239 



2 4 



U/Uo u= 



6 8 10 



' SHIP, MODEL SPEED 



FIGURE 24. Measured boundary layer velocity profiles 

 for R/V ATHENA and wind tunnel model 5366 with and 

 without propeller at locations 2 and 6. 



and the greatest deviation from the model-scale 

 wake. 



In part, this scatter is also due to the fact 

 that the longitudinal velocity component ratios 

 presented are an average of the longitudinal velocity 

 component as measured in the tangential plane and in 

 the radial plane. Therefore, any scatter error in 

 either the tangential or radial plane measurements 

 will influence the calculation of the longitudinal 

 component. Another factor which probably contrib- 

 uted to increased scatter at the innermost radius 

 is the close proximity of the pitot tube to the 

 strut bossing. 



The longitudinal velocity component ratio at the 

 innermost radius is about 10 percent lower for the 

 ship than for the model, while the peaks of the 

 tangential and radial velocity component ratios are 

 about 10 percent higher for the ship than for the 

 model. Although there are undoubtedly scale effects 

 on the shafting and strut bossing at this radius, 

 another significant factor is that the bossing on 

 the ship is proportionately much longer than on 

 the model. This is due to the collar to which the 

 pitot tube rakes ahead of the struts were attached. 



At the outer radii the longitudinal velocity 

 component ratios for the ship are 2-4 percent lower 

 than those for the model. The peaks of the radial 

 and tangential velocity component ratios at the 

 outer radii are 8-10 percent higher for the ship 

 than for the model. At the two innermost radii, 

 the shift in the radial and tangential velocity 

 component ratios indicate that there is a stronger 

 upf low on the ship than the model , in the region 

 under and outboard of the propeller hub. This 

 effect is much weaker, and has shifted to the inside 



on the two outer radii. One possible cause of the 

 shift at the outer radii is the fact that the full 

 scale trial was performed with a propeller operating 

 on the port shaft, while the model data were col- 

 lected without the propeller present. However, the 

 most likely source of the increased upward flow is 

 a difference in attitude between the ship and model. 



The models were run at a number of Reynolds num- 

 bers in the towing tank and wind tunnel and the 

 longitudinal velocity component was measured at a 

 single location near the hull for these various 

 Reynolds numbers. The results of these measure- 

 ments are plotted in Figure 27. These results in- 

 dicate that for a Reynolds number greater than 10 

 there is very little effect of either Reynolds num- 

 ber or Froude number on the longitudinal velocity 

 component. Therefore, in cases where it is desir- 

 able to obtain accurate longitudinal velocity 

 component measurements , the model should be run at 

 the correct Froude trim, at a Reynolds number 

 greater than 10 . 



A comparison of the boundary layer profiles 

 presented in Figures 20, 21, and 22 shows that, as 

 might be expected, the model velocity profile is 

 not as fully developed as the full-scale velocity 

 profile at Locations 1 and 3. This is clearly a 

 consequence of the one decade difference in Reynolds 

 number between the model and ship. However, at 

 Location 2, the model- and full-scale boundary 

 layer velocity profiles almost coincide. This is 

 clearly an anomalous situation, particularly be- 

 cause even at 0.46 meters from the hull full scale, 

 the velocity has not reached the free-stream 

 velocity, let alone the potential flow velocity 

 which is even higher. The most likely explanation 

 for the low full-scale velocity profile is a mal- 



02 04 06 08 



U/Uo U= • SHIP. MODEL SPEED 



FIGURE 25. Measured boundary layer velocity profiles 

 for R/V ATHENA and wind tunnel model 5366 with and 

 without propeller locations 3 and 7. 



