Constant effort and research are devoted to the development and improvement of naval sonar and associated equipment. As more efficient equipment is perfected old equipment is replaced as rapidly as possible consistent with production and money available. There is an unavoidable time lag between the development and installation of new equipment in ships of the fleet and inclusion of such equipment in a publication such as this. At the time you use this book, some equipment discussed in this and succeeding chapters may not be the equipment currently being used in the fleet. However in most cases circuits and operating principles will be similar to those found in current equipment. Equipment manuals for specific equipment should be studied thoroughly by personnel responsible for repair maintenance and operation of that equipment.


The model QGB sonar equipment, shown in figure 6-1, is typical of modern searchlight equipment. This type of equipment is designed for installation aboard destroyer escorts and destroyers. The QGB consists of a transmitter, a receiver, an indicating range recorder, a bearing deviation indicator, and a transducer with an associated hoist-train mechanism.

The system operates at a frequency determined by the resonant frequency of the magnetostriction transducer. Available frequencies are 20, 22, 23, and 26 kc. The transmitter and receiver both cover the frequency range of from 17 to 26 kc. To change the operating frequency of the system, and still have it operate efficiently, the transducer must be changed. Because magnetostriction transducers have sharply resonant characteristics, the system must be adjusted to the resonant frequency of the transducer.

The receiver, bearing-deviation indicator, indicating range recorder, range indicator, remote

  training control, and the keying control unit are contained in a console, which is the sonarman's operating station. The indicator panel of the console is shown in figure 6-2. This console is housed in the sonar control room. The transmitter and its power supply are contained in a cabinet in the sonar equipment room. The hoist-train mechanism is located in the lower sound room.

Receivers and transmitters will be discussed in chapters 7 and 8, respectively.

The transducer, with its training and hoisting mechanism, generally is installed so that the unit is parallel to the fore-and-aft axis of the ship. The transducer, which can be rotated through a maximum of about 840°-2 1/3 revolutions-operates over the entire port and starboard sectors of the vessel.

The transducer itself is mounted in a sea chest and the transducer and its hoisting units are raised and lowered inside this chest. When in the raised position, the sound dome seals off the sea chest so that water cannot enter the ship if the top of the chest were removed. Sealing off the chest in this manner permits installation, removal and servicing of the transducer and its training mechanism while the ship is at sea or in port. All other units of the system are accessible from the interior of the ship.

The bearing scale on the transducer shaft in the lower sound room is adjusted to read zero when the transducer is directed dead ahead. The bearing is remotely repeated at the operating position in the sonar control room by a synchro repeater system. The repeater on the operating console indicates either true or relative bearings.

The transducer is trained by rotating a handwheel on the console, geared to a control transformer. The output of this control transformer is the error signal for the training control amplifier that controls the amplidyne generator.


Pictorial diagram of the QGB system showing Transmitting, Driving and Training circuits.
Figure 6-1. -Pictorial diagram of the QGB system.

After the maintenance-of-true-bearing (MTB) feature is switched on any change in the ship's heading causes the transducer to rotate by the same amount but in the opposite direction. Thus, the transducer remains on the same true bearing regardless of changes in the ship's heading.

Keying of the equipment can be controlled by any one of the four means-(1) the range recorder located in the console of the equipment, (2) an external tactical range recorder, (3) a hand key on the console, or (4) a multivibrator that is a part of the keying unit.

The keying intervals are arranged and controlled as follows. When keying is controlled by the sound-range recorder, two scales are available-one at 1,500 yards and the other at 3,750 yards. To conserve recorder paper when there is no target, a multivibrator takes control. The multi-vibrator has two time rates-one at a scale of 3,000 yards and the other at a scale of 5,000 yards.

  During search operations one of the last two scales is selected depending upon sound conditions.

The keying interval and the conditions of manual keying or listening are selected by a rotary switch on the console.

The cycle of operation of the QGB is as follows: (1) A keying signal is delivered to the keying unit which energizes the keying relay. (2) The relay transfers the transducer from the receiver input to the transmitter output. (3) The transmitter impresses an r-f voltage of the correct frequency across the transducer at the proper energy level. (4) The transducer converts this energy into sound power, and emits it in a narrow beam along the bearing to which the transducer is trained. (5) Immediately after the keying period the relay restores the transducer connections to the receiver input. (6) Any sound energy that is returned to the transducer is converted into electric energy and applied to the receiver input.

Indicator console of the QGB system.
Figure 6-2. -Indicator console of the QGB system.

Pictorial diagram of the QJB system.
Figure 6-3. -Pictorial diagram of the QJB system.

In the QGB, the receiver output is supplied to a loudspeaker or headphones to allow the operator to identify the signals. This signal voltage may also be used to mark the paper in the recorder and make a permanent record on the chart. A third use of the signal is to deflect the beam of the bearing deviation indicator (BDI).


The model QJB, shown in figure 6-3, is a searchlight sonar equipment also intended for installation in destroyers and destroyer escort vessels. This equipment is smaller than the QGB, and all its electronic units are housed in a single console. The QJB transmitter is much smaller than the QGB transmitter, because of the lower power requirements of the QJB. Although the transmitted power is less, the resultant echoes are equal to those of the QGB due to the higher sensitivity of the crystal transducer of the QJB. The transmitter is mounted in the lower right section of the console. The other units in this console are the receiver, keying unit, bearing deviation indicator, indicating range recorder, remote training control, and power supplies.

The receiver is of the sum and difference type and has both time-varied-gain (TVG) and reverberation-controlled-gain (RCG) features.

The QJB utilizes the unicontrol-oscillator system for tuning the receiver and transmitter. The operating frequency need not be tuned carefully to the frequency of the crystal transducer because the transducer has a rather flat resonance curve-about 6 kc wide.

The chief difference in the appearance of the QJB console as compared with the QGB is the manner in which the BDI cathode-ray tube is mounted. The QJB mounting is shown in figure 6-4. The tube is located in the center of the bearing circle so that the operator does not have to shift his eyes away from the bezel indicating the transducer bearing when he wishes to see the BDI indications. To mount the BDI cathode-ray tube in the center of the bearing indicator-which consists of a true-bearing circle and a bezel ring that indicates the transducer bearing-a complicated mechanical system is required.

The keying-control unit and indicating range recorder are conventional.

  BDI and remote bearing indicator of the QJB
Figure 6-4. -BDI and remote bearing indicator of the QJB console.

The transducer is made up of ADP crystals mounted on a steel plate. These crystals project perpendicularly from it a distance equal to one-quarter wavelength of sound in the crystal medium. Opposite the crystals on the reverse side of the mounting plate are steel rods extending a distance equal to one-quarter wavelength of sound in the steel medium. These steel rods are longer than the crystal units because the velocity of sound is greater in steel than in crystal. This system of mounting the crystals with the resonating rods results in greater power output than if the crystals were used alone. It has the disadvantage of making the transducer frequency sensitive.

This combination forms a half-wave system rigidly mounted in the center. In this type of system the mounting plate is stationary, and the crystals vibrate in a direction perpendicular to the plane of the mounting plate. The crystals are connected so that they all vibrate in time phase-all crystal surfaces move in the same direct ion at the same time. The results of this arrangement approach the theoretical results of the piston type of vibrating source.


