719 
ai as 
(iv) Calibration. The amount of charge developed by a given gauge 
per unit change in applied pressurc was determined by means of a "static" 
microcoulometer technique [2], The assembled but unmounted gauge element 
was subjected to 3000 1b/ing pressure which could be suddenly released, 
and the charge developed by the gauge read on a microcoulometer.: Calibra~ 
tion of unmounted elements by this technique was reproducible to better 
than 2 percent but owing to the following factors the absolute value of the 
gauge constant in actual use was not necessarily represented by this pre-~- 
cision: (1) the static calibration did not subject the gauge to a pro= 
gressive shock wave simulating that which the gauge is used to measure; (2) 
the gauge Pahoa may have been modified during mounting and coating; (3) 
cable signal4/ may have caused a variable apparent gauge constant diifcrent 
from the actual gauge constant; () gauges may have been unlmowingly dam 
aged during use thus altering their constants. 
(b) Electrical cables. — The transmission of the gauge signal from 
the gauge to the recording vessel was a major problom in studying large cx- 
plosive charges, The clectrical signal produced by the gauge was small, 
and the long cables necessary caused attenuation of this signal and did not 
have a linear frequency response, In addition to this it was found that 
concentric cables themselves gave off a signal when subjected to mechanical 
stress. These problems have been dealt with in detail by R. H. Cole clse- 
whore [] but are presented below particularly with view to the RELIANCE 
worke 
(i) Cable signal. When a coaxial cable is struck by a shock wave an 
electrostatic charge is develoved between the shicld and central conductor. 
In a high impedance system such as encountered in using piezoelectric 
gauges the magnitude of the charge so developed may be large relative to 
the charge developed by the gauge itself, If the cable is leading directly 
away from the gauge, this signal does not affect the peak pressure value 
obtained, since the shock wave will not have struck the cable before the 
peak has been recorded. However, if it is desired to obtain the complete 
pressure-time curve this signal may introduce serious distortion, either 
positive or negative, As it was not practicable to construct gauges with a 
higher voltage output, it was necessary to overcome this problem by the use 
of cables with very small cable signal. 
The cable used at UERL for this purpose was the type developed by the 
Taylor Model Basin [16]. This consisted of a copper tube of 1/8<in. outer 
diameter and 1/16~in, inner diametor through which was led a fiber~glass 
insulated enameled wire which served as the central conductor, Hot ceresin 
wax was dravm into the space between the central conductor and the tube. 
The cable signal developed by this tube was small compared with the gauge 
signal, The piczoelectric gauges were attached to 20 ft of the tube for 
most work which, since it lead directly awey from the gauge in the direction 
of the propagation of the shock wave, gave a h-msee interval during which 
the cable signal was as low as possible. When pressure—time curves over 
Weee Sec. 7(bi). 
