CHAPTER 3

INTRODUCTIO TO TRANSMISSION LINES N AND WAVEGUIDES

A TRANSMISSIO N LINE is a devic e designed to guide electrica l energ y from one poin t to another. It is used , for example , to transfe r the outpu t rf energy of a transm itter to an antenna. This energy will not travel through normal electrical wire without great losses . Althoug h the antenn a can be connected directly to the transmitter, the antenna is usually located some distance away from the transm itter. On boar d ship , the transmitte r is locate d insid e a radio room, and its associated antenna is mounted on a mast. A transmission line is used to connect the transm itter and the antenna. The transmissio n line has a singl e purpos e for both the transmitter and the antenna. This purpose is to transfe r the energ y outpu t of the transm itte r to the antenn a with the leas t possibl e powe r loss. How well this is done depend s on the specia l physica l and electrical characteristics (impedance and resistance) of the transmission line. TRANSMISSIO N LINE THEORY flow tha t may be expecte d throug h the insulation, If the line is unifor m (all value s equa l at each unit length) , the n one smal l sectio n of the line may represen t severa l feet. Thi s illustratio n of a two-wire transmissio n line will be used throughou t the discussion of transmission lines; but, keep in mind that the principle s presente d apply to all transmissio n lines. We will explain the theories using LUMPED CONSTANTS and DISTRIBUTED CONSTANTS to further simplify these principles. LUMPE D CONSTANTS A transmissio n line has the propertie s of inductance, capacitance, and resistance just as the more conventional circuit s have . Usually , however , the constants in conventional circuits are lumped into a singl e device or component . Fo r example , a coil of wire has the propert y of inductance . When a certain amoun t of inductanc e is neede d in a circuit , a coil of the prope r dimension s is inserted . The inductance of the circui t is lumpe d into the one component . Two meta l plate s separate d by a smal l space , can be used to suppl y the require d capacitanc e for a circuit . In such a case, most of the capacitance of the circuit is lumpe d into this one component . Similarly , a fixed resisto r can be used to suppl y a certai n valu e of circuit resistance as a lumped sum. Ideally, a transmission line would also have its constant s of inductance, capacitance, and resistance lumped together, as shown in figure 3-1. Unfortunately, this is not the case. Transmission line constants are as described in the followin g paragraphs. DISTRIBUTE D CONSTANTS Transmission line constants, called distributed constants , are sprea d along the entir e lengt h of the transmissio n line and canno t be distinguishe d separately. The amount of inductance, capacitance, and resistanc e depend s on the lengt h of the line, the size of the conductin g wires , the spacin g betwee n the

The electrica l characteristic s of a two-wire transmission line depend primarily on the construction of the line. The two-wir e line act s like a long capacitor. The change of its capacitive reactance is noticeabl e as the frequenc y applie d to it is changed. Since the long conductor s have a magneti c field about them when electrica l energ y is being passe d through them, they als o exhibi t the propertie s of inductance. The value s of inductanc e and capacitanc e presented depen d on the variou s physica l factor s tha t we discusse d earlier . For example , the type of line used, the dielectri c in the line, and the lengt h of the line must be considered . The effects of the inductiv e and capacitiv e reactanc e of the line depen d on the frequenc y applied . Since no dielectri c is perfect, electron s manag e to move from one conducto r to the othe r throug h the dielectric . Eac h type of two-wire transmission line also has a conductance value. This conductanc e valu e represent s the valu e of the current

3-1

Figure 3-1.Two-wire transmission line.

Figure 3-3.Distributed capacitance.

wires, and the dielectric (air or insulating medium) betwee n the wires . The followin g paragraph s will be usefu l to you as you stud y distribute d constants on a transmission line. Inductance of a Transmission Line

R esistance

of a Transmission

Line

When current flows through a wire, magnetic lines of force are set up aroun d the wire. As the current increase s and decrease s in amplitude , the fiel d around the wire expand s and collapse s accordingly . The energ y produce d by th e magneti c lines of force collapsin g back into the wire tend s to keep the current flowing in the same direction. This represents a certain amoun t of inductance , which is expresse d in microhenrys per unit length. Figure 3-2 illustrates the inductanc e and magneti c fields of a transmission line. Capacitance of a Transm ission Line

The transmission line shown in figure 3-4 has electrical resistance along its length. This resistance is usuall y expresse d in ohms per uni t lengt h and is show n as existin g continuousl y from one end of the line to the other.

Figure 3-4.Distributed resistance.

Leakage

Current

Capacitanc e also exist s betwee n the transmission line wires, as illustrated in figure 3-3. Notice that the two paralle l wire s act as plate s of a capacito r and that the air between them acts as a dielectric. The capacitanc e betwee n the wire s is usuall y expressed in picofarad s per uni t length . This electri c field betwee n the wire s is simila r to the field tha t exists betwee n the two plate s of a capacitor.

Since any dielectric , even air, is not a perfect insulator , a smal l curren t know n as LEAKAGE CURREN T flows betwee n the two wires . In effect, the insulator acts as a resistor, permitting current to pass betwee n the two wires . Figur e 3-5 show s this leakag e path as resistor s in paralle l connecte d between the two lines . Thi s propert y is called CONDUCTANCE (G) and is the opposit e of resistance. Conductanc e in transmissio n lines is expresse d as the reciproca l of resistanc e and is usuall y given in micromhos per unit length.

Figure 3-2.Distributed inductance.

Figure 3-5.Leakage in a transmission line.

3-2

ELECTROMAGNETI

C FIELD S

CHARACTERISTI C

IMPEDANCE

The distributed constants of resistance, inductance, and capacitanc e are basic propertie s commo n to all transmission lines and exist whether or not any current flow exists . As soon as curren t flow and voltag e exist in a transmissio n line , anothe r propert y become s quite evident . This is the presenc e of an electromagnetic field, or lines of force, abou t the wire s of the transmission line. The lines of force themselves are not visible ; however , understandin g the force tha t an electro n experience s while in the field of thes e lines is ver y importan t to your understandin g of energy transmission. Ther e are two kind s of fields; one is associated with voltag e and the othe r wit h current . The field associate d with voltag e is calle d the ELECTRI C (E) FIELD . It exert s a force on an y electri c charg e placed in it. The field associate d wit h curren t is called a MAGNETI C (H) FIELD , becaus e it tend s to exert a force on any magneti c pole place d in it. Figur e 3-6 illustrate s the way in which the E fields and H fields tend to orien t themselve s betwee n conductor s of a typical two-wire transmission line. The illustration shows a cross section of the transmission lines. The E field is represente d by solid lines and the H field by dotte d lines . The arrow s indicat e the directio n of the lines of force. Both fields normall y exis t together and are spoke n of collectivel y as the electromagnetic field.

You can describe a transmission line in terms of its impedance . The rati o of voltag e to curren t (Ein/I in) at the input end is known as the INPUT IMPEDANCE (Z in). This is the impedanc e presente d to the transmitter by the transmission line and its load, the antenna. The rati o of voltag e to curren t at the outpu t (E OUT /I OUT ) end is known as the OUTPUT IMPEDANCE (ZOUT ). This is the impedanc e presente d to the load by the transmission line and its source. If an infinitely long transmissio n line could be used , the rati o of voltage to current at any point on that transmission line would be some particula r valu e of impedance . This impedance is know n as the CHARACTERISTI C IMPEDANCE. The maximum (and most efficient) transfer of electrica l energ y take s place when the sourc e impedance is matche d to the load impedance . Thi s fact is very important in the study of transmission lines and antennas. If th e characteristi c impedanc e of the transmissio n lin e and the load impedanc e are equal, energy from the transmitter will travel down the transmissio n line to the antenn a with no powe r loss caused by reflection. LINE LOSSES

The discussion of transmission lines so far has not directly addressed LINE LOSSES; actually some losses occur in all lines. Line losses may be any of three typesCOPPER , DIELECTRIC , an d RADIATION or INDUCTIO N LOSSES. NOTE: Transmission lines are sometimes referred to as rf lines . In this text the term s ar e used interchangeably. Coppe r Losses One type of coppe r loss is I 2R LOSS. In rf lines the resistanc e of the conductor s is neve r equa l to zero. Wheneve r curren t flows throug h one of thes e conductors , some energ y is dissipate d in the form of heat. This heat loss is a POWER LOSS. W ith copper braid, which has a resistance higher than solid tubing, this power loss is higher.

Figure 3-6.Fields between conductors.

