A COMMON CORE BOOKv>

-i

XJOmm

THE NEW MODEL ILLUSTRATED COURSE

OF ELEMENTARY TECHNICIAN TRAINING

basic

COMMON-CORE

By VAN VALKENBURGH, NOOGER & NEVILLE, INCAdapted for British and Commonwealth usage

by a special Electronics Training Investigation Team of .

the ROYAL ELECTRICAL & MECHANICAL ENGINEERS

THE TECHNICAL PRESS

BASIC ELECTRONICS

Part 3

A Course of Training Developed forTHE UNITED STATES NAVY

by the New York firm ofManagement Consultants and Graphiological Engineers

VAN VALKENBURGH, NOOGER & NEVILLE, INC.

Adapted to British and Commonwealth Usageby a Special Electronics Training Investigation Team of

the Royal Electrical & Mechanical Engineers

LONDON

THE TECHNICAL PRESS, LTDNEW YORK

THE BROLET PRESS

British and Commonwealth Edition first published 1959Second Edition revised 1961

Reprinted with Amendments 1962

©Copyright 1959, 1961, 1962 by

VAN VALKENBURGH, NOOGER & NEVILLE, INC.New York, U.S.A.

All rights reserved

American Edition first published 1955

©Copyright 1955 by

VAN VALKENBURGH, NOOGER & NEVILLE, INC.New York, U.S.A.

U.S. Library of Congress Catalog Card No. 55-6984

All rights reserved

The words "COMMON-CORE", with device and without device,are trade-marks of the Copyright owners

Spanish-Language Edition published 1958

Portuguese-Language Edition published 1960

Dutch Edition published 1960

Made andprinted by offset in Great Britain byWilliam Clowes and Sons, Limited, London and Beccles

PREFACE

In thesesix volumes on BASIC ELECTRONICS and the five which have pre-

ceded them on BASIC ELECTRICITY, there lies the heart of the uniqueCOMMON-CORE Series of Illustrated Training Manuals developed some years agoat the request of the United States Navy by a distinguished New York firm oftechnical education consultants and graphiological engineers,

VAN VALKENBURGH, NOOGER & NEVILLE, INC.Carefully planned and programmed for elementary technician training, brilliantlysimplified, and radically new, the COMMON-CORE Manuals soon became stan-dard texts in U.S. Navy Training Schools. More than 100,000 men have used themas an essential part of their training to technician level in 14 different Navy trades,and their average training time has been cut by half.

"The educational-engineering methodology employed in developing the originalCOMMON-CORE Program", the Authors have written, "was the culmination ofextensive research and practical experience with thousands of students. This

methodology can be described, briefly, as follows:

"1) Job analysis was used to identify the subject-matter (or core of knowledge)

required to achieve the necessary level of job performance;

"2) This subject-matter, once identified, was broken down into a learningsequence of small, easily-assimilated steps;

"3) These steps (or frames) were then presented in a format comprising the

single page of a book, each being described both visually and verbally.

"The completed materials were proved out on individuals of the appropriateaudience level; and then tried out on a group basis in selected schools.

"Although termed (for popular reference) a New-Model Way of Learning, thesewidely-acclaimed training courses owe their success and effectiveness in manycountries around the world to the fact that they were among the first programedinstructional materials ever produced."

Negotiations for a British edition of the COMMON-CORE Manuals were openedwith the American Authors; and while these were in progress, word reached theBritish publishers that there had recently been set up, under command of TrainingHeadquarters, Royal Electrical and Mechanical Engineers, at Arborfield inBerkshire, a special "Electronics Training Investigation Team" whose task was todevise solutions for some of the training problems which would face the British Armywhen National Service ended, and when the Army's increasingly elaborate electricaland electronic gear would have to be manned and serviced by recruits entering theArmy with none of the technical knowledge which many National Servicemen hadhitherto brought with them into the Forces.

It seemed possible that most of the REME requirements for a new-style, yettechnically sound, instructional approach could be met by a suitably edited British

version of the VVN&N Manuals. A visit to Arborfield was accordingly arranged;and after a careful evaluation of their merits and potential suitability had been made,War Office consent was secured to a proposal that the work of adapting text andillustrations to British notation and terminology should be undertaken by the

Electronics Team at Arborfield.