The crystal array is housed inside a rubber sound window. The space inside the sound window that is not occupied by crystals is filled with castor oil that has been treated to remove all air and moisture. This castor oil protects the crystals from damage by moisture, because it excludes water.

The only other unit of this equipment that is not located in the console is the retracting gear, which is in the hull of the ship near the keel. This retracting gear is similar to that of the QGB equipment. However, it is slightly smaller because the QJB transducer is smaller.


Model QGA (figure 6-5) is another searchlight sonar system designed for installation on destroyers. The system has two complete sonar equipments that are practically identical. One operates on a frequency of 14 kc; the other, on a frequency of 30 kc. The 30-kc transducer can be tilted downward from an angle of 0° to an angle of 45° below the deck. This feature is of value when the sonar vessel is approaching a deep target.

The QGA consoles are similar to the QGB console. They are installed side by side in the sonar control room. The two equipments of the QGA are capable of independent operation, or they may be slaved by a control on the 14-kc console. An

  Block diagram of the QGA system.
Figure 6-5. -Block diagram of the QGA system.

external tactical recorder can be used to control the transmission of either equipment or both of them.

The receiving system for each console consists of an audio receiver and a BDI receiver. The transmitters are conventional r-f amplifiers. A unicontrol-oscillator system tunes the receivers and transmitter of each unit.

The magnetostriction transducers are mounted on concentric shafts that are hoisted and lowered together. The 30-kc transducer is smaller because of its higher frequency. It is mounted over the 14-kc transducer. The training mechanisms are arranged so that the transducers can be trained independently of each other.

Scanning Sonar Equipment

The model QHB-a scanning sonar equipment (figure 6-6) is an ultrasonic, magnetostrictive, echo-ranging-listening equipment that provides a video presentation of acoustic reception from all directions and an audio presentation of reception on any selected bearing.

In the echo-ranging condition, the QHB-a transmits a pulse of sound power in all directions, and then scans or samples all echoes so as to produce on the screens of associated cathode-ray tubes a plan-position indication of all echoes received. Simultaneously, the audio-channel sensitivity pattern may be trained in any desired direction for aural recognition of the characteristics of any of the echoes, as well as for determination of the range by means of a range recorder.

In the listening condition, automatic transmission is omitted. However, the video channel is

  still alert in all directions. It scans 26-kc ultrasonic frequency noise and produces radial patterns on the screens of the cathode-ray tubes, from which the true bearing of any noise source can be obtained. Simultaneously, the audio-channel sensitivity pattern may be trained on any noise source for determination of its character.

System Line-Up

The system employs a single transducer for both transmission and reception. It contains 48 electrically independent hydrophones, which are arranged symmetrically along the periphery of a nontrainable cylinder. During the transmission pulse of 35 milliseconds these hydrophones are connected in parallel by the receive-transmit switching relay so that the acoustic power is transmitted simultaneously in all azimuth directions. Immediately after the transmission pulse


Pictorial diagram of the QHB-a system.
Figure 6-6. -Pictorial diagram of the QHB-a system.

the receive-transmit switching relay returns the circuit to normal so that the hydrophones are again independent. Any reflected acoustic intelligence is incident only on those hydrophones that face its path. The output of each of these hydrophones is connected through their individual preamplifiers to corresponding stator segments on both the audio and the video scanning switches. The scanning switches do not effect a direct contact but utilize a capacitive connection. The stator plates connect to their corresponding hydrophone units, and the rotor plates-18 in number-connect to a lag line to form the acoustic beam. Scanning switches are needed to interpolate bearings; otherwise the bearing could be obtained only in steps of 7.5° and the accuracy of the equipment would be impaired.

The video scanning switch is driven at a continuous rate of 1,750 rpm. Geared at a 1-to-1 ratio with this switch is a control transformer that positions the electron beam of the

  cathode-ray tube so that it remains in synchronism with the true bearing of the scanning switch.

The rotor of this control transformer is excited by a d-c voltage that is varied linearly with time to produce a slowly expanding spiral sweep. The picture of the cathode-ray tube therefore indicates plan position with the ship at the center. The audio switch is identical to the video switch but is not continuously rotated. The rotor of the audio switch is positioned by a servo system that is controlled from the console. This servo system drives another control transformer, which feeds the bearing information back to the cursor line. The cursor line appears on the screen of the cathode-ray tube during the transmission period and indicates the true bearing to which the audio channel is trained. Two scanning switches are needed because the video switch is rotated so rapidly that audio signals from it occupy too small a time duration to be heard. The inputs to the scanning switches are parallel, but their

Block diagram of the QHB-a system.
Figure 6-7. -Block diagram of the QHB-a system.

outputs are separate and each supplies a receiver. The circuits just discussed are called the directional-sensitivity circuits.

The transmitter is a conventional pulse-type amplifier.

A block diagram, figure 6-7, illustrates the over-all operation of the QHB-a equipment. Transmission is initiated by a keying pulse, which originates in the keying-pulse generator circuit of the sonar indicator control and which either functions automatically or is triggered by a range recorder. This pulse operates the relay for the transmit-receive switching and produces the transmitted power at a frequency derived from the unicontrol-oscillator system of the receiver. This transmitting electric power is transformed into acoustic power by the transducer and transmitted simultaneously in all directions. At the end of the transmission the keying relay restores the circuits to the condition for receiving, and the transducer then acts as a hydrophone and produces electric signals from acoustic reflections or noise sources.

The video scanning switch, rotating continuously at 1,750 rpm, produces a signal voltage whenever its acoustic beam sweeps past an echo signal. This voltage is then delivered to the video channel of the receiver, a conventional superheterodyne, the rectified output of which supplies brightening signals for the grids of cathode-ray tubes. The beam deflection in these tubes is synchronized with the video scanning switch so that the brightening occurs at the correct indicator bearing. Approximate range is shown by causing the axial deflection of the beam to increase at an appropriate rate with respect to time. This procedure produces a slowly expanding spiral sweep.

The audio scanning switch, which can be positioned by the training control, receives echo signals from a particular bearing and delivers them to the audio channel of the receiver. This channel also is a conventional superheterodyne with a beat-frequency oscillator for producing audio notes from the ultrasonic echo signals. The output of this channel supplies echo signals to the loudspeakers and the range recorder.

In addition to the echo scanning, the video presentation includes (1) a radial solid line, or bearing cursor, and (2) a radial dotted line, which

  indicates direction of the ship's stern. The bearing cursor appears automatically on the screen of the cathode-ray tubes at a bearing that corresponds to the direction in which the acoustic beam of the audio scanning switch is trained. Cursor time is confined normally to a portion of the transmission interval but may be extended when the OKA-1 sound range recorder is in operation. When slewing, the bearing cursor automatically appears continuously. The stern-line indicator appears only during the sweep interval and progresses with the sweep from a small arbitrary range to the maximum range use.

Description of Components

Transducer. -The transducer is the underwater element that performs the fundamental function of reciprocal conversion of acoustic energy into electric energy. It is a simple cylinder, as shown in figure 6-8.

