Rf Communications

8/14/2019 Rf Communications

1/130

Base Station Antennas

Combiners

Duplexers

Lightning

RF Transmission Lines

R F

C O M

M U N I C A

T I O N S

A Non-Technical Approach

Presented by Decibel Products

A b o u t :

Selective Cavities


2/130

ARF Communicat io

A Non -Techn ical A

Presented by Decib el P


3/130

First published as a series of individual booklets by

Decibel Products 8635 Stemmons Freeway, Dallas, Texas 75247-3701P. O. Box 569610, Dallas, Texas 75356-9610 Tel: (214) 631-0310 Fax: (214) 631-4706

Editing and Design: Steven Patterson Graphic Illustration: Carla Brown, Steve Claxton Marketing Communications Manager: Devra D. Fuller Print ing: Ussery Printing Company, Irving, Texas

All rights reserved.No part of this book may be reproduced in any way without the expressed written permission of the publishe

Printed in the U. S. A.Copyright 1964, 1965, 1970, 1973, 1975, 1996 by Decibel Products, a division of the Allen Telecom Grou

Any trademarked names appearing in this book are the property of their respective holders. They are used only foeditorial purposes and the publisher wishes no infringemupon them.

Limits of Liability and Disclaimer of Warranty The publisher has used its best efforts in preparing this boand the information it contains. The publisher makes no warranties of any kind, expressed or implied, with regardthe instructions and suggestions contained in this book. Tpublisher shall not be liable in the event of incidental or consequential damages in connection with, or arising ou


4/130

t ocInt rod uctio n ...........................................................

CHAPTER 1Base Stat ion Anten nas ..............................................What is an antenna? ........................................................................................Matching and VSWR......................................................................................Wavelength, Frequency, and Velocity ...............................................................Antennas Work the Same

When Transmitt ing or Receiving ..............................................................The Half-Wave Dipole .....................................................................................Antenna Polarity.............................................................................................Vertical and Horizontal Antenna Radiation Patterns........................................No True Omnidirectional Antennas...............................................................Antenna Gain .................................................................................................What About That Word Decibel? ...................................................................How Do We Get Antenna Gain?....................................................................Squashing the Doughnut ................................................................................

Omnidirectional Pattern Gain Antenna..........................................................In Phase and Out of Phase ............................................................................. 1Aperture ........................................................................................................Spacing of Dipole Elements

in the Vertical Collinear Array .................................................................Feeding the Collinear Array ............................................................................Maximum Gain from aCollinearArray ..........................................................

Table of Con te


5/130

About RF COMMUNICATIONS ii

Phased Radiators for Directivity .......................................................................A Bidirectional Array or Figure-Eight Pattern .................................................Additional Gain by Stacking Directional Antennas ........................................Side Mounted Antennas ................................................................................Then How Do We Get an Omni Pattern? .......................................................Noise .............................................................................................................

What Do We Mean by Noise? ........................................................................What Happens to Noise with Gain?...............................................................What About the 450 MHz Band?..................................................................What About the Countryside? .......................................................................Bandwidth .....................................................................................................

CHAPTER 2 Comb iners ...............................................................Combiner Requirements ..................................................................................Cavity Type Combiners ...................................................................................All Bandpass Combiners..................................................................................Notch Filter Combiner ....................................................................................Bandpass-Notch Combiners ...........................................................................Transmitter Intermodulation ..........................................................................Typical System Analysis ....................................................................................The Ferrite Isolator .........................................................................................Low-Loss Type Transmitter Combiner ............................................................Low-Loss Transmitter Combiner

For Close Frequency Spacing ...................................................................The Hybrid Coupler ........................................................................................The Hybrid Combiner .....................................................................................The Receiver Multicoupler or Combiners........................................................A Complete System Configuration ................................................................Combiner Selection ........................................................................................

CHAPTER 3 Selective Cavit ies .................. .................... ...............What is a Selective Cavity and What Does it Do? ............................................How it Works .................................................................................................Selectivity ........................................................................................................Insertion Loss..................................................................................................How to Obtain Selectivity ..............................................................................Internal Losses ...............................................................................................


6/130

TABLE OF CONTENT

Cavity Installation ...........................................................................................Tuning ............................................................................................................

CHAPTER 4 Dupl exers ................................................................The Simplex System ..........................................................................................The Duplex System .........................................................................................Why No Duplexer?.........................................................................................The Need for Isolation ....................................................................................Receiver Selectivity ...........................................................................................Receiver Desensitization ..................................................................................Transmitter Noise ............................................................................................Isolation Between Transmitter and Receiver....................................................Obtaining Isolation ........................................................................................Horizontal Antenna Separation ......................................................................Vertical Antenna Separation ...........................................................................Using a Duplexer ...........................................................................................The Advantages of a Duplexer .......................................................................Things a Duplexer Must Do ..........................................................................Losses Through the Duplexer .........................................................................The Bandpass Cavity ......................................................................................The Bandpass Duplexer .................................................................................The Cable Harness ......................................................................................... 4Band-Reject Filters and Band-Reject Duplexers ...............................................The Band-Reject Duplexer ..............................................................................The Cable Harness ......................................................................................... 4Other Facts About Duplexers .........................................................................Use of Duplexers As Combiners .....................................................................Power ............................................................................................................Temperature ................................................................................................... 4Frequency Separation ..................................................................................... 4

CHAPTER 5 RF Transmission Lin es .................. .................... ........Types of Cables ................................................................................................Mechanical Elements of a Coaxial Cable.........................................................Electrical Properties of Coaxial Cable...............................................................Characteristic Impedance ................................................................................Velocity of Propagation ...................................................................................


7/130

About RF COMMUNICATIONS iv

Corrosion ........................................................................................................Galvanic Series .................................................................................................Connectors ......................................................................................................Installation of Coaxial Cable ..........................................................................Selecting the Proper Coaxial Cable..................................................................The Installation Environment .........................................................................

Field Testing Coaxial Cable..............................................................................Standing Waves or VSWR................................................................................Problems and Solutions ..................................................................................High VSWR.....................................................................................................Preventive Maintenance .................................................................................

CHAPTER 6 Light ning ...............................................................Exposure Factors..............................................................................................Evaluation of Lightning Exposure .................................................................Lightning Current ...........................................................................................Types and Incidence of Thunderstorms ...........................................................Cycle of a Lightning Strike .............................................................................Introduction to Grounding ............................................................................Grounding Philosophy....................................................................................Equalizing Potentials ......................................................................................Antennas and Supporting Structures ..............................................................Two-Way Radio Antennas...............................................................................Microwave Antennas .....................................................................................Two-Way Radio Antenna Support Structures ..................................................Microwave Antenna Support Structures.........................................................Antenna Support Structures on Buildings......................................................Coaxial Lines and Waveguides........................................................................Lightning Safety Tips ......................................................................................


8/130

This book has been written many people engaged in RFcommunications who are nengineers. A non-technical at the complex aspects of RFcommunications is presentedbring about a better undersof the world of two-way rasystems.

While we cant possibly expanyone to become an expertcommunications simply by this material, if some of the unveiled and the picture seeclearer, we will feel amply rfor our efforts.

Introduct


9/130

vi About RF COMMUNICATIONS


10/130

Its obvious that the antenna system is an important an RF communication system without it the system wwork. Equally obvious is the fact that the antenna syscommon to both the transmitter and the receiver; any cmade in the antenna system affects both transmissionreception.

This brings us quickly to consider the economics ofsystem. We can help the talk-out range base station tobile by doubling the transmitter power, but this help the talk-back range mobile to base in the le

On the other hand, if we can change the antennafectively double the transmitter power (and well see his done a bit later in our discussion), then the powermobile transmitters will also be effectively doubled.

By changing the antenna system we help both bamobile and mobile-to-base ranges. And, generally, it

expensive to change the antenna system than it is to chthe transmitter power of the base and mobile units.

What is an an tenna?The antenna is the portion of the radio system fo

Base Stat ion Antenn


11/130

1-2 About RF COMMUNICATIONS

feeds the received energy down the transmission line receiver. To oversimplify, an antenna is designed to radiadio energy into space and collect radio energy from sp

We already know that an antenna changes radio efrom the transmission line into radiated energy and vice What is remarkable, though, is how efficiently this oc

A common household light bulb is only about 20 efficient in changing electrical energy into light (anothof radiation), whereas a two-way antenna is nearly 10cent efficient.

