MEASUREMENT OF PRESSURES 181 



calibrations at much higher pressures is within their combined experi- 

 mental error, the results being shown in Table 5.1. Calibrations made 

 at three different laboratories by similar, essentially static methods are 

 included in the tabulation and indicate the satisfactory agreement ob- 

 tained by different observers and equipment. 



C. Use of transient waves. A fundamentally more desirable method 

 than either static or steady state calibration would be one in which a 

 shock wave, or discontinuity of pressure, with known amplitude and 

 long duration passes the gauge to be calibrated. One possible method 

 consists in releasing the pressure of a closed chamber, by bursting of a 

 diaphragm for example, and initiating a compression wave in fluid at 

 atmospheric pressure external to the chamber. Experiments of this 

 kind attempted with thin copper diaphragms and oil as the pressure 

 fluid met with only limited success, the difficulties being that, with the 

 arrangement employed, the pressure wave developed had a rounded 

 outhne and was not very reproducible. This difficulty was the result 

 of the time of about 100 ^tsec. required for rupture of the diaphragm 

 and the failure of the wave so initiated to develop into a shock front in 

 the rather unfavorable geometrical arrangement improvised for the 

 tests. A further objection was the impossibility of obtaining an abso- 

 lute calibration of the pressure; the method as tried was therefore useful 

 only for comparison purposes. 



A more promising line of approach is use of a plane shock wave in air 

 formed by rupture of a plastic diaphragm separating a region at known 

 gauge pressure from a tube containing air at atmospheric pressure. 

 This method has been extensively employed in air-blast work^^ and has 

 the advantage that pressure waves so initiated in air develop into shock 

 fronts much more rapidly than in water. A further advantage of the 

 method is the fact that the increase in shock wave velocity with pressure 

 difference is very considerable in air even at moderate pressures ; for ex- 

 ample, an excess pressure of 20 Ib./in.^ in air initially at atmospheric 

 pressure and a temperature of 20° C. travels with a speed of 1,620 

 ft. /sec, a value 47 per cent greater than the speed of an acoustic wave. 

 This increase in velocity is, moreover, directly related to the pressure 

 difference by the Rankine-Hugoniot equations (section 2.5) and velocity 

 measurements of the progressive wave therefore permit absolute pres- 

 sure determinations.^^ A gauge to be used in underwater measurements 

 should of course be subjected to a pressure wave in water or a liquid, 

 and this can be done by allowing the air shock wave to be reflected from 



14 A comprehensive report, with references to earlier work, is that of Fletcher et 

 al(36). 



1^ Experimentally determined pressures from velocity data are found to lie sys- 

 tematically below values predicted from the pressure released, indicating imperfect 

 formation of a plane shock wave in the experimental tubes employed. See Reference 

 (36). 



