136 



USRL TEST STATIONS 



and a tentative design is partially completed. The 

 curve tracer mechanism is based on a light-beam and 

 photoelectric cell null-balancing scheme. 



Transient Wave Analyzer 



In the study of wave forms, the use of the Henrici 

 analyzer for the measurement and analysis of tran- 

 sients has proved both time-consuming and expen- 

 sive. Therefore, USRL later developed an instrument 

 which not only may be constructed from inexpensive 

 and easily available parts but also speeds up the proc- 

 ess of analysis. (See Figure 81.) This transparent 

 cylinder is rotated by a motor at about 1,800 rpm 

 running between a light source on the inside and a 

 narrow slit outside. The light passing through the 

 slit impinges on a photoelectric cell of the vacuum 

 photomultiplier type. The associated tube circuit for 

 amplification is shown at the right of the figure and 

 the power supply at the left. The usual recording sys- 

 tem for any steady-state signal such as system 2 is 

 used beyond this point. 



In operation, an oscillogram of the transient is re- 

 produced as an opaque stencil and attached to the 

 surface of the cylinder. It is then rotated between the 

 light source and the photoelectric cell producing an 

 electric signal corresponding to the transient which 

 is repeated some thirty times each second. The en- , 

 velope of the amplitudes of the Fourier components 

 of this signal is proportional to the spectral distribu- 

 tion of energy in the transient. This em elope may be 

 obtained directly as a function of the frequency by 

 using system 2 with the 300-c band which will average 

 several adjacent harmonics. The record obtained will 

 be independent of the lowest Fourier frequency as 

 long as it is small compared to 300 c, since changing 

 the frequency and amplitude of the linear sweep cor- 

 responds to changing the scale factor in a Henrici 

 analysis. (See references 58 and 79.) Calculations 

 made from this record are identical with those from 

 the Henrici analyzer. 



After a careful adjustment to eliminate distortion, 

 several transient sounds from Navy devices were ana- 

 lyzed by this method. The results were carefully 

 checked to determine their validity. The accuracy of 

 the reproduction was tested by viewing it on the 

 CRO, and the broad-band rms level of the original 



transient was compared with the one delivered by 



this apparatus. As an additional check, a square wave 

 was analyzed. Since the Fourier analysis of this wave 



is mathematically known, the analysis of the instru- 



Ficurf. 81. Optical signal generator of apparatus for 

 transient analysis. 



ment cotdd be readily compared with the theoretical 

 values. Good agreement was found to exist between 

 the two. 



There are minor improvements which could be 

 made, but the instrument as it stands is workable 

 and has adequate precision. The speed and facility 

 in the analysis of pulse signals have been very much 

 improved. Obviously, the method may be applied to 

 electrical pulses from any source. 



If no analyzer such as system 2 is available, the am- 

 plitudes of the Fourier components may be deter- 

 mined with a commercial electric harmonic analyzer. 

 The phases of the components, however, cannot be 

 determined by either method. 



Acoustic Phase Measurements 



For a more complete characterization of transduc- 

 ers, an instrument is desired which measures the 

 phase between the acoustic signal and the electric 

 signal. This would be of advantage, particularly in 

 the analysis of transient wave forms. 



Phase bridges are available" 07 ' 1 that measure the 

 relative phase between two electric signals. This 

 limits measurements in acoustic tests to the difference 

 in phase between the current into a projector and the 

 voltage generated by the hydrophone. However, the 

 reciprocity relation of a transducer indicates that the 

 phase shift between the current and the generated 

 pressure when acting as a projector, minus the phase 

 shift between the applied pressure and the open-cir- 

 cuit voltage when acting as a hydrophone, is either 

 180 or degrees for a magnetostrictive or a piezoelec- 