3-3

Anothe r type of coppe r loss is due to SKIN EFFECT. When dc flows through a conductor, the movement of electrons through the conductors cross section is uniform, The situation is somewhat different when ac is applied . The expandin g and collapsing fields abou t eac h electro n encircl e othe r electrons. This phenomenon, called SELF INDUCTION, retards the movemen t of the encircle d electrons . The flux density at the center is so great tha t electro n movement at this poin t is reduced . As frequenc y is increased, the oppositio n to the flow of curren t in the cente r of the wire increases . Curren t in the cente r of the wire become s smalle r and most of the electro n flow is on the wire surface . When the frequenc y applie d is 100 megahertz or higher, the electron movement in the cente r is so smal l tha t the cente r of the wire could be remove d withou t any noticeabl e effect on current. You shoul d be abl e to see tha t the effectiv e crosssectiona l are a decrease s as the frequenc y increases. Sinc e resistanc e is inversel y proportiona l to the cross-sectiona l area , the resistanc e will increas e as the frequenc y is increased . Also, since powe r loss increase s as resistanc e increases , power losse s increase with an increas e in frequenc y becaus e of skin effect. Coppe r losses can be minimize d an d conductivity increased in an rf line by plating the line with silver. Since silver is a better conductor than copper, most of the current will flow through the silver layer. The tubing then serves primarily as a mechanical support. Dielectri c Losses

The atomic structure of rubber is more difficult to distor t tha n the structur e of some othe r dielectric materials. The atoms of materials, such as polyethylene, distor t easily . Therefore , polyethylen e is often used as a dielectri c becaus e less powe r is consumed when its electro n orbit s ar e distorted. Radiation and Induction Losses

RADIAIO N and INDUCTIO N LOSSE S are simila r in tha t both are cause d by the field s surrounding the conductors. Induction losses occur when the electromagnetic field about a conductor cuts through any nearby metallic object and a current is induced in tha t object. As a result , powe r is dissipate d in the object and is lost. Radiatio n losses occur because some magneti c lines of force abou t a conducto r do not retur n to the conducto r when the cycle alternates . Thes e lines of force are projecte d into space as radiation , and this result s in powe r losses . Tha t is, powe r is supplied by the source , but is not availabl e to the load. VOLTAG E CHANGE In an electric circuit, energy is stored in electric and magneti c fields . Thes e fields mus t be brought to the load to transmi t tha t energy . At the load, energy containe d in the fields is converte d to the desire d form of energy. Transm ission of Energy

DIELECTRI C LOSSES result from the heating effect on the dielectri c materia l betwee n the conductors. Power from the source is used in heating the dielectric. The hea t produce d is dissipate d int o the surrounding When ther e is no potentia l difference medium. betwee n two conductors , the atom s in the dielectric materia l betwee n them ar e norma l and the orbit s of the electrons are circular. When there is a potential differenc e betwee n two conductors , the orbit s of the electrons change. The excessive negative charge on one conducto r repel s electron s on the dielectri c toward the positiv e conducto r and thu s distort s the orbit s of th e electrons . A chang e in the pat h of electrons requires more energy, introducing a power loss.

When the load is connecte d directl y to the source of energy, or when the transmission line is short, problem s concernin g curren t and voltag e can be solved by applying Ohms law. W hen the transmission line become s long enoug h so the tim e differenc e between a change occurring at the generator and a change appearin g at the loa d become s appreciable , analysis of the transmission line becomes important. Dc Applied to a Transmission Line

In figur e 3-7, a batter y is connecte d throug h a relativel y long two-wir e transmissio n line to a load at the far end of the line. At the instan t the switch

3-4

is closed, neither current nor voltage exists on the line. When the switc h is closed , poin t A become s a positive potential , and poin t B become s negative . Thes e points of differenc e in potentia l mov e down the line. However , as the initia l point s of potentia l leave points A and B, they are followed by new point s of difference in potential , which the batter y add s at A and B. This is merely saying that the battery maintains a constant potential differenc e betwee n point s A and B. A short time afte r the switc h is closed , the initia l point s of differenc e in potentia l have reache d point s A and B; the wir e section s from point s A to A and point s B to B ar e at the sam e potentia l as A and B, respectively. The point s of charg e are represente d by plus (+) and minus (-) signs along the wires, The directions of the current s in the wire s are represente d by the arrowheads on the line, and the directio n of trave l is indicated by an arro w below the line . Conventional lines of force represen t the electri c field tha t exists betwee n the opposit e kind s of charg e on the wire sections from A to A and B to B. Crosse s (tail s of arrows ) indicat e the magneti c field create d by the electri c field movin g dow n the line. The moving electri c field and the accompanyin g magneti c field constitute an electromagnetic wave that is moving from the generato r (battery ) towar d the load. This wave travels at approximatel y the spee d of ligh t in free space . The energ y reachin g the load is equa l to that develope d at the batter y (assumin g ther e are no losses in the transmissio n line). If the loa d absorb s all of the energy , the curren t and voltag e will be evenly distribute d along the line. Ac Applied to a Transm ission Line

Figure 3-8.Ac voltage applied to a line. line at the spee d of light . The actio n is simila r to the wave create d by the battery , excep t the applie d voltage is sinusoida l instea d of constant . Assum e tha t the switc h is closed at the momen t the generato r voltage is passin g throug h zero and tha t the nex t half cycle make s poin t A positive . At the end of one cycle of generato r voltage , the curren t and voltag e distribution will be as show n in figur e 3-8. In thi s illustratio n the conventiona l lines of force represen t the electri c fields . For simplicity , the magneti c fields are not shown . Point s of charg e are indicate d by plus (+) and minu s (-) signs , the larger signs indicatin g point s of highe r amplitud e of both voltage and current. Short arrows indicate direction of curren t (electro n flow) . The wavefor m draw n below the transmissio n lin e represent s the voltag e (E) and curren t (I) waves . The line is assume d to be infinite in length so there is no reflection. Thus, traveling sinusoidal voltage and current waves continually travel in phas e from the generato r towar d the load, or far end of the line. Wave s travelin g from the generator to the load are calle d INCIDEN T WAVES . Waves travelin g from the load back to the generato r are called REFLECTE D WAVES and wil l be explaine d in later paragraphs. STANDING-WAV E RATIO The measurement of standing waves on a transmission line yields information about equipment operating

When the batter y of figur e 3-7 is replace d by an ac generator (fig. 3-8), each successive instantaneous valu e of the generato r voltag e is propagate d down the

Figure 3-7.Dc voltag e applie d to a line.

3-5

conditions . Maximu m powe r is absorbe d by the load when Z L = Z0 . If a line has no standin g waves , the termination for that line is correct and maximum power transfer takes place. You have probabl y notice d tha t the variatio n of standing wave s show s how nea r the rf line is to being terminated in Z0. A wide variatio n in voltag e along the length means a termination variation means termination far from Z0. A small Z0. Therefore, the near

voltage . Since powe r is proportiona l to the square of the voltage , the rati o of the squar e of the maximum and minimu m voltage s is calle d the powe r standing-wave ratio. In a sense, the name is misleading becaus e the powe r along a transmissio n line does not vary. Curren t Standing-W av e Ratio

ratio of the maximum to the minimum is a measure of the perfectio n of the terminatio n of a line. This rati o is calle d the STANDING-WAV E RATIO (swr) and is alway s expresse d in whole numbers . For example , a rati o of 1:1 describe s a line terminate d in its characteristic impedance (Z0). Voltag e Standing-W av e Ratio

The ratio of maximum to minimum current along a transmissio n line is calle d CURREN T STAND- ING- WAVE RATIO (iswr) . Therefore:

The ratio of maximum voltage to minimum voltage on a line is calle d the VOLTAG E STANDING-WAVE RATIO (vswr) . Therefore:

This ratio is the same as that for voltages. It can be used wher e measurement s are mad e with loop s that sampl e the magneti c field along a line. It give s the same results as vswr measurements. TRANSMISSIO N MEDIUMS The Navy uses man y differen t type s of TRANSMISSIO N MEDIUM S in its electroni c applications. Each mediu m (line or waveguide ) has a certain characteristi c impedanc e value , current-carrying capacity , and physica l shap e an d is designe d to meet a particular requirement. The five types of transmission mediums that we will discuss in this topic include PARALLEL-LINE, TWISTED PAIR, SHIELDE D PAIR, COAXIAL LINE, and WAVEGUIDES. The use of a particular line depends , amon g othe r things , on the applied frequency , the power-handlin g capabilities , and the type of installation.

The vertical lines in the formula indicate that the enclosed quantities are absolute and that the two values are take n withou t regar d to polarity , Dependin g on the nature of the standing waves, the numerical value of vswr range s from a valu e of 1 (ZL = Z0, no standing waves ) to an infinit e valu e for theoreticall y complete reflection . Since ther e is alway s a smal l loss on a line, the minimum voltage is never zero and the vswr is always some finite value. However, if the vswr is to be a usefu l quantity . the powe r losse s along the line must be small in comparison to the transmitted power.

Pow r Standing-Wav e

e Rati o

Parallel

Line

The squar e of the vswr is calle d the POWE R STANDING-WAV E RATIO (pswr) . Therefore :

This ratio is useful because the instruments used to detect standin g wave s reac t to the squar e of the

One type of paralle l line is the TWO-WIR E OPEN LINE, illustrated in figure 3-9. This line consists of two wire s tha t are generall y space d from 2 to 6 inches apart by insulating spacers. This type of line is most often used for powe r lines , rura l telephon e lines , and telegraph lines. It is sometimes used as a transmission line betwee n a transmitte r and an antenn a or between an antenn a and a receiver . An advantag e of thi s type

3-6

Figure 3-11.Twisted pair. Figure 3-9.Two-wire open line.

of line is its simpl e construction.

