42 



30. The theoretical curves are computed from the exact theories and are taken 

 from Reference 68. It is seen that for small cavitation numbers this method 

 of comparison may be used as an approximation. It should be pointed out, how- 

 ever, that the two-dimensional theory 

 does not give values for the length 

 of the cavities which may be used to 

 estimate the length in the three- 

 dimensional case, the two-dimensional 

 cavities being much larger than three- 

 dimensional ones for the same cavita- 

 tion number. The question of the 

 length of cavitating regions, in 

 venturi-shaped nozzles is discussed in 

 the following section. 



The effect of wall or free 

 surface proximity on finite cavities 

 has been considered by Simmons. 73 

 This problem is of interest in the de- 

 sign of water tunnels. 



More exact results for engin- 

 eering use must await development of 

 the three-dimensional free-streamline 



Figure 30 - Comparison of Area Ratios theory for bodies with fixed separa- 



for Two-Dimensional Mathematical . . . . . . ■ . . , 



Models and Three-Dimensional tl0n P°mts and both two- and three- 

 Cavities Behind the Discs dimensional theory for curved surfaces, 

 of Figure 23 



MECHANISM OF STEADY- STATE CAVITATION IN VENTURI-TYPE NOZZLES 



In the case of the cavities studied at the Taylor Model Basin, it 

 has been suggested that the oscillations are associated with the pressure and 

 velocity fluctuations in an unstable wake. Although, on the average, such 

 cavities have been successfully described (in two dimensions) from purely hy- 

 drodynamical considerations, the processes in the actual cavity are still not 

 very clear. 



In the cavitating venturi of Fottinger 8 and Hunsaker 59 the condi- 

 tions are somewhat clearer. The conditions at the end of the cavitating re- 

 gion may be likened to the conditions in a nozzle with an abrupt expansion 

 since there is apparently always a central jet of liquid which flows through 

 the cavitating region. In this case, the conditions at the "jump" may be such 

 that its position is stable or the downstream pressures may be altered in such 

 a way that this point may move upstream in analogy to the moving hydraulic jump. 



