383 



1 1 1 1 1 1 1 1 1 r 



a, (DEG) P„ (kPa) Vco (m/s) 



T 1 



• 



o 

 O 



D 

 A 



( °) 



0.95 

 0.95 

 0.95 

 1.00 



1.55 

 1.55 



76.3 



124.3 

 158.8 

 165.7 



76.3 

 127.7 



11.49 



14.78 

 16.42 

 16.42 



11.49 

 14.78 



= FOIL ANGLE AT K 



FIGURE 26. Influence of V 



1.0 

 REDUCED FREQUENCY, K 



(£m/c) . 



and K on cavity length 



Nevertheless, if the inception angle ciig is known 

 from the steady model tests, the unsteady effect 

 on cavitation inception, to the first order, may 

 be estimated by the present method. Since the 

 present tests were carried out with only one foil 

 shape and only one pitch axis location , further 

 experiments are required, and in particular, the 

 range of variables should be extended. 



Based on photographic observations of the leading 

 edge sheet cavitation instabilities, it appears 

 that the free shear layer and near-wake stability 

 concepts reviewed by Wu (1972) give a reasonable 

 qualitative description of the physical process. 

 The inherent instability of the free shear layer 

 and associated vortex shedding appear to provide 

 a reasonable model for the breakup of a sheet 

 cavity. However, the detailed hydrodynamics 

 associated with the near-wake closure region can 

 still only be postulated. The commonly held concept 

 of a reentrant jet, Wu (1972) , may provide a reason- 

 able description applicable to the closure of the 

 near-wake region during the actual shedding of 

 vapor. For sheet cavitation extending over only a 

 portion of the foil chord this reentrant jet may 

 not actually penetrate the cavity itself but pene- 

 trate only a locally separated region just down- 

 stream of the sheet cavity trailing edge. In any 

 event, the presence of a reentrant jet is not 

 required to explain the inherent instability and 

 breakup of the sheet cavity. 



For the conditions of the experiments reported 

 here, where the gross flow is nominally two dimen- 

 sional, the cavity instability is not coherent to a 

 significant extent along the foil span. In other 



words, the cavity instability is highly three- 

 dimensional and appears to be principally dependent 

 on conditions in the immediate upstream free shear 

 layer flow. The most extreme foinn of cavity insta- 

 bility is manifest as a large shed cloud of vapor 

 and thus referred to in the literature as "cloud" 

 cavitation. 



Within the context of the experimental results 

 reported here, the principle parameters controlling 

 the formation of cloud cavitation are reduced fre- 

 quency, K, cavitation number, a and foil oscillation 

 amplitude, aj. The maximum cavity length, (P-m/c) » 

 is a function of these three parameters. However, 

 it has been shown that predictions of ^m/c at finite 

 reduced frequencies cannot be based on the cavitation 

 observations at zero reduced frequency. With a con- 

 stant, the results show that it is possible to have 

 no cloud cavitation at finite reduced frequencies - 

 even though it was present on a stationary foil set 

 to the maximum unsteady angle. However, if the 

 steady foil is set to the mean angle of oscillation, 

 Uq, and no cloud cavitation is present, then it is 

 easily shown that at finite reduced frequencies 

 cloud cavitation will be present. Thus, Ito's con- 

 clusion that there exists a "critical" reduced fre- 

 quency for the onset of cloud cavitation appears to 

 be the result of the specific chosen values of the 

 parameters, K, a, and aj. 



The implication of the above results is that the 

 prediction of the occurrence of cloud cavitation 

 for hydrofoils in waves and propellers in wakes can- 

 not be based solely on the performance in calm 

 water or uniform flow. 



