It is judged that the dominant cause of the discrepancy is the effect of 
the propeller. 
Based on these results it was concluded that the downstream body 
reduced the mean velocity into the propeller by 5 percent at the self- 
propulsion condition. This is somewhat smaller than the 10 to 14 percent 
reduction in effective wake that was obtained in Reference 2 in which the 
same dynamometer boat was used behind a single screw model hull. It was 
assumed that the 5 percent reduction in the present experiment occurred 
at all conditions at which experiments were conducted. Therefore, after 
the effects of centrifugal force were subtracted from the measured loading 
components as discussed previously, the time-average value per revolution 
of each hydrodynamic loading component was corrected for the effect of the 
downstream body as follows: From the measured hydrodynamic blade thrust 
CE ) and hydrodynamic blade torque (M ), effective advance coefficients 
H 
based on thrust identity (Jp) and torque identity (J) were deduced from 
the open-water data (Figure 11). These values were multiplied by (1/0.95) 
to obtain corrected values of Jn and Jos i.e., without the downsteam body. 
The corrected values of La and M were then obtained from the open-water 
H 
data at the corrected advance coefficients Jr and Jae respectively. It 
was assumed that the downstream body did not affect the radial center of 
thrust F_ and tangential force F_. Therefore, 
Vy 
M corrected = (F_ corrected/F measured) (M_ measured) 
Vy aH Vy 
2 
-F corrected corrected/M measured) (F_ measured) 
Vy *y aH vy 
The spindle torque (M, ) was corrected by the same procedure as used 
for F_ and Gos using unpublished hydrodynamic spindle torque data for 
the DD-963 propeller. No corrections were made to Ee for the effect of 
H 
the downstream body; however, BD is very small for all experimental con- 
H 
ditions, as discussed later. 
No correction for the effect of the downstream dynamometer boat was 
made to the measured circumferential variation of the loading components. 
30 
