Most people think of sonar, or asdic as it has traditionally been called in Britain, only in its wartime antisubmarine role.

There are, however, many uses for sonar in nonmilitary fields such as fisheries research and operation, hydrographic surveying and geophysical and oceanographic research.

Sonar and underwater acoustical engineering

of underwater acoustics were, until comparatively recently, largely restricted to naval operations. It was during the 1914-18 War that a system of underwater echo ranging was developed under the name asdic (said to represent the initials of the Allied Submarine Detection Investigation Committee). A better term than echo ranging, though less used, is echo location. The principle was that a pulse of sound was transmitted into the water, and any reflection (echo) from a submarine or other underwater object was received by a hydrophone, which is the underwater equivalent of a microphone. The received signal could be heard on headphones, and the time delay between transmission and reception used as a measure of the range of the submarine. If the sound transmission could be made directional, an idea of the direction of the target could also be obtained. If the sound pulse was directed vertically downwards, an echo would be received from the sea bottom, so that the depth of water beneath a ship was thus easily determinable. Such a system acquired the name echo sounding, and gradually came into use in purely civil applications as a navigational instrument. Since the Second World War, asdic systems have come into ever-increasing civil use in various fields of activity, notably in thefishingindustry both as a research instrument and as an everyday tool for locating fish shoals for catching. Other applications are in whale catching, in hydrographic surveying, in oceanography, and in civil engineering (for observing scour around piers etc.). Ranges obtainable vary from a few metres to many kilometres, and frequencies from a few kilocycles per second to nearly 1 Mc/s.
APPLICATIONS 220

Underwater communication is possible by acoustic waves a good deal more efficiently than in air. It is usual for underwater communication systems to use a modulated carried wave. Successful systems have been developed for both speech and telemetry (i.e. instrument-data transmission), particularly over relatively short ranges, say a few hundred metres. Speech systems are important in, for example, communicating with divers; data High-power pulses An application of underwater systems are important in, for example, acoustics that has become more and oceanographical research, where inmore important in connection with struments may be placed in midwater prospecting for oil, as well as other or on the ocean floor. All the applications of underwater geophysical research, is the use of high-power pressure pulses (e.g. gen- acoustics mentioned above are active erated by a small explosion) for systems, in which a source of acoustic exploring the sediments and rocks energy forms part of the system. There underneath the sea. In the simplest are, however, many uses for a passive conception of such a system, we may system, which involves only 'listening' think of a very-high-power echo to acoustic emissions from indepensounder; the acoustic waves are dent sources. These sources may reflected, of course, from the sea include various kinds of marine bottom, but, owing to their high organism (some species of fish, parpower, penetrate the sediments and ticularly tropical ones, emit strong and are reflected also from any discon- characteristic noises) or manmade tunities in thesee.g. from the boun- noises, e.g. from cavitating propellors. dary between a sediment and a hard It will be clear that the term rock. 'acoustic' must be taken to include In a more complicated system, the whole range of frequencies from
Electronics & Power July 1965

In the United States, the term asdic became displaced by the more obvious term sonar, which is derived from sound navigation and ranging. This term lines up very suitably with the analogous electromagnetic echo-location system radar, which was developed in Great Britain between the wars (and is thus a good deal more recent than the underwater acoustic system). It is becoming evident that the name 'sonar' is rapidly displacing 'asdic' throughout the world, and, since 'asdic' has rather narrow naval implications, it is now far preferable to use 'sonar', which is a more general term, for all civil applications. Sonar systems occur in nature. The porpoise has a very effective sonar system by which it navigates itself. In air, many species of bat use sonar in place of vision, and some show a degree of refinement and accuracy that is scarcely credible and not yet fully understood.

the receiver may be remote from the source, so that waves detected by the former may travel by different paths, e.g. directly through the water, or partly through the sea bottom. Since the velocity or propagation is different in different media (being much higher in solids than in water, and being higher in some rocks than in others), it is possible to sort out the various components of the received signal and thus learn a good deal about the nature of the sediments and rocks below the sea.
Modulated carrier wave

zero up to many megacycles per second and is thus more general than the term 'ultrasonic' which seems to exclude the frequency range below, say, 16kc/s. It is perhaps necessary to point out, at this stage, that the importance of acoustic waves for underwater detection and communication systems is mainly due to the fact that electromagnetic waves (which, in air, are more useful than acoustic waves over any distance in excess of, say, 100m) hardly propagate at all in water except when their wavelength is so great (e.g. several kilometres) that they are useless for forming directional beams and resolving the usual objects that have dimensions of the order of a few metres or even centimetres.

