Nowaekt and Sharma 
was needed for calculating the combined potential and wave thrust 
deduction force 6,Ry,,, see Equation (B 65), from which followed 
the thrust deduction fraction (th + t,,) by Equation (B 66). The po- 
tential component t, alone was obtained from a simple degenerate 
case (zero Froude number wake) of this calculation, see remark fol- 
lowing Equation (B 67). The final results of this calculation are 
shown in Fig. 30 in comparison to the measured total thrust deduction 
t- replotted from ‘Fig. 25: 
Let us try to interpret the salient features of Fig. 30. First, 
the wave component of thrust deduction t,, is small, but not negligi- 
ble compared to the potential component t,. Second, the oscillations 
in calculated thrust deduction are not due to t,, , but are already 
present in t, . This can be understood by reference to Equation (B 63) 
which defines thrust deduction as the Lagally force on the hull sources 
due to the axial flow induced by the propeller sources. Since our hull 
sources were assumed independent of Froude number and since the 
flow induced by a source upstream of itself is almost monotonic with 
Froude number, the observed oscillations of calculated thrust deduc- 
tion can only be due to variations of propeller source strength with 
Froude number. This is indeed the case, for by Equation (B 15) the 
source strength depends on loading and wake, which were both found 
to oscillate with Froude number, Asa result the calculated thrust 
deduction t, (as wellas t,,) correlates strongly with advance coef- 
ficient J,;, and effective wake wp (compare Fig. 25). Third, the 
oscillations in the measured thrust deduction t are much stronger 
than in the calculated (t, + t,). This means that either the residual 
viscous component of thrust deduction t, , see Equation (2) , oser= 
lates appreciably with Froude number or that our assumption of the 
hull sources being independent of Froude number was invalid. This 
point cannot be decided at the moment. But in any case it points toa 
significant interaction of viscous and wave effects at the stern, pre- 
sumably intensified by propeller suction. For instance, if the line of 
boundary layer separation is pulled rearward by the propeller, the 
result would be a negative viscous thrust deduction as well as a rela- 
tive increase in the effective sink strength of the afterbody. Fourth, 
specifically the steep variation of measured thrust deduction around 
Yo = 5 cannot presently be explained, except as a possible viscous 
effect, i.e. a reduction in the extent of boundary layer separation 
under the combined influence of a negative wave wake (Fig. 28) anda 
high propeller loading (Fig. 25). Fifth, the thrust deductions calculat- 
ed from the Hough and Ordway sink disk are significantly higher than . 
those calculated from the Dickmann sink disk and are in better agree- 
ment with measurements. This is a direct consequence of the signifi- 
cant difference between the two sink disks, both in average intensity 
and in its relative distribution over propeller radius, see Fig. 29. 
1868 
