Lin, Paulling, and Wehausen 



Fig. 5 - Photograph of the two models tested 



minimizes R^/pgY for the design Froude and Reynolds numbers. The various 

 steps required are incorporated in Table 2 and are explained below. The suc- 

 cessive lines in the table are obtained as follows. Fixing l/h fixes L/B for each 

 of the three available values of B/H. Then Cp is fixed by the given value of c^ 

 [see Gertler (1954), pp. 10-12]. Since there is no hull form with B/H = 3.75, 

 L/H = 20, Cy = 0.003, this column now drops out. The associated values of 

 c^ = SL-^c^-'^ and of c^ - R/VapSv^are read directly from Gertler [1954]. The 

 value of Cf = Rf/VzpSv^ is the Schoenherr coefficient for Re = 1.182 x 10 ^^ cor- 

 responding to a 400' ship in salt water at 63 °F. with a ship's speed correspond- 

 ing to 7^ = 5. Then c, = C + C. and 



pgV 



C.Fr^C C 



s 'V 



ic 



c c. 



The hull with B/H = 3 and Cp = 0.536 is evidently the best within Taylor's 

 Standard Series which meets the constraints L/H = 20 and C^ = 0.003. Although 

 this is not an "equivalent" hull, it does seem to be also a legitimate one to use 

 in a comparison with the optimum symmetric ship for y =5. 



TEST PROCEDURE 



The models were each tested in the Ship Towing Tank of the University of 

 California. The models were attached to the dynamometer so that they were 

 free to both heave and trim. Figure 6 shows the symmetric model being towed 

 at a Froude number of 0.316. 



Each model was tested both with and without a tripwire. In the region for 

 which data are presented there was a small constant difference in the resistance 

 coefficients R^/!4pSv^ with and without the tripwire. This was taken as evidence 



1054 



