy as np
  from io import StringIO
  
  infile = 'EmptyTunnel_TotalPressureBL_X486mmProfileB_M0p10.dat'
  input_data = open(infile,'r')
  input_text = ''
  comment = '%'
  
  for line in input_data:
      if not line[0] == comment:
          input_text += line
  data = np.genfromtxt(StringIO(input_text),delimiter='\t',names=True)

Data are then accessed from the structure using data['VariableName'], e.g. data['U'].
 

5.2.4. Importing .mat files into Python

  import scipy.io
  mat = scipy.io.loadmat('filename.mat')

5.2.5. Importing .dat files into Tecplot

Open the .dat file in a text editor and note the line number with the variable names and the line number of the first row of the data table.

Within Tecplot click File > Load Data... and navigate to the .dat file. Under Files of type select All Files (*) if not already in that mode and click on the .dat file and then Open. Under loader type select General Text Loader and then OK. Click Variables button and in the variables window select Scan for Variable Names radio button and enter the line number of the variables in both the Start Line and End Line boxes then press OK (the delimiter can be left as Auto). Next click the Data button and in the data window enter the line number of the first row of the data table under Start Line and select End of File radio button for the end line then click OK (delimited can be left as Auto). Under Data Preview select the View Processed Data radio button and then click Scan File. The data table should appear in the display window. Click OK. The data should now be available for plotting in Tecplot under the Mapping Style dialog.
 

5.2.6. Importing .dat files into Microsoft Excel

Within Excel click File > Open and navigate to the .dat file location. Change the file type drop down from All Excel Files to All Files (*.*). Click on the desired .dat file and click Open. Under Original data type select Delimited then click Next. Under Delimiters check only the Tab box then click Next. Under Column data format leave the radio button on General then click Finish.

Alternatively, the .dat file can be opened in a text editor (e.g. notepad, notepad++, etc) and the contents selected and copied. Next within Excel, right click in the cell A1 and select paste. Excel will automatically apply a tab delimiter when pasting the .dat file contents.
 

To achieve the core objective of this program, a series of flow diagnostics were conducted with the speed bump model installed in the splitter plate. A high-level tabular summary of all experimental data obtained with the bump model installed is provided in the following table which shows the flow speeds tested using each of the diagnostic tools. The table also specifies the bump configuration tested at the reference condition (A, B, or both). The quantities measured and corresponding experimental techniques employed include: the boundary layer profiles using the hot-wire anemometry system (HW), the flow fields upstream of the bump using SPIV (SPIV), the surface streamlines in the separated flow region using flow visualization oil (FlowViz), the mean and instantaneous static pressure over the bump (Cp and Kulite, respectively), the mean skin friction over the bump (Cf ), the separated flow field downstream of the bump using PIV (PIV), and the separated flow field downstream of the bump in cross-planes using SPIV (SPIVcross). Additional data summary tables are provided at the start of each of the following subsections which indicate the specific locations where data were obtained within the flow field for the measurement type being discussed.

Section 2.3 of the test report provides information on the coordinate systems and notation for the bump configurations tested. See Section 4.1 of the test report for details on the experimental test conditions.
 

5.5.1. Hot-wire: Upstream Boundary Layer Profile Development

The hot-wire anemometry and traversing system described in Section 2.4.3 of the test report was used to obtain mean velocity and turbulence profiles of the boundary layer upstream of the bump apex on centerline. See Section 4.2.1 in the test report for more information on the data obtained.

The following table outlines where hot-wire data were obtained and for what conditions:

The following figure visually depicts where the hot-wire data were obtained relative to the bump:

A zip file containing the hot-wire data obtained upstream of the bump can be downloaded here:

• HotwireBoundaryLayerProfiles.zip

 

5.5.2. Stereoscopic PIV (SPIV): Upstream Flow Field Development on Centerline

The upstream SPIV setup described in Section 2.4.6.2 of the test report was used to measure the mean velocities and Reynolds stresses upstream of the bump apex. See Section 4.2.2 in the test report for more information on the data obtained.

The following table outlines where SPIV data were obtained and for what conditions.

The following figures visually depict where the SPIV data were obtained relative to the bump.

