The measured values of the mean velocity components are listed in Tables 3 



through 9 along with other measured quantities. Tables 3 through 9 give the 



measured data along the 0, 67, 80, 83, 86, 87, and 90-degree planes, respectively, 



for various axial locations along the model. The velocity components are non- 



dimensionalized by the free-stream velocity U . As shown in the tables, the mean 



o 



axial velocity is the largest of the three measured components. Measured mean 

 velocity profiles in the x and n directions are shown in Figures 7a through 7c. 

 Each figure presents the profiles at various angular positions for a particular 

 axial location. Figure 7a shows that the mean axial velocity profiles vary only 

 slightly with angular position on the model at x/L = 0.719. Also, the boundary 

 layer is thin, with an overall thickness of less than 1 in. Little variation in 

 the normal velocity component is noted. Examining the profiles further aft on the 

 model, the boundary layer thickens with increased angular position. Little varia- 

 tion in profile occurs for angles less than or equal to 80 degrees. However, 

 profiles between 80 and 90 degrees become increasingly fuller with increased angular 

 location. From repeated measurements, the accuracies of the experimental measure- 

 ments of u /U and v /U are estimated to be about 0.5 percent and 1.0 percent, 

 X o n o ^ 



respectively. 



Comparisons of the measured and predicted mean axial velocity profiles are 

 shown in Figure 8 at selected positions along the model. The circular symbols 

 represent the "X" hot-film measurements and the solid curves represent the theoreti- 



o ^ 



cal results of the C K method using the displacement body concept. Calculations 



2 

 using the C K computer code were made using the initial velocity profiles generated 



within the computer code. Calculations were begun at 1.5 percent of the body length 



with the transition located at 3 percent of the body length. Use of a limited, 



discrete set of offsets to define the model for computational purposes forced the 



2 

 use of this transition location. As shown in Figure 8a, the C K method, used with 



and without the displacement body, predicted the same profile at x/L = 0.719 and 



degrees. For the basic body geometry, prior to using the displacement body concept, 



2 

 the C K method experienced excessive boundary-layer growth and aborted prematurely, 



giving no predictions for axial locations x/L >^ 0.81 and angles greater than 80 

 degrees. The agreement between the computed and measured mean axial velocity pro- 

 files is good at x/L = 0.719 and degrees where the boundary layer is thin. 



12 



