ANYONE WHO HAS SPENT ANY TIME swimming underwater will recall how quiet everything seems beneath the surface. Even in waters that teem with fish, or where currents cause marine plants to vibrate with obvious violence, few sounds are heard by the swimmer other than the faint rush of blood inside his own head. The illusion of submarine silence is explained by the ineffectiveness of the human ear as a detector of underwater sound, a fact that became apparent with the invention of the hyrdrophone. This instrument disclosed that all of the oceans, many lakes, rivers even brooks are filled with sounds least as varied and interesting as those in the air. Few amateurs have been privileged to hear or record submarine sounds, however, because hydrophones are not generally available. This barrier is now removed. Frank Watlington, a specialist in underwater sound-detection who is currently abroad on a Government assignment, has designed a hydrophone that amateurs can build without special tools at a cost of less than $10. In the following discussion he reviews some of the acoustic phenomena opened to amateur investigation by the hydrophone and explains how to build and use the instrument.

"Underwater sounds," Watlington writes, "were opened to investigation by the development of electronics and associated technologies. Instruments have been devised that are far more effective than the ear for the perception of mechanical vibrations. With these 'transducers' it became apparent that the oceans and lakes are not the silent world that they were once considered to be.

Submarine sounds originate in disturbances caused by physical phenomena, by marine animals and by men. They are called ambient noise when the sources are not identified.

"The water noises that originate from physical phenomena may be induced by the breaking of surf on the shore, the action of tidal currents on stones or small pebbles, the movement of water in eddies through reefs and rock pockets, by earthquakes, volcanoes and even by whitecaps in the open ocean.

"Marine organisms are sometimes extremely vociferous. The croaker, a fish that abounds in both the Atlantic and Pacific, makes a noise like a persistent woodpecker. Classified as a drumfish, it is equipped with a gas-filled bladder. It collects in certain areas in large numbers during the summer months, at which time little else can be heard. The noise emitted by the croaker is produced by the vibration of a muscle beside the swim bladder of the fish. A similar noise can be produced artificially in air by rubbing a toy balloon with a slightly moistened finger. The snapping shrimp makes a noise resembling the snapping of fingers. The shrimp has one prominent claw, and it makes this noise by clicking the claw rapidly. There are areas where the population density of these animals approaches 20 per square foot; in such numbers they produce a high-pitched roar that sounds like a giant pan of frying bacon. A weird wailing sound heard off Bermuda during March and April is attributed to hump back whales migrating in large numbers to northern waters, usually with their calves. The sounds are not unlike ' those from a herd of cattle, but are characterized by a greater range in pitch. No one knows how the whale produces these sounds, but they have aroused much interest.

"These are only a few of the many sounds heard in the oceans of the world. Why and how do marine animals make them? It is possible that some animals have evolved a method of echo-ranging on their prey in the same way that ships measure the depth of the water: a sound is emitted from the ship's bottom, and the time required for this sound to travel from the ship to the bottom and back is a measure of the depth. It is also possible that some marine-animal sounds represent mating calls or even the communication of information.

"Among man-made sounds the most prominent come from ships. These originate mostly in the propellers and the propulsion machinery. Other frequently heard man-made noises are emitted by explosions, by machinery on barges, by pile drivers and in some locations even by vehicular traffic on land.

"A sound is anything that the ear can hear. But many vibrations are emitted by mechanical disturbances above and below the range to which the ear is sensitive. Technically speaking, sound can be defined as a longitudinal disturbance traveling progressively from particle to particle through a material medium.' The medium may be a solid, a liquid or a gas. As is well known, sound travels in the form of waves. A long wave is heard as a sound of low pitch, and a short wave as one of high pitch. A complete wavelength is called a cycle, and is the distance measured from the crest of one wave to the crest of the next. If the frequency is below 16 cycles per second or above 16,000 cycles per second, it cannot be heard by most people.

"The numerical product of frequency and wavelength in a given medium is the velocity of the sound in this medium. At 68 degrees Fahrenheit the velocity of sound in air, for example, is 1,127 feet per second, and that in water is 4,900 feet per second.

