0% found this document useful (0 votes)

27 views

Radio and Radar Principles

Uploaded by

Cyprian Obota

Available Formats

Download as DOC, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

27 views

Radio and Radar Principles

Uploaded by

Cyprian Obota

Available Formats

Download as DOC, PDF, TXT or read online on Scribd

You are on page 1/ 324

AIR

SERVICE
TRAINING
(ENGINEERING)
LIMITED
A Subsidiary of Perth College

RADIO AND RADAR PRINCIPLES

Part 66 – B2/015a

PERTH COLLEGE
BRAHAN BUILDING
CRIEFF ROAD
PERTH PH2 1NX
TEL: 01738 877105
FAX: 01738 553369
©A IR SERVICE TRAINING (ENGINEERING) LTD

Aeronautical Engineering Training Notes

These training notes have been issued to you on the understanding that they
are intended for your guidance, to enable you to assimilate classroom and
workshop lessons and for self-study. Although every care has been taken to
ensure that the training notes are current at the time of issue, no amendments
will be forwarded to you once your training course is completed. It must be
emphasised that these training notes do not in any way constitute an
authorised document for use in aircraft maintenance.

All Rights Reserved

The copyright in these technical training notes remain the physical and
intellectual property of Air Service Training (Engineering) Ltd, (AST).
Copying, storing in hard copy or electronic format, transmission to third parties
and use for teaching by establishments other than AST is forbidden, except
with the written permission of the AST General Manager.

M Haufe
Training Manager November 2006
© Air Service Training (Engineering) Limited
Part 66 – B2/015a Radio and Radar Principles

Contents
Page

CHAPTER 1:

Issued November 2006 Contents 1

2 Contents Issued November 2002

CHAPTER 1 : RADIO FUNDAMENTALS

SECTION 1 : PROPAGATION OF RADIO WAVES
TRANSVERSE ELECTROMAGNETIC WAVE (TEM)
Radio waves, propagating in an unbounded medium, are transverse electromagnetic
waves. They consist of oscillating electric (E) and magnetic (H) fields, which are at
right angles to each other and to the direction of propagation. A stationary E field
can exist in space, without an accompanying H field, for example in the space
surrounding a static charge. Similarly, a stationary H field can exist without an E
field, for example in the space surrounding a permanent magnet. It is not possible
for changing E and H fields to exist in isolation. A changing E field is associated with
a changing H field and vice versa.
Radio waves are generated in the space surrounding a conductor carrying a varying
current. The E and H field components are mutually supporting. If the current is
sinusoidal, the E and H fields vary sinusoidally, with a rotation of E following H this
would give a direction of propagation following a corkscrew rule and the E and H
field ripples, which radiate away from the source, constitute the radio wave. When
the radio wave falls upon another conductor, a current is induced, which is a time
delayed replica of the current which generated the wave.
An EM wave is made up of an electric (E) field and a magnetic (H) field as shown,
such that:
Both fields vary sinusoidally with time and are of the same frequency and phase.
The fields are at 900 to the direction of propagation.
The plane of polarisation refers to the E field direction.

FIG 1 TRANSVERSE ELECTROMAGNETIC WAVE (TEM)

Issued November 2006 Chapter 1 : Section 1 1

VELOCITY OF ELECTRO-MAGNETIC RADIATION

An electro-magnetic wave is not supported by a medium and can therefore travel
through a vacuum. The speed in a vacuum is a constant c = 3 x 108 metres/sec.
The speed in free air is very nearly equal to c but the velocity is determined by the
nature of the medium.
and metres/sec in free space
These electric and magnetic fields are confined not just close to the wire but radiate
outwards in all directions as electro magnetic (EM) waves. Low frequencies do not
travel very far but higher frequencies above the AF (Audio Frequency) max of 20
kHz ie. RF (Radio Frequencies) will travel many thousand of miles if the frequency
and power are high enough.
The electric field will travel parallel to the wire and the magnetic field will travel out at
right angles to the wire or aerial as we will now call it. Both E and H fields must be
present in the electro-magnetic radio frequency wave.
Frequencies from 20 kHz to 300,000 Mhz (300 GHz) are used for radio and radar
systems. Radio waves like light can travel through empty space at a speed of 300 x
106 metres/sec or 186,000 miles/sec.
Velocity of a wave = frequency x wavelength

Thus a radio wave of 1 Mhz has a wavelength of:  = 300 metres

Examples:

BBC Radio 4 wavelength 1500 metres has a frequency of: Khz

A VHF broadcast of 90 Mhz has a wavelength of:  = 3.33 metres

A radar installation working with a wavelength of 10 cm has a frequency of:

= 3000 Mhz
= 3 Ghz

2 Chapter 1 : Section 1 Issued November 2006

THE RADIO FREQUENCY SPECTRUM

Band Frequency Wavelength Typical Use Method of
Propagation
VLF – Less than 30 kHz Greater than World wide Ground wave
Very low 10,000m telegraphy, only
frequency submarine
communications
LF – Low 30 - 300 kHz 10,000 – Broadcasting, Ground wave
frequency 1,000m DF beacons for up to some 400
use with shipping miles
MF – 300 kHz – 3 MHz 1,000 – 100m General Ground wave
Medium broadcasting up to some 300
frequency miles. Some
sky wave at HF
end of band at
night-time
HF – High 3 – 30 MHz 100 – 10m World wide Sky wave.
frequency telephony, Ground wave
telegraphy for some 40
miles only
VHF – 30 - 300 MHz 10 – 1m Short range R/T ‘Line of sight’
Very high Television Radar
frequency (Long range)
Police, Fire and
Ambulance
services, etc.
UHF – 300 MHz – 3 GHz 1m – 10cm Television Radar ‘Line of sight’
Ultra high Short Range R/T
G = Giga
frequency Navigational aids
= 1000 MHz
SHF – 3 - 30 GHz 10 cm – 1 cm Precision radar ‘Line of sight’
Super Microwave links
high Satellite
frequency communications
EHF – 30 - 300 GHz 1 cm – 1mm High definition Less than line
Extremely radar (Short of sight due to
high Range) atmospheric
frequency attenuation

FIG 2 RADIO FREQUENCY SPECTRUM

Issued November 2006 Chapter 1 : Section 1 3

THEORY OF PROPAGATION – AN IDEAL RADIATOR

A useful concept when studying the behaviour of radio waves is the isotropic radiator
which is a hypothetical, truly omnidirectional source of Radio Frequency (RF). The
radiation field of such a radiator suspended in free space, ie. unaffected by the
proximity of the earth and its atmosphere, may be represented by an expanding
sphere (Fig 3).

FIG 3 THE ISOTROPIC RADIATOR

As the wave expands, its power is distributed over an increasing area. If the
strength of the signal is such that 1 watt is distributed over an area of 1 square metre
at range D, then a doubling of the distance would produce a power density of ¼ watt.
Put another way, four times the power is needed to double the range and the energy
distribution is said to follow the inverse square law. This attenuation, or weakening
of the signal with increasing range is called the Free Space Loss, and will affect all
forms of radiation. There is in practice no such thing as a truly omnidirectional
radiator, all practical aerials display some directional properties, and hence some
gain.

OVERVIEW OF RADIO WAVE PROPAGATION

When radio waves travel in proximity to the earth, and through the atmosphere,
additional attenuation factors must be considered. The behaviour of the radio waves
may be best studied by grouping them by mode of propagation. A convenient
grouping is between those waves that make use of the layers of ionised gases
surrounding the earth, and those that do not.

4 Chapter 1 : Section 1 Issued November 2006

FIG 4 SUMMARY OF PROPAGATION METHODS

Issued November 2006 Chapter 1 : Section 1 5

IDEAL RADIATION PATTERN OF A HALF-WAVE DIPOLE

A simple vertical aerial radiates RF energy in all directions except from its ends
(Fig 5).

FIG 5 IDEAL RADIATION PATTERN OF A HALF-WAVE DIPOLE

Some of the energy travels parallel to the ground and is called a surface wave, and
some is radiated upwards in the form of sky waves or space waves. The lower the
frequency the more easily will the waves follow the earth’s curvature.

DIFFRACTION
The process of diffraction causes bending of rays of light over and around obstacles,
and radio waves near the surface will similarly bend around an obstacle – in this
case the earth itself (Fig 6). Diffraction increases as frequency is reduced.

FIG 6 SURFACE AND SKY WAVES

As the wave front moves away from the aerial there is a loss of energy to the ground
below. This causes that part of the wave front in contact with the surface to be
retarded, the wave front is tilted forward assisting diffractive bending. (Fig 7).

FIG 7 DIFFRACTION OF WAVE FRONT

6 Chapter 1 : Section 1 Issued November 2006

The practical radiation pattern of the half-wave dipole, used near the earth’s surface,
will be modified from the ideal, reflections producing secondary lobes as shown
(Fig 8).

FIG 8 RADIATION PATTERN FROM A REAL AERIAL (VERTICAL PLANE)

GROUNDWAVES
Strictly, the groundwave consists of surface wave and the space wave (to be
discussed later), but for most practical purposes surface wave and groundwave are
synonymous. The groundwave is the primary mode for frequencies below 2 MHz
and down to 20 kHz. A great benefit of ground wave propagation, unlike sky wave,
is that it is little affected by seasonal or diurnal variations. Possible disadvantages
are the requirement for very powerful transmitters, large aerials, and the inability to
beam the energy effectively, as with higher frequencies. The range achieved by the
groundwave depends on a number of factors.

FREQUENCY
Shorter wavelengths imply more penetrations of the underlying terrain, each
penetration contributes to the overall attenuation. (Fig 9).

FIG 9 EFFECT OF FREQUENCIES ON ATTENUATION

POLARISATION
At the frequencies associated with the ground wave, horizontally polarised signals
suffer severe attenuation, so vertical polarisation is usually employed.

Issued November 2006 Chapter 1 : Section 1 7

TYPE OF SURFACE
Surfaces – High conductivity favours the passage of the signal, whilst rock and
desert surfaces, having low conductivity, cause marked losses (Fig 10).

FIG 10 RANGE OF GROUND WAVES OF 1KW EFFECTIVE RADIATED POWER FOR SENSITIVE COMMUNICATION
RECEIVER

A formula closely approximating to the range of the space wave in nautical miles is:

Where = Height of transmitter in feet. (amsl)

= Height of receiver in feet (amsl)
= Range in nautical miles
Thus an aircraft at 10,000 feet may expect to receive a station at 400 feet at a range
of:
= 150 nautical miles
This limited range is of benefit since it allows facilities to use the same radio
frequencies without interference, provided they are sufficiently far apart. For
instance television stations in Scotland, London and Spain may all occupy the same
channel, and under normal conditions there will be no interference.
Probably the greatest benefit of those frequencies primarily propagated by space
wave, is that they are little affected by time of day, weather or noise.

8 Chapter 1 : Section 1 Issued November 2006

TRANSMITTER POWER
Both free-space losses, and surface absorption will reduce the signal strength, but
providing there is no restriction on available power then global ranges are possible
using the VLF band.

RECEIVER SENSITIVITY
This is really no longer an issue since most modern receivers will detect signals until
the point at which their strength has diminished to, or below the level of noise.

NOISE AND INTERFERENCE

As noise level increases, the signal to noise ratio will be reduced, so limiting the
usable range. It is worth noting that the lower frequencies are most affected by
noise.
In summary – for maximum ground wave range – use low frequency (for maximum
diffraction, and least attenuation), and vertical polarisation.

SPACE WAVES
At frequencies above about 50 MHz, attenuation is so severe that the surface wave
is virtually non-existent, also frequencies are too high to be refracted by the ionised
layers aloft. Transmission on these frequencies is by waves travelling in straight
lines – direct waves.
In addition there may also be present a reflected wave (Fig 11).

FIG 11 THE TWO COMPONENTS OF THE SPACE WAVE

The combination of direct wave and reflected wave is called the space Wave.

Issued November 2006 Chapter 1 : Section 1 9

Emissions following different paths will sometimes be in phase and sometimes out of
phase, producing lobes and nulls, particularly close to the station where the
difference in path length is most marked (Fig 12).

FIG 12 EFFECT ON COMBINATION OF DIRECT AND INDIRECT WAVES

The range of the space wave would appear to be line of sight, and hence determined
by the height of the aerials (Fig 13).

FIG 13 TANGENTIAL RANGE

In practice the lower layers of the atmosphere cause refraction of the wave, bending
it beyond the optical horizon (Fig 14).

FIG 14 RANGE OF THE REFRACTED WAVE

10 Chapter 1 : Section 1 Issued November 2006

A further small increase in range will result from diffraction when the wave is
tangential to the earth’s surface (Fig 15).

FIG 15 EFFECT OF DIFFRACTION ON SPACE WAVE RANGE

The final range is described as Quasi-Optical.

ATTENUATION OF THE SPACE WAVE

In addition to the free space loss there will be further loss due to:
 Atmospheric absorption
 Water droplets causing absorption and scattering.

EFFECT OF POWER ON SPACE WAVE RANGE

Since range is limited to the quasi-optical, increasing power has little effect. An
aircraft VHF transmitter produces about 25 watts of power to the aerial, and this is
adequate for ranges of about 250 miles at jet cruising levels. Radar equipment that
relies on reflected energy from the target will of course benefit from high power.

SKYWAVES AND THE IONOSPHERE

So far, frequencies below 2 MHz and above 50 MHz have been examined. In each
case very high power was required to achieve significant ranges. It is possible using
frequencies between these limits to communicate over considerable distances
employing only modest power levels by making use of the ionosphere.

Issued November 2006 Chapter 1 : Section 1 11

IONISATION PROCESS
In the high atmosphere there is little mixing and the constituent gases tend to occupy
separate levels depending on their mass.
Radiation from the sun covers a wide spectrum of frequencies, but principally around
the ultra-violet, radiation causes electrons to be freed from their parent molecules
producing positively charged ions and free electrons. The overall effect is for a
number of discrete regions, or layers of ionisation to be formed as shown. These
layers are designated D, E, and F in order of height and are collectively known as
the ionosphere, the F layer being of the greatest intensity (Fig 16).

FIG 16 IONISATION

12 Chapter 1 : Section 1 Issued November 2006

The actual heights and intensity of ionisation depend primarily on time of day,
season and latitude, and on activity on the sun’s surface, measured by visible spots
on the solar disc. This activity has been observed to follow an eleven year cycle and
is to an extent predictable (Fig 17).

FIG 17 SUNSPOT ACTIVITY 1750 TO 1940

There are also random occurrences on the sun’s surface in the form of solar flares
that disturb the normal pattern of ionisation. These cause especially intense
ionisation of the D-region, lasting about half an hour. Obviously only the sun-lit side
of the earth will be affected. The solar flares as well as increasing radiation may also
emit particles that concentrate in the earth’s magnetic field, bringing disruption of the
ionosphere particularly in polar regions, accompanied by visible aurora. This latter
effect may persist for several days. The solar disc rotates over approximately 27
days, and the effects may be repeated after this interval, albeit at reduced intensity.

DIURNAL VARIATION
In the higher levels of the atmosphere, ionisation persists throughout the night. The
reason for this is the very low density, making recombination of free electron and
positive ion difficult. Nearer the surface the reverse is true and ionisation quickly
disappears when the sun sets (Fig 18).

FIG 18,19 DIURNAL VARIATION OF THE IONOSPHERE

Issued November 2006 Chapter 1 : Section 1 13

SEASONAL VARIATION
Broadly speaking, since ionisation intensity is dependent on radiation from the sun, it
may be expected that maximum intensity will occur in summer.
The F2 layer is noticeably higher during summer (Fig 20).

FIG 20 SEASONAL VARIATION OF THE IONOSPHERE

The effect of the decrease in insulation causes dispersal of the D region by night,
and the two distinct F1 and F2 layers combine to form a single F layer. Dawn and
dusk, as may be inferred, are periods of Ionospheric instability.

GEOGRAPHICAL VARIATION

FIG 21 THE EFFECT OF LATITUDE ON IONISATION

At high latitudes the radiation is spread over a greater area than in the equatorial
regions, producing less ionisation as the poles are approached (Fig 21).

14 Chapter 1 : Section 1 Issued November 2006

EFFECTS OF IONISATION ON RADIO WAVES

When a radio wave passes through an ionised region it experiences attenuation and
a change of velocity. Lower frequencies suffer the severest attenuation, so there is
an incentive to use a higher frequency to overcome this difficulty. There is however
a problem involved with the use of too high a frequency. To be explained shortly,
attenuation is most apparent at low levels in the atmosphere, and refraction at the
higher levels.
As the waves pass through the layers their plane of polarisation may be somewhat
changed. This may be of importance when satellite communications are discussed.
The effect is most marked at VHF frequencies and above.

FIG 22 WAVE REFRACTION

REFRACTION
When a ray of light passes through glass or water, the ray is bent towards the normal
as it is retarded in passage through the denser medium.
A radio wave entering an ionised layer accelerates and bends away from the normal.
Under some circumstances the bending is sufficient to return the wave towards the
earth’s surface (Fig 22).

Issued November 2006 Chapter 1 : Section 1 15

It is for all practical purposes as if the wave had been reflected by a mirror type
surface. The height of the surface is described as the natural height of reflection and
the wave may often be spoken of as being reflected, but is in fact refracted. (Fig 24).

FIG 23 REFRACTION SIMPLIFIED

Such total refraction depends on a number of variables.

THICKNESS OF LAYER AND IONISATION INTENSITY

If the wave penetrates half-way through the layer before being bent parallel to the
layer it will bend back in the opposite direction to emerge from the top of the layer it
will bend back in the opposite direction to emerge from the top of the layer as an
Escape Ray (Fig 24).

16 Chapter 1 : Section 1 Issued November 2006

FIG 24 ESCAPING AND REFRACTING RAYS

This is likely if ionisation is insufficiently intense.

FREQUENCY
The higher the frequency the less is the refraction.

ANGLE OF INCIDENCE
If the angle of entry is too acute the waves will be insufficiently refracted to return to
the surface.

Issued November 2006 Chapter 1 : Section 1 17

CRITICAL ANGLE
For one particular frequency, and degree of ionisation, it is possible to define a
critical angle below which total refraction will not take place. Establishing this critical
angle will also define a critical ray, and an associated minimum range called the skip
distance. Rays travelling away from the aerial at greater than the critical angle will
be freely refracted down to about 5 above the horizon, at which point ground losses
become excessive.
Another area can be identified called the Dead Space between the limit of ground
wave reception and the first returning skywave. (Fig 25).

FIG 25 CRITICAL ANGLE, SKIP DISTANCE AND DEAD SPACE

18 Chapter 1 : Section 1 Issued November 2006

CRITICAL FREQUENCY
If a radio signal is directed vertically upwards at a frequency of 50 MHz it will almost
certainly penetrate all the ionised layers. If the frequency is progressively reduced,
eventually a signal will be returned to the surface (Fig 26). This frequency is defined
as the Critical Frequency for that particular layer. Obviously the higher the frequency
the greater is the level of ionisation. If the time interval between transmission and
reception can be measured, the height of the layer can be calculated.

FIG 26 CRITICAL FREQUENCY

FREQUENCY AND SKIP DISTANCE

At a fixed level of ionisation, an increase in frequency will cause the ray previously
the critical ray, to become an escape ray, and as can be seen this will increase the
skip distance (Fig 27).

Issued November 2006 Chapter 1 : Section 1 19

FIG 27 EFFECT OF FREQUENCIES ON SKIP DISTANCE

IONISATION AND SKIP DISTANCE

At a fixed frequency if ionisation decreases, the effect will be the same. What was
the critical ray will now penetrate the layer and skip distance will increase (Fig 28).

FIG 28 EFFECT OF CHANGING IONISATION ON SKIP DISTANCE

MAXIMUM USABLE FREQUENCY (MUF)

For a given angle of incidence appropriate to a communication link between two
points, and using a particular layer, there will be a maximum useable frequency
(MUF) (Fig 29).
This frequency is equal to the critical frequency multiplied by the secant of the angle
of incidence.

20 Chapter 1 : Section 1 Issued November 2006

FIG 29 THE MAXIMUM USABLE FREQUENCY (MUF)

There are a number of radio observatories established around the world, the function
of which is to monitor Ionospheric conditions. Mathematical models may be devised
based on the variables so far discussed and on historic information, and these
models may be used to produce forecasts of MUF for particular paths.
Examples of forecast MUF’s show pronounced peaks roughly aligned with the time
of local noon at the point of refraction (Fig. 30).

FIG 30 MUFS AT SUNSPOT MAXIMUM AND MINIMUM

Although the requirement exists for the use of as high a frequency as possible to
overcome absorption losses, the use of the MUF itself is not satisfactory.
Any reduction in ionisation intensity or error in forecasting would place the receiver
within the skip distance – result – no communication.

Issued November 2006 Chapter 1 : Section 1 21

USE OF OPTIMUM WORKING FREQUENCY (OWF)

It is usual to operate on a frequency equal to 85% of the MUF to ensure continuous
communication, and this frequency is described as the optimum working frequency
(OWF) (Fig 31).

FIG 31 MUFS FOR THE LONDON –NEW YORK CIRCUIT SHOWING OWF ADDED

International radio regulatory authorities assign various small parts of the total RF
spectrum for authorised uses, such as aviation. The aeronautical agencies in turn
assign various spot frequencies within these sub-bands for use on a regional basis.
A listing for the North Atlantic reveals the availability of the following frequencies (all
in KHz) 2899, 5616, 8864, 13281 and 17946 for use by various air traffic service
units within the region.

FIG 32 MUFS FOR THE LONDON-NEW YORK CIRCUIT SHOWING OWF ADDED ALSO BEST FIT OF AVAILABLE
FREQUENCIES

22 Chapter 1 : Section 1 Issued November 2006

The combination of forecast MUF’s and available spot frequencies enables the
selection of the most appropriate channel for the time of day (Fig 32).
There are a number of providers of HF communication services for purposes other
than air traffic control. One example of such, is Portishead Radio operated by British
Telecom.
Portishead produce a forecast for 2 monthly periods, recommending the use of
particular frequencies for communication with Portishead from various geographical
locations (Fig 33).
Phone-patch facilities are available for national or international connection,
permitting for example conversation between flight deck and airline operations office.

FIG 33 MUFS FOR PHONE-PATCH

Issued November 2006 Chapter 1 : Section 1 23

MEDIUM FREQUENCIES AND THE IONOSPHERE

Daylight Conditions

FIG 34 D REGION ATTENUATION OF MF BY DAY

During the day the presence of the D region precludes reception of skywaves at
medium frequencies.
This means that broadcast stations are unlikely to be heard beyond 200 to 300
nautical miles during daylight hours.

Night Time

FIG 35 MF SKY WAVE AT NIGHT

By night in the absence of the D region, distant stations may cause interference to
one another, and equally troublesome, is mutual interference of sky wave and
groundwave from the same source (Fig 35).

24 Chapter 1 : Section 1 Issued November 2006

FADING
In almost all forms of radio reception the amplitude of the received signal is subject
to fluctuation. The effect of this can be overcome to some extent by automatic gain
control circuits in the receiver. In the case of skywaves the effect may be so severe
that audible fading results.
In all cases if should be remembered that the areas of ionisation are generally
unstable, and the level of attenuation is changeable. The layers are especially
unstable in the transitional periods around dawn and dusk.
Five different types of fading can be identified some or all of which may be present at
any one time.

SELECTIVE FADING
Selective fading affects broadband signals. The different frequencies contained in
the transmission will penetrate the ionised layers to different depths causing phase
differences in the refracted signal, and varying degrees of absorption. Together
these effects produce fading and distortion. Use of SSB with its narrower bandwidth
will mitigate the effect.

INTERFERENCE FADING
Already noted is the possibility of receiving a surface wave and a skywave at the
same time. At ranges where the amplitudes are matched there may at times be total
cancellation when the difference in path lengths is such as to produce 180 degrees
of phase difference (Fig 36).
A similar effect is possible when surface waves interact with delayed reflections from
terrain in mountainous regions.

FIG 36 INTERFERENCE FADING

POLARISATION FADING
As mentioned earlier, a radio wave may have its polarisation modified when passing
through an ionised layer. If the effect is marked, and changeable, it will cause
differences in signal level at the receiver.

Issued November 2006 Chapter 1 : Section 1 25

SKIP FADING
Skip Fading is experienced when the frequency in use is too close to the MUF. A
small reduction in ionisation intensity will place the receiver within the increased skip
distance causing loss of signal (Fig 37).

FIG 37 SKIP FADING

ABSORPTION FADING
The D region is mainly attenuating, and as the degree of attenuation changes,
particularly at dawn and dusk, significant fluctuations in amplitude occur. The effect
is most marked in polar regions.

VERY LOW FREQUENCIES (VLF) AND THE IONOSPHERE

In earlier discussion on ground waves it appeared that the lower the frequency the
greater was the diffraction, and the less the attenuation. Power requirements
however become prohibitive when ranges over 2000 miles are required.
Fortunately, at VLF frequencies, multiple reflections between the earth’s surface and
the ionosphere considerably enhance the reception range to global levels.
The earth’s surface and the lower ionised layers form the walls of a virtual wave
guide, the dimension of which favours the propagation of VLF narrow band signals
(Fig. 38).

FIG 38 VLF DUCTING

26 Chapter 1 : Section 1 Issued November 2006

EARTH-SPACE PROPAGATION
It should by now be apparent that none of the propagation modes so far reviewed
are suitable for totally reliable broadband long-range communications. This
deficiency has been overcome by the use of man-made orbiting satellites operating
outside the earth’s atmosphere. The orbits and type of satellite are many and varied,
but those used for meteorological purposes exemplify two orbital extremes.

LOW ALTITUDE POLAR ORBITERS

The earth rotating within the orbit presents narrow vertical strips on successive
orbits. Distance from the earth is typically 500 Kms (Fig 39).

FIG 39 LOW ALTITUDE POLAR ORBITS

Issued November 2006 Chapter 1 : Section 1 27

GEOSTATIONARY SATELLITES
This type of satellite is positioned over the equator at such a distance that its orbital
speed matches the rotational speed of the earth. This causes it to appear stationary,
relative to an observer on the earth’s surface (Fig 40).

FIG 40 GEOSTATIONARY ORBIT

Because of the distances involved in earth-space propagation free-space losses are

so significant that other path losses become increasingly important.
The up-path is not a problem, since there is no restriction on radiated power. The
situation is quite different on the down-link where the power is limited to that
produced by solar panels.
A satisfactory signal from space will require the use of a high-gain aerial implying
narrow beam width.
A geostationary satellite is virtually unusable in Arctic and Antarctic regions because
of the oblique view of the satellite. This is very noticeable in pictures of the earth’s
disc viewed from space sometimes shown on televised weather programmes.
In passing through the ionosphere/atmosphere the signal will suffer absorption, and
changes in velocity and polarisation. The lower the angle the more marked these
effects (Fig. 41).

28 Chapter 1 : Section 1 Issued November 2006

FIG 41 EFFECT OF SATELLITE ELEVATION

The frequencies used must obviously be high enough to ensure penetration and
avoid excessive refraction by the ionosphere, but not so high as to suffer excessive
absorption. The range of frequencies extends from 100 MHz to 20 GHz. The higher
frequencies being more suited to moderate sized parabolic aerials. The implications
of beamwidth versus frequency should also be considered.

IONOSPHERIC SCATTER

FIG 42 IONOSPHERIC SCATTER

The mechanism of ionospheric scatter is similar to that of tropospheric scatter,

referred to earlier (Fig 42).
The scattering process this time, takes place in the ionosphere, thus the range
achieved is greater than that from the tropopause.
An important difference is in the frequencies affected, in this case between 30 MHz
and 50 MHz. Again the requirement for high power and ultra sensitive receiving
equipment makes this mode unsuitable for aeronautical use.

Issued November 2006 Chapter 1 : Section 1 29

REVISION QUESTIONS (short answer)

1 A TEM wave consists of……………………………………………………………..?
2 A TEM wave is said to be vertically polarised when……………………………..?
3 The velocity of a TEM wave in free space is………………………………….…..?
4 The velocity of a TEM wave within a transmission line (eg. coax cable) will be?
(slower, faster, the same?)………………………………………………….…..…..?
5 Define the range of frequencies in the HF band…………………………………..?
6 Which radio frequency band is most suitable for reliable long range
communication (ie. not susceptible to changing Ionospheric conditions)….…..?
7 An Isotropic radiator has what kind of radiation pattern………………………….?
8 What is meant by ‘the Inverse Square Law’ and how does it apply to TEM
waves………………………...………………………………………………………..?
9 A half-wave dipole has what kind of radiation pattern…………………………....?
10 Low frequency radio waves generally travel around the curvature of the earth
and are known as……………………………………………………………...……..?
11 Why do low frequency radio waves travel/follow the curvature of the
earth……………………………………………………………………………………?
12 Radio waves travel…………………? Over the sea; Why……………………….?
13 What is the range (in nautical miles) of a space wave when the transmitter is at
40,000 ft. and the receiver is at 40,000 ft…………………………………….…....?
14 What are the principle disadvantages of using very low frequency radio
waves………………………………………………………………………………….?
15 Transmissions above approximately 30 MHz are said to be…………………….?
16 During daylight hours the Ionosphere consists of four layers – describe their
location and name relative to each other…………………………………………..?
17 What effect does seasonal variation have on the Ionosphere…………….…….?
18 Why is there less Ionisation over the Poles (North & South) than in the
equatorial regions…………………………………………………………………….?
19 What are ‘sky waves’ and to what range of radio frequencies do they most
apply…………………………………………………………………………………...?
20 What is meant by ‘critical angle’ as it applies to radio propagation……………..?
21 Define ‘critical frequency’…………………………………………………………....?
22 What is meant by Maximum Useable Frequency’ (MUF)………………………..?
23 Define OWF (Optimum Working Frequency) ……………………………………..?
24 Selective Fading can be reduced if what type of signals (modulation etc) are
used…………………………………………………………………………………....?
25 Define: Interference Fading, Polarisation Fading, Skip Fading and Absorption
Fading………………………………………………………………………………….?

30 Chapter 1 : Section 1 Issued November 2006

26 What range of frequencies are most suitable for Satellite Communication and
Why……………………………………………………………………………………?

Issued November 2006 Chapter 1 : Section 1 31

REVISION QUESTIONS II
1 VLF waves are used for some types of services because:

2 Frequencies in the UHF range propagate by means of:

3 An advantage of using microwaves compared with other frequencies for

communication is that:

4 Selective fading will not occur in a region of:

5 For a given state of ionisation and given angle of incidence, the critical
frequency is defined as the frequency:

6 The propagation of radio waves at VHF is:

7 Normal MW local broadcasting makes use of the:

8 Attenuation of the ground wave of a transmission:

32 Chapter 1 : Section 1 Issued November 2006

9 Long distance communication using sky waves reflected from the ionosphere
is usually carried out at frequencies ranging from:

10 Ionospheric scatter may be used at:

11 For HF communication the ‘optimum working frequency’ is:

12 A tropospheric scatter link:

13 Selective fading is due to:

14 ‘Skip distance’ may be defined as the:

15 Wireless communication is usually achieved by a sky wave path via the

ionosphere at frequencies:

Issued November 2006 Chapter 1 : Section 1 33

16 The range at which the sky wave first returns to earth can be increased by:

17 The maximum usable frequency of a radio link is the highest frequency:

18 With radio waves, the range of the ground wave is:

19 A feature of the HF band of radio frequencies is that:

34 Chapter 1 : Section 1 Issued November 2006

SECTION 2 : TRANSMISSION LINES

INTRODUCTION
Information can be sent from one point to another along cables or wires. At low
frequencies, the construction of the cable is unimportant. To keep losses to a
minimum in radio systems, radio frequency energy is transferred by means of special
cables called transmission lines. The purpose of these transmission lines, or
feeders, is to transfer energy from a transmitter to an aerial or from an aerial to a
receiver with minimum loss. For maximum power to be transferred, the source must
be matched to the load, eg. transmitter to aerial, aerial to transmission line and
transmitter to transmission line.

TYPES OF LINE

Open (Twin Wire) Feeder

The open wire feeder consists of 2 parallel wires maintained at a fixed distance apart
by insulating spacers. In ground installations the wires are only a few centimetres
apart and usually run about 3 metres above the ground, supported on insulators
spaced at intervals of 20 to 30 metres. The open wire feeder has certain advantages
over other types of feeder, but it also has some limitations.

FIG 1 OPEN TWIN FEEDER

Issued November 2006 Chapter 1 : Section 2 35

Advantages
 High transmitter power outputs can be handled without danger of ‘flash
over’ between conductors.
 Standing waves can be easily measured and maintenance of line is easy.
 The line is balanced, ie. the impedance between each line and earth is
the same, this property is useful when matching to aerials

Disadvantages
 They are bulky and rigid and can only be used in static installations.
 The practical upper frequency limit is about 100 MHz.
 The line must be kept clear of ground and walls

COAXIAL CABLE.
This is a transmission line in which one conductor is in tubular form and completely
surrounds and screens the inner conductor. The inner conductor is a wire running
along the axis of the outer tube, supported in position by insulating washers or by
completely filling the space between the conductors with a low loss dielectric (eg.
polyethylene). The outer conductor may be a metal tube or braided metal. The
whole assemblage is covered by a plastic material.

FIG 2 COAXIAL CABLE

In relation to the twin wire feeder, the coaxial feeder has certain advantages and
certain limitations.

36 Chapter 1 : Section 2 Issued November 2006

Advantages
 The coaxial line is a screened cable and radiation losses are negligible.
It can be used at frequencies up to 1000 MHz.
 It is flexible.

Disadvantages
 The power handling capacity is less than the open wire feeder.
 The coaxial line is unbalanced, one conductor being at earth potential
and the other isolated from it. This can introduce aerial matching
problems.