Of course, we dont quite get all of the energy out put in. Factors affecting this include a coaxial line that perfectly match the input to the antenna and powdue to such things as skin effect, insulator dielectricurrents, etc.

But, since we can typically claim that an antenna rbetter than 95 percent of the watts it receives from the cline provided it matches the line an antenna is efficient device when compared to most other energyting things we know.

Ma tchi ng a nd VSWRIf we consider an automobile, we know that the

gears must match if we are to transmit maximum pothe wheels. Similarly, to get maximum power from osystem, we must match the transmitter to the coaxial linantenna.

We match the transmitter to the coaxial line by tuadjusting its output circuit. And, since standard coaxiare fixed at 50 ohms for the two-way industry, we m

sign and adjust the antennas to be reasonably close tohms.

If we were to connect our transmitter to a 50 ohm line, then put a 50 ohm dummy load at the end of thwe could use a watt meter to read: (1) the power gointhe line from the transmitter and, (2) the power reachi


12/130

BASE STATION ANTEN

will be returned to the transmitter. All of the poweconsumed, either in the line or in the dummy load.

Now, lets suppose we change the dummy loadohms. This would be like an automobile gear with half imissing; it cant accept all the power presented to it. the power that would have been transferred from gear gear B remains in gear A; in other words, its sent bapoint of origin.

In a radio system, when the elements arent pmatched, the element that is mismatched rejects part signal and sends it back down the line. This rejected sigthen reflected back and forth between the load or ttenna and the transmitter. This sets up a fixed, measuwave pattern along the line; we call this the StandingRatio (SWR) or the Voltage Standing Wave Ratio (VSWR

The SWR, or VSWRas it is typically referred to, is exas a ratio. This ratio expresses the degree of match betwthe line and the antenna, or load. When the VSWR is oone (expressed as 1.0:1) a perfect match exists. If the VSW1.5:1, the percentage of reflected power is only 4%, or, another way, 96% of the power that gets to the end line goes into the antenna.

Waveleng th, Freq uency, and VelocityWe know that a radio wave travels at the same speed

beam of light around 186,000 miles per second, or a billion feet per second. This is its speed or velocity.

We also know that a radio wave oscillates or altfrom plus to minus, back to plus, back to minus, etc. Thetion from plus to minus and back to plus is called one c

one hertz since, like a wheel, it repeats itself (see Fig1).

The number of cycles the signal goes through second is called the frequency. If we know the freq(which we can measure), we can then find out how wave travels in one cycle by dividing the speed by t


13/130


ranges of frequencies used in two-way communicatshown in Figure 1-2.

An idea of the size of a wavelength, or a half waveis useful when we consider how antennas radiate in por out of phase.

Antennas Work the Sam eWhen Tra nsmitt in g or Receivin g

Fortunately, for two-way radio users, antennas equally as well as they transmit. This is true even thodoesnt always seem that way in actual performance. Dsuch things as noise and interference, the antennas trueformance in the receive mode can be masked.

Lets start with a basic type of antenna and use it as aping stone to understanding more complex gain array

The Half- Wave Dip oleThe most basic antenna we use in two-way base s

is the half-wave dipole radiator. The half-wave dipole isa straight conductor made of wire, rod, or tubing that, cally, is one-half wavelength long. Generally, the feed ltaches at the middle. It radiates at maximum intensitymiddle of the dipole, at right angles to its length; thmum intensity is at its ends (see Figure 1-3).

In two-way mobile radio services, the half-wave dcommonly referred to simply as a dipole is the acreference standard when we state performance or gaiother types of antennas. Gain is to antennas what horsepis to automobiles; it is a measure of performance powe

Since dipole antennas radiate best when at a res

length relative to the desired frequency, they are generalor adjusted in length to a desired frequency. Furthermorelectrical half wavelength is generally a few percent than the physical half wavelength. This is done to alwhat is called end effect of the conductor.

A rule of thumb for the length of the half-wav


14/130

BASE STATION ANTEN

Frequency (MHz)1/2 Wavelength

(Feet)Frequency (MHz)

1/

25 19.7 150

30 16.4 160

35 14.1 170

40 12.3

45 10.9 450

50 9.8 460

74 6.7 470

Figure 1-1: Relationship of Time, Frequency, and Wavelength

Figure 1-2: 1/2 Wavelengths of Two-Way Frequencies

one foot long. Its good to keep these lengths in mindwe start talking about antenna arrays or stacking dipotowers.

Antenna Polar ityAntenna polarity simphow the antenna is orientetioned. If the radiating eleoriented vertically, then ivertical polarization; if the el

+ + +

A cycle completes itself in going from plus to minus back to plus.

The number of cycles in one second giv es the fr equency.

The speed dividefrequency gives the w ave travelsThis is called the

Maximum Radiatio n


15/130


since it is easier to install a vertical whip on a vehicle thto install a horizontal one. If our mobile antenna is vpolarized, then, of course, our base station must be verpolarized if we are to obtain maximum efficiency anfrom the combination.

Vertical a nd Hor izontal Antenna Radi ation PatternAll antennas have a given three-dimensional ra

pattern. If the radiation patterns were equal in all directwould be that of a round ball or a sphere.

If we cut the sphere vertically we would see the pattern, which would be a circle. On the other hand, if the sphere horizontally we would see the horizontal pand it too would be a circle. We could then say that thecal pattern was omnidirectional and the horizontal patteomnidirectional, and that the two were equal.

No True Omni dir ectiona l AntennasIn the previous paragraph, we stated that the an

patterns would be omnidirectional. In actual practiceever, there is no truly omnidirectional antenna it exisin theory. Our half-wave dipole antenna, mounted ve

as used in two-way communications, has a three-dimenpattern as shown in Figure 1-4a. It appears as a largdoughnut. Its horizontal pattern (Figure 1-4b) is circuits vertical pattern (Figure 1-4c) looks like a fat figure-ing on its side.

Dipole in center of doughnut-shaped pattern.


16/130

BASE STATION ANTEN

In two-way mobile communications we really arecerned so much about the antennas vertical pattern. field, we are always working at horizon elevations, eva tall tower on the top of a hill.

For example, if our antenna is mounted on a tower200 feet tall and the tower is on a hill that is 200 feet hiwould consider the antenna to be 400 feet high.

At a distance of, lets say, 20 miles, the angle betwebase antenna and the mobile unit would be less thadegree or, for practical purposes, the horizon level itsin the greater ranges where we need maximum gaiwhere were concerned with the antenna pattern and

Antenna GainAntenna gain and pattern shape are interrelated;

change one, you generally change the other.Just as we need a starting or reference point wh

survey land, we must have a reference point to start from we talk about gain. In addition, just as inches, feet, and (or centimeters, meters, and kilometers), etc., are usedunit of measure in surveying land, we need a unit of mwhen we talk about gain.

The reference we use in two-way base station antenthe half-wave dipole and the unit of measure is the decibthe point of reference, we use the half-wave dipole andhas a gain of one (or unity), or stated in decibels, it has aof zero decibels.

What Ab out Tha t Word Decib el?The decibel (usually abbreviated as dB), is used to

pare one power level to another. For example, if we haantenna that has twice the power gain of the half-wapole, we can find the power ratio of 2.00 in the table ure 1-5 and see that the our antenna will have a 3 dBover a half-wave dipole.

H D W G A G i ?


17/130


tern shape with greater intensity. Unfortunately, howevcant increase the power, so this cant be done.

2. The other option is to change the shape of the pso it radiates more of the antennas signal in a particular tion. This is something we can do.

Since we dont need radiation in all directions, increase the signals intensity by changing the shape antenna pattern. We do this by designing the antenna soit radiates the same amount of total power, but we cthe shape of the pattern so that it directs the radiation we want it.

To get a bett er idea of how this works, lets comparlawn sprinkler. The sprinkler head, which would represantenna, is attached to a water hose; the water hose wrepresent the transmission line. If we adjust the sprinkleso that it covers a full circle and turn the water on full fthat is, we set it at full power we distribute the wacircular pattern around the sprinkler head.

Now, if we readjust the head so that it only covcorner of the lawn and turn the water on full force agoutput the same amount of water as before, but this

Power Ratio dB Power Ratio d

0.10 -10 1.00

0.13 - 9 1.26

0.16 - 8 1.58

0.20 - 7 2.00

0.25 - 6 2.50

0.32 - 5 3.16

0.40 - 4 4.00

0.50 - 3 5.00

0.63 - 2 6.30


18/130

BASE STATION ANTEN

because we have limited the pattern to one direction, the or our antenna signal goes further and outputsinto a given direction.