The

principal

Shielde d Pair

disadvantages of this type of line are the high radiation losses and electrica l noise picku p becaus e of the lack of shielding . Radiatio n losses ar e produce d by the changing fields created by the changing current in each conductor. Anothe r type of paralle l line is the TWO-WIRE RIBBON (TWIN LEAD) LINE, illustrated in figure 3-10. This type of transmission line is commonly used to connec t a televisio n receivin g antenn a to a home televisio n set. This line is essentiall y the sam e as the two-wir e open line excep t tha t unifor m spacin g is assure d by embeddin g the two wire s in a low-loss dielectric , usuall y polyethylene . Sinc e the wires are embedde d in the thin ribbo n of polyethylene , the dielectri c space is partl y ai r and partl y polyethylene. Twiste d Pair The TWISTED PAIR transmission line is illustrated in figure 3-11. As the name implies, the line consists of two insulate d wires twiste d togethe r to form a flexible line withou t the use of spacers . It is not used for transmitting high frequency because of the high dielectric losses that occur in the rubber insulation. When the line is wet, the losses increase greatly.

The SHIELDE D PAIR, shown in figur e 3-12, consist s of paralle l conductor s separate d from each othe r and surrounde d by a soli d dielectric . The conductor s are containe d withi n a braide d copper tubing that acts as an electrical shield. The assembly is covered with a rubbe r or flexibl e composition coatin g tha t protect s the line fro m moistur e and mechanical damage. Outwardly, it looks much like the power cord of a washin g machin e or refrigerator.

Figure 3-12.Shielded pair. The principa l advantag e of the shielded pair is that the conductors are balanced to ground; that is, the capacitance between the wires is uniform throughout This balanc e is due to the the lengt h of the line. unifor m spacin g of the grounde d shiel d tha t surrounds the wires along thei r entir e length . The braide d copper shield isolates the conductors from stray magnetic fields. Coaxial Lines Ther e are two types of COAXIAL LINES , RIGID (AIR) COAXIAL LINE and FLEXIBL E (SOLID) COAXIAL LINE. The physical construction of both types is basically the same; that is, each contains two concentric conductors.

Figure 3-10.Two-wire ribbon line.

3-7

The rigid coaxial line consists of a central , insulated wire (inne r conductor ) mounte d insid e a tubula r outer conductor. This line is shown in figure 3-13. In some applications, the inner conductor is also tubular. The inner conductor is insulated from the outer conductor by insulating spacers or beads at regular intervals. The spacer s are made of Pyrex , polystyrene , or some other material that has good insulating characteristics an d low dielectri c losses at high frequencies.

Flexibl e coaxia l lines (fig. 3-14) are mad e with an inne r conducto r tha t consist s of flexibl e wire insulate d from the oute r conducto r by a solid, continuous insulating material. The outer conductor is made of metal braid, which gives the line flexibility. Early attempts at gaining flexibility involved using rubbe r insulator s betwee n the two conductors. However , the rubbe r insulator s cause d excessiv e losses at high frequencies.

Figure 3-13.Air coaxial line.

Figure 3-14.Flexible coaxial line. Becaus e of the high-frequenc y losse s associated with rubbe r insulators , polyethylen e plasti c was developed to replace rubbe r and eliminat e these losses. Polyethylene plastic is a solid substance that remains flexible over a wid e rang e of temperatures . It is unaffecte d by seawater , gasoline , oil, and most other liquid s tha t may be found aboar d ship. The use of polyethylen e as an insulato r result s in greater high-frequenc y losses tha n the use of ai r as an insulator . However , thes e losses are stil l lower than the losses associate d with most othe r solid dielectric materials. This concludes our study of transmission lines. The rest of this chapte r will be an introductio n into the stud y of waveguides. WAVEGUID E THEORY The two-wire transmissio n line used in conventional circuits is inefficient for transferring electromagnetic energy at microwave frequencies . At these frequencies, energ y escape s by radiatio n becaus e the fields are not confine d in all directions , as illustrate d in figure 3-15. Coaxia l lines are mor e efficien t tha n two-wir e lines for transferring electromagnetic energy because the fields are completel y confine d by the conductors , as illustrate d in figure 3-16. Waveguide s are the most

The chief advantag e of the rigid line is its ability to minimize radiation losses. The electric and magnetic fields in a two-wir e paralle l line exten d into space for relativel y grea t distance s and radiatio n losses occur. However, in a coaxial line no electric or magnetic fields exten d outsid e of the oute r conductor . The fields are confine d to the spac e betwee n the two conductors, resultin g in a perfectl y shielde d coaxia l line. Another advantag e is tha t interferenc e from othe r lines is reduced. The rigid line has the followin g disadvantages: (1) it is expensiv e to construct ; (2) it mus t be kept dry to preven t excessiv e leakag e betwee n the two conductors ; and (3) althoug h high-frequenc y losses are somewha t less tha n in previousl y mentione d lines, they are still excessive enough to limit the practical lengt h of the line. Leakag e cause d by the condensatio n of moisture is prevente d in some rigid line application s by the use of an inert gas, such as nitrogen, helium, or argon. It is pumpe d into the dielectri c space of the line at a pressur e tha t can vary from 3 to 35 pound s per square inch. The inert gas is used to dry the line when it is first installe d and pressur e is maintaine d to ensure that no moisture enters the line.

3-8

efficien t way to transfe r electromagneti c energy. WAVEGUIDE S are essentiall y coaxia l line s without They ar e constructe d from center conductors . conductive material and may be rectangular, circular, or elliptica l in shape , as show n in figur e 3-17.

Figure 3-17.Waveguide shapes. of a coaxia l cable is large , but the surfac e are a of the inner conductor is relatively small. At microwave frequencies, the current-carrying area of the inner conductor is restricted to a very small layer at the surfac e of the conducto r by an actio n calle d SKIN EFFECT. Skin effect tends to increas e the effectiv e resistance of the conductor . Althoug h energ y transfe r in coaxial cable is cause d by electromagneti c field motion , the magnitud e of the field is limite d by the size of the current-carrying area of the inner conductor. The small size of the cente r conducto r is even furthe r reduced by skin effect, and energy transmission by coaxial cable become s less efficien t tha n by waveguides. DIELECTRI C LOSSE S are also lower in waveguides tha n in two-wir e and coaxia l transmissio n lines. Dielectri c losse s in two-wir e and coaxia l lines are cause d by the heatin g of the insulatio n betwee n the conductors. The insulation behaves as the dielectric of a capacito r forme d by th e two wire s of the transmissio n line. A voltag e potentia l acros s the two wire s cause s heatin g of the dielectri c and result s in a powe r loss . In practica l applications , the actual breakdown of the insulatio n betwee n the conductors of a transmissio n line is mor e frequentl y a problem than is the dielectric loss. This breakdown is usually caused by stationary voltag e spike s or nodes, whic h ar e cause d by standing waves. Standin g wave s ar e stationar y and occur when par t of the energ y travelin g down the line

Figure 3-15.Fields confined in two directions only.

Figure 3-16.Fields confined in all directions.

WAVEGUID E ADVANTAGES Waveguide s have severa l advantage s over two-wire and coaxial transmission lines. For example, the large surfac e are a of waveguide s greatl y reduce s COPPER (12R) LOSSES. Two-wire transmission lines have large coppe r losse s becaus e they have a relativel y small surfac e area . The surfac e are a of the oute r conductor

3-9

is reflecte d by an impedanc e mismatc h with the load. The voltag e potentia l of the standin g wave s at the point s of greates t magnitud e can becom e larg e enough to brea k down the insulatio n betwee n transmission line conductors. The dielectric much lower insulating in waveguides loss is air, which has a than conventional are als o

WAVEGUID E DISADVANTAGES Physica l size is the primar y lower-frequency limitatio n of waveguides . Th e widt h of a waveguide mus t be approximatel y a half wavelengt h at the frequenc y of the wave to be transported . For example, a waveguid e for use at 1 megahert z would be about 700 feet wide. This make s the use of waveguide s at frequencie s below impractical. 1000 megahert z increasingly range of any system The lower frequency

dielectric

materials.