THE INFINITE LINE

Introduction
If a generator is connected to a line stretching to infinity, voltage and current waves
will travel along the line in phase as shown in Fig 3. If it is assumed that there are no
losses, the waves will continue to infinity and there will be no reflection of energy
back toward the generator. It appears that the generator is connected to a resistor of
value given by:

FIG 3 VOLTAGE AND CURRENT WAVES ALONG AN INFINITELY LONG LINE

Issued November 2006 Chapter 1 : Section 2 37

CHARACTERISTIC IMPEDANCE
The resistance seen by the generator is a characteristic of the line and depends on
the diameter of the conductors, their spacing and the type of dielectric between
them. The input impedance of an infinite line is called the characteristic impedance
. When a current passes down a line, magnetic fields are set up around the
conductors and electric fields between them. A transmission line can therefore be
represented in terms of inductance and capacitance (assuming no losses).

FIG 4

The characteristic impedance can be calculated from the formula:

Where = characteristic impedance

= inductance (henrys)/unit length

= capacitance (farads)/unit length
Characteristic impedance is a property of the line itself and is independent of the:
 frequency of the RF energy that it carries
 length of the line
Typical values of are:
 open wire feeder between 150 and 1000 ohms, typical 300 or 600 ohms.
 coaxial line between 40 and 300 ohms, typical 45, 50, 75 or 300 ohms.
It is not possible to have a line of infinite length, but since the input impedance of an
infinite line is equivalent to a resistor value any line terminated in such a resistor
will act in the same way as an infinite line.. Whatever the distance from the
generator to the resistor, all the energy leaving the generator is absorbed.

38 Chapter 1 : Section 2 Issued November 2006

RESONANT AND NON-RESONANT LINES

The Effects of Line Termination

When a generator is connected to a load by a transmission line and all are of equal
impedance, maximum transfer of energy takes place between the generator and line
and the energy travels down the line (the forward or incident wave) to the load where
it is totally absorbed. A line terminated by a load equal to is said to be non-
resonant.

FIG 5

When the load is not matched to the line, not all the energy travelling down the line is
absorbed by the load and some is ‘reflected’ back up the line to the generator. This
wave is called a ‘reflected’ or ‘backward’ wave and its amplitude is determined by the
degree of mismatch between the load and .

FIG 6

Issued November 2006 Chapter 1 : Section 2 39

RESONANT (MISMATCHED) LINES

In general, mismatches increase losses in communication systems, although there
are occasions when they are deliberately introduced. We will consider these points
further in the following paragraphs.

SHORT CIRCUITED LINE

If a generator is connected to a line terminated in a short circuit, the line at the
generator end acts in a similar way to that of an infinite line, ie. the voltage and
current are in phase. When the waves reach the shorted end, the current is a
maximum as a short circuit has no resistance, it cannot absorb any energy and so all
the energy travelling down the line must be reflected back to the generator end. The
reflected energy must be carried by voltage and current waves and so there are now
4 waves on the line:
 2 voltage and 2 current.
The voltages cancel at the shorted end and the current waves add. As the incident
wave travels down the line, there will be a reflected wave returning towards the
generator. These 2 waves are continuously changing in phase relation with each
other and eventually a resultant of these 2 waves is formed on the line. The
maximum and minimum points of this wave are always in the same place, the
resultant wave is therefore called a standing wave. There will be a current standing
wave set up in a similar manner, but the current standing wave will be in phase
opposition to the voltage standing wave.

FIG 7 RMS VALUES OF VOLTAGE AND CURRENT ON A SHORT CIRCUITED LINE

Fig. 7 shows the voltage and current standing wave patterns. In theory, the current
should be infinite ( being ), but the generator can only provide a limited amount.
At the generator end, the voltage is maximum and the current zero, the conditions
we would expect if the generator were open circuit. The input impedance at that
point is infinite and if the generator were connected at any point which is an odd
number of quarter wavelengths away from the short circuit, it would ‘see’ an open
circuit.

40 Chapter 1 : Section 2 Issued November 2006

Since an open circuit is the equivalent of a parallel tuned circuit, any short circuit
transmission line will behave as a parallel tuned circuit if it is an odd number of
quarter wavelengths long. At a point between the short circuit and the quarter
wavelength point, the voltage is leading the current and that point acts as an
inductive reactance. At a point a half wavelength from the short, it will act as another
short(series tuned circuit). At a point between the quarter and half wavelength
points, the current leads the voltage and the line acts as a capacitive reactance.
These conditions are repeated every half wavelength and by altering the length of a
short circuit transmission line, we can alter its input impedance.

OPEN CIRCUIT LINE

The open circuit line is similar to the shorted line in a number of respects. Since the
resistance at the end of the line is infinite, the current is zero and the voltage
maximum. The current and voltage waves are reflected without change in amplitude
and so current and voltage standing waves are set up. When the generator is an
odd number of quarter wavelengths away from the open circuit, it will see a short
circuit, for an even number, it will see an open circuit and at points in between, a
capacitive or inductive reactance, depending on position.

FIG 8 RMS VALUES OF VOLTAGE AND CURRENT ON AN OPEN CIRCUITED LINE

Resistive Termination
If the load resistance is not equal to , some (but not all) of the energy will be
reflected back to the generator. The greater the proportion of reflected energy, the
greater will be the standing wave. The worst cases are when the load is a short-
circuit or an open-circuit, the best case when there is a matched load .

Reactive Termination
If the line is terminated in a pure capacitance or pure inductance, the standing wave
pattern will be similar to that of the open and short circuit terminations, except that
the standing waves are displaced to produce a capacitive or inductive region at the
end.

Issued November 2006 Chapter 1 : Section 2 41

STANDING WAVE RATIO

Standing waves on a transmission line are undesirable for the following reasons:
 At points of maximum voltage there is an increased risk of arcing due to
insulation breakdown.
 Dielectric, and radiation losses increase.
 Transmitter tuning may be affected.
The degree of matching (or mismatching) is indicated by the standing wave ratio,
defined as:

In the case of a perfectly matched line:

and

In the case of a completely mismatched line (ie. open or short-circuited):

and
(infinite)

MEASUREMENT OF SWR
SWR may be calculated by measuring the amplitudes of the voltage (or current) at
the nodes and antinodes. Alternatively, it is often more convenient to measure the
amplitude of the incident (forward) wave (Vf) and the reflected wave (Vr), then:

Sometimes it is more convenient to measure powers, eg. with an in-line wattmeter.

As power is proportional to the square of the voltage:

and dividing top and bottom by

In practice, lines should be matched as accurately as possible although this is not

easy over a wide frequency range. A high degree of matching ensures that little
power is wasted in the reflected wave, and high voltages are not developed at the
antinodes. With an aerial as the load, it is sometimes difficult to obtain a very low

42 Chapter 1 : Section 2 Issued November 2006

VSWR and limits are laid down for each type of transmitting system in use. See
notes on Systems later.

USE OF OPEN AND SHORT CIRCUIT LINES

Short and open circuited lengths of transmission line have many uses and several of
these applications are described in the following paragraphs. A quarter wave section
of short-circuited line is particularly versatile and can be used as an insulator, a
matching transformer and as a RF tuned circuit. For matching purposes any length
of open or short-circuited line can be used when positioned carefully at points on a
transmission line where a particular reactance is present. Such lengths are termed
matching stubs’.

METALLIC INSULATORS
A short-circuited, quarter-wave, resonant transmission line, known as a stub, has a
very high input impedance. It can therefore be used as a metallic insulator to
support an open wire transmission line without short circuiting the line.

FIG 9 SHORT-CIRCUIT STUB USED AS METALLIC INSULATOR

HARMONIC FILTER
Since the impedance across the ends of the stub is very high, the main
transmission line is not affected in any way. This is only the case at one frequency
and if the frequency of the transmitter changes, the stub is no longer a quarter
wavelength long. The stub can also be used to filter out the second harmonic of the
signal being carried as, at this frequency, the stub is half a wavelength long and
shorts out the undesirable harmonic.

Issued November 2006 Chapter 1 : Section 2 43

REVISION QUESTIONS I
A radio frequency transmission line is used to……………………………………. The
RF energy from a……………………………..to a…………………………….
For maximum power transfer to take place in any system, the………………… must
be matched to the……………………………………………………………..
Identify the type of line and label the diagram.

4 The line shown in question 3 has certain advantages. Indicate the ones that
apply.

5 Identify the type of line and label the diagram.

6 The line shown in question 5 has certain advantages, indicate the ones that
apply:

7 The input impedance of an infinitely long line is called the……………………….

impedance.

44 Chapter 1 : Section 2 Issued November 2006

8 Select typical values of characteristic impedance for the lines below

Unbalanced Lines Characteristic Impedance Balanced Lines
300 OHMS

75 OHMS

57 OHMS

600 OHMS

45 OHMS

22 OHMS

9 Calculate the characteristic impedance of a transmission line with

inductance/unit length of 1 H/metre and a capacitance/unit length of 11.11
pF/metre.

10 If, with an infinitely long line, power is continually drawn from the supply, the
infinitely long line acts as a…………………………………………equal in value
to…………………………………………………
11 A transmission line can be represented by ‘discrete components’. Identify the
correct circuit by indicating your answer in the box provided.

Issued November 2006 Chapter 1 : Section 2 45

12 A non-resonant line is a line terminated by a resistance equal to

its……………………………….……. ………………………………………….
13 A resonant line has……………………………………………..……..waves on it,
resulting from a……………………………………………..load.
14 A finite length of line can be made to act like an infinite length of line by
terminating it with a resistance equal to the………………………………………..
15 Standing waves are set up on a line due to a combination
of…………………………………..and reflected waves.
16 Explain why you think standing waves are undesirable by placing a tick
opposite your choice in the answer panel.

They increase risk of insulation breakdown.

Losses are increased.
A large diameter cable must be used.
Transmitter tuning may be affected.
Receiver tuning may be affected.
It causes instability of signal.

17 Draw in the voltage and current patterns for the lines shown.

46 Chapter 1 : Section 2 Issued November 2006

18 The standing wave ratio of a line is the ratio of……………………………………

standing voltage to…………………………………………………………standing
voltage.
19 In a certain line, the value of standing voltage wave is 250 volts maximum and
100 volts minimum. Calculate the VSWR.

20 Match the correct responses to the diagrams.

Diagrams

Responses

Issued November 2006 Chapter 1 : Section 2 47

REVISION QUESTIONS II
The characteristic impedance of a transmission line depends upon the:

In a perfectly matched transmission line the SWR is:

If a transmission line is shorted at the far end:

A ‘quarter wave transformer’ could be used to:

A transmission line consisting of 2 parallel wires has inductance of 0.05 microhenries

and a capacitance of 10 picofarads. The characteristic impedance of the line
is approximately:

6 The diagram below shows a matching stub before attachment to a

transmission line, for successful matching the input impedance of the main
transmission line, at the point of attachment would be:

48 Chapter 1 : Section 2 Issued November 2006

7 The characteristic impedance for maximum power transfer would be:

8 is:

9 is:

10 The input impedance of a finite length of transmission line terminated in a load

may be expressed as:

11 The input impedance of a line, infinite in length:

12 A transmission line is terminated by a load equal in value to the line

characteristic impedance. Sketch the graphs of voltage and current against
length of line.

Issued November 2006 Chapter 1 : Section 2 49

13 In a perfectly matched transmission line:

14 Standing waves should normally be avoided since:

15 The combination of incident and reflected currents in a mis-matched line gives

rise to:

16 A serious mismatch exists on a Tx line when the measured ratio of to

is:

17 List 5 properties of a coaxial feeder which makes it useful for installation in

aircraft.

50 Chapter 1 : Section 2 Issued November 2006

18 Standing waves are deliberately introduced on lines for the purpose of:

19 If a S/C line is long at 60 MHz, what impedance does it offer to a

generator connected at the input if the frequency is increased slightly?

20 Refer to the drawing below. Which is correct?

21 A quarter wave matching transformer is used to match a 300  transmission

line to a 75  aerial. The characteristic impedance of the quarter wave
transformer is:

Issued November 2006 Chapter 1 : Section 2 51

GLOSSARY OF TERMS
Baluns Balanced – unbalanced units
C Capacitance
 Current
 Infinity
L Inductance
 Wavelength
MHz Megahertz
OC Open Circuit
 Ohms
R Resistance
RF Radio frequency
RMS Root mean square
SC Short circuit
VSWR Voltage standing wave ratio
XC Capacitive reactance
XL Inductive reactance
ZG Impedance of the generator
Zin Input impedance
ZL Impedance of the load
Zo Characteristic impedance

52 Chapter 1 : Section 2 Issued November 2006

SECTION 3 : AERIAL PARAMETERS

INTRODUCTION
Aerials (or antennae) are metallic or dielectric structures for transmitting or receiving
radio waves. They are connected to the transmitter or receiver by a transmission
line or waveguide. The properties of aerials are the same for transmission or
reception and in many communication and radar equipments, the same aerial is
used for both. Aerials take a wide variety of forms, including rods, wires, slots,
loops, arrays, parabolic reflectors, horns, lenses, helixes, spirals and printed circuit
aerials. Those applicable to aircraft communication will be covered in the following
notes.

NEAR AND FAR FIELDS

The electromagnetic field radiated from an aerial consists of a number of
components which vary inversely with powers of the distance of the aerial. Those
which vary inversely with the square and higher powers of the distance fall off fairly
rapidly and are known as the near fields. At some distance from the aerial, only the
field which is inversely proportional to distance will be significant, and this is known
as the far field. We will be concerned only with the far field in these notes, although
information about the near field may be essential for safety reasons in the vicinity of
aerials used for high power transmission.
RADIATED PATTERN
In general, aerials do not radiate equally in all directions. The radiation pattern which
is a plot of relative field strength (by convention the E field) as a function of direction
will indicate in which directions the field strength is greatest. For complete
information the radiation pattern will be 3-dimensional, but is usually given as two 2-
dimensional radiation patterns which are mutually orthogonal, the planes chosen
being that of the E-field and that of the H-field since for a traverse electromagnetic
wave these fields are at 90 to one another, as shown in Fig 1.

FIG 1

Issued November 2006 Chapter 1 : Section 3 53

The radiation pattern may be plotted in Cartesian or polar co-ordinates and typically
might be as shown in Fig 2.

FIG 2(A) - POLAR FIG 2(B) - CARTESIAN

AERIAL GAIN (DIRECTIVITY)

An aerial which radiates in all directions is called omni-directional. The gain of the
aerial is a measure of its ability to radiate energy in a particular direction. To quantify
gain the aerial is compared with a hypothetical isotropic aerial which radiates equal
energy in all directions and which is fed with the same input power. Gain is defined
by:

The gain is usually quoted in decibels.

54 Chapter 1 : Section 3 Issued November 2006

BEAMWIDTH
The beamwidth of an aerial is defined as the angle between the half-power points of
the main lobe. On the radiation pattern diagram (Fig 2) these points correspond to
the points where E = 0.707 E max. It can be seen that the beamwidth is also a
measure of the directivity of the aerial and is therefore related to the gain.

POLARISATION
The radiation from an aerial may be linearly, elliptically or horizontally polarised. In
fact the polarisation may vary from one part of the radiation pattern to another. The
simplest aerials radiate (and receive) linearly polarised waves usually orientated so
that the polarisation (direction of the E-vector) is either horizontal or vertical. The
choice of polarisation is determined by a number of factors. For example, below 1
MHz vertical polarisation is preferred since it propagates more successfully and
suffers less cancellation due to ground reflection, for television, however, horizontal
polarisation is preferred since it maximises signal to noise ratio due to the fact that
most man-made noise is vertically polarised, for airborne radar horizontal
polarisation tends to give small clutter echoes from the sea. For yet other purposes,
circular polarisation may be preferred, eg. to reduce clutter echoes due to rain or to
receive signals whose linear polarisation may vary such signals received from
satellites or by ionospheric reflection.
SIDELOBES
An unavoidable feature of aerials in general is the existence in the radiation pattern
of small subsidiary beams known as sidelobes which are generated in directions
other than that of the main lobe. Their existence is undesirable since they represent
wasted power and may receive unwanted signals, eg. deliberate jamming. Their
magnitude is usually quoted relative to the main lobe in decibels, eg. if the biggest
sidelobe (usually adjacent to the main lobe) in the radiation pattern is 10 times
smaller than the main lobe, then:

If the aerial has a large back lobe in the opposite direction to the main lobe, a back-
to-front ratio may be quoted, ie.:

Issued November 2006 Chapter 1 : Section 3 55

NOTES:

56 Chapter 1 : Section 3 Issued November 2006

SECTION 4 : AERIAL MATCHING

INTRODUCTION
In this section, the problem of matching aerials to transmission lines is considered.
The need for matching aerial loads to the feeder has already been shown in the
earlier chapter on transmission lines. Here, the methods of matching the dipole (2-
element) aerial and the effects of mismatch are dealt with, and the way in which the
efficiency of an antenna can be measured.

INPUT IMPEDANCE
The voltage and current values along the length of an aerial vary. The ratio of
voltage to current at any point is the impedance at this point. The impedance that an
aerial presents to the source of energy is called its input impedance. It follows that
the input impedance must depend upon the point at which the source is connected to
the aerial.
For maximum transfer of power from the transmitter to the aerial (or from receiving
aerial to a receiver) it is important that correct impedance matching is achieved.
 In order to achieve maximum transfer of power, the internal resistance of the
generator must be matched to the resistance of the aerial at the point of feed.
If this is done the generator ‘sees’ a resistance equal in value to its own
resistance, and the travelling wave reflected from the end of the aerial is
absorbed without reflection, by the generator. Maximum power transfer takes
place under these conditions.
 If a transmission line is used to convey the energy from transmitter to aerial,
the internal resistance of the transmitter must be matched to the characteristic
impedance of the line, at the aerial end the characteristic impedance of the
line must be matched to the input impedance of the aerial. This often means
employing a matching transformer.

Issued November 2006 Chapter 1 : Section 4 57

METHODS OF FEED
If the generator is connected at the centre of the aerial it is feeding into a point of low
impedance and is said to be current fed or centre fed. If connected at a point of
maximum voltage, the aerial is said to be voltage fed or end fed. Both these
methods of feeding are illustrated in Fig 1 below.

FIG 1 METHODS OF FEEDING A DIPOLE

The input impedance of a current fed half-wave dipole is 73 ohms while for a voltage
fed half-wave dipole it is approximately 2,500 ohms, thus the input impedance and
radiation resistance of a half-wave dipole are of the same value only when the aerial
is centre fed.

RADIATION RESISTANCE
The purpose of an aerial is to dissipate energy to the surrounding space in the form
of electromagnetic radiation. Thus, the aerial converts energy from one form to
another in the same way that a resistor converts electrical energy into heat energy
when a current flows through it. Therefore an aerial can be considered as a fictitious
resistor which dissipates energy in the form of EM radiation. This effective
resistance is known as the radiation resistance of the aerial.
In general the input impedance of an aerial is a complex value, the real part of which
represents both radiation resistance and ohmic resistance. The ohmic resistance is
small and often considered negligible. The rms current supplied by the generator
driving a half-wave dipole, and necessary to overcome the radiation losses,
constitutes energy dissipated in the radiation resistance of the aerial. In the case of
a half-wave dipole the radiation resistance is approximately constant at 73.

58 Chapter 1 : Section 4 Issued November 2006

ENERGY LOSS
If the aerial was perfect, all the energy supplied to it would be converted into EM
radiation. However, losses occur and some of the supplied energy is wasted. This
constitutes a loss resistance which added to the radiation resistance, increases the
total aerial resistance. The sources of these losses are summarised as follows:
 Dielectric losses. These are due to the energy dissipated in the aerial
insulators and nearby objects which are affected by the radiation fields. High
quality low loss insulators have been developed, which reduce these losses to
a minimum but to be fully effective the insulators must be kept clean.
 Brush Discharge Losses. These are caused by a discharge due to ionisation
of the air near the aerial. Brush discharge is most apparent from high voltage
points on the aerial and can be reduced by rounding the ends of the aerial.
 Copper Losses. The inherent ohmic resistance of the aerial wire converts
electrical energy into heat energy, causing a loss of supplied energy. These
losses increase with frequency, due to skin effect.
 Eddy Current Losses. Any nearby metal bodies will have currents induced in
them by the aerial energy. These currents produce heat, the energy for which
has come from the aerial. Thus to reduce these losses, the aerial should be
mounted as far away as possible from metal objects.

Aerial Efficiency
The efficiency of an aerial is:

Where is the radiation resistance of the aerial, and the total loss resistance.
This formula represents the fraction of the total power supplied to the aerial which is
converted into radiated power. For high efficiency, loss resistance must be small
compared with radiation resistance. For practical aerials the efficiency can vary from
15% at low frequencies to 90% and above at higher frequencies.

Issued November 2006 Chapter 1 : Section 4 59

MATCHING METHODS
A variety of impedance transformers are used to achieve matching, at low
frequencies these tend to be coupled coils of wire or coil and capacitor circuits, at
higher frequencies they may be transformers, short circuit stubs or waveguide
matching devices such as posts or irises. In the relatively new ‘active’ aerials used
for reception only, the matching device is a transistor amplifier connected to the
aerial whose input impedance matches the aerial and whose output impedance
matches the transmission line.
Mismatches between an aerial and its transmission line results in standing waves
being set up on the line. These give rise to radiation losses on the line and arcing.
In addition the efficiency of the aerial is reduced.

The Folded Dipole

In cases where it is necessary to use a transmission line of high characteristic
impedance (of the order of 300 ohms) to feed a centre fed half wave dipole,
matching is made easier by using a folded dipole. The important feature of this
aerial is its high input impedance which is approximately 4 times that of a simple
dipole, ie. . Thus it presents a suitable load to a 300 ohm feeder.
The current and voltage distribution on an aerial one wavelength long is shown in Fig
3 (a). When the portion of is folded over the portion , the current in these 2
halves is in the same direction and the current standing waves will be in phase and
reinforce each other – Fig 3(b).

FIG 2 THE FOLDED DIPOLE

60 Chapter 1 : Section 4 Issued November 2006

SECTION 5 : SIMPLE RESONANT AERIALS

INTRODUCTION
In dealing with the elementary principles of radiation a simple half-wave dipole can
be used to illustrate the principles. The half-wave dipole is a resonant or standing
wave aerial, non-resonant or travelling wave aerials are also used. The 3 simple
resonant aerials which will be dealt with are the half-wave dipole, the Marconi
quarter-wave aerial and the slot aerial, all in common use in ground and air radio
installations.

STANDING WAVE OR RESONANT AERIALS

An aerial of this type is basically a series resonant circuit in which electrons surge
backwards and forwards along the wire which forms the aerial. The electrons can be
considered as effectively forming one large charge of electricity which is continuously
being accelerated and decelerated by the existing voltage of the generator. The
result is that radiation occurs, maximum radiation takes place perpendicular to the
aerial and zero from the ends.

THE HALF-WAVE DIPOLE – PRODUCTION OF STANDING WAVES

A half-wave dipole consists of a wire half a wavelength long at the frequency
considered, with an alternating generator at its centre. It is shown in Fig 1. When
terminal A is negative, a travelling wave of current, accompanied by a travelling
wave of voltage moves from towards . When the current reaches it suffers a
reversal of phase and is reflected back towards the generator. When it arrives at the
generator after having travelled half a wavelength, it will be in phase with the incident
wave of current. The voltage wave travels the same distance, but is reflected
without a change of phase and so arrives back at the generator in anti-phase to the
incident voltage. Similar conditions apply for waves travelling from to

FIG 1 PRINCIPLES OF THE HALF-WAVE DIPOLE

Issued November 2006 Chapter 1 : Section 5 61

The incident and reflected waves travelling along the aerial combine at each instant
throughout the cycle to form resultant voltage and current standing waves. Fig 2
illustrates the build up and collapse of the standing waves at one-eight of a cycle
intervals over half a period. The current standing wave is a maximum at the centre,
forming a current antinode and zero at the ends forming current nodes while the
voltage standing wave has antinodes at the ends and a node at the centre. It has
already been mentioned that a resonant aerial is an open series resonant circuit.
The voltage at the ends of the aerial is equivalent to the voltage across the capacitor,
and is many times the applied voltage. Were there no losses in the aerial, its Q
would be infinity and there would be voltage at the ends but zero voltage at the
centre. In practice, the aerial does suffer losses because it radiates energy, and a
relatively small driving voltage is required to maintain standing waves, in the same
way that a small voltage is required to overcome the resistance losses in a series
resonant circuit.

FIG 2 STANDING WAVES ON A HALF-WAVE DIPOLE

62 Chapter 1 : Section 5 Issued November 2006

As shown in Fig 2 (a), the voltage and current standing waves on the aerial are 90
out of phase in time, ie. when voltage is a maximum current is zero and a quarter of
a cycle later, when current is a maximum voltage is a minimum. However, when
depicting standing waves on an aerial, graphs of the rms values of voltage and
current are shown as illustrated in Fig 2 (b). These waves can be measured readily
by using a suitable meter.
We have seen earlier that the input impedance of a dipole varies with the point of
connection. For a current (ie. centre fed) dipole it is 73 while a voltage fed dipole is
2500.
So far resonant aerials with half a wavelength long have been discussed. It is
possible to have resonant aerials of greater length, ie. or or even longer.
However, for simplicity, ease of construction and ease of matching, a half
wavelength aerial is by far the most common.

POLAR DIAGRAMS
 A vertical aerial in space does not radiate fields of equal intensity in all
directions. It cannot, for example, radiate any energy from the ends. Thus, if
a field strength meter were placed directly above or below a vertical aerial it
would read zero. If the meter, still in the vertical plane, were moved around
the aerial and readings taken at regular intervals, a diagram could be
constructed from these readings. An indication of the strength of the field
radiated in all directions in the vertical plane could thus be obtained. The
same could be done for the horizontal plane and another diagram drawn.
 These diagrams are called polar diagrams and are very useful in determining
the radiation characteristics of an aerial. A polar diagram does not show the
limits of reception in various directions. This depends, among other things,
upon the strength of the transmitter feeding the aerial, and the sensitivity of
the receiver.

Issued November 2006 Chapter 1 : Section 5 63

POLAR DIAGRAMS OF A HALF-WAVE DIPOLE SUSPENDED IN FREE SPACE

The polar diagrams of a half-wave dipole removed from any reflecting surface such
as the earth, are shown in Fig 3. Fig 3(a) shows the horizontal and vertical polar
diagrams for a vertical half-wave dipole in free space and Fig 3 (b) shows the same
for a horizontal dipole. A composite picture of the radiation pattern for a vertical
half-wave dipole is given in Fig 3 (c). Notice that if a line is drawn on the diagram in
any direction, from the aerial to the curve – AB in Fig 3(a) – its length indicates the
relative strength of the wave radiated in that direction. Thus maximum radiation
occurs along AC (90 to the aerial) and zero along AD (from the end of the aerial).
Equal all-round radiation occurs in the horizontal plane for a vertical half-wave
dipole, and in the vertical plane for a horizontal half-wave dipole.

FIG 3 POLAR DIAGRAMS FOR HALF-WAVE DIPOLES SUSPENDED IN FREE SPACE

64 Chapter 1 : Section 5 Issued November 2006

The polar diagram of a dipole aerial varies not only with its orientation, but with its
height (usually expressed in wavelengths) above the ground, which is considered to
be a conducting plane. Fig 4(a) shows the vertical radiation pattern for a horizontal
dipole mounted one wavelength above the ground. Note that there is no radiation in
the horizontal plane. The total number of lobes is 4 for each wavelength above the
ground. As the height increases, the number of lobes increases.
Fig 4(b) shows the vertical radiation pattern for a vertical dipole suspended at a
height of . Compared with the horizontal dipole, there is a horizontal radiation.
The number of lobes increases with aerial height, but not in an easily predictable
way.

FIG 4

Issued November 2006 Chapter 1 : Section 5 65

THE MARCONI AERIAL

The aerial consists of a vertical conductor, the lower end being just above the earth.
One of the feeders is connected to the bottom of the aerial, the other to earth.

FIG 5

To improve the earth connection with ground based aerials, a ‘mat’ of copper wires
radiating out from the aerial is buried in the ground. On an aircraft the metal skin
acts as a ground plane. It has an input impedance of 36. Its polar diagrams are
shown in Fig 6.

FIG 6

66 Chapter 1 : Section 5 Issued November 2006

THE LOOP AERIAL

The H-field of the electromagnetic field links with the loop, inducing a voltage in it.

FIG 7

Since the H-field is horizontal and at right angles to the direction of propagation,
maximum voltages will be induced in the loop when it is in position (a) relative to the
transmitter. Minimum voltage will be induced when the aerial is in position (c)
relative to the transmitter. The loop may consist of many turns.

FIG 8

FERRITE AERIAL
Ferrite is a magnetic material which is an insulator. Therefore the magnetic field of
an electromagnetic wave is concentrated within the ferrite and a strong voltage is
induced in a coil would round the ferrite. The voltage induced in the coil is greatest
when the ferrite rod is parallel with the magnetic field of the electromagnetic wave.
Its polar diagram is therefore similar to that of the loop.

FIG 9

Issued November 2006 Chapter 1 : Section 5 67

DIRECTIONAL AERIAL ARRAYS

The simple aerials so far considered do not possess directional properties. For
example, the field radiated from a single vertical aerial is of equal amplitude in all
directions in the horizontal plane. An aerial system designed to radiate most of its
energy in one direction would have many uses. The operational range of the
equipment could be increased or the transmitter power could be reduced. Further,
such a system, used as a receiver aerial, would indicate the direction in which a
transmitter lies.
Directional properties can be obtained by using combinations of aerials and systems
such as these are called aerial arrays. The common television receiving aerial is
one for example, and many installations employ directional aerial arrays.

PARASITIC ELEMENTS
If a metal rod is placed in the field radiated from a driven dipole it will have current
induced in it, and it will radiate. The rod obtains its energy from the driven dipole and
so is called a parasite element (Fig 10) The phase of the energy radiated by the
parasite will depend on its length, and its distance (S) from the dipole. The length
of the parasite determines the phase difference between the current and induced
voltage. The distance of the parasite from the driver determines the phase of the
voltage induced in the parasite relative to the field radiated by the driver.

FIG 10 THE PARASITIC ELEMENT

Both the parasitic element and the driven element radiate EM energy. These fields
will reinforce each other in one direction, while they will weaken each other in the
opposite direction. In practice it is found (by trial and error), that for radiation to be
increased in one direction the parasite has to be reactive. This is achieved simply by
making the parasite longer or shorter than half a wavelength.
Best results are obtained when an inductive parasite longer than the dipole is placed
a quarter of a wavelength behind the dipole. This results in the horizontal polar
diagram shown in Fig 11 (a). This parasitic element is called a reflector.

68 Chapter 1 : Section 5 Issued November 2006

Similar results are obtained if a parasite shorter than the dipole is placed in
front of the dipole. The parasite is now called a director and the combination
produces a horizontal polar diagram as shown in Fig 11 (b). The aerial array shown
in Fig 11 (a) is often used as roof-top television receiver aerial.

FIG 11 DIPOLE WITH PARASITES

THE YAGI ARRAY

A combination of reflector and director used with a driven dipole, gives greater
forward radiation than either used separately. This arrangement is shown in Fig 12,
and is known as a Yagi array. The spacing and lengths of the elements are usually
found by trial and error, typical figures are given in Fig 12(a). the horizontal polar
diagram is shown in Fig 12(b).

FIG 12 THREE-ELEMENT YAGI ARRAY

The directivity can be increased if the number of directors used is increased. The
further the director is from the dipole, the greater is the capacitive reactance required
to obtain correct phasing of the parasitic current. The lengths of the directors
therefore taper off as shown in Fig 13 (a).

Issued November 2006 Chapter 1 : Section 5 69

FIG 13 SIX-ELEMENT YAGI ARRAY

INPUT IMPEDANCE OF YAGI ARRAY

Because of the proximity of the parasitic elements and the driven element of a Yagi
array, there is considerable inductive coupling between them. This reduces the input
impedance.

