between bodies, such as between a ship hull and unsteady propeller cavi- 

 tation, where the nuclei content affects the gas void fraction of the 

 medium directly, a different kind of problem to which we now, momentarily, 

 defer. 



To make full use of the nuclei distribution information, we need an 

 appropriate physical understanding of just how the nuclei themselves lead 

 to cavitation in the flows of interest. (Direct observation of this process 

 is still lacking.) And of course, one would like to predict from labora- 

 tory tests (given full control over all laboratory variables) what cavita- 

 tion inception indices and forms might be on previously untried prototype 

 bodies and conditions. We think this is still not quite possible except 

 for a certain kind of cavitation, namely travelling-bubble cavitation. To 

 appreciate the mechanism of this form of bubble growth, we digress briefly 

 to review the key features of bubble mechanics. 



Some Bubble Mechanics 



We lean heavily, here, on the monograph of Knapp et al. We are con- 

 cerned with the growth of a small bubble of initial radius, R , immersed in 



o 



a pressure field, p (t) , i.e., the pressure far away is time dependent. 

 The vapor pressure, p , depends on the temperature of the bubble wall, T(R), 

 and there may be a gas partial pressure, p . We need the equation for the 

 motion of the bubble radius, R(t) , which is (from Knapp et al. (1970)) 



p L 



RR + o R - zz\ p„„, - 4 y dMT-P,, 



2 p I cav R R 



(12) 



where p is the liquid density, u the viscosity, and O the surface tension 

 and 



) = p (T(R)) + p (13) 



cav v a 



The partial pressure of the gas may be expressed as 



59 



