Data Set Description


THREE DIMENSIONAL BOUNDARY LAYER APPROACHING A WEDGE


Data Originators: Shawn D. Anderson and John K. Eaton
Department of Mechanical Engineering, Stanford University

Primary References: 

Anderson, S.D. and Eaton, J.K. "An Experimental Investigation of Pressure Driven Three-Dimensional Boundary Layers, Rept. MD-49, Thermosciences Division, Dept. of Mechanical Engineering, June 1987.

Anderson, S.D. and Eaton, J.K. "Experimental Study of a Pressure-Driven, Three-Dimensional, Turbulent Boundary Layer," AIAA J. Vol. 25, August 1987, pp. 1086-92.

Anderson, S.D. and Eaton, J.K. "Reynolds Stress Development in Pressure-Driven Three-Dimensional Turbulent Boundary Layers," J. Fluid Mech., Vol 202, 1989, pp. 263-294.

Abrahamson, S.D. and Eaton, J.K. "Heat Transfer through a Pressure-Driven Three Dimensional Boundary Layer," AIAA/ASME Thermophysics and Heat Transfer Conference, Seattle, June 1990.

General Description of Experiment:

An initially two-dimensional boundary layer was skewed and distorted by the pressure field ahead of an upstream-facing wedge as sketched in Figure 1.  This produces a strongly three dimensional boundary layer and separation ahead of the wedge tip.  Extensive measurements of the mean velocity and pressure fields document the overall flow development.   Previous experience suggests that a computation of the entire field using the RANS equations is required as opposed to boundary-layer calculations.  Therefore, the full field measurements are suggested as the primary test data.  Additional data including the Reynolds stresses, the skin friction, and the heat transfer coefficient are documented along the centerline and one selected streamline.

Geometry Specification:

The test section consists of two parallel planes separated by 11.9 cm.  All of the measurements were made on one of the plane walls designated as the "test wall".  The low-momentum region of the boundary layer on the opposite wall was removed by a two-dimensional scoop located at X = -14.5 cm.

The origin of coordinates is the centerline of the test wall at the beginning of the test section.  The total wedge angle is 90 degrees and it is positioned symmetrically so the freestream is turned by 45 degrees.   The wedge tip is located at X = 51.8 cm and Z = 0 cm.  The wedge extended to X = 121.9 cm and Z = +/- 68.6 cm.  The pressure field was also strongly affected by the position of the fairings.  The fairings were set by cutting a template to match a fourth-order polynomial then pushing the fairings into position against the fairing.  The polynomial was:

Z = A + BX + CX^2 + DX^3 + EX^4

where Z and X are in centimeters and:

			A = 30.004 cm
			B =  0.018957
			C = -0.0036287/cm
			D =  0.00020397/cm^2
			E = -0.00000119/cm^3

The fairings extended beyond the end of the wedge.

Initial Conditions:

The first measurement station for the test-wall boundary layer was at X = 7.6 cm.   The mean velocity data at this station are contained in the file 0305D1.DAT and attached as Table 1.  Turbulence data at the same position are contained in file 11TCUA.110 and attached as Table 2.  The boundary layer is mildly three dimensional by this point so the calculation should probably be started upstream.  The opposite wall boundary layer initial condition was measured at X = 2.0 cm.  The boundary layer thickness (delta99) was 2.64 cm, the momentum thickness Reynolds number 2654 and the skin friction coefficient 0.00388 at this point.  The tunnel geometry prevented easy access to the near-wall region of the boundary layer on the fairing.  Initial boundary layer measurements on the fairing were difficult.  Outer layer measurements indicated a boundary layer thickness of approximately 5 cm.

Data Tables:

The primary measurements were all made using probes inserted through seven spanwise slots in the opposite wall of the test section.  These slots are referred to as S1-S7 in the data tables and figures.  The axial positions of the slot are shown in the following table.
------------------------------------------------------------
Slot Position

			S1	 S2	  S3	   S4    S5	S6	 S7  
          X=  7.6   22.9  38.1  45.7  53.3  61.0  68.6 cm.
------------------------------------------------------------

Static Pressure:
The static pressure data are presented as the pressure coefficient.  The data are referenced to the pressure at the centerline at X = 7.6 cm and normalized by the dynamic pressure at that same point.  The data are contained in a single data file NCPC1.DAT.  The data file is printed a Table 3 and plotted as Figure 2.

Mean Velocity Profiles:
All velocity data are presented in a raw form, that is no normalization has been applied.  An upstream reference velocity was recorded with each measurement and may be used to normalize the data.  Generally, the reference velocity was held constant within less than 1% by the automatic wind tunnel controller so no normalization is needed.  Mean velocity profile data were acquired using a conventional three hole probe along the centerline and along a freestream streamline beginning at X = 0 and Z = 15 cm.  The file names for each profile are shown in Table 4.  The file extension for every file is .DAT.  The data formats are the same as that for the initial velocity profile shown as Table 1.  The actual positions of the profiles are compiled in Table 5.

Mean Velocity Maps:
Full planes of two-component mean velocity data were acquired at 5 different Y values.  The data are contained in 35 files as shown in Table 6.  The file name extension for each file is .DAT.  A typical plane of data is shown in Figure 3.  The angle DELTA reported for each measurement is yaw angle relative to the x axis.

Skin Friction:
The skin friction was measured using a rotatable surface fence and was also calculated on the centerline by fitting to Coles velocity profile.  The agreement between the two techniques was excellent.  The skin friction data are compiled in Table 7 and plotted in Figure 4.  The data in the table are presented in dimensional form. The data in the figure are normalized by the local freestream dynamic pressure.

Heat Transfer Data:
The heat transfer coefficient was measured on a constant heat flux heat transfer surface.  A preheater plate was installed in the development section to produce a developed thermal boundary layer at the test section entrance.  The preheater plate began at x = -120 cm and was operated at the same heat flux as the test plate installed in the test section.  The typical heat flux was 630 W/m2.  The heat flux was slightly smaller near the wedge but the variation was not significant.  The main heat transfer data are compiled in Table 8.  The dimensional temperature profiles measured at the same positions as the velocity profiles along the selected streamline are contained in the file CASE1T.DAT.