The construction and harnessing of the units are such that the mechanical Q is approximately 12. When the transducer is connected to the system load, the effective Q is reduced to approximately

Cut-away view of the QHB-a transducer.
Figure 6-8. -Cut-away view of the QHB-a transducer.


8 ½, thus providing a 3-kc operating frequency band.

The transducer is composed of 48 transducer units mounted radially in the transducer. A cut-away view of a single unit is shown in figure 6-9.

Directly above the array of 48 transducer units is a similar ring of 48 smaller units, each of which is but 1 3/4 inches high. These units are series-connected and are employed only for transmission when echo ranging on deep, nearby targets. This manner of transmission is called maintenance of close contact (MCC). The short vertical dimension of the units provides a broad vertical transmission pattern, assuring that sound energy reaches a target at a large depth angle. The vertical response of the main portion of the transducer has a gain of approximately 11 decibels over a nondirectional radiator. The two-way loss in echo ranging with this pattern makes contact with a deep target unlikely, so the MCC units are keyed with the main units in order to distort the beam into a broad vertical pattern.

Scanning-switch assembly. -The scanning-switch assembly (figure 6-10) is concerned primarily with

  the preamplification of the 48 signals from the transducer and the formation of acoustic beams from these signals. It contains the send-receive switching provisions and the means for changing over from normal transmission to MCC transmission.

The preamplifier unit contains 48 identical resistance-coupled amplifiers, each of which is associated with a specific transducer unit. Plate and filament supplies are obtained from the power-supply chassis. Each amplifier consists of an input transformer and a twin triode, 6SL7, with associated capacitors and resistors. The output of each amplifier is at low impedance, and permits connection to the scanning switch without use of a twisted pair as required in the input.

The video scanning switch is identical to the audio scanning switch in all respects except the method of rotation of the rotor. This video switch is mounted directly below the audio scanning switch. Signal connections to the video switch are direct from the audio-switch terminal board. Instead of being driven by a servo, the rotor shaft is continuously driven at a 2-to-1 reduction by a capacitor-type induction motor that is rated at

Cut-away view of QHB-a transducer unit.
Figure 6-9. -Cut-away view of QHB-a transducer unit.

QHB-a scanning-switch assembly, showing
Figure 6-10. -QHB-a scanning-switch assembly, showing interior.

3,500 rpm and 1/20 hp. The drive is accomplished through helical gears to ensure smoothness of rotation and reduction of noise. A 5HCT control transformer is driven at a 1-to-1 ratio by the scanning-switch shaft. The rotor of this control transformer is excited by a direct current that is proportional to the sound range. Therefore, a 3-phase a-c voltage is induced in the stator at slightly less than 30 cps.

The magnitude of the 3-phase voltage is proportional to the range, and its phase relation is constant with the instantaneous angular position of the scanning-switch rotor. This polyphase signal is connected to the deflection coils of a cathode-ray tube. The plan position picture is fairly accurate if the beam of the tube is precisely synchronized in angle with the scanning switch rotor and in radius with range. The output of the scanning-switch rotor is connected through suitable amplifiers to the grid of the indicator cathode-ray tube.

  The range and bearing of an echo thus may be identified by the appearance of a bright arc on the screen of the cathode-ray tube.

It may be noted in passing that the angular speed of the scanning-switch rotor and the pulse length of the transmitter are such that the rotor is "trained" on a possible returning echo at least once during the time the echo is incident on the transducer. Shorter transmitter pulses would possibly result in failure to detect echoes because all sound energy might return during the time the rotor was trained in other directions. The pulse length of the equipment is 35 milliseconds, or 1/28.5 seconds, the returning echo has the same time duration, and as the angular speed of the scanning switch is slightly less than 30 revolutions per second, any

QHB-a sonar receiver-transmitter cabinet, showing interior.
Figure 6-11. -QHB-a sonar receiver-transmitter cabinet, showing interior.


returning sonar intelligence is incident on the transducer during a scanning cycle.

Receiver-transmitter unit.-The receiver-transmitter unit (figure 6-11) contains a dual-channel receiver, which amplifies the signal from the two scanning switches to a level that is suitable for operation of the video and audio indicators of the system. It also contains the complete impulse-type transmitter, which provides the high-level electric signal employed in echo ranging. On the outside, this unit is a simple box without any controls.

When the upper door of the cabinet is open, only the receiver-converter is accessible. When the lower door is open, the complete transmitter is accessible. In the bottom of the cabinet are mounted four large oil-paper capacitors, which constitute the energy storage for the high-power pulse. On the cabinet structure directly above the capacitors are mounted two interlock switches operated by the door. One disconnects the 3,700-volt power supply, and the other short-circuits the capacitor bank, thereby minimizing the hazard from extremely high potentials.

Directly above the storage capacitors is a drawer that contains the high-voltage supply for the transmitter. Directly above the power-supply chassis is another drawer that contains the transmitter-power amplifier. To prevent possible hazard to maintenance personnel, the upper portion of the cabinet is separated mechanically from the transmitter section by an expanded metal grill. The receiver-converter unit is the vertical chassis mounted in the front part of the upper portion of the cabinet. The chassis contains the complete twin-channel amplifier plus the converter section, which provides the output-frequency signal at a suitable level for driving the power amplifier. The tuning range provided is from 22 to 29 kc. The transducer that is furnished with the equipment is useful only over the range of from 24 to 27 kc.

Other operational controls on the chassis are (1) audio gain control, which provides for adjustment of the level of the audio output with respect to the video; (2) doppler-nullifier gain control, which provides for the "stiffness" of response of the doppler-nullifier circuit; (3) converter gain control, which adjusts the level of the signal to the transmitter,

  thereby governing the power-output level of the transmitter; and (4) target-doppler nullification on-off switch, which completely disables the target-doppler nullification circuit without disabling own-doppler nullification. All the adjustments described here are of the screw-driver locking type.

The transmitter power amplifier is contained in a drawer in the upper portion of the transmitter section of the cabinet. In it are mounted 3 type-715C beam-power tetrodes, the filament transformer, the input transformer from the converter, and the output transformer tuning capacitors.

QHB-a sonar indicator control, showing interior.
Figure 6-12. -QHB-a sonar indicator control, showing interior.


The unit is extremely simple and contains a minimum of internal components. No adjustments are involved in the circuit.

Electrically, the function of the unit is to accept from the converter an r-f signal pulse of relatively low power. The unit amplifies this pulse to a power level of approximately 7 kilowatts maximum, which attenuates approximately 3 ½ db during the pulse time. This attenuation is characteristic of a pulse-type plate supply.

The transmitter power supply has the single function of charging the storage capacitors during the interval following the echo-ranging transmitting pulse. This power supply consists of a large power transformer, the primary of which is in series with a current-limiting reactor; the secondary supplies the plates of two type-866A rectifier tubes. The cathodes of these tubes are connected directly to the storage capacitors and to a resistance network, which serves as a bleeder. These resistors are mounted on the underside of the chassis, and a portion of the combination is paralleled by the voltmeter on the power-supply plate.

Sonar indicator control. -The sonar indicator-control unit (figure 6-12) is the sonarman's station and contains all the operating controls.

