649 



6, Tank Results 



The results of applying the analytical methods of Sec. U to the tank 

 shots are summarized in Tables III, IV, and V, The eimiLirlty curve for 

 Engineer's Special Detonators derived from this work appears in Fig. 13. 

 The results of the sympathetic detonation series «u:e superimposed on this 

 similarity curve in Fig. 14 and those obtadned in solutions of lowered 

 surface tension are superimposed on the similarity curve in Fig. 15. 

 Typical film records are shown in Figs. 10, 11, and 12. Discussion of 

 these results will be presented in Part IV. 



7, Methods for Ctaculatlng T 



The principle of obtaining T from Eq, (2) is very simple. Any 

 procedure by which Vq can be made equal to zero will allow T to be 

 calculated . 



V - — ^— (P - -) - 0- 

 /*e ^ ^ 



P -^-0 

 '^m 2 



T - 2P. 



The only requirement is that P^ be knovm from some other experimental 

 or theoretical evidence. 



Four separate methods have been used to obtain values of T. The 

 first method was discussed in Sec. 3. Measurements of the "critical 

 depth" by lowering the detonator until spray was no longer formed were 

 made on two separate occasions, the first at a time when the water sur- 

 face was smooth and the second when the surface had about 2 - 3 in. 

 ripples. The results obtained on these two days are shown in Table VII. 

 Other estimates of T made from data in the literature also appear in 

 Table VII. No attempt has been made to include in Table VII all 

 estimates of T that might be made from the data in the literature. 



A second method is the extrapolation of curves of dome velocity 

 versus depth to Vq - 0. Of the dome velocity results in this report, 

 only those in the tank series allow a justifiable extrapol'ition to Vq « 0. 

 For this case, Vq - at a depth of about 18 in. or a wV3/r value 

 of 0.092 corresponding to a peak pressure of 1800 lb/in. 2 or tension of 

 3600 lb/in. 2. While this method is interesting it can obviously be 

 applied only to a series of dome velocity results which is very c-i^iolete ,„ 

 and subject to little experimental error, especially for lov: values of W^' V'"^ 