A zip file containing the SPIV data obtained upstream of the bump can be downloaded here:

• SPIVupstream.zip

 

5.5.3. Fluorescent Oil Flow Visualization: Bump Surface Streamlines Within the Separation

The fluorescent oil mixture described in Section 2.4.1 was applied to the downstream region of the bump to visualize the time-mean surface streamlines produced by the separated flow. See Section 4.3 in the test report for more information on the images obtained.

The following figure is an example fluorescent oil flow image obtained at M=0.2 for Configuration A. In the image the apex of the bump is indicated by the annotation and the bump centerline is delineated with three parallel orange lines.

The location of the surface foci at M=0.2 for Configuration A were extracted by applying a photogrammetry technique to map the fluorescent oil flow image onto the 3D bump surface. The foci locations are provided below in local bump-relative coordinates and have an estimated uncertainty of +/-2mm, based on how the center of the foci were visually identified.

F1: (x,z) = (124.7, 113.8) mm
F2: (x,z) = (129.3, -109.6) mm

Fluorescent oil flow images are provided for both configuration A and B at M=0.1 and M=0.2. A zip file containing the images can be downloaded here:

• OilFlowVis.zip

 

5.5.4. Pressure Taps: Bump Surface Static Pressure

The static pressure taps described in Section 2.4.7.1 of the test report were used to measure the mean surface pressure field along the bump. The ports were aligned on five arrays, three arrays in the streamwise direction and two arrays in the spanwise direction. See Section 4.4.1 in the test report for more information on the data obtained. The following table outlines where the static pressure data were obtained and for what conditions.

The following figure visually depicts where the static pressure data were obtained on the bump.

A zip file containing the static pressure data obtained on the bump can be downloaded here:

• SurfacePressure_timeaverage.zip

 

5.5.5. Kulites: Bump Surface Dynamic Pressure

The Kulite dynamic pressure sensors described in Section 2.4.7.2 of the test report were used to measure instantaneous pressure fluctuations downstream of the bump apex. See Section 4.4.2 in the test report for more information on the data obtained. The following table outlines where the dynamic pressure data were obtained and for what conditions.

The following figure visually depicts where the Kulite sensors were installed on the bump.

A zip file containing the dynamic pressure data obtained on the bump can be downloaded here:

• SurfacePressure_instantaneous.zip

 

5.5.6. Oil Film Interferometry (OFI): Bump Surface Skin Friction

The photogrammetric OFI technique described in Section 2.4.4 in the test report was used to measure the skin friction coefficient along the bump. See Section 4.5 in the test report for more information on the data obtained. The following figure depicts an example distribution of OFI measurements along the splitter plate and bump.

The OFI measurements were made at M=0.05, 0.1 and 0.2 for bump configurations A and B. A zip file containing the skin friction measurements obtained on the bump using OFI can be downloaded here:

• SkinFriction.zip

 

5.5.7 Particle Image Velocimetry (PIV): Downstream Flow Field on Streamwise Planes

The PIV setup described in Section 2.4.5 of the test report was used to measure two-component mean velocity and Reynolds stresses in the separated flow region downstream of the bump apex. See Section 4.6 in the test report for more information on the data obtained.

The following table outlines where the downstream PIV data were obtained and for what conditions.

The following figure visually depicts on what planes PIV data were obtained on the downstream side of the bump.

A zip file containing the downstream PIV data can be downloaded here:

• PIV.zip

 

5.5.8 Stereoscopic PIV (SPIV): Downstream Flow Field on Cross-Planes

The cross planar SPIV setup described in Section 2.4.6.1 was used to measure three-component mean velocity and Reynolds stresses downstream of the bump apex. See Section 4.7 in the test report for more information on the data obtained.

The following table outlines where the downstream SPIV data were obtained and for what conditions.

The following figure visually depicts on what planes SPIV data were obtained on the downstream side of the bump.

A zip file containing the downstream SPIV data can be downloaded here:

• SPIVcrossplanes.zip

 

---

 

6. List of Related CFD Papers

 

---

The following is a list of known CFD papers that report scale-resolving computations of the SBSE configuration (including both the full configuration as well as a centerplane or center-region "slice").

