BASIC RADAR SYSTEMS

LEARNING OBJECTIVES Upon completing this chapter, you should be able to do the following: 1. Describe the two types of energy transmission used by radar systems. 2. Describe the different seaming techniques used in radar systems. 3. Describe the major components in todays radar transmitters. 4. List the basic design requirements of an effective radar receiver.

Introduction
The two basic types of radar systems, pulse radar sand continuous-wave radars, use pulse and continuous wave energy transmission. As a Fire Control-man, you need to know how these systems work. This chapter discusses radars, scanning methods, transmitters, and receivers in detail. For further information on radar systems, refer to Microwave Principles, Module 11, Navy Electricity and Electronics Training Series, NAVEDTRA 172-11-00-87; and Radar Principles, Module 18, Navy Electricity and Electronics Training Series, NAVED-TRA 172-18-00-84.

RADAR SYSTEMS
This section discusses the two types of pulse radars and the two types of continuous-wave radars.

PULSE-RADAR SYSTEMS
The pulse-radar systems include the basic pulse radar and the pulse-Doppler radar.

Basic Pulse-Radar System

The signal of a basic pulse-radar system is generated by the transmitter and is radiated into space by the antenna. Intermediate frequencies from 30 to 60MHz are commonly used because signal handling is easier to accomplish at the lower frequencies than at the transmitter frequencies. The duplexer enables the use of a single antenna to transmit and receive the radar signal. The return echo signal is then mixed with a local oscillator (LO)signal to produce an intermediate frequency (IF)signal at a low er frequency than the transmitter. The IF filter conditions the echo signal through amplifying and filtering extraneous signals. The IF signal is then sent to the second detector where the IF is converted to a lower frequency video signal. At that time, the video signal is processed for display by a video amplifier. The display is usually a cathode-ray tube (CRT) that is monitored by an operator. The timer/synchronizer controls the repetition frequency of the transmitter. It can also provide a zero range start signa l for the display device. Basic pulse-radar systems are rather complex in their composition, but they all contain the same basic functional areas, with additional equipment included for specific purposes. For instance, a search radar requires additional circuitry to indicate antenna azimuth position coincident with a particular target echo. Additional circuitry might also be added to a search radar for moving-target indication (MTI) and to filter out stationary targets, landmasses, and clutter from weather and the sea state. A tracking radar, such as a firecontrol radar, requires additional circuitry to measure target range, azimuth, and elevation. Since circuitry is also requi red to keep the antenna pointed at the target, fire-control radars have ranging and angle tracking systems included. Figure 2-1 shows a basic pulse radar.

Figure 2-1 shows a basic pulse radar. Pulse-Doppler Radar System A pulse-Doppler radar has certain advantages over a basic pulse radar or a continuous-wave radar. It can detect both stationary and moving targets and can also determine range. In addition, it can distinguish be-tween two targets with the same radial velocity, but at different ranges. The radial velocity is the apparent speed that the target is closing on or going away from the radar. Pulse-Doppler radar, however, has some disadvantages, too: blind target ranges and velocities, range and delivery ambiguities, and reduction in maximum range capabilities. These disadvantages can be compensated for by using additional circuitry. A pulse radars echo signal also contains velocity information in the Doppler frequency information, but the signal is not normally used by a basic pulse radar. By using the Doppler signal available on an echo signal, a basic pulse radar can detect a weak signal from a moving target in the presence of strong signals from large targets, such as landmasses and heavy seas. Pulse-Doppler radars use the Doppler shift signal. These radars can detect an aircraft flying close to a hill or mountain where the strong landmass echo would block detection with a basic pulse radar. Circuitry in the pulse-Doppler radar normally would reject the stationary target, allowing easy detection of the weak signal from the moving target. The ranging system of pulse-Doppler radar is more complex than that of a pulse or frequency-modulated, continuous-wave (FMCW) radar. A pulse-Doppler radar senses both range and velocity by timesharing its waveform between these functions. To detect a Doppler frequency from the target echo, most pulseDoppler radars use a much higher pulse-repetition frequency (PRF) than basic pulse radars. Higher PRF decreases the pulse-repetition time(PRT) between pulses, resulting in the possibility of a target echo returning at the time of the next transmission. This, in turn, results in a blind spot in the range. If the echo from the first pulse returns after the second pulse is transmitted, then a range ambiguity occurs. The range blind spot and ambiguities can be compensated for by changing the PRF over a wide range. For example, the fire-control computer could adjust the radar PRF based on the expected range of a designated target. If the designation were for a target at 50 nmi (100 kyd), the PRF could be changed so that