70 Chapter 1 : Section 5 Issued November 2006

REVISION QUESTIONS
1 In a radiating half wave dipole, the voltage and current distribution is as
follows:

2 The ‘radiation resistance’ of an aerial is:

3 An energised dipole is suspended in free space. The radiation will be:

4 At its resonant frequency, the impedance at the centre of a half-wave

dipole is:

5 One method of increasing the bandwidth of a half-wave dipole is to:

6 One of the main differences between the Marconi quarter wave aerial and the
half wave dipole is the:

Issued November 2006 Chapter 1 : Section 5 71

7 An aerial suitable for direct matching to a 300 ohm feeder is a:

8 The current in a radiating half wave aerial is:

9 Reflections from the surface of the earth modifies the radiation pattern (ie.
polar diagram) of an aerial. For a vertical half wave dipole about 2
wavelengths above the ground, reflections from the ground:

10 It is required to construct a half wave dipole to receive the BBC VHF FM

radio transmissions at 100 MHz. The aerial would have a length of:

11 A certain quarter wave aerial is designed to resonate at 10 MHz. If this

aerial is used to transmit a signal at 8 MHz, the input impedance of the
aerial will be:

12 At its resonant frequency the input impedance of a quarter wave Marconi

aerial is about:

72 Chapter 1 : Section 5 Issued November 2006

13 A Marconi aerial is to be used to transmit a frequency at which it is less than

quarter of a wavelength. In order to radiate the maximum power from the
aerial:

14 An electromagnetic wave induces an emf in a dipole receiving aerial:

15 The length of a quarter wave aerial for use at 3 MHz is approximately:

16 Power radiated from a correctly matched aerial is proportional to:

17 Aerial gain is given by:

18 The bandwidth of an aerial can be increased by:

Issued November 2006 Chapter 1 : Section 5 73

19 A centre-fed dipole aerial is:

20 A parasitic element, slightly longer than a radiating dipole is placed a

quarter of a wavelength away from it. This results in:

21 The directivity of a Yagi array is dependent on:

22 The fundamental frequency of a dipole 5 metres long is approximately:

23 Placing a parasitic director in front of an aerial:

24 The input impedance of a folded dipole is approximately:

25 A Marconi aerial can be electrically lengthened by:

74 Chapter 1 : Section 5 Issued November 2006

26 An aerial is not absorbing all the power which reaches it via the feeder
system. This results in:

27 Increasing the number of elements in a Yagi:

28 The vertical radiation pattern of a vertical half dipole is:

Issued November 2006 Chapter 1 : Section 5 75

SHORT ANSWER QUESTIONS

1 Sketch the relationship of the vectors of an EM wave……………………
2 State which vector is used as the plane of polarisation of an
aerial……………………………………………………………………………
3 State at which point(s) of the main lobe of a directional aerial the bandwidth is
measured……………………………………………………….
4 State the reasons why sidelobes should be kept to a
minimum………………………………………………………………………..
5 Draw the voltage and current waveforms for a centre fed
dipole…………………………………………………………………………...
6 Define radiation resistance…………………………………………………..
7 Calculate the efficiency of an aerial which has a radiation resistance of 80
ohms and a loss resistance of 10 ohms………………………………..
8 State the input impedance for a dipole when:
a centre fed………………………………………………………………
b end fed………………………………………………………………….
9 State an advantage of a folded dipole………………………………………
10 Sketch the polar diagrams for:
a vertical aerial…………………………………………………….
b horizontal aerial…………………………………………………
11 Sketch the polar diagrams for a Marconi vertical aerial…………………..
12 State the required positioning of a loop aerial relative to the transmitter to
induce:
a maximum voltage……………………………………………………..
b minimum voltage………………………………………………………
13 State an advantage of using a ferrite aerial………………………………..

76 Chapter 1 : Section 5 Issued November 2006

CHAPTER 2 : TRANSMITTERS
CHAPTER 1 : ANALOGUE MODULATION
INTRODUCTION
Modulation is the process of modifying a carrier frequency in such a way that an
information signal (speech, music, data, etc) which otherwise is incapable of being
transmitted, may be sent over a transmission link. The form of the transmission link
is not important in the present discussion. What is of concern is the processing of
the signal, perhaps a speech waveform, into a form which can travel down the link.

TYPES OF MODULATION
There are three fundamentally different ways in which modulation may be carried
out:
 Analogue Modulation. Analogue modulation is a process whereby some
feature of a continuous wave is modified in direct proportion to the modulating
signal.
 Pulse Modulation. Pulse modulation is similar to analogue modulation in the
sense that a feature of the wave is modified in proportion to the modulating
signal. The essential difference is that the modulated wave is not a
continuous sinusoid, but a pulse train. In radar systems the meaning of pulse
modulation is somewhat different, being the switching on and off of a
continuous wave (CW) transmitter for pre-defined intervals.
 Digital Modulation. Digital modulation involves the encoding of the
modulating signal as a sequence of binary digits (for example) according to
the value of the modulating signal at discrete intervals.

AMPLITUDE MODULATION
In amplitude modulation (AM) the amplitude of the carrier is made proportional to the
instantaneous amplitude of the modulating signal. The frequency of the modulating
signal, , is always much lower than the frequency of the carrier wave, . The
effect on the carrier wave form can be seen in Fig 1.

Issued November 2006 Chapter 2 : Section 1 77

FREQUENCY MODULATION
In frequency modulation (FM), it is the frequency that is modulated directly by the
modulating signal, though this also affects the phase of the wave. This is shown in
Fig 1(d) which exaggerates the effect for clarity. In practice is always less than
by a factor of many thousands.

FIG 1 AM AND FM WAVEFORMS

78 Chapter 2 : Section 1 Issued November 2006

PHASE MODULATION
In phase modulation (PM), it is the phase that is modulated directly by the
modulating signal, though this also affects the frequency of the wave.
Because FM and PM are so very similar they are often treated as one and the same.
Nevertheless there are differences between them to the disadvantage of PM, and
FM is the technique which is almost always chosen to implement angle modulation.
All future references will be to FM.
The abbreviations AM and FM may be used for either of the terms amplitude
modulation or amplitude modulated, and frequency modulation or frequency
modulated, according to context.

PRODUCTION OF AN AM WAVEFORM
There are many ways in which AM may be achieved and some of these will be
examined in Module 8 which deals with AM transmitters. Despite these differences,
all methods work in what is essentially the same manner. This amounts to the
simultaneous application of the carrier and modulating waveforms to a non-linear
device.
If two frequencies, (the carrier) and (the modulation) are applied to the input of
a linear device the output is their simple sum, as shown in Fig 2. the carrier can be
seen to rise and fall in time with the modulation. This does not achieve the objective
of modulating the amplitude, it just varies the average value of the carrier.

FIG 2

Issued November 2006 Chapter 2 : Section 1 79

On the other hand, if the same frequencies and are applied to the input of a
non-linear device, the output of the system will contain not just the original
frequencies but also components at frequencies and . Fig 3
emphasises the difference in behaviour of the non-linear and linear devices.

FIG 3

In Fig 1(c) the amplitude of the carrier (A) changes to A + a or A – a in sympathy with
the amplitude of the modulating signal as shown in Fig 1(a). The change in the
carrier amplitude (a) is proportional to the change in . These changes are in
synchronism and are caused by the three frequency components , and
continuously moving in and out of phase relative to each other. Over some intervals
they give more or less constructive interference (the carrier amplitude becomes
bigger than A), and over other intervals they give more or less destructive
interference (the carrier amplitude becomes less than A).

80 Chapter 2 : Section 1 Issued November 2006

PRODUCTION OF AN FM WAVEFORM
All methods of producing a FM wave depend upon changing the effective
capacitance or inductance of a tuned circuit, using the modulating signal. The tuned
circuit is part of an oscillator so that the modulating signal has direct control over the
frequency of oscillation. This is shown schematically in Fig 4.

FIG 4

The relationship between the modulating signal and its effect on the carrier is shown
in Fig 5. It is quite different to the case of AM. The amplitude and frequency of the
modulating signal have the following effects upon the carrier.
 Amplitude. The amplitude of the modulating signal controls the frequency
deviation of the carrier. In Fig 5, under the influence of waveform a, the 80
MHz carrier has its frequency changed between the limits 79.95 and 80.05
MHz. For waveform b, the limits are 79.9 and 80.1 MHz. In general the
deviation is proportional to , the instantaneous value of the modulating
signal.

Issued November 2006 Chapter 2 : Section 1 81

FIG 5 FREQUENCY MODULATION WAVEFORM

 Frequency. For modulating signals of the same amplitude but different

frequencies, the time taken for the carrier to go through one complete cycle of
deviation is different. Because the actual deviations are the same (the
modulating signals have the same amplitude) the different cycle times imply
that the rate of change of is greater for the higher modulating
frequency.
So far, the descriptions given have been general, but there are a number of
parameters associated with analogue modulation systems which must be dealt with.
AM and FM systems will be dealt with separately, though there are some terms
which are common to both.

82 Chapter 2 : Section 1 Issued November 2006

AM PARAMETERS
Side frequencies
When a carrier is modulated by a single modulating frequency, 4 frequencies are
produced by the system, the 2 originals plus the sum and difference frequencies. In
AM, the modulating frequency is usually much lower than the carrier frequency. This
causes the carrier and the sum and difference frequencies to be closely grouped,
well above the modulating frequency. This is shown in Fig 6 for and
though normally is much greater than this. It is easy therefore to filter
out the 3 wanted frequencies with a resonant circuit of suitable centre frequency and
selectivity, in this case 10 kHz.

FIG 6

In this example the carrier frequency is and the 2 other wanted frequencies are
and , which are called the side frequencies. At 9 kHz is the lower side
frequency (LSF) and at 11 kHz, the upper side frequency (USF).

Issued November 2006 Chapter 2 : Section 1 83

BANDWIDTH
The difference between the USF and the LSF is called the bandwidth. In the
example given, and shown in Fig 7, the USF is 11 kHz and the LSF is 9 kHz, so the
bandwidth B is 2 kHz. This is the minimum bandwidth needed by any resonant
circuit used to select the modulated carrier.

FIG 7

84 Chapter 2 : Section 1 Issued November 2006

MODULATION INDEX
Refer to Fig 1. It was stated in paragraph 10 that the change in the carrier amplitude
(a) was proportional to the amplitude of the modulating waveform . As
varies, so the ratio of a to A varies. If increases, a increases in step, though it
cannot be greater than A.
The ratio of a to A is called to modulation index (m). It can have a value between (or
equal to) 0 and 1, or it may be expressed as a percentage in the range 0% to 100%,
when it is called the depth of modulation. The modulated waveform is shown in figs.
8(a) and 8(b) for m = 0.4 and m = 1.0. If m = 0 then the carrier is unmodulated. If an
attempt is made to increase m beyond 1, then the carrier becomes interrupted and
the envelope of the modulation is a distorted version of the modulating signal. This
is called over-modulation and is shown in Fig 8(c).

FIG 8 PERCENTAGE MODULATION

Calculating m from a diagram where a and A are know is straight forward. In

practice, where the waveform is displayed on an oscilloscope identifying a and A is
very difficult. It may be simplified by adopting the procedure described below.

Issued November 2006 Chapter 2 : Section 1 85

 Refer to Fig 9 which represents an oscilloscope display of an AM carrier. It is

easy to measure the maximum peak-to-peak of the waveform and also its
minimum peak-to-peak value. These are marked as H and h respectively. It
can be readily shown that:

FIG 9

The calculation of m does not require an estimate of the zero volt axis and further
estimates of a and A, so source of error are eliminated.
 For example, if H is found to be 8 V and h to be 4 V then:

 Often it is required to set up a wave with a certain value of m. This can be

difficult, needing much trial and error. The procedure is simplified by use of
the expression:

For example, if m is to be 0.6, then the ratio of the maximum to minimum

is:

On the oscilloscope it is straightforward to use the variable Y sensitivity to set

the maximum to some convenient value and adjust the modulation to give the
correct ratio.

86 Chapter 2 : Section 1 Issued November 2006

 The measurement of m, or setting of m, can be carried out by means of the

so-called trapezoidal display, shown in Fig 10. the actual values of H and h
are not important, just their ratio (as shown in the preceding sub-paragraph).
The display controls may be adjusted to give a convenient value to one or the
other. The vertical deflection is as normal, ie. the modulated carrier is fed to
the plates. The modulating waveform is fed to the X deflection plates and
the mode selector (or whatever the manufacture calls it) TO X-Y. The
trapezoidal display can be adjusted with the usual controls, though they may
not work in quite the expected manner.

FIG 10

Issued November 2006 Chapter 2 : Section 1 87

FM PARAMETERS
Side Frequencies
A FM carrier has side frequencies just as an AM carrier does, but the number and
position of these side frequencies are not so easy to predict. The analysis is heavily
mathematical, but may be summed up by saying that each modulating frequency
generates an infinite number of side frequencies. Fortunately, the amplitude of these
side frequencies falls off rapidly with displacement from the nominal carrier
frequency, and most of them can be ignored without any serious effect on the
performance of the system.
The use of the terms LSF and USF is not appropriate here, because a single
modulating frequency gives rise to a number of side frequencies, symmetrically
disposed about , as shown in Fig 11, which shows the spectrum of a FM carrier
modulated with a single frequency.

FIG 11

BANDWIDTH
It is immediately obvious from Fig 11 that the bandwidth requirement for a FM carrier
is much greater than that for an AM carrier. This is surprising and runs against
intuition, which would suggest that the bandwidth would be just . Despite the
difficulty in analysing such a spectrum, there is an easy way to calculate
approximately the required bandwidth, it is known as Carson’s rule, and it states that
the bandwidth is twice the sum of the frequency deviation and the modulating
frequency, or:

88 Chapter 2 : Section 1 Issued November 2006

MODULATION INDEX
The modulation index (M) for a FM system (sometimes actually referred to as the
frequency modulation index) relates the maximum frequency deviation from to the
amplitude of the modulating voltage. For a sinusoidal modulating signal of amplitude
, producing a frequency deviation of :

MODULATION BY SPEECH
So far the modulation process has been restricted to a single modulating frequency.
The applications of this kind of modulation are mostly restricted to data transmission
and of more interest in the present discussion is the use of a complex modulating
signal, such as speech.

AM
Referring back to Fig 6 or 7, it can be seen that a single modulating frequency
generates a pair of discrete side frequencies, the LSF and the USF. If the
modulating signal is itself a spectrum of frequencies, then each component of that
spectrum will generate its own LSF and USF, and Fig. 12 shows that the spectrum of
the modulated carrier becomes a band (or 2 bands) or frequencies around the
carrier.

FIG 12

Fig 12 (a) shows the spectrum of an AM wave when the modulating signal is a
continuous spectrum of frequencies from 0 to 10 kHz, and is 100 kHz. Note the
use of the triangular symbol to represent a band of frequencies. It does not imply
that the amplitude of the higher modulating frequencies is greater than that of the
lower ones. It is just a representation of the band.
The bands either side of are referred to as the lower and upper sidebands (LSB
and USB) and the bandwidth is now the separation of the highest component in the

Issued November 2006 Chapter 2 : Section 1 89

USB and the lowest component in the LSB. Because these components are both
generated by the highest frequency component of the modulating signal, it can be
seen that the bandwidth is twice the maximum modulating frequency. Fig 12 (b)
shows the spectrum resulting from modulation by a band limited signal such as that
used for speech in a radio communication system. The start of the LSB and the
USB is displaced from by an amount equal to the minimum frequency in the
modulating signal.

FM
For an FM carrier, remarks similar to those for AM apply, in that each frequency
component of the modulating signal contributes to the final frequency spectrum.
This is correct, though the final spectrum is extremely complex because each
frequency component of the modulating single may produce many side frequencies.
Carson’s rule for the bandwidth still applies, but it must be modified to take into
account the spectrum of the modulating signal:

90 Chapter 2 : Section 1 Issued November 2006

REVISION QUESTIONS
1 An RF carrier wave of frequency 10 MHz is amplitude modulated by an
audio signal of 4 kHz. The lowest side frequency is:

2 If an RF carrier of 500 kHz is amplitude modulated with an audio signal of

3000 Hz, the bandwidth is:

3 If an RF carrier of 500 kHz is amplitude-modulated and has an associated

bandwidth of 16 kHz, the:

4 A simple AM transmitter is modulated with a band of audio frequencies

extending from 420 Hz to 1800 Hz. The transmitter therefore requires a
bandwidth of:

5 A 5 MHz signal is amplitude modulated by a band of frequencies between 1

and 2 kHz. The separation between lowest and highest frequencies transmitted
is:

6 The peak-to-peak amplitude of an AM carrier varies between 120 volts and 80

volts. The percentage modulation is:

Issued November 2006 Chapter 2 : Section 1 91

7 When measuring modulation depth by an oscilloscope, the long side of the

trapezium is found to measure 9 cm and the short side 6 cm. The modulation
depth is:

8 Increasing the frequency (but not the amplitude) of the modulating signal
applied to an FM transmitter will cause the:

9 Increasing the amplitude of the modulating signal to an FM transmitter will

cause:

REVISION QUESTIONS (SHORT ANSWER)

1 How is the information carried in an FM transmission?
Answer………………………………………………………………………….
2 What causes the carrier frequency to vary?
Answer………………………………………………………………………….
3 What causes the rate of frequency change?
Answer………………………………………………………………………….
4 An RF carrier is being frequency modulated by a fixed AF frequency. What
will be the changes in the resultant output waveform if the modulating frequency:

92 Chapter 2 : Section 1 Issued November 2006

SECTION 2 : RADIO SYSTEM PRINCIPLES

INTRODUCTION
Radio systems may be classified in many different ways. In whatever way they are
classified, they are concerned with the transmission of information. It is the nature of
the information and the means by which it is transmitted which differentiates between
them. This section will outline those differences.

ELEMENTS OF A BASIC RADIO SYSTEM

Fig 1 shows the minimum requirements for a radio system. It is not specific to any
particular type of system. Its elements are:
 Information Source. The information source may be one of many
types. For example, in a broadcast radio system it would be speech or
music, in a telemetry system it would be data, usually in digital form, in a
navigation aid it might be information on range and bearing.
 Transmitter. The transmitter has to modify the input from the source so
that it is in a form appropriate to the channel through which it is to pass.
It has to provide enough energy to overcome the losses in the channel.
 Channel. The channel can be one of 2 forms, either free space or air for
wireless transmissions, or a landline. A landline can support
transmission at frequencies below those in the radio band, eg. a public
address system, but this is not of concern here. It can also support
transmission at frequencies above the radio band as in optical fibre
systems which will be dealt with in another module. In many systems the
channel supports transmission in both directions simultaneously.
 Receiver. The receiver has to extract the original information from the
output of the channel. Generally speaking it has to undo the processes
carried out in the transmitter and prepare the signal for use at its output,
eg. driving a loudspeaker.

FIG 1

Issued November 2006 Chapter 2 : Section 2 93

Output. The output from the system may be just that, sound, vision or
perhaps printer matter. On the other hand it may be an outcome, in that
the information may be fed to another system which carries out a function
on the basis of that information, as in an autopilot which may use
navigation information from radio beacons.
The basic system may be modified in a number of ways. Most commonly, it may be
a 2-way system in which there is a transmitter and receiver at each end of the
channel. This gives obvious advantages by making the process supported by the
system an interactive one.

CLASSIFICATION OF RADIO SYSTEMS

Radio systems may be classified in many different ways but for our purposes they
are:
 radio telephony.
 radio telegraphy.
 radio navigation.
Each category is briefly described in the following paragraphs.

RADIO COMMUNICATIONS
The expression ‘radio communications’ has traditionally referred to speech
transmission, though it includes much more than this:
 Broadcast. Broadcast covers all commercial ‘radio’ and television systems
and they are characterised by being one-way systems. It is not important to
this course except in providing examples of standard technical practices in the
field of radio engineering.
 Radio Telephony. Radio telephony (properly abbreviated to RT but
traditionally written as R/T) is the 2-way, ground-to-ground, ground-to-air and
air-to-air radio communication of speech. This is of particular interest and
much of the following work on this course assumes that this is the kind of
system being used. It should be noted that conventional RT is gradually
being replaced by digital data links.
 Radio Telegraphy. Radio telegraphy refers to the transmission of data over
a radio communication link as, for example, in a telemetry system. It is a term
which is obsolescent and the phrase ‘data transmission’ is to be preferred to
‘telegraphy’.
Radio telephony and radio telegraphy are the important sub-divisions and
traditionally have used similar but distinct techniques to achieve their particular ends.
 Radio Telephony. The basic principle behind radio telephony is that a RF
carrier is speech modulated by one of the techniques dealt with in Module 4,
using either amplitude, frequency or pulse modulation. These methods will be
dealt with at some length in later sections of the course and no further
description of them is given here.

94 Chapter 2 : Section 2 Issued November 2006

RADIO TELEGRAPHY
Radio telegraphy uses RF signals as the carrier of data through the channel. If the
channel is free space, the older term ‘wireless telegraphy’ is sometimes used, but
radio telegraphy may use a landline as the channel.

FIG 2

Typically, a data transmission system uses some form of encoding such as ON-OFF
keying (OOK) – the classic example of which is Morse code, phase-shift keying
(PSK) or frequency-shift keying (FSK). Examples of these are shown in Fig. 2(a), (b)
and (c) respectively. Notice that PSK and FSK are cases of PM and FM taken to an
extreme. The Morse code example is distinctly different from the others because the
ON-OFF cycling of the transmitter contains the information, the other 2 simply use
the different phases or frequencies to encode 1 or 0. Because the output from a
PSK or FSK system is not immediately understandable to an observer, it has to be
converted to a human readable form. By contrast, trained operator can understand
Morse code directly.
Whether OOK represents Morse code or some other code is not the issue. What is
important is that for the on-period, the wave is continuous. This gives rise to the
name of the technique, which is continuous wave (CW). When the transmitter is
keyed on, the carrier is unmodulated and in a conventional receiver this would give
no sound from the loudspeaker. In order to hear when the transmitter is keyed on,
there needs to be a tone produced in the receiver. This is done with a beat
frequency oscillator (BFO) whose output is mixed with the incoming carrier to
produce the audible tone.

Issued November 2006 Chapter 2 : Section 2 95

An alternative method is for the carrier to be modulated before it is transmitted. This

is the modulated continuous wave (MCW) system and it allows OOK transmissions
to be received on a standard domestic broadcast receiver. CW and MCW
transmitted waveforms are shown in Fig. 3.

FIG 3

The pace of modern developments means that the boundaries between

speech/telephony and data/telegraphy have become blurred, especially in view of
widespread use of pulse code modulation (PCM). When comparing traditional
systems, however, they have their advantages and disadvantages.

ADVANTAGES OF TELEPHONY
No skilled operator is needed (to perform encoding/decoding) and the final output is
immediately intelligible to the person receiving. Telegraphy does need skilled
operators to prepare messages in a form which may be transmitted, eg. operators
trained in Morse code or in using teleprinters.
Information can be assimilated by the person receiving it as it arrives, ie. it is a real-
time process. Thus it is a fast means of communication. On the other hand
telegraphy is relatively slow in terms of the time it takes to generate the message,
encode it, transmit it, receive and decode it. Extremely large amounts of data can be
transmitted very quickly, but there can be a built-in time lag due to the need to
encode/decode and present the data.

96 Chapter 2 : Section 2 Issued November 2006

ADVANTAGES OF TELEGRAPHY
The content of a message is not readily understood unless the encoding process is
known, so confidentiality is maintained. With a speech transmission, then so long as
the language is understood there is no bar to understanding the content of the
message.
It is usually possible to have a narrower bandwidth when transmitting data, but much
depends upon the precise form of modulation used.

RADIO NAVIGATION
Radio navigation generally refers to the use of radio equipment to enable aircrew to
find their way from point to point or as aids to landing an aircraft. If it is extended to
include radio equipment which helps determine the position of an object (be it the
aircraft itself, a target, the ground or meteorological phenomena) then it also covers
the majority of radio systems as well.
Some radio navigation systems are airborne based relying on the reception of
signals from one or more beacons on the ground. Others may be ground based
systems which require transmissions from an aircraft. A third type of system has
appeared in recent years based on signals received from 3 or more satellites. All
those systems are discussed in more detail in later modules.

Issued November 2006 Chapter 2 : Section 2 97

REVISION QUESTIONS
1 List 3 types of communication channel.

2 List 3 functions of radio communications

3 One advantage of ON-OFF keying using Morse code compared to PSK and
FSK is:

4 A major advantage of telephony over telegraphy is that it:

98 Chapter 2 : Section 2 Issued November 2006

SECTION 3 : CW TRANSMITTERS
INTRODUCTION
Fig. 1 shows the BSD of a CW transmitter. There are 3 essential stages:
 master oscillator
 buffer amplifier
 power amplifier
There are often 2 others:
 frequency multiplier
 driver or intermediate power amplifier
If information is to be transmitted then some means of keying the transmitter will be
needed. This is usually included at a low power stage and involves interrupting the
current in one of the devices. Whichever combination is used an aerial is essential
as an output transducer.

FIG 1

Issued November 2006 Chapter 2 : Section 3 99

STAGE FUNCTIONS
 Master Oscillator. The master oscillator (MO) is the source of RF energy
which determines the final output frequency of the transmitter. There are
many different kinds of oscillator which can be used and some of them will be
dealt with later in this chapter. In order to produce a sine wave at a constant
frequency and with a constant amplitude it must operate under stable
conditions.
 Buffer Amplifier. As its name suggests this stage acts as a buffer between
the MO and the power amplifiers. It is able to deliver RF energy (under the
control of, but at greater power levels than, the MO) to give an adequate drive
signal to the next stage.
 Frequency Multiplier. This stage is able to generate a harmonic of the input
RF, so that the transmitted frequency is higher than that generated by the
MO. This is because it is usually easier to generate a stable oscillation at low
RF than at a high one. There may well be more than one stage of
multiplication. It is sometimes called a harmonic amplifier.
 Driver Amplifier. The driver or intermediate power amplifier (IPA) may not
be needed in low power systems but is essential where high output powers
are wanted. It provides the necessary drive power to the input circuit of the
output stage. There may be some frequency multiplication included in this
stage.
 Power Amplifier. The radio frequency power amplifier (RFPA) has to provide
sufficient output power to overcome the losses in the transmission channel. It
must also be as efficient as possible, especially in high power systems, so as
to minimise the heat dissipation in the circuit components.
 Aerial. The aerial (or antenna) acts as an efficient transducer between the
electrical energy in the transmitter circuitry and the electromagnetic energy
which is propagated. It also has an impedance matching function and usually
has the ability to concentrate the radiated energy in certain directions.

100 Chapter 2 : Section 3 Issued November 2006

Fig 2 shows the waveforms to be expected at the output of each stage.

FIG 2

Issued November 2006 Chapter 2 : Section 3 101

NOTES:

102 Chapter 2 : Section 3 Issued November 2006

SECTION 4 : OSCILLATORS
INTRODUCTION
The operation of basic oscillators has been dealt with at an earlier stage of this
course. However, it is useful to recall some fundamental principles which apply to all
oscillators, of whatever type.
 Oscillators are simply amplifiers which derive their input from their output via a
feedback network. They have to have enough gain to overcome the
attenuation in the feedback loop. To sustain oscillations the loop gain must
be unity. Ie. for a voltage gain and feedback fraction . From
cycle to cycle the loop gain might not be exactly unity, but on average it must
be so.
 The phase of the fed back signal must be such that it sustains the oscillations,
ie. it must be positive feedback.
 There must be a frequency determining network in the loop, which controls
the output frequency.
 In a signal generator, the sinusoidal output must not be distorted to any
significant degree. This is not so important in transmitters where the output of
the oscillator is often deliberately distorted in frequency multiplier stages.
It should be noted that although a circuit may be operated in Class C bias, it needs
to be in Class A to get it working initially. Many transistor oscillators work in Class A
all the time. They achieve a loop gain of unity by operating out to the extremities of
their load line. Under these conditions the of the transistor falls to a low value so
that the average over the whole cycle is affected. The basic is greater than
that needed to give a loop gain of unity, but it is reduced because of the proportion of
the cycle time which the device spends at the ends of the load line.

Issued November 2006 Chapter 2 : Section 4 103

FREQUENCY STABILITY
The frequency stability of oscillators is of major importance in transmitters, especially
in applications where there is limited channel spacing. There are a number of
factors which affect it. The main ones are:
 Loading. Variations in the load on an oscillator cause the output frequency
and amplitude to vary, indeed, overload it and it will stop oscillating. They are
therefore operated under as near constant conditions as possible. They are
under-run, ie. with a light load, by using them to drive buffer stages which are
designed to take as little input current as possible.
 Temperature. Changes in temperature have small but important effects on
the dimensions, and hence the values, of inductors and capacitors. It also
affects the parameters of semiconductor devices, including inter-electrode
capacitances (IECs), which may be used at UHF as part of the tuning of the
transmitter. Any changes in L and C affect the resonant frequency, they can
be minimised by one or more of the following:
o The temperature may be kept constant by placing the sensitive
components in an oven whose temperature is thermostatically
controlled within close limits. This may introduce the need for a short
warm up period before the full stability is achieved, but in modern
equipments this is not a serious problem.
o Aurtomatic frequency control circuits may be used but these need an
extremely stable reference oscillator to be effective.
o Capacitors with negative temperature coefficients of capacitance may
be used to keep the LC product constant, but this is only effective over
a limited range of frequencies

 Device Changes. The IECs of devices (normally enhanced by the Miller

effect) are usually in parallel with the tuned circuit. Changes in these will
affect the tuned frequency and the amplitude of the oscillation. There are 2
main reasons why these IECs might change.
o Like most transistor parameters, IECs are affected by the dc conditions
under which the device is operated. It is important to keep the
operating point steady and this can be done primarily by keeping the
supply voltage constant. With modern power supplies this should not
be a problem because they are usually well regulated. This may be
supplemented by the generous use of decoupling networks in the
supply lines to the oscillator
o If the device becomes unserviceable and has to be replaced, the IECs
of the replacement will almost certainly be different to those of the
original. Most circuits will oscillate at a different frequency. This is
especially a problem at the higher end of the VHF band and at UHF.
At lower frequencies (in the HF band) circuit design can overcome this
problem because the IECs are much smaller than the tuning
capacitors.

104 Chapter 2 : Section 4 Issued November 2006

 Vibration. Vibration and shock can have undesirable effects on oscillators.

The disposition of components might change slightly, having an effect on
stray capacitance and inductance. This is usually overcome by good
mechanical design or, if necessary, mounting critical components (or an entire
equipment) on shock and vibration absorbing supports.
 Screening. Electrostatic and electromagnetic screening of tuned circuit
components may be used to protect them from the effects of other high power
equipment operating in close proximity. However, this is an extremely
complicated problem falling within the realm of electromagnetic compatibility
(EMC) , which is a lifetime’s study in itself!

CRYSTAL OSCILLATORS
One of the piezo-electric materials commonly used in oscillators is quartz, a naturally
occurring dielectric mineral. The piezo-electric effect is simple in concept:
 A potential difference is developed between a pair of opposite faces of a
crystal when pressure is applied between a different pair of opposite faces.
Conversely, if a potential difference is applied between the pair of opposite
faces, there is a mechanical strain between the other pairs of opposite faces,
as one pair moves apart the other pair moves together. These movements
are proportional to the voltage but are nonetheless extremely small.
 If the crystal vibrates mechanically, an alternating voltage is produced
between opposite faces. On the other hand, if an alternating voltage is
applied between a pair of opposite faces, it produces an alternating
displacement of the other faces.
In an oscillator, the crystal can be shock excited into vibration by simply switching of
a suitable circuit. (The sudden change in dc potentials around the crystal is exactly
analogous to hitting a bell with a hammer). Its internal losses are extremely small,
which is mainly responsible for the high Q. If these losses can be made up by using
a fraction of the output voltage to excite the crystal with the correct phase (positive
feedback) then the circuit becomes a crystal-controlled oscillator.
The frequency stability of a quartz crystal is very high compared with that of an LC
circuit. The table below compares the stability of LC and crystal-controlled
oscillators.

Ambient Air Constant Temperature Oven

LC 1 in 103 1 in 105
Crystal 1 in 106 Better than 1 in 108

TABLE 1

Issued November 2006 Chapter 2 : Section 4 105

OVERTONE OPERATION
When a crystal vibrates it does so as a whole, ie. opposite sides go in and out
together. In an overtone crystal, adjacent planes of the crystal vibrate in antiphase
to each other. These 2 situations are shown in Fig. 1 where the overtone (harmonic)
used is the 3rd. The response to the higher order harmonics declines but even so
crystals are available which give up to about the 5th harmonic. Crystals for harmonic
operation are specially cut to inhibit the fundamental mode, but, extra circuit
components are often included to ensure that not only the fundamental but other
unwanted harmonics are suppressed.

FIG 1

Overtone operation is very common, particularly at frequencies above a few

megahertz. Fundamental operation can be achieved up to nearly 100 MHz, but this
is only a recent development and most current equipments do not use such crystals.

106 Chapter 2 : Section 4 Issued November 2006

REVISION QUESTIONS
1 Instability in the oscillator of a radio transmitter could cause:

2 List the 5 major causes of oscillator instability:

3 In an oscillator:

4 A common cause of an amplifier running into oscillation is:

5 A quartz crystal used as an oscillator makes use of:

6 The main factor affecting the frequency of a quartz crystal is the:

7 One disadvantage of a crystal controlled oscillator is:

Issued November 2006 Chapter 2 : Section 4 107

8 Crystal controlled oscillators are frequently used in transmitters because:

9 The cause of parallel resonance in a crystal oscillator is:

10 When used in an overtone mode the highest harmonic normally used is:

11 Overtones are produced in a crystal by:

108 Chapter 2 : Section 4 Issued November 2006

SECTION 5 : FREQUENCY SYNTHESIS

INTRODUCTION
A conventional oscillator is not capable of generating the wide range of frequencies
needed in a modern transmitter, with the necessary stability. A crystal controlled
oscillator has the stability but not the tuning range. Frequency synthesis is the name
given to the various techniques which give both the accuracy of crystal control and
the variability which comes from the use of LC tuning. The only limitations is that the
tuning is in discrete steps, rather than being continuous. There are 3 methods which
are of interest:
 single crystal direct frequency synthesis (no longer used)
 multiple crystal direct frequency synthesis (no longer used)
 indirect frequency synthesis

INDIRECT FREQUENCY SYNTHESIS

Fig 1 shows the basic arrangement. The output from the generator at hertz, drives
the PLL. It is compared with the output of the circuit ( is an integer) in the phase
comparator. In order that the output of the circuit also be at hertz, the voltage
controlled oscillator (VCO) must produce its output at hertz. Changing changes
the output frequency of the VCO. The spacing between the possible frequencies
from the VCO (the channel spacing) is set by the crystal oscillator output frequency.
For close spacing, say 9 kHz, the crystal oscillator would have to be at 9 kHz, which
is too low for a conventional crystal controlled oscillator.

FIG 1

Issued November 2006 Chapter 2 : Section 5 109

This can be overcome by dividing the output of a suitable crystal controlled oscillator
by (also an integer), as shown in Fig 2. By dividing down to the required channel
spacing, a suitable choice of gives an output frequency from the VCO of .

