15 inches per second were used. In a third method this 

 same signal was fed to the vertical deflection amplifier of 

 a Tektronix Model 512 oscilloscope. This display was 

 photographed using a 35- mm Fairchild oscillograph camera. 

 The horizontal sweep or time base was achieved by the 

 movement of the film past the vertically deflected spot on 

 the scope face. The film was set to run by the deflected 

 spot at 2 5 inches per second; to insure accuracy of final 

 reading, timing marks were placed on the film at 100- msec 

 intervals. 



During the laboratory analysis of the data it was found 

 that the tape recordings were of limited value. This was 

 because the data were amplified and put on the tape as 

 amplitude changes. The low- frequency response of an AM 

 recording distorts and loses much of the detail in this type 

 of complex signal. A frequency- modulated type of record- 

 ing and playback would have corrected this deficiency; how- 

 ever it was not possible to conduct further experiments 

 after this shortcoming was noted. Fortunately the two 

 types of photographically recorded data appear to contain 

 the necessary detail. Figure 13, with its associated descrip- 

 tion, provides pertinent information on a typical underwater 

 spark signal. It should be noted that the signal from the 

 underwater spark is very similar to the signal received 

 from underwater explosions of TNT. 



The signal in figure 13 covers a period of slightly more 

 than 13 milliseconds with the source at 100- foot depth. 

 During the first 2 msec the shock wave amplitude was so 

 high that the signal was off the film and cannot be seen. At 

 about 3. 8 msec the bubble has expanded to its greatest 

 diameter and the pressure is falling. At about 7. 6 msec 

 from the start the pressure rises sharply as predicted, ^ 

 corresponding in time with the near maximum contraction 

 of the bubble. The bubble then expands to a second max- 

 imum at about 9. 2 msec and the pressure falls until it 

 reaches a point at about 10. 8 msec, where again the pres- 

 sure rises sharply, corresponding to another maximum in 

 the contraction of the bubble. The bubble then tries to 

 expand again but begins to disperse. 



In addition to the oscillograph trace of figure 13, 

 pertinent frames from the high-speed record are shown 

 with identification of their time- position relative to the 

 oscillograph trace. 



2 

 Cole, R. H. , Underwater Explosions, Princeton University 



Press, 1948 



27 