In the upper portion of the cabinet is mounted the assembly of the cathode-ray tube. Immediately to the left of this assembly is a small control panel. On it a toggle switch provides for selection of peak or band filter in echo-ranging operations. In the peak position it provides target-doppler nullification if this feature has not been disabled by the switch in the receiver. It is undesirable to echo range with this switch in the peak position if the target-doppler nullifier circuit has been disabled in the receiver, because the peak filter is so sharp that a target with appreciable doppler provides very little audio indication. In the band position an RC band filter is inserted, and target-doppler nullification is eliminated. On this same control panel are two potentiometers, one governing the threshold signal of the cathode-ray tube and the other governing the intensity of the electronic bearing line or cursor. In the rear of the upper portion of the cabinet is mounted the second-anode supply of the cathode-ray tube.

  Remote indicator. -The remote range and azimuth indicator is essentially a cathode-ray tube repeater. Associated with it is a loudspeaker. This remote installation gives a complete duplication of the visual and audible indications available to the sonarman. The design of the QHB-a is such that two range and azimuth indicators may be installed with each equipment.

The upper portion of the front panel of the remote indicator contains a circular opening covered by amber filter glass through which the cathode-ray tube is viewed. Surrounding this opening is an azimuth ring identical with that on the sonar-indicator control. Below the panel to the right are various controls-the threshold-intensity and focus adjustments of the cathode-ray tube, the video-signal level adjustment, and the loudspeaker volume control. These controls make possible the complete and independent adjustment of visual and audible indications at the remote station as long as the sonarman operates the equipment at a reasonable level.

Transmit-Receive Switching

The 48-transducer unit, the transmitting capacitors, and the signal transformers from the connection for transmission to those for reception are switched by two contacts on keying relay K401. For transmission, each capacitor should be in series with a transducer unit, and the preamplifier input transformers should be connected so that the voltage developed across each of them by the transmission pulse is small. For reception, each transformer should be connected to its respective transducer unit, which is paralleled by a capacitor.

These switching connections are shown in condensed form in figure 6-13, A, where the connections for transmission occur when the relay contacts are open and where the connections for reception occur when these contacts are closed. The 48 transmitting capacitors, C401 through C448, are connected in series with their respective transducer unit, Z1 through Z48. One terminal of each of the 48 input transformers also is connected to the corresponding transducer unit; the other terminals are connected together during transmission and are separated from ground by keying relay K401.


QHB-a transmit-receive switching.
Figure 6-13 -QHB-a transmit-receive switching.

If the capacitors, transducer units, and transformer input impedances were identical in each of the 48 circuits, the a-c potentials applied to the transformers would be equal and the return circuit from the transformer common connection would be open. This condition would result in no voltage across the transformer primaries. Hence, the voltages existing during transmission in any of the 48 transformers have a value that depends on the inequalities of the 48 capacitors, transducer unit,

  and transformers. As a result, the circuit for normal transmission can be shown functionally, as in figure 6-13, B.

Each transducer unit, Z1 through Z48, including 40 feet of cable, has a nominal impedance of 58+ j82 ohms at center frequency, and the series capacitors, C401 through C448, produce a total effective impedance for each unit of 58-j400 ohms. The 48 parallel circuits then present, at 25.5 kc, an impedance of 1.21-j8.35 ohms. The tuning inductor, L402, in parallel with this combination, has a reactance of 8.8 ohms. These values produce a load that is equivalent to approximately 50 ohms on the output transformer, T713.

Figure 6-13, C, shows the equivalent circuits for reception. The impedance of the signal source, which consists of a transducer unit paralleled by one capacitor, becomes equivalent to 81+j86 ohms. Hence, the transformers are designed to reflect the input impedance of each preamplifier circuit and produce a primary impedance of 81-j86 ohms, which results in a conjugate impedance match to the transducer circuit and the greatest transfer of energy.

To provide a broader vertical beam pattern and to reduce the transmitted acoustic intensity in the horizontal plane, a condition called MCC transmission can be established for purposes of maintaining close contact.

In addition to the main units, the series-connected ring of 48 short units-located in the upper end of the transducer and called the MCC ring-is used to transmit with a broad vertical pattern. This ring is not involved in reception. Some power is supplied to the remaining transducer unit, with phase and amplitude relations between these units and the MCC ring sufficient to cancel the transmitted intensity along the horizontal axis. This cancellation reduces surface-reverberation effects. The circuit with the MCC connections is shown in figure 6-13, D.

Directional Sensitivity Circuits

General -The directional sensitivity circuits cover the signal-receiving function from the transducer, through the preamplifiers and scanning switches to the receiver.

In the receiving condition the de-energized keying relay connects the 48 transmitting capacitors in parallel with their respective transducer units.


The keying relay also bridges the primaries of the 48 preamplifier input transformers across the transducer units. Each of these transformers provides a conjugate impedance match between (1) the transducer-unit circuit, including the capacitor, and (2) the input impedance of the preamplifier tube. Because the 48 preamplifier circuits are identical, this discussion deals only with one of them. The signal from the transformer is connected to the control grid of the left side of the twin triode, which is connected as a voltage amplifier with a gain of approximately 20. The output is connected to the right side of the twin triode, which is operated as a cathode-follower. Thus, it is a low-impedance source for transmission to its corresponding segment on the stator plate of the audio and video scanning switches. Similarly, the remaining 47 preamplifiers deliver the amplified transducer-unit signals to the other 47 capacitor segments on each scanning-switch stator.

Beam-pattern formation and rotation. -One function of the equipment is to produce electrically an acoustic pattern, continuously rotatable through 360°, using a fixed cylindrical array of hydrophones or transducer units. This function is accomplished by two devices-(1) the electric circuits necessary to produce the beam pattern and (2) the electro-mechanical means for rotating this pattern.

The production of an optimum beam pattern from a fixed array of radiators or receptors is a fairly well known art and consists either in (1) choosing amplitudes and phases for the currents in the radiators or (2) modifying the voltages from the individual receptors-depending on the geometry of the array. The phasing requirements are imposed because this array of transducer units is cylindrical. Therefore, the signals received by each unit from a plane sound wave, unlike the signals that exist in a plane-faced transducer, differ in phase in proportion to their physical displacement. The total voltage from a group of units facing the sound source is a maximum when all the signals have been shifted so as to be in phase with one another. The resulting beam pattern is similar to that of a plane-faced transducer of approximately equivalent dimensions. This phasing requirement is accomplished by the use of a linear phase-shift "lag line" in order that the phasing may remain correct when the frequency is changed.

  The usual beam pattern for this type of transducer consists of a major lobe, accompanied on each side by minor lobes that decrease in sensitivity as the angle from the main lobe increases. Altering the relative amplitude of the side-lobe signals (shading) can result in a reduction of the level of the more adjacent minor lobes at the expense of increasing the very small lobes, which are at a large angle from the main beam. The optimum signal-to-noise ratio occurs for such shading when the minor lobes have all been brought to the same level, and consequently the main beam is widened only slightly. Design engineers select the proper fraction of each of the signals that are to be phased and added so as to provide this shading.

Lag-line phasing and shading. -The beam-pattern formation can be analyzed by (1) inspection of the voltages produced by the transducer units, (2) the required phasing, and (3) the choice of shading.

