379 



SIDE VIEW 



TOP VIEW 



P'IGURE 18. Cloud cavity sepa- 

 ration from leading edge sheet 

 cavity (example 1) , K = 0.99, 

 V^ = 11.5 m/s, P^ = 76.2 kPa, 

 a = 3.25 + 1.55 sin tot. 



-TOP VIEW 



SIDE VIEW 



FIGURE 19. Cloud cavity sepa- 

 ration from leading edge sheet 

 cavity (example 2), K = 0, 

 V^ = 14.8 m/s, P_^ = 124.1 kPa, 

 a"= 3.25°. 



for K = 0) shedding occurrence alternated between 

 several spanwise locations. This is clearly seen 

 in Figure 23. 



Several other aspects of the cavity shedding 

 process were apparent. The shed vapor had an 

 initial gross rotation with the same direction as 

 occurs in the free shear layer. This was evident 

 from the high speed movies viewing the cavitation 

 along the span (ie., a side view), and can also be 

 inferred from the pulse camera photographs taken 

 from the same view. The gross volume of the shed 

 vapor had relatively little dispersion prior to its 

 collapse but frequently developed within it regions 

 of apparent bubble coalesence prior to collapse, 

 as can be seen in Figures 14 and 24. 



On the basis of the previously described defini- 

 tion of cloud cavitation, its occurrence was 

 determined from available photographs. The presence 

 of cloud cavitation as a function of the ratio of 



maximum sheet cavity length, 1^, to chord length, 

 C, and reduced frequency K is shown in Figure 25. 

 The data used to define the condition for the 

 occurrence of cloud cavitation were all taken at 

 nominally the same value of a. Figure 25 shows 

 that for a given (S-m/c) value, cloud cavitation 

 can occur at nonzero K values whereas none would 

 be apparent for K = 0. For example, if test 

 conditions were adjusted such that li'm/c ~ 0.36, at 

 0.3 < K < 0.4, then one could conclude as did I to 

 (1975) that there was a "critical" reduced frequency 

 associated with the onset of cloud cavitation. 



Figure 25 also shows two curves representative 

 of the influence of the value of aj on cloud cavita- 

 tion. It is readily apparent from the data in 

 Figure 25 that the conditions for cloud cavitation 

 cannot be simulated by quasi-steady experiments. 

 As shown in Figure 25, cavity length is strongly 

 dependent on K. If the angle of a stationary foil 



