104 



the impedance of the sensitive element can be kept low. High quality hydro- 

 phones operating at high frequencies with low self noise are normally designed 

 to be resonant hydrophones operating in the neighborhood of resonance. At 

 frequencies below 30 cps there is no considerable body of quantitative experience 

 at sea. The principal problems in this region appear to be in the proper quiet 

 handling of the hydrophone rather than electronic design. 



Whenever directionality is desirable in the hydrophone system, which is 

 frequently the case, single hydrophones having large directivity indices are 

 available only in the region above about 10 kc. Of this type a well known exam- 

 ple is the early World War II QBE, a directional transducer designed for an 

 echo ranging system operating at 24 kc. It has all the disadvantages of a tuned 

 system, but is not so sharply peaked that it is completely useless at other fre- 

 quencies. For the low frequencies the designer is on his own and may construct 

 any of a wide variety of hydrophone arrays to suit his particular requirements 

 from the simple "figure 8" pattern of a two element array to the sharply direc- 

 tional multi-element arrays. There is no circuitry generally available with any 

 considerable body of experience behind it for the use of arrays in the low fre- 

 quency field. This lack is becoming a serious shortcoming as field observa- 

 tions and understanding of this part of acoustical research progress. 



The choice of amplifier and recorder used in taking field data is deter- 

 mined by a number of factors, some of which are not the least bit obvious at the 

 outset except to experienced investigators who have suffered from past failures. 

 For example, shall, a high quality recording of the sound be attempted for rela- 

 tively short periods, or will it be more advantageous to obtain a recording of 

 lower quality which can, nevertheless, be made continuously over long periods 

 of time? Is it wise to use some photographic process for recording that re- 

 quires the records to be developed a considerable time after the tests? High 

 quality records may result but you are assured that you won't know your data 

 until it's too late to change a faulty procedure or try a new idea. Shall rather 

 complete reliance be placed on the "quick and dirty" answer to be obtained from 

 visual observations on a sound level meter or perhaps some such crude record- 

 er as the type which records sound intensity as relative blackness on a chemical 

 recording paper? The list could be continued. The state of understanding in 

 the particular problem and the object of the particular observations are usually 

 supposed to be the principal determining factors. Actually acoustical research 

 groups as well as individuals seem to have developed rather arbitrary prefer- 

 ences in recording and subsequent analysis. For such standards as there are, 

 underwater acoustics is wholly dependent on such related fields as air acoustics 

 and seismology. 



In passive listening one is usually interested in the quality as well as the 

 intensity of the sound, and earphones or loudspeakers are apt to be used for 

 observations somewhat more than an output meter. Except for the crudest sort 

 of tests it is desirable to make some sort of recording that can be played back 

 at a later time for analysis. The usual hydrophones, such as those mentioned 

 above, have so low an output for sounds ordinarily to be heard that some sort of 

 pre-amplifier is required between the hydrophone and any commercial record- 

 er. While it appears on the surface that a stage or two of amplification with 

 perhaps a third stage of cathode follower for matching is all that is required for 

 the job, it has been our experience that a sufficiently versatile pre-amplifier for 

 more than a very restricted use ends up as a rather high gain amplifier incor- 

 porating within itself provision for covering a very wide dynamic range and a 

 wide choice of filtering, as well as built-in conveniences for electrical calibra- 

 tion. Such an amplifier is not required to take high quality observations, but it 

 is a great time-saver in a continuing and somewhat varied acoustical research 

 program. We started the development of such an amplifier in 1947 and called 