However , waveguide s

subjec t to dielectric breakdow n cause d by standin g waves. Standing waves in waveguides cause arcing, which decrease s the efficienc y of energ y transfe r and can severel y damag e th e waveguide . since the electromagneti c fields Also are completel y

using waveguides is limited by the physical dimensions of the waveguides. Waveguide s are difficul t to instal l becaus e of their rigid, hollow-pip e shape . Specia l coupling s at the joint s are require d to assur e prope r operation . Also, the insid e surface s of waveguide s ar e often plate d with silve r or gold to reduc e ski n effect requirement s microwave increas e the costs frequencies. and losses . These decreas e the

containe d within the waveguide , radiatio n losse s are kep t ver y low. Power-handlin g capabilit y is anothe r advantage of waveguides . Waveguide s can handl e more power tha n coaxia l lines of th e sam e siz e becaus e power-handlin g capabilit y is directl y relate d to the distanc e betwee n conductors . Figur e 3-18 illustrates the greate r distanc e betwee n conductor s in a waveguide.

practicalit y of waveguid e system s at any othe r than

DEVELOPIN G THE WAVEGUIDE FROM PARALLE L LINES You may bette r understan d the transitio n from ordinar y transmissio n line concept s to waveguide theorie s by considerin g th e developmen t of a waveguide from a two-wire transmission line. Figure 3-19 show s a sectio n of a two-wir e transmissio n line supporte d on two insulators . At the junctio n wit h the line, the insulators must present a very high impedance A low to groun d for prope r operatio n of the line.

impedanc e insulato r woul d obviousl y short-circui t the line to ground , and thi s is wha t happen s at very high frequencies. Ordinary insulators display the characterincreases, the istic s of the dielectri c of a capacito r forme d by the wire and ground. Figure 3-18.Comparison of spacing in coaxial cable and a circular waveguide. In view of the advantage s of waveguides , you would thin k tha t waveguide s shoul d be the onl y type of transmissio n lines used . However , waveguide s have certain disadvantages only at microwave that make them practical for use frequencies. insulator As the frequency section overal l impedanc e decreases . A bette r high-frequency is a quarter-wave of transmission line shorte d at one end. Such an insulato r is shown in figur e 3-20. The impedanc e of a shorte d quartransmission line. This type of ter-wav e sectio n is very high at the open-en d junction with the two-wire insulator is known as a METALLIC INSULATOR

and may be place d anywher e along a two-wir e line.

3-10

Figure 3-19.Two-wire transmission line.

Figure 3-21.Metallic insulator on each side of a two-wire line.

Figure 3-20.Quarter-wave section of transmission line shorted at one end.

Note tha t quarter-wav e section s are insulator s at only one frequency. This severely limits the bandwidth, efficiency , and applicatio n of this typ e of two-wire line. Figure 3-21 shows several metallic insulators on each side of a two-wir e transmissio n line. As more insulators are added, each section makes contact with the next, and a rectangular waveguide is formed. The lines become par t of the walls of the waveguide , as illustrated in figur e 3-22. Th e energ y is then conducte d withi n the hollo w waveguid e instea d of along the two-wire transmission line. The compariso n of the way electromagneti c fields work on a transmission line and in a waveguide is not exact. During the change from a two-wire line to a waveguide , the electromagneti c field configurations also undergo changes, the many changes. As a resul t of these waveguide does not actuall y operate

Figure 3-22.Forming a waveguide by adding quarter-wave sections. like a two-wir e line tha t is completel y shunte d by quarter-wave sections. If it did, the use of a waveguide would be limite d to a single-frequenc y wave lengt h tha t was four time s the lengt h of the quarterwave sections. In fact, waves of this length cannot pass efficiently through waveguides. Only a small rang e of frequencie s of somewha t shorte r wavelength (highe r frequency ) can pass efficiently.

3-11

As show n in figur e 3-23, the wides t dimension of a waveguid e is called the a dimensio n and determine s the rang e of operatin g frequencies . The narrowes t dimensio n determine s the power-handling capabilit y of the waveguid e and is called the b dimension.

ENERG Y PROPAGATIO N IN WAVEGUIDES Since energ y is transferre d throug h waveguides by electromagnetic fields, you need a basic understanding of field theory . Bot h electri c (E FIELD ) and magneti c fields (H FIELD ) are presen t in waveguides, and the interactio n of thes e fields cause s energ y to travel through the waveguide. This action is best understoo d by firs t lookin g at the propertie s of the two individua l fields. E Field An electri c field exist s when a differenc e of potentia l cause s a stres s in the dielectri c betwee n two points . The simples t electri c field is one tha t forms betwee n the plate s of a capacito r when one plat e is made positiv e compare d to the other , as show n in view A of figur e 3-24. The stres s create d in the dielectric is an electri c field. Electric fields are represente d by arrows tha t point from the positiv e towar d the negativ e potential . The numbe r of arrow s shows the relativ e strengt h of the field. In view B, for example , evenl y space d arrows indicat e the field is evenl y distributed . For ease of explanation , the electri c field is abbreviate d E field, and the lines of stres s are called E lines. H Field The magneti c field in a waveguid e is mad e up of magneti c lines of force tha t are cause d by curren t flow throug h the conductiv e materia l of the waveguide. Magnetic lines of force, called H lines , ar e continuous closed loops, as show n in figur e 3-25. All of the H lines associate d with curren t are collectivel y called a magneti c field or H field . The strengt h of the H field, indicate d by the numbe r of H lines in a given area, varies directly with the amount of current. Although H lines encircle a single, straight wire, they behav e differentl y when the wire is forme d into a coil, as show n in figur e 3-26. In a coil the individual H lines tend to form aroun d each tur n of wire. Since

Figure 3-23.Labeling waveguide dimensions, NOTE:

This metho d of labelin g waveguide s is not standar d in all texts , Differen t method s ma y be used in other texts on microwave principles, but this method is in accordance with Navy Military Standards (MIL-STDS). In theory, a waveguide could function at an infinite numbe r of frequencie s highe r tha n th e designed frequency; however, in practice, an upper frequency limi t is cause d by modes of operation , which will be discussed later. If the frequenc y of a signa l is decrease d so much tha t two quarter-wavelength s are longe r tha n the wide dimensio n of a waveguide , energ y will no longe r pass through the waveguide. This is the lower frequency limit , o r CUTOF F FREQUENC Y o f a g ive n waveguide . In practica l applications , the wide dimensio n of a waveguid e is usuall y 0.7 wavelength at the operatin g frequency . This allows the waveguide to handl e a smal l rang e of frequencie s both above and below the operatin g frequency . The b dimension is governe d by the breakdow n potentia l of the dielectric, which is usually air. Dimensions ranging from 0.2 to 0.5 wavelength are common for the b sides of a waveguide.

3-12

Figure 3-24.Simple electric fields.

waveguid e is confine d to the physica l limit s of the guide. Two conditions , know n as BOUNDARY CONDITIONS , mus t be satisfie d for energ y to travel through a waveguide. The firs t boundar y conditio n (illustrate d in fig. 3-27, view A can be state d as follows: For an electric field to exist at the surface of a conductor, it m ust be perpendicular to the conductor.

Figure 3-25.Magneti c field on a singl e wire.

the H lines tak e opposit e direction s betwee n adjacent turns , the field betwee n the turn s is canceled . Inside and outsid e the coil, wher e the directio n of each H field is the same , the fields join and form continuous H lines around the entire coil. A similar action takes place in a waveguide.

Figure 3-27.E field boundary condition. The opposit e of thi s boundar y condition , shown in view B, is also true. An electric field CANNOT exis t paralle l to a perfec t conductor. The second boundar y condition , which is illustrated in figur e 3-28, can be state d as follows: For a varyin g magneti c field to exist , it must form close d loop s in paralle l w h the it conductors and be perpendicular to the electric field.

Figure 3-26.Magneti c field on a coil.

BOUNDAR Y CONDITION S IN A WAVEGUIDE The trave l of energ y down a waveguid e is similar, but not identical , to the trave l of electromagneti c waves in free space . The differenc e is tha t the energ y in a

3-13

intervals, as illustrated in figure 3-30. Angle is the directio n of trave l of the wave with respec t to some referenc e axis.

Figure 3-28.H field boundary condition.

Since an E field causes a current flow that in turn produce s an H field, both fields alway s exis t at the sam e time in a waveguide . If a syste m satisfie s one of thes e boundar y conditions , it mus t also satisf y the othe r since neithe r field can exis t alone. WAVEFRONT S WITHI N A WAVEGUIDE Electromagnetic energy transmitted into space consists of electric and magnetic fields that are at right angle s (90 degrees ) to each othe r and at righ t angles to the directio n of propagation . A simpl e analog y to establis h this relationshi p is by use of the right-hand rule for electromagneti c energy , base d on the POYNTIN G VECTOR . It indicate s tha t a screw (right-han d thread ) wit h its axis perpendicula r to the electri c and magneti c fields will advanc e in the directio n of propagatio n if the E field is rotate d to the righ t (towar d the H field). This rule is illustrated in figure 3-29.

Figure 3-30.Wavefronts in space. The reflectio n of a singl e wavefron t off the b wall of a waveguid e is show n in figur e 3-31 . The wavefront is shown in view A as small particles, In views B and C particl e 1 strike s the wall and is bounced back from the wall without losing velocity. If the wall is perfectl y flat , the angl e at which it the wall, know n as the angl e of incidenc e is the same as the angl e of reflectio n An instan t afte r particle 1 strikes the wall, particle 2 strikes the wall, as shown

Figure 3-29.The Poynting vector. The combine d electri c and magneti c field s form a wavefron t tha t can be represente d by alternate negativ e and positiv e peak s at half-wavelength

Figure 3-31.Reflection of a single wavefront.