the second pulse would not occur until enough time had elapsed for the target echo to return for that range, plus an additional range interval for the acquisition and tracking gates. Varying the PRF over a wide range by computer control can resolve range ambiguities and blind ranges.

DOPPLER SHIFT THEORY. A Doppler shift allows distinguishing between the target and the transmitter leakage. The amount of Doppler shift is determined by the radial velocity of the target since the radial velocity is the apparent speed that the target is closing on or going away from the radar. A target can move in any direction and in a wide range of speed; therefore, the radial velocity can change considerably. If the target is moving at a 900angle to the radar, then no Doppler shift is produced. However, if the target moves straight at or away from the radar, radial velocity will equal the actual target speed. The amount of Doppler shift is also dependent on the wavelength resulting from the transmitter frequency. A target radial velocity that produces a specific Doppler s hift at 5,000 MHz would produce twiceas much at 10,000 MHz. DOPPLER SHIFT DETECTION. Pulse-Doppler radars can detect moving targets by the Doppler shift. Movingtarget indication (MTI) is used primarily to detect moving targets with pulse-Doppler search radars. Since stationary targets produce no Doppler shifts, the return signal echo has the same frequency and phase as the transmitted pulse. However, moving targets do produce Doppler shifts; therefore, the return signal echo has a different phase from that of the transmitted pulse. To use this principle, pulseDoppler radars must be able to compare the echo signal with a reference signal that is in phase (coherent) with the transmitted signal. A means of storing or controlling the transmitted phase is required to provide coherent detection. One method uses a magnetron for the transmitter. This requires that the local oscillator be stable for a small fraction of a cycle during one pulse period. A sample of the transmitted and stable local oscillator (STALO) signals is fed to the coherent oscillator (COHO). This locks in the phase of the COHO until the next transmitter pulse. Figure 2-2 is a diagram of coherent MTI with a phase-locked COHO oscillator.

Figure [Link] MTI with a phase-lockedCOHO oscillator.

With the development of power amplifier kly-strons, traveling-wave tubes, and crossedfield amplifiers, a much better method evolved for coherency. A pulsepower amplifier fed by a stabilized master oscillator (STAMO) makes up the transmitted signal. The STAMO signal is mixed with the IF oscillator to provide an RF mixer input. The receiver then detects the Doppler shift and produces the video signal. The video signal from either type of coherent radar is usually bipolar (both positive and negative).The bipolar video that is detected in a coherent receiver is caused by the phase and frequency differences of the return signals. Stationary targets, such as land, produce the same phase/frequency return on each pulse, whereas moving targets produce changing phase returns on each pulse. The MTI systems use pulse-to-pulse cancellation to suppress the stationary target returns by subtracting the previous return from the current return. Figure 23 is a diagram of a coherent MTI with a STAMO oscillator.

CONTINUOUS-WAVE RADAR SYSTEMS

The continuous-wave radar systems include the basic continuous-wave radar and the FM-CW radar.

