240 



1 -Z 

 1 .1 



C.9 

 O.S 

 3.7 

 0.5 

 0.5 

 1 

 3 



a.z 



-0.1 



-0.: 



FIGURE 26. Velocity component 

 ratios for DTNSRDC model 5365 at 

 0.633 radius for model speeds of 

 5.22 knots and 13.5 knots. 



3C 103 120 MO \K 13C COG Z20 240 



ftN&l. E IN DEGREES 



!50 280 303 3J0 3*0 3G3 S'JC 



function in the instrumentation but a check of the 

 data records indicated no obvious errors in the 

 data. 



The results of boundary layer profile measure- 

 ment with the propeller operating, plotted in 

 Figures 23, 24, and 25 indicate that the data at 

 positions 1 and 8, just ahead of the propeller, 

 show a slight increase in velocity profile due to 

 the propeller suction. The increases are about 

 the same at both model- and ship-scale . The data 

 at positions 3 and 7, behind the propeller, show 

 rather significant increases in the velocity pro- 

 file for both scales. This is undoubtedly due to 

 the wake of the propeller. From the model-scale 

 data, at Locations 2 and 6, there is no noticeable 

 difference in the data obtained with or without 

 the propeller operating. This is consistent with 

 the separation between the boundary layer probe and 

 the propeller. There is no ship-scale data ahead 

 of the operating propeller at location 6 due to the 

 failure of that boundary layer probe. 



In order to evaluate our ability to predict the 

 boundary layer of the hull, a series of boundary 

 layer calculations were instituted. For these 

 calculations, the ship was approximated as a body 

 of revolution, and the boundary layer was calculated 

 using the standard DTNSRDC method for bodies of 

 revolution [Wang and Huang (1976) ] . Two methods 

 for generating the bodies of revolution were tested. 

 In one, the body was generated with radii equal to 

 the square root of twice the sectional area of the 



ship; and in the other, the body was generated 

 using circumferences equal to twice the girth of 

 the ship. The boundary layer calculations using 

 the body of revolution based on sectional area 

 agreed best with the experimental data. 



The results of the equivalent body of revolution 

 calculations are plotted with the experimental data 

 on Figures 20, 21, and 22. The calculations for 

 the ship at Locations 1 and 3 agree reasonably well 

 with the full-scale data. However, at the model- 

 scale, the calculations do not agree nearly as 

 well. This is probably due to the fact that at 

 lower Reynolds numbers, the boundary layer is much 

 more sensitive to errors in the flow velocity and 

 pressure gradient than at higher speeds. As stated 

 previously, the data at Location 2 is anomalous, 

 as is shown by a comparison with the calculated 

 boundary layer profile. 



8. PREDICTION OF NOMINAL WAKE 



Although the model- and full-scale wake of the R/V 

 ATHENA both agree qualitatively, there are some 

 substantial quantitative differences between the 

 model- and full-scale velocity components. To 

 develop an understanding of the origins of these 

 differences, it was necessary to predict the wake 

 of both the model- and full-scale ship analytically. 

 Since the hull of the ATHENA showed no separation, 

 it appeared that the presence of the hull could be 