3-14

in view C, and reflects in the same manner. Because all the particles are traveling at the same velocity, particle s 1 and 2 do not chang e thei r relativ e position with respec t to each other . Therefore , the reflected wave has the sam e shap e as the original . The remaining particles as shown in views D, E, and F reflect in the same manner. This process results in a reflecte d wavefron t identica l in shape , but opposite in polarity , to the inciden t wave. Figure 3-32, views A and B, each illustrate the directio n of propagatio n of two differen t electromagnetic wavefront s of differen t frequencie s being radiated into a waveguid e by a probe . Not e tha t only the directio n of propagatio n is indicate d by the lines and arrowheads . The wavefront s ar e at righ t angle s to the directio n of propagation . Th e angl e of incidence and the angl e of reflectio n of the wavefronts vary in size with the frequenc y of the inpu t energy, but the angle s of reflectio n are equa l to eac h other in a waveguide. The CUTOFF FREQUENCY in a waveguid e is a frequenc y tha t would caus e angle s of incidenc e and reflectio n to be perpendicula r to the walls of the guide . At any frequenc y below the cutoff frequency , the wavefront s will be reflecte d back and forth acros s the guide (settin g up standin g waves ) and no energ y will be conducte d down the waveguide.

The velocit y of propagatio n of a wave alon g a waveguid e is less tha n its velocit y throug h fre e space (spee d of light) . This lower velocit y is cause d by the zigzag pat h take n by the wavefront . The forward-progres s velocit y of the wavefron t in a waveguid e is called GROU P VELOCIT Y and is somewha t slowe r tha n the spee d of light. The group velocit y of energ y in a waveguid e is determined by the reflectio n angl e of the wavefronts off the b walls . The reflectio n angl e is determined by the frequenc y of the inpu t energy . This basic principl e is illustrate d in figur e 3-33. As frequency is decreased . the reflectio n angl e increases , causing the grou p velocit y to decrease . The opposit e is also true ; increasin g frequenc y increase s the group velocity.

Figure 3-33.Reflection angle at various frequencies.

WAVEGUID E MODE S OF OPERATION The waveguide analyzed in the previous paragraphs yields an electri c field configuratio n know n as the half-sine electric distribution. This configuration, called a MODE OF OPERATION, is shown in figure 3-34. Recal l tha t the strengt h of the field is indicated by the spacing of the lines; that is, the closer the lines, The region s of maximum the stronge r the field. voltag e in this field mov e continuousl y down the waveguid e in a sine-wav e pattern . To mee t boundary conditions . the fiel d mus t alway s be zero at the b walls.

Figure 3-32.Different frequencies in a waveguide.

3-15

Figure 3-34.Half-sine E field distribution. The half-sin e field is only one of man y field configurations, or modes, that can exist in a rectangular waveguide . A full-sin e field can als o exis t in a rectangular waveguide because, as shown in figure 3-35, the field is zero at the b walls.

Figure 3-36.Magnetic field caused by a half-sine E field.

Of the possibl e mode s of operatio n availabl e for a given waveguide , the dominan t mode has the lowest cutoff frequency . The high-frequenc y limi t of a rectangular waveguide is a frequency at which its a dimensio n become s larg e enoug h to allow operation in a mode highe r tha n tha t for which the waveguide has been designed. Circula r waveguide s are used in specific area s of radar and communications systems, such as rotating joints used at the mechanical point where the antennas rotate . Figur e 3-37 illustrate s the dominan t mode of a circular waveguide. The cutof f wavelengt h of a circula r guide is 1.71 time s th e diamete r of the waveguide . Since the a dimensio n of a rectangular waveguid e is approximatel y one half-wavelengt h at the cutof f frequency , the diamete r of an equivalent circular waveguide must be 2/1.71, or approximately

Figure 3-35.Full-sine E field distribution. The magnetic field in a rectangular waveguide is in the form of closed loops paralle l to the surfac e of the conductors. The strength of the magnetic field is proportiona l to the electri c field. Figur e 3-36 illustrate s the magneti c field patter n associate d with a half-sin e electri c fiel d distribution . The magnitude of the magneti c fiel d varie s in a sine-wav e pattern down the cente r of the waveguid e in time phase with the electri c field . TIME PHAS E mean s tha t the peak H lines and peak E lines occur at the same instant in time, although not necessarily at the same point along the lengt h of the waveguide. The dominan t mode is the most efficien t mode. Waveguides are normall y designe d so tha t only the dominan t mode will be used . To operat e in the dominan t mode, a waveguid e mus t have an a (wide) dimensio n of at leas t one half-wavelengt h of the frequenc y to be propagated . The a dimensio n of the waveguid e mus t be kep t nea r the minimum allowable value to ensure that only the dominant mode will exist. In practice, this dimension is usually 0.7 wavelength.

Figure 3-37.Dominant mode in a circular waveguide.

3-16

1.17 time s the a dimensio n of a rectangular waveguide. MODE NUMBERIN G SYSTEMS So far, only the most basic type s of E and H field arrangement s have been shown . Mor e complicated arrangement s are often necessar y to mak e possible coupling, isolation , or othe r type s of operation . The field arrangement s of the variou s modes of operation are divide d into two categories : TRANSVERSE ELECTRI C (TE) an d TRANSVERS E MAGNETIC (TM). In the transverse electric (TE) mode, the entire electric field is in the transverse plane, which is perpendicular to the waveguide, (direction of energy travel) . Par t of the m agneti c fiel d is paralle l to the length axis. In the transverse m agnetic (TM) mode, the entire m agnetic field is in the transverse plane and has no portio n paralle l to the lengt h axis. Since ther e are severa l TE and TM modes, subscript s are used to complet e the descriptio n of the field pattern. In rectangular waveguides, the first subscrip t indicate s the numbe r of half-wav e patterns in the a dimension, and the second subscript indicates the numbe r of half-wav e pattern s in the b dimension. The dominan t mode for rectangula r waveguides is shown in figur e 3-38. It is designate d as the TE mode becaus e the E field s are perpendicula r to the a walls. The first subscript is 1, since there is only one half-wav e patter n acros s the a dimension . There

are no E-field pattern s acros s the b dimension , so The complet e mode the second subscrip t is 0. descriptio n of the dominan t mode in rectangular Subsequen t descriptio n of waveguide s is TE 1,0 . waveguid e operatio n in this tex t will assum e the dominan t (TE 1,0) mode unles s otherwis e noted. A simila r syste m is used to identif y the mode s of circular waveguides. The general classification of TE and TM is true for both circular and rectangular waveguides. In circular waveguides the subscripts have a different meaning. The first subscript indicates the numbe r of fill-wav e pattern s aroun d the circumference of the waveguide . The second subscrip t indicates the numbe r of half-wav e pattern s acros s the diameter. In the circular waveguide in figure 3-39, the E field is perpendicula r to the lengt h of the waveguide with no E lines paralle l to the directio n of propagation. Thus, it must be classified as operating in the TE mode. If you follow the E lin e patter n in a counterclockwis e directio n startin g at the top, the E lines go from zero, throug h maximu m positiv e (tail of arrows), back to zero, through maximum negative (head of arrows) , and then back to zero again . This is one full wave, so the first subscript is 1. Along the diameter , the E lines go from zero through maximum and back to zero, makin g a half-wave variation. The second subscript, therefore, is also 1. TE1,1 is the complet e mod e descriptio n of the dominant mode in circula r waveguides . Severa l mode s are possible in both circular and rectangular waveguides. Figur e 3-40 illustrate s severa l differen t modes that can be used to verify the mode numbering system.

Figure 3-38.Dominant mode in a rectangular waveguide.

Figure 3-39.Counting wavelengths in a circular waveguide.

3-17

Figure 3-40.Various modes of operation for rectangular and circular waveguides.

WAVEGUID E INPUT/OUTPUT METHOD S A waveguide , as explaine d earlie r in thi s topic, operate s differentl y from an ordinar y transmissio n line. Therefore , specia l device s mus t be used to put energy into a waveguid e at one end and remov e it from the othe r end. The thre e device s used to injecto r remov e energy from waveguides are PROBES, LOOPS, and SLOTS. Slots may also be called APERTURES or WINDOWS. When a smal l probe is inserte d into a waveguide and supplied with microwave energy, it acts as a quarter-wav e antenna . Curren t flows in the probe and sets up an E field such as the one shown in figure 3-41, view A. The E lines detac h themselve s from the probe . When the probe is locate d at the poin t of highest efficiency , the E lines set up an E fiel d of considerable intensity. The most efficien t place to locate the probe is in the cente r of the a wall, paralle l to the b wall , and one quarter-wavelengt h from the shorte d end of the waveguide , as shown in figur e 3-41, view s B and C. This is the point at which the E field is maximum in the dominant mode. Therefore, energy transfer (coupling) is maximum at this point. Note that the quarter-wavelength spacing is at the frequency required to propagat e the dominan t mode.

In man y application s a lesse r degre e of energy transfer , called loose coupling , is desirable . The amount of energy transfer can be reduced by decreasing the lengt h of the probe , by movin g it out of the center of the E field, or by shieldin g it. Wher e the degree of couplin g mus t be varie d frequently , the probe is made retractabl e so the lengt h can be easily changed. The size and shap e of the probe determine s its frequency , bandwidth , and power-handlin g capability. As the diamete r of a probe increases , the bandwidth increases . A probe simila r in shap e to a door knob is capabl e of handlin g much highe r power and a larger bandwidt h tha n a conventiona l probe . The greater power-handlin g capabilit y is directl y relate d to the T w o example s o f broad-bandwidt h probe s are illustrate d in figur e 3-41, view D. Remova l of energ y from a waveguid e is simpl y a reversa l of the injectio n proces s using the same type of probe. Anothe r way of injectin g energ y into a waveguide is by settin g up an H field in the waveguide . This can be accomplishe d by insertin g a smal l loop that carries a high current into the waveguide, as shown in figur e 3-42, view A. A magneti c field build s up aroun d the loop and expand s to fit the waveguide , as shown in view B. If the frequenc y of the curren t in the loop is withi n the bandwidt h of the waveguide, energ y will be transferre d to the waveguide. increase d surfac e area .