"The three principal characteristics of sea water that influence the speed of sound are temperature, pressure and salinity. Of these the most important is temperature. Warm water is less dense than cold water. Consequently warm water floats above cold water. As an underwater swimmer moves from the surface toward the bottom, he often encounters a region that is distinctly colder than the region above. Sounds are refracted by this temperature gradient, because a sound vibration traveling from warm water into cold water is slowed down. The action is analogous to a line of marching soldiers proceeding from a hard-surfaced path to a sandy beach. If the direction of march is oblique to the soft surface, the soldiers who first encounter the sand are slowed down, and the line of march takes a new direction. Because of refraction effects, it is not always easy to predict the location of an underwater source of sound merely by observing the direction of the received waves. Much interesting work remains to be done in the field of underwater sound ranging

"Numerous transducers have been developed to sense the sound of the sea. They parallel the instruments used for the same purpose in air.

"A transducer, or hydrophone, that lends itself nicely to home construction is based on the fact that the magnetic properties of certain materials (e.g., nickel) vary when the material is subjected to mechanical stress. The phenomenon is known as magnetostriction, and can be demonstrated by wrapping a coil of insulated copper wire around an appropriately shaped core of nickel. When current flows through the coil, a magnetic field is induced in the nickel. Many of the molecular magnets of the nickel immediately rotate into alignment with the induced field, and they retain this orientation after the field is removed. The nickel thus becomes weakly but permanently magnetized. If the core is now stressed mechanically, some of the molecular magnets are forcefully rotated away from the direction of the permanent field and no longer contribute to its strength. In other words, stressing a permanent magnet of nickel weakens its field. When the stress is removed, the disturbed molecular magnets return to their former orientation, and the magnet regains its strength.

"The effect can be observed by connecting a voltmeter of adequate sensitivity across the terminals of the coil and stressing the magnetized nickel core. When stress is applied, the field strength drops and induces a voltage in the coil. When the stress is removed, the field strength increases to its normal value and induces a second voltage pulse (of opposite sign) in the coil. The physical deformation of the core need not be large: a deformation measured in thousandths of an inch is more than sufficient to cause the observed effect. The corresponding voltage induced in the coil will vary with the number of turns of the wire comprising the coil. In the case of small cores wound with a few hundred turns, the potential is measured in millionths of a volt. But this minute voltage can be amplified by any desired amount. If one wants to hear sounds in the sea through a loudspeaker, the voltages from the usual magnetostriction hydrophones must be amplified from 100,000 to a million times, depending upon the characteristics of the instrument.

"Any amplifier having an input for either a dynamic (moving coil) microphone or a variable-reluctance record-playing cartridge will have sufficient gain to operate from a magnetostriction hydrophone. Any high-fidelity amplifier or modern tape-recorder that includes a preamplifier should do nicely.

"Nickel is only one of several materials that exhibit high magnetostriction. Others include 2V Permendur and various alloys containing nickel and iron. The magnetostrictive properties are affected by the gross composition, by impurities, by strains incurred in fabrication and by the crystal structure. 'Grade A' nickel is 99.45 per cent nickel, .25 per cent manganese, .15 per cent iron, .005 per cent sulfur, .05 per cent silicon and .05 per cent copper. It can be purchased in the form of rods, sheets or tubing. In some eases alloys obtained from commercial sources already have adequate magnetostrictive properties, but if adequate properties do not exist, annealing will generally provide them. A 'soft anneal' from the mill will provide moderately good magnetic properties, but for the best results additional annealing is usually required, for this will relieve strains that have been incurred during fabrication.

