McCormick 



flow quantity was calculated to be 1.37 cfs for a design velocity of 50 fps. The 

 pump was estimated to require 8.34 hp. 



In order to integrate the Karman momentum equation it is necessary to 

 know the body radius and static pressure distribution. The body shape of TRI-B 

 is composed of three parts: (a) a modified ellipsoid nose, (b) a parallel mid- 

 section and (c) the afterbody of the DTMB series 4166 body. The final shape is 

 similar to a Reichardt constant pressure body. 



For this shape, the pressure is nearly constant over 80% of its length be- 

 ginning 5% back from the nose. The measured pressure distribution obtained 

 from wind tunnel tests is presented in Fig. 2. Included on the figure are empir- 

 ical expressions which were used in the numerical integration. 



At 50 fps, the laminar skin-friction drag on the body was estimated to be 

 11.7 lbs. The drag of the ring tail was estimated at 29.4 lbs giving a total drag 

 of 41.1 lbs. 



If laminar flow were not achieved, the body drag was estimated to be 132 

 lbs giving a total drag of 173.1 lbs. The pump was designed to eject the suction 

 flow at 50 fps; hence in evaluating the drag from the terminal velocity, thrust 

 (or drag) from the pump must be considered. In terms of equivalent flat plate 

 area, f = D/q, f was predicted to be equal to 0.069 for a turbulent boundary 

 layer and 0.0164 for a laminar one. 



The method by which the body was designed has been presented only briefly 

 because of various shortcomings in the method which became obvious as the 

 project progressed. These will be discussed later in the paper. 



TESTING OF TRI-B 



The first tests of TRI-B began October 1962 at the U.S. Naval Torpedo Sta- 

 tion, Keyport, Washington. Over a period of two months, 12 runs were per- 

 formed of which 7 yielded valid data. A photograph showing the body exiting 

 from the water is presented in Fig. 3. For these runs, only the velocity as a 

 function of time was measured. From the results, the disappointing conclusion 

 was reached that the expected laminar flow had not been achieved. 



The fact that the boundary layer with the pump operating was turbulent was 

 substantiated by running with a trip ring on the nose for which the body attained 

 the same terminal velcoity as without the ring. 



There were several possible reasons at this time why laminar flow was not 

 being achieved. First, a calibration of the suction pump showed that at the de- 

 sign hydraulic pressure of 2000 psi, it was delivering only 0.85 cfs instead of the 

 design value of 1.37 cfs. Secondly, the suction slots were not continuous around 

 the circumference, instead they were interrupted by small, structural, carry- 

 through bridges. Finally, the body exhibited a tendency to depart from the ver- 

 tical in its travel to the surface. 



1004 