3-18

Figure 3-41.Probe coupling in a rectangular waveguide.

For the most efficien t the loop is inserte d at one magneti c field will be of those point s are shown in

couplin g to the waveguide, of severa l point s wher e the greates t strength . Four of figur e 3-42, view C.

When less efficien t couplin g is desired , you can rotate or move the loop unti l it encircle s a smaller numbe r of H lines . When the diamete r of the loop is increased , its power-handlin g capabilit y also increases. The bandwidt h can be increase d by increasin g the size of the wire used to mak e the loop. When a loop is introduce d into a waveguid e in which an H field is present, a current is induced in the loop . When this conditio n exists , energ y is remove d from the waveguide. Slots or aperture s are sometime s used whe n very loose (inefficient ) couplin g is desired , as shown in figure 3-43. In this method energy enters through a smal l slot in the waveguid e and the E field expands into the waveguide . The E line s expan d first across the slot and then acros s the interio r of the waveguide.

Figure 3-42.Loop couplin g in a rectangular waveguide.

3-19

WAVEGUID E IMPEDANCE MATCHING Waveguide transmission systems are not always perfectl y impedanc e matche d to thei r loa d devices. The standing waves that result from a mismatch cause a power loss, a reductio n in power-handlin g capability, and an increas e in frequenc y sensitivity . Impedance-changin g device s are therefor e place d in the waveguid e to matc h the waveguid e to the load. These device s ar e place d nea r the sourc e of the standing waves. Figure 3-43.Slot coupling in a waveguide. Minimum reflections occur when energy is injected or remove d if the size of the slot is properl y proportione d to the frequenc y of the energy. After learnin g how energ y is couple d into an d out of a waveguide with slots, you might think that leaving the end open is the most simpl e way of injectin g or removing energy in a waveguide. This is not the case, however , becaus e when energ y leave s a waveguide, fields form aroun d the end of the waveguide . These fields cause an impedance mismatch which, in turn, causes the development of standing waves and a drastic Variou s method s of impedance loss in efficiency. matching and terminating waveguides will be covered in the next section. Figure 3-44 illustrates three devices, called irises, tha t are used to introduc e inductanc e or capacitance into a waveguide. An iris is nothing more than a metal plate that contains an opening through which the waves may pass. The iris is located in the transverse plane of eithe r the magneti c or electri c field. An inductive iris and its equivalent circuit are illustrated in figure 3-44, view A. The iris places a shunt inductive reactance across the waveguide that is directl y proportiona l to the size of the opening. Notice that the inductive iris is in the magnetic plane. The shunt capacitive reactance, illustrated in view B, basically acts the same way. Again, the reactance is directl y proportiona l to the size of the opening , but the iri s is place d in the electri c plane . The iris, illustrated in view C, has portions in both the magnetic

Figure 3-44.Waveguide irises.

3-20

and electri c transvers e plane s and forms an equivalent parallel-L C circui t acros s th e waveguide . At the resonant frequency, the iris acts as a high shunt resistance . Above or belo w resonance , the iris acts as a capacitive or inductive reactance. POST S and SCREW S mad e fro m conductive materia l can be used for impedance-changin g devices in waveguides . View s A and B of figur e 3-45, illustrat e two basi c method s of usin g posts and screws. A post or scre w tha t only partiall y penetrate s into the waveguide acts as a shunt capacitive reactance. When the pos t or screw extend s completel y throug h the waveguide, making contact with the top and bottom walls, it acts as an inductive reactance. Note that when screws are used, the amount of reactance can be varied. Figure 3-46.Waveguide horns. As you may have antennas . They have impedance-matchin g bandwidth and simple noticed , horn s are reall y simple severa l advantage s over other devices , such as thei r large construction.

Figure 3-45.Conducting posts and screws. WAVEGUID E TERMINATIONS

A waveguid e may also be terminate d in a resistive load tha t is matche d to the characteristi c impedance of the waveguide . The resistiv e load is most often called a DUMMY LOAD, becaus e its only purpose is to absor b all the energ y in a waveguid e without causing standing waves. Ther e is no place on a waveguid e to connec t a fixed terminatio n resistor ; therefore , severa l special arrangement s are used to terminat e waveguides . One metho d is to fill the end of the waveguid e with a graphite and sand mixture, as illustrated in figure 3-47, view A. When the field s ente r the mixture , they induce a current flow in the mixture that dissipates the energ y as heat . Anothe r metho d (view B) is to use a high-resistanc e rod place d at the cente r of the E field. The E field cause s curren t to flow in the rod, and the high resistanc e of the rod dissipate s the energy as a powe r loss, agai n in the form of heat. Still anothe r metho d for terminatin g a waveguide is the use of a wedge of highl y resistiv e material , as show n in view C of figur e 3-47. Th e plan e of the wedge is place d perpendicula r to the magneti c lines

Electromagneti c energ y is often passe d through a waveguid e to transfe r the energ y from a sourc e into space. As previousl y mentioned , the impedanc e of a waveguid e does not matc h the impedanc e of space, and without proper impedance matching standing waves caus e a larg e decreas e in the efficienc y of the waveguide. Any abrupt change in impedance causes standing waves , but when the chang e in impedanc e at the end of a waveguid e is gradual , almos t no standin g waves are formed . Gradua l change s in impedanc e can be obtaine d by terminatin g th e waveguid e with a funnel-shaped HORN, such as the three types illustrated in figur e 3-46. The typ e of horn used depend s upon the frequenc y and the desire d radiatio n pattern.

3-21

to carry liquids or other substances. The design of a waveguid e is determine d by the frequenc y an d power level of the electromagnetic energy it will carry. The followin g paragraph s explai n th e physica l factors involve d in the desig n of waveguides. W aveguide Bends

The size, shape , and dielectri c materia l of a waveguide must be constant throughout its length for energ y to move from one end to the othe r without reflections . Any abrup t chang e in its size or shape can caus e reflection s and a los s in overal l efficiency. When such a change is necessary, the bends, twists, and joint s of th e waveguide s mus t mee t certain condition s to preven t reflections. Waveguide s mayb e ben t in severa l ways tha t do not caus e reflections . One way is the gradua l bend shown in figure 3-48. This gradual bend is known as an E bend becaus e it distort s the E fields . The E bend mus t have a radiu s greate r tha n two wavelengths to preven t reflections.

Figure 3-47.Terminating waveguides. of force. When the H lines cut through the wedge, curren t flows in the wedge and cause s a powe r loss. As with the other methods, this loss is in the form of heat . Since very littl e energ y reache s the end of the waveguide, reflections are minimum. All of the termination s discusse d so fa r are designe d to radiat e or absor b the energ y without reflections. In many instances, however, all of the energ y mus t be reflecte d from th e end of the waveguide . The best way to accomplis h this is to permanentl y weld a meta l plat e at the end of the waveguide , as show n in view D of figur e 3-47. WAVEGUID E PLUMBING Since waveguide s are reall y only hollo w metal pipes , the installatio n and the physica l handlin g of waveguide s have man y similaritie s to ordinary plumbing. In light of this fact, the bending, twisting, joining, and installation of waveguides is commonly called waveguide plumbing. Naturally, waveguides are differen t in desig n from pipes tha t ar e designed

Figure 3-48.Gradual E bend. Anothe r commo n bend is the gradua l H bend (fig. 3-49). It is called an H bend becaus e the H fields are distorted when a waveguide is bent in this manner. Again , the radiu s of the bend mus t be greate r than two wavelength s to preven t reflections . Neithe r the E bend in the a dimensio n nor the H bend in the b dimensio n change s the norma l mode of operation.

Figure 3-49.Gradual H bend.

3-22

A shar p bend in eithe r dimensio n may be used if it meet s certai n requirements . Notic e th e two 45-degre e bend s in figur e 3-50; the bend s ar e 1/4 apart . The reflection s that occur at the 45-degre e bends cancel each other , leavin g the fields as thoug h no reflections have occurred.