Later on, when practical experience with the Manuals in classroom use had beenacquired by a wide variety of instructors, there were received a large number ofsuggestions for improvement, clarification or development of the first-edition text

—

particularly from the Royal Air Force and from the REME Battalion responsible fortraining technicians in Radar and Guided Missiles. All these criticisms were col-lated and weighed by the Electronics Team at Arborfield; and much painstakingrevision of text and illustrations was then undertaken by the Team, working in closecollaboration with the editorial staff of the Publishers. An entirely new Section onthe Cathode Ray Tube was added in its appropriate position in the text towards theend of Part 5.

Despite their Services background, the Manuals have proved entirely suitable

for civilian use. Their purpose, however, is limited to the training of technicians

at the operator level, NOT of qualified engineers. They aim to turn out men capableof operating, maintaining, and carrying out repairs to the equipment described

—

NOT men capable of inventing or improving it.To the intending electrical engineer, however, the Manuals can be of value in the

early stages of his training by giving him an easy-to-follow overall view of his

subject, sound and accurate within its limits—a framework into which he canconfidently fit the more detailed and advanced work he must undertake later on.

No approach to the teaching of electricity and electronics can afford to be whollynon-mathematical. But every effort has been made throughout the Series to restrict

the use of arithmetic, and of elementary geometry and trigonometry, to cases where a

mathematical approach to the problem is either unavoidable or actively helpful to

the student; and nowhere is anything but the most elementary knowledge of basic

mathematical convention assumed without careful prior explanation of the point

involved.

Despite such reasonable concession to the ordinary man's dislike of figures,

however, there has been in these Manuals no shirking of essentials, even when they

are difficult; and students with higher qualifications and educational background

find nothing in the Manuals to irritate. They merely pass on to the next subject

quicker than the rest.

It has been the aim, in short, to present in these Manuals a unique simplification

of an ordinarily complex set of subjects; and to ensure that in them first things shall

come first—and only the essentials shall come anywhere.

August, 1962

TABLE OF CONTENTS

Section Page

1 Video Amplifiers 3.2

2 R.F. Amplifiers 3.13

3 Tuned Circuits 3.17

4 The Single-stage R.F. Amplifier 3.26

5 The Two-stage R.F. Amplifier 3.32

6 Oscillators 3.42

7 The Armstrong, the Hartley, and the Colpitts Oscillators 3.54

8 The "Tuned-Anode Tuned-Grid", and Crystal Oscillators 3.65

9 The Electron-coupled Oscillator 3.74

10 Other Oscillator Circuits 3.78

11 General Review of Oscillators 3.89

12 Introduction to Transmitters 3.92

Index 3.93

This Course in

BASIC ELECTRONICS

comprises 6 Parts

This is PART 3

It is preceded by a Course in

BASIC ELECTRICITY

comprising 5 Parts

all uniform with this volume.

Part 1 explained the General Principles of Electricity.

Part 2 described and discussed D.C. and D.C. Circuits.

Parts 3 and 4 described and discussed A.C. and A.C. Circuits.

Part 5 described and discussed A.C. and D.C. Machines.

tf&m/UdjieM

3.1

3.2 § I : VIDEO AMPLIFIERS

Introduction to Video Amplifiers

Video amplifiers are very similar in principle to the i?C-coupled audio amplifiers

you have already studied in Part 2 of Basic Electronics.

One very important difference between them, however, is this. Video amplifiersare designed so as to be able to amplify without distortion a number of odd-looking

waveforms which an ordinary audio amplifier would distort.

Some of the waveforms that video amplifiers are required to handle are called

"pulses" or "square waves."

A French mathematician, Fourier, showed that such waveforms can be consideredto be the sum of many sine waves of different frequencies.Some of these sine waves which the mathematicians say can be found in a square

wave are many times higher in frequency than the square wave itself. Therefore,the amplifier needed to amplify the square wave without distortion must have a fre-

quency range which covers not only the frequency of the square wave (fundamental)itself, but also the frequencies (harmonics) of all the other sine waves which make upthe square wave as well.

However, all this about "harmonic frequencies" need not make you afraid thatyou will need an advanced course in mathematics before you can understand videoamplifiers. For the mathematician can go right on defining a video amplifier as"a wide-band amplifier capable of amplifying signal in the frequency range of from

30 cycles all the way up to several million cycles (megacycles) per second." Youcan look at it in a much simpler, yet just as accurate, way—as follows.A video amplifier must be able to amplify signals such as square waves without

adding distortion. Therefore, if you put a square wave input into a video amplifier,

you ought to get a square wave output out of it—and if you don't, something iswrong with the amplifier. It is just as simple as that

!