DNS (Direct Numerical Simulation) Papers:

• Uzun, A., Malik, M. R. (2021). Simulation of a turbulent flow subjected to favorable and adverse pressure gradients. Theoretical and Computational Fluid Dynamics, 35, 293-329, https://doi.org/10.1007/s00162-020-00558-4.
  DNS data from this paper is provided on the page: DNS: Gaussian Bump at Two Reynolds Numbers
• Shur, M. L., Spalart, P. R., Strelets, M. K., Travin, A. K. (2021). Direct numerical simulation of the two-dimensional speed bump flow at increasing Reynolds numbers. International Journal of Heat and Fluid Flow, 90, 108840, https://doi.org/10.1016/j.ijheatfluidflow.2021.108840.
• Balin, R., Jansen, K. E. (2021). Direct numerical simulation of a turbulent boundary layer over a bump with strong pressure gradients. Journal of Fluid Mechanics, 918, A14, https://doi.org/10.1017/jfm.2021.312.
• Uzun, A., Malik, M. R. (2022). High-fidelity simulation of turbulent flow past Gaussian bump. AIAA Journal, 60(4), 2130-2149, https://doi.org/10.2514/1.J060760.
  DNS data from this paper is provided on the page: DNS: Gaussian Bump at Two Reynolds Numbers

WRLES (Wall-Resolved Large-Eddy Simulation) Papers:

• Wright, J. R., Balin, R., Jansen, K. E., Evans, J. A. (2021). Unstructured LES_DNS of a turbulent boundary layer over a Gaussian bump. In AIAA Scitech 2021 Forum (AIAA-2021-1746), https://doi.org/10.2514/6.2021-1746.
• Uzun, A., Malik, M. R. (2022). A Dynamic Nonlinear Subgrid-Scale Model for Large-Eddy Simulation of Complex Turbulent Flows, NASA/TM-20220013891, https://ntrs.nasa.gov/citations/20220013891.
• Rizzetta, D. P., Garmann, D. J. (2023). Wall-resolved large-eddy simulation of flow over a three-dimensional gaussian bump. In AIAA SCITECH 2023 Forum (AIAA-2023-0286), https://doi.org/10.2514/6.2023-0286.
• Rizzetta, D. P., Garmann, D. J. (2023). Wall-Resolved Large-Eddy Simulation of Flow over a Parametric Set of Gaussian Bumps. In AIAA AVIATION 2023 Forum (AIAA-2023-3983), https://doi.org/10.2514/6.2023-3983.
• Rizzetta, D. P., Garmann, D. J. (2023). Arch Vortex in Flow over a Three-Dimensional Gaussian Bump. AIAA Journal, 61(11), 5176-5179, https://doi.org/10.2514/1.J062876.

WMLES/DDES (Wall-Modeled Large-Eddy Simulation/Delayed Detached Eddy Simulation) Papers:

• Balin, R., Jansen, K. E., Spalart, P. R. (2020). Wall-modeled LES of flow over a Gaussian bump with strong pressure gradients and separation. In AIAA Aviation 2020 Forum (AIAA-2020-3012), https://doi.org/10.2514/6.2020-3012.
• Iyer, P. S., Malik, M. R. (2021). Wall-modeled LES of flow over a Gaussian bump. In AIAA Scitech 2021 Forum (AIAA-2021-1438), https://doi.org/10.2514/6.2021-1438.
• Gray, P. D., Gluzman, I., Thomas, F., Corke, T., Lakebrink, M., Mejia, K. (2021). A new validation experiment for smooth-body separation. In AIAA Aviation 2021 Forum (AIAA-2021-2810), https://doi.org/10.2514/6.2021-2810.
• Whitmore, M. P., Griffin, K. P., Bose, S. T., Moin, P. (2021). Large-eddy simulation of a Gaussian bump with slip-wall boundary conditions. Center for Turbulence Research Annual Research Briefs, 45-58, https://web.stanford.edu/group/ctr/ResBriefs/2021/06_Whitmore.pdf.
• Agrawal, R., Whitmore, M., Griffin, K., Moin, P. (2021). Dynamic modeling of non-Boussinesq subgrid-scale models for large-eddy simulations. Center for Turbulence Research Annual Research Briefs, 31-43, https://web.stanford.edu/group/ctr/ResBriefs/2021/05_Agrawal.pdf.
• Prakash, A., Balin, R., Evans, J. A., Jansen, K. E. (2022). Wall-modeled large eddy simulations of a turbulent boundary layer over the Boeing speed bump at ReL= 2 million. In AIAA SciTech 2022 Forum (AIAA-2022-0338), https://doi.org/10.2514/6.2022-0338.
• Iyer, P. S., Malik, M. R. (2022). Wall-modeled LES of turbulent flow over a two dimensional Gaussian bump. ICCFD11 Paper, 204, 2022, https://www.iccfd.org/iccfd11/assets/pdf/papers/ICCFD11_Paper-0204.pdf.
• Agrawal, R., Bose, S. T., Moin, P. (2022). Wall modeled LES of the Boeing speed bump using a non-Boussinesq modeling framework. Center for Turbulence Research, Annual Research Briefs, 43-58, https://web.stanford.edu/group/ctr/ResBriefs/2022/06_Agrawal.pdf.
• Agrawal, R., Whitmore, M. P., Griffin, K. P., Bose, S. T., Moin, P. (2022). Non-Boussinesq subgrid-scale model with dynamic tensorial coefficients. Physical Review Fluids, 7(7), 074602, https://doi.org/10.1103/PhysRevFluids.7.074602.
• Iyer, P. S., Malik, M. R. (2023). Wall-modeled LES of the three-dimensional speed bump experiment. In AIAA SCITECH 2023 Forum (AIAA-2023-0253), https://doi.org/10.2514/6.2023-0253.
• Gray, P. D., Lakebrink, M. T., Thomas, F. O., Corke, T. C., Gluzman, I., Straccia, J. (2023). Experimental and Computational Evaluation of Smooth-Body Separated Flow over Boeing Bump. In AIAA AVIATION 2023 Forum (AIAA-2023-3981), https://doi.org/10.2514/6.2023-3981.
• Arranz, G., Ling, Y., Lozano-Duran, A. (2023). Wall-modeled LES based on building-block flows: Application to the Gaussian Bump. In AIAA AVIATION 2023 Forum (AIAA-2023-3984), https://doi.org/10.2514/6.2023-3984.
• Zhou, D., Whitmore, M. P., Griffin, K. P., Bae, H. J. (2023). Large-Eddy Simulation of Flow over Boeing Gaussian Bump Using Multi-Agent Reinforcement Learning Wall Model. In AIAA AVIATION 2023 Forum (AIAA-2023-3985), https://doi.org/10.2514/6.2023-3985.
• Iyer, P. S., Malik, M. R. (2023). Assessing Reynolds Number Effects for flow over a Gaussian bump using Wall-modeled LES. In AIAA AVIATION 2023 Forum (AIAA-2023-3982), https://doi.org/10.2514/6.2023-3982.

 

---

Acknowledgements

The Boeing bump model geometry was developed by Philippe Spalart (Senior Technical Fellow, the Boeing Company, Retired) in close collaboration with Jeffrey Slotnick (Technical Fellow, the Boeing Company) and the New Technologies and Services (NTS) group under Professor Strelets in St. Petersburg. The wind tunnel experiments were conducted by the University of Notre Dame. Program oversight and accompanying CFD analysis performed by Boeing Research & Technology.

This work was funded through a cooperative agreement between the Office of Naval Research, the National Aeronautics and Space Administration, the Army Research Office, the Air Force Research Lab, and The Boeing Company. Government funding for the work was provided through ONR Cooperative Agreement No. N00014-20-2-1002. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the Office of Naval Research.
 

---

 

Return to: Data from Experiments - Intro Page

Return to: Turbulence Modeling Resource Home Page

 
 

---

Recent significant updates:
02/09/2024 - updated the experimental test report, and added the CFD test report
08/28/2023 - posted final data

Privacy Act Statement

Accessibility Statement

Responsible NASA Official: Ethan Vogel
Page Curator: Clark Pederson
Last Updated: 10/08/2024