ships' track

depth

Sonar systems The majority of sonar systems, like most radar systems, use the transmission and reflection of a pulse of 2 Typical echo-sounder records taken on the chemical recorder, showing fish above energy as their basis. The system is the sea bottom typified by the arrangement shown in Fig. 1. Individual systems show many variations, such as the use of a single electroacoustic transducer for both received information on sensitised trated in Fig. 2, where two portions of transmission and reception, or the use paper, which may be a dry type or a record taken with a vertical-beam of an oscillatory discharge from a damp type, in which the paper is echo sounder are shown. The solid capacitor into the transmitter so that impregnated with potassium iodide. part represents the echoes from the sea no separate continuous oscillator and In this instrument the time base is bottom, and as the ship has steamed gating unit is required. The pulses are mechanical; the recording stylus is along, the edge of the dark marking always bursts of carrier frequency, drawn across the paper. Since the mark portrays the profile of the bottom although the extension of the term is made when an echo is received, the along the ship's track. (The reason why 'sonar' to include marine seismic work position of the mark across the the bottom echoes extend so far down using explosive pulses is becoming traverse indicates the range of the the depth scale is simply that the sound more common. target. The paper is moved slowly in a beam was very wide, and echoes The distance from the transducers direction perpendicular"to the traverse received at angles other than the to a particular reflecting object (or of the stylus, so that successive vertical obviously travel a greater target) is indicated by the time traverses lie side by side. If the range distance and mark at a later instant on elapsing between transmission of the of the target from the transducers does the stylus traverse.) The small dark pulse and reception of the echo. The not vary, the line is produced parallel patches above the sea bottom repredisplay therefore always includes a to the direction of motion of the paper. sent the echoes from largefish(actually time base, the traverse of which is If the range changes, e.g. owing to the cod of length approaching lm). The initiated by the transmitted pulse. For motion of the target, or of the ship on depth of water was about 220 m, simple systems of the type shown in which the equipment is fitted, the line but the zero of the recorder has been offset so that the chart covers a range Fig. 1, the commonest type of display is sloped relative to the paper motion. is the chemical recorder, a remarkably The kind of record which is ob- of depth of only 100 m. efficient instrument which records the tained on a chemical recorder is illusThe upper record was taken in daylight, and it can be seen that the fish form agglomerated shoals. The lower record was taken at night in the same locality, and the fish are clearly seen oscillator to to be spread out in a diffuse shoal. generate gating transmitting amplifier carrier unit It will be observed that the trace followed by the stylus is a circular arc. pulse It is obviously very convenient to .of sound energy travelling carry the stylus at the end of a rotating through arm, and this is the most common water arrangement. But there are also various kinds of straight-line recorder.
target
1 Schematic of a typical simple sonar system: most sonar systems, like most radar systems, are based on the transmission and reflection of a pulse of energy

P.P.I, technique

display

A quite different kind of display can be obtained by using the p.p.i. (planposition indicator) technique, which is the normal thing in radar but is not very common in sonar. This uses a
221

Electronics & Power

July 1965

cathode-ray tube, and the trace corresponding to a pulse travelling out along a beam in a particular direction is displayed radially from the centre of the tube, with the direction of the trace corresponding exactly with the direction of the axis of the beam relative to some datum axis, which may be either a compass bearing or the fore-and-aft axis of the ship on which the equipment is fitted. As the transducers are rotated to search in different directions, a true plan of targets is displayed. But this kind of display is not much used in sonar, because the velocity of sound in water is too low to permit an allround search to be made in a reasonably short time. It is clearly necessary to wait until a transmitted pulse has travelled out to the maximum range and any echo pulse has returned before rotating the transducer to a new position. With a beamwidth of 5 and a maximum range of 750 m, say, it would require a minimum of 72 s to sweep all round. Another kind of sonar system (as distinct from merely a different display), which appears to correspond to that used by certain species of bat, is the frequency-modulated system, but this is little used in practice and will not be discussed any further here. In the simple systems outlined above, a single beam is produced. To search a large sector, however, requires the transducer(s) to be swung mechanically so that successive transmissions, or groups of transmissions, look in different directions. As already pointed out in connection with p.p.i.type displays, the rate of search is very slow in comparison with the corresponding rates of radar, since the velocity of sound in water is only about l-5km/s.