The first of these methods, involving a single-unit pattern, may be presented in several ways, one of which is shown in figure 6-14. This illustration shows that the total lagging phase shift for the signal from unit 1 is at least 680° in order to bring it into phase with the signal from unit 8. The

QHB-a transducer-unit voltages.
Element Amplitude Phase
1 1.00 0
2 0.99 31
3 0.97 84
4 0.92 170
5 0.79 264
6 0.62 373
7 0.54 520
8 0.45 680
9 0.36 867
10 0.25 1031
11 0.18 1224
12 0.12 1422
Figure 6-14 -QHB-a transducer-unit voltages.


QHB-a lag line and equivalent circuits.
Figure 6-15 -QHB-a lag line and equivalent circuits.

phase shift for the voltage from unit 2 is 31° less than that from unit 1.

To accomplish the necessary phase shifts, the simplest lag line is a uniform line of as many sections as required to match the desired angles with whole or half sections. One limitation is that the phase shift per section must be kept below approximately 60° in order to approach linear phase shift with frequency. The final lag line as designed has a phase shift of 52° per full section at 26 kc and consists of 14 sections. The physical arrangements for three sections are shown in figure 6-15, A. The proper choice of entry points for the signal voltages allows a good approximation to the required phase shifts. The electrical equivalent of this circuit for a single section is shown as a bridged-T network in figure 6-15, B, and as the equivalent lattice network used in design in figure 6-15, C. This circuit results in a characteristic impedance of 16,300 ohms at 26 kc.

  A method of injecting the signal voltages from low-impedance generators has been devised. This method does not impose any loading of the lag line or mismatching at any point, which would result in standing waves of voltage on the line. Because the scanning switch must introduce these voltages through the capacitance of the segments on the stator and rotor plates, this capacitance is made a part of a network of three capacitances, through which is injected a voltage (figure 6-15, D).

Two units of the scanning-switch circuit that are symmetrically disposed about the center of the beam-forming network (such as 3R and 3L in figure 6-17), introduce signals at the same point on the lag line. The complete circuit for signal injection is shown in figure 6-15, E. The values are related because the total capacitance must equal the value required (1) by the lag line at the point of signal injection and (2) by the desired fraction of the signal voltage that is to appear on the lag line.

The optimum beam pattern requires an attenuation or shading of the transducer-unit voltages. The desired total attenuations are shown in figure 6-16, with the attenuation already present, that is due to the single-unit pattern. The ratio of the single-unit pattern to the total attenuation for any unit determines the attenuation that must be introduced by the beam-forming network. This

Figure 6-16 -Shading curves.
Figure 6-16 -Shading curves.


attenuation is accomplished by the choice of capacitance values shown in figure 6-15, E. The combination of these circuits results in a complete scanning-switch circuit, illustrated in figure 6-17. Because the phase shift from unit 7 to unit 8, and from unit 8 to unit 9, is approximately 180°, a   voltage also is used from unit 9 and is introduced into the same point as that from unit 7.

A typical resultant beam pattern has (1) a major lobe 11° wide at a level that is 6 db below the peak sensitivity and (2) minor lobes that are at least 25 db below the same reference level.

QHB-a scanning-switch circuit.
Figure 6-17 -QHB-a scanning-switch circuit.

Scanning switches. -The video and audio switches are identical both structurally and electrically. Structurally, they each consist of two large cast-iron cups accurately doweled and bolted together at their open ends. At the left end of the assembly is a circular terminal board, to which are connected the 48 output leads from the preamplifiers. These leads are connected directly by short conducting loops to pins that protrude through the left-hand surface of the scanning switch proper. The pins are insulated by rubber grommets.

A glass disk, 11 inches in diameter and 7/8 of an inch thick, is bolted securely to the machined surface of the left-end bell. The outer plane surface of the disk is coated with silver and scribed with 48 radial divisions, which form separate conducting segments. After 48 holes have been drilled completely through the segments and the glass, they are metalized. In this way the segments are connected electrically to the back of the hole, where connection pins are soldered, thereby providing a complete circuit from any specific preamplifier to a corresponding segment. Each of the rotors consists of a glass plate (identical to the stator plate), which is secured to a large steel hub that has been shrunk on the rotor shaft. With the rotor shaft mounted in place, there is an air gap of approximately 0.004 inch between the glass plates.

With the segments of both rotor and stator lined up, 48 equal capacitors of from 80 to 100 micromicrofarads are formed. These capacitors constitute the means for connecting the outputs of the preamplifiers to the electrical network carried on the rotor assembly. This network provides the proper phasing and attenuation of the signals from any 18 consecutive stator elements in order to form an acoustic beam. The network is mounted in a cast-aluminum can, and the output connections are carried to suitable slip rings mounted on the rotor shaft. Carbon-silver brush members inserted through the rear wall of the right-end bell engage these slip rings, thus making accessible the electrical-signal equivalent to the acoustic information that is obtained at any instant.

Because of the necessary high rate of rotation of the video switch the signal from an echo is not sufficiently long for audio presentation. Therefore, two switches are needed. The inputs of

  these switches are parallel-connected, and their outputs are connected to individual receivers-one for video and the other for audio presentation. The system lined-up for the receiving functions is as follows:

1. Each transducer unit connects through its own preamplifier to a stator segment on the audio scanning switch and to the corresponding stator segment on the video scanning switch.

2. The output of the video scanning switch connects to the input of the video receiver, the rectified output of which is used as the brightening signal for the control grids of the cathode-ray tubes.

3. The output of the audio scanning switch connects to the input of the audio receiver, the output of which is used to drive the loudspeakers and to mark the tactical range recorder.

The azimuth angle of the video scanning-switch rotor at any instant is indicated on the cathode-ray tube by a 5HCT control transformer. This control transformer is used as a 3-phase sweep generator-a unique employment of a synchro.

In the usual synchro system the rotor is energized by a single-phase a-c voltage. Therefore, the stator coils remain in time phase but vary in magnitude, depending on the angle between the axis of the magnetic field and the axis of each stator coil. When these voltages are connected to a synchro receiver they duplicate the magnetic axis of the transmitter. The 5HCT synchro used to generate the sweep voltage is excited by a direct current, the magnitude of which is proportional to the range of the active volume at any instant. The output therefore is a true 3-phase voltage with the peak magnitude of the voltage in each phase increasing with time. If the excitation remained at a constant value this arrangement would be the same as that in any 3-phase generator the rotor of which is d-c-excited and rotated in a 3-phase stator.

The rotor of the 5HCT synchro that is used to generate the sweep voltage is geared at a 1-to-1 ratio to the video scanning-switch rotor. When the system is in motion a 3-phase voltage proportional to range and synchronized in bearing is available at the stator terminals. Because the rotational speed is about 1,750 rpm, the output


frequency is slightly less than 30 cps. This polyphase signal provides relative bearing of the video beam in the deck plane.

