16 RESISTANCE OF BILGE KEELS. 



It is interesting to note that the areas of the normal 3-inch keels are 0.057 of 

 the wetted surface of the hull and if the same values for / and n can be used, the 

 frictional resistance is that fraction of the frictional resistance of the hull. 



The surface of the fore extensions of the 3-inch bilge keels is 4.5 square feet 

 or .015 of that of the bare hull. The frictional resistance at 6.0 knots may be 

 computed as 1.35 pounds, which is less than the limit of accuracy of our thrust 

 mechanism. 



The curve of thrust on speed for the 3-inch keels with both bow and stern 

 extensions as plotted was found to be hardly distinguishable from that for the 

 normal 3-inch keels; at lower speeds it appeared to fall slightly below and at 

 higher speeds to rise slightly above that curve; but the differences were easily 

 within the limit of accuracy of our observations. Consequently our observa- 

 tions, so far as they give any evidence, indicate that the bilge keels may be produced 

 so far in proportion as shown on Plate 13 without undue increase of resistance. 

 It will be noted that the extension at the stern does not much interfere with stream 

 lines, but at the bow the extension crosses the stream lines at appreciable angles. 



The experimental determination of the power required to drive the Fulton with 

 and without bilge keels is given in kilowatts input to the electric motor by the 

 curves on Plate 20. It is immediately evident that these curves in general charac- 

 teristics and distribution conform to the curves of thrust on Plate 19, and that 

 consequently they confirm those curves, and in particular justify the thrust curves 

 for the 9-inch keels. The curve for the 9- inch keels has been chosen as most con- 

 venient for analysis ; inspection of Plate 20 shows that the conclusions drawn from 

 such an analysis may be applied to the narrower keels and in particular to the 

 normal 3-inch keels. 



At the several speeds in column i of Table III the kilowatts input to the 

 motor was read and set down in columns 2 and 3 for the bare hull and for the 

 9-inch keels. From the brake tests of 19 12 the equivalent brake horse-powers 

 were taken and set down in columns 4 and 5. These brake tests of 1912 were 

 made at length and with suitable precautions; the motor and shaft were in sub- 

 stantially the same condition during the experiments of this season as when the 

 brake tests were made; and no possible change of condition would have an appre- 

 ciable effect on our conclusions. 



The efificiency of the propeller was taken from tests made at the Washington 

 Model Basin by Naval Constructor D. W. Taylor, U. S. N., on a 12-inch model of 

 the propeller used. The real slip (required for use with the report of such a test) 

 was found from the apparent slip on the assumption of a ten per cent wake. Had 

 a larger wake factor been used the efificiency would appear to be smaller than 

 here given ; but no possible change in propeller efificiency would affect our conclu- 

 sions to any considerable degree. 



Multiplying the brake horse-power by the propeller efificiency gives the power 

 delivered by the propeller as set down in columns 7 and 8 of Table III ; the differ- 

 ence, set down in column 9, is shown as the basis of comparison with the power re- 