Basic Continuous-Wave Radar System

A basic continuous-wave (CW) radar employs a continuous transmission that results in a continuous echo signal from a target. With a continuous echo signal, determination of the target range is impossible, since there is no distinguishing start and stop of the echo signal. Leakage from the transmitter into the

receiver has the same form and could be classified as a target. If the target is moving radially with respect to the transmitter, then a shift in frequency occurs. This shift is called the Doppler shift or Doppler frequency. Most CW radars use a separate antenna to receive, since there is no transmit rest period as in pulse radars. This also improves the isolation between the transmitted signal and the received echo signals. Basic CW radars with a single antenna use a ferrite circulator to act as a duplexer. These circulators are limited to lower-power CW radars, because the amount of leakage from the transmitter to the receiver is about 20 to 30 decibels (dB) down from the transmitter power level. For a 1-watt transmitter, this would be 1 milliwatt of leakage. More leakage than this could easily damage the receiver.

The simple CW radar has the same basic components found in pulse radars; the main difference is the use of a separate antenna to receive and the use of filters to detect a Doppler shift. The filters are normally designed for the IF range, since working at the transmitter frequencies is more critical in the construction of circuits. The filters are set up to detect a particular narrow frequency band. The bands are set so they are adjacent to each other and cover the expected Doppler frequencies above and below the zero shift. The narrower the filter bandwidth, the more filters are required and the more discrete velocities can be determined within the receiver bandwidth. With the basic CW radar, target range cannot be determined. However, a target can be tracked by the angle method. Angle resolution is determined the same way for both CW and pulse radars. CW radars have an advantage over pulse radars when detecting moving targets in clutter. This is very useful when the clutter is caused by chaff, since detection is based on the Doppler frequency and not on are turn pulse. With a simple CW radar, it is almost impossible to detect a stationary target because of clutter and leakage at the transmitter frequency. Moving targets are easily detected, but range determination is more difficult. A diagram of a basic CW radar is shown in figure

[Link] [Link] CW radar.

FM-CW Radar System

The limitation of being unable to determine target range with C W radars can be overcome by frequency modulating the transmitted signal. The resulting radar is the frequency-modulated, continuous-wave (FM-CW) radar. The transmitter still transmits continuously, but the frequency is changed in a predetermined fashion. The modulation can be in the form of a saw tooth, triangular, sinusoidal, or other shape as long as it produces a frequency change of known rate.

SCANNING METHODS
For a radar to track a target, some means of keeping the radar beam pointed at the target is required. The radar system must be able to determine in which direction the radar beam must be moved so that the target remains in the center of the beam. A visual indication on the CRT can depict range and angle of the target in respect to the beam center. The operator can then move the beam by positioning the antenna center on the target. Todays radar systems use computer-aided automatic tracking systems. Their two basic scanning methods are mechanical and electronic scanning.

MECHANICAL SCANNING
Mechanical scanning can be flexible in that the antenna can be moved in one of two desired patterns:(1) The feed horn can be moved relative to a fixed reflector, or (2) the reflector can be moved relative to a fixed feed horn. The most common mechanical scan technique used by fire-control radars is a movable feed horn relative to a fixed reflector, which is called conical scanning. Included in mechanical scanning are nutation, nutating waveguide, and angle tracking.

Nutation
The nutation process is difficult to describe inwords, but easy to demonstrate. Hold a pencil hand at the eraser end and the otherhand at the point. While holding the eraser end as stillas oint in a circle to form a [Link] motion of the pencil is called nutation. The pencilpoint corresponds to the transmitting end of a waveguide [Link] important point is that the polarization of theRF beam does not change during s means that the axis of the moving feed horn mustnot change horizontal or vertical orientation ing. The movement might be compared tothat of a Ferris wheel; the orientation of the seats remains vertical, regardless of where they are on thewheel. withboth hands, one possible, move the p

a nutation [Link] while thefeed is mov

Nutating Waveguide
The waveguide feed is a metal pipe, usually rec-tangular in cross section. The open end of the waveguide feed faces the antenna disk reflector, and theenergy it emits is bounced from the reflector surface.A conical scan can be generated by nutation of thewaveguide feed. In this process, the axis of thewaveguide feed is nut ated through a narrow [Link] movement is fast (from 30 to 60 hertz) andsmall in amplitude. To an o bserver, the waveguidefeed would appear to be vibrating slightly. The amplitude of the nutation determines the angle of the [Link] amplitude is kept small so that sufficient poweris present i n the center of the conical pattern for targettracking. Figure 25 shows the conical pattern produced by nutating the RF beam.