For conversion to true bearing with stabilization, the signal is taken to the data converter and is applied to the stator terminals of a 5SCT synchro, which is used as a polyphase phase-shifter. The rotor of the 5SCT is positioned primarily by the ship's gyro order with a stabilizing component

  related to azimuth sonar train. The output at the rotor terminals of the 5SCT synchro is therefore a 3-phase true bearing sweep signal that is stabilized in a horizontal plane, with respect to the line of sight. The three components must be amplified by three identical feedback amplifiers so as to provide sufficient signal for the deflection coils of the cathode-ray tube. These coils are the stator coils of a 5SCT synchro.
Depth-Determining Equipment



The model QDA depth-determining equipment is an ultrasonic echo-ranging equipment operating in the frequency band of 50 to 60 kc. It is primarily an attack instrument and is installed in conjunction with an OKA-1 sonar resolver and an azimuth sonar equipment, which may be either the QHB-a or the QGB. The differences between the QDA equipment and a standard azimuth search equipment lie chiefly in the transducer and the recording mechanism.

As with ordinary echo ranging, range is determined by the ping-to-echo time lapse and the velocity of sound. For determining the depression angle, the QDA transducer, which has a sharp beam in the vertical plane, is pivoted on an athwartship axis to permit the beam to be tilted to any necessary depression angle. In depth search, pings are sent with the beam tilted at various angles. When the beam is directed toward a target, echoes are detected if the target is within range, and the tilt of the beam at the time such echoes are received corresponds approximately to the target depression angle Eq. After depth contact with a target, the alignment of the beam with the target depression can be indicated more accurately by a depression deviation indicator (DDI), which is analogous to the bearing deviation indicator employed in azimuth sonar systems.

Depth is indicated automatically by a depth recorder, which is similar in principle and construction to the tactical range recorder and the indicator range recorder. In the depth recorder, however, the stylus travels at a speed proportional to the sine of the target depression angle, which is the vertical component of the velocity of sound,


  Vz. The stylus moves at a rate corresponding to the slope of the sound beam. If the beam is steeply inclined the stylus moves rapidly from left to right; if the beam is nearly horizontal the stylus moves slowly. The ping is transmitted just as the stylus moves away from its zero position, and the stylus marks the recorder paper at the instant the echo returns.

For a distant target the depression angle is small, and correspondingly, the stylus moves slowly but for a relatively long period before the echo mark is recorded. For a nearby target the depression angle is relatively large, and as a result the stylus moves rapidly for a short period before the echo returns. In both cases the echo is recorded at the same distance from the starting position of the stylus. A linear depth scale, reading in feet, extends across the recorder chart. The stylus speed is controlled by the OKA-1 resolver, but the basis of this speed is determined by the QDA beam depression angle and the velocity of sound in water. When the bypass switch is in the search position the azimuth transducer only is stabilized. When the bypass switch is in the attack position both the azimuth and the QDA transducers are stabilized.

Keying and Controlling the Recorders

The keying interval for both the azimuth and the QDA echo-ranging systems is controlled by the sound-range recorder, a unit of the OKA-1 equipment. When the fly-back contacts of the sound-range recorder are closed, the action of the keying circuit in the azimuth equipment and of a similar circuit in the OKA-1 resolver are both initiated. The latter circuit controls the depth recorder and keys the QDA. These timing circuits cause the stylus clutch to release in the


indicator range recorder of the azimuth equipment, and of the depth recorder so as to allow the respective styli to fly back and dwell for a brief period. The clutches are then re-energized, and almost simultaneously the two equipments are keyed.

The general arrangement and sequence of operations of the units of the different equipments are shown in the block diagram of figure 6-18.

If the depth recorder reaches the end of its travel before the fly-back contacts of the sound-range recorder are operated, the stylus flies back and dwells until the sound-range recorder initiates another keying cycle. When the sound-range recorder is turned off, the depth recorder is no longer slaved, and it controls its own keying cycle through the OKA-1 resolver timing circuit.

Horizontal Range and Computed Target Depression

The sound-range recorder is provided with an adjustable cursor, the position of which represents sound range, Rq. Rq is transmitted by synchro to the predictor resolver circuit of the OKA-1 resolver unit. The depth recorder likewise is

  equipped with a cursor, the position of which represents target depth below the transducer, H'q, which also is transmitted by synchro order to the predictor resolver circuit. From these two inputs the OKA-1 resolver computes the following:

1. Horizontal sound range, Rhq, which is transmitted to the horizontal recorder and to the remote indicators of the QDA equipment.

2. Sonar target depth, Hq, which also is transmitted to the remote indicators.

3. The computed sonar target depression, cEtq, which is transmitted to the tilt-control differential generators in the QDA console as an aid in maintaining contact.

The operator of the QDA console modifies the order cEtq by an adjustment of the tilt wheel under guidance of the DDI. The order introduced by the operator is known as adjustment of computed sonar target depression, jEtq. If the azimuth beam is centered on the target, as indicated on the DDI, the order leaving the console represents apparent depression of the acoustic path to the target, Eq. The order Eq is the sum

Sequence of operations for determining target depth.
Figure 6-18. -Sequence of operations for determining target depth.

of cEtq and jEtq. This order is indicated on the console by a synchro that positions an indicator.

Correction for Bending of the Sound Beam

The order Eq is transmitted to the stabilization computer by way of the bypass switch and the Snell's law resolver circuit of the OKA-1 resolver unit. The purpose of the Snell's law resolver circuit is to establish the speed of the depth-recorder stylus so that its excursion rate is proportional to Vz. If the beam is vertical, the vertical component of the velocity of sound along the beam is simply the velocity of sound in water at the prevailing temperature. If the beam is horizontal, Vz is zero.

The Snell's law resolver computes and establishes Vz from the input information consisting of (1) the Eq order; (2) the velocity of sound, Vo, which is injected manually into the circuit; (3) the difference of velocity in the mixed layer and the refracting layer, V, also manually injected; and (4) the layer-depth timing introduced by the closure of the layer-depth contacts in the depth recorder at the appropriate point in the stylus excursion. The layer-depth contacts in the depth recorder are adjusted manually, and information for this setting of Vo and V in the Snell's law resolver and for Vo in the sound-range recorder are derived from bathythermograph information.

Tilt-Order Synchro Circuit

As has been stated, the depression angle order originates in the OKA-1 resolver, where it is computed and transmitted by synchro transmitters at 2 and 36 speed. For on-target conditions, this order is called cEtq. It is delivered to a pair of DG synchros to permit the operator to modify the computed order so as to center the beam on the target as indicated by the DDI. The signal introduced by the operator is jEtq. The output of the DG synchros is the sum of the two inputs and is equal to Eq, provided the azimuth and depth beams are on the target. A synchro receiver, which operates the tilt dial in the console, is connected to the output of the 2-speed DG to indicate this adjusted order.

It is important to understand the significance of the tilt dial indication. Whether or not the two acoustic beams are centered on the target, the indication represents the depression of the QDA beam below the horizon along the bearing of the

  azimuth beam, provided the computer is not bypassed. For on-target conditions, this angle is equal to Eq, the true depression at the transducer of the acoustic path to the target. In the absence of refraction or bending of the QDA beam, Eq is equal to Etq, the true depression of the target.

The true tilt angle of the transducer below the horizon is equal to the indicated angle, Eq, provided the azimuth transducer is trained dead ahead; but the true tilt angle exceeds the indicated angle by increasing amounts as the azimuth transducer is trained farther off the bow. When the bypass switch is in the bypass position, the tilt indicator shows the actual tilt of the QDA transducer with respect to the deck plane, and the transducer tilt is unaffected by the position of the azimuth beam.

