| 
EXPERIMENTS ON THE FROUDE. 93 
The electrical input to the motor was computed from the readings of 
the voltmeter and the ammeter. From this gross input there was sub- 
tracted (1) an allowance for the electrical losses as determined from the 
rating of the motor in the Electrical Engineering Laboratory and (2) an 
allowance for the friction of motor, propeller-shaft and driving gear from 
the brake tests which were made as stated on page 107. 
It will be noted that two curves are given for both revolutions and 
thrust, one with the wind and one against the wind. It is immediately 
evident that the mean revolution and thrust curves which should represent 
results that would be expected were there no wind, cannot be obtained by 
halving the differences, because the boat received comparatively little 
assistance from the wind when running with it, while there was a consider- 
able resistance when running against the wind; the ordinary convention 
is that the wind resistance varies with the square of the apparent velocity 
of the wind observed from the deck. Making use of this convention the 
probable real velocity of the wind is computed by a process derived by 
Professor Everett and explained in Appendix C; afterwards an equation 
given by William Froude is applied to locate the mean thrust curve. The 
same proportional correction is applied to the curve for speeds and revo- 
lutions. There was then a preliminary curve drawn (not shown in this 
paper) of powers (corrected for electrical and frictional losses) on revo- 
lutions and from it the powers on speeds were determined and plotted as 
shown in the Figures 6 to 19, Plates 4o to 50. 
A comparison of the thrust and power curves of the figures just described 
will show that the location of the propeller with regard to the stern of the 
boat and the broad rudder-post had a large effect on the thrust and power 
required to drive the boat. The same things may be seen more conveniently 
from the table on page 95; it is even more evident on Fig. 11, Plate 45, 
which is plotted with distances abaft the stern post for abscissa and with 
thrust in pounds for ordinates. Each curve of which represents the thrust in 
pounds at a certain speed and appears to consist of two parts: (1) that for 
settings near the rudder post and (2) that for settings with a good clearance 
from the rudder-post. The curves (not the points) for settings 6 inches or 
more abaft the stern-post appears to be well located; those parts which 
represent settings less than 6 inches astern are open to question, especially 
at the abrupt change of form where one branch of a curve is connected with 
the other branch. The mind at once jumps at the conclusion that a large 
waste of power is due to setting the propeller in the eddy astern of the 
rudder post, and that when the propeller is once free from that eddy (as at 
the six-inch settings) there is only the normal and regular gain that may be 