Figure [Link] lobe.

Angle Tracking
The radars angle error-detecting circuits provide correction signals to the antenna and director drive circuits. The correction signals are proportional to the target displacement from the nutation axis of the antenna. Target displacement is detected by first locking onto the target in range, so that only that target is gated and used for angle tracking. A two-phase reference generator in the antenna establishes the position of the feed horn relative to the nutation axis of the antenna. The generator provides the reference voltages for the angle error-detection circuits. The elevation reference voltage is 900 out of phase with the azimuth reference voltage. Each errordetector circuit compares the phase ofits reference voltage with the phase of its video envelope. The phase comparison indicates the directionof the error. The amplitude modulation of the videoenvelope is compared to the amplitude of the refer-ence voltage to determine the amount of error.

ELECTRONIC SCANNING
Several techniques are used for electronic scanning. The two most common techniques used in firecontrol radars are monopulse (simultaneous lobing) scanning and phasing scanning. Monopulse scanning does not move the trans-mitted beam; instead, the echo signal is scanned or compared. Phasing scanning causes an actual movement of the radar beam with respect to the antennaaxis. In addition, electronic scanning includes conical-scan-on-receive-only (COSRO) scanning techniques. Monopulse Scanning With monopulse (simultaneous lobing) scanning, range, bearing, and elevation angle information of a target is obtained from, as the name implies, a single pulse. This type of tracking radar normally produces a narrow circular beam of pulsed RF energy at a high pulse-repetition rate (PRR). Each pulse is divided into four signals that are equal in amplitude and phase. The four signals are radiated at the same time from each of four feed horns that are grouped in a cluster. The radiated energy is focused into a beam by a

microwave lens. In turn, energy reflected from the target is refocused by the lens into the feed horns. The amount of the total energy received by each horn varies, depending on the position of the target relative to the beam axis. The four targets are at different positions with respect to the beam axis. A phase inversion takes place at the microwave lens similar to the image inversion in an optical system. The amplitude of returned signals received by each horn is continuously compared with those received in the other horns, and error signals are generated that i ndicate the relative position of the target with respect to the axis of the beam. Angle servo circuits receive these error signals and correct the position of the r adar antenna and the director to keep the beam axis on target. An important advantage of a monopulse-tracking radar over a radar using conical scan is that the instantaneous angular measurements are not subject to errors ca used by target scintillation. Scintillation is the rapid fluctuation of the echo signal amplitude as the target maneuvers or moves, resulting in radar beams bouncing off different areas of the target and causing random reflectivity, which may lead to tracking errors. A monopulse-tracking radar is not subject to this error because each pulse provides an angular measurement without regard to the rest of the pulse train; therefore, scintillation does not affect the measurement. An additional advantage of monopulse tracking isthat no mechanical action is required, such as a scanner. Figure 2-6 (on page 2-8) shows monopulse variations of received energy with target positions. Phasing Scanning Phased array antennas use the phasing scanning method. This method controls the phase of the RF signals fed to multiple feed horns, dipoles, or radiators. The angular position of the beam is determined by the relative phase of the signals at each element. When the phase of the signals applied is changed, the beam can be steered without moving the antenna. Another method of phasing uses changes in the transmitter frequency. Changing the frequency changes the wavelength; thus, with a fixed length of waveguide between the elements, the phase relationship changes as the frequency changes. Phased arrays can be used for tracking by a monopulse-receiving technique. The array is divided into quadrants, with each quadrant equivalent to one of the four horns. The sum of all four quadrants is compared to the sums and differences of different quadrants, just as in monopulse scanning. This technique is also adaptable to receive-only antenna systems. COSRO Scanning Scanning the received signal by electronically switching between the antenna elements (feed horns)to produce a conical antenna receive pattern is called conical-