"The most effective magnetic circuit is one that has no air gap. That is, there is no break in the metal. If there is no air gap, most of the magnetic flux will be confined to the metal, which has a higher permeability than its surroundings, usually air. A piece of tubing meets this requirement if the coil is wound on it in the form of a toroid, as shown in the accompanying illustration. Commercial magnetostrictive hydrophones usually employ tubular construction because they are relatively easy and inexpensive to build. Most are priced in the vicinity of $100. These hydrophones have a pressure sensitivity of around -80 decibels with respect to one volt per dyne per square centimeter at 1,000 cycles per second; this means that if a pressure of one dyne per square centimeter deforms the magnetic material at the rate of 1,000 cycles per second, 100 micro-volts will appear across the terminals of the coil at a frequency of 1,000 cycles per second. This may seem like a very small voltage, but it is more than adequate for underwater-sound detection. A satisfactory hydrophone can be constructed with Grade A nickel tubing if the metal is annealed in an ordinary kitchen oven at a temperature of approximately 600 degrees Fahrenheit in air for three hours. The heat is then shut off and the tube permitted to cool in the closed oven. (The material should be cleaned of all grease in advance of annealing.) Annealed nickel tubing as shipped by the manufacturer will exhibit a sensitivity not too far below optimum. Tubing for the instrument to be described can be purchased from the Edmund Scientific Company of Barrington, N. J.

"The edges of the tube as received from the supplier are usually sharp. To protect the winding from damage it is necessary to cover them. The easiest way to accomplish this is to coat the ends of the tubing with a heavy varnish such as Insl-X, the preparation commonly used for coating hand tools such as insulated screwdrivers and pliers. This material and all others required for the construction can be ordered through a local hardware or electrical-supply dealer. Arrange the tubing for dipping as shown in the accompanying illustration [above]. The can of Insl-X is raised under the tubing until the end is submerged to a depth of about a quarter of an inch. Watch for bubbles and break with a matchstick any that form. Two dips (with a drying interval of one hour between) will provide a coating of adequate thickness.

"A bobbin for winding the toroid coil on the tubing is made-from a piece of sheet aluminum about a 16th of an inch thick as shown in the accompanying illustration. Cut the channels with a hacksaw and round off all corners and burs with a file followed by sandpaper. The small sections of scotch tape applied as shown provide a cushion for the wire.

"The toroid coil is wound by hand. Transfer 200 turns of No. 80 AWG single silk enamel magnet wire to the bobbin. To facilitate the transfer, insert the blade of a screwdriver in the hole through the center of the spool and clamp the end of the blade in a bench vise. This frees both hands for the tedious winding job. Tape the end of the wire to the center of the bobbin and wind with a hand-over-hand motion. Be sure that no kinks or twists occur. No. 30 wire is recommended, but the size is not critical. If smaller wire is used, greater care must be exercised to prevent breakage; on the other hand, larger sizes reduce the number of turns that can be wound on the tube, and therefore limit the output voltage.

"When the bobbin is loaded, secure the outer end of the wire to the nickel tube with a small piece of scotch electrical tape. Then transfer the wire to the nickel tube by successively threading the bobbin through the tube as illustrated. The threading operation is continued until the tube has been wrapped with one full layer. Be sure that each turn is placed as closely as possible to the preceding one and that all turns are parallel to the axis of the tube.

"When the winding is in place, clean each end of the wire with a small piece of sandpaper as shown [below]. Enamel insulation can be deceiving. It resembles bare copper. Clean copper shines brightly. Failure to clean the wire invites a faulty connection and the risk of having to rebuild the unit after it has been sealed in its watertight housing.

"Stranded hook-up wire of the type used for building radio receivers will suffice for flexible leads. Prepare two 18-inch lengths by stripping a quarter of an inch of insulation from each end. Use red insulation for one and black for the other.

"Connect the leads to the coil ends and solder, using rosin flux. Insulate the joints with a two-inch length of closely fitting plastic radio spaghetti. Tape the leads securely to the nickel tube as shown, using scotch electrical tape, and check the unit for continuity. A continuity test can be made with an ohmmeter or with a battery and flashlight bulb. The direct-current resistance of the completed coil should come to about 30 ohms.

"Use the leads for suspension and dip the entire coil in the Insl-X varnish. This secures the wire to the tube. (The vibration of a loose turn will cause noise.) Hang the unit by its leads to dry, one end down, for at least an hour. Break any bubbles that form.

