646 



The third series, with caps fired at depths of 8, 10, and 13^ in., 

 was not an intended experiment but is included because of the deviation 

 of the results from the normal similsirity curve. Experiments with some 

 spreading agent such as oleic acid were planned but because oxidized fuel 

 oil was accidently introduced into the tank during the course of the 

 similarity curve determination, it was decided to continue with the 

 shooting to ascertain if the spreading oil had any effect. Fortunately, 

 the oil did spread easily, probably into a monomolecular layer, and when 

 measured on the ring tensiometer (in the absence of an Adam-Langmuir trough) 

 allowed the ring to separate from the surface at a tension of about 

 52 dynes/cm thus showing a spreading pressure of betireen 25 and 30 dynes/cm, 

 since the surface tension of pure water at these temperatiires, as measured, 

 is about 80 dynes/cm. 



3. Critical Depth Experiments 



A direct experimental approach to the value of T in Eq. (2) is the 

 determination of the "critical depth" at which the explosive must be placed 

 so that Vq becomes zero. Above this "critical depth" water is forced from 

 the surface and below, it is not. At this depth, where V© - 0, T can be 

 calculated from Eq. (2) if a reliable value of Pm is available. The 

 "critical depth" can be determined by suspending Engineer's Special 

 detonators at various depths below the suirface of the water in the harbor 

 and photographing the resulting explosion by means of a 4 x 5 Speed Graphic 

 camera. The detonator was fired by means of the camera's synchronized flash 

 attacV-^ent and the shutter speed was maintained at the relatively low value 

 of 1/25 sec in order to catch any surface effects. A typical print is shown 

 in Fig. 4. 



III. EXPERIMENTAL RESULTS 



U* Methods of Computation 



The initial velocity of the rising dome was obtained by the analysis 

 of strip prints made from the Fastax and Streak negatives from the fxill 

 scsde experiments conducted in Nonomessett Pond and from the E«istman High 

 Speed negatives of the tank experiments. Examples of these strip prints 

 appear in Figs. 5, 6, 7, 10, 11, and 12. Figures 6 and 10 have been 

 reproduced photographically. 



For the Fastax and Eastman High Speed prints, a straight edge was 

 used to draw a line through the tips of smoothed dome contours on 

 successive frames, starting with the Initial appearance of the dome and 

 continuing until the rise was no longer linear with time or until the image 

 left the print £u:ea. Since a carpenter's level was used in mounting the 

 cameras to insure horizontal travel of the film, the edge of the sprocket 

 holes could be used as a base line, and it was therefore easy to calculate 

 the slope of the rise line, i.e., dome velocity, in terms of arbitrary 

 units depending upon the film speed and enlargement for each print. This 

 dome velocity, in arbitrary units, is measured from thb Streak prints in 

 much the same manner, the rise line being the best straight line which can 

 be drawn through the edge of the image which corresponds to the path of 

 the marl mum dome height. 



-5- 