FIG 2

The BSD of its frequency synthesiser is shown in Fig 3. The crystal controlled
oscillator runs at 3.2 MHz and is divided by 512 to give 6.25 kHz. Effectively this is
the channel spacing of the synthesiser, but because of the use of a pair of frequency
doublers after the synthesiser, the transmitter channel spacing is 25 kHz.

FIG 3

110 Chapter 2 : Section 5 Issued November 2006

The phase comparator gives a dc control voltage (when the system is in lock) which,
after passing through the low pass filter, drives the VCO in the range 2 MHz to
2.05625 MHz. This corresponds to a transmitted range of 8 MHz to 8.225 MHz.
After buffering, the synthesiser is fed to the block. The value of is
programmable over the range 320 to 329. the programming is carried out in
response to the 4 inputs which carry the channel selection (0 to 9) in binary coded
form.
In a combined transmitter and receiver (transceiver) the frequency synthesiser would
be used as part of both the transmitter and receiver, as shown in Fig 4. Most of the
amplification in the receiver is carried out at a fixed frequency, lower than the
incoming RF, but higher than the frequency of the message signal. This is the
intermediate frequency (IF). It is fixed despite the fact that the RF may be at one of
a large number of different channel frequencies.
The transmitter frequency, , is the sum of the synthesiser frequency, , and the IF,
. The received frequency , is the same as . It is mixed with a sample of ,
and the difference frequency, , is selected with a suitable tuned circuit.

FIG 4

Issued November 2006 Chapter 2 : Section 5 111

REVISION QUESTIONS
1 Frequency synthesisers are preferred to conventional oscillators in
transmitters because they:

112 Chapter 2 : Section 5 Issued November 2006

SECTION 6 : BUFFER AMPLIFIERS

INTRODUCTION
The simplest transmitter is just an oscillator connected to an aerial via a switch.
When the switch is closed, the load presented by the aerial to the oscillator would, at
best, reduce the amplitude of the oscillation and change its frequency, and at worst,
stop the oscillator from oscillating. Oscillators work best with a high constant load.
This can be achieved by feeding the output of the oscillator into a buffer amplifier,
and then feeding the aerial from the output of the buffer.

USING FET’S
For satisfactory operation the buffer should have a high input impedance and
enough gain to drive its load. The need for a high input impedance suggests that a
FET would be the best device, and for linear operation without distortion, it should be
biased in Class A. Such a circuit is shown in Fig 1.

FIG 1

Issued November 2006 Chapter 2 : Section 6 113

USING BIPOLAR TRANSISTORS

If a bipolar transistor is used then it will need to use a step-down input transformer to
give the necessary impedance matching. This would reflect a high impedance into
the primary winding. The bias resistors would be chosen so as not to seriously shunt
the already low impedance of the transistor. This is shown in Fig. 2.

FIG 2

The circuit in Fig 1. has a series fed output tuned circuit but is more common to have
shunt fed loads as shown in Fig 2. This will reduce to a minimum, the effect of stray
and hand capacitance on the tuning of the load. It also reduces to a minimum the
RF currents circulating in the PSU and the supply lines.
The disadvantages of shunt connected loads is that the output voltage is developed
across an RF coil rather than a tuned circuit and both the gain and selectivity are
less than could be achieved with a series connected load. In most applications,
however, this is not a serious drawback.

114 Chapter 2 : Section 6 Issued November 2006

MODERN APPLICATIONS
In a number of modern applications, it is possible to use an untuned source follower.
When isolation is critical either cascaded buffers or a dual gate FET buffer may be
used. This latter circuit is shown in Fig 3. By replacing a signal earth on the second
gate, the isolation between the source and the input gate can be made extremely
high. The dc voltage on the second gate gives a measure of control over the gain of
the state, a stage gain of 10 (20 dB) can easily be achieved. Additional isolation
can be gained by using a low value coupling capacitor between the oscillator and the
buffer and offsetting the additional attenuation against the stage gain.

FIG 3

Issued November 2006 Chapter 2 : Section 6 115

REVISION QUESTIONS
1 One purpose of a buffer amplifier in a CW transmitter is to:

2 In the basic 3-state transmitter the buffer amplifier would:

3 Shunt fed tuned circuit loading of a transistor buffer amplifier is preferred to

series fed because it:

d minimises the effect of hand and other stray capacitances

4 A FET is preferable to a bipolar transistor as a buffer amplifier because it:

5 Which of the following could be used as a buffer amplifier?

116 Chapter 2 : Section 6 Issued November 2006

SECTION 7 : FREQUENCY MULTIPLICATION

INTRODUCTION
It is often the case that the transmitted frequency is at too high a value for
satisfactory operation of the oscillator. Particularly with crystal controlled oscillators,
there is difficulty in operating them beyond about 80 MHz without overtone operation,
and in any case their best stability is achieved at lower frequencies. In these cases it
is necessary to run the oscillator at a suitably low frequency and then, after buffering,
use frequency multiplier stages.
The common feature of all frequency multipliers is that they usually use Class C
fixed bias to distort the input signal, sometimes automatic bias may be used. This
gives an output current which is rich in harmonics. The output tuned circuit then
filters out the unwanted harmonics to leave the required frequency. Although the
output current is heavily distorted, the filtering action of the tuned circuit reduces the
proportion of unwanted frequencies sufficiently to make the output voltage almost
sinusoidal, as explained previously.
The greater the distortion of the input signal, the richer the harmonic content of the
output current, but the energy level of the higher order harmonics diminishes rapidly.
For this reason it is usual to use only the 2nd or 3rd harmonic, though it would be
possible to isolate and amplify up to the 6th or 7th.
Fig 1 shows a basic transistor frequency doubler. The transistor has no base bias
network, being switched on and off by the input signal. The input circuitry comprises
a dc blocking capacitor followed by an asymmetric, unbalanced, low pass, pi-
filter, and .

FIG 1

Issued November 2006 Chapter 2 : Section 7 117

Though not essential as part of the multiplier, the filter has 2 functions which are of
practical importance and are commonly seen in HF/VHF/UHF circuitry.
 Firstly it matches the output impedance of the buffer to the low input
impedance of the transistor.
 Secondly it has a sharp cut off above the input frequency, , avoiding the
presence of unwanted frequencies at the input of the transistor. These could
inter-modulate with (due to the non-linearity of the input characteristic) to
give unwanted frequencies which fall within the pass band of the output tuned
circuit.
is an RF choke which has a high reactance at but gives the base a low
resistance dc return to the PSU. Without it, would charge negatively and alter the
bias from its design point. Both the primary and secondary tuned circuits in the
output filter out the unwanted harmonics.
Sometimes a series tuned filter for the fundamental is included (shown dotted) to
help eliminate the largest amplitude component. The circuit could be modified to
give the 3rd harmonic rather than the 2nd , but the output would be lower in
amplitude.
An FET may be used as a frequency multiplier but because the nature of its bias
requirements it usually has the gate taken to a fixed negative bias line. This is
shown in the following 2 circuits which are of special note, the push-push frequency
doubler and the push-pull frequency trebler are particularly useful because of the
harmonic structure of the output currents.

THE PUSH-PUSH FREQUENCY DOUBLER

The push-push doubler is shown in Fig 2. The FETs are biased beyond cut-off by
the negative bias voltage fed via the input transformer secondary. is an RF choke
which blocks the RF from the bias supply. The gates are driven in anti-phase so that
they produce positive-going pulses of current alternatively as shown in Fig 3.
Because the pulses occur twice per cycle, the combined current waveform has its
fundamental at twice the input frequency. This combined current is ‘pushed’ through
the output tuned circuit by each FET in turn.

FIG 2

118 Chapter 2 : Section 7 Issued November 2006

The output tuned circuit, only responds to the fundamental of the output current
which is at twice the input frequency. is to decouple the RF current from the
positive supply rail. The advantage of this form of doubler is that the unwanted
frequency with the largest amplitude, which normally exists in the circuit of a simple
multiplier is not present. In fact there are no odd-numbered harmonics in this
waveform, they are self-cancelling (assuming matched FETs and accurately centre-
tapped transformers). This ensures the minimum distortion to the final output
waveform.

FIG 3

Issued November 2006 Chapter 2 : Section 7 119

DIODE MULTIPLIERS
It is possible to use a diode or a pair of diodes in a multiplier. The function of the
diode is to distort the waveform of the input signal, which it does by virtue of its
exponential input characteristic, and the fact that it rectifies the input signal. Fig 4
shows a simple multiplier using a germanium point contact diode. Its efficiency is
low and operation is limited to HF and the lower VHF band. For higher frequencies,
up to UHF, Varactor (varicap) diodes and step recovery diodes may be used with
great efficiency and these have the ability to multiply by factors of 10 or more.

FIG 4

OSCILLATOR MULTIPLIERS
It is often convenient to combine the multiplier with the oscillator. This gives a saving
in both weight/space and component count. Fig 5 shows an inverted crystal
controlled Colpitts oscillator. The sole difference is that the output tuned circuit
resonates at the selected harmonic of the input frequency.

BUFFER MULTIPLIERS
Buffer multipliers are used for the same reason that oscillator multipliers are used.
The buffer input must still present a light constant load to the oscillator, but it must be
biased in Class B or preferably Class C to give the maximum energy at the wanted
harmonic. However, if Class C is used it must be by fixed bias as automatic bias
would load the oscillator during those parts of the cycle when base/gate current is
drawn.

120 Chapter 2 : Section 7 Issued November 2006

FIG 5

Issued November 2006 Chapter 2 : Section 7 121

REVISION QUESTIONS
1 The reason for using frequency multipliers in a transmitter is:

2 If Class C amplifier is used as a harmonic generator, the power output:

3 A power amplifier is used as frequency multiplier. Class C bias is used rather

than A or B because:

4 The collector current waveform of a single transistor frequency multiplying

stage contains:

5 Class C bias is used in frequency multiplier stages of a transmitter in order

that:

6 The reason why frequency multipliers normally only multiply by a factor of

or is because:

7 The reason why point contact diodes can be used as frequency multipliers is
that:

122 Chapter 2 : Section 7 Issued November 2006

SECTION 8 : RF POWER AMPLIFIERS

EFFICIENCY
This is a topic of great importance with high power RF amplifiers. In general, Class
C gives the greatest efficiency, whilst Class A gives the greatest linearity with the
minimum produce of unwanted harmonics.
Because efficiency is defined as the ratio of RF output power to dc input power, it
can be seen that the higher the average supply current in the device, the lower the
efficiency will be.

Example
When operated in Class A a transistor produces 0.535 A of RF current in a 70 load.
It draws 1 A from a 50 V supply. Its efficiency is:
RF power out
dc power in

When operated in Class C it draws 0.14 A from the supply, producing 0.267 A of RF
current in the same load. Its efficiency is:
RF power out
dc power in

This illustrates 2 points:

 Firstly that efficiency increases in Class C
 Secondly that the available output power goes down.
In order to cope with the requirements for different power outputs, amplifiers may be
constructed in different ways, single-ended and push-pull (double-ended).

SINGLE-ENDED AMPLIFIERS
Fig. 1 shows a typical single-ended, Class C power amplifier. The circuit appears
much more complicated than it really is. It is characterised by an apparent lack of
bias and no obvious load for the transistor. There are also a number of components
whose function would appear to be obscure.
The heart of the amplifier is, of course, the transistor. Its load is resonating with
its own output capacitance. Most of the other components are for impedance
matching. As mentioned in the section on frequency multipliers, an RF transistor
may have an exceptionally low input impedance, less than 1  is quite common.
Source and load impedances are commonly 50 , so there needs to be impedance
matching networks feeding into the base and from the collector.
At the input and provide this matching. an RF choke, provides a low
resistance path to earth from the base, while having a very high reactance to avoid
shunting the matching network. At the output and match the 50  load to
the aerial. works with to keep the RF off the supply line. In addition there

Issued November 2006 Chapter 2 : Section 8 123

would be, in a practical circuit, various decoupling capacitors and filter networks may
be included at critical points to prevent the signal from appearing where it is not
wanted, eg. the supply line, or to prevent instability due to accidental positive
feedback.

FIG 1

124 Chapter 2 : Section 8 Issued November 2006

PARALLEL OPERATIONS
If operated in Class C, FETs (and bipolar devices) have relatively low outputs,
despite higher efficiency. This can be offset by using multiple devices in parallel.
The manufacturer uses this technique to produce high power devices. The internal
structure is organised with emitters and collectors in parallel, in what are described
as inter-digital structures. Individual devices of this nature may still be paralleled,
and an example of this is shown in Fig 4. The nature of this circuit is identical to that
shown in Fig 1. The only difference is that the RF choke, , is connected to the
input side of and , rather than the bas sides, so that it may act for both
transistors but retain symmetrical loading of the input matching circuitry. The effect
of paralleling any devices are:
 The load on the previous stage is doubled and it must be able to source the
necessary drive current. This implies that its output impedance must be low.
 If sufficient drive power is available, then the output power is proportional to
the number of devices.

FIG 4

 Output impedance is reduced in proportion to the number of devices.

 Inter-electrode capacitances are increased. This is not a problem with bipolar
transistors, but with FETs it can lead to reduced stability (see the later section
on neutralisation). It will also affect the maximum frequency of operation with
FETs, but with bipolar types, the shunt reactances are very low in any case
and any further reduction is not too significant.

Issued November 2006 Chapter 2 : Section 8 125

DOUBLE-ENDED (PUSH-PULL) AMPLIFIERS

The other common method of operation, particularly with Class C bias, is to operate
the devices in push-pull. Such a circuit biased in Class A or AB and suitable for use
throughout the HF band, is shown in Fig 5. Impedance matching can be
accomplished entirely by the use of the input and output transformers which, while
not as efficient as the matching networks in Fig 1 for example, are adequate. At
higher frequencies the iron losses would become progressively worse.

FIG 5

Resistors and are of low value to swamp any difference in the input
impedance of the individual transistors so as to maintain a balanced load on the
input transformer. The bias supply could be a potential divider chain, but to get a
stable bias voltage, a very low value of resistance would be used. Better results
would be had from the type of circuit shown in Fig 2. and are RF chokes to
keep the RF from the bias circuitry.
A similar circuit using FETs is shown in Fig 6. This is very similar to its valve
predecessor and uses the same type of automatic Class C bias using and
. Like all push-pull circuits, the use of centre-tapped output transformer avoids
the tendency for saturation which affects single-ended designs, so allowing higher
powers to be achieved. The tuning of ensures that the harmonics of the
current waveform are suppressed and only the fundamental is passed.

126 Chapter 2 : Section 8 Issued November 2006

FIG 6

AERIAL COUPLING CIRCUITS

At RF the matching between the RFPA and the aerial becomes a major factor in the
overall performance of the system. Mismatches not only prevent maximum power
from being delivered to the aerial from the simple fact that the amplifier load is the
wrong value, but the standing wave generated on the transmission line can create
conditions at the amplifier output which result in poor performance or worse, damage
being done to the output stage.
The output of the power amplifier has to be fed to the aerial via a suitable
transmission line. This line will normally be chosen to be co-axial cable for an
unbalanced (end-fed) aerial, and twin wire feeder for a balanced (centre-fed) aerial.
Coupling circuits from the RFPA to the transmission line can take a number of forms.

Issued November 2006 Chapter 2 : Section 8 127

BALANCED AERIAL
Fig 7 shows a balanced aerial fed from a series-loaded amplifier (Fig 7a) and a
shunt-loaded amplifier (Fig 7b). The isolation afforded by inductive coupling allows
the 2 lines to have the same impedance to earth, ie. to be balanced. The turns ratio
gives the necessary impedance matching.

FIG 7

128 Chapter 2 : Section 8 Issued November 2006

UNBALANCED AERIAL
Fig 8 shows 2 forms of an unbalanced coupling circuit. In both circuits the turns ratio
sets the matching, but for the series-fed load (Fig. 8a) the separate windings are
needed to give the dc isolation of the supply. In the shunt-loaded case (Fig 8b)
gives the isolation and a tapping on the tuning coil sets the impedance match.

FIG 8

PI COUPLING
The pi coupler, shown in Fig 9 has the advantage over the other types that it acts as
a low pass filter, so minimising the harmonic content of the transmission. It can be
used with either a balanced (Fig 9a) or unbalanced (Fig 9b) aerial. The tuning and
the input impedance matching is mainly effected by and the output impedance
matching by though there is some interaction of these capacitors.

FIG 9 (A)

Issued November 2006 Chapter 2 : Section 8 129

FIG 9(B)

In Fig 9b, the tuning is by with the parallel combination of the transistor’s output
capacitance with . is a dc block and its reactance will be negligible compared
with that of . For maximum power transfer, the input impedance to the pi network
should be the same as the output impedance of the transistor.
If the aerial current is monitored and adjusted for maximum current, it will indicate
when the tuning is correct. Adjustment of will match the output impedance of the
network to the input impedance of the line (the same as the aerial). As the
impedances become matched the current should rise further. However, any change
in will affect the tuning slightly, so that will need some further adjustment to
bring the system back on tune. Unfortunately this may slightly upset the output
impedance matching so that the adjustment process needs to be repeated until no
further improvement is possible.
It may be used with aerials that work over a wide frequency range, but because the
aerial’s impedance is sensitive to frequency, the network needs to be set up again at
every major frequency change. This can lead to a large number of switched, preset
components in the pi coupler, to cater for the range of transmitted frequencies.

130 Chapter 2 : Section 8 Issued November 2006

REVISION QUESTIONS
1 To increase the efficiency of a Class C RF PA the:

2 The PA stage of a Tx is 75% efficient and has a standard (quiescent) voltage

of 500 V. If the RF output is 2 amps rms into a 50 load, the direct component of
the anode current of the PA is:

3 The RF power amplifier in a transmitter is normally operated with Class C bias

because:

4 Parallel operation of FETs may be used in RF amplifiers because:

5 High power class C RF amplifiers must be driven by power amplifiers. This is

because:

6 A power amplifier gives an output of 12 watts. The is 60 volts and the

average current is 300 mA. The efficiency of the stage is:

Issued November 2006 Chapter 2 : Section 8 131

7 One advantage of a pi coupling is that it:

132 Chapter 2 : Section 8 Issued November 2006

SECTION 9 : PROBLEMS AT VHF AND UHF

VHF AND UHF EFFECTS
At frequencies which fall within the VHF and UHF bands, effects which are present
but of little consequence at HF, can create problems which become more severe as
frequency increases. Although these are often referred to as VHF or UHF effects,
there is no hard and fast division between HF and VHF and UHF, other than for our
convenience in classifying the bands. The changes from one band to the other are
gradual, and whether or not an effect is a problem depends as much on the use for
which an equipment is designed as on the frequency at which it is operated.

SIGNIFICANT EFFECTS
 Skin Effect. The skin effect is due to the inductance which every conductor
has. A straight conductor has about 15 nH of self inductance per centimetre.
The rapidly changing flux due to alternating current induces a back emf, which
because of the symmetry of a circular conductor, is a maximum at the centre.
As the frequency increases the proportion of the conductor’s cross-section
affected also increases. Eventually this back emf restricts the movement of
charge to the periphery of the cross-section, so increasing the conductor’s
resistance. The increase in resistance results in reduced current capacity and
greater losses.
 Stray Capacitance. Stray capacitance may shunt the signal away from
loads, resulting in a reduction in signal level. As frequency increases and
reactance falls, this effect become severe.
 Lead Inductance. As seen above, even a short lead length has self
inductance which at a high enough frequency can produce large reactances.
This can lead to reduced currents or increased voltage drops resulting in
reduced signal levels.
 Stray Resonances. It is possible for resonances to occur between the stray
capacitances (or capacitors) and unwanted lead inductances. These
resonances are usually at higher frequencies than the working frequency.
They can give rise to shunting or blocking of the signal depending on whether
a series or shunt circuit is formed. Of greater likelihood is the generation of
parasitic oscillations.
 Parasitic Oscillation. The generation of parasitic oscillations can occur for a
number of reasons. At low frequencies this can be due to poor decoupling of
supply lines. Generally, however, parasitics occur much higher than the
working frequency due to the presence of stray resonances. These can
create the right conditions for oscillation. Parasitics can reduce the power
available at the working frequency or overload the system causing component
damage.

Issued November 2006 Chapter 2 : Section 9 133

 Component Size. The physical size of components for use at VHF and UHF
can pose problems of miniaturisation. In particular, inductors are difficult to
make without high losses, and this results in low Q circuits. While this is not
necessarily a drawback as it allows a wide range of channels, it is usually
accompanied by a low efficiency.

134 Chapter 2 : Section 9 Issued November 2006

REVISION QUESTIONS
1 List 6 significant problems associated with electronic circuits at VHF and UHF.

2 Skin effect can be reduced by:

3 What changes would you make to the connecting leads of a VHF transmitter
to overcome signal losses on those leads?

4 Lead inductance can have the effect of:

5 Parasitic oscillations:

Issued November 2006 Chapter 2 : Section 9 135

NOTES:

136 Chapter 2 : Section 9 Issued November 2006

SECTION 10 : TRANSMITTER TUNING

CHANNEL SELECTION
The procedure for aircrew to select the wanted frequency has been greatly simplified
in the current generation of equipments. The control unit accessible to the aircrew is
usually remote from the actual transmitter equipment. Channel selection in a
frequency synthesised equipment is by electrically signalling the channel number to
a programmable frequency divider in the transmitter. The actual selection is usually
made via a series of rotary switches on the control unit in the cockpit.
Although modern channel selection is performed by push button keypads linked to
frequency synthesising circuits other older systems still exist. These include the use
of gear trains, push button switches, simple synchronous control system or a
Wheatstone drive circuit.

Issued November 2006 Chapter 2 : Section 10 137

CW TRANSMITTER ABBREVIATIONS
AP Air Publication
Voltage gain

Beta
BFO Beat frequency oscillator
BSD Block schematic diagram

C Capacitance
Capacitance between base and emitter
Capacitance between collector and base
Capacitance between collector and emitter
Feedback capacitor
Capacitance of holder
CW Continuous wave or clockwise

dB Decibel
dc Direct current

EMC Electro-magnetic compatibility

Parallel resonance
FM Frequency modulation
Natural frequency
Series resonance
FSK Frequency shift keying

HF High frequency

I Current
IEC Inter-electrode capacitance
IF Intermediate frequency
IPA Intermediate power amplifier
kHz Kilohertz

138 Chapter 2 : Section 10 Issued November 2006

kW Kilowatts

L Inductance
Lambda

m milli
MCW Modulated continuous wave or modulated carrier wave
MHz Megahertz
MO Master Oscillator

n Nano
Efficiency

OOK On-off keying

P Pico
PA Power amplifier
PCB Printed circuit board
PCM Pulse code modulation
PLL Phase locked loop
PM Pulse modulation
PSK Phase shift keying
PSU Power supply unit

Q Selectivity

R Resistance
RF Radio frequency
RFC Radio frequency choke
RFPA Radio frequency power amplifier
R/T Radio telephony

 Micro
UHF Ultra High Frequency
V Voltage

Issued November 2006 Chapter 2 : Section 10 139

Voltage at the base

Voltage at the collector
Voltage at the emitter
Voltage between the base and emitter
Voltage between the collector and base
Supply voltage
Voltage between the collector and emitter
VCO Voltage controlled oscillator
VHF Very high frequency

Capacitive reactance
Inductive reactance or crystal

Z Impedance

140 Chapter 2 : Section 10 Issued November 2006

SECTION 11 : POWER DISTRIBUTION IN AN AM WAVEFORM

INTRODUCTION
Previously it was seen how the modulation process results in the production of
sidebands. This section looks at the way the energy is distributed between the
carrier and sidebands.

MODULATION DEPTH
The subject of modulation index/depth was dealt with in Module 4 but is now
summarised by way of revision. There are 2 basic methods of finding the depth of
modulation. In Fig. 1 it can be found from:

FIG 1

Issued November 2006 Chapter 2 : Section 11 141

whereas in Fig 2 (representing the same modulated carrier it is found from):

The limits of m are 0 for no modulation, and 1 for maximum or 100% modulation.

FIG 2

OVER MODULATION
If an attempt is made to modulate a carrier at a level greater than 1, the carrier will
be over-modulated and there will be distortion when the signal is demodulated in the
receiver. This is shown in Fig. 3. In effect, over modulation is an attempt to make h
(in the second expression) less than zero, which, of course, is impossible, the result
is the interrupted carrier shown.

FIG 3

142 Chapter 2 : Section 11 Issued November 2006

UNDER MODULATION
It is more difficult to define under modulation. In theory 100% modulation is the ideal
because it maximises the signal-to-noise ratio in the receiver. In practice, it is not
easy to achieve 100% modulation without introducing distortion or risking over
modulation on unexpected audio peaks.
A typical speech signal is so variable in amplitude that most of the time it is well
below the peak level. Even if the peaks corresponded to 100% modulation most of
the waveform would be about 20-30%, so it can be seen that under modulation is the
normal state of affairs. It is usual to use a peak value for m of about 0.8 or some
such figure, so as to minimise distortion yet still be able to cope with unexpected
peaks without overloading the modulator.
The information content of a modulated carrier is contained in the upper and lower
sidebands. If the amplitude of the modulating signal is very small compared to the
carrier then carrier amplitude variations may be difficult to detect and this would
result in less or degrading of the information transmitted. It is obvious that the power
of the modulating signal will always be chosen so that it imprints itself sufficiently on
the carrier and the real problem is to avoid overdoing it.

POWER DISTRIBUTION
The relationship between the transmitted power, carrier power, , and sideband
power, is given by:

For example, if a 25W carrier is modulated to a depth of 60%, the total transmitted
power is:

Issued November 2006 Chapter 2 : Section 11 143

The increase of 4.5 W does not seem to be a great deal of sideband power and it is
useful to look at how it varies as the depth of modulation changes. This is given by
the expression:

It is shown graphically in Fig 4. The most obvious result is that the maximum
proportion of the transmitter power in the sidebands is one-third which occurs when
m is 1. If m is reduced then the proportion reduces rapidly, reaching 10% when m is
just less than 0.5.

FIG 4

This is very inefficient when it is borne in mind that the combined upper and lower
sideband power can never be more than one-third of the total. Even worse is the
fact that both sidebands carry the same information. This is not deliberate, just a
side effect of the way the modulation is carried out. The maximum proportion of
useful power transmitted can never exceed one-sixth of the total.
The importance of sideband power is that it is possible with correct filtering and
receiver design to dispense with the carrier altogether and even one of the
sidebands. This is the basis of single sideband systems.

144 Chapter 2 : Section 11 Issued November 2006

SECTION 12 : THE STRUCTURE OF AM TRANSMITTERS

INTRODUCTION
The BSD of a CW transmitter was described previously. The BSD of an am
transmitter is very similar except for the provision for modulating the carrier. Fig. 1
shows a typical (simplified) arrangement.

FIG 1

The functions of the oscillator, buffer amplifier, driver amplifier and final power
amplifier are just as described in the chapter on CW techniques. The modulator has
a function not necessary in a CW transmitter. Its purpose is to take the signal (in this
case speech input from a microphone), amplify it and use to control the amplitude of
the carrier.
It is important to understand that all the power in the sidebands of the am wave
comes from the modulator. As seen in the previous chapter, the sideband power is
in addition to the carrier power and it is all provided by the modulator which is an
audio power amplifier. It is normally integrated into the RFPA (in the example
shown) and so perhaps should not be featured as a separate block.

TYPICAL WAVEFORMS AND LEVELS

Fig. 2 shows the waveforms to be expected at each stage of the transmitter. Note
that the frequency ratio between the AF and RF has been exaggerated for clarity. In
practice, there would be at least 44 RF cycles for each audio cycle. This assumes a
carrier frequency of 198 kHz (Radio 4) modulated by a maximum AF of 4.5 kHz. At
the higher end of the MF broadcast band (about 1.6 MHz) the same AF cycle would
last for more than 350 RF cycles.

Issued November 2006 Chapter 2 : Section 12 145

FIG 2

The power levels in a transmitter can vary widely depending on the stage at which
they are considered, and the type of transmitter. Fig. 3 shows a double sideband
amplitude modulated (DSB AM) transmitter capable of producing 100W of carrier.
Typical gains are shown against each block. It should be borne in mind that each
block may contain a number of stages – particularly the modulator. In practice the
modulator must be capable of producing more than one-third of the carrier power.
This excess power enables losses to be overcome and minimise distortion in the
modulator.

FIG 3

146 Chapter 2 : Section 12 Issued November 2006

Airborne equipments in current service produce a wide range of output powers, from
about 10 W up to about 400 W. Although their modulation systems are not the same
as that described so far, similar gains will apply and the powers will be broadly in
proportion to those shown.
The modulator stage must produce a wide range of output powers which, from
instant to instant, will depend upon the nature of the modulating signal. If this
variation is avoided by considering a sinusoidal modulating signal, there are only 2
factors which determine the power required from the modulator:
 The maximum depth of modulation which the system is designed to operate
with.
 The carrier power which has to be modulated.

These factors have already been identified as

Issued November 2006 Chapter 2 : Section 12 147

NOTES:

148 Chapter 2 : Section 12 Issued November 2006

SECTION 13 : TECHNIQUES OF AMPLITUDE MODULATION

INTRODUCTION
All amplitude modulators act in the same fundamental way, that is by controlling the
amplitude of the RF output current of the stage where modulation occurs. This can
be done by control of the base-emitter/gate-source voltage. Paragraphs 2 and 3
below explain that the modulators can also be classified according to the point in the
transmitter where the modulation is applied.

LOW LEVEL MODULATION

Low level modulation describes the technique whereby the modulation is applied at a
low power stage in the transmitter, generally before the final RFPA but after any
buffer and multiplier stags. In Fig 1 it is applied to the driver stages.

FIG 1

Issued November 2006 Chapter 2 : Section 13 149

Advantage
The modulator has to provide only a small amount of AF power because the
sideband power produced will be further amplified in the final RFPA.
Disadvantages
 All the stages following the one at which modulation takes place must be
linear, ie. be operated as single ended Class A or push-pull Class A or B.
Class C operation is to be avoided as it can lead to unwanted distortion
unless the modulation depth is low but, as seen previously, low depths of
modulation are to be avoided.
 Because Class C cannot be used, the efficiency of the higher power stages is
compromised. Whether this is a severe disadvantage or not depends upon
the power transmitted. It is not likely to be a major problem with airborne
communications transmitters with their limited output power, but for higher
power ground installations it is a most important factor.

HIGH LEVEL MODULATION

This is the system assumed in the previous chapter, where the modulation takes
place in the final RFPA. It is shown in Fig 2.

FIG 2

150 Chapter 2 : Section 13 Issued November 2006

Advantage
Because all the stages up to the final RFPA are unmodulated, they may be operated
in Class C if their function allows this. If they develop significant power (as in the
driver stages) there will be benefits of efficiency. The most significant advantage
however, is that the final RFPA may be a Class C stage, and in a high power
transmitter this is of great significance.
Disadvantages
 The modulator itself must be capable of delivering considerable AF power.
 Achieving modulation depths which approach 100% may only be possible if
modulation is applied to the previous stage as well as to the RFPA.

METHODS OF MODULATION
Base or Gate Modulation
This technique is shown in Fig 3 overleaf, where the dc, AF and RF signals are all
combined at the base of the transistor. The power needed from the AF amplifier is
low because of the small voltages and currents which the base can handle.
The waveforms are in Fig 4 which shows how the combination of signals at the base
causes the RF pulses of collector current to vary in amplitude according to the Af.
Thus, the energy fed to the output tuned circuit varies with the AF and so does the
amplitude of the output voltage. The amplitude of the AF must not exceed that of the
carrier or else the RF output will be interrupted.

Issued November 2006 Chapter 2 : Section 13 151

FIG 3

FIG 4

152 Chapter 2 : Section 13 Issued November 2006

The same technique may be used with FETs and such a circuit is shown in Fig 5. It
is shown with a fixed bias applied to the gate but it could be operated with just
source bias. The source resistor is still included for bias stabilisation.

FIG 5

EMITTER AND SOURCE MODULATION

This is essentially the same as base/gate modulation, but the signal is applied to the
low input impedance at the emitter/source. Fig 6 shows emitter modulation and the
dominant feature is the step down AF transformer feeding the emitter. This is
needed to give the relatively high emitter AF current, ie. it is impedance matching
between the emitter and the AF amplifier output impedance. Apart from this, the
operation is the same as for base/gate modulation.

COLLECTOR OR DRAIN MODULATION

Fig 7 illustrates drain modulation. The modulating FET, is in series with the RF
output transistor which becomes modulated.
The AF input to is between the gate and the fixed dc potential of the power rail,
ie. across . This is so that one side of the input is at signal earth. If had been
the other way up the input would have to be applied between the gate and the top of
the RF choke and this latter point is not at a fixed potential. A P-channel device has
now to be used to get the same direction of current in both FETs.

Issued November 2006 Chapter 2 : Section 13 153

FIG 6

FIG 7

154 Chapter 2 : Section 13 Issued November 2006

SECTION 14 : SPEECH CLIPPING

INTRODUCTION
The effectiveness of an am transmission depends upon the readability of the
received signal. The signal strength must be sufficiently high so that the quietest
parts of the message are audible above the background noise, but not so high that
the receiver circuits are overloaded. This is taken care of, in part, by making the
gain of the receiver dependent upon the received signal strength. In the transmitter
the dynamic range of the modulation can be reduced by use of the technique of
speech clipping.

DYNAMIC RANGE
The dynamic range of a signal is simply the ratio of the greatest intensity to the least
intensity. Normal speech has a wide dynamic range, perhaps 50 dB or so
(10,0000:1), depending on the person speaking. Fig 1(a) shows a signal where the
peaks are 50dB above the lowest level. In practice, this lowest level might be set by
noise generated within the microphone amplifier, rather than the quietness of the
sound.