Electronic sector scanning

It is being found increasingly necessary to raise the rate of scan, and electronic sector-scanning systems have been developed for this purpose. In such a system, the whole of the sector to be examined is insonified by a transmitted pulse which comes from a wide-beam transducer. A relatively narrow receiving beam is swung by electronic means across the sector within the time required for the pulse to travel its own length in the water. All directions within the sector are therefore effectively sampled on the one transmission. The schematic of the arrangement of one particular electronic scanning system is shown in Fig. 3. The receiving transducer is n times the length of the transmitting transducer, and is divided into n sections, where n is the number of beamwidths (as measured between points where the power response has fallen to half) it is desired to contain in the scanned sector. If these n sections were connected to n corresponding uniformly spaced taps on the delay line, the beam would be deflected by an amount dependent on the phase shift in the delay line. In the scanning system, frequencychanger equipment is inserted between the transducer sections and the delay line. The local oscillator, which feeds all the frequency changers, is swept in frequency by the bearing time base, so that the signal frequency received by the delay line varies over a range during every sweep of the bearing time base. If then the delay line is made to have a phase shift which varies over this frequency range from negative values to positive values, the beam will be swept from left to right during each sweep of the bearing time

base. The latter also deflects the spot on the cathode-ray tube from left to right, so that signals received on any particular bearing are recorded on that bearing on the display. The range time base works in the usual way, so that the position of an echo spot on the tube indicates the position of the echoing object on rectangular axes of bearing and range. (This kind of diaplay is called the B scan.) If the bearing scan is so rapid that it is completed within the duration of the pulse and is immediately repeated, no information is lost, and all directions in the sector are effectively looked at simultaneously, but with the angular resolution corresponding to the beam-width of the receiving transducer. A typical result using a system of this type is shown in Fig. 4.
Transducer design

transmitting transducer

frequency control a.g.c. I amplifier detector intensity modulation range ^transmitting oscillator Jand pulsing unit range time base

La
222

bearingTL. time base

' bearing

While the design of the electronic part of a sonar system is usually straightforward when the principles are understood, the design of the electroacoustic transducers is often quite difficult. Although there is a good deal of theory relating to transducers of various kinds, the limitations of practical materials are such that design has inevitably a very large empirical element. The three main types of transducer in use for underwater applications are magnetostrictive piezoelectric electrostrictive. Electrostrictive transducers are becoming much more widely used and seem likely to displace the other types for the majority of applications. Examples of electrostrictive materials used for transducers are barium titanate and lead zirconate in ceramic form. The change of dimensions is dependent on the magnitude but not on the polarity of the applied electric field, and thus polarisation is needed as with magnetostrictive transducers. Electrostrictive transducers generally have very convenient impedances of a few hundred ohms. With all these types of transducer it is possible to achieve efficiencies of up to 50% and Q factors as low as about 5. The Qfactor should be low, as it controls the bandwidth of the system. The beamwidth of a transducer is 3 Schematic showing the basic prin- inversely proportional to the dimenciples of an electronic sions of the transducer (measured in sector-scan ni n g wavelengths) on the appropriate axis. sonar system: a relatively narrow receiv- Thus, at low frequencies, severe ing beam is swung directional requirements necessitate by electronic means very large transducersor more accuacross the sector rately a large array of transducer within the time required for the pulse elements. For this reason, low-freto travel its own quency echo sounders, which normally length in the water operate at about 15kc/s and obtain
Electronics & Power July 1965

