166 



ePULL SCALE MEASUREMENTS 

 + MODEL EXPERIMENTS 



1-6 



1-5 



1-4 



1-3 



^ 1-2 



l-l 



o/o PROPELLER RADIUS 

 FROM HULL. 



12-5 



MODEL PREDICTIONS 



250 



FULL SCAI 

 PREDICTIONS 



l-O 



0-9 



25 /, 



FULL SCALE 

 PREDICTIONS. 



12-5 >25F - = f|7^- -.MODEL PREDICTIONS 



FULL SCALE 

 TUNPOWERED PROPULSION 



o L_^^_l I I — I 



MODEL SELF 

 PROPULSION 



J I 



= u 

 2> 



1-0 l-l 1-2 1-3 1-4 

 PROPELLER DIFFUSION RATIO 



FIGURE 16. Relative magnitude of the secondary flow 

 and velocity defect for vehicle A. 



inflow to the propeller at model and full-scale; 

 however, it should not be regarded as a general 

 conclusion on the basis of this one experiment. 

 Additionally although the propeller inflow may be 

 similar at self-propulsion the propeller thrust 

 loading, as indicated by the diffusion ratio, will 

 be different. 



Comparison between the wake defect and secondary 

 flow model measurements for the two vehicles 

 (Figures 16 to 18) show generally similar magnitudes 

 for the former, but a much larger secondary flow in 

 the case of the body with the fuller afterbody. 



The latter effect can also be seen in the velocity 

 profiles given in Figures 8 and 11. 



Comparison between Predicted and Measured Results 



It can be seen from Figure 14 that the mean circum- 

 ferential velocity predictions for the powered model 

 of vehicle A are always higher than measured. The 

 maximum differences occur at model self-propulsion 

 conditions and are 7 percent and 4 percent for the 

 positions 12.5 percent and 25 percent of the pro- 

 peller radius from the hull respectively. Both 

 the measured data and the predicted velocities can 

 be seen to vary linearly with propeller diffusion 

 ratio. For the blunter stern. Figure 15 indicates 

 that for radial positions between 23 percent and 44 

 percent of the propeller radius from the hull the 

 predictions of mean circumferential velocity are 

 generally in good agreement with the measured data. 

 For the two outer radii the predictions tend to 

 be high as in the case for body A, the maximum 

 errors at model self-propulsion being of the order 

 of 4 percent. However, for the innermost radial 

 position, the powered predictions are up to 14 per- 

 cent below the measured values. It is apparent 

 for Figure 15 that, in contrast to the other radii, 

 the model results for this position are not linear 

 with propeller diffusion ratio because of the low 

 velocity obtained in the unpowered condition. Since 

 the measured data was linear at a similar radial 

 position for body A this suggests that the poor 

 powered prediction of velocity for body B is due to 

 the low unpowered velocity measurement which is 

 used as the datum for the prediction. This low 

 measured velocity may be the result of flow separa- 

 tion on the vehicle with the blunt afterbody which 

 is suppressed by the favourable pressure gradient 

 produced when the propeller is operating. 



Comparison between the full-scale and predicted 

 mean circumferential velocities in Figure 14 show 

 the latter to be less accurate than for the model 

 case, the predicted values being 15 percent and 9 

 percent high for the inner and outer positions 

 respectively. However, correlation of propulsion 

 data from sea trials and model experiments on 

 vehicle A suggest an equivalent full-scale hull 

 Reynolds number of one-tenth of the true value and 



TABLE 1 Experimental and Trial Conditions 



Hull 



Reynolds Diffusion 

 Vehicle Conditions Number Ratio Remarks 



Model 



1.3 X 10^ 1.308 

 1.226 

 1.130 



Self propulsion 



A 

 B 



Trial 



5.5 X 10° 1.160 



Self propulsion 



Model 



1.2 X 10' 1.233 

 1.175 

 1.111 



Self propulsion 



