30 



If the surface of the specimen Is completely uniform, It appears 

 that separation would occur at a fixed, straight line around the specimen and 

 that the strlatlons would be uniformly spaced. However, the photograph of 

 the sphere drop In Figure 23 by no means represents the typical distribution 

 on all sphere drops. Although the strlatlons always appear, they are sel- 

 dom as uniformly distributed as on the floats and sphere shown herein. It 

 is In connection with this point that the statement was made in the foregoing 

 that the surface roughness is undoubtedly a variable which must be considered 

 in explaining the flow configuration. If the roughness of the surface is not 

 uniform, the idea of a random distribution of nuclei explains the ragged line 

 of separation. The result would be a vortex of nonuniform strength with re- 

 sulting distortion of the strlation pattern. 



So far the discussion has been concerned only with examples in which 

 the pressure in the "cavity" behind the specimen is atmospheric. However, the 

 same results should be manifested when the cavity is closed and the pressure 

 within it is lower. That the same results actually occur has been observed 

 in photographs of torpedo models under cavltating conditions in the variable- 

 pressure water tunnel. 



ANALYSIS OF THE DATA IN THE PLANING REGIME 



The principal purpose of the present investigation was to extend 

 the available data for the TMB planing float Into the planing regime. Owing 

 to the techniques of experimentation used, any attempt to deduce criteria 

 from the data below the hump speed would prove fruitless because of the number 

 of variables that must be taken into account when the float is completely 

 submerged at undetermined positions below the surface. Furthermore, the re- 

 sults reported previously for maximum loading (2) are sufficient for design 

 purposes below the hump speed. 



With the methods used, it is not possible to determine precisely 

 the speed at which true planing begins. Inasmuch as the effects of displace- 

 ment and of dynamic lift on the float occur together in a region^ of transi- 

 tion from typical displacement flow to true planing, the definition of the 

 boundary between the displacement regime and the planing regime is arbitrary. 

 For purposes of the reduction of the TMB float data, this dividing line is 

 defined as the speed determined by the position of the horizontal tangent to 

 the drag curve at the hump. This definition is not entirely satisfactory 

 because of the ambiguity existing in the hysteresis region for the heavy load 

 conditions, but the data can be reduced to a unique relationship on this basis. 



Using the horizontal-tangent method on all the drag curves of Figure 

 10, the dividing line between the displacement and planing regime is shown by 