FIG 1

Issued November 2006 Chapter 2 : Section 14 155

CLIPPING
The peaks in Fig 1(a) are generally due to what are called the explosive consonants,
which are produced when letters such as b, d, k, p and t are sounded. These letters
are said to be unvoiced, ie. they cannot tell anything about the person uttering them,
such as age, gender or accent. If the corresponding peaks were reduced in
amplitude there will be a loss of information but it will be minimal and will not affect
the intelligibility of the signal.
By allowing the peaks to run into the non-linear region of the modulator they will
automatically be limited. By amplifying the signal sufficiently, not only can this
limiting be accomplished, but the average level of the signal can be increased,
causing the average depth of modulation to rise from, say 30% to 50%. This is
shown in Fig 1(b). This would give less than 8% increase in the transmitted power
but the sideband power would be increased by a factor of 2.8 giving a proportional
increase in the output of a receiver.
Unfortunately the use of the modulator’s non-linearity would give rise to the
generation of unwanted RF harmonics in the transmitter output which would overlap
into adjacent channels. It is not a satisfactory way of limiting the signal. Instead, the
signal is fed to a dedicated clipping circuit. The maximum amplitude from this circuit
is then at a known value and this can be made to correspond to the maximum
undistorted level in the modulator. The transmitter output is then as clean as
possible, ie. the unwanted harmonics are kept to a minimum.
Fig 2 shows a simple speech clipping circuit comprising a pair of diodes acting as
positive and negative shunt limiters. Silicon diodes allow a maximum peak-to-peak
voltage of about 1.5V. Different levels of speech would give rise to different levels of
clipping. To ensure that the average level is reasonably standard, automatic gain
control (AGC) may be used prior to clipping. AGC is covered in Section 3, Chapter
12, on AM Receivers.

FIG 2

156 Chapter 2 : Section 14 Issued November 2006

The limiting action gives rise to the generation of AF harmonics which would
seriously impair the intelligibility of the signal. These are filtered out with a low pass
filter which has its bandwidth set to about 3 kHz for normal voice communication
systems. Broadcast systems use a wider bandwidth (up to 4.5 kHz on AM broadcast
transmissions) and a system which compresses the dynamic range rather than just
slicing off the peaks of the signal.
Fig 3 shows another circuit which does the same job as that of Fig 2 but gives a
somewhat higher output voltage. A dc bias supply of 30 V is potentially divided by
and to give 3.2 V at the cathode of . has a low reactance and prevents
the signal affecting this level. and divide the 3.2 V equally across the 2 diodes
so that the output voltage is 1.6 V in the absence of an input signal. This means that
neither diode is conducting.

FIG 3

Issued November 2006 Chapter 2 : Section 14 157

Any input signal which does not cause either diode to conduct will appear at the
output unchanged, apart from any level shifting if the mean input voltage is not 1.6 V.
If the input signal goes above 3.8 V (3.2 V + 0.6 V) will conduct and below –0.6 V
will conduct. Either condition will limit the output voltage to a peak-to-peak value
of 4.4 V. Fig 4 shows the effect of the circuit on 2 different amplitude waveforms.

FIG 4

158 Chapter 2 : Section 14 Issued November 2006

AM TRANSMITTERS – REVISION QUESTIONS

1 State the formula for calculating the depth of modulation.
2 Calculate the depth of modulation for the waveform shown in Fig 5.

FIG 5

3 State the formula for calculating the power distribution of an AM wave.

4 A transmitter with a carrier power of 100 watts is modulated to a depth of
40%. Calculate the total power radiated.
5 What fraction of the total power is contained in one sideband of question 4.
6 A transmitter is modulated to 100%. What useful power will be contained in
the upper sideband.
7 Label the BSD of the AM transmitter shown in Fig 6.

FIG 6

8 State the function and bias of each stage of question 7

9 a Explain the difference between high level and low level modulation.
b List the advantages/disadvantages of both types
10 Explain what is meant by the dynamic range of a signal.
11 State the reason for using speech clipping in an R/T transmitter.
12 Calculate the voltage at pointers A and B of Fig 12.

Issued November 2006 Chapter 2 : Section 14 159

FIG 7

13 Explain the circuit action of Fig 12 for input signal levels of:
a 1 volt peak
b 4 volt peak
14 Low level modulation is applied to a:

15 High level modulation has the advantage that:

16 In a low level AM system, amplifiers following the modulated stage will be:

160 Chapter 2 : Section 14 Issued November 2006

17 An amplitude modulated signal has a modulated index of 0.75 and develops

90 watts at the aerial. How much power is contained in each sideband?

18 A CW transmission is converted to MCW by employing 50% amplitude

modulation.

19 An AM signal develops 150 watts at the aerial. If the depth of modulation is

25% the amount of power in each sideband is:

20 A class C power amplifier working in an aerial system is collector modulated

with a sinusoidal audio tone. As a result the aerial ammeter shows an
increase whilst the mean collector current of the PA transistor remains
constant. This is due to:

21 The unmodulated output power of a transmitter is 100 watts. If the modulation

depth is 25%, the total power contained in the sidebands is:

22 If an AM transmitter PA stage had a very high Q tuned circuit series load:

Issued November 2006 Chapter 2 : Section 14 161

23 A transmitter radiates an unmodulated CW carrier from an aerial with a

radiation resistance of 70. The RF aerial current is 2 amp. The transmitter is
then amplitude modulated 60%. The power in:

24 Speech clipping circuits are used to:

162 Chapter 2 : Section 14 Issued November 2006

CHAPTER 3 : RECEIVERS
SECTION 1 : BASIC RECEIVER REQUIREMENTS
BASIC POINTS
The purpose of the receiver is to exactly reproduce the information used to modulate
a given transmitter carrier signal.
The basic requirements of the receiver can be grouped under the following headings:
 Selection. With many signals present at the input to a receiver, selection
of the required one is necessary to prevent distortion of the output. All other
signals must be rejected or attenuated to a level at which they produce no
interference at the receiver output.
 Amplification. Signals picked up by an aerial and fed to a receiver are of
very small amplitude, usually microvolts. To obtain a usable output, therefore,
a large amount of amplification is required.
 Demodulation. All signals received are at radio frequency. To obtain the
intelligence from the signal, an appropriate method of demodulation is
essential.
A block schematic diagram showing the reception process is shown below.

FIG 1 RECEPTION PROCESS BLOCK SCHEMATIC DIAGRAM

Issued November 2006 Chapter 3 : Section 1 163

RECEIVER PARAMETERS
To describe how effective a receiver is in meeting the above requirements, the
following parameters are used:
 Selectivity. Selectivity is the ability of the receiver to select the desired
signal frequency and reject, or attenuate to a low level, other, unwanted,
signal frequencies. The effective Q factor of the tuned circuits will determine
the sharpness of frequency selection.
 Sensitivity. Sensitivity is the level of RF input signal required to give a
standard power output level from the AF amplifier. Sensitivity is usually
expressed in terms of microvolts input (RF) to watts of audio power output.
 Signal to Noise Ratio. The maximum sensitivity of any receiver is limited by
the ratio of the desired signal’s strength to the strength of all interfering noise
rather than by the actual gain of the receiver. Signal-to-noise ratio may be
defined simply as:

Although it is more commonly quoted in dB. Since appreciable noise can

be generated within the receiver, another common parameter quoted is a
receiver’s noise factor which will be dealt with later.
 Fidelity. Fidelity may be defined as the ability of a receiver to reproduce an
output which is an exact copy of the modulation of the input signal. The
bandwidth of the receiver must be sufficiently wide to allow all of both
sidebands of the transmitted signal to pass through. Too narrow a bandwidth
will result in a loss of the higher modulating frequencies on AM or clipping of
the modulation signal on FM. In either case, the transmitted signal will not
have faithfully reproduced in the receiver.
Note that fidelity requires a wide enough bandwidth to allow all the transmitted
signals through, whereas selectivity requires a narrow bandwidth to reject adjacent
signals.

164 Chapter 3 : Section 1 Issued November 2006

REVISION QUESTIONS
1 The 3 main requirements of a receiver are:
a……………………………………………… Choose from: modulation,
b……………………………………………… demodulation, sensitivity,
selectivity, amplification,
c……………………………………………….
attenuation

2 Selectivity of a receiver may be defined as the ability of a receiver

to……………………….…..the desired signal and……………………………
or…………………………………..……...all others.

Choose from: signal signal demodulate

amplify reject accept
modulate attenuate

3 For an input signal of 2W, a receiver under test should give an output of 5W.
If the output is only 1W from the above input signal, the receiver has poor:

4 A receiver in use is found to have a good quality output at the correct level,
but with a high, constant noise level. The receiver has:

Issued November 2006 Chapter 3 : Section 1 165

NOTES:

166 Chapter 3 : Section 1 Issued November 2006

SECTION 2 : TUNED RADIO FREQUENCY (TRF) RECEIVER

TRF RECEIVER
The TRF receiver is a basic receiver operation on one frequency only. A typical
structure is shown in Fig 1.

FIG 1 BASIC RECEIVER BLOCK SCHEMATIC DIAGRAM

As only one frequency is required selectivity is provided by the RF amplifiers and

these also provide high gain. After demodulation, the audio frequency signal is
amplified in the AF stages to produce an audible output.
The TRF receiver has proved useful in the past for several purposes ranging from
cheap single channel domestic radios to highly expensive receivers in avionic
equipment such as the instrument landing system (ILS). However it suffers severe
limitations caused by a combination of 2 or more of the following:
 Low and uneven sensitivity.
 Poor selectivity.
 Instability (a tendency to oscillate).
Using several RF amplifiers to obtain high gain with good selectivity gives rise to
some difficulty in ensuring that each amplifier is tuned to exactly the same frequency.
Components are not identical in all respects so frequency alignment can be a
problem. Even when correctly aligned small amounts of positive feedback via
couplings, stray capacitance or radiation creates oscillation and hence instability.
In TRF receivers made tuneable over a frequency range (usually 500-1600 kHz),
couplings to make all the RF amplifiers change in step, also cause instability and
alignment problems are worse.
It would be noted that the TRF receiver started to become obsolescent in 1927 but
has persisted in specialised applications to the present day.

Issued November 2006 Chapter 3 : Section 2 167

REVISION QUESTIONS
By reference to Fig 2, complete table 1 by inserting the name and function of each
stage.

FIG 1

Stage Name Function

TABLE 1

2 A major disadvantage of TRF receivers is that, if used for wide frequency

coverage, they:

3 A major feature of TRF receivers designed to have a wide frequency

coverage and good sensitivity is that:

168 Chapter 3 : Section 2 Issued November 2006

SECTION 3 : THE SUPERHETERODYNE RECEIVER

THE SUPERHET RECEIVER
The limitations of the TRF receiver can be overcome by carrying out the amplification
at a fixed frequency. By converting all input signals to this fixed frequency, as much
amplification as is required can be obtained without instability, but with good
sensitivity and selectivity. This is the principle of the superheterodyne (superhet)
receiver, for which a simple BSD is shown in Fig 1 below.

FIG 1 SUPERHET RECEIVER BSD

FUNCTIONAL STAGES
The functions of the stages of the superhet receiver in Fig 1 are:
 RF Amplifier. The RF amplifier, usually only one stage, improves both
signal-to-noise ratio and selectivity. Not all receivers use an RF amplifier, but
its inclusion does provide certain advantages.
 Local Oscillator. The local oscillator (LO) produces a frequency which is
different to the incoming signal by a constant amount. The difference
frequency is the intermediate frequency (IF).
 Mixer. The mixer (frequency changer) combines the incoming signal and LO
frequencies to produce the IF.
 IF Amplifier. The IF amplifier, usually 2 or more stages, amplifies the
difference frequency from the mixer. It is the IF amplifier(s) which provide
most of the gain and selectivity of a receiver.
 Demodulator (Detector). The demodulator extracts the intelligence from the
RF signal. It may also provide a dc voltage for gain control purposes.
 AF Amplifier. The AF amplifier raises the signal to the required level.

Issued November 2006 Chapter 3 : Section 3 169

Typical symbols used in BSDs, together with typical waveforms.

FIG 2 RECEIVER BSD

Not shown in the BSD, but equally as important as the other circuits, is the aerial
coupling circuit. This is used to couple the aerial to the receiver so that optimum
bandwidth and/or matching is achieved. Several types of circuit may be used and
examples of these are included in an annex to this section.
Although the superhet receiver overcomes the limitations of the TRF, it still has
problems associated with its operation and alignment.
Whatever method of tuning to a particular transmission is adopted the LO is tuned to
a higher frequency than the signal. This allows reception down to zero hertz making
full use of the overcrowded frequency spectrum. It will be seen later that in radar
receivers the LO is tuned below the signal. This is because oscillators are more
stable at lower frequencies and there is less problem with the spectrum.

170 Chapter 3 : Section 3 Issued November 2006

REVISION QUESTIONS
1 A receiver in which selected signals of any frequency are converted to a
single common frequency is called a:

By reference to Fig 3, complete table 1 by inserting the name and function of each
stage and sketching the waveforms expected at the output of each stage for an AM
signal input.

FIG 3

Stage Name Function Output Waveform

TABLE 1

3 Identify the frequency bands at which the LO of a receiver is tuned above or

below the selected signal. Insert your answer in table 2 below.

Frequency Band LO Above/Below

VHF

UHF

TABLE 2

Issued November 2006 Chapter 3 : Section 3 171

4 Tuning the LO above the signal frequency in some receivers is essential to:

5 An advantage of tuning the LO below the signal frequency at VHF and above
is that:

6 A receiver is tuned to receive a signal on 1.7 MHz when its LO is tuned to

2165 kHz.

7 What RF signal would (ideally) be received by a receiver having an IF of 500

kHz if its LO is set to 1955 kHz.

172 Chapter 3 : Section 3 Issued November 2006

SECTION 4 : AERIAL COUPLING

TYPES OF COUPLING
Because signals arriving at the aerial are of low amplitude, it is important that they
should be transferred to the RF amplifier (or mixer) with minimum loss. The prime
requirement is for maximum power transfer combined with suitable
sensitivity/selectivity characteristics. Several types of aerial coupling circuit are
used, each having its own characteristics. The design used will depend on:
 Frequency.
 Whether the aerial is resistive or reactive.
 Whether RF amplifier is fixed, tuned or variable tuned.
There are several basic methods of aerial coupling:
 Transformer.
 Auto transformer.
 Capacitive.
 Stagger tuned (to give wide bandwidth).
In general, transformer coupling will give best selectivity, whereas auto transformer
or capacitive coupling will give best sensitivity.

FIG 1 AERIAL COUPLINGS

Issued November 2006 Chapter 3 : Section 4 173

NOTES:

174 Chapter 3 : Section 4 Issued November 2006

SECTION 5 : INTERFERING SIGNALS

ADJACENT CHANNEL
In radio communication any form of unwanted radio signal is termed interference. In
the TRF receiver, the main source of interference is from a signal frequency which is
close to the required signal frequency – the ADJACENT CHANNEL. The superhet
receiver, however, because of the present of the LO frequency, is suspectable to
interference from a wider range of frequencies in addition to the adjacent channel. It
should be noted here that even adjacent channel interference in a superhet receiver
is less than in a wide frequency range TRF because of the IF amplifiers.

Issued November 2006 Chapter 3 : Section 5 175

SECOND AND IMAGE SIGNALS

 Second or Image Channels. With a receiver tuned to, say 1000 kHz and its
LO tuned to 1500 kHz, and IF of 500 kHz is produced. A signal of 2000 kHz
will also produce an IF of 500 kHz if allowed to ‘mix’ with the LO frequency.
This signal, twice the IF away from the required signal, is called the second
or image channel. As can be seen from Fig 1, the image frequency will only
produce an IF if it is allowed to reach the mixer. Thus, an RF amplifier will
reduce image channel interference to acceptable levels. However, if the IF is
very low, the image and signal frequencies are close and rejection is less
effective. Conversely, a high IF helps to provide good image channel
rejection.

FIG 1 SECOND CHANNEL INTERFERENCE

(A) LOW IF – HIGH OUTPUT AT IMAGE (B) HIGH IF – LOW OUTPUT AT IMAGE
FREQUENCY FREQUENCY
FIG 2

176 Chapter 3 : Section 5 Issued November 2006

An often quoted parameter for a receiver which indicates its rejection of image
channel interference is image channel ratio which is defined as:

 Signals at IF. Since the mixer stage responds to IF signals fed into it, any
signal at the same frequency as the IF may produce interference in the
output. However, as can be seen from Figs 2(a) and 2(b), provided the IF is
widely different from the RF signal frequency and an RF amplifier is used, this
form of interference is reduced to negligible levels.
 LO Second Harmonic Causes. If the LO has an output with a large second
harmonic content, this can mix with some other signal and again produce an
interfering signal at the IF. This form of interference is again reduced
considerably if an RF amplifier is used.
The foregoing are examples of the prime interfering signals present in a superhet.
However, there are many combinations of signals which can mix with each other and
produce either a signal to which the receiver is tuned or the IF. Fortunately these
forms of interference rarely produce major problems in practice.

NOISE INTERFERENCE
In addition to the interference caused by radio signals, the received signal can be
distorted by one of several forms of electrical noise. This electrical noise can be
separated into noise entering the receiver from an outside source – external noise –
and the noise generated within the receiver – internal noise.

EXTERNAL NOISE
Noise entering a receiver from external sources may be sub-divided into 3 main
types:
 Man Made. Man made noise is the interference produced by arcing switches,
motors, etc. and is most pronounced below about 12 MHz.
 Atmospheric. Electric discharges during storms (especially in tropical zones)
produce an appreciable amount of atmospheric noise at frequencies below
VHF, particularly at LF.
 Cosmic. Many objects in space radiate over a wide range of frequencies –
many of which reach the receiver and produce interference. Most noticeable
above about 12 MHz. This interference is called cosmic noise.
The approximate relationship between the effect of some types of external noise and
frequency is shown overleaf.

Issued November 2006 Chapter 3 : Section 5 177

FIG 3 EXTERNAL RECEIVER NOISE SOURCES/LEVELS

INTERNAL NOISE
The internal noise generated in a receiver comes from several dependent sources.
The principle noise types are:
 Thermal Noise. Also called Johnson noise. Thermal noise is due to the
random movement of electrons in a conductor. The movement varies with the
temperature of the conductor, hence its name. The equivalent noise voltage
of a device can be approximately calculated by the formula:

 Semiconductors. Semiconductors, because of their inter-electrode

resistances, generate thermal noise. Additionally, random injection of charge
carriers across depletion layers generates shot noise which can be reduced
by operating at low current values. A third type, called flicker noise, arises
primarily from surface leakage. Flicker noise is inversely proportional to
frequency and is therefore most noticeable at low frequencies.
Because signals arriving at a receiver are usually of very small amplitude, the
amount of noise present with the signal (either internally or externally generated)

178 Chapter 3 : Section 5 Issued November 2006

must be kept to the lowest possible level. This is especially important in the first
stage. Although noise cannot be completely eliminated, it can be reduced, mainly by
the designer but also by correct servicing of a receiver. Additionally, since the noise
spectrum is infinitely wide, a reduction in receiver bandwidth by the operator will also
help to reduce noise effects. However, it is important that bandwidth reduction does
not produce sideband clipping of required signals.
Since noise levels in a receiver play such an important role in the receiver’s ability to
provide useful outputs from weak signals, it is vital that they are kept to a minimum.
This can be achieved by accurate measurement of the signal-to-noise
characteristics, and more important, the receiver noise factor to quickly determine
any reduction in these qualities.

FIG 4 RECEIVER SIGNAL-TO-NOISE RATIO

 Signal-to-Noise Ratio. The signal/noise ratio at the output of a receiver is

the ratio of signal power to noise power. In an ideal receiver the input
signal/noise ratio would equal the output signal/noise ratio since no internal
noise would be added by the receiver. In practice, internal noise is added in
the receiver at all stages, so the input signal/noise ratio will be lower than the
input signal/noise ratio. If the input signal is 10W and the input noise 1 W,
the input signal to noise ratio is 10:1. this may also be written as:

Issued November 2006 Chapter 3 : Section 5 179

 Noise Factor. The noise factor of a receiver gives an indication of the noise
introduced by the receiver itself.

since, in an ideal receiver, output S/N ratio = input S/N ratio

For an ideal receiver, the noise factor is unity. For all practical receivers, it must be
greater than unity. The lower the noise factor the better the receiver, since the
output signal-to-noise ratio is nearer to the input signal-to-noise ratio.

with an input S/N ratio = 10:1 and

output S/N ratio = 4:1

this can also be expressed in dB.

Noise factor (dB) = input S/N ratio (dB) – output S/N ratio (dB) using the above
figures.

10:1 = 10 dB 4:1 = 6dB

Noise factor = 10 dB – 6 dB (=10 log 2.5)

180 Chapter 3 : Section 5 Issued November 2006

Worked Example:
A receiver has a noise factor of 5. If the input signal-to-noise ratio (SNR) is 20 dB
and the noise power output 100 W, what is the signal power output?
NF (dB) = input SNR (dB) – output
SNR (dB)
but NF = 5
NF (dB) = 10 log 5
= 6.99 (say 7)
therefore 7 = 20 – output SNR (dB)
therefore output SNR (dB) = 20 – 7 dB
= 13 dB

but output SNR (dB) =

therefore 13 =

therefore 1.3 =

therefore antilog 1.3 =

therefore 20 =

therefore output signal power =

The measurement of signal-to-noise ratio and noise factor will be dealt with later
together with other receiver measurements.

Issued November 2006 Chapter 3 : Section 5 181

RF AMPLIFIER
In reception where the signal received is very weak, eg. satellite signals, radar
echoes and long range communications, it is sometimes necessary to amplify the
signal before heterodyning. RF amplifiers in such cases must give high gain but to
prevent the introduction of even more noise than is already present in the signal the
amplifier must be a low noise device. It is the first stage which determines the
overall noise factor for the receiver and the RF amplifier if suitably chosen can assist
in achieving a low value for noise factor.
At the same time the inclusion of an RF amplifier has other advantages:
 Reduction of image channel interference – the Q of the tuned circuits (usually
 15) can reduce the effect of image channels.
 Prevention of local oscillator radiation, the oscillations inserted to the mixer
can, in the absence of an RF amplifier and due to feedback, become radiated
by the aerial circuit causing interference. This problem and the following are
not important in modern solid state receiver.
 Reduction of LO second harmonic interference, feedback of the second
harmonic may also occur if isolation between the LO and aerial circuit is
insufficient.
Where higher transmission powers or frequencies above HF are used the RF
amplifier is not essential unless required as a low noise front end. In the absence of
the RF amplifier selectivity is provided by a tuned circuit together with filters or wave
traps in the aerial circuit.

182 Chapter 3 : Section 5 Issued November 2006

REVISION QUESTIONS
1 During tests on a receiver having an IF of 500 kHz, the following
measurements and corresponding inputs for standard output were recorded.
Determine the image channel ratio of the receiver required.

Signal Generator Receiver Input Signal

Frequency Frequency Level
900 KHZ 900 KHZ 2 V + 0 dB
1400 KHZ 900 KHZ 2 V + 25 DB
1900 KHZ 900 KHZ 2 V + 60 DB

ICR =……………………………………….dB
2 A receiver having an IF of 465 kHz is tuned to 1300 kHz. Which of the
following signals could interact with LO second harmonic to produce a receiver
output?

3 What is the equivalent noise voltage of an amplifier having a bandwidth of

200 kHz an equivalent noise resistance of 1 k operating at a temperature of
290C? Take Boltzman’s constant to the

4 If the bandwidth of the amplifier in question 3 were reduced to 50 kHz, the

effective noise voltage would be:

5 For a given overall receiver gain the main limitation to the sensitivity of the
receiver is:

6 Noise factor can be defined as:

Issued November 2006 Chapter 3 : Section 5 183

NF (dB) = ……………………………….(dB) - …………………..…………….(dB)

7 A receiver has an input SNR or 8:1. If the output signal power is 2 W and the
output noise power 0.5 W, what is the receiver noise factor?

8 If the signal power input of the receiver in question 7 is 2 W, what is the
noise power input?

9 One advantage of having RF stage in a superhet receiver (as against no RF

stage) is that:

10 A receiver’s noise factor is kept small by:

11 One of the advantages of using an RF amplifier in a superhet receiver (as

against no RF stage) is that:

12 The use of an RF amplifier in a superhet receiver reduces LO radiation by:

184 Chapter 3 : Section 5 Issued November 2006

13 The use of an RF amplifier in a superhet receiver reduces interference on the

intermediate frequency by:

14 Why should the RF amplifier in a superhet receiver by a low noise stage?

Issued November 2006 Chapter 3 : Section 5 185

NOTES:

186 Chapter 3 : Section 5 Issued November 2006

SECTION 6 : TYPICAL RF AMPLIFIERS

RF AMPLIFIERS
The operation of RF amplifiers has been considered in the chapter on CW
techniques. The following circuits are included to show typical receiver RF
amplifiers.

FIG 1 BASIC SINGLE STAGE RF AMPLIFIER

Modern communication receivers have noise factors of about 10 (10 dB) and better
values (<10) are possible but are not really justified because of the following 2
reasons:
 The pick up of atmospheric and man-made noise (external noise) is such that
a noise factor of less than 30 (15 dB) is not really justified.
 Secondly, in the RF amplifier (which has most effect on the receiver’s overall
noise factor, being the first stage), it has been found that the best noise factor
is obtained at some optimum operating current for the FETs but the best
linearity and hence the best intermodulation results are obtained at high
operating currents.
 Clearly then, a choice must be made between the best noise factor and
intermodulation results. A satisfactory level of intermodulation occurs at
operating currents which give an overall noise factor of about 10 but not at
lower currents which result in lower noise factors. So it turns out that
intermodulation performance is the crucial factor and the RF stage is
designed to optimise this whilst keeping the overall noise factor to about 10
(10 dB).

Issued November 2006 Chapter 3 : Section 6 187

 RF amplifier vary in circuit design and are usually adapted to the particular
requirements of the receiver. The variations are chiefly found in:
 input circuit coupling
 output circuit coupling
 number of stages
Figs 2 and 3 show examples of typical input and output couplings.

FIG 2 TYPICAL TRANSISTOR RF AMPLIFIER AS IN AD 260 ILS RECEIVER

FIG 3 TYPICAL TRANSISTOR RF AMPLIFIER

188 Chapter 3 : Section 6 Issued November 2006

FET AMPLIFIER
An RF amplifier may be realised using an FET. The FET has the advantage that
cross modulation distortion (caused by a strong undesired signal within the pass
band of the input tuned circuit of the RF amplifier) is completely absent because of
its near square law characteristic.
An FET amplifier may be single or dual gate. Dual gate FETs are preferred
because:
 they have low capacitance
 the second gate can be used for gain control

FIG 4 RF AMPLIFIER USING AN FET

Issued November 2006 Chapter 3 : Section 6 189

CASCODE RF AMPLIFIER
The term ‘cascode’ was originally used to describe a 2-stage (cascaded) valve
amplifier in which triode valves were used because of their relatively low noise
property. The 2 stages gave a high gain equal to the much noisier pentode valve.
The first stage was a low noise low gain stage followed by a second low noise high
gain stage. The second stage was a grounded grid amplifier.
Similar advantages can be obtained with solid state devices where a low noise FET
stage is cascaded with a high gain transistor. The grounded base transistor gives
greater stability to the 2-stage amplifier.

FIG 5 SOLID STATE CASCODE AMPLIFIER

190 Chapter 3 : Section 6 Issued November 2006

SECTION 7 : FREQUENCY CHANGING (MIXING)

INTRODUCTION
Frequency changing (mixing) like the modulation process in the AM transmitter, is
carried out by mixing one signal (the received signal) with another (the local
oscillator) and from the resulting complex wave selecting the required frequency(ies).
There are 2 mixing processes available, each having its own characteristics. They
are:

ADDITIVE MIXING
In additive mixing, the RF and LO signals are ‘summed’ at the input to the mixer
device which is then biased to class B. Provided that the mixer device has a square
law characteristic ( ….), the resulting distortion of the signal
will produce frequencies of:
 RF
 LO
 RF + LO
 LO – RF (difference frequency)
In the mixer anode/collector current. The difference frequency (LO – RF) is known
as the intermediate frequency which is then selected and amplified further before
demodulation.

FIG 1 BASIC ADDITIVE MIXERS

Issued November 2006 Chapter 3 : Section 7 191

MULTIPLICATIVE MIXING
In multiplicative mixing the signals are fed to an active device in such a way that the
output waveform at any instant is proportional to the product of the 2 input signal
magnitudes. This cannot be achieved with a transistor as it has insufficient input
electrodes. Although easily achieved with multi-grid valves, the only solid state
device which gives a multiplicative output is the dual gate FET. The class of bias
used for multiplicative mixing is class A.

FET MIXERS
Fig 2 shows an FET injection mixer in which LO and signal are added via an input
transformer before being applied to the FET gate. The output tuned circuit is
resonant at the IF.

FIG 2 FET MIXER

192 Chapter 3 : Section 7 Issued November 2006

DUAL GATE FET MIXER

Because both gates of a FET control the drain current, the 2 signals to be mixed can
each be fed to separate gates to achieve multiplicative mixing. The very low
capacitance between gate 1 and drain make the dual gate FET ideal for high
frequency work. The device is biased in Class A to avoid distortion of the modulated
signal.

FIG 3 DUAL GATE FET MIXER

CHOICE OF MIXER TYPE

The decision to use an additive or a multiplicative mixer depends on:
 Noise.
 Frequency.
 Gain.
 Type of active device.
 Value of the IF.
Noise becomes important above 50 MHz so mixers at VHF and above are usually
additive particularly in modern equipments.
At high frequencies amplifier gain is low and although the conversion gain (ie. IF
output for a given signal input) of an additive mixer is lower than that of a
multiplicative one, the former is preferred because of its better noise performance.
Transistors cannot give a multiplicative output so additive mixers are normal in solid
state circuits.
With a low intermediate frequency the signal and LO values are close together. In
additive mixers this can give rise to interaction between the signal and LO inputs.
This interaction is termed ‘pulling’ and varies with the relative strengths of the input.
The effect is reduced by using minimum coupling between the inputs so feeding at
base and emitter is preferred.

Issued November 2006 Chapter 3 : Section 7 193

DIODE MIXER
A diode has a non linear characteristic and can be used as a mixer to produce an IF
from RF and LO inputs.

FIG 4 BASIC DIODE MIXER

The diode is operated on the bend in the characteristic as an additive mixer.

Disadvantages
 There is no amplification in the circuit and it has low conversion conductance.
 Coupling between LO and input RF signal.
 Low impedance presented to the LO will absorb power from the LO and could
lead to frequency instability.

194 Chapter 3 : Section 7 Issued November 2006

Advantages
Advantages are:
 Low noise.
 Low inter-electrode capacitance.
 Small size.
 No power supplies needed.
Above 150 MHz, crystal diodes must be used.

FIG 5 DIODE BALANCED MIXER

The balanced modulator used to produce SSB can also be used as a mixer in a
receiver. The IF being the difference in frequency, corresponds to the lower
sideband produced by the modulator.
A 2-diode mixer is shown in Fig 5. The LO is fed in-phase to the 2 diodes, which
feed equal but opposite currents through . Thus, there is no LO frequency in the
output (of suppression of carrier in balanced modulator).
The RF input is fed in anti-phase to the 2 diodes so that one of them is always
biased off. The RF input does appear at the output.
With both RF and LO applied, the RF creates an imbalance in the biasing of and
, so that one of them conducts more of the LO to than the other.

Issued November 2006 Chapter 3 : Section 7 195

FIG 6 ACTION OF BALANCED DIODE MIXER

The output as in the balanced modulator, consists of sum and difference frequencies
only. is tuned to the difference frequency and gives the IF output.
The mixer has the advantage that, not only is the LO frequency suppressed, but also
any noise produced by the LO> Also, it can be used at frequencies up to the
microwave regions.
The only disadvantage is that it provides no gain.

196 Chapter 3 : Section 7 Issued November 2006

LOCAL OSCILLATOR
Just as important for good reception of signals as the other stages, the local
oscillator(s) can be one of many types. The main factor which determines the local
oscillator design is stability, of both frequency and amplitude.
Although only of very low power output, the LO must still be isolated from changes in
load, vibration and power supply variations. The degree to which the design is taken
depends very much on the intended use of the receiver. A high quality
communications receiver will have a correspondingly higher quality oscillator than a
portable MW receiver, consequently many differing designs may be met, often
crystal controlled and sometimes synthesised.
At UHF, receiver mixers are achieved in micro strip using the hybrid ring (rat-race)
balanced mixer. Fig 7 shows the construction with associated diodes. Signal and
LO inputs allow the difference frequency to be taken at the output.

FIG 7

Issued November 2006 Chapter 3 : Section 7 197

REVISION QUESTIONS
1 A mixer stage has internally generated noise to the value of 4 W and a gain
of 2. If the input signal-to-noise ratio is 100:1 with a signal level of 100 W,
find the signal-to-noise ratio at the output of the mixer.

2 If the mixer stage referred to in question 1 above is preceded by an RF stage

with a gain of 10 and negligible internal noise, find the signal-to-noise ratio at
the mixer output if the input signal at the aerial is unchanged.

3 Name one of the characteristics of the additive type mixer.

4 Why does interaction (or pulling) between the local oscillator and RF carrier
frequencies occur in additive mixer circuits?