Figure 3-52.Flexible waveguide. constructe d in sections and the sections connecte d with joints . The thre e basic type s of waveguid e joint s are the PERMANENT, the SEMIPERMANENT, and the ROTATING JOINTS. Since the permanent joint is a factory-welde d join t tha t require s no maintenance, only the semipermanen t and rotatin g joint s will be discussed. Section s of waveguid e mus t be take n apar t for maintenance and repair. A semipermanent joint, called a CHOKE JOINT, is most commonly used for this purpose . The choke joint provide s good electromagnetic continuit y betwee n the section s of the waveguide with very littl e power loss. A cross-sectiona l view of a choke joint is shown in figur e 3-53. The pressur e gaske t show n between the two metal surfaces forms an airtight seal. Notice in view B tha t the slot is exactl y from the a wall of the waveguide . The slot is also deep, as shown in view A, and becaus e it is shorte d at point 1, a high impedance results at point 2. Point 3 is from poin t 2. The high impedanc e at poin t 2 results in a low impedance, or short, at point 3. This effect create s a good electrica l connectio n betwee n the two section s tha t permit s energ y to pass with very little reflectio n or loss. Whenever a stationary rectangular waveguide is to be connecte d to a rotatin g antenna , a rotatin g joint must be used. A circular waveguide is normally used in a rotating joint. Rotating a rectangular waveguide would caus e field patter n distortion . The rotating sectio n of the joint , illustrate d in figur e 3-54, uses a choke joint to complet e the electrica l connectio n with The circular waveguide is the stationary section. designe d so tha t it will operat e in the TM0,1 mode.

Figure 3-50.Sharp bends. Sometime s the electromagneti c fields mus t be rotated so tha t they are in the prope r phas e to match the phas e of the load. This may be accomplishe d by twistin g the waveguid e as show n in figur e 3-51. The twist must be gradual and greater than

Figure 3-51.Waveguide twist.

The flexible waveguid e (fig. 3-52) allow s special bends , which some equipmen t application s might require . It consist s of a speciall y woun d ribbo n of conductive material, the most commonly used is brass, with the inner surface plated with chromium. Power losses ar e greate r in the flexible waveguid e because the inne r surface s are not perfectl y smooth . Therefore, it is only used in shor t section s wher e no other reasonabl e solutio n is available. W aveguide Joints

Since an entir e waveguid e syste m canno t possibly be molde d into one piece, the waveguid e mus t be

3-23

The rectangular sections are attached as shown in the illustration to prevent the circular waveguide from operatin g in the wrong mode. Distanc e O is so tha t a high impedanc e will be presente d to any unwanted modes. This is the most common design used for rotatin g joints , but othe r type s ma y be used in specific applications. WAVEGUID E MAINTENANCE

The installatio n of a waveguid e syste m presents problem s tha t are not normall y encountere d when dealin g with othe r type s of transmissio n lines . These problem s often fal l withi n the technician s are a of responsibility. A brie f discussio n of waveguide handling , installation , and maintenanc e will help prepar e you for this maintenanc e responsibility, Detaile d informatio n concernin g waveguid e maintenance in a particular system may be found in the technical manuals for the system. Since a waveguid e naturall y has a low los s ratio, most losses in a waveguid e syste m are cause d by other factors . Improperl y connecte d joint s or damage d inner surface s can decreas e the efficienc y of a syste m to the point that it will not work at all. Therefore, you must take great care when working with waveguides to preven t physica l damage . Since waveguide s are made from a soft, conductive material, such as copper or aluminum, they are very easy to dent or deform. Even the slightes t damag e to the inne r surfac e of a waveguid e will caus e standin g waves and , often, internal arcing. Internal arcing causes further damage to the waveguid e in an actio n tha t is often self-sustaining until the waveguide is damaged beyond use. Part of your job as a technician will be to inspect the waveguid e syste m for physica l damage . The previousl y mentione d dent s are only one type of physica l damag e tha t can decreas e the efficienc y of the system . Anothe r proble m occurs because waveguides are made from a conductive material such as copper while the structure s of most ship s are made from steel . Whe n two dissimila r metals , such as copper and steel , are in direc t contact , an electrical action calle d ELECTROLYSI S take s place tha t causes ver y rapi d corrosio n of the metals . Waveguide s can be completel y destroye d by electrolyti c corrosio n in a relativel y shor t perio d of time if they are not isolated from direct contact with other metals. Any inspection

Figure 3-53.Choke joint.

Figure 3-54.Rotating joint.

3-24

of a waveguid e syste m shoul d includ e a detailed inspectio n of all suppor t point s to ensur e tha t electrolytic corrosion is not taking place. Any waveguide tha t is expose d to the weathe r shoul d be painte d and all joint s sealed . Proper painting prevents natural corrosion , and sealing the joints prevent s moistur e from enterin g the waveguide. Moistur e can be one of the wors t enemie s of a waveguid e system . As previousl y discussed , the dielectric in waveguides is air, which is an excellent dielectric as long as it is free of moisture. Wet air, however , is a very poor dielectri c and can caus e serious internal arcing in a waveguide system. For this reason, care is take n to ensur e tha t waveguid e system s are pressurized with air that is dry. Checking the pressure and moistur e conten t of the waveguid e air may be one of your daily system maintenance duties. More detaile d waveguid e installatio n an d maintenanc e informatio n can be found in th e technical manuals that apply to your particular system. Another good sourc e is the Electronic s Installatio n and Maintenanc e Handbook s (EIMB ) publishe d by Naval Sea Systems Command. Installation Standards (EIMB) Handbook , NAVSE A 0967-LP-000-0110 , is the volume that deals with waveguide installation and maintenance. WAVEGUID E DEVICES The discussio n of waveguides , up to thi s point, has been concerne d only with the transfe r of energy from one poin t to another . Man y waveguid e devices have been developed , however , tha t modify the energy in some fashion during the transmission. Some devices do nothin g more tha n chang e the directio n of the energy . Other s hav e been designe d to chang e the basic characteristic s or powe r level of the electromagnetic energy. This sectio n will explai n the basi c operating principle s of some of the mor e commo n waveguide devices , such as DIRECTIONA L COUPLERS , CAVITY RESONATORS, and HYBRID JUNCTIONS. Directional Couplers

for measuremen t or use in anothe r circuit . Most coupler s sampl e energ y travelin g in one directio n only. However , directiona l coupler s can be constructe d that sampl e energ y in both directions . Thes e are called BIDIRECTIONA L coupler s an d are widely used in radar and communications systems. Directional couplers may be constructed in many ways. The couple r illustrate d in figur e 3-5 5 is constructe d from an enclose d waveguid e sectio n of the sam e dimension s as the waveguid e in which the energ y is to be sampled . The b wall of this enclosed sectio n is mounte d to the b wall of the waveguide from which the sample will be taken. There are two holes in the b wal l betwee n the section s of the coupler . Thes e two holes are apart . The upper sectio n of th e directiona l couple r has a wedg e -o f energy-absorbin g materia l at one end and a pickup probe connecte d to an outpu t jack at the othe r end. The absorbent material absorbs the energy not directed at the probe and a portio n of the overal l energ y that enter s the section.

Figure 3-55.Directional coupler. Figur e 3-56 illustrate s two portion s of the incident wavefron t in a waveguide . The wave s trave l down the waveguid e in the directio n indicate d and ente r the couple r sectio n throug h both holes . Sinc e both portions of the wave trave l the sam e distance , they are in phase when they arriv e at the picku p probe . Becaus e the wave s are in phase , the y add togethe r and provid e a sampl e of the energ y travelin g down the waveguide. The sampl e take n is only a smal l portio n of the energy tha t is travelin g down the waveguide . The magnitude of th e sample , however , is proportiona l to the magnitud e of the energ y in the waveguide . The absorben t materia l is designe d to ensur e tha t the ratio

The directiona l couple r is a device tha t provides a metho d of samplin g energ y from withi n a waveguide

3-25

betwee n the sampl e energ y and the energ y in the waveguid e is constant . Otherwise , the sampl e would contain no useful information. The ratio is usually stamped on the couple r in the form of an attenuation factor.

and the probe are in opposit e position s from the directiona l couple r designe d to sampl e the incident energy . This positionin g cause s the two portion s of the reflecte d energ y to arriv e at the prob e in phase, providin g a sampl e of the reflecte d energy . The transmitte d energ y is absorbe d by the absorbent material.

Figure 3-56.Incident wave in a directional coupler designed to sample incident waves.

Figure 3-58.Directional coupler designed to sample retlected energy. A simpl e bidirectiona l couple r for samplin g both transmitte d and reflecte d energ y can be constructed by mountin g two directiona l coupler s on opposit e sides of a waveguide , as show n in figur e 3-59.

The effect of a directiona l couple r on an y reflected energy is illustrate d in figur e 3-57. Not e tha t these two wave s do not trave l the sam e distanc e to the picku p probe . The wav e represente d by the dotted line travel s furthe r an d arrive s at the probe 180 degree s out of phas e wit h the wave , represente d by the solid line. Becaus e the wave s are 180 degrees out of phas e at the probe , they cance l each othe r and no energ y is induce d into the picku p probe . When the reflecte d energ y arrive s at the absorben t material, it add s an d is absorbe d by the material.

Figure 3-59.Bidirectional coupler. Cavit y Resonators

Figure 3-57.Reflected wave in a directional coupler.