Squa shin g th e DoughnutLets look at the doughnut pattern of the dipole aga

Figure 1-6a). If we squash it down on top, it will flattena round, flatter shape (see Figure 1-6b; and the more we sqit, the flatter it gets and the larger in diameter it becomother words, the horizontal area is increased at the expenthe high vertical area. And, this is exactly what we want.

Since we dont have any mobile units running arspace, we dont want the radiation to be wasted up theflattening, or squashing, the signal we now have an orectional or horizontal pattern which has more gain thoriginal dipole pattern. This is the desired antenna patta base station that needs maximum range in all directio

Omnid irectional Pattern Gain AntennaTo achieve greater gain with an omnidirectional

we can stack multiple vertical dipoles above each o

Ap ert ur e of Dip oles Vertical Pattern Horizon tal Pat

Sing le Dipole

Four Dipoles Vert ically Stacked

(a) (b) (c)


19/130


shown in Figure 1-6b. This increases the vertical size orture of the signal.

We then feed the dipoles with power in such a wthe radiation from the individual dipoles will add togetdistant point. Also, we must connect the dipoles togetthey match the cable if we want to get the most rpower out of the antenna. The most radiated power tained when the dipoles are: (1) lined up vertically (cwith optimum spacing between them and, (2) fed withpower that arrives at the dipoles at the same instant (in ph

This type of antenna is called a collinear (vertical) array. And, since it is the most common and most ptype of base station gain antenna, its important that we ine it a little closer.

In Pha se an d Ou t of PhaseReferring to Figure 1-7, lets look at the things tha

the radiation from two or more dipoles to be either in or out of phase. When we look at these examples, weline we are standing at a particular point that is locateddistance from the dipoles.

In the first example (see Figure 1-7a), we have two d

A and B. They are positioned vertically, one above tand have a separated distance of about one wavelengthtween their centers. Several miles away, an observer wantenna and receiver is located at point X. Point Xsame distance from each dipole.

Dipoles A and B are each connected to a transmtwo coaxial cables of equal length. This means that the from the transmitter arrives at A and B at the same

Also, since each dipole is connected to its transmission the same manner, the dipoles start radiating in the same dtion at the same time.

Now, since the observers antenna at point X is thdistance from each dipole, it will receive the two sigexactly the same time. And, together, the two signals w


20/130

BASE STATION ANTEN

Trans- mitter

Cable A

Cable B

Dipole A

Dipole B

2 miles

Radiation from Dip ole A arrives at Observer X at tinstant as that f rom Dipo le B. Since signals A and B astronger overall signal is produced.

In-Phase Radiation

Trans- mitter

Cabl e A

C a b l e B

Dipole A

Dipole B

Cable A is th e same length as Cabl e B

2 miles

Radiation f rom Dipole A arrives at Observer X out ofwith the radiation from Dipole B because Dipole A andin d istance by 1/2 wavelengt h. Signal A cancels signal B

Out-of-Phase Radiation

1/ 2 wavelength

Trans- mitter

C a b l

e A

Cabl e B

Dipole A

Dipole B

Cable Ais 1/2 wavelength short er than Cabl e B

2 miles

Radiation f rom Dipole A arrives at Observer X out of pwit h radiation from Dip ole B because Cable A is 1/2 wavshorter (electrically) than Cable B. Signal A cancels signal


Cable A is th e same length as Cabl e B

Cable A is

Cable ADipole A

2 miles

R di ti f Di l A i t Ob X t f


Trans- mitter

(a)

(b)

(c)

(d)


21/130


wavelength. Now, the signal from dipole B arrives X one-half cycle sooner than the signal from dipole AA is going minus, B is going plus and the signaceled. We say the two are out of phase.

Now, suppose observer X had a hot air balloon ango up in the air and find some spot where the distancthe same between himself and the two dipoles; here tdiation would again add together, they would be in phthis situation, we would say the beam was tilted up.

We could also say this would be a poor base statitenna. The radiation was not destroyed or lost by phasecellation, instead it was just repositioned or reshaped itern.

Figure 1-7c shows how the same out-of-phase results

be obtained if one of the transmission cables were shorone-half wavelength. Actually, the cable would not hbe just one-half wavelength shorter; any odd-half such as 1/2, 3/2, 5/2, 7/2, etc., would also cause the sto be out-of-phase.

To be in phase the cables dont have to be the exact length. They can be of different lengths but they musby multiples of a full wavelength for example: one fu

length, two full wavelengths, three full wavelengths, etcboth transmission cables are at the proper length, the cuflowing in cable A will arrive at dipole A at the samthe wave cycle that the current flowing in cable B ardipole B. Since there is one cycle per wavelength currents arrive at the dipoles at the same time or they difa multiple of one full wavelength or cycle the currsaid to arrive in phase. As a result, the signals radiate

the dipoles will start out together, or in-phase.Figure 1-7d shows the third way that the signals fpoles A and B can be out of phase.

For purposes of simplicity, it is assumed that the innductor of cable A connects to the upper half of dipwhile the outer conductor of cable A connects to th


22/130

BASE STATION ANTEN

connected to the upper half of dipole B. This is exactlsite from the connections made in cable A.

Because of these mismatched cable-to-dipole connecwhen the current in A goes positive on the upperdipole A, the current in the upper half of dipole negative; and vice versa for the lower half of each dip

As a result, as far as point X is concerned, the twocancel each other because they arrive out-of-phase.

To summarize, the three most common ways thattion gets out of phase with respect to a distant antennreceiver are:

(1) Dipole radiators are displaced in distance; (2) Feed cables are not of equal lengths or in mult

a full wavelength;

(3) Dipole radiators are not connected to the feed in the same way.

ApertureAs the aperture or opening size of a valve contr

amount of water that flows through a pipe, the aperbeam width determines the gain of the antenna. The effaperture actually takes in something more than the ph

size. We think of the aperture as the signal surroundiantenna in all directions and extending out a given di(such as one-half wavelength) from the sides and ends. Tfore, it can be said that the aperture is a volume of spac

As an example, a smaller aperture or beam width, degrees, will have a greater gain than a larger aperture90 degrees. The radiation pattern in the smaller beam wprojected farther forward along the horizontal plane an

along the vertical plane; this results in a higher gainversely, the radiation pattern in the larger beam widmore of the signal projected along the vertical plane analong the horizontal plane; this results in a lower gain

Spa cin g of Dipol e Elem ents


23/130

1-14 About

RF COMMUNICATIONS

aperture that will give maximum gain. This separation erally somewhat less than one wavelength between ce

As the dipoles are spaced closer together, the gain fabecause the coupling between the ends cancels some oeffective radiation.

Feeding the Collin ear Arr ayIn a collinear array of two or more dipoles, the tw

common means of feeding the power from the coaxialmission line to the individual dipoles are by (1) serieand (2) parallel or shunt feed.

In series feed, the power flows up through a singlto the first dipole, then to the second dipole, then to theetc. In this arrangement, the top dipoles are being fed co

erably less power than the bottom dipoles; as a result, thdipoles do not contribute as much to the gain of theFurthermore, this arrangement tends to tilt the lobearray off the horizon, further decreasing the gain. Thethe number of dipoles that can be fed in series reaches a

Parallel or shunt feed is generally accomplished bning a separate feed cable to each dipole. Then, using ming transformers and junctions, the cables are connec

the transmission line that runs down the tower. Thisthe array to be fed in the center. In this arrangement, beais avoided because essentially equal amounts of powebe delivered to each dipole. In addition, each dipole ially as effective as the others since they all receive similar input.

Whether series or parallel feeding is used, approxthe same gain can be obtained in the same physical leng

aperture of the array.This brings us to the conclusion that with proper and we must emphasize proper design the effectivof a collinear array is dependent upon the aperture; andin turn, is largely dependent upon the physical lengtharray.


24/130

BASE STATION ANTEN

to get this increased range, it would also seem that we stack as many dipoles as our tower would hold, therebting the most signal out of our array. As good as thsound, there is obviously a physical and economic end approach.