4 These four photographs are of the display of the high-resolution electronically scanned sonar during the observation of fish movements around the cooling-water intake at Kincardine power station. They were taken at short intervals, and show clearly the movements and breaking up of a small fish shoal. The marks along the top edge represent fixed piers supporting the intake grille, and the scattered marks at the right-hand side are due to the shingle on the river bottom. Each picture represents a kind of plan view, since the horizontal axis represents distance from the transducer (the whole width representing 20m) and the vertical axis represents bearing angle (the whole height representing 30). The resolution is about 7cm in range by about 0-5 in bearing

research is necessary into the behaviour of very narrow beams (e.g. 0-1), since it is already known that scattering and refraction by inhomogeneities such as turbulence cause a widening of the beam which is obviously greater at greater ranges. For longer-range sonar systems and for seismic systems, the development of low-frequency high-power highly directional acoustic sources is very desirable, and the possibilities are being explored of producing large virtual arrays (giving high directionality) by nonlinear interaction between two coincident narrow high-frequency beams or by nonlinear response to a single modulated high-frequency beam. Similarly, parametric action in the water may offer a low-frequency high-gain highly directional receiver necessitating only small high-frequency transducers.
Radio and radar parallels

ranges of several kilometres, operate with rather wide conical beams of perhaps 30. High-resolution sonars, such as those used for studying the movement and behaviour of fish, need transducer arrays of tens or even hundreds of wavelengths, and therefore operate on much higher frequencies, e.g. 400kc/s, with ranges of perhaps 100m.
Problems of propagation

Although the design of a sonar system, as described above, involves many difficult problems and is the subject of much research, the greatest problems in the use of sonar undoubtedly arise in connection with the propagation of acoustic signals in the sea. (For sonar systems operating in air, corresponding problems exist but are much less well understood.) When the object of the sonar system is to detect small targets (as distinct, for example, from the sea bottom), the target signal has to be detected against a random background due to reverberation; this is the sum of all the numerous small echoes produced by back scattering from sand and stone particles on the sea bottom, from minute air bubbles and other inhomogeneities in the water, from waves on the surface and so on. Although background noise in the sea, e.g. noise due to waves breaking, may limit the maximum range of detection, reverberation on the other hand may be a limitation of performance at all ranges. This is because it arises from the signal transmission
Electronics & Power July 1965

and, at any time interval thereafter, has a power level closely related to that of the signal. Thus for detection of small objects in areas where reverberation level is high, owing for example to a shallow sea bottom, it is essential to use beams that are as narrow as possible in the relevant dimension. Because reverberation is normally the limiting factor in detection, an increase of transmitted power is not effective in improving detection over most of the range; consequently most sonar systems operate at acoustic powers of fairly low peak level, e.g. lW-lkW.
Present research and future developments

The present position in civil sonar is that individual small fish (say 15 cm), not too close to the sea bottom, can be detected and displayed with the electronic sector-scanning sonar now going into commercial production, at ranges up to about 100 m using a frequency of 400kc/s, pulses of 50 /AS duration and a receiving beam of 1 width. This performance is probably adequate for ordinary studies of fish behaviour. To achieve much longer ranges, such as are necessary in fish catching, either very much lower frequencies are necessary (with consequently much wider beams, and detection and resolution only of shoals instead of individualfish)or very much narrower beams (requiring larger transducers) and higher powers may be used to maintain high resolution at longer ranges at the high frequency. Thus

The advantages of digital signalprocessing are worth studying further, as simpler, more reliable and more flexible electronic equipment may be obtained in this way; there is also the possibility of presenting the target information in numerical form (e.g. target strength and positional coordinates). Development of improved signal processing is taking place in other directions also. Sonar systems with a very wide bandwidth (e.g. 10 : 1 frequency ratio) are still in the development stage, and the problems of designing (indeed, even of conceiving) transducers that are efficient and uniformly directional over a very wide frequency band have still to be solved. Such systems could display not only the range and bearing of targets, but also their frequency response. There is reason to think that this would be very helpful in the identification of fish species. Research into propagation problems continues all the time, of course. Fluctuations in received signals owing to turbulence, thermal effects, multiplepath propagation etc. limit the performance of sonar and other underwater communication systems, and must therefore be understood and, if possible, overcome. The special features of propagation and reverberation over long ranges in certain kinds of situation, such as under ice, also receive attention. In many respects, underwateracoustic research parallels radio and radar research, as may be seen, for example, from the last paragraph. It has for long been my view that research (and even some aspects of development) in the two fields should proceed hand-in-handas indeed it does in my own university department.
223