§1] 3.3

Introduction to Video Amplifiers (continued)

To many of you, the word "video" is synonymous with television; it will thereforecome as no surprise to learn that video amplifiers are indeed used in televisionreceivers.

The picture is sent out by the station in the form of r.f. signals amplitude-^modulated by a series of electrical pulses which represent the dark and light portionsof the picture. These signals are de-modulated in your receiver, amplified by thevideo amplifier, and applied to the picture tube so as to duplicate the picture sentout by the station. In this process, the quality of the picture received depends inpart on the quality of the video amplifier. If this amplifier distorts the pulses, thepicture will lack the sharpness and detail it would otherwise have had.

Another important application of video amplifiers is in the oscilloscope. Theseinstruments need amplifiers which can amplify an input signal without introducingany distortion in the process. Even if the input signal is itself a distorted sinewave, or a square wave, that is how it must appear on the 'scope screen if the 'scopeis doing its job properly.

Because video amplifiers are capable of amplifying almost any waveform withoutdistortion, they are much used in oscilloscopes.

Another very important application of video amplifiers is in radar equipment, asyou will find out when you come to study the several Parts of Basic Radar.

Every radar, whatever the use to which it is to be put, contains a video amplifier.Radar echoes are sharp pulses; and the amplifier must be able to preserve the shapeof these pulses if the radar operator is to obtain accurate information about thetarget which is sending back the echo.

Video amplifiers are used whenever pulses or square waves have to be amplifiedwithout distortion. No other amplifier comes close enough to meeting these require-ments in oscilloscopes, in television, in photo facsimile equipment, or in radar.

3.4 B

Distortion of Square Waves

One reason why a square wave is distorted in an audio amplifier is because the

audio amplifier has a poor low-frequency response. This shows up in the output

as an unfaithful reproduction of the flat portion of the square wave.

frequency response

Distortedoutput dueto poorer

low frequencyresponse

This is what happens. During the flat portion of the square wave, between

time (a) and time (b), the grid voltage, the anode current and the anode voltage all

remain constant. Then at time (b), the grid voltage drops suddenly to a new value,

and the anode voltage rises just as suddenly to its new value.

When this happens, a current will flow through R-2 in the direction shown in

the diagram below, thus causing an output voltage to appear across this resistor. The

initial amount of current flowing in R-2 depends, of course, on the new value of the

anode voltage, and on the value of R-2. This current will charge up C-l, and

thereby change the voltage across the capacitor. As the voltage across C-l in-

creases, the current through R-2 decreases, resulting in the distorted output shown

above.

HT +

OUTPUTO

INPUT

If C-l has a small value, the voltage across it will change more rapidly; and the

output will be more distorted. Distortion will also increase if R-2 is decreased;

for this, too, will result in a more rapid change in voltage across the coupling

capacitor. In order to improve the low-frequency response of an RC-coupled

amplifier, therefore, and thereby to reduce the distortion which occurs in the flat

portion of the square wave, R-2 and C-l should be made as large as possible.

§1] 3.5

Distortion of Square Waves (continued)

Even with good low-frequency response, however, there might still be anothercause for distortion—poor high-frequency response, which appears in the output asan unfaithful reproduction of the steep portions of the square wave.

The effective stray capacitance, C-2, is the cause of this distortion. A capacitorcannot change its voltage instantaneously; and, since C-2 is directly across the out-put, the output voltage Cannot change instantaneously either.

The stray capacitance, C-2, charges and discharges through R-l and R-2, whichare effectively in parallel, since the reactance of the coupling capacitor is negligible.

If the parallel combination of R-l and R-2 allows C-2 to charge and dischargequickly, the output will show little or no distortion. But if R-l and R-2 are largeresistances, C-2 will require a relatively long time to charge and discharge, and thesteep sides of the square wave will not be perfectly vertical.The larger these resistances become, the worse the distortion becomes. When this

becomes very severe, C-2 will never be able to charge and discharge enough toreach the flat portion of the square wave, and the output will resemble the triangu-larly shaped wave shown above.

- HT+c-i

wtnt-»-o

1-2 »k.