5 The transistor is suitable for what type of mixing?

6 The approximate upper frequency limit at which multiplicative mixing can

usually be used is:

7 One reason why multiplicative mixing can be used at frequencies below VHF
is because:

198 Chapter 3 : Section 7 Issued November 2006

SECTION 8 : INTERMEDIATE FREQUENCY AMPLIFIERS

INTRODUCTION
The IF amplifier stages carry out the main amplification and selection functions for
the receiver, and the design of such stages is determined primarily by the need for
good stability and selectivity. In the superhet receiver the constant value of IF for all
signal frequencies allows these requirements to be easily met. Unfortunately
conflicting requirements exist in choosing the IF value.
High
A high IF gives:
 good second channel rejection.
 good bandwidth for reasonable Q-tuned circuits.
 good harmonic rejection.
Low
A low IF gives:
 easier tracking (LO closer to SF).
 good adjacent channel rejection.
 low LO frequency, hence stable operation.
 narrow bandwidth (good selectivity).

MAIN CHARACTERISTICS
The major characteristics for high/low IF’s are shown in Table 1

Factor High IF Low IF

2nd (image) channel rejection Good POOR

Adjacent channel rejection POOR GOOD

Gain POOR BETTER

LO stability NOT CRITICAL VERY CRITICAL

Sideband clipping NONE PROBABLE

Tracking DIFFICULT EASY

TABLE 1 MAJOR CHARACTERISTICS OF HIGH/LOW IFS

Issued November 2006 Chapter 3 : Section 8 199

For many purpose a compromise solution can be reached and an ’intermediate’ IF

used which will give quite good all round results. Typical compromise IFs for
communications are:
 AM broadcast 465 kHz
 FM broadcast 10.7 MHz
However, for high quality reception, the solution is to use 2 IFs, the first one using a
high frequency and the second one a low frequency. Apart from some tracking
difficulties with the first local oscillator, the overall result has all the advantages of
both sets of requirements. A receiver using this method is called a double
superhet, the block diagram for which is shown in Fig 1. This structure overcomes
the practical difficulties of heterodyning signals above 30 MHz down to 465 kHz
when the LO frequency is too close to the signal.

FIG 1 DOUBLE SUPERHET RECEIVER

The use of IF amplifiers in cascade, as shown in Fig 2, may however, cause

problems when a high gain/wide bandwidth amplifier is required. It can be shown
that, if a number of identical tuned amplifiers ‘n’ each of bandwidth ‘B’ are connected
in cascade, the overall bandwidth is reduced.
Thus, to obtain a high gain with wide bandwidth, each amplifier may need to be very
flatly tuned. This particular problem may be overcome by one of several methods.
We will also see how bandwidths can be made narrower for the reception of narrow
band signals combined with minimum noise.
An example of bandwidth narrowing by use of identical amplifiers is shown below.

FIG 2 BANDWIDTH OF CASCADED AMPLIFIERS

200 Chapter 3 : Section 8 Issued November 2006

VARIABLE SELECTIVITY
To ensure the bandwidth is large enough to allow all the sideband frequencies
present in the signal, it may be necessary to be able to increase the bandwidth of the
IF amplifier.
For certain facilities, it may be necessary to reduce bandwidth:
 Reception of CW
 Reception where a large amount of noise is present
 Adjacent channel rejection
Typical circuits which show differing methods of obtaining the required
characteristics are shown in the following diagrams. Note also that since distortion
must be minimised and each stage of relatively high gain, all the circuits (as with all
IF amplifiers) are biased to Class A.

FIG 3 SELECTIVITY CONTROL BY VARIABLE DAMPING

Issued November 2006 Chapter 3 : Section 8 201

SELECTIVITY CONTROL BY CRYSTAL FILTER

The split electrode Xtal eliminates the need for neutralising, which is essential when
using a normal Xtal. From the simplified diagram it can be seen that the system is
balanced. and now effectively add to and .

FIG 4 SPLIT CRYSTAL FILTER EQUIVALENT CIRCUIT

202 Chapter 3 : Section 8 Issued November 2006

SELECTIVITY CONTROL BY SWITCHED VARIABLE COUPLING

With as shown, coupling is from to . In the alternative position, is
introduced, which increases the mutual inductive coupling and so broadens the
bandwidth.

FIG 5 VARIABLE COUPLING OF IF AMPLIFIER

TYPICAL SELECTIVITY VALUES

Some typical approximate IF bandwidths used in receiver systems are listed in
table 2 below.

Mode Bandwidth
CW 1 KHZ
MCW 2 KHZ

Broadcast-AM 9 KHZ

SSB-R/T 2.7 KHZ

DSB-AM 6 KHZ
BROADCAST-FM 180 KHZ

TABLE 2

Issued November 2006 Chapter 3 : Section 8 203

NOTES:

204 Chapter 3 : Section 8 Issued November 2006

SECTION 9 : DEMODULATION
INTRODUCTION
The process of demodulation is, as we have noted earlier, similar to the modulation
and mixing processes. However, since we already have 3 frequencies present
(carrier + USB + LSB), there is no need to use a separate oscillator. The signal can
just be fed into a non-linear device and the appropriate signal filtered off.
The most common method of demodulating the AM signal is the so called ‘linear’
method. This involves feeding the signal, appropriately amplified, to a simple
rectifier and then filtering – an almost identical process to power supply rectification
and filtering.
Consider the ‘ideal’ rectifier characteristic below with the signal input/outputs as
shown.

FIG 1 LINEAR DEMODULATION

Examination of the output wave shows that the mean value of the output current
follows the same pattern as the positive half of the input AM signal. This, of course,
is the same variation as the original modulating signal, thus the rectifier is effectively
demodulating the AM signal. It is still, however, in the form of IF (rectified), together
with the AF component and, as with all rectifiers, a dc component. Furthermore
since the frequencies of the output are widely different (IF, AF, DC), and because the
detector is used to drive a high input impedance device, filtering can be done most
easily and cheaply using C-R circuits.

Issued November 2006 Chapter 3 : Section 9 205

FIG 2 DIODE DEMODULATOR – BLOCK SCHEMATIC DIAGRAM

FIG 3 SERIES DIODE DEMODULATOR – PRACTICAL CIRCUIT DIAGRAM

A PRACTICAL DEMODULATOR – SERIES TYPE

Although and are called load and reservoir capacitor respectively, their
function is to increase the efficiency of the circuit. Instead of just having the mean of
the input signal to produce an output is added to charge up to the peaks of the
input signal. Resistor enables the charge to leak away at a controlled rate to
ensure that can follow all variations in the input signal due to the modulation
present. It is important that is not too small, otherwise will discharge too
rapidly and this would nullify the purpose of . Conversely, being too large
would prevent following the variations due to modulation. The effects of
differing CR times on the demodulation process are shown below. Note that the
frequency relationship used in the diagrams is unrealistic, but is used for
diagrammatic purposes.

206 Chapter 3 : Section 9 Issued November 2006

FIG 4 EFFECT OF CR TIME CONSTANT

The choice of load resistor and reservoir capacitor is thus fairly critical for best
efficiency of the demodulator, especially if one considers the driving stage. If, for
example, is too large, the damping effect on the final IF amplifier can be
excessive. Values found in detectors can be widely different, much depends on the
IF, driven and driving stages, whether FETs or transistors are used and also the
greatest depth of modulation expected to be received. For this reason, component
values are not given. The results of incorrect time constant, however, should be
remembered. The time constant is chosen to be usually at least  10 the
value of the intermediate frequency periodic time and is typically  25.
In addition to the distortion shown above, there will be harmonic distortion due to the
non-linearity of the diode characteristic, especially if the signal level is small and the
modulation depth high. Thus the diode detector is most effective when used with
large signals having a small depth of modulation.

Issued November 2006 Chapter 3 : Section 9 207

SHUNT DIODE DETECTOR

An alternative to the series diode detector which is used extensively, is the shunt
diode detector. It is a type of clamping detector.

FIG 5 SHUNT DIODE DETECTOR

This arrangement has a lower input impedance than the series detector, but can be
used where the detector is required to connect direct to the collector of the previous
stage, as shown in Fig 6 below.

FIG 6 APPLICATION OF SHUNT DIODE DETECTOR

There is no RF isolation between input and output as in the series-diode detector,

consequently the output at point A on the above diagrams contains a much greater
RF ripple than that of the series-diode circuit.

208 Chapter 3 : Section 9 Issued November 2006

During the small part of each RF signal input cycle, the diode conducts, charging the
reservoir and at the same time acts as a short circuit across the output terminals.
Thus, for this period, the detector output is zero. This occurs at the positive peaks of
the input signal. The detector output is the RF input waveform displaced vertically so
that each positive peak touches the zero volts line as shown in Fig 7 below.

FIG 7 SHUNT DIODE DETECTOR WAVEFORMS

Note that the mean or average value of the output signal varies with the modulation
signal and, if the RF ripple is filtered out, consists of the modulation waveform
superimposed on a negative (in this case) component. This form of detector is an
example of negative clamping in which positive peaks of the input signal are
clamped at zero volts.

COMPARISON: SERIES V SHUNT DETECTOR

Series
 In basic circuit good RF filter action.
 Relatively high input impedance.
 Relatively high output impedance.
 Audio output level slightly less than on modulated RF input waveform.
 Reservoir and load values quite critical if distortion is to be avoided.
Shunt
 Virtually no RF filter action (requires separate circuit).
 Relatively low input impedance.
 Relatively low output impedance.
 Audio output level greater than on modulated RF input waveform (up to , but
depends upon CR values).
 Reservoir and load values not critical.

Issued November 2006 Chapter 3 : Section 9 209

In practice, the components in the circuit of Fig 2, may have different positions, it is
common, for example, to have the series diode positioned in the bottom leg, and
in parallel and, in some cases, fewer components. Typical variations are:

FIG 8 ALTERNATIVE DIODE DETECTOR CIRCUITS

 Major limitations of diode detectors are that although they are simple and
provide a good quality output, they have no amplification and tend to damp
the final If transformer. When these limitations are unacceptable, transistor
FET detectors may be used.

PRODUCT DETECTOR
Another type of detector takes the form of a balanced diode mixer. Although suitable
for DSB AM detection the circuit is best applied in SSB receivers and for this reason
is not described here but is dealt with in detail later on in the course.

210 Chapter 3 : Section 9 Issued November 2006

DETECTION DISTORTION
In addition to the distortion produced by detector non-linearity and incorrect filter
component values, distortion can be caused if the following stage is of low input
impedance. This type of distortion, called ‘peak clipping’, is also dependent upon the
depth of modulation. It is obviated in the design stage by ensuring that the input
impedance is large compared with the diode load resistance. A typical output
waveform from such a distorted circuit would appear as shown in Fig 11 below.

FIG 9 EFFECT ON WAVEFORM OF EXCESSIVE LOADING OR HIGH DEPTH OF MODULATION

Issued November 2006 Chapter 3 : Section 9 211

DETECTION OF CW SIGNALS
The demodulation of CW signals presents a major problem in ordinary receivers,
since there is no modulation present as such. However, if the received signal can be
‘mixed’ with another one such that their difference frequency is an audio frequency,
this can then be demodulated in the normal manner. Fortunately, because the
mixing is a non-linear process, we can use the detection device to do both detection
and mixing.

FIG 10 SIMPLIFIED CW RECEIVER

To ensure that interference is not caused when receiving normal AM signals, the
beat frequency oscillator (BFO) as it is called, is switched on only for CW reception.
It is switched off, usually by disconnecting the HT. Additionally it is usually made
variable (IF to  1 kHz) to allow the operator to adjust the AF tone for his/her own
comfort.
Typical BFOS are Hartley or Colpitts types, loosely coupled to the receiver chain to
prevent interaction.

212 Chapter 3 : Section 9 Issued November 2006

REVISION QUESTIONS
IF Amplifiers
1 The stage(s) in a superhet receiver providing most gain and selectivity is
(are):

2 Given that the signal level is relatively small and the need to avoid distorting
the modulated carrier wave, identify the class of bias used in an IF amplifier.

3 One reason why a receiver should have a high IF is because:

4 Give one reason why a receiver should have a low IF.

5 For a communications receiver, the double superhet design is preferred to the

single superhet because the double superhet gives:

6 State why the IF of a MW/LW broadcast receiver is typically 470 kHz.

Issued November 2006 Chapter 3 : Section 9 213

7 State a typical IF bandwidth requirement for a MW broadcast receiver.

8 What is the effect on the overall bandwidth when identical IF amplifiers are
connected in cascade? The bandwidth is:

Variable Selectivity
9 One method of providing variable selectivity in a receiver is by:

10 Give another method of providing variable selectivity in a receiver.

Detectors
11 The complement of modulation is:

12 State one advantage of the shunt diode detector as compared with the series
diode detector.

214 Chapter 3 : Section 9 Issued November 2006

13 State one disadvantage of the series diode detector as compared with the
shunt diode detector

14 A simple series detector circuit operates from an IF amplifier of 47 kHz. A

suitable time constant for the load and reservoir capacitor would be:

15 What results if the diode detector load time constant is made too short?

16 What results if the diode detector load time constant is made too long?

17 How will the output of a diode detector be affected if the input signal level from
the last IF amplifier is too small?

18 Why is it preferable to have a reasonably large signal appearing at the input of

the series diode detector circuit?

Issued November 2006 Chapter 3 : Section 9 215

19 One effect of excessive shunt loading of a detector circuit is:

20 In the series diode detector circuit, in what order are the 3 components of the
detector output processed?

BFO and CW Operation

21 Why is there no output from a domestic receiver when it is tuned to a CW
transmission?

22 State which stage must be added to a receiver to provide an audio output for
CW input signals.

23 If the intermediate frequency of a superhet receiver is 465 kHz, what will be

the approximate frequency of the BFO when receiving a CW signal on 1.1 MHz?

216 Chapter 3 : Section 9 Issued November 2006

SECTION 10 : CONTROL OF GAIN

INTRODUCTION
Signals received by a receiver are all at different levels depending upon transmitter
power aerial systems, range and propagation path. It is essential, therefore, that the
gain of any receiver should be controllable to allow adjustment of its output for the
complete range of inputs. Furthermore, to counteract the effects of fading, which
causes continual variation in signal input levels, it is advantageous to have an
automatic system so that the receiver output is maintained at its desired level.
A typical receiver has 2 completely separate gain control systems:
 Manual gain control
 Automatic gain control (AGC)
With the exception of the simple manual AF gain control, which is a variable
potentiometer placed across the detector output, gain control relies upon
non-linearity of active device transfer characteristics. Where this non-linearity is not
sufficient in normal devices, such that a change in bias causes a change in gain,
special devices have been designed. A typical FET transfer characteristic is shown
below.

FIG 1 TRANSFER CHARACTERISTIC – N CHANNEL FET

For the transfer characteristic shown, gm at point B is clearly larger than at point A.
therefore, for a given signal input and output load, the gain is dependent, to a large
extent, on bias point.
Typical block schematic diagrams of receivers employing gain control are shown in
Fig 2 overleaf. are all low pass filters with direct coupling
between IF, RF and detector stages. Instability will result if filtering (decoupling) is
ineffective.

Issued November 2006 Chapter 3 : Section 10 217

FIG 2 RECEIVER WITH AGC

FIG 3 RECEIVER WITH MANUAL RF/IF GAIN CONTROL

Since the RF stages do not provide as much gain as the IF stages, and also since
reducing the RF gain too much will reduce the signal/noise ratio, a reduced variation
in bias is applied to the RF stages.
The last IF stage does not normally have gain control applied to it because:
 The last IF stage is much smaller than the other IF stages due to detector
damping of the stage.
 The amplitude of the signal is large and a vari- characteristic would distort
the signal.
Multiplicative mixers may have AGC applied to them, but not additive mixers (initial
bias condition).

218 Chapter 3 : Section 10 Issued November 2006

EFFECT OF VARIABLE BIAS

If variable bias is applied to an FET RF or IF amplifier, the reverse biased gate
depletion layer would vary in width and cause a change in input capacitance. This
can result in the stage going off tune.
The effect is less troublesome in bipolar transistor circuits because of the lower input
impedance of the forward biased base-emitter junction. This lower input impedance
can, however, have a damping effect on the output of the previous stage.
The effects of detuning can be minimised by:
 Making the tuning capacitor large so that changes in input capacitance of the
amplifier are negligible compared with it.
 Aligning the receiver on weak signals at maximum gain.
 Avoiding stray input capacitance.

AUTOMATIC GAIN CONTROL

Automatic gain control in a receiver is achieved by feeding a bias derived from the
carrier signal back to the RF and IF stages. It may also be fed to the AF stages.
The polarity of the bias and whether it is in current or voltage form is determined by
the amplifying devices used, for example:
 FETs. Positive or negative voltage, depending on type and circuit
 Bipolar Transistor. NPN – negative current, PNP – positive current.
The most convenient place to obtain a dc voltage or current which is proportional to
the carrier input level, is the detector stage. Consider the circuit in Fig 4 below.

FIG 4 DIODE DETECTOR

At point A, the signal has 2 main components: AF signal and a dc voltage. The dc is
filtered off to allow the AF to pass on for further amplification. By taking an output
from A through the low pass filter, only dc will result and this is proportional to the
input carrier level.

Issued November 2006 Chapter 3 : Section 10 219

FIG 5 DETECTOR WITH LOW PASS FILTER

At first glance it would seem that fulfils exactly the same function as .
However, remembering the demodulation process, filter out the IF, a very
much higher frequency than the AF present at point A. The component values are
therefore very much different. Typical component values for a transistor circuit could
be
As the receiver input signal level rises, the output rises and so the AGC voltage
becomes more negative (or positive if diode connection is reversed) and when
applied to the appropriate receiver stages, it reduces the gain and hence the receiver
output tends to remain constant. The AGC voltage is applied via a long CR filter
(time constant 0.1 – 0.2 second), as described above.

220 Chapter 3 : Section 10 Issued November 2006

SIMPLE AGC
Since AGC can only reduce the gain of a receiver if it is applied all the time as in
simple AGC, the gain of the receiver will be reduced all the time, even on weak
signals. Since any output from the above detector will produce an AGC voltage
which will reduce the gain, this a real disadvantage with weak signals when the
receiver gain is required to be at maximum. It is quite a common occurrence to
combine signal and simple AGC demodulation in one relatively straightforward circuit
as shown in the example at Fig 6. Note, that with one or 2 re-arranged components,
the circuit is very similar to the one illustrated in Fig 5. effectively isolates the
effects of the AGC line from the signal detection circuitry. provides the AGC
line time-constant with operating as an IF/AF decoupling capacitor.

FIG 6 COMBINED SIGNAL AND AGC DETECTION

Below, receiver output/incoming signal strength is illustrated, with and without AGC.

FIG 7 EFFECT OF AGC

The disadvantages of this system can be summarised:

 weak signals further weakened since Rx gain reduced
 AGC voltage always less than peak IF voltage.

Issued November 2006 Chapter 3 : Section 10 221

DELAYED AGC
No AGC is produced until the receiver output reaches a pre-set minimum level. The
receiver gain is then at a maximum on weak input signals. Once the receiver output
exceeds the pre-set level, AGC is produced. It should be noted that the delay
referred to here is concerned with voltage, not time.
Consider the simple delay AGC circuit illustrated in Fig 8 below.

FIG 8 DELAYED AGC CIRCUIT

A positive bias is applied to the cathode of the diode (the delay bias) and its
value is determined by and the values of . The diode cannot conduct and
produce AGC until the IF signal at anode exceeds the delay bias, which is
developed across and . When conducts, a voltage is developed across ,
negative at the anode of (wrt earth). This voltage is distributed between R3 and
, the AGC line being taken from their junction and is negative wrt earth.
With simple AGC, the signal detector can also be used to produce AGC. With
delayed AGC, this is no longer possible and a separate AGC diode is required. Like
simple AGC, the AGC voltage can never be greater than the peak IF voltage.
Connecting the AGC line to the collector rather than to the secondary of the IF
transformer makes the AGC response smoother, since it has a reduced damping
effect on the last IF transformer.

222 Chapter 3 : Section 10 Issued November 2006

The delay may be achieved by biasing a transistor as shown in Fig 9 below.

FIG 9 DELAYED AGC USING BIAS THRESHOLD

Below receiver output/incoming signal strength response is illustrated and shows:

 No AGC
 Simple AGC
 Delayed AGC

FIG 10 EFFECT OF DELAYED AGC

Issued November 2006 Chapter 3 : Section 10 223

DELAYED AMPLIFIED AGC

In some equipments (eg. ILS) it is essential that the output level of the receiver be
constant. In order to do this, the AGC must have a high level of control, so the AGC
voltage must be larger than the peak IF. Therefore, some form of amplification of the
AGC voltage must be used.
There are 2 main methods
a Use of an Additional IF Amplifier. This has the disadvantage that the
amplification is carried out at high frequency.

( A)

b Use of Low Frequency or dc Amplification of AGC Detector Output. This

has the advantage that dc amplifiers do not require the screening an IF
amplifier would, so this method is more common.

( B)

FIG 11 ALTERNATIVE DELAYED AMPLIFIED AGC METHODS

224 Chapter 3 : Section 10 Issued November 2006

Illustrated in Fig 12 are responses showing:

 No AGC.
 Simple AGC.
 Delayed AGC.
 Delayed amplified AGC.

FIG 12 EFFECT OF DELAYED AMPLIFIED AGC

MISCELLANEOUS AGC CIRCUITS

The following AGC circuits are included as typical examples.

FIG 13 AGC CIRCUIT FROM AD 36- ILS/VOR RECEIVER

acts as the AGC detector, and form a dc amplifier.

Issued November 2006 Chapter 3 : Section 10 225

FIG 14 AGC AMPLIFIER

An extra stage of IF amplification is used here to give amplified AGC.

FIG 15 AGC CIRCUIT FROM AD 260 ADF RECEIVER

226 Chapter 3 : Section 10 Issued November 2006

FIG 16 AGC LINE

POST DETECTOR AGC

Since the AGC bias is derived from the stage whose amplification it controls (IF
amplifier), it is impossible to obtain a perfectly constant output. The mere fact that
the AGC voltage can vary shows the output varies, since the AGC is derived from
the output of the IF amplifier.
In order to obtain a more constant output, some of the amplifying stages after the
detector may also be controlled by the AGC.
Conventional AGC is termed back acting since it only controls stages before the
AGC detector. Where stages after the detector are controlled, the AGC is called
forward acting, audio or post detector AGC.
In order to accomplish this, care must be taken to prevent distortion, especially with
large signal amplitudes, due to curvature of the transistor characteristic. The gain of
transistors is controlled by varying the bias current and distortion will be produced if
the signal amplitude is too large.

Issued November 2006 Chapter 3 : Section 10 227

REVISION QUESTIONS
AGC and Control of Gain
1 If the gain of the IF amplifiers is to be controlled either automatically or
manually, what special feature must the active device possess?

2 Which of the output signal components from a simple detector circuit can be
used to provide simple AGC?

3 The stages of a receiver always controlled by an AGC system are:

4 A simple AGC would be defined as a system which:

5 One limitation of simple AGC is:

6 One limitation of delayed AGC is:

7 Define the term post detector AGC:

228 Chapter 3 : Section 10 Issued November 2006

SECTION 11 : NOISE LIMITING

SERIES AND SHUNT NOISE LIMITERS
In general, noise produces large amplitude, but short duration changes of the signal,
producing pulse type outputs from the detector. These pulses may be reduced in
amplitude by a limiter circuit.

FIG 1 NOISE PEAKS

SHUNT NOISE LIMITER

FIG 2 SIMPLE NOISE LIMITER

On positive half cycles of the input signal conducts, but only when the voltage
across it is greater than approximately 0.3V germanium diodes (Ge) or 0.6 V silicon
diodes (Si). Thus, any positive going noise pulses above 0.6 V (or 0.3 V) are
shunted to earth and do not reach the next stage of the receiver.
Similarly, on negative half cycles shunts to earth any signal voltages greater than
its forward bias voltage.
The combined effect of and is to restrict the signal amplitude to peak-to-peak
voltages of 0.3 V (Ge diodes) or 1.2 V (Si diodes). Noise pulses greater than these
voltages are shunted to earth and not passed to the next stage.
The limiting level can be varied by biasing the diodes, as shown in Fig 3 below.

Issued November 2006 Chapter 3 : Section 11 229

FIG 3 PRACTICAL NOISE LIMITING

SERIES NOISE PEAK LIMITER

FIG 4 SERIES NOISE PEAK LIMITER

The detector load is formed by and which, if their values are identical, will be
half the detected voltage and negative wrt earth.
The detector output is also fed to the long CR (47 mS) consisting of . Hence, at
C there will be a dc level proportional to the carrier level.
Under normal signal conditions, will be charged to the mean carrier level. Point B
is at half the detected output. will always be biased on. So the AF signal at B will
pass onto the AF amplifiers.

230 Chapter 3 : Section 11 Issued November 2006

Noise pulses at the detector output will not affect the dc potential at C because of the
long time constant .
At B, however, the voltage will become more negative and cut off for the duration
of the noise pulse, reducing the amplitude of noise fed to the AF amplifier.
The potential divider determines the depth of modulation at which limiting
occurs.
forms a low pass filter which attenuates the level of intermediate frequency
components appearing at anode. is the load for across which the audio
output is developed before being coupled to the audio amplifier stages.

NOISE LIMITING CIRCUITS

Under quiescent conditions, the forward resistance presented by each diode is
relatively high with 0.6 V developed across each diode and 0.1 V across and .

FIG 5 NOISE LIMITER IN ELLIOT 21

Positive half cycles of input signal reduce the current through and cause a
corresponding increase in current through and increase in voltage across .
Conversely, the negative half cycles increase the current through and reduce the
current through and . Consequently for normal signals, the voltage across ,
the output voltage, follows the input voltage.
The signal level available at point A is approximately 0.3 V peak. If a positive noise
voltage at point A exceeds 0.6 V peak, cuts off and has no further effect on the
voltage at point C. If the noise voltage is negative and exceeds 0.6 V peak, cuts
off and point C falls to zero. Thus noise signals greater than 0.6 V peak are limited.

Issued November 2006 Chapter 3 : Section 11 231

FIG 6 NOISE LIMITER

The noise limiter diode is forward biased by the combination. The

incoming signal is positive going.
Positive noise pulses in speech signal cause an increase in positive potential at the
junction of and at . Due to the time constant of the anode potential
cannot follow this sudden change. The diode will therefore be cut off, removing the
pulse from the speech waveform.

MUTING
A high gain receiver will produce considerable noise when:
 No signal is being received.
 A new channel is being selected.
This is due to the fact that the noise level may not be strong enough to produce
AGC.
This high noise level can produce fatigue in the operator. To overcome this, the
receiver output is cut off when no signal is being received.
This process is known as:
 Muting.
 Squelch.
 Quiescent AGC.

232 Chapter 3 : Section 11 Issued November 2006

Muting is generally carried out by:

 Applying a cut-off bias to an AF stage.
 Operating a relay which earths the receiver output when only noise is
received.
Muting may be derived from:
 Carrier Level. When the carrier level at the detector is above a certain level,
the signal is fed to the AF amplifier. When the carrier level is below that level,
the signal is blocked.
 Signal/Noise Level. Only when the signal/noise level is above a preset level
is the signal fed to the AF amplifier.
Consider the circuit in Fig 7. the action of this circuit is very similar to delayed AGC
detection. The demodulating diode is biased off by the positive voltage on its
cathode.
The voltage produced from wiper is taken via AF gain and IFT to the demodulator
anode so that when wiper is raised, the muting threshold voltage level is reduced.
Only signals overcoming the bias across the demodulator will be detected and
produce an audible output. Thus, only the receiver is muted when the input to the
demodulator is not of sufficient amplitude. This is termed ‘muting threshold level’.
An alternative form of muting is to mute the signal when the signal-to-noise ratio falls
below a certain level.

FIG 7 SIMPLE BIASED DEMODULATOR

Issued November 2006 Chapter 3 : Section 11 233

SQUELCHING
Carrier Squelch Circuit
This is operated from the AGC line. The voltage on this line is proportional to the
strength of the carrier and when the carrier exceeds a minimum level conducts.
(For the purposes of explanation, assume that the AGC line goes more negative as
the signal level increases).

FIG 8 CARRIER SQUELCH CIRCUIT TAKEN FROM RECEIVER T6401

234 Chapter 3 : Section 11 Issued November 2006

With no input from either the AGC line or the squelch AF and noise amplifiers,
and are off and is on. The collector of is thus on the positive peak and
diode is biased on, shunting the output of the detector to earth, therefore there is
no input to the AF amplifier and the receiver is muted.
An input from as the carrier level rises will cause and to cut on, to cut off
and a negative voltage to be applied to the anode of . is then biased off and
the detected signal is fed to the AF amplifier.

SQUELCH CIRCUITS
Low and high pass filters separate signal and noise, which are separately detected in
opposite senses and then compared. The resultant output is fed via a dc amplifier to
a Schmitt trigger, which operates the AF gate in the signal line. The present control
at the comparator input determines the signal-to-noise ratio at which the gate
opens.

SIGNAL/NOISE SQUELCH CIRCUIT

The detected signal has a full bandwidth of 50 kHz at this point and is split into 2
paths:
 Through an AF filter, allowing 0-3 kHz AF only through to the squelch AF
amplifier. The output of this amplifier is rectified to produce a negative voltage
proportional to the AF signal amplitude.
 Through a high pass filter which cuts out the AF, leaving only noise signals.
This is fed to the squelch noise amplifier and is rectified to produce a positive
voltage proportional to the noise amplitude.
These negative AF and positive noise signals are added and which is bigger controls
the base of the comparator transistor . The noise signal keeps biased off. The
AF signal, if larger than the noise, will bias on. This will, in turn, cut and on
and off, which will bias off and allow the AF signal through to the AF amplifier.
The noise signal, if larger, will keep and off and on. This puts a positive
bias on which biases it on and shunts the detector output on earth.

SQUELCH OVERRIDE CIRCUIT

When several transmitters are operating on nearby frequencies, heterodyning may
result in spurious frequencies being produced which may be interpreted as noise, ie.
signal-to-noise ratio may decrease the squelch level. If the desired signal output
from the AGC circuit is of sufficient amplitude to overcome the delay set by the
Schmitt trigger is operated to open the gate despite the output from the comparator.

Issued November 2006 Chapter 3 : Section 11 235

REVISION QUESTIONS
Noise Limiters
1 In the series noise peak limiter, the signal detecting circuit is essentially:

2 Which of the following statements is correct regarding the series noise peak
limiter?

Muting
3 The purpose of a muting circuit in a communications receiver is to:

4 Muting is normally carried out in a communications receiver by:

5 From which of the following reference signal levels may muting be derived?

236 Chapter 3 : Section 11 Issued November 2006

SECTION 12 : AF AMPLIFIERS
INTRODUCTION
The AF amplifier stages are usually standard amplifiers working over the 300 Hz to 3
kHz range although the upper frequency limit may be extended in specialised
equipment. Normally 2 or 3 stages are included depending on the output power
required, and the detector output.

FIG 1

CHIEF CHARACTERISTICS
The chief characteristics which determine the design of the AF stages are:
 Maximum output power requirement, determined by the load.
 Sensitivity, determined by the detector output.
 Power gain, determines the number of stages and the amplifier efficiency.
 Gain – frequency response, to prevent distortion an even response is required
over the designed audio bandwidth.

Issued November 2006 Chapter 3 : Section 12 237

 Distortion, may be caused by:

o Harmonics, produced in the driver stage due to non linearity
(particularly the second).
o Intermodulation, frequencies are produced which are the sum or
difference of frequencies present in the input. This distortion affects
the second and subsequent stages. As a rule harmonic and
intermodulation distortion have the same cause and to reduce non
linearity to a minimum, strong negative feedback is normal.
o Incorrect loading, maximum power transfer requires a particular load.
Other load values than this will reduce efficiency and introduce
distortion. Af power amplifiers work in an optimum load which allows
the highest power to be developed while distortion is kept below
designed level.
o Waveform shape, distortions in the AM envelope before detection are
simply amplified in the AF amplifiers.

238 Chapter 3 : Section 12 Issued November 2006

Four separate inputs are provided, 3 for use in transmit mode, the other being used
in the receiver mode. The AF output is fed to earphones on receive and balanced
modulator on transmit.
Since low level low power modulation is used, the output of this amplifier is adequate
for modulation purposes. If high power modulation of an RFPA anode was required,
then a much higher AF power output would also be required.
acts as a buffer stage between the complex input mixing and the input.
and a conventional AF amplifier with negative feedback from collector to
emitter (via ).
The frequency range need only be 300-3000 Hz, since that is all that is transmitted.
The transformer in the output stage will have a turns ratio determined by what it is
matched onto, eg. headphones, balanced modulated or the anode of an RFPA
stage. The ratio being given by:

= turns ratio
= secondary impedance
= primary impedance

ADDITIONAL FUNCTIONS
 Volume Control. The volume of sound needs to be adjusted for comfortable
listening. Control over volume is achieved in the output stages using a
potentiometer to vary the signal applied at the input to the first AF amplifier.
 Post Detector ACG. Automatic control of gain is normally achieved in the IF
stages in order to maintain constant output against varying signal power. In
some receivers the AGC voltage may be fed forward to the AF stages in order
to improve constancy in the output but it is important that with large signals
distortion is not introduced.