"While the varnish is hardening, cut a piece of cork gasketing material from 16th-inch stock (such as Armstrong cork DC100) so that when it is rolled into a cylinder with butted ends, the cylinder makes a snug fit with the inner surface of the coil. The air pockets of the cork cells provide a cushion inside the tube so that pressure exerted on the unit by fluid will deform the tube inwardly. Without this provision, variations in sound pressure would exert equal force on the outer and inner walls of the tube, no movement of the metal would occur and no signal voltage would be generated. In effect the gasket material creates the essential pressure differential across the walls of the tube. Unfortunately it also limits the depth to which the unit may be submerged. Beyond a critical depth, which varies with the strength of the gasket material, the static pressure of the water crushes the cells of the cork and destroys their cushioning effect. This renders the unit inoperative. The accompanying illustration shows the assembly with the gasket in position.

"A final continuity check and another over-all coat of Insl-X varnish complete the assembly. The core must now be magnetized. This is accomplished by passing a direct current of approximately 1.5 amperes through the coil for a period of about half a second. I use a heavy-duty 45-volt battery as the power source. If the continuity test has been made with a lamp and battery, the core will have been partially magnetized by the incidental magnetic field, and it is therefore advisable to connect the magnetizing circuit so that the current flows in the same direction as during the test. The coil absorbs power at the rate of about 100 watts during the magnetizing interval, considerably beyond the capacity of the wire to dissipate energy with out dangerous heating. If the magnetizing current is permitted to flow for more than half a second, the coil may be damaged.

"The pickup operates in a bath of castor oil. (Do not use substitute oils. The electrical properties of some are inferior to castor oil, and others attack the insulating materials chemically.) Sound vibrations are transmitted to the nickel core through the oil. For maximum sensitivity the container must therefore be completely filled. Care must be taken to exclude bubbles from the oil. This can be accomplished by completing the final assembly with all parts fully submerged in oil. Fill an open container, such as a saucepan, with enough castor oil to cover the unassembled parts. The pickup unit is then inserted part way into the housing and immersed in the oil. The parts are manipulated while submerged to assure the escape of all bubbles, and the nozzle cap is screwed home.

"The unit is now removed and wiped clean. Seal the joints at the end of the nozzle and between the cap and the body of the jug with a wrapping of rubber tape and cover the rubber tape with scotch electrical tape. A connector that mates with the input of the amplifier is attached to the free end of the microphone cable. The hydrophone is now ready for use.

"The sensitivity of the instrument varies inversely with the frequency and is low in the audio range (16 to 16,000 cycles per second). The output rises at the rate of about six decibels per octave. For this reason you will need high amplification for satisfactory listening in the audio-frequency range. The home high-fidelity amplifier designed to operate from a variable reluctance cartridge will have enough amplification.

"A 50-ohm 'line-to-grid' transformer is used to match the impedance of the hydrophone to the input of the amplifier. Connect one terminal of the secondary coil directly to the grid of the input tube and ground the other. Connect the hydrophone across the 50-ohm coil. Tape recorders that have an input for a low-level microphone will also have sufficient gain to operate from the hydrophone. Usually these recorders include a transformer for low-impedance transducers and can therefore accept signals directly from the hydrophone. In either case the chassis must be well grounded to reduce 60-cycle hum and other background noise.

"Lower the hydrophone into the nearest river or off a convenient pier. Listen for passing motorboats and noises made by the local fish population. If there is an aquarium in your locality, perhaps the director will allow you to record the noises in his tanks. He will have to shut off the air or circulating water in the tanks during intervals of listening, or the background roar will mask the fish noises."

Bibliography

THE CHARACTER AND SIGNIFICANCE OF SOUND PRODUCTION AMONG FISHES OF THE WESTERN NORTH ATLANTIC. Marie Poland Fish in The Bulletin of the Bingham Oceanographic Collection, Vol. 14, No. 3, pages 1-109; 1954.

FUNDAMENTALS OF SONAR. Joseph Warren Horton. United States Naval Institute, 1959.

Suppliers and Organizations

The Society for Amateur Scientists
5600 Post Road, #114-341
East Greenwich, RI 02818
Phone: 1-877-527-0382 voice/fax

Internet: http://www.sas.org/