First of all, the United States Federal Communicationmission (the FCC) says that if an antenna structure exmore than 20 feet above the top of a tower, building,tank, etc., it must have lighting on the tower to alert airits presence. This implies that the support structure mstrong enough to support the antennas and the lights, amust have climbing steps to access the lights for mainten

From an economic standpoint, this pretty much ruantennas that extend more than 20 feet above the top

tower. Therefore, given these constraints, the gain of annidirectional, top-mounted antenna is limited to approx6 dB at 150 MHz or 10 dB at 450 MHz. (Youll noteomnidirectional. Its actually possible to obtain greater gaa directional array.)

In the 25-50 MHz band, specially designed dipoles mounted on the side of a tower to obtain an omnidirepattern. These antennas would have cross-sectional di

sions in the order of one-eighth wavelength or less. towers measuring several hundred feet, up to six dipolebe stagger-mounted down the tower. Mounted on opcorners or faces of the tower, results can be achieved approach the collinear gain antennas at 150 and 450 M

Directional Gain AntennasWeve talked about omnidirectional gain antennas

tical collinear arrays. These antennas have a circular, hotal pattern. We have also stated that they achieve gain by pressing the vertical pattern down to the horizon using tically stacked, collinear dipole array. Now, lets consitennas that primarily shape the horizontal pattern to again.


25/130

1-16 About

RF COMMUNICATIONS

The Di pol e a nd ReflectorIf you will remember, a vertical half-wave dipole h

cular horizontal pattern. If we place it in front of a screenof metal or wire mesh (see Figure 1-8) it is evident thattion going to the rear will be blocked. If this blocked ris redirected, the resulting pattern will no longer be cir

We know from theory and experiment that if the dspaced a quarter wavelength in front of the reflecting sthe radiation that would normally go to the rear is redto the front to form a directional lobe hence a dirantenna. Also, the larger the screen (to a point) the narthe directional lobe becomes. In this arrangement, wsay the antenna has narrowed its beam width or increasgain. In effect, then, the dipole serves to illuminate the

with the screen radiating on transmit and collectingceive.

The Corner Reflector an d Par ab oli c ReflectorIf an antennas reflecting screen is formed into a righ

or a V of the proper size, and if the dipole is located a distance from the screen angle intersection, the antenncomes a corner reflector directional antenna (see Figure

The beam width and gain will depend upon the relatiof the screen size, the angle of the screen, and the dposition. The screen also can be formed into a parabolireflector with the dipole at the focal point (see Figure

To reduce wind drag, instead of making reflectowire mesh screen or solid metal, we can use closely svertical rods (see Figure 1-8). The use of vertical rods also mfor a more rugged structure.

The Yag i Anten naPerhaps the most widely used directional gain ante

the Yagi. The Yagi has many forms and variations but ally it consists of at least two elements, and more oftenelements (see Figure 1-9).


26/130

BASE STATION ANTEN

behind the radiator. In general, the director elementshortest element while the reflector is the longest. The of the elements and the distances between them deterthe radiated power that goes into a directional lobe.these factors ult imately determine the Yagis gain.

Due to its high gain, low weight, low wind dragrelatively low cost, the Yagi antenna is considered to bsuited for use in two-way radio communications.

How M an y Elem ents?To increase the gain of the three-element Yagi, w

add additional directors in front of the first director butpractical standpoint, there is a limit to this arrangeme

Figure 1-8: The Omnidirectional Pattern of a Dipole can be made Directi

Corner Screen The screen behi nd t he

dipole cuts off radiation to the b ack and reflects

i t forward to form a beam.

Corner Reflector Vertical rods replace the screen o f th e Corner Screen t o mak e a corner reflector o f stronger mechanical design.

Flat Screen Radiation t o the rear is blocked.

1 /4 W.L.

Parabolic Screen The parab oli c screen focuses the beam.


27/130

1-18 About

RF COMMUNICATIONS

thermore, for applications where the antenna must oon multiple frequencies, the additional elements narrband width and make it less useful.

Another way to increase the directivity or the gYagi antenna is to position two antennas side by side. done by determining the proper spacing between the a

nas, attaching them to a horizontal support bar, then, ming the support bar to the tower. This has the advantabringing the center of the antenna array into the momast or tower. This is especially important in the 30-5band.

Pha sed Rad ia tor s for Di rectivi tyIn our discussion of in-phase and out-of-phase con

of base station gain antennas, we said that when ObsX (see Figure 1-7) was positioned an equal distanctwo radiators, the transmitted currents would add up in-and increase the gain. But, when Observer X was poswere there was a difference of one-half wavelength bethe radiators, the currents canceled each other. This pri

Reflector

Figure 1-9: A Two-Element and Three-Element Yagi Directional Anten

Reflector Reflector Director

Director

Direction of beam


28/130

BASE STATION ANTEN

A Bidirectiona l Ar ra y or Figur e-Eigh t PatternA bidirectional, figure-eight pattern can be form

two radiators that are spaced one-half wavelength aparfed in phase. Looking at Figure 1-10, it can be seen that the line AB the radiators will be equal distances. Ththe radiation from them will arrive along this line at thtime, or in-phase, to produce an increase.

Along line CD, however, there is a difference of wavelength, which is out-of-phase. The radiation aloline will cancel and make a null. The resultant patternfore, is a figure-eight.

Addi tiona l Gain by Stack ing Dir ectiona l AntennasSince directional gain antennas obtain gain prim

compressing the horizontal pattern, it becomes evidenadditional gain can be obtained by stacking such antenna vertical line as we did with the omnidirectional collin

antennas. The net effect, then, is to compress both the zontal and vertical patterns.

This double compression gives optimum utilizatiopower in a certain direction where maximum gain is dOf course, they must be phased together correctlmatched to the transmission line to achieve maximum

Figure 1-10: This figure illustrates how two dipoles spaced horizontally ap1/2 wavelength and fed in phase, produce a Figure 8 bidirectional patte

A

C

Dipoles ar e 1 / 2

w av e lengt h apar t a

n d

cance l a long t h is line .

Dipole 1

Dipole 2

D i p o l e s a r e t h e d i s t a n c e f r o m t h

a n d a d d t o g e t h e


29/130


Again, it must be remembered that the effective ground height of the antenna is to the center of the Therefore, in the 30-50 MHz band, where the antenntypically quite large, stacking additional elements mimprove performance unless the tower is tall enough commodate them.

Side Moun ted Antenna sMany directional antennas, and so-called omnidir

antennas, are mounted on the sides of towers. For the tional antenna this poses no problem. Since the radiatdirected out from the tower, the tower has little effecthe pattern or gain. This is not true with antennas desfor omnidirectional patterns because the tower will b

excited by the currents radiated into it. This will cradiation to be redirected in a directional pattern.

Then How Do We Get an O mn i Patt ern ?When an omnidirectional pattern is desired from

mounted antenna, two or more radiators of proper can be placed around the tower in a manner that will psevere cancellation between the radiators.

This is practical in the 30-50 MHz band through thea folded dipole radiator. When placed on opposite sidetower where the horizontal displacement distance bethe two radiators is no more than 3/16 wavelength, thtern will essentially be omnidirectional. By stacking thtors apart vertically the effective aperture of the anteincreased and gain is obtained. On towers that are tall enoadditional gain may be obtained by adding radiator

vided the radiators are positioned properly and matcthe transmission line.Limitations in this arrangement include: a tower t

tall enough to be effective, signal loss in the extra conncables, and the costs associated with the additional antenn


30/130

BASE STATION ANTEN

high levels compared with the noise level at the mobilewill make the antenna appear to be inferior when recversus when it is transmitting. On the other hand, if thelevel at the base station is low compared to the mobiletion, it will make the antenna appear superior when reversus when it is transmitting.

What Do We Mean by N oise?When receiving, a antenna is a collector. It colle

only the desired signal but, because it cant differentiatween a valid signal and noise signals, it collects anysignals that fall within its pattern and bandwidth. Itdiscriminate or select. Furthermore, since the antenna is less selective than cavity filters or receivers, it collect

over a fairly broad frequency range.

What Hap pens to Noise with Gai n?Sometimes, when a unity-gain (0 dB) omnidirectio

tenna is replaced with a high-gain directional antenperformance does not meet expectations when receWhen the higher gain antenna compresses the verticatern, the noise level often increases along with the incr

gain. In terms of signal-to-noise improvement, the reimprovement in gain is not as apparent upon receive aupon transmit. This is especially true in the 30-50 MHz 148-174 MHz bands in metropolitan or industrial area

Directional gain antennas could hurt or help the sidepending upon whether the directive beam looks away from a area of high noise. For example, if a 4directional antenna looks out across a busy metropolitan

it will likely show poor performance upon receive. Howits turned away from the city and directed toward a resior rural area, it may show an improvement in receptiongreater than its gain. This would be due to its front-ratio being used to discriminate against the noise and imthe desired signal.


