EQUIPMENT AT MOUNTAIN LAKES 



95 



MEASURING 

 APPARATUS 

 _ * = I35" 



-CALIBRATING RESISTOR 

 B 



Figure 32. Typical circuit arrangements for hydrophone 



coupling characteristic measurements: (A) observation of 

 input signal, (B) observation of output signal. 



quency for this amplifier. This relation may be ob- 

 tained by using the circuit arrangements shown in 

 Figure 33. 



Low-sensitivity, high-impedance hydrophones are 

 usually tested in conjunction with the underwater 

 preamplifier described in Section 6.2.1. This arrange- 

 ment permits the hydrophone calibration to be re- 

 ferred to the ends of extremely short leads and results 

 in an essentially open-circuit calibration. 



An alternative method of calibration, particularly 

 adapted to high-impedance tourmaline gauges, ex- 

 presses the hydrophone output in terms of the charge 

 generated rather than the open-circuit voltage. This 

 method, described in Chapter 4, requires minor modi- 

 fications in the underwater preamplifier. 



Measurements o/ Inherent Noise. The inherent 

 noise level of a hydrophone is measured with the in- 

 strument in quiet water. When open water conditions 



FROM 

 OSCILLATOR 



10 DB 

 I3S" PAD 

 BALANCED 



TO MEASURING 

 APPARATUS 



I3S : .034 M 



FROM OC 



OSCILLATOR 3 

 8=l 3S"|j 



PIER 

 COUPLING 

 AMPLIFIER 



10 DB 

 135" PAD 

 BALANCED 



TO MEASURING 

 APPARATUS 



Figure 33. Circuit arrangements for determination of 

 gain frequency characteristic of coupling amplifier: (A) 

 observation of input signal, (B) observation of output 

 signal. 



are not sufficiently quiet, a bucket with suitable anti- 

 shock mounting or an acoustically treated tank is 

 used. A highly efficient unit, the noise level of which 

 is greatly atfected by changes in the radiation impe- 

 dance, should not be tested in the bucket because of 

 probable standing waves. In fact, even an acoustically 

 treated tank may allow the formation of standing 

 waves sufficient to prevent exact noise level measure- 

 ments. Standing waves cause variations in the im- 

 pedance which the medium offers to the diaphragm. 

 In well-designed radiators, the mechanical impedance 

 of the diaphragm more or less matches the impe- 

 dance of the medium to which the energy is trans- 

 ferred. The more efficient the device, the less will be 

 the loss between the electric power supplied and the 

 acoustic power radiated and, therefore, the closer will 

 be the coupling between the impedance of the device 

 and the impedance of radiation. Since thermal noise 

 generally is proportional to the resistance, small con- 

 fined areas which produce standing waves are not 

 conducive to accurate measurements on the more 

 efficient devices. 



In order to minimize mechanical vibration, the test 

 units are suspended in low-period antishock mounts 

 and every effort is made to reduce the background 

 noise to a minimum during the observations. 



Measurements of the hydrophone noise level and 

 its distribution through the operating frequency 

 range are made on system 2 with each of the accept- 

 ance band widths, 10, 300, and 6,000 c, in such a 

 manner that adequate overlap is obtained. To deter- 

 mine whether the noise level of the system is suffi- 

 ciently high to affect the measurement of the hydro- 

 phone noise level, observations are made with a re- 

 sistor connected in place of the test instrument. The 

 resistance is selected by trial to be small enough so 

 that the thermal noise generated in it is negligible. 

 Progressively smaller values are tried until there is 

 no observable change in the output noise level. 



RecehnngDirectivity Patterns. Directivity patterns 

 of transducers may be obtained rapidly and with good 

 angular accuracy by means of the rotator and record- 

 ing turntable assemblies described in Section 6.2.1. 

 Exploratory observations are first made to check the 

 angular orientation and to adjust the system gain 

 so that the pattern traced will lie within the bound- 

 aries of the recorder paper. Whenever possible, the 

 signal level at zero angle is adjusted to the upper 

 limit of the chart in order to utilize its full 50-db 

 range and to facilitate subsequent chart comparisons 



