558 



Tae precision of measurement of the periods depended on the precision 

 with which the film record coiild be read. This is estimfited to be about 

 0,5/6. Much greater precision is not very likely to l.a attained by this 

 method because of the finite width of the trace on the film and because of 

 th« difficulty of locating the exact position of thf bubble pulse maximum on 

 the film. 



Depths were found from measured lengths of cable, or, in the deeper 

 shots, from readings of a Bourdon depth gauge. From the various data 

 collected, it is estimated that the precision of measurement is about 1$ and 

 may be as poor as 2$ in the worst cases. Less reliance can be placed on 

 those shots where the cable angle was gaeat, i.e., greater than 20® from 

 the vertical, and where no depth gauge was used. 



Since the radius of the bubble depends upon the cube root of the depth, 

 it Is not very sensitive to errors in the latter. However, the period depends 

 upon the 5/6 power of the depth, so only slightly more precision is attainable 

 in the period than is inherent in the depth meas\u*ement. 



3. Charge Weight 



The half-pound charges used were weighed to the nearest gram, the 

 1 oz charges to 0.1 gm. Since cht4i"ge weight copears in the equations as a 

 cube root, the error made here is quite negligible. 



A correction had to be made for the booster in each case. This correction 

 was made by reducing (or increasing) the weight of booster used in the cal- 

 culation to an equivalent weight of the explosive being studied. The factor 

 employed rra.s found from the ratios of bubble period constants, by 

 successive approximations if necessary. Any error made in this correction 

 will introduce a negligible error in the final weight, since the weight of 

 booster is only a small fraction of the total. 



4. Badiua Measurements 



Because of the relatively narrow angle of view of the high-speed cameras 

 uged, the bubble at maximum size ordinarily came very close to the edges of 

 the photograph. In the very shallow shots using 25 gm cliarges, the narrower 

 dimension of the picture was too small to include an entire bubble diameter. 



The possibility of optical distortion due to the lens was first 

 investigated. Photographs of a grid taken with the Fastax camera showed no 

 appreciable distortion {<2%) over the whole usable field. Calculation2) 

 of the distortion introduced by the glass-air interface before the lens 

 showed that in the worst case the correction would be entirely negligible. 



Measurements of the radius of the bubble were made on photographic prints. 

 Diameters were measured in at least three directions and averaged. Care was 

 taken to avoid including irregularities outside what was believed to be the 

 true surface of the bubble of gases. As the minimum is approached, the 

 bubble is progressively obscvirad by opaque streamers of explosion products. 