The 2-speed output of the differential transmitters in the console is connected to the OKA-1 resolver to operate the Snell's law resolver, which controls the stylus speed of the depth recorder. This 2-speed output also is connected to the stabilization computer, which corrects it for the roll and pitch of the ship. This corrected order for on-target conditions is E'q, which represents the depression of the beam relative to the deck, measured in a plane through the line of sight perpendicular to the deck. E'q also is the order that would control the tilt of the transducer if the transducer were trained to the bearing of the target. For the QDA system, however, the E'q order is transmitted to the tangent solver of the computer, where it is converted into a transducer tilt order that causes the fan-shaped beam to pass through a target, which is at the bearing of the azimuth transducer. The transducer tilt order is E'q's.

E'q's is transmitted at 2 and 36 speed, by way of the tangent solver, to the control transformers on the tilt-control mechanism. The signals from these control transformers are connected to the tilt-control amplifier, which supplies the power to the tilt motor that drives the transducer to wipe out the signals and thus bring the transducer to the ordered angle, E'q's.

The tilt of the QDA transducer is controlled by the following factors:

1. Factors controlling the depression order from the console. These factors are combined and then are indicated by the tilt indicator on the console.


a. The computed target-depression order is based on the position of the depth-recorder cursor and that of the cursor of the sound-range recorder. If either of these positions is changed, the order from the resolver to the QDA console changes.

b. The position of the two differential-transmitter rotors located in the console is controlled by the tilt wheel.

2. Factors controlling the correction of the depression order from the console. These

  factors affect the transducer tilt but not the position of the tilt indicator.

a. Roll and pitch correction orders. These stabilization orders are incorrect if the azimuth beam is not centered on the same target as the QDA beam.

b. The alteration of depression order by the tangent solver to supply additional depression for targets on either side of the bow. The correction is necessitated by the non-variable fan-shaped beam.

Block diagram of the QDA transceiving system.
Figure 6-19 -Block diagram of the QDA transceiving system.

Transceiving System

The block diagram of the QDA transceiving system is shown in figure 6-19. The signal transmitted through the water is produced by inter-modulating the outputs of three oscillators shown in the figure. The 190-kc i-f oscillator is a fixed-frequency generator labeled "IF" because it establishes the intermediate frequency in the receiver. The unicontrol oscillator with a frequency range of from 240 to 250 kc (1) establishes the r-f carrier and (2) tunes the receiver to that frequency. The 0.8-kc audio oscillator originates the audio frequency that is heard when an echo or reverberation is received.

The oscillator with a frequency of from 240 to 250 kc is heterodyned with the 190-kc oscillator to produce a difference frequency of between 50 and 60 kc at the output of the first modulator. This difference frequency is the r-f carrier frequency. Because the 190-kc source is blocked except while the transmitter is keyed, the carrier exists only during the key-down condition. The carrier then is modulated with the audio frequency to produce the signal that is transmitted through the water. In the modulation process, negative half cycles of the relatively high level of the 0.8-kc oscillator blocks the modulator. The resulting modulated signal then consists alternately of a group of oscillations having a frequency of from 50 to 60 kc followed by an equal period of zero signal (figure 6-19). This modulated signal is composed of the carrier frequency and the side-band frequencies spaced at frequency intervals of 0.8 kc. The modulated carrier is amplified in the transmitter amplifier unit and is conducted through the transducer keying relays to the two halves of the transducer, which are connected in parallel.

The echo returning through the QDA transducer is similar to the outgoing signal except that the frequency of the echo has been shifted by Doppler.

In the upper channel, marked "sum," the signals from the two halves of the transducer are combined to produce the signal that would be obtained if the transducer were not split. This signal is amplified in two r-f stages and passes through a modulator, where it is heterodyned against the frequency between 240 and 250 kc. In the absence of Doppler, the difference frequency is 190 kc-exactly the frequency of the i-f oscillator, although the signal shape is different because of the 0.8-kc

  modulation. The Doppler always shifts this frequency to 190-D, the direction of shift being reversed by the heterodyning action.

In passing through the tuned i-f circuits, the higher-order side frequencies are filtered out and the modulated signal assumes a more nearly sinusoidal envelope. The detector rectifies this signal to produce an 800-cps signal similar to the envelope of the i-f signal. The frequency of the audio signal is 0.8 kc-D X(0.8/55); that is, the audio frequency is shifted by only about 1.5 percent of the original Doppler shift.

The reason for this shift can be understood if the modulated signal in the water is thought of as consisting of only three frequencies-the 60.000-kc carrier, the 59.200-kc lower side band, and the 60.800-kc upper side band. Because the Doppler effect shifts the frequencies by 0.7 cycles per kilo-cycle per knot of range rate, a range rate of 20 knots, closing, shifts the named frequencies to 60.840, 60.029, and 61.651 kc, respectively. The same frequency differences are preserved through the i-f circuit. The audio frequency produced when the signal is demodulated in the detector is the difference between the carrier and the side bands, or, in this example, 0.788 kc. Thus, the audio shift is only 12 cps (Doppler shift of the 0.8-kc modulation component), whereas the original Doppler shift is 840 cps. Obviously, the doppler would be very troublesome if the full shift of 840 cps were carried over into the audio signal, as it is in azimuth systems working in the frequency band of from 20 to 30 kc.

The virtual elimination of Doppler in the audio permits the use of a narrow pass filter in the audio circuit with significant benefits in noise reduction.

The signal in the sum channel, traced in the foregoing description, is formed vectorially by adding the signal outputs of the two halves of the transducer. The signal in the diff channel-the lower channel in figure 6-19-is formed from the vector difference between the signals in the two halves of the transducer. The diff signal is zero if the transducer beam is centered on the target in the vertical sense. The diff signal, which has the same character as the sum signal, is heterodyned in the same way as the sum signal. After suitable amplification, the diff signal, together with a portion of the sum signal, is fed into the conjugate


QDA transducer.
Figure 6-20 -QDA transducer.

rectifier. As a result two d-c outputs are produced-one proportional to the sum signal plus the diff signal, and the other proportional to the sum signal minus the diff signal. These two d-c signals are transmitted to opposite ends of the vertical-deflection coil of the DDI oscilloscope in the console. If the sum-plus-diff signal is greater, the electron beam is deflected upward, indicating that the target is above the transducer beam. If the sum-minus-diff signal is greater, the spot is deflected downward, indicating that the target is below the beam. The two d-c signals are equal when the beam is on the target because the diff signal is then zero.

The two d-c signal-return currents, added together, flow out of the mid-point of the deflection coil, and this combined signal (1) supplies voltage to operate the intensity grid of the cathode-ray

  tube to cause brightening when an echo is received, and (2) supplies current through the recorder-stylus chart-paper circuit to produce marks on the chart when echoes are received. These circuits are discussed more thoroughly in chapter 7.


