Cavttatton (Influence of Free Gas Content) 
the surface, it is particularly important to have some means of di- 
scriminating between vaporous and gaseous cavities. If the cavity is 
truly vaporous, and grows "'explosively"', then the collapse should 
also be far from equilibrium with the local pressure field and should 
produce noise. This definition has been used by innumerable investi- 
gators. For headform studies, Saint Anthony Falls Hydraulic Labora- 
tory (SAFHL) [18] , [19] and NSRDC have regularly been using 
the acoustic radiation as the criteria for inception, 
It can be shown that the velocity of cavity collapse for nominal- 
ly hemispherical cavities scales approximately as the square root of 
the pressure difference across the cavity wall [20] ; [2 1] eins, if 
one assumes that the shape of the cavity during collapse is essentially 
the same for slight changes in this pressure difference, then the col- 
lapse velocities also would only experience slight changes. On this 
basis it will be assumed in this paper that the noise produced by the 
collapsing cavities will not be significantly affected by small changes 
in the pressure of the water tunnel, 
When the acoustic impedence between the water and the head- 
form material is changed, then the amplitude of the noise detected by 
the hydrophone will be affected. As one would expect, the peak noise 
amplitude will vary over a finite range. This has been experimentally 
shown by Brockett fi 0] for a headform made of Delrin which is a good 
impedence match to water. When the headform material is a metal, 
such as copper, then there is a poor acoustic impedence match and 
one would expect to detect a lower peak noise amplitude from the col- 
lapsing cavity. This material influence cannot be entirely cancelled 
out by adjusting the detection threshold of the electronics on the basis 
of background noise. Therefore, it is expected that the DELRIN head- 
forms will indicate cavitation at a somewhat higher water tunnel pres- 
sure than for a metal headform. This aspect is not of significant con- 
cern here because direct comparison of the cavitation inception num- 
ber, o;, for the two types of materials is not intended. The most im- 
portant concern is to determine how variations in the free gas bubble 
distributions affect the inception on a headform of the same material. 
B. Free Stream Bubble Size and Distribution Measurements 
The microscopic gas bubbles immediately upstream of the 
headform at inception conditions have been measured with a high 
speed holographic technique. This technique was selected because 
it appeared to be unique in its ability to (1) make direct measure- 
ments with no calibration required, (2) discriminate between bubbles 
1139 