By definition , a resonan t cavit y is any space A directiona l couple r designe d to sampl e reflected energy is shown in figure 3-58. The absorbent material completel y enclose d by conductin g - walls that can contain oscillatin g electromagneti c fields an d possess resonant properties. The cavity has many advantages

3-26

and uses at microwav e frequencies . Resonan t cavities have a very high Q and can be buil t to handle relativel y large amount s of power . Cavitie s with a Q value in excess of 30,000 are not uncommon. The high Q gives thes e device s a narro w bandpas s and allows very accurate tuning. Simple, rugged construction is an additiona l advantage. Althoug h cavity resonators , buil t for different frequenc y range s and applications , hav e a variet y of shapes , the basic principle s of operatio n are the same for all. One example of a cavity resonator is the rectangular box shown in figure 3-60, view A. It may be thought of as a sectio n of rectangula r waveguid e closed at both ends by conducting plates. The frequency at which the resonan t mod e occurs is of the distance betwee n the end plates . The magneti c field patterns in the rectangular cavity are shown in view B. There are two variable s that determin e the primary frequency of any resonant cavity. The first variable is PHYSICA L SIZE. In general , the smalle r the cavity, the higher its resonant frequency. The second controlling factor is the SHAPE of the cavity. Figure 3-61 illustrates several cavity shapes that are commonly used . Remembe r from the previousl y state d definition of a resonan t cavity tha t any completel y enclosed conductive surface, regardless of its shape, can act as a cavity resonator. Energ y can be inserte d or remove d from a cavity by the same method s tha t are used to coupl e energy into and out of waveguides . The operatin g principles of probes , loops, and slots are the sam e whethe r used in a cavity or a waveguide . Therefore , any of the three method s can be used with cavitie s to injec t or remove energy. The resonan t frequenc y of a cavity can be varied by changin g any of the thre e parameters : cavity volume , cavity capacitance , or cavity inductance. Changing the frequencies of a cavity is known as TUNING. The mechanical methods of tuning a cavity may vary with the application, but all methods use the same electrical principles.

Figure 3-60.Rectangular waveguide cavity resonator. W aveguide Junctions

You may have assumed that when energy traveling down a waveguid e reache s a junctio n it simpl y divides and follows the junction. This is not strictly true.

3-27

Figure 3-61.Types of cavities.

Differen t type s of junction s affect the energ y in differen t ways. Since waveguid e junction s ar e used extensivel y in most systems , you need to understand the basic operatin g principle s of thos e most commonly used. The T JUNCTIO N is the most simpl e of the commonly used waveguide junctions. T junctions are

divide d into two basic types , the E TYPE and the H TYPE. HYBRID JUNCTIONS are more complicated development s of the basic T junctions . The MAGIC-T and the HYBRID RING are the two most commonly used hybrid junctions. E-TYPE T JUNCTION. An E-typ e T junction is illustrated in figure 3-62, view A.

Figure 3-62.E fields in an E-type T junction.

3-28

It is called an E-typ e T junctio n becaus e the junction arm extend s from the main waveguid e in the same directio n as the E field in the waveguide. Figure 3-62, view B, illustrates cross-sectional views of the E-typ e T junctio n wit h input s fed into the various arms. For simplicity, the magnetic lines tha t are always presen t with an electric field have been omitted . In view K, the inpu t is fed into arm b and the outputs are taken from the a and c arms. When the E field arrive s betwee n point s 1 and 2, poin t 1 become s positiv e and poin t 2 become s negative . The positiv e charg e at poin t 1 then induce s a negative charg e on the wall at poin t 3. The negativ e charge at poin t 2 induce s a positiv e charg e at poin t 4. These charge s caus e the fields to form 180 degree s out of phas e in the main waveguide ; therefore , the outputs wil l be 180 degree s out of phas e with each other. In view L, two in-phas e input s of equa l amplitud e are fed into the a and c arms . The signal s at point s 1 and

2 have the sam e phas e and amplitude . No difference of potentia l exist s acros s the entranc e to the b arm, and no energ y will be couple d out . However , when the two signal s fed into the a and c arm s are 180 degree s out of phase , as show n in view M, points 1 and 2 have a differenc e of potential . Thi s difference of potentia l induce s an E field from poin t 1 to point 2 in the b arm, and energy is coupled out of this arm. Views N and P illustrat e two method s of obtaining two output s with only one input.

H-TYPE T JUNCTION. An H-type T junction is illustrate d in figur e 3-63, view A. It is called an H-typ e T junctio n becaus e the long axi s of the b arm is paralle l to the plan e of the magneti c lines of force in the waveguide. Again, for simplicity, only the E lines are shown in this figure. Each X indicates an E line movin g awa y from the observer . Each dot indicate s an E lin e movin g towar d the observer.

Figure 3-63.E field in an H-type T junction.

3-29

In view 1 of figur e 3-63, view B, the signa l is fed into arm b and in-phas e output s are obtaine d from the a and c arms . In view 2, in-phas e signal s are fed into arm s a and c and the outpu t signa l is obtained from the b arm becaus e the fields add at the junction and induc e E lines into the b arm . If 180-degree-out-of-phas e signal s are fed into arm s a and c, as show n in view 3, no outpu t is obtaine d from the b arm becaus e the opposin g fields cance l at the junction. If a signal is fed into the a arm, as shown in view 4 , output s will be obtaine d fro m the b and c arms. The reverse is also true. If a signal is fed into the c arm , output s will be obtaine d from the a and b arms. MAGIC-T HYBRI D JUNCTION. A simplified version of the magic-T hybrid junction is shown in figure 3-64. The magic-T is a combination of the H-type and E-type T junctions. The most common applicatio n of thi s type of junctio n is as the mixer section for microwave radar receivers.

Figure 3-65.Magic-T with input to arm b. In summary, when an input is applied to arm b of the magic-T hybrid junction, the output signals from arm s a and c are 180 degree s out of phas e wit h each other, and no output occurs at the d arm. The action that occurs when a signal is fed into the d arm of the magic-T is illustrated in figure 3-66. As with the H-type T junction, the signal entering the d arm divide s and moves dow n the a and c arm s as output s tha t are in phas e wit h each othe r and with the input . The shap e of the E fields in motio n is shown by the numbere d curve d slices . As the E field moves down the d arm , point s 2 and 3 are at an equal potential . The energ y divide s equall y into arm s a and c, and the E fields in both arm s become identica l in shape . Sinc e the potential s on both sides of the b arm are equal, no potentia l difference exists at the entrance to the b arm, resulting in no output. 3-30

Figure 3-64.Magic-T hybrid junction. If a signa l is fed into the b arm of the magic-T, it will divide into two out-of-phas e components . As show n in figur e 3-65, view A, thes e two components will move into the a and c arms. The signal entering the b arm will not ente r the d arm becaus e of the zero potentia l existin g at the entranc e of the d arm . The potentia l mus t be zero at this poin t to satisf y the boundar y condition s of the b arm . This absenc e of potentia l is illustrate d in views B and C wher e the magnitude of the E field in the b arm is indicate d by the lengt h of the arrows . Since the E line s ar e at maximum in the center of the b arm and minimum at the edge wher e the d arm entranc e is located , no potentia l differenc e exist s acros s the mout h of the d arm.

destroy the shape of the junctions. One method is shown in figure 3-68. A post is used to match the H plane, and an iris is used to match the E plane. Even though this method reduces reflections, it lowers the power-handling capability even further.

Figure 3-66.Magic-T w input ith

to arm d. Figure 3-68.Magic-T im pedance m atching.

When an input signal is fed into the a arm as shown in figure 3-67, a portio n of the energ y is coupled into the b arm as it would be in an E-type T junction. An equal portion of the signal is coupled through the d arm because of the action of the H-type junction. The c arm has two fields across it that are out of phase with each other. Therefore, the fields cancel, resulting in no output at the c arm. The reverse of this action takes place if a signal is fed into the c arm, resulting in outputs at the b and d arms and no output at the a arm.

HYBRID RING. A type of hybrid junction that overcomes the power limitation of the magicT is the hybri d ring, also called a RAT RACE. The hybrid ring, illustrated in figure 3-69, view A, is actually a modification of the magic-T. It is constructed of rectangular waveguides molded into a circular pattern. The arms are joined to the circular waveguide to form E-type T junctions. View B shows, in wavelengths, the dimensions required for a hybrid ring to operate properly. The hybri d ring is used primaril y in highpowered radar and communications systems to perfor m two functions . Durin g the transmit period , the hybri d ring couple s microwav e energy from the transmitter to the antenna and allows no energ y to reac h the receiver . Durin g the receive cycle, the hybri d ring couple s energ y from the antenna to the receive r and allows no energ y to reach the transmitter. Any device that performs both of these functions is called a DUPLEXER. A duplexer permits a system to use the same antenna for both transmitting and receiving. SUMMARY This concludes our discussion on transmission lines and waveguides. In this volume you have been given a basic introduction on wave propagation from the time it leaves the transmitter to the point of reception. In volume 8 you will be introduced to a variety of electronic support systems.

Figure 3-67.Magic-T w input ith

to arm a.

Unfortunately, when a signal is applied to any arm of a magic-T , the flow of energ y in the output arms is affected by reflections. Reflections are caused by impedance mismatching at the junctions. These reflections are the cause of the two major disadvantages of the magic-T. First, the reflections represent a power loss since all the energ y fed into the junctio n does not reac h the load that the arms feed. Second, the reflections produce standing waves that can result in internal arcing. Thus, the maximum power a magic-T can handle is greatly reduced. Reflections can be reduced by using some means of impedance matching that does not 3-31

Figure 3-69.Hybrid ring with wavelength measurements.

3-32