612 -2- 



cnosen generally between iCi and 60J of the maximum raoius of the dome. The fact that the done 

 is limited in diameter is itself an indication that tne variation of velocity assumeo Is only 

 -iooroxlmate. 



Notes on the oractical details of the method are given in the aooenoix. 

 Bast s of method . 



The Oasic assumotion In the method is that tt\e detonation wave exceot In the imirsdiate 

 neighbourhood of the exoloslon, Behaves aoproximately as a sonerical sound wave. For a soherical 

 sound wave the strain enerjy in the wave is constant, I.e. 



It follows that where the wave length, •_ is constant and small comoared with the radius 



=^ r^ = constant 



and the oressure is therefore inversely Drocortional to tne radius. 



At the surfac-.; the detjnatijn :;ressure wave is reflected as '-i suction wave and the 

 oarticle velocity at tht surface which is assumed to oe the initial velocity of the spray is 

 given by 



p vc =20 CCS 8 at °, 



p V c = 20 at 0. 



where p is the density and c is the velocity of sound in sea water.. 



It follows that ■-■ = ■ ' . ^°- " 



Results obtained so far with the method . 



The method has been aaolied to the analysis of results oOtaineo with deoth charges in 

 tests during Sectember, I9i*l. The depth charges were oroDceo from aircraft in a wide range of 

 height and sceed conditions* 



Particulars of these tests and the results of the analysis are given in Table 1, There 

 is no direct check on the Cilculateo aeoths of detonation, but some in:iication of their accuracy 

 can be obtained. In Figures 3 an,i u the calculateo Jeoth is olottej against two time intervals 

 which are relateo to the oecth anj yet are measured quite inaeoenjently of the jecth calculation. 



It will be seen that with one exception the depths obtained are all within 10 feet of a 

 mean curve and the majority are within 5 feet. Both of the time scales used are influenced 

 Dy other factors which cannot be allowed for with any certainty. The time unoerwater before 

 detonation (see Figure 3) must depend on the impact conditions which vary widely an:! the time 

 from ''etonation tc the Appearance of tne plum: (in Figure 4) is bound tc be somewhat erratic 

 if only .:n accrunt ^f the irregularity ;f the olume itself. It will be seen, however, that in 

 Figure u In which the time base Is least oeoen:!ent on the external conoiti;ns of the test, 

 there Is the closest grcuoing ;f the- oolnts. There is :ne p:int in Figure- 3 which may be noted. 

 In one drop the depth charge bounced on Impact so that by the second impact it had lost the 

 greater part of its Initial velocity. The time underwater from the second impact to detonation 

 was 5.7 seconds and the calculated depth was 57 feet. This corresdonds to a mean sinking 

 soeeJ of 10 f.c.s, which Is the accerto.? figure for these decth charges. 



It is reasonable tc suoooso therefore, that the variations shown in Figure 3 an3 u ire 

 real variations In the aeoth ,f Jetonatlon ano that the method in practice gives tne ceoth to 

 within a v-:ry few feet. For c;nclusiv» ;r:of control tests with Jeetn charges oetaiate: at 

 known deoths are required. 



It has 