Issued November 2006 Chapter 3 : Section 12 239

REVISION QUESTIONS
1 The minimum upper frequency of the signal in the AF amplifier of an R/T
receiver is normally:

2 Two essential characteristics of an AF amplifier are:

3 The optimum load on an AF amplifier allows:

4 A potentiometer in the input circuit to the first AF amplifier of a receiver is for:

5 The reason why post detector AGC may be used instead of back-acting AGC
is that it:

6 The reason why aerial tuning is not a very satisfactory way of ensuring a
receiver is on tune is because:

240 Chapter 3 : Section 12 Issued November 2006

AMPLITUDE MODULATED RECEIVERS

Abbreviations
AF Audio frequency
AGC Automatic gain control
AM Amplitude modulation
Amp Amplifier
B Bandwidth or base
BFO Beat frequency oscillator
BSD Block schematic diagram
Bn Overall bandwidth
C Capacitance or collector
Cp Padder capacitor
Ct Trimmer capacitor
Cn Neutralising capacitor
CW Continuous wave or carrier wave
D Drain or diode
DB Decibel
∆ Delta
dc Direct current
Demod Demodulator
Det Detector
DSB Double sideband
E Emitter
f frequency
FC Frequency changer
FM Frequency modulation
G Gate
gc Conversion conductance
I Current
IF Intermediate frequency
I/P Input
ILS Instrument landing system
K Kelvin or Boltzman’s constant
kHz Kilohertz
L Inductance

Issued November 2006 Chapter 3 : Section 12 241

LF Low frequency
LO Local oscillator
LPF Low pass filter
Max Maximum
MHz Megahertz
Min Minimum
MCW Modulated carrier wave
m milli
mix Mixer
NF Noise factor
O/P Output
PA Power Amplifier
Q Amplification or selectivity of tuned circuit
R Resistance
R/T Radio telephony
Rx Receiver
RF Radio frequency
S Source or Siemen
SF Signal frequency
SSB Single sideband
S/N Signal-to-noise ratio
Superhet Superheterodyne
T Temperature, or time, or transformer, or turns ratio
TR Transistor or field effect transistor
TRF Tuned radio frequency
 Micro
V Volt
Vcc Supply voltage
Vgs Voltage between gate and source
Vo Voltage output
VHF Very high frequency
W Watts
Zp Impedance of primary
Zs Impedance of secondary

242 Chapter 3 : Section 12 Issued November 2006

SECTION 13 : MICROPHONES
INTRODUCTION
A transducer is a device which converts one form of energy into another. A
microphone is therefore a type of transducer as it converts sound energy into
electrical energy.

CARBON MICROPHONES
Carbon microphones, moving coil microphones, dynamic microphones and
condenser microphones are used in aircraft.

FIG 1 THE CARBON MICROPHONE

From Fig 1 it is seen that the carbon microphone consists of a cup shaped metal
container C having a carbon ‘back’ electrode E fixed inside. A carbon plunger P is
positioned at the entrance of C, and is attached to a light aluminium or paper
diaphragm D. The space between P and E inside C, is filled with carbon granules G.
The whole is contained in a bakelite or metal case, having a grill screwed over the
front.

Issued November 2006 Chapter 3 : Section 13 243

OPERATION
For its operation the carbon microphone utilises the fact that when any two adjacent
carbon granules are compressed as in Fig 2(b), their contact area is increased and
thus their electrical contact resistance is reduced.

FIG 2

Conversely, when they are released as in Fig 2(a), their contact area decreases and
thus their electrical contact resistance increases.
A dc polarising battery V is required to drive current  around the circuit through R,
and through the granules in the cup C.
When the diaphragm is steady this current is steady dc. When the operator speaks
into the microphone, the sound waves strike the diaphragm and cause it to vibrate in
sympathy, and this causes the plunger P to move to and fro accordingly. Thus the
carbon granules are alternately compressed and released at the speech frequencies
and their resistance varies in a like-wise manner, consequently current  rises and
falls in a similar manner, and is an electrical facsimile of the incoming sound waves.
Current  is thus varying dc and the voltage across R will be an audio voltage
superimposed on a steady dc voltage component. This dc component may be
removed by a capacitor as shown in Fig 3 or a transformer as shown in Fig 4.

FIG 3

244 Chapter 3 : Section 13 Issued November 2006

FIG 4

CHARACTERISTICS
 Requires an energising voltage (polarising supply)
 Is noisy – produced by the movement of the carbon granules
 Poor frequency response
 High voltage output (approximately 100 mV to 500 mV)
 Reasonably robust
 Low impedance (typically 20 ohms to 100 ohms)
 Single ended output (ie. microphone + ve and earth)

THE MOVING COIL MICROPHONE

FIG 5 MOVING COIL MICROPHONE

From Fig 5 this can be seen to consist of a specially shaped permanent magnet M in
the form of a ring. In the centre of this is a central pole or limb. A strong magnetic
flux exists in the gap between this central pole and the ring magnet.
Making a loose fit over the central pole is a light paper tube P which is attached to a
light metal or paper diaphragm D. On the outside of the tube P is a winding C of fine
insulated copper wire known as the SPEECH COIL. It is positioned in the strong
radial flux in the air gap between the magnetic poles. The ends of the speech coil
form the output terminals of the microphone.
When no sound waves are striking the diaphragm, it is steady, and so the moving
coil is stationary in the magnetic flux. Thus no voltage is induced into the coil, and
no output occurs. When the operator speaks, the incoming sound waves cause the
diaphragm to vibrate in sympathy, and so the moving coil is moved backwards and
forwards in the magnetic flux between the poles. Consequently an alternating signal
voltage is induced into the coil, and this voltage is an exact representation of the
incoming sound waves.

Issued November 2006 Chapter 3 : Section 13 245

MOVING COIL CHARACTERISTICS

 No energising voltage required
 Good signal/noise ratio
 Good frequency response (reasonably constant to 10 kHz)
 Low voltage output (approximately 2mV to 5mV)
 Low impedance (typically 20 ohms to 300 ohms)
 Balanced output normally connected to the primary of an audio transformer
with its centre tap at earth.

THE DYNAMIC MICROPHONE

This type of microphone is a combination of the carbon type and the moving coil type
microphones.
The dynamic microphone uses the moving coil microphone capsule which gives
good frequency response and low noise. The output is then fed to an integral
amplifier supplied by what would be the microphone polarising supply to increase the
output amplitude up to the level for carbon microphones (ie. approximately 250mV to
500mV).
The dynamic microphone output connections are similar to the carbon microphone,
ie. a single ended output.

NOISE CANCELLATION
As you will appreciate, the ambient noise level in an aircraft is usually high due to
engine noise, etc. One method used to minimise this noise is to use noise cancelling
microphones. They use the principle of equal pressure on either side of the
microphone diaphragm to cancel the effect of this ambient noise. (Equal and
opposite forces cancel).
Speech will not be seriously affected as the speech only comes from one side of the
diaphragm.

246 Chapter 3 : Section 13 Issued November 2006

THE CONDENSER OR ELECTROSTATIC MICROPHONE

FIG 6(A)

Fig 6(a) shows the physical construction of the condenser microphone. When sound
waves move the diaphragm, the capacitance between diaphragm and back plate
changes.
An energising voltage supply shown in Fig 6(b) is connected to the capacitance via a
large value of resistor R. the result is an output voltage proportional to the
movement of the diaphragm. This type of microphone is used as the area or cockpit
microphone in some cockpit voice recorders.

CHARACTERISTICS
Frequency response poor at very low frequencies, (which is desirable for an aircraft
environment) and then relatively constant up to approximately 10 kHz.
Energising or polarising supply voltage needed.
Low voltage output (typically 5mV to 10mV), and high impedance with the result that
the microphone is normally used with an integral amplifier or an amplifier physically
close to the microphone.

Issued November 2006 Chapter 3 : Section 13 247

NOTES:

248 Chapter 3 : Section 13 Issued November 2006

SECTION 14 : TELEPHONES/LOUDSPEAKERS
INTRODUCTION
 A telephone is a transducer which converts electrical energy into sound
energy.
 Moving coil telephones and moving iron telephones are used on aircraft.
 First consider the moving coil telephone.

MOVING COIL TELEPHONE

The construction here is identical to that of the moving coil microphone described
earlier. The action is also similar though remember we are now considering the
reverse case. Figs 1(a) and (b) show the basic construction and action.

FIG 1(A) FIG 1(B)

The audio speech current is passed through the speech coil C. One half cycle make
the top end of the speech coil a N pole, and so it is attracted downwards to the S
central pole of the magnet as shown in (a). the next half cycle makes the top end of
the speech coil a S pole, and so it is repelled upwards away from the central S-pole
of the magnet. Thus the diaphragm vibrates to and fro, generating sound waves in
the process.

Issued November 2006 Chapter 3 : Section 14 249

MOVING IRON TELEPHONE

The permanent magnet exerts a constant pull on the thin diaphragm which is slightly
‘concaved’ when no current is passing through the coils. When audio frequency
signal current passes through the coils one half cycle will energise the pole-pieces
such that they AID the magnetism of M and thus pull the diaphragm in further. The
other half cycle magnetises the pole-pieces such that they OPPOSE the magnetism
of the permanent magnet. Consequently the pull on the diaphragm decreases, and it
moves out. Thus an AF current passing through the coils causes the diaphragm to
vibrate to and fro at the same frequency. In doing so it generates sound waves.
These sound waves cause the operator’s ear drum to vibrate in sympathy and thus
he hears the signal.
The permanent magnet is essential. If it were not present to ‘polarise’ the device,
then the generated sound waves would be TWICE the frequency of the audio
currents flowing in the coils. It is called the POLARISING MAGNET.

FIG 2 MOVING IRON TELEPHONE

250 Chapter 3 : Section 14 Issued November 2006

STANDARD MICROPHONE AND TELEPHONE CONNECTORS

These are shown in Fig 3, 4 and 5. They are labelled to help you (on the NATO plug
R = red, B = blue, G = green, and Y = yellow, which is the colour of the respective
wires).

FIG 3

FIG 4

Issued November 2006 Chapter 3 : Section 14 251

FIG 5

LOUDSPEAKERS
A loudspeaker is used in many receivers to convert audio electrical signals into
sound waves. In an aircraft there are often one or more loudspeakers fitted for use
by the aircrew or passengers.
Several features and characteristics of loudspeakers are similar to those of
telephones.

252 Chapter 3 : Section 14 Issued November 2006

MOVING COIL LOUDSPEAKER

FIG 6 LOUDSPEAKER CONSTRUCTION AND ACTION

The construction of a moving coil loudspeaker is shown in Fig 6.

An intense radial magnetic field is produced in an annular air gap between the poles
of a magnet which has a cylindrical centre pole surrounded by an outer pole.
This magnet may be of the permanent type or an electro-magnet, which requires a
separate dc supply to energise the coil.
A coil of fine wire, known as the moving coil, or speech coil, is wound on an
insulating collar, firmly attached to a cone shaped diaphragm having a flexible
suspension around its edge. The coil is supported so that it is free to move axially, in
the air gap between the poles of the magnet, at right angle to the magnetic flux.
Audio frequency currents flowing in the speech coil produce axial forces (in
accordance with FLEMING’S LEFT HAND RULE) which set the cone in vibration
with frequency and amplitude corresponding to those of the currents in the coil.
Connections to the coil are brought out by flexible leads.

Issued November 2006 Chapter 3 : Section 14 253

A baffle board is necessary for the correct operation of a loudspeakers. It prevents

the sound wave from the front of the cone cancelling that from the back, since a
decrease in air pressure is created at the back, simultaneously with an increase at
the front and vice versa. The baffle board isolates the front of the speaker from the
back as shown in Fig 7. The larger the baffle board the better is the isolation, but of
course there is obviously a limit to its physical size. In practice, the baffle size is
determined by the lowest frequency which must be reproduced without loss of
power, and it is found that the baffle must extend at least one quarter of a
wavelength (/4) at this frequency around speaker as shown in Fig 7.

FIG 7

254 Chapter 3 : Section 14 Issued November 2006

CHAPTER 4 : SSB AND FM COMMUNICATIONS

SECTION 1 : SINGLE SIDEBAND TRANSMISSION
WHAT IS SSB

FIG 1 CONVENTIONAL AM DSB SYSTEM

FIG 2 SSB SUPPRESSED CARRIER SYSTEM

SSB differs from AM transmission in that:

 Only one sideband is transmitted.
 The carrier power can be very reduced or virtually non-existent.
SSB has particular advantages over DSB AM in certain circumstances:
 Crowded frequency spectrum conditions, eg. Citizen Band transceivers.
 Noisy conditions, eg. radio-telephone systems.
 Limited available power source, eg. mobile radio.

Issued November 2006 Chapter 4 : Section 1 255

FORMS OF SSB

SUPPRESSED CARRIER
The carrier is completely suppressed and only one sideband is sent. This is the
most efficient form of SSB since all the transmitted power contains information,
however, detection at the receiver is more complicated since the carrier is missing.

FIG 3

PILOT CARRIER
The carrier is sent at reduced power together with one sideband. The presence of
the carrier makes detection easier.

FIG 4

256 Chapter 4 : Section 1 Issued November 2006

CONTROLLER CARRIER
The carrier is sent at full power during breaks in the modulation (ie. when no
information is being sent). When information is being sent then the system reverts to
suppressed carrier operation. The power consumed by the transmitted is dictated by
the amount of information sent. The more information transmitted, the less time the
carrier is ‘on’, so the more efficient the system.

FIG 5

Issued November 2006 Chapter 4 : Section 1 257

NOTES:

258 Chapter 4 : Section 1 Issued November 2006

SECTION 2 : BALANCED MODULATOR

OPERATION
The balanced modulator is a common circuit in SSB transmitters where it is used to
produce an output consisting of upper and lower sidebands without the carrier.

FIG 1

In the FET balanced modulator circuit diagram shown at Fig 1, the RF is fed in phase
to the 2 FETs so the outputs from the 2 transistors at the RF or carrier frequency will
cancel in the drain tuned circuit since it is fed in antiphase by the 2 FETs.

Issued November 2006 Chapter 4 : Section 2 259

The drain tuned circuit does not provide a load for the AF so there is no output at AF.
When both AF and RF are applied, the AF signal makes one FET conduct more than
the other (AF signal fed in antiphase to the 2 FETs), so there will then be a resultant
RF output, the amplitude of which will depend on the amplitude of the AF as shown.

FIG 2

This envelope is not a normal amplitude modulated waveform. The curve ABCDE is
not a sine wave as it would be in an AM waveform.

FIG 3

This is the result of the carrier being absent. The only frequencies present in the
output being the (carrier + AF) and the (carrier – AF).

260 Chapter 4 : Section 2 Issued November 2006

This modulator gives the 2 sidebands with no carrier.

FIG 4

DIODE QUAD OR RING BALANCED MODULATOR

FIG 5

NOTE: In the following circuits where a diode is not conducting it is omitted from
the circuit.

Issued November 2006 Chapter 4 : Section 2 261

CIRCUITS WITH CARRIER (RF) ONLY APPLIED

In both of the cases below, each diode conducts by an equal amount and resultant
output voltage is zero.

FIG 6 FIG 7

CIRCUITS WITH AF ONLY APPLIED.

In this case on alternate half cycles of AF input the primary of the output transformer
is open circuit and there is no AF output.

FIG 8 FIG 9

262 Chapter 4 : Section 2 Issued November 2006

CIRCUIT WITH BOTH INPUTS APPLIED

On the half cycle of AF shown conducts more than by an amount dependent
on the AF amplitude. Thus there is an output. On the other half cycle of AF input
conducts more than again giving a resultant output proportional to the AF
amplitude.

FIG 10

FIG 11

The output waveform is the characteristic double sideband signal with carrier
suppressed. The output consisting of RF + AF and RF – AF frequencies only.

Issued November 2006 Chapter 4 : Section 2 263

PRODUCTION OF SSB BY FILTERS

FIG 12 BLOCK DIAGRAM OF SSB FILTER SYSTEM

The output from the balanced modulator consists of both sidebands. The sideband
filter allows only one sideband through.
The sideband filter can consist of:
 Simple LC filters – many sections may be needed to get the required
characteristic
 Electro-mechanical filters.
 Crystal lattice filters.
 Ceramic filters.

264 Chapter 4 : Section 2 Issued November 2006

CHAPTER 3 : MECHANICAL FILTERS

FIG 1

MAGNETOSTRICTION EFFECT.
Materials, such as nickel and ferrites, elongate or shorten in the presence of a
magnetic field. In the mechanical filter, the input signal is converted to mechanical
vibrations which are applied to the first section of coupling wires. These vibrate
longitudinally and apply torsional oscillations to the first disc. The oscillations are
then transmitted to the remaining sections of the filter and the output transducer
converts them back to electrical oscillations.

CHARACTERISTICS
 Mechanical elements electrically equivalent to a very high Q circuit.
 Frequency response depends on the dimensions of the elements.
 Small, rugged and stable (mechanically and temperature wise).

FIG 2 RESPONSE CURVES FOR 2 MECHANICAL FILTERS

Issued November 2006 Chapter 4 : Section 3 265

NOTES:

266 Chapter 4 : Section 3 Issued November 2006

SECTION 4 : CRYSTAL FILTERS

FIG 1

CHARACTERISTICS
The dip without neutralisation via Ch is caused by the high impedance the crystal
presents at its parallel resonance. This can be countered by Cn.
If Cn is adjusted to equal the holder capacitance, which causes the parallel
resonance, then the current in the output coil due to Ch will be cancelled out by an
equal current due to Cn. Hence the effect of Ch is cancelled out and the dip in the
response curve due to the parallel resonance of the crystal will disappear.
The resulting response curve is extremely narrow and falls off rapidly each side of
resonance because of the high Q of the crystal. A response curve having a wider
pass band will be required normally and crystal band pass filters of the above type
are not commonly used. Circuits using 2 or more crystals can give greater
bandwidths.

Issued November 2006 Chapter 4 : Section 4 267

PRACTICAL CRYSTAL FILTERS

Refer to Fig 2. The input and output coils are tuned to the centre frequency and the
series resonant frequencies of the crystal will be at the edge of the pass band
and as shown. The sharp cut-off is obtained by capacitors and . If
adjusted correctly they can cause the dip of one crystal to appear above and the
dip of the other crystal to appear below . In some cases a flatter top may be
obtained by letting each crystal consist of 2 crystals in parallel each with a difference
resonant frequency.

FIG 2 FIG 3

Crystals and are a matched pair, and so are and . The series frequency
of one pair is the same as the parallel frequency of the other pair.

FIG 4

268 Chapter 4 : Section 4 Issued November 2006

Fig 5 shows a crystal band pass filter in which the crystal frequencies are slightly
staggered in order to obtain the required frequency.

FIG 5

Filters of the magneto-striction and crystal types provide a well defined pass band.
They are smaller than LC types and their response curve shows less alteration with
temperature and age than LC filters. There is no need for alignment and they are
robust.

Issued November 2006 Chapter 4 : Section 4 269

NOTES:

270 Chapter 4 : Section 4 Issued November 2006

SECTION 5 : SSB (SINGLE SIDE BAND) TRANSMISSION

BASIC SSB TRANSMITTER

FIG 1

Fig 1 shows a BSD of a basic SSB transmitter. The single sideband from the filter is
mixed with an HF oscillator frequency to produce an output at the final transmitter
frequency. If the final Tx frequency is VHF then more than one stage of mixing may
be required.
Frequency multiplication cannot be used since it would increase the bandwidth of the
signal, by the multiplication factor.

SSB TRANSMITTERS

FIG 2

Issued November 2006 Chapter 4 : Section 5 271

SUPPRESSED CARRIER
 500 kHz Oscillator. This is a crystal oscillator whose frequency is
maintained as constant as possible by use of a temperature controlled
oven which is periodically checked against a standard crystal.
 Balanced Modulator. This gives a double sideband output with the
carrier (500 kHz) suppressed. The 2 sidebands are:
500 kHz + (300 to 3000 Hz) = (500.3 to 503 kHz) and
500 kHz - (300 to 3000 Hz) = (499.7 to 947 kHz)

 Band Pass Filter. A highly selective electromechanical filter allows the

upper sideband 500.3 to 503 kHz through but rejects the lower sideband.
 Mixers. The output, varying between 500.3 – 503 kHz is mixed with a
higher frequency, eg. 6 MHz. The output of the mixer is tuned to about
6,500 kHz thus selecting the sum frequency and converting the signal to
the frequency band 6,500 kHz. A further mixer converts this to a still
higher frequency band 26,500 kHz – 26,504 kHz by use of a 20 MHz
input.
 Power Amplifier. The previous stages worked at lower power. The
output is now amplified to feed the aerial system. As the signal,
consisting of a mixture of frequencies is not of constant amplitude, Class
A or Class B stages must be used.
 The 6 MHz and 20 MHz would be variable in frequency allowing tuning
through a range of output frequencies.

272 Chapter 4 : Section 5 Issued November 2006

PILOT CARRIER

FIG 3

The basic transmitter is similar to the suppressed carrier SSB Tx except that after
the sideband has been filtered out, the carrier (500 kHz) is added in by the
combining circuit at reduced amplitude.

CONTROLLED CARRIER

FIG 4

When there is no AF and therefore no sideband being produced, the control circuit
allows the carrier (500 kHz) to be fed to Mixer 1 as shown. When there is AF, the
control circuit cuts the feed of carrier of Mixer 1 and only the sideband from the filter
is fed to the mixer.

Issued November 2006 Chapter 4 : Section 5 273

REVISION QUESTIONS
1 The output of a balanced modulator consists of:

2 In a diode quad balanced modulator with only the RF carrier applied:

3 In the filter method of SSB transmission the purpose of the filter is to:

4 Frequency multiplication cannot be used in a SSB transmitter because:

5 The advantage of producing SSB transmission by the outphasing method is:

6 In an SSB transmitter the carrier is transmitted only when the sideband is not
being transmitted. Such a carrier is termed:

7 In controlled carrier SSB:

274 Chapter 4 : Section 5 Issued November 2006

8 Class C bias cannot be used in the PA stage of SSB transmitters because:

9 The most efficient method of SSB transmission is:

10 One advantage of using SSB transmission in preference to AM DSB is that:

11 Single sideband transmission uses:

12 Which of the following is an SSB system susceptible to:

13 In a standard AM signal, audio intelligence is contained in:

14 In a single sideband suppressed carrier system, the receiver must:

Issued November 2006 Chapter 4 : Section 5 275

15 In pilot carrier SSB, the carrier is:

16 The output of a balanced modulator circuit is:

17 A crystal lattice may be used in a:

18 The output of the filter circuit of a single sideband transmitter is 500.3 kHz. To
convert this to a single sideband transmission on 4.5 MHz, it would be necessary
to:

19 The amplifying stages following the balanced modulator in an SSB Tx do not

work in Class C because:

20 Compared with DSB transmissions, SSB transmissions:

276 Chapter 4 : Section 5 Issued November 2006

SECTION 6 : SSB RECEIVERS

SIMPLE SSB RECEIVER
Fig 1 shows a simple schematic of a SSB/AM receiver. Most of the stages are
standard superhet stages.

FIG 1

RF Amp
 provides amplification of the aerial signal.
 selects the required frequency.
 keeps the receiver noise factor at a minimum.
 reduces oscillator radiation.

Local Oscillator
 provides a stable output thus maintaining a constant IF.

Mixer
 combines the RF digital with the LO output to produce the IF.

Issued November 2006 Chapter 4 : Section 6 277

IF Amplifier
 provides most of the gain the receiver.
 determines the overall selectivity of the receiver.

Detector
 extracts the required information from the signal as faithfully as possible
AFC
 compensates LO frequency for changes in RF input carrier or LO
changes

Crystal Filter and Amplifier

 Filters out the carrier at the IF for separate amplification and for AFC

OPERATING PRINCIPLES
The BSD shows a simple pilot carrier SSB receiver. The receiver differs from a
conventional DSB AM receiver in the way in which the carrier frequency is treated.
Because the carrier is attenuated it has to be restored to a full value before detection
of the signal. At the same time it is important to ensure that the receiver is kept on
tune, particularly in the IF amplifier stages. This requires a filter circuit which detects
changes in carrier frequency and varies the LO output to maintain a constant IF, ie.
automatic frequency control.
It should be noted that in suppressed carrier SSB receivers AFC based on the carrier
is obviously not possible. Alternative techniques are used in this case which are
described later.

TRIPLE SUPERHET RECEIVER

FIG 2

The triple superhet structure has high gain and selectivity which makes it most
suitable for the reception of the narrow band SSB signal.

278 Chapter 4 : Section 6 Issued November 2006

The final IF is at a relatively low frequency of, say, 85 kHz or 100 kHz, and at this
frequency it is easier to achieve the required gain and bandwidth in order to:
 assist in achieving the necessary output signal-to-noise ratio, and
 give good adjacent channel rejection characteristics.
If the final IF is very stable, the third IF amplifier bandwidth can be made exactly the
same as the bandwidth of the incoming signal, thus minimising noise interference.
Any change in frequency of the local oscillators employed in the receiver will cause
the final IF frequency to vary with SSB. The carrier reinserted in the detector will no
longer be correct and the receiver output will be distorted. The IF signal will not be
correctly aligned with the last IF amplifier’s bandwidth and so part of the sideband
may be lost.
When automatic frequency control is used, the final IF will be stable against
frequency variations in the local oscillators and frequency drift in the transmitter.

PRODUCT DETECTOR

FIG 3

This form of detector is ideally suited for use with SSB. It is really a form of balanced
mixer, the AF being the lower sideband produced (RF-LO). This type of detector has
to be used with suppressed carrier SSB, but is also suitable for DSB AM application.

Issued November 2006 Chapter 4 : Section 6 279

Assume a frequency of 500 kHz

applied at input 1 and a
frequency of 501 kHz applied at
input 2. The combination of the
2 signals produces waveforms
at points A and B and shown.

Any positive-going half cycle

at A causes D1 to conduct
depositing electrons on the
top plate of C2 and
effectively shorting A to E,
giving no output across R2.
At the same time, a positive-
going signal at B causes D2
to cut off and current to be
drawn from E, through R2
and R4, to the bottom of C2
– thus producing an output
across R2.

Any negative going half cycle

at A causes D1 to cut off and
current to flow from the top
plate of C1 through R3 and R2
to E. At the same time, a
negative-going signal at B
causes D2 to conduct, passing
current from the bottom plate of
C2 to E. This effectively shorts FIG 4 RESULTANT OUTPUT ACROSS R2 AF
COMPONENT SHOWN DOTTED
B to E, giving no output across
R2.

Hence, current flows in opposite directions through R2 on alternate half cycles and
the final output voltage is the mean of the 2 voltages produced by these currents – in
this case, 1 kHz.

280 Chapter 4 : Section 6 Issued November 2006

SUPPRESSED CARRIER SSB RECEIVER (SCSSB Rx)

FIG 5

Issued November 2006 Chapter 4 : Section 6 281

The BSD in Fig 5 shows the triple superhet receiver modified for use in the SCSSB
mode.
Note:
 absence of incoming carrier based signals
 internally generated automatic frequency control (AFC)
 AF derived automatic gain control (AGC)

NOTE: Carrier based AFC and AGC is pointless since, with SCSSB, a carrier
signal is not transmitted.

For stable frequency operation, the receiver on SCSSB mode relies on:
 the transmitted signal remaining precisely on frequency.
 a very stable, locally generated carrier signal being incorporated in the
circuitry.
The carrier oscillator drives a frequency synthesiser which, through a system of
frequency multipliers and dividers, controls the frequency of the local oscillators. To
avoid distortion in the receiver output, this re-inserted carrier must be within a few
hertz (Hz) of the original nominal carrier frequency.
Note that the AGC is derived from the audio stages and not from the demodulator as
in double sideband receivers. The audio signal is rectified in the AGC detector
before being passed to the RF and IF stages as a variable dc bias voltage.
Incorporated in the low frequency IF amplifiers (but not shown in the BSD) will be a
pair of switched sideband filters (USB and LSB).
Demodulation is normally done by a product detector, which may take form similar to
the balanced demodulator already described under AM detectors. This stage mixes
the sideband with the output of the reinserted carrier oscillator to produce the
difference frequency – the audio output. It should be noted that the normal envelope
type of detectors (eg. collector bends, infinite impedance, series diode, etc) are
entirely unsatisfactory for use with SSB signals.
It is not necessary for the SSB receiver to be a triple superhet, in practice most
airborne types are double superhets. Fewer mixers means less noise generated and
a consequent lowering of the noise factor.
Shown in Fig 6 is a BSD of a double superhet SCSSB receiver showing sideband
filters, amplifiers and reinserted carrier oscillators with its control of the local
oscillators.

282 Chapter 4 : Section 6 Issued November 2006

FIG 6

Issued November 2006 Chapter 4 : Section 6 283

NOTES:

284 Chapter 4 : Section 6 Issued November 2006

SECTION 7 : COMPARISON OF SSB AND DSB AM SYSTEMS

ADVANTAGES OF SSB
SSB has a number of advantages over DSB transmissions. These are chiefly:
 Power saving.
 Reduced bandwidth.
 Noise reduction.
 Reduction in multipath or selective fading.
The amount of power which can be saved using SSB depends upon the form of SSB
transmission used, ie. the amount of carrier present and the depth of modulation.
Example:
A 2 kW carrier modulated to a modulation index of 0.7 gives:

Total transmitted power of DSB AM:

Percentage power saving with one sideband removed:

Percentage power saving of a suppressed carrier SSB transmission:

In most applications the SCSSB transmission allows minimum power saving of 75%
compared to DSB AM.
NARROW BANDWIDTH
In crowded frequency spectrum conditions, SSB can allow almost twice as many
channels to be used within a given bandwidth. This can be seen in Fig 1 below.

Issued November 2006 Chapter 4 : Section 7 285

FIG 1

Because noise is directly proportional to bandwidth, SSB transmissions with reduced

bandwidth contain reduced noise.
One of the most difficult problems to overcome in radio reception is that of fading.
This can be caused in 2 ways.
 Multipath Fading. Using skywaves the signal may take different reflective
paths and arrive at the receiver out of phase. The resultant destructive and
constructive interference at the receiver aerial results in a varying audio
output.
 Selective Fading. In this case fading is usually caused by the interference of
adjacent channels in the selectivity circuit of the receiver.

REDUCED FADING
SSB transmissions help in reducing the fading problem because of the narrower
bandwidth.
Although the advantages of SSB communications are considerable, there are a
number of drawbacks which result in SSB only being used when it is superior in
performance.
The chief disadvantages of SSB are:
 more complexity circuitry in both transmitter and receiver
 frequency stability is more critical in SSB systems
 weaker SSB signal can be overpowered at the receiver by stronger
transmissions, and
 carrier re-insertion requires accurate generation of a stable frequency of the
correct frequency and amplitude.

286 Chapter 4 : Section 7 Issued November 2006

SECTION 8 : AUTOMATIC FREQUENCY CONTROL

AFC IN SSB RECEIVERS

FIG 1 SSB RX WITH AFC

In SSB systems when the carrier is absent or weak, it is very important in the
receivers to ensure that the carrier frequency in the transmitter is very precise and
stable. In SC SSB systems where the carrier is generated and re-inserted within the
receiver the local oscillator used must be similarly very precise.
To ensure accuracy in the frequency of the re-inserted carrier, AFC is applied to the
second LO. After the carrier has been filtered, it is fed to a frequency discriminator
circuit. This circuit will give an output proportional to the deviation of the carrier from
the frequency to which the discriminator is tuned (positive if carrier too high, negative
output if carrier too low).
This output is fed to a variable reactance device (reactance transistor or Varactor
diode). The variable reactance varies the frequency of the second LO so that the
carrier is always at the correct frequency.

Issued November 2006 Chapter 4 : Section 8 287

THE RATIO DISCRIMINATOR

Fig 2 shows the circuit of the ratio discriminator

FIG 2

The diodes conduct simultaneously (when anode is positive and cathode is

negative). The components and are as before except for the output
being taken from between their respective junctions, instead of the junctions being
joined and connected to earth.
has a critical effect on the circuit operation. It has a very large value. It charges
up to the peak-to-peak value of the carrier and the very long time constant,
prevents the voltages on changing except in response to long term
changes in the carrier amplitude. As far as the detection of the audio is concerned,
could be replaced by a battery of the correct voltage. cathode might be at,
say +6V and anode at –6V.
The symmetry of and means that when the carrier is unmodulated,
and have equal charges and the potentials at points A and B are both zero.
Frequency above IF
When the frequency of the carrier rises above the IF, is driven harder than .
The voltages on and increase and decrease respectively as their charging
currents alter. However, the sum of their voltages is fixed so the voltage at point A
must fall relative to that at point B, which remains at zero.
Frequency below IF
When the frequency of the carrier falls below the IF, is driven harder than .
This time the voltages on and decrease and increase respectively as their
charging currents alter and the voltage at point A falls relative to that at point B.

288 Chapter 4 : Section 8 Issued November 2006

ADVANTAGES OF THE RATIO DISCRIMINATOR

The ratio discriminator has 2 main advantages over other types:
 It is insensitive to input amplitude variations. Any modulation of the carrier
amplitude is filtered out by the time constant . This network acts
as a LPF and AF signals fall within its stop band, perhaps > 10 Hz.
 The fixed voltage across the network can be used as a source of AGC voltage
because it changes only in sympathy with carrier amplitude changes which fall
within the passband of the LPF. To use this feature either the top or bottom
rails or point B is earthed or connected to some other specific fixed voltage.

DISADVANTAGE OF THE RATIO DISCRIMINATOR

The main disadvantage of the ratio discriminator is that it is not as linear as some
others, therefore the distortion levels are greater.

Issued November 2006 Chapter 4 : Section 8 289

REVISION QUESTIONS
SSB Receivers
1 Why is the double superhet receiver suitable for SSB signal reception?

2 The SCSSB receiver differs from the AM broadcast receiver in that SSB
receivers require:

3 In a communications receiver, the BFO can be used for the following

operations:

4 The result of small change in the frequency of the first local oscillator
employed in a double superhet receiver when tuned to a SCSSB signal could be:

5 Which of the following, when referring to a double/triple superhet receiver,

results in the use of a relatively low final IF?