The transducer is of the ADP crystal type. It is shown in figure 6-20. The crystals are in an array approximately 20 inches long and 1 1/8 inches wide and are mounted in groups-each group ½ inch long and 5/16 inch wide-on a steel base-plate that has integral resonators directly behind the crystals. This arrangement provides efficient half-wavelength units with a nodal point at the face of the base plate. The array is connected electrically into separate halves about a horizontal centerline. Each half is connected to the high side of a transformer, the low sides of which provide impedance matches to the transmitter and receiver amplifier circuits. Gold-to-gold contacts between the crystals and their electrodes provide low interface electric resistance and consequently reduced heating. The gold surface is applied by evaporating gold onto the crystal faces and then bonding gold-plated metal foil to these faces.

The crystal array and resonator-plate assembly are mounted, with their long axis vertical, inside a streamline corrosion-resistant steel housing having a thin corrosion-resistant steel window that is almost sound-transparent. Also mounted inside the housing are two transformers, as well as a laminated baffle that attenuates extraneous signals through the back of the transducer. A blanket of the baffle material is assembled along the sides and back of the resonator assembly and helps to reduce the effects of reflections within the housing. At the top of the housing is an integral pivot block, on which are mounted a sector gear and trunnion bolts for attachment to the hoist-tilt mechanism. A cover plate near the top of the housing provides access to a cable seal and filling plug. Another cover at the bottom of the housing provides access to a second filling plug in a rubber-gasketed plate and ring assembly, which seals the housing.

The transducer is vacuum-filled with approximately 1 gallon of electrical-grade castor oil, from which the air and water vapor have been removed.

The transducer is a highly efficient reciprocal converter of electric into acoustic energy over a frequency range of 50 to 60 kc. This wide range


makes possible a choice of frequency that permits simultaneous operation of equipment by several ships in the same area. The beam pattern is very sharp in the vertical plane so as to permit the accurate determination of target depth. In the horizontal plane the pattern is broad so that contact with the target can be maintained over a wide angular range on either side of the bow of the ship. Figures 6-21 and 6-22 illustrate typical beam patterns in both planes.

The crystal array of the transducer is divided electrically into top and bottom halves for obtaining depression deviation indications.


From the output of the second modulator a voltage of from 50 to 60 kc ±0.8 kc is delivered to the grid of the transmitter preamplifier, a type 6SJ7 tube. From the preamplifier output this voltage is used to excite the control grids of the transmitter-amplifier stage, which consists of six type-807W beam-power tetrodes connected in push-pull with three tubes in parallel on each side. The output of the transmitter is connected to the transducer through a tuning retard coil, which has several taps that facilitate tuning out the

Typical vertical beam pattern of a QDA transducer at 55 kc.
Figure 6-21. -Typical vertical beam pattern of a QDA transducer at 55 kc.

  Typical horizontal beam pattern of a ODA
transducer at 55 kc.
Figure 6-22 -Typical horizontal beam pattern of a ODA transducer at 55 kc.

capacitance of the transducer and approximating an impedance match between the amplifier and the transducer. The plate-to-plate impedance of the output circuit should be 3,300 ohms for maximum power output, which is 150 watts.


The QDA target depth-determining equipment and the azimuth sonar equipment when operating together provide complete and continuous information regarding the location of a submerged object. The sole function of the QDA is to determine the depth of a submerged object, whereas the azimuth equipment determines the range and bearing of the target.

The QDA determines target depth indirectly because it is an echo-ranging equipment that measures the time lapse between transmission and echo. The vertical velocity of the sound beam is measured by a chemical recorder that has a variable-speed stylus. The stylus speed is a function of the sine of the depression angle times the velocity of sound. The stylus speed is controlled by an associated equipment-the OKA-1 recording-resolving equipment-but the basis of the


stylus speed is the QDA beam-depression angle and the velocity of sound in water. To obtain accurate depth solutions the equipment has a cathode-ray tube used as a depth-deviation indicator.

Because the target-depth information is employed at various locations on the ship, a means for transmitting the information is provided by an optical cursor that positions a synchro transmission system.

While the sonar vessel is closing a target, the depression angle increases-slowly at first, and more rapidly as the range is decreased. To relieve the operator of most of the burden of following the target, the OKA-1 provides aided tracking by (1) computing the depression angle theoretically required-known as the computed target depression, cEtq-and (2) transmitting this synchro order to the QDA tilt-order system. The QDA operator then adds corrections, jEtq, to this order, guided by the indications of the DDI.

The QDA beam must be stabilized against roll and pitch, and the depression orders originated by the QDA operator must be modified accordingly. This action is accomplished by the stabilization computer, with which a stable element is associated.

The transducer cannot be trained in azimuth. When the target is dead ahead, the target depression corresponds to the transducer tilt-if the sound path in the water is a straight line. When the target bears off the bow, contact still can be maintained because the transducer beam is very broad in an athwartships direction, but a greater tilt is required to keep the fan-shaped beam on the target. This correction to the depression angle is made by the tangent solver, which is a unit of the stabilization computer.

The sound path through the water generally undergoes some bending, principally because of unequal temperatures at various depths. The OKA-1 equipment makes the correction for bending by varying the stylus speed of the depth recorder. While the sound energy is passing through the mixed layer, the path does not bend and the stylus speed is constant. During this period the stylus speed is based on the sound velocity near the surface and the ordered depression angle. Below the mixed layer the sound energy travels at a speed that varies with depth, and as a result the beam is bent.

  If the beam is bent downward, the stylus speed is increased for a corresponding period because the depression angle is greater than the ordered angle. Layer-depth contacts, closed by the stylus carriage, are adjustable and bring about the transition from the mixed-layer travel rate to the refraction-layer rate. The layer-depth contacts in the depth recorder and the two sound-velocity controls in the OKA-1 resolver are set according to the information from bathythermograph readings.


The sonar system installed on antisubmarine vessels is composed of several sonar and fire control equipments operating in a reciprocal-information and control network. The purpose of the system is to fix a submarine's position once contact has been established and to solve the necessary fire control problems to assure a kill. The system is an attack system and not a search system. There has always been a need for a fire control system in antisubmarine warfare, but the need did not become acute until the advent of the high-speed, deep-diving submarine. At the outbreak of World War II the depth of the submarine was approximated by the conning officer from the range at which contact was lost because of the sound beam passing over the U-boat. By the time the lost-contact range was reached the anti-submarine vessel was on its attack course, and was already starting to lay the depth-charge pattern.

In spite of this difficulty, the method was fairly effective against the old type of submarine with riveted construction because the pressure hull could be ruptured by near-misses. With the advent of the modern welded-hull construction, which can stand terrific pressures and stresses, it became necessary to score a direct hit on the submarine to do a reasonable amount of damage. The underwater fire control system furnishes the precise information necessary to score these direct hits. Furthermore, it supplies the information until a very late stage in the attack. The older single-sonar search system frequently lost contact at ranges up to 600 yards. As a result, the submarine had ample time to take evasive action which could not have been detected by the attacking vessel.

A typical installation aboard an antisubmarine vessel may consist of a QGB or QHB, a QDA, an OKA-1, a Mark 4 director, and a stabilization computer with its associated stable element.


Previous Chapter
Previous Chapter
Sonar Home Page
Sonar Home Page
Next Chapter
Next Chapter


Copyright © 2013, Maritime Park Association.
All Rights Reserved.
Legal Notices and Privacy Policy
Version 3.00