31/130


What About the 450 MHz Band ?Theres noise at 450 MHz, but its greatly reduced fr

of the 150 MHz and 40 MHz bands. It must be rememthat noise is actually a combination of many smaller Principally, among these are man-made noises such as mobile and truck ignitions, electric power distributio

and electrical machinery. These noises are much stronthe 30-50 MHz band; there is a considerable reduction MHz, and even further reduction at 450 MHz.

What Ab out the Count ryside?Noise prevails not only in the city and industrial ar

often in rural areas far removed from the freeways aelectrical machinery. Generally, the offender is a relative

voltage electric power system used in supplying pofarms. The type of construction used in the power systemthe lack of a suitable earth ground can play havoc with a nreceiver.

A similar condition can exist along sea coasts whereline insulators have become encrusted with salt depositsresults in corona and voltage breakdowns across the intors on the high voltage power lines. Fortunately, a goo

can often clear up this condition.Noise is a great problem and careful surveys shomade when low noise is essential to a systems perform especially where high gain antennas are used. Also, blankers can be quite effective against impulse noise fromignition systems, but they are less effective against pownoise.

Bandwidth Theres one more thing about antennas that we consider: bandwidth. Earlier in our discussion, we statean antenna had to be resonant at the operating frequencwork properly. We also stated that to be resonant, it hadcut to the right length. While these statements are true, it


32/130

BASE STATION ANTEN

even a megahertz, above or below the tuned frequency at least so well you couldnt tell the difference.

By designing a antenna with a bandwidth in mipending upon the frequency band, some types of antcan be made to operate equally well over a broad ran15 to 20 megahertz, for example. These are the ones we

to as broadband antennas.While theres a lot more to antenna design than

covered in this section, wed like to emphasize is that no magic in antenna design there arent any electronnets, wave concentrators, or signal intensifiers. An antea passive device; it can only radiate the power sent to ithe transmitter or furnish the receiver with energy it from the air. But, with a minimum efficiency of 96 pe

does a remarkable job of these two functions.


33/130



34/130

The need for combiners has long been known anrecognized importance continues to grow at an accelrate. More and more land mobile radio systems are equipped for simultaneous operation on several frequfrom a common site, and a combiner can eliminate thefor separate antennas for each radio system. In additireducing the number of antennas, better performancusually be realized if the highest antenna site is selected

used with the optimum combiner.A single master antenna and its transmission line shared by two or more transmitters, receivers, or simplexstations by connecting them to the antenna through abiner. Sharing of a single antenna is not limited to asystem operator. When several base stations, operated byferent users, are located at the same site, they can often sa common antenna, depending upon the frequencies u

Most radio systems that operate at a common siutilize independent antennas and transmission lines wquire multiple interference protective devices. These are ferrite isolators for reducing transmitter intermodulan acceptable level, bandpass or band-reject cavity filtestalled between transmitters and antenna) for reduct

Combin


35/130


dio systems are to be achieved, these losses can apprthose of a combiner and yet not afford an optimum RFantenna site.

Combiner RequirementsA combiner that enables the use of a common anten

two or more transmitters should cause a minimum oftion loss (transmitter power loss) and should provide degree of isolation between the transmitters. This ensurepotential transmitter-produced intermodulation frequencminimized.

Transmitter intermodulation is the primary factor thbe considered when two or more transmitters are cominto a common antenna. In addition to the above, a

biner that enables a number of transmitters and receivuse a common antenna must also ensure that any recdesensitization caused by the transmitters and any transmnoise at the various receiver frequencies are reduced acceptable level.

When the transmitters and receivers share a commotenna through a combiner, the only practical method otection for transmitter noise and receiver desensitization

use of resonant cavity filters between the transmitters aceivers. Should the frequencies be separated by a reasonamount, a simple cavity-filter-combiner configurationused for two or more systems the radio manufactureplex operation curves can provide the proper isolation refor any given frequency separation. If the transmitter frcies are extremely close, the hybrid/ferrite isolator combnormally used.

Cavi ty Type Com bi ner sThe cavity-type combiner is one of the most commo

biners used to couple transmitters and/or receivers into a antenna. This type of combiner is generally more econand affords less insertion loss than hybrid/ferrite combi


36/130

COMB

Also, as mentioned previously, if the systems to be comcontain receivers, cavities must be used since hybridcombiners have unidirectional devices (isolators) as cnents.

The cavity-type combiners can be composed of all bancavities, all notch (band-reject) cavities, or a combina

bandpass and notch cavities.

All Band pa ss Comb in ersThe bandpass combiner is used when a fixed num

stations with a relatively wide frequency separation arebined into a single antenna. Figure 2-1 shows a blocgram of a bandpass combiner. Two transmitters and twceivers are combined into a single antenna through th

of bandpass cavities and a five-way connector. The bancavities in the transmitter lines protect the receivers frommitter sideband noise radiation by attenuating the outthe transmitters at the receiver frequencies. Those in the tmitter lines also mutually isolate the transmitters, thereducing the possibility of transmitter intermodulatibandpass cavities in the receiver lines protect the receiversreceiver desensitization by attenuating the transmitter c

before they reach the receivers.The number of cavities in each combiner system is dent upon the frequency separation between the systemFigure 2-1, if the frequencies were closer together, four osibly five bandpass cavity filters would be needed to misolate the systems. The length of the interconnect tran

T1451 MHz

Antenna


37/130


38/130

COMB

The block diagram in Figure 2-2 shows three scoupled by use of a notch filter combiner. Note that tquency separations are 540 KHz and 615 KHz in the VHFEach system is mutually protected from the other by a setwo cavity-notch filters having about 50 dB isolation. sertion loss of each system measured at the antenna port

put) is approximately 2.5 dB.The addition of another system to the above exampossible but economically impractical since the number oties would nearly double. The reason for this is that eactem must be protected from the new system, while thsystem needs protection from the existing systems. Doinwould add eight more cavities to the setup. Because othe band-reject type combiner is best suited for com

two or three, but rarely four, systems.

Antenna

Reject 161.415

Reject 160.875

Reject 161.415

160.8

Reject 160.260

Reject 160.875

160.2


39/130


The advantage of the band reject combiner is its combining systems with close frequency separations. minimum separations are approximately 150 KHz in band, 500 KHz in the 150 MHz band, and 1 MHz in tMHz band. An important disadvantage is its lack of ence protection from other systems located nearby.

Bandpass-Notch CombinersThe bandpass-notch class of cavity-type combiner

the combination of bandpass and notch filters to allowtransmitters and/or receivers to be used with a single antThis is a modular-type system in that each channel consa bandpass filter and a notch filter, as shown in Figure

The purpose of the bandpass filter is to isolate the s

from others in the multicoupler while the notch filter

mainly for matching the input terminal to the outputlow insertion loss from port 1 to port 2 and vice ver

number of cavities in the bandpass filter will be depupon the frequency separation between the systems anmaximum insertion loss that can be tolerated.

Figure 2-4 shows how these modules fit together ta four-system combiner. The last system does not need a nfilter for matching since it is the terminal system. How

Figure 2-3: Bandpass - Notch Module

Bandpass Filt er Notch Filter

2

3


40/130

COMB

readily expandable system is needed and the minimu

quency separations are approximately 500 KHz in the low1 MHz in the 150 MHz band, 2 MHz in the 450 MHand 5 MHz in the 806-960 MHz band.

Like the all-bandpass combiner, the bandpass-notchbiner has the advantage of added interference protectionth t i th i di t It di d t g i

Figure 2-4: Bandpass / Notch Combiner

Antenna


41/130


Tra nsmi tter Interm odulati onThe all-bandpass combiner offers excellent protec

transmitter noise and receiver desensitization. But what transmitter intermodulation products and/or rintermodulation products? As we stated earlier, transmiis the primary factor that must be considered when

more transmitters are combined into a single antenna.All intermodulation effects produced by mixing linear outputs of either transmitter, or the non-lineastages of the receiver, are a function of the frequency ence between the combined frequencies for any numsums and the differences of the fundamental frequencietheir harmonics. This is illustrated for the familiar third order intermodulation products in Figure 2-5.

