230 - " - 



|f b is smill the relationship Betwet'n :i' ind t' will be little altered by viscosity, so that the 

 value of the integral in f.quation (lO) c^n be estimated using the values calculated when /i = o. In the 

 case calculated in Report 4, *hcre z'^ = 2, the v^lue of r'' ^, / da'Nj,^, taden over the whole of the 



first period of the bubble is 0.«, so that the value of b at the end of the first period is 

 " °"" (U) 



(pVg) ^ 



In some of our experiments, in which transformer oil of viscosity C.3 poises and density 0.875 



gm./cc. was used, z' was of the order of 2; the value of d calculated from equation (ii) was 0.006, so 



that less than one per cent of the 'Energy of the bubble is dissipated by viscous forces during the first 

 period, if the bubble remains spherical. 



5. The choice of operating liquid . 



With explosions in open water the pressure at points far distant from the bubble is never less 

 than 1 atmos; the minimum pressure in the bubble, though considerably less than 1 atmos. , does not fall 

 as low as the saturation vapour pressure (S.V.P.) of water. in experiments with witer under a surface 

 pressure which is low compared with normal atmospheric pressure but well above the S.V.P. the minimum 

 pressure near the surface of the bubbl-' may be U-ss th-,n the S.V.P. The water then boils near the bubble. 



Figure 1 shows two superposed photographs, on the same plate, of •-'. bubble made by a spark under 

 water at surface pressure rather graat. r than the S.V.P., (l) at an early stage of the expansion, when it 

 is smooth and spherical, and (2) at a later stage, when the small vapour bubbles due to boiling have had 

 time to develop. It will be seen that the surface has beC'.me pitted by the boiling. 



In ..rder tj avoid boiling when experiment iny at low pressures it is necessary tc use a liquid with 

 a low S.V.P. For this reason transformer oil was used. Though the viscosity of this liquid is about 

 30 times that of water, it has been shown that the expansion of the gas bubble is not likely to be materially 

 different from what it would be in water, at any rate while it is approxim-.tely sphtrical. 



6. D escription of the apparatus . 



The airtight vessel in which the bubbles are produced is shown in elevation in Figure 2; it 

 consists of a vertical glass cylinder, », 12 inches diameter, j, inch wall, 15 inches high, cemented between 

 two horizontal steel end-plates, 8 and C, 18 inches diameter, i inch thick. This vessel is filled with 

 water or oil to a depth of about n inches, and it is connected to a vacuum pump and a pressure gauge 

 through the tube D. 



A bubble is produced in the liquid by the discharge of a condenser, charged to a potential difference 

 of u.OOO volts, across a gap of about lO/lOOOth inches between the rounded ends of the horizontal brass 

 rods E and F, ^th incn diameter and 3 inches long. The electrode E is connected electrically to the upper 

 steel plate B, which is earthed; the electrode F is insulated and is connected to the condenser and the 

 timing pendulum by a lead passing through the insulated bush, G. The electrodes are fixed to the brass 

 frameworks shown in the diagram so that they can be moved independently, without admitting air to the vessel, 

 by means of a brass push tubr and two br^ss push rods p.-.ssing through stuffing boxes fixed to the upper 

 steel plate. 



The framework supporting the electrode F can be moved verticalfy or rotated by means of the push 

 rod passing through the right stuffing box shown in the diagram. 4 hinge H allows the electrode to fly 

 back during an explosion, A'hilst the horizontal membsr of th= framework is supported and is prevented from 

 moving at right angles to the plane of the diagram by a wire staple J fixed to a block of insulating 

 material attached to the pusn rod. 



The framework supporting the electrode E is connected through the hinged joint H to the push 

 tube passing through the lower stuffing bjx, shewn ^n the left in the diagram; a push rod passes through 

 this tube and through the upper stuffing b.x and carries at its I wer end a horizental brass bar to which 



fixed 