scan-on-receive-only

(COSRO). COSRO scanning is used with monopulse transmissions and with single-beam transmissions. Angle errors are produced in much the same way as mechanical conical scanning. However, COSRO scanning is less effective than monopulse scanning, since many pulses are required to produce the targetecho envelope. While COSRO scanning is used more effectively with CW radars that use a separate receive antenna, it may also be used as an alternate tracking method, such as an electronic protection (EP) technique.

Figure [Link] variations of received energy with target position.

TRANSMITTERS Search and fire-control radars require high-powered oscillators and amplifiers to produce the transmitted RF signal. The high-power levels required for these radars enable the detection and tracking of targets at extended ranges. Solid-state transmitters have replaced most of the radar transmitters that once used vacuum tubes to provide high-power RF energy. The high-power tubes used fall into two categoriescrossed field and linear beam. Each type has different characteristics, making one tube more suited for one application than another. CROSSED-FIELD TUBES

Crossed-field tubes get their name from the fact that the dc electric field and the magnetic field are crossed at right angles. One of the first crossed-field tubes used in early radars was the magnetron, and it was the only one available for quite a few years. Crossed-field tubes are also known as M-type devices, since they deal with propagation of waves in a magnetic field. There are two types of crossedfield tubes: resonant and non resonant.. Resonant Tubes: Resonant tubes are oscilla-tors and generate the RF signal. The most commonresonant crossedfield tube used in radars is themagnetron.. Nonresonant Tubes: Nonresonant crossedfield tubes are amplifiers and generally will not oscillate, but, instead, will amplify an applied RF signal. The amplifiers are subclasses as to whether they use the forward or backward wave and whether they are reentrant. (Reentrant means whether the electrons emitted by the cathode that return to the cathode can reenter the charge that travels to the anode or are then lost [wasted].) Only one type of nonresonant crossedfield tube has found wide use in radar application: the crossed-field amplifier (CFA). The CFA, discussed in a later subsection, is nonresonant, backward wave, and reentrant. Magnetrons Basically, the magnetron is a diode and has no grid. A magnetic field in the space between the plate(anode) and the cathode serves as a grid. The plate of a magnetron does not have the same physical appearance as the plate of an ordinary electron tube. Since conventional inductance-capacitance (LC) networks become impractical at microwave frequencies, the plate is fabricated into a cylindrical copper block containing resonant cavities that serve as tuned circuits. The magnetron base differs greatly from the conventional base, as it has short, large-diameter leads that are carefully sealed into the tube and shielded. The cathode and filament structure is at the center of the tube and is supported by the filament leads, which are large and rigid enough to keep the structure fixed in position. The output lead is usually a probe or a loop extending into one of the tuned cavities and coupled into a waveguide or a coaxial line. The plate structure is a solid block of copper. The cylindrical holes around its circumference are resonant cavities. A narrow slot runs from each cavity into the central portion of the tube and divides the inner structure into as many segments as there are cavities. Alternate segments are strapped together to put the cavities parallel to the output. These cavities control the output frequency. The straps are circular metal bands that are placed across the top of the block at the entrance slots to the cavities. Since the cathode must operate at high power, it must be fairly large and be able to withstand high operating temperatures. It must also have good emission characteristics, particularly under back bombardment, because much of the output power is derived from the large number of electrons emitted when high-velocity electrons return to strike the cathode. The cathode is indirectly heated, and is constructed of a high-emission material. The open space between the plate and the cathode is the interaction space, where the electric and magnetic fields interact to exert force upon the electrons. The magnetic field is usually provided by a strong, permanent magnet mounted around the magnetron so that the magnetic field is parallel with the axis of the cathode. The cathode is mounted in the center of the interaction space. The direction of an electric field is from the positive electrode to the negative electrode. The law governing the motion of an electron in an electric (E) field states that the force exerted by an electric field on an electron is proportional to the strength of the field. Electrons tend to move from a point of negative potential toward a positive potential. In other words, electrons tend to move against the E field. When an electron is being accelerated by an E field, energy is taken from the field by the electrons. The law of motion of an electron in a magnetic(H) field states that the force exerted on an electron in