6 The final IF stage in a SSB receiver should be stable in order to:

290 Chapter 4 : Section 8 Issued November 2006

7 How is AGC produced in a SCSSB receiver?

8 One method used to control the frequency of the local oscillators is a double
superhet receiver is to:

9 In a filter system SSB receiver, the carrier frequency is re-inserted at the:

10 To detect the signal in a suppressed carrier SSB superhet receiver, use can
be made of a:

11 In detecting a suppressed carrier single sideband transmission:

12 Many single sideband receivers use a triple superhet. This is because:

13 With a suppressed carrier single sideband signal, AFC is obtained from:

Issued November 2006 Chapter 4 : Section 8 291

NOTES:

292 Chapter 4 : Section 8 Issued November 2006

SECTION 9 : FREQUENCY MODULATION TECHNIQUES

FREQUENCY MODULATION (FM)
In FM, the carrier is transmitted at constant amplitude but its frequency is controlled
by the AF signal.
The amount by which the carrier deviates from its nominal (centre) frequency
depends on the amplitude of the AF signal.
The rate at which the carrier deviates depends on the frequency of the AF
signal.

FM FREQUENCY SPECTRUM
The frequency content of an FM wave is very much more complicated than that of an
AM waveform. The FM signal contains many more sidebands spaced at intervals
determined by the modulating signal frequency. In addition the number of sidebands
and their amplitudes vary with the FM modulation index.
It must be noted that the FM modulation index (M) is not the same as the Am
modulated factor (m). For a given modulating signal frequency and frequency
deviation the FM modulation index is given by:

……………..(1)

For example, if a modulating signal of 15 kHz is used to produce FM with a deviation

of 75 kHz and the modulation factor is 0.5 then the FM modulation index is:

Notice that whereas m is always< 1, M can be any value and is sometimes very
large.
We can illustrate the effect of modulation index (m) on sideband amplitudes by a
family of curves as shown in Fig 1. Here the carrier frequency is shown as a
damped oscillation as are the sidebands ( etc). Their relative amplitudes are
represented by the instantaneous values of these waves at any given value of M.
(Note that the sidebands etc. are more usually labelled as etc). These
curves (called Bessel functions) show that:
 The carrier signal or a particular sideband may not exist at certain values of
M.
 The total power in the FM signal remains constant but at different modulation
depths the energy is distributed differently among the sidebands. (Note that
the negative relative amplitude values are squared to give positive power
values).

Issued November 2006 Chapter 4 : Section 9 293

FIG 1

It can be seen from equation 1 that in FM signals the modulation index M varies:
 directly with the frequency deviation
 inversely with the modulating frequency
Because deviation varies with the amplitude of the modulating signal, M can be
varied in 2 ways:
 by keeping the frequency of the modulating audio signal constant and varying
its amplitude.
 by keeping the deviation constant and changing the frequency of the
modulating signal
In both cases, higher values of modulation index cause more sidebands and
changes in their relative amplitudes. The effect is shown in Fig. 2.

FIG 2 FM FREQUENCY SPECTRUM VARIATION

294 Chapter 4 : Section 9 Issued November 2006

It is important to remember that in FM both amplitude and frequency of the

modulating signal are changing so the resulting frequency spectrum is a complex
resultant in which the number and relative amplitudes of sidebands are both
determined by the modulation index. A typical frequency spectrum of an FM wave
with a modulation index M = 5 would appear as shown in Fig 3 below.

FIG 3 FREQUENCY SPECTRUM FOR M=5

FM PRODUCTION BY PHASE MODULATION

In phase modulation the instantaneous phase of a carrier is varied in accord with a
modulating signal. If the modulating voltage is an audio signal then the output of a
phase modulation will be FM.
FM and PM are forms of angle modulation and in both cases the frequency and
phase of the carrier are made to deviate. The difference between them is in the way
in which the parameters of the modulating signal are encoded into the carrier and
this is summarised in Table 1 below.

Frequency Deviation Phase Deviation

FM Proportional to the modulating Directly proportional to the
signal voltage modulating signal voltage and
inversely proportional to its
frequency
PM Directly proportional to the Proportional to the modulating
voltage and frequency of the signal voltage
modulating signal.

TABLE 1 COMPARISON OF PHASE AND FREQUENCY MODULATION

Issued November 2006 Chapter 4 : Section 9 295

BANDWIDTH OF FM
Bandwidth in FM is calculated as the frequency range contained between the
outermost sidebands. Since the sidebands are spaced apart and are in pairs, the
bandwidth is given by:

For example, a 15 kHz modulating signal at a modulating signal at a modulating

index M = 5 gives 6 pairs of sidebands with significant amplitudes. The bandwidth is
thus:

This value which is typical of good quality FM broadcasting can be compared with
speech communications in mobile systems where a maximum of 3 kHz is used
with a frequency deviation of 15 kHz giving a bandwidth (for m = 1 and M = 5) of
51 kHz.
The above values would achieve very high quality FM broadcasting but in practice
lower values are used for economy of wavelength usage.
The standard deviation for VHF/FM broadcasts is and the approximate
bandwidth as calculated by Carson’s rule is:

To reduce the possibility of adjacent channel interference in domestic FM, the BBC
allows a total bandwidth of 200 kHz. Since the required bandwidth is 180 kHz, this
leaves 20 kHz, 10 kHz either side, therefore the separation between 2 adjacent BBC
transmitters will be 20 kHz.
Each channel is therefore allocated 200 kHz bandwidth. Such systems are terms
BROADBAND to distinguish them from NARROWBAND FM systems. Narrow band
systems normally operate with a standard deviation of eg. CB radio.
Obviously the bandwidth of such systems is much less than for broadcast systems.
It should be noted that Carson’s Rule is only an approximation and the actual
bandwidth varies with depth of modulation and the number of significant side
frequencies generated. The actual value is calculated in a different way as shown
above.

296 Chapter 4 : Section 9 Issued November 2006

FIG 4 BROADBAND FM CHANNEL SPACING

FM TRANSMITTERS

FIG 5 BASIC FM TRANSMITTER

From Fig 5 it can be seen that the modulating signal is applied to the master
oscillator via:
 microphone
 audio amplifier
 variable reactance device
The modulating signal varies the oscillator frequency only, not its amplitude.

Issued November 2006 Chapter 4 : Section 9 297

CRYSTAL STABILISED FM TRANSMITTER

We have seen that frequency stability is important in FM transmitters because any
spurious movement of oscillator frequency will result in an audible note being
received.
The frequency stability of the transmitters is determined primarily by the crystal
oscillator. Since the crystal oscillator is very stable, any variations in the frequency
output of the mixer will be due to the MO.

FIG 6

The mixer output is applied to a discriminator circuit which produces a dc output

proportional to the output frequency of the mixer. The dc is applied to a reactance
device to alter the effective value of C or L across the master oscillator tuned circuit.
The AFC adjusts the frequency of the MO and brings it back to the correct centre
frequency.
A low pass filter can be used to stop the modulation from affecting the reactance
device so that its bias is only affected by a drift in the master oscillator’s centre
(average) frequency.

298 Chapter 4 : Section 9 Issued November 2006

FM RECEIVERS
FM domestic receivers differ from AM in a number of ways:
 Higher operating frequencies, eg. VHF and above compared to HF and below
for AM.
 Construction – shorter leads and smaller components.
 Greater bandwidth in the received signal.
 The detector is usually a frequency discriminator (eg. Foster Seeley or ratio
detector).
 Larger signal bandwidth requires wideband tuned circuits particularly in the If
amplifier stages.
 Different IF frequencies are used.
 The stages required are shown in the following BSD.

FIG 7 FM RECEIVER WITHOUT AFC

The following stages perform the same functions as similar stages in an AM receiver.
 RF amplifier
 Mixer
 Local oscillator
 IF amplifier
 AF amplifier
The mixer and IF amplifiers perform the same function but require different circuit
techniques to deal with the FM signal. These techniques are described in the
following section. The remaining stages are:
 Limiter. If the frequency discriminator is amplitude sensitive then the signal
amplitude must be made constant before detection of frequency variation.
The limiter is thus not always required.
 Frequency discriminator. Detects frequency variation in the FM signal and
produces and audio frequency output.

Issued November 2006 Chapter 4 : Section 9 299

If AFC is applied then an output from the detector stage is first filtered to give a
control voltage to a variable reactance circuit which may be incorporated in the local
oscillator.

FIG 8 FM RECEIVER WITH AFC

In conditions where a weak signal is being received which is adjacent to the strong
signal the AFC tends to lock onto the strong station. This is prevented by an AFC
DEFEAT switch which open-circuits the AFC loop allowing manual tuning to the
weaker station.

INTEGRATED CIRCUIT FM RECEIVER

Modern FM receivers reduce the circuitry by using integrated circuit.

FIG 9 IC FM RECEIVER

300 Chapter 4 : Section 9 Issued November 2006

AUTOMATIC GAIN CONTROL

AGC in an FM receiver is normally derived from the amplitude of the input signal to
the limiter stage and is applied, as in AM receivers to the earlier IF amplifier stages
and the RF amplifier. This is shown in Fig 10 below.

FIG 10 FM RECEIVER WITH AGC

The de-emphasis stage shown in Fig 16 is dealt with in detail in the next section.
De-emphasis is needed in receivers which take in signals from transmitters in which
a pre-emphasis circuit has been employed to give gain to high frequencies.

ADVANTAGES OF FM
The advantages of FM over AM are:
 Less susceptible to adjacent channel interference due to the large bandwidth
which each channel occupies
 Since the signal is of constant amplitude the system can be made to be less
influenced by external noise which mainly consists of amplitude variations.
 High quality transmissions are possibly (a property really of the large
bandwidth available at VHF and the wider audio range used 20 Hz 
15 kHz).
 Class C stages operating at maximum efficiency can be used since the signal
is of constant amplitude.
 Because of the number of sidebands which exist in the FM spectrum, the
information contained in the sidebands is less easily lost through interference
and processing.
 In AM any 2 signals transmitted with the same carrier frequency will interfere
in a receiver. In FM the stronger signal is received and a weaker signal is
eliminated. This phenomenon is called the capture effect.

Issued November 2006 Chapter 4 : Section 9 301

DISADVANTAGES OF FM
 FM uses a larger bandwidth than an AM signal of comparable quality.
 Can only be used at VHF and above due to large bandwidth required.
 Transmissions are limited to line of sight. This is because VHF waves are
very easily reflected from large objects, eg. hills, buildings, trees, etc. and the
reflections cause signal strength fluctuations (watt for watt however, FM
transmitter range is about 4 times that of an AM transmitter).

FM TECHNIQUES
It has been seen that FM transmitters and receivers have high operating frequencies
and require frequency control in both transmitters and receivers.
The following circuits are commonly employed in FM systems to produce the
required modulation and frequencies.
 Phase modulator
 Reactance transistor modulator
 Varactor diode modulator
 Ferrite modulator
 Pre and de-emphasis circuits
 Frequency multipliers
 Frequency synthesisers
These circuits are described in the following paragraphs.

PRE AND DE-EMPHASIS

While FM systems are almost immune to amplitude modulated noise by the use of
limiters or ratio detectors in the receiver, they are still prone to noise. Random noise
frequencies, when mixed with the incoming frequency, produce a noise output from
the discriminator. This noise is more troublesome at high audio frequencies than at
low audio frequencies.

FIG 11

302 Chapter 4 : Section 9 Issued November 2006

As can be seen from the diagram, the signal/noise ratio at high audio frequencies is
not as good as it is at low audio frequencies. In order to improve the signal/noise
ratio of the high audio frequencies, they are transmitted at a higher level than the low
audio frequencies by using a PRE-EMPHASIS CIRCUIT in the transmitter. The
output of the discriminator now has the same signal/noise ratio for both high and low
audio frequencies.

FIG 12

In order to flatten the overall frequency response, the output of the discriminator is
fed to a DE-EMPHASIS CIRCUIT before being passed onto the AF amplifier.

FIG 13

Issued November 2006 Chapter 4 : Section 9 303

FREQUENCY MULTIPLICATION IN FM
Frequency multiplication is necessary in FM transmitters where it serves 3 purposes:
 It achieves the high operating frequency normal in FM (eg. 88-108 MHz) from
a low frequency
 It allows the MO to operate at low frequency which helps its stability
 It enables the large frequency deviation in FM to be obtained
There may be more than one stage of multiplier in the FM transmitter as the
multiplication factor of each stage is kept low. (Normally  2  3). The multipliers
are operated in Class C which produces:
 an output rich in harmonics
 maximum efficiency

FIG 14

FREQUENCY SYNTHESIS
Frequency synthesis has many applications in communications, transmitters and
receivers. In FM, indirect synthesis where a VCO is used in a PLL, has the
advantage that should the crystal oscillator fail then an output of reduced stability is
still available. This is important where safety or emergencies are involved. This
approach can be applied in a crystal stabilised FM Tx or as the LO in an FM Rx
where it performs a similar function to the AFC loop.

FIG 15 INDIRECT SYNTHESISER

One major advantage of the circuit in Fig 15 that it combines the good short-term
stability of the VCO with the good long term stability of the crystal oscillator.
In digitally controlled FM the indirect synthesiser is combined with a microprocessor
which allows remote control of stations and storage of complex search and channel
selector patterns together with BIT routines.

304 Chapter 4 : Section 9 Issued November 2006

SECTION 10 : FM RECEIVER STAGES COMPARED TO AM

INTRODUCTION
It was mentioned in the last chapter that many FM stages perform the same function
as they would in AM receivers but the circuit techniques employed are usually
different.

RF AMPLIFIERS
Gain is always required in the early stage of FM receivers whereas in AM receivers
the RF amplifier can sometimes be omitted. The gain increases the receiver
sensitivity. Using an RF amplifier in an FM receiver assists in improving image
channel rejection characteristics and the noise factor of the receiver.

MIXER STAGE
This stage produces the wanted IF which is the sum or difference between signal
and local oscillator frequencies. The mixer may be a VHF mixer as shown in Fig 1
below. In this circuit and have values which:
 adjust the relative strengths of the 2 inputs thus isolating them and preventing
pulling.
 match the input circuits to the mixer.
and acts as an input tuned circuit. and act as a filter selecting the
difference frequency from the mixer output. (RFC) isolates the mixer output
frequencies from .

FIG 1 VHF MIXER

More commonly FM receivers make use of a combined oscillator mixer. Fig 2 shows
a typical converter circuit in which a Meissner oscillator arrangement is combined

Issued November 2006 Chapter 4 : Section 10 305

with a non-linearly biased amplifier. The output tank circuit is tuned to the difference
frequency between the LC oscillator and the RF signal input.

FIG 2 OSCILLATOR-MIXER

IF AMPLIFIERS
The IF amplifiers in FM receivers differ from AM IF amplifiers in 3 mains respects.
 There may be more stages of IF amplification
 The last stage may also act as a limiter
 The bandwidth is much greater
In AM receivers using a bandwidth of 9 kHz the IF amplifiers can be simply critically
tuned to give the required bandwidth. If FM, however, the IF amplifiers are centred
on 10.7 MHz and have a flat response between 10.6 and 10.8 MHz, ie. a bandwidth
of 200 kHz. This is achieved using broadband amplifiers in which bandwidth is
widened by the following methods.

306 Chapter 4 : Section 10 Issued November 2006

 Stagger Tuning. Three stages are progressively tuned to give wide flat
response. This method is not favoured for FM as it does not increase gain
and the amplifiers are difficult to align.

FIG 3 STAGGER TUNED IF AMPLIFIER STAGES

 Flat Tuning. Each stage is tuned to the same frequency but each is
resistively damped. The overall gain is low but the response is useably flat.

FIG 4 RESISTIVELY DAMPED FREQUENCY RESPONSE

 Overcoupling. The second IF amplifier stage is overcoupled to give a double

peak response. The first and second stages are flatly tuned by resistive
damping. The overall response becomes flatter than that obtained from flat
tuning alone but the stages are difficult to align.

FIG 5 OVERCOUPLING

Issued November 2006 Chapter 4 : Section 10 307

FM LIMITER
A limiter is necessary to clip or suppress noise peaks which have been added to the
FM transmission. In addition the Foster-Seeley discriminator is amplitude sensitive
so a limiter is needed before detection where this type of discriminator is used. The
ratio detector does not require as much limiting and in this case the limiter may be
omitted. Noise peaks are then removed by the last IF amplifier.
A typical FET form of FM limiter is shown in Fig 6 below.

FIG 6 FET FM LIMITER

In this circuit gate leak bias causes the positive peaks of the input to drive the FET
gate positive which clips the positive peaks. The negative peaks are removed by
operating at a low voltage so that the bias is beyond cut-off.
Since the limiter needs an input in the range 5 to 10 volts at least 2 stages of IF
amplifiers are needed before the limiter stage. Even then using only one limiting
stage may not operate adequately so the second stage is required which makes use
of the gain of the first stage.
Note that limiting does not affect the information contained in the modulated signal.
Fig 7 shows that even a high depth of limiting still leaves the signal with sufficient
amplitude for adequate detection.

FIG 7 FM LIMITING

308 Chapter 4 : Section 10 Issued November 2006

REVISION QUESTIONS
Frequency Modulated Transmitters
1 An RF carrier is being frequency modulated by a fixed AF frequency. What
will be the changes in the resultant output waveform if the modulating frequency:

2 Why is the use of FM normally confined to frequency ranges of VHF and

above?

3 Why is an FM signal affected less by external noise than an AM signal?

4 State 2 reasons for employing frequency multiplication in an FM transmitter.

5 Why is it not normally possible to use a crystal controlled oscillator as the

fundamental oscillator in an FM transmitter?

6 How can a crystal oscillator be used to control an FM transmitter?

7 The mean frequency of an FM transmitter increases, what circuits are

required in the AFC loop and how would they be connected to return the
frequency to its correct value?

8 Explain briefly what is meant by pre-emphasis.

9 An FM signal with a deviation  Hz is passed through a frequency tripler.

The wave in the output of the tripler will have a deviation of:

Issued November 2006 Chapter 4 : Section 10 309

10 An FM signal with a deviation of 5 kHz is passed through a mixer, and has its
frequency reduced 5-fold. The deviation in the output of the mixer is:

11 A pre-emphasis signal provides extra noise immunity by:

12 Which of the following statements is correct?

13 FM can broadcast with higher fidelity than AM mainly because:

14 The reason for using an AFC circuit in an FM transmitter is to:

15 In an FM transmitter, with an oscillator followed by 2 trebler stages and a PA

stage, the modulated stage would be:

16 The bandwidth required for broadcast FM is approximately:

310 Chapter 4 : Section 10 Issued November 2006

17 Increasing the frequency (but not the amplitude) of the modulating signal
applied to an FM transmitter will cause the:

18 Increasing the amplitude of the modulating signal to an FM transmitter will

cause:

19 The purpose of the discriminator in a crystal stabilised FM transmitter is to:

20 The amplifiers following the modulated state in an FM transmitter would for

maximum efficiency work in class:

21 A mixer stage in an FM transmitter will:

Issued November 2006 Chapter 4 : Section 10 311

REVISION QUESTIONS
FM Receivers
1 The rule relating the amplitude of a modulating signal to the deviation of the
carrier frequency is approximately that the amplitude of modulating signal is:

2 The rule relating the frequency of a modulating signal to the rate of change of
carrier deviation is approximately that the frequency of the modulating signal
is:

3 In an FM receiver, the local oscillator usually operates below the incoming

signal frequency for the following reason:

4 Why is the tracking between signal and LO circuits in a VHF/FM receiver

easier than for a receiver tuned to the MF band?

5 How may the required wide bandwidth in FM receivers be achieved?

6 The discriminator must be preceded by a limiter in an FM receiver because:

312 Chapter 4 : Section 10 Issued November 2006

7 Why do FM receivers usually have more IF stages than equivalent AM

receivers?

8 How does the degree of AGC in VHF/FM receivers compare with conventional
MW/AW receivers? The level of AGC is:

9 In a ratio detector:

10 The purpose of stagger tuning in an FM receiver is to:

11 If the bandwidth of the IF amplifiers in an FM receiver is too narrow, then the

receiver output will be distorted on:

12 Frequency modulation gives an improved signal-to-noise ratio compared with

amplitude modulation because:

Issued November 2006 Chapter 4 : Section 10 313

13 A limiter is used in an FM receiver because the:

14 The required bandwidth in the IF stages of a receiver may be obtained by

using:

15 An advantage of the ratio detector over other types of discriminator circuits is

that:

16 The de-emphasis circuit in an FM receiver consists of a resistor and capacitor

in series connected so that the:

314 Chapter 4 : Section 10 Issued November 2006

SSB AND FM COMMUNICATION SYSTEMS ABBREVIATIONS

Ae aerial
AF audio frequency
AFC automatic frequency control
AGC automatic gain control
AM amplitude modulation
Amp amplifier
Attn attenuation
beta
bal mod balanced modulator
BSD block schematic diagram
C capacitance
CB citizen band
Ch capacitance of holder
Cn control
d diode
delta
dB decibel
dc direct current
det detector
demod demodulator
disc discriminator
diff difference
DSB double side band
E earth
Eo RF voltage of oscillator
fc frequency carrier
fd frequency deviation
fm modulating frequency
FM frequency modulation
fo frequency of oscillator
freq frequency
freq syn frequency synthesiser
gm mutual conductance
HF high frequency

Issued November 2006 Chapter 4 : Section 10 315

Hz hertz
IC integrated circuit
Id drain current
IF intermediate frequency
I/P input
kHz kilohertz
kW kilowatts
L inductance
LO local oscillator
LPF low pass filter
L/S loud speaker
LSB lower sideband
M modulation index
m modulation factor
MHz megahertz
mic microphone
mix mixer
MO master oscillator
O/P output
osc oscillator
PA power amplifier
PC pilot carrier
Pc carrier power
PLL phase locked loop
PM phase modulation
Q amplification or selectivity
R resistance
rel relative
res resultant
RF radio frequency
RFC radio frequency choke
Rx receiver
Sb sideband
SC suppressed carrier
sprkr speaker

316 Chapter 4 : Section 10 Issued November 2006

SSB single sideband

sw switch
t time
T transformer
temp temperature
Tr transistor or field effect transistor
Tx transmitter
USB upper sideband
V voltage
Vc capacitor voltage
Vcc supply voltage
VCO voltage controlled oscillator
Vd voltage at drain
Vdd supply voltage
Vg voltage at gate
VHF very high frequency
XF frequency multipliers
X crystal

Issued November 2006 Chapter 4 : Section 10 317

NOTES:

318 Chapter 4 : Section 10 Issued November 2006

Beatty NBS137
No ratings yet
Beatty NBS137
323 pages
Unit 1 Homework SOLUTIONS - V2
100% (1)
Unit 1 Homework SOLUTIONS - V2
14 pages
BFP740 SPICE v3
No ratings yet
BFP740 SPICE v3
2 pages
Radar Frequency Bands Band Name Frequency Range Wavelength Range Notes
No ratings yet
Radar Frequency Bands Band Name Frequency Range Wavelength Range Notes
4 pages
Spanish Army Perspective: BG Enrique Silvela
No ratings yet
Spanish Army Perspective: BG Enrique Silvela
9 pages
Radar Data and Relationships: Radar Bands Linear DB Scale
No ratings yet
Radar Data and Relationships: Radar Bands Linear DB Scale
2 pages
Radio Frequency (RF) Data Communications
No ratings yet
Radio Frequency (RF) Data Communications
9 pages
Scattering Parameters 1
100% (1)
Scattering Parameters 1
5 pages
Quectel MC60 at Commands Manual V1.0
100% (1)
Quectel MC60 at Commands Manual V1.0
256 pages
725 Intro-F12
No ratings yet
725 Intro-F12
98 pages
Analysis of Micro-Doppler Signatures
No ratings yet
Analysis of Micro-Doppler Signatures
6 pages
Public NJFA Nato 02-2017
No ratings yet
Public NJFA Nato 02-2017
28 pages
Antennas Patterns
No ratings yet
Antennas Patterns
17 pages
Microwave Filter Cross - Coupling
100% (1)
Microwave Filter Cross - Coupling
12 pages
Speed Flex
No ratings yet
Speed Flex
10 pages
VNA Measurement Guide PDF
No ratings yet
VNA Measurement Guide PDF
256 pages
Essentials of Radio Wave Propagation - Christopher Haslett - The Cambridge Wireless Essentials Series, 1, 2008 - Cambridge University Press - 9780511371127 - Anna's Archive
No ratings yet
Essentials of Radio Wave Propagation - Christopher Haslett - The Cambridge Wireless Essentials Series, 1, 2008 - Cambridge University Press - 9780511371127 - Anna's Archive
215 pages
SWR - The Persistent Myth: Fielding Zs5Jf
No ratings yet
SWR - The Persistent Myth: Fielding Zs5Jf
66 pages
Communications Receivers DSP Software Radios and Design PDF
No ratings yet
Communications Receivers DSP Software Radios and Design PDF
2 pages
Smith Chart
100% (2)
Smith Chart
47 pages
FMCW Principe
No ratings yet
FMCW Principe
21 pages
(1967) Digital Pulse Compression Radar Receiver
No ratings yet
(1967) Digital Pulse Compression Radar Receiver
9 pages
ICom IDAS
No ratings yet
ICom IDAS
12 pages
Review of Antennas
No ratings yet
Review of Antennas
11 pages
HFIC Chapter 7 Low-Noise Amplifier Design
No ratings yet
HFIC Chapter 7 Low-Noise Amplifier Design
50 pages
(MWJ0306) RF Test Fixture Basics
100% (1)
(MWJ0306) RF Test Fixture Basics
6 pages
Frequency Bands
No ratings yet
Frequency Bands
6 pages
Enhancement of Microwave Absorption Using Frequency Selective Surface With Electronic Waste
No ratings yet
Enhancement of Microwave Absorption Using Frequency Selective Surface With Electronic Waste
37 pages
Exam 2016 Radar
50% (2)
Exam 2016 Radar
13 pages
FMCW Radar Is A Special Type of Radar Sensor Which Radiates: Frequency-Modulated Continuous-Wave Radar
No ratings yet
FMCW Radar Is A Special Type of Radar Sensor Which Radiates: Frequency-Modulated Continuous-Wave Radar
8 pages
Quad - Channel - Software - Defined - Receiver For Passive Radar
No ratings yet
Quad - Channel - Software - Defined - Receiver For Passive Radar
12 pages
Analog To Digital Conversion
No ratings yet
Analog To Digital Conversion
37 pages
Group Report Arduino Radar Shield
No ratings yet
Group Report Arduino Radar Shield
12 pages
EITN90 Radar and Remote Sensing Lecture 4: Characteristics of Clutter
No ratings yet
EITN90 Radar and Remote Sensing Lecture 4: Characteristics of Clutter
56 pages
Keysight
No ratings yet
Keysight
94 pages
Understanding The Signal Structure in DVB-T Signals For Passive Radar Detection
No ratings yet
Understanding The Signal Structure in DVB-T Signals For Passive Radar Detection
6 pages
EITN90 Radar and Remote Sensing Lecture 5: Target Reflectivity
No ratings yet
EITN90 Radar and Remote Sensing Lecture 5: Target Reflectivity
61 pages
Introduction To Antenna Test Ranges Measurements Instrumentation
100% (1)
Introduction To Antenna Test Ranges Measurements Instrumentation
13 pages
Microwave FMCW Radar
100% (3)
Microwave FMCW Radar
48 pages
Two-Ray Ground-Reflection Model
No ratings yet
Two-Ray Ground-Reflection Model
5 pages
F33 Service Manual
No ratings yet
F33 Service Manual
34 pages
Guide to RF Coaxial Connectors and Cables
No ratings yet
Guide to RF Coaxial Connectors and Cables
20 pages
Modelling RF Power Amplifiers With DPD Using Matlab White Paper PDF
No ratings yet
Modelling RF Power Amplifiers With DPD Using Matlab White Paper PDF
31 pages
Surveillance Mode S European Mode S Functional Specification 20050509
100% (1)
Surveillance Mode S European Mode S Functional Specification 20050509
212 pages
Antenna Lecture 2, 3, 4
No ratings yet
Antenna Lecture 2, 3, 4
101 pages
Radar 9 MIMO Radar
No ratings yet
Radar 9 MIMO Radar
39 pages
Linear Prediction Coding Vocoders: Institute of Space Technology Islamabad
No ratings yet
Linear Prediction Coding Vocoders: Institute of Space Technology Islamabad
15 pages
ACP194
No ratings yet
ACP194
32 pages
Terrestrial Propagation
No ratings yet
Terrestrial Propagation
30 pages
An Introduction To Passive Radar - 2017
No ratings yet
An Introduction To Passive Radar - 2017
110 pages
Frequency Nomenclature
No ratings yet
Frequency Nomenclature
1 page
Field Strength Measurement: by Surender Singh, Director (Engg.) Staff Training Institute (Technical)
100% (1)
Field Strength Measurement: by Surender Singh, Director (Engg.) Staff Training Institute (Technical)
19 pages
Antenna and Wave Propagation
100% (1)
Antenna and Wave Propagation
73 pages
Microwave & RF April 2013
100% (1)
Microwave & RF April 2013
136 pages
RF Blockset Getting Started Guide
No ratings yet
RF Blockset Getting Started Guide
92 pages
Basics of Radio Wave Propagation
No ratings yet
Basics of Radio Wave Propagation
14 pages
Upkar Radar Engineering
No ratings yet
Upkar Radar Engineering
9 pages
ECE 5233 Satellite Communications: Prepared By: Dr. Ivica Kostanic
No ratings yet
ECE 5233 Satellite Communications: Prepared By: Dr. Ivica Kostanic
11 pages
Voice over LTE Standard Requirements
From Everand
Voice over LTE Standard Requirements
Gerardus Blokdyk
3/5 (1)
Aerials & Radio Frequency Propagation
No ratings yet
Aerials & Radio Frequency Propagation
30 pages
Introduction 2011
No ratings yet
Introduction 2011
13 pages
EM Wave Propagation
No ratings yet
EM Wave Propagation
7 pages
3.electromagnetic Theory NET-JRF VKS
No ratings yet
3.electromagnetic Theory NET-JRF VKS
31 pages
Inspiring Analog System Design and Innovation at Entry-Level
No ratings yet
Inspiring Analog System Design and Innovation at Entry-Level
22 pages
ST3 6kVA 10kVA UPS Service Manual
100% (1)
ST3 6kVA 10kVA UPS Service Manual
15 pages
10A Rectifier Features
No ratings yet
10A Rectifier Features
2 pages
The Ultimate Guide To GaN
No ratings yet
The Ultimate Guide To GaN
35 pages
Arc Welding (SMAW)
100% (1)
Arc Welding (SMAW)
89 pages
For New Designs: SIM6827M
No ratings yet
For New Designs: SIM6827M
52 pages
Car Stereo
No ratings yet
Car Stereo
1 page
AN-1252 Application Note: How To Configure The
No ratings yet
AN-1252 Application Note: How To Configure The
12 pages
HV Reference Manual V2 May 2011
100% (1)
HV Reference Manual V2 May 2011
101 pages
Application of A Partial Element Equivalent Circuit Method To Lightning Surge Analyses
No ratings yet
Application of A Partial Element Equivalent Circuit Method To Lightning Surge Analyses
4 pages
13-04-2025_Sr.Super60_Elite, Target & LIIT-BTs_EAPCET_EGT-02_ Second Year Syllabus_Q.PAPER
No ratings yet
13-04-2025_Sr.Super60_Elite, Target & LIIT-BTs_EAPCET_EGT-02_ Second Year Syllabus_Q.PAPER
21 pages
Energy Management System in HEV Using PI Controller
No ratings yet
Energy Management System in HEV Using PI Controller
5 pages
Connection Cabinets: Os1, Os2, Os3, Os4, Os5, Os6 (With Extension For Mechanical Interlock)
No ratings yet
Connection Cabinets: Os1, Os2, Os3, Os4, Os5, Os6 (With Extension For Mechanical Interlock)
3 pages
Observations and Calcula Tions: Temperature (K) Voltage, V (MV) Current, I (Ma) Resistivity (Ohm CM)
No ratings yet
Observations and Calcula Tions: Temperature (K) Voltage, V (MV) Current, I (Ma) Resistivity (Ohm CM)
3 pages
Alpine MRV-F900 PDF
No ratings yet
Alpine MRV-F900 PDF
45 pages
Product
No ratings yet
Product
25 pages
Lab 8
No ratings yet
Lab 8
14 pages
SEE - IEE-Model QP 2
No ratings yet
SEE - IEE-Model QP 2
3 pages
Facts PPT Seminar
100% (4)
Facts PPT Seminar
22 pages
Electrostatic Microplastic Removal
No ratings yet
Electrostatic Microplastic Removal
12 pages
Bi 100 00 00 Ic SPC 0005
No ratings yet
Bi 100 00 00 Ic SPC 0005
5 pages
PD810 Series
No ratings yet
PD810 Series
3 pages
2N2218 2N2219
No ratings yet
2N2218 2N2219
5 pages
7.3 6T SRAM Cell: 7.4 Ring Oscillator 7. Case Studies 7.2 CMOS Inverter
No ratings yet
7.3 6T SRAM Cell: 7.4 Ring Oscillator 7. Case Studies 7.2 CMOS Inverter
4 pages
IPU43K4-B3A21 PSU Specification
No ratings yet
IPU43K4-B3A21 PSU Specification
2 pages
NS2000 Manual
No ratings yet
NS2000 Manual
16 pages
Compound Manual
No ratings yet
Compound Manual
28 pages