Typi cal System Ana lysisThe following example illustrates the thiintermodulation protection provided by the all-bandpasbiner represented in Figure 2-5: A is frequency T1, Bquency T2, and their frequency difference is S. Thirintermodulation frequencies 2A-B and 2B-A and f3A-2B and 3B-2A are depicted. There are an infinber of intermodulation frequencies having a frequency

ence of S. However, normally only the third order are of sufficient power levels to cause interference probThe attenuation to each intermodulation product

triple bandpass cavities can be determined from the cavsponse curves. The energy coupled into transmitter Atransmitter B is attenuated approximately 70 dB (a

Figure 2-5: Intermodulation Frequencies

431 MHz 441 MHz 451 MHz 461 MHz 471 MHz

3A-2B 2A-B A B 2B-A

S S S S


42/130

COMB

by more than 70 dB. Consequently, the thirdintermodulation products are greater than 140 dB belotransmitter carrier(s) a most insignificant level.

However, suppose this is not the only system at thition. Further, suppose another 9 dBd gain antenna rad250 watts, at a frequency two channels higher, is located

zontally from our systems 9 dBd gain antenna, as shoFigure 2-6.

Figure 2-6: Close Coupled Systems with 3rd Order Intermodulation Em

6 ft .

9 dBd Gain Omnidirectional

Antenna

9 dBd Gain Omnidirectional

Antenna

Ta 451 MHz

Tb 461 MHz

Ra 456 MHz

Rb 466 MHz

T451.0

MH


43/130


nas and transmission line loss. The cavities offer practicaattenuation other than the insertion loss at this frequency ration.

Let us determine the approximate power of the rathird order intermodulation frequency from transmitte

Coupling loss: Antenna space coupling loss = 25 dB (Antenna 1 to 2Transmission line loss = 1 dB (Transmitter A and B)Cavity loss at + 50 KHz = 2.5 dB (A)Total = 28.5 dB

Third Order Intermodulation Product radiated fromTransmitter Power (250 watts) = + 24 dBw (from C)

Coupling loss = 28.5 dB (C to A)Power in final (Tx A) = 4.5 dBw Transmitter conversion loss = 6.0 dB (assumed)Cavity loss at + 100 KHz = 3.0 dB Antenna gain = + 9.0 dB (Antenna A)Transmission line loss = 1.0 dB (A)ERP of lM product = 5.5 dBw or 0.28 watts (reference wave dipole)

This radiated power could be a problem to a mobtem that might be in the vicinity of the antenna site. Ieasily occur if our site were a building roof top.

The Ferr ite IsolatorThe ferrite single junction isolator (see Figure 2-7) is

port circulator (see Figure 2-8) with a matched resistivconnected to port 3. The ferrite circulator is a three-porreciprocal device consisting of ferrite material, magnethree short lengths of transmission line terminated at amon junction. The basic ferrite material commonly useoxide of yttrium iron garnet (YIG) commonly calledThe garnet is cut to the proper shape in a manner sim


44/130

COMB

response for the opposite direction is flat. It is from theposite effects that isolation is permitted between the pthe circulator. Power entering port 1 is rotated and emat port 2. Power entering at port 2 emerges at port 3 absorbed by the resistive load. The load at port 3 shou

ways be of an adequate wattage rating to absorb the mmum expected reflected power from the subsequent ansystem.

The ferrite isolator is the most effective solution tmitter-produced intermodulation when the frequency stion between the transmitter frequencies is as close as cent channels. When installed between the transmittesubsequent combining network, or between the transand antenna, the ferrite isolator acts like an RF diode. It ptransmitter power from its input (port 1) to its outputwith very little loss perhaps 0.5 dB but attenuaergy in the opposite direction by 25 to 30 dB for thestage isolator.

Low- Loss Type Tra nsmi tter Com bi nerTo prevent radiation of t ransmitter intermodulati

our all-bandpass cavity combiner of Figure 2-6, we canpair of tunable ferrite isolators to each transmitter leg as sin Figure 2-9. The coupling loss of Figure 2-9 is now incby an additional 60 dB, and, with this added insertiothe power radiated by the third order intermodulatiquency is 0.28 microwatts an insignificant amount

Figure 2-7: Ferrite Isolator Figure 2-8: Ferrite Circu

2

3

1

2

1


45/130


Low-Loss Transmitter CombinerFor Close Frequen cy Spacin g

Utilizing extremely high Q bandpass cavities anisolators, the all-bandpass combiner shown in Figure 2be used for extremely close transmitter frequency separaA four-channel low-loss transmitter combiner is showmatically by the block diagram in Figure 2-10.

Figure 2-9: Close Coupled system with Ferrite Isolators

C = Cavity, I = Ferrite Isolator

9.2 dBd Gain Omnidirectional

Antenna

9.2 dBd Gain Omnidirectional

Antenna

6 f t .

Ta 451 MHz

Tb 461 MHz

Ra 456 MHz

Rb 466 MHz

Tc 451.050

MHz

C

C

C

C

C

C

C

C

C

C

C

C

I

I

I

I

I

I

5-Way Connector


46/130

COMB

tivity the closer the frequency separation betweennels can be and still maintain low insertion loss per ch

In an extremely high Q coaxial cavity the input imrapidly approaches a low impedance as the frequency moff resonance. In the 150 MHz frequency band, the ance at 60 KHz off resonance is low enough that by usi

proper length of cable between the cavity and the N-waytion (a quarter-wavelength less the electrical length of thpling loop) a high impedance is presented at the juThis allows all of the signals from the other transmitters through the junction at a minimum loss.

A definite advantage that the low loss combiner in2-10 offers is the additional filtering from the high Q cathe line. At 5 MHz from the transmitter frequencies the receiver frequencies usually fall in the UHF band mitter noise and spurious responses are usually reducleast 50 dB from the output of the transmitter.

The low loss type combiner in Figure 2-10 is expto a 5-, 6-, 7-, or 8-channel unit. The loss through thebiner does not appreciably increase unless the frequency sration between channels is decreased.

5-Way Connector

Antenna

High Q Cav

Load Is

Load Iso


47/130


The Hybr id Coupl erFigure 2-11 represents a four-port, stripline, 3 dB

tional coupler. When the hybrid is properly terminatmatched loads, isolation is provided between ports 1 anup to 40 dB. Energy entering port 1 and/or port 2 splhalf going to port 3 and half to port 4. Consequen

termination at port 3 must be able to absorb half otransmitters power.

The Hybr id Com bi nerThe hybrid two-transmitter combiner illustrated i

2-12 utilizes 3 dB, stripline, hybrid ferrite isolators apass harmonic filters to combine two transmitters. Themitters should be at adjacent channels or at channels upfrequency separation determined by the bandwidth of tbrid and/or ferrite isolators. The harmonic filter is r

Figure 2-11: 3 dB Stripline Hybrid Coupler

Figure 2-12: Hybrid Transmitter Combine

Antenna

Input Input

Load

2

4

3

1

70 dB Isolator Isolator

Hybrid

Insertio n Loss 4.0 dB

Insertio n Loss 7.5 dB

H


48/130

COMB

because the ferrite isolator may itself produce harmonicbandpass cavity rejects harmonic energy in the low-loscombiner.

The use of additional hybrid couplers and two-tracombiners allows expansion to either a four-channel trater combiner (see Figure 2-13) or an eight-transmitte

biner (see Figure 2-14). The nominal isolation betweenmitters is 65 to 70 dB when a single isolator is used; the nal isolation can approach 100 dB when two isolatoused per channel. However, as the number of transmincrease, the insertion loss for each transmitter increasesexample, the two-channel combiner has a loss of approxim4 dB, the four-channel has a loss of approximately 7.5 dBthe eight-channel has a loss of approximately 11 dB.

A disadvantage of the hybrid-type combiner the transmitter power loss is that the isolation of theis a function of how well the hybrid is matched to tenna system. An input VSWR of 1.5:1 on the transmline to the antenna system can cause an isolation loss of 25 dB. Most hybrid couplers, however, have built-in cimatch the 1.5:1 VSWR, thereby maintaining at least a isolation.

The main advantage of a ferrite combiner is its aisolate and combine any channel assignments within quency band. Its relatively small size is another advantafeature. Multiple channel combiners can be packaged in adard 19-inch rack.