a magnetic field is at right angles to both the field and the path of the electron. The direction of the force is such that the electron trajectories are clockwise when viewed in the direction of the magnetic field. Magnetron oscillators are divided into two classes: negative resistance and electron resonance.. Negative Resistance: A negative resistance magnetron oscillator operates by a static negative resistance between its electrodes and has a frequency equal to the natural period of the tuned circuit connected to the tube. The split-anode negative resistance magnetron is a variation of the basic magnetron, which operates at a higher frequency and is capable of more output.. Electron Resonance: An electron resonance magnetron oscillator operates by the electron transittime characteristics of an electron tube (the time it takes electrons to travel from the cathode to the plate).It is capable of generating very large peak-power outputs at frequencies in the thousands of megahertz. The electron resonance magnetron is most widely used for microwave frequencies. Modern designs have a reasonably high efficiency and relatively high output. However, one disadvantage of the electron resonance magnetron is that its average power is limited by the cathode emission. Furthermore, the peak power is limited by the maximum voltage that it can withstand without damage. The operation, coupling methods, tuning, and seasoning of magnetrons are discussed in the following subsections. Since this information is general in nature, the recommended times and PMS procedures in equipment technical manuals should be followed when baking-in a specific type of magnetron. Figure 2-7 shows a magnetron.

Figure [Link].

MAGNETRON OPERATION. The electricfield in the electron resonant oscillator consists of thealternating current (at) and direct current (de) field [Link] action of electronics taking energy away from theac fields is an undesirable effect. The dc field extendsradially b etween adjacent anode segments by the RFoscillations induced in the cavity tank circuits of theanode block. For mor

e information on magnetronoperation, refer to your appropriate operating proce-dures for your firecontrol system. MAGNETRON COUPLING METHODS. The RF energy can be removed from a magnetron bya coupling loop. At frequencies lower than 10,000MHz, the coupl ing loop is made by bending the innerconductor of a coaxial cable into a loop and solderingthe end to the outer con ductor so that the loop projectsinto the cavity. To obtain sufficient pickup at higherfrequencies, the loop is located at the end of the cav-ity. MAGNETRON TUNING. A tunable magnetron permits the system to be operated at a precisefrequency anywhere within a band of frequencies, asdetermin ed by the magnetrons characteristics. Theresonant frequency of a magnetron can be varied byvarying the inductanc e or capacitance of the resonantcavities. MAGNETRON SEASONING. During initial operation, a highpower magnetron arcs from thecathode to the plate and must be properly broken-in orbakedin. Actually, arcing in magnetrons is very common. It may occur with a new tube or following along period of idleness. One of the prime causes of arcing is the liberationof gas from tube elements during idle periods. Arcingmay also be c aused by sharp surfaces within the tube,mode shifting, and excessive current. Although thecathode can withstand co nsiderable arcing for short periods of time, continued arcing will shorten the lifeof the magnetron and may destroy it entirely. Hence,each time e xcessive arcing occurs, the tube must be baked-in again until the arcing ceases and the tube isstabilized. The bakingin procedure is relatively [Link] voltage is raised from a low value untilarcing occurs several times a second. The voltage isleft at that value until arcing dies out. Then the voltage is raised further until arcing again occurs, and it isleft at that value until the arcing again dies out. Whenever the arcing becomes very violent and resembles acontinuous arc, the applied voltage is excessive andshould be reduced to permit the magnetron to [Link] the normal rated voltage is reached and themagnetron remains stable at the rated current, thebaking-in is complete. It is a good maintenance prac-tice to bakein magnetrons left idle in the equipment,or those used as spares, when long periods of nonoperating time have accumulated. Crossed-Field Amplifier The crossedfield amplifier (CFA) is constructedvery similar to a magnetron. The major difference isthat the CFA requires an RF i nput in addition to theelectrical input and the magnetic [Link] CFA anode is very similar to that of a regulartrapezo id block magnetron, with the exception of thedrift tube section. The drift tube section serves twopurposes: (1) to dampen out electron bunches, andthus oscillations, as the spokes move past the outputport, and (2) to provide a pat h for RF energy from theinput to the [Link] CFA is normally used in a chain of two ormore CFAs in series. Wh en the CFA is not pulsedwith high voltage, the tube presents a low-impedance-toinput RF and passes it through to the output [Link] used in a chain, one or more CFAs can beenergized, depend ing on the output power [Link] example, for closein targets, the final stage orstages may not be needed, but all CFAs are requiredfor detecting small targets at maxi mum [Link] cathode of a CFA is cylindrical, just as inmagnetrons; however, most are cold cathodes (no filaments or heaters). With a cold cathode, sufficientelectron emission is available to form a space chargewhen an el ectric field is applied. The electrons willnot achieve sufficient velocity to escape the regionclose to the cathod e without an RF field [Link], when an RF field is applied, the electronsgain some energy from the RF and move farther outfrom the cathode. While some will gain even moreenergy and reach the anode, others will r eturn to thecathode and strike with sufficient energy to causesecondary emission and increase the electrons in the space charge. Therefore, the major source of electronsis secondary emission from the reentrant [Link] e mission generates heat in the cathodeand further increases emission. If the cathode be-

comes too hot, thermionic emission may becomesufficient to allow the tube to oscillate or to producenoise, even t hough an RF field has not been [Link] prevent the spurious generation of noise, thecathode is usually watercooled. Another means ofreducing noise output is to make the RF pulse slightlywider than the highvoltage [Link] control is found on some tubes with a biaselectrode (a cutoff electrode). The bias electrode hasa positive pulse (with respect to the cathode) appliedjust before the end of the RF pulse. The cutoff pulseallows the el ectrode to collect the electrons in thedrifttube region and the tube to shut off. The pulsemust be wide enough to extend past the end of the [Link] do not have the gain of some of the linearbeam tubes. However, they do have three advantages.1. The CFAs, when used in a multistage chain,can produce equal or greater overall gain at lowerhigh-voltage requirements than a linear- beam amplifier, such as a multicavity klystron. For example, atypical CFA can deliver a 1megawatt RF peak poweroutput with a 40-kilovolt, 50-ampere peak pulse,where a klystron would require a 90kilovolt, 40-ampere peak pulse to produce the same RF [Link], the CFA would require a higher power levelinput drive signal than the klystron. 2. The CFA is a cold cathode, which normallyhas a much longer operating life than a heated cathode.3. The CFA produces far less X-rays than linear-beam tubes; therefore, lead shielding is not required.

LINEAR-BEAM TUBES
A linearbeam tube uses a magnetic field that isparallel to the electron beam and is used to focus thebeam. Some tubes do not use a magnetic field; instead, electrostatic focusing is used to hold the beamtogether while it travels the length of the tube. Thetwo most co mmon types of linear-beam tubes used infire-control equipment are the klystron amplifier andthe travelingwave tube.

Klystron Amplifiers
The basic theory of a klystron amplifier is quitesimple. The klystron amplification principle may bereadily explained with an analogy to a simple triodeamplifier with tuned plate and grid circuits. Klystronamplifiers include twocavity power klystrons andmulticavity power klystrons.