H

H H

H H H H

Insert ion Loss 11.0 dB 8-Transmitter Comb


49/130


The Receiver Mul ticoupler or Comb in ersNormally, each transmitter, unless it is for paging on

a receiver associated with it. If the transmitters are cominto a single antenna, then logically all the receivers shocoupled from a common antenna.

Usually, the receiver multicoupler for close frequenc

ing will consist of a 2, 4, 8, 16, etc., hybrid splitter amplifier.The critical aspects of a receiver multicoupler are th

gent requirements for an active device that has enoughto recover the splitting losses and yet has an extremely gain characteristic. Otherwise, a non-linear mixing spresent to generate receiver intermodulation which couldthe desired signals.

A Com pl ete System Confi gur ati onFigure 2-15 depicts a two-channel hybrid-type co

that is operating through a duplexer with a two-port rmulticoupler. The criterion with this configuration is duplexer should have a bandwidth that will accomthe required frequencies.

For the combiner to operate efficiently, the transmiquencies must be spaced close enough to allow the ntype duplexer to adequately isolate the transmitters aceivers. That is, the notch width of the duplexer at

Antenna

Notch-Type Duplexer

Hybrid Receiver Multicoupler


50/130

COMB

quired isolation will dictate the maximum transmitter frseparation. This criterion also applies to the receiver frecies.

In many systems such as the RCC, taxi, and mobiphone channels bandwidths up to 600 KHz are needcombine all or most of the available channels. In cases su

these, the low Q, rack mount, notch-type duplexers are nein the combiner because their wide notch widths allequate isolation across the entire frequency band. Abandpass duplexer would not be practical since a relbroad band of frequencies are to be passed throuduplexer.

Com bi ner SelectionThe optimum type of combiner for a particular sys

pends upon separation of frequencies, antenna system the number of transmitters and their power output, theber of receivers (if any), the proximity to other systemother local factors. A combiner should be designed tothe requirements of a specific system.


51/130



52/130

Select ive Cavit

As the demand for land mobile radio services steacreases, the problems caused by frequency congestion, recdesensitization and intermodulation, grow rapidly. Thetive cavity will help in solving these problems. In this we will take a look at selective cavities, how they whow they can be used.

What is a Selective Cavi ty an d Wha t Does it Do?A selective cavity is a rather simple device that serv

filter for radio frequencies. It has the ability to let a band of frequencies pass through while frequencies othis narrow band are attenuated. Stated differently, thwanted and unselected frequencies are rejected and fiout while the desired frequency is passed through witslight attenuation. The narrow band of frequencies thathrough the cavity are within a few thousand hertzcavitys resonant frequency.

The selective cavity, with this filtering action, is in the land mobile radio services as more and more newtions are crowded into the same area. As new stations the air, they can, and frequently do, cause interference to stations that exist in the immediate area. In addition


53/130


transmitter noise), and intermodulation. Without goilengthy discussions of these two problems, let it be sareceiver desensitization occurs when a nearby transmitter powers a receiver. Intermodulation occurs when two onearby transmitters mix within the RF stages of a and generate new frequencies, with one of the new frecies being the same as the receiver frequency.

In both cases, a selective cavity can be used to helpthe problem. In the case of transmitter noise, the cavity cused at the transmitter to reduce transmitter noise sidebMore specifically, it will increase receiver selectivity anthe receiver less sensitive to nearby transmitters. In the caintermodulation, the cavity will filter out the unwantemissions, thus keeping them from reaching the receiver

Input Connector

Tuning Knob Screw

Output Connector

Outer Conductor (Shell)

Coupl

Inner (fixed

Inner(

External View


54/130

SELECTIVE CAV

How it Work sAs RF energy is fed into the cavity by means of a c

loop, the energy excites the resonant circuit formedcavitys inner and outer conductors; the other loop cenergy from the resonant circuit to the output. The rescircuit is formed by the inner conductor and an equal

of the outer conductor. To change the resonant frequenthe cavity, the length of the inner conductor is changedally this is done by making a portion of the inner comovable and attaching it to a screw which can be turnetuning knob. The coupling loops dont affect the resonaquency but they do help determine the selectivity of thity.

SelectivitySelectivity is the ability of a cavity to select one fr

and reject the rest. The ideal selective cavity would hresponse curve like that shown in Figure 3-2. (The frequare shown along the horizontal axis of the chart and atttion values are shown along the vertical axis.)

Attenuation 0 dB

-10 dB

-20 dB

-30 dB

-40 dB

-50 dB

d


55/130


According to Figure 3-2, a very narrow band of cies passes through the cavity with very little attenuationwe need a little more room for the modulation, we narrow band of frequencies to pass through instead ofing it to a single frequency. This narrow band of freq(centered on f 0 in Figure 3-2), represents the frequency the cavity is tuned; this is referred to as the cavitys resfrequency. All other frequencies outside this narrow bandget through at all just what we need for real selUnfortunately, no one has ever been able to build thefect cavity that is, one that can be tuned to pass tremely narrow band but, excellent results can be obwith the technology that is available.

Figure 3-3 shows the response curve for a typical The chart is arranged as before: the frequencies are shalong the horizontal axis (with the resonant frequencysented as f 0 in the middle) and the attenuation vashown along the vertical axis.

For the frequency to which the cavity is tuned (f a little bit above and below this frequency, there is ve

Attenuation 0 dB

-10 dB

-20 dB

-30 dB

-40 dB

-50 dB

60 dB


56/130

SELECTIVE CAV

Attenuation 0 dB

-10 dB

-20 dB

-30 dB

-40 dB

-50 dB

60 dB

attenuation, only about a half of a dB. (A half of a power reduction, or attenuation, of about 11 percent.

For frequencies outside of the narrow band, hothings get more difficult. For frequencies 0.6 megahertkilohertz) above or below f 0 the attenuation is 10 dtenuation of 10 dB is a reduction of 10 to 1.) For frequ2 megahertz from f 0 , the attenuation is 20 dB, a red100 to 1. And at 5 MHz from f 0 , the attenuation is reduction of 1000 to 1. To put it another way, a signal5 MHz away from f 0 will only be 1/1000 as stronleaves the cavity as it was when it entered.

The attenuation of half a dB at the center frequenccates what is referred to as a tightly coupled cavity. ConvFigure 3-4 shows the response curve of a loosely coupledity. Here the f 0 attenuation is 3 dB (a 2 to 1 reductnotice how much steeper the sides of the response curvand how much lower they go before they start to flare only 0.4 MHz (400 KHz) away from f 0 the attenuatiAt 1 MHz away its 30 dB and at 5 MHz away the atteis 50 dB a reduction of 100,000 to 1.


57/130


Inserti on LossIf we compare Figures 3-3 and 3-4, we can say tha

ening the coupling increases the selectivity but wcreased attenuation of the desired frequency. This attenuat f 0 , the desired frequency, is referred to as insertion losobjective when using a cavity is to obtain the optimum nation of selectivity and insertion loss that is, havienough selectivity to solve the interference problemkeeping insertion loss at a minimum.

How to Obt ai n Selectivi tyThe coupling loops, or the degree of coupling, are

that govern the selectivity of a cavity or that make onedesign better than another. The other major factors arvolume of the cavity, the internal RF losses, and the freqof operation.

The volume of the cavity depends, of course, on itsand diameter. For proper operation, a cavity must be amum of 1/4 wavelength long, electrically. This workbe about seven inches long at 450 MHz, about 20 inch150 MHz, and about six feet at 30 MHz. There are wmake cavities shorter, however, by using end-loading ocal center conductors. Also, if preferred, a cavity can wavelength long, electrically. This becomes practical foing a 450 MHz cavity the same length as a 150 MHz c

Since the length is fixed we can increase the volthe cavity only by increasing its diameter. But, beyond aeter of about 1/4 to 1/3 the length of the cavity, we gailittle increase in performance. In addition, as the sizecavity increases, so does the cost and the inconvenienhandling. As a result, the diameter of most cavities is ty1/4 to 1/3 their length.

Int ern al LossesThe internal losses are determined by the kinds of m

als used in the cavity, with materials having the lowes


58/130

SELECTIVE CAV

difference at frequencies above 500 MHz. The most imprequirement if we use copper or aluminum is to keepfrom corroding. This can be accomplished by silver and/or lacquer-coating them.

Determine the Q of the CavityThe volume of the cavity and the internal losses det

the Q of the cavity. The greater the volume and/or ththe internal losses, the higher the Q of the cavity. A his preferred over a low Q because the higher the more selective the cavity. (There are actually two Q Unloaded Q and Loaded Q. These are not impoour discussion here; however, they may be importantcomparing cavity specifications.)

Theres not much we can do about the cavitys freof operation since that depends upon the system in the cavity will be operating. Unfortunately, the higher quency the more difficult it is to obtain selectivity. This ia rule that says: bandwidth, or selectivity, is directly tional to frequency and inversely proportional to thethe cavity. Algebraically stated this would be: BandwFrequency divided by Q.

We can go up in frequency rather easily, but we ccrease the Q as quickly. So, its more difficult to be selechigh frequencies than at low frequencies.

Obta ini ng Gr eater SelectivityLets consider what can be done when a cavity wo

vide enough selectivity without excessive insertion loexample, to solve an interference problem caused by a that is 2 MHz off from our receiver frequency, it mus

duced by 32 dB before it reaches our receiver. Figure 3-5 sus that one cavity with 1/2 dB loops (or a 1/2 dB iloss) will provide only 16 dB of attenuation to the insignal. And, at 1/2 dB, were losing about 11 percentsignal power inside the cavity.


59/130


power inside the cavity.There is another way to obtain the attenuation w

without losing half of our signal power. If we put twoin series, each with 1/2 dB loops, the unwanted signal attenuated by 16 dB in each cavity for a total of 32 dbecause the desired signal is attenuated by only 1 dB (1in each cavity), we only give up 20 percent of our signal Obviously, thats much better than the 50 percent loss wif we use the 3 dB loops.

Simply stated, as we can see from this example, ibetter to obtain the desired attenuation using several coupled cavities instead of one loosely coupled cavity. Tha limit to this procedure, however. Because of other factousually not practical to use more than three cavities inand, its almost never practical to use more than four caviseries.

Attenuation 0 dB

-10 dB

-20 dB

-30 dB

-40 dB

-50 dB

-60 dB

-70 dB 4 3 2 1 f 0 1 2 3 4

Figure 3-5: Cavity with 1/2 dB Loops


60/130

SELECTIVE CAV

a single frequency (there are typically low-power noisbands) we can use cavities to sharpen the output of a tmitter.

Attenuation 0 dB

-10 dB

-20 dB

-30 dB

-40 dB

-50 dB

-60 dB

-70 dB 4 3 2 1 f 0 1 2 3 4

Figure 3-6: Cavity with 3 dB Loops

When the cavity is in series with the transmitter wto be concerned with the amount of power the cavhandle. With 1/2 dB loops in the cavity, about 11 perthe power will be lost in the cavity and this lost powbe dissipated by the cavity. For example, with a 100-wattmitter the cavity will have to dissipate 11 watts, or withwatt transmitter it will have to dissipate 28 watts.

Dissipating this power doesnt present much of a pto a typical cavity with its large metal conducting surfaif we consider a 3 dB loop cavity being used with a 3transmitter, a 3 dB insertion loss means that 50 percent transmitter power, or 175 watts, must be dissipated cavity. This is more than most cavities even with themetal conducting surface can handle.


61/130


Mor e About Coupl ing LoopsSo far, we have discussed using cavities with various

tion losses. We have also stated that insertion losses, or pling loop losses, and selectivity vary with the degree opling. Loose or tight coupling can be obtained in seveferent ways.

Some cavities have permanent coupling loops. Incavities, the only way we can get different selectivity teristics and different insertion losses would be to changthe cavities. On the other hand, some cavity designsreplaceable coupling loops. In these designs, if we wanchange from 1 dB loops to 3 dB loops, for example, wesimply remove a few screws, take out the old loops, putnew ones, and replace the screws. Ordinarily, these repable loops come in several sizes 1/2 dB, 1 d3 dB. Since its usually difficult to figure out in adcisely how much selectivity will be required, its often hhave different sizes of loops available when trying to particular interference problem. The overall goal, of should be to use the lowest insertion loss possible ansolve the problem by rejecting the interfering signal.

The next alternative is to use cavities with variabpling loops. In this type of cavity, the loops are not chInstead, they are moved closer to or further away frocenter conductor. This movement is usually accomplishrotating the loops. For tighter coupling they are movedto the center conductor; for looser coupling they are away from the center conductor. With this method it sible to obtain the lowest insertion loss and still obtrequired selectivity to solve the problem. This also elimthe need to use different sized loops.

We should mention that the coupling loops musthe right size and shape to match the impedance of thenecting coaxial cables. The cavity manufacturer takes cthis detail, however.


62/130

SELECTIVE CAV

going to be more of a hindrance than a help. When tuna particular frequency, the cavity must stay tuned to thquency.

Mechanical stability merely requires an application omechanical engineering principles with regard to sizes, shand materials. Stability in spite of temperature change rethat the length of the inner conductor remain constvery nearly so, when the temperature varies. Expanscontraction of the outer conductor isnt of great conceronly that part of the outer conductor equal in lengthinner conductor is in the resonant circuit. The excess the outer conductor is actually not being used electricallyif the diameter of the cavity changes because of expanscontraction, it only changes the volume of the cavity only slightly at that so just the Q of the cavity cnot the resonant frequency.

To keep the length of the inner conductor constant, ais needed that doesnt expand or contract with variatitemperature since the movable part of the center conduattached to the tuning screw. This problem can be easily sif the metal used for the tuning screw is one that isnt aby temperature changes.

Fortunately, a nickel-steel alloy known as invar, wh

a low temperature coefficient of linear expansion, can beThis means that if a cavitys inner conductor is made othere will be little expansion of the alloy with changes perature. (To be more precise, a 2-foot length of invar wchange by 0.00145 inches in length with a tempechange of 100 degrees Fahrenheit.)

Cavi ty Install ati on

Typically, the installation of a cavity is quite easy scavity can be mounted in any position. The biggest conation is to leave room to access the tuning knob.

The cavity is placed in the circuit by means of coaxiand the appropriate connectors. And, since theres no


63/130


TuningIf the cavity is to be used with a transmitter, a wa

should be used on the output side of the cavity; the should then be tuned for a maximum power reading.

In tuning the cavity, when it is used in series with mitter, its necessary that a load of some kind either tenna or a dummy load be connected to the outputerwise, the cavity will have to dissipate all of the tranpower which will probably result in damage to the cavtransmitter, of course, should be set to the tune powetion for the initial cavity adjustments.

If the cavity is being used in series with a receiver, thest way to tune it is to have a low level, on-frequencyfrom the antenna. Then adjust both the cavity and tceiver antenna coil for maximum limiter current. Yostart with a signal generator but be sure to touch ucavity and receiver antenna coil after the antenna has connected.

A selective cavity is both highly useful and easy-to-uits the solution to many troublesome interference prob


64/130

Duple

Consider for a moment how difficult an ordinarysation would be if our telephone systems were simpletems. With a simplex system, you talk to the other partylisten, then talk again; but you cant talk and listen at thetime. So, if you miss the first part of the other personsage, you must wait until he finishes talking before yask him to repeat it. This built-in delay slows the process and makes an exchange of information more di

Fortunately, we arent confronted with this probcause our telephone systems are duplex systems. This mwe can carry on a normal conversation with each partytalking back and forth.

But what about our radio systems? If a duplex sysuperior to a simplex system, why are most of our radtems simplex? There are several reasons, the most commwhich are:

(1) A duplex system requires the use of two freq

(one for transmit, one for receive) and only certain radiocan obtain authorization for use of a second frequency

(2) A duplex system requires additional equipmentsystem is typically more costly than a simplex system.

(3) With a duplex system, certain unique technic


65/130


The Sim plex SystemThe typical simplex system consists of a transmit

receiver operating on the same frequency (see FigureSince you cant transmit and receive at the same timantenna change-over relay (located within the radswitches the antenna to the transmitter whenever the mphone button is depressed; then, it switches to the rewhenever the microphone button is released.

If the transmitter and receiver shown in Figure 4both connected directly to the antenna system, bypassinantenna change-over relay, the receiver would be severelsensitized and its performance degraded. In addition, it probably be physically damaged by the output powertransmitter. And, not only would the receiver suffer, themitter would be less efficient since a portion of its power would go directly into the receiver.

The Dup lex SystemThere are numerous types of

duplex systems, but they all have one thing in common: the transmitter operates on one frequency and the receiver operates on another.The duplex system generally is installed to accomplish one or more of the following goals:

(1) to replace the wire-line control circuit betwbase station and the remote control

Documents

Rf Communications