TWO-CAVITY POWER KLYSTRON AM-PLIFIERS. Figure 2-8 shows a simplified schematic of a triode amplifier with resonant circuits atboth the input and the output. Such resonant circuitsrestrict t he bandwidth of the amplifier and increasethe gain.

Figure [Link] schematic of a triode amplifierwith resonant circuits at both the input and the output.

A triode tube consists of three elements: (1) acathode that emits a stream of electrons, (2) a grid thatcontrols the electron stream, and (3) a plate thatattracts the electrons and catches them after they passthrough the grid.1. Cathode: The cathode emits a stream of electrons.2. Grid: The grid acts as a valve, opening orclosing the current path according to the voltage applied to it. The RF input signal comes to the grid as aweak alternating voltage. This voltage modulates theelectron flow through the tube at the radio frequency.3. Plate: The electron stream then delivers, at theplate, an alternating current, which is an amplifiedreproduction of th e input signal. This alternatingcurrent flows through the resonant plate circuit andexcites alternating voltages ac ross it. These voltagesconstitute the RF output from the [Link] time it takes electrons to cross the tube is approximately onebillionth of a second. This transittime is short compared to the period of a cycle of aradio wave below the micr owave range (approxi-mately one-millionth of a second). Hence, the electrons are slowed down or speeded up by the voltageon the grid at a given instant. The flow of electrons,therefore, c an follow the voltage fluctuation on [Link] the case of microwaves, however, the oscillations are so rapid (the cycle period is so short) thatthe voltage on the grid may go through several complete oscillations while a particular quantity ofelectrons travel across the tube. In other words, thegrid voltage changes too rapidly for the electrons tofollow the fluctuation. There are other reasons whythe conventional triode tube fails at microwave frequencies, but the most fundamental reason is that thetransit time of the electrons is long when compared tothe peri od of one cycle of the microwave [Link] klystron amplifier makes a virtue of the verything that defeats the trio dethe transit time of theelectrons. The klystron amplifier modulates the velocity of the electrons, so that as the electrons travel through the tube, electron bunches are formed. Thesebunches deliver an oscillating current to the outputresonant circuit of the klystron. The klystron amplifierconsists of three separate sections: the electron gun,the RF section, an d the collector.. Electron Gun: The electron gun consists of aheater, a cathode, a control grid, and an anode. Electrons are emitted by the cathode and drawn toward theanode, which is operated at a positive potential withrespect t o the cathode. The electrons are formed intoa narrow beam by either electrostatic or magneticfocusing techniques. The control grid is used to con-trol the number of electrons that reach the anode region. It may also be used to turn the tube completelyon or off in certain pulsed-amplifier [Link] by the time it reaches the anode, theelectron beam passes through a hole in the anode andon to the RF sect ion of the tube, eventually strikingthe collector. The electrons are returned to the cathodethrough an external power supply. It is evident thatthe collector of a klystron amplifier acts much like theplate of a triode insofar as the coll ecting of electronsis concerned. However, there is one important difference. The plate of a triode is normally connected,in some fashion, to the output RF circuit, whereas ina klystron a mplifier, the collector has no connectionto the RF circuitry at all.. RF Section: The RF section of a basic klystron amplifier is quite different from a conventionaltriode amplifier. The resonant circuits used in a klystron amplifier are reentrant cavities.. Collector: The collector is normally insulatedfrom the RF section of large klystron amplifiers topermit separate metering of the electrons interceptedby the drift tubes and those interce pted by the collector. The electrons intercepted by the RF section arecalled body current, whereas electrons intercepted bythe collec tor are referred to as collector current. Obviously, the sum of the body current and the collectorcurrent is equal to the total current in the electronbeam, which is called beam current. Klystron amplifier specifications often place a maximum limit onallowable body current. The collector of most highpower klystrons is insulated from the body of thetube, This allows separate metering and overload pro-