C-2

Stray capacitance

due to capacity of

valve and wiring

If you recall the discussion about improving the high-frequency response of audioamplifiers, you will remember that there are two different ways of doing this. Thefirst is to reduce the stray capacitance C-2 by using special pentode valves withvery low values of input and output capacitance, and by using special wiring tech-niques to reduce the stray capacitance between the wiring and earth.The second way is to reduce the time it takes C-2 to charge and discharge. This

is done by using lower values of R-l and R-2.You know, however, that reducing R-2 would harm the low-frequency response;

so this is not done. As for reducing R-l instead, it is true that this reduces thegain of the stage; but this disadvantage is overcome in video amplifiers by addingmore stages, each with low gain but good frequency response.

Special valves are used, such as the 6AC7, 6SH7 and the 6AG7. These valvesare designed for high gains, and for low input and output capacitances; they aretherefore ideally suitable for video amplifiers.

3.6 [§l

Compensating Networks—High-frequency CompensationOne common way of improving wave shapes in video amplifiers is to decrease

the effect of the things which cause distortion. You saw on the last page one methodof doing this—the adjustment of the values of R and C in the coupling networkso as to reduce distortion. Another way is to introduce into the circuit, deliberately,other distortions of a nature and an amplitude exactly opposite to those of the dis-

tortions which are causing the trouble, and so to cancel the latter out.

c

__

_ _ _ _ t

r

y¥a

'

_c_ —— --

Ideal Distortion due Back EMF"""output "" toC-2alone " across L

/Distortion Output-- due to L --duo to L -

alone and C-2

To improve the steepness of the vertical portion of the square wave, an inductor(L) is placed in series with the anode load resistor (R-l). The back e.m.f. set up in

this inductor every time the anode current changes suddenly will be in such a direc-

tion as to cause a greatly increased

change in the anode voltage.

When the grid voltage is swingingpositive, the anode current increases.

The anode voltage decreases; and atthe same instant, a back e.m.f. is set upacross L. This back e.m.f. tends to

oppose the increase of current, and has

the direction shown at (a) in .the dia-gram above. This negative back e.m.f.lowers the anode voltage below the

value to which it would fall under the

action of the grid alone, thus causing

a peak to appear on the square wave.

When the grid voltage swings negative, the anode current decreases, and theback e.m.f. across L is in the opposite direction. This then adds to the anodevoltage, causing another peak to appear at (b).

Now consider the effect of these peaks on the distortion caused by the effectivestray capacitance. These higher values of voltage—higher because of the peaks

—

will cause C-2 to charge and discharge faster than it would if the coil L wereomitted.

It follows that, if the proper value of L is used, the final output can be made analmost perfect square wave, as you will see from the waveforms illustrated at the

top of the page.

Because of its effect, this inductor L is called a "compensating coil."

§1] 3.7

Compensating Networks—Low-frequency CompensationThe most common compensating circuit used for correcting low-frequency dis-

tortion is one which resembles a decoupling network. Like the compensating coil,

this circuit introduces a distortion which is opposite to, and so counter-balances,

the distortion caused by poor low-frequency response. The distortion for whichthis circuit is designed to compensate is the distortion of the flat region of the square

wave caused by the voltage changes across the coupling capacitor.

(\) 00

2I bc

2 bc

Ideal Distortedoutput output duo

to C-l a Ion

Voltage^npoint

Distortedoutput due

to C-3 alone

Output d:to C-land C

HT+At (a) the anode current increases as the

grid goes positive, and the voltage-dropsacross R-l and R-3 tend to increase. Thevoltage at point 3 (see circuit diagram) can-

not fall immediately the current through the

valve changes, because C-3 must discharge.The change in voltage at point 3 is illustratedat (iii) above.

The voltage at point 1 is also affected bythe discharge of C-3, and in the absence ofC-l it would vary as illustrated at (iv) above.However, as you saw on page 3.4, the

effect of C-l is to cause the voltage at point 1

to rise between (a) and (b), as illustrated at(ii) above. So the effects of C-l and C-3 onthe anode voltage cancel each other out, andthe output voltage (point 2) becomes the perfect square wave shown at (v) above.At (b), on the other hand, the current suddenly decreases; and the voltage at

point 1 rises as the capacitor C-3 charges up. Thus between time (b) and time (c),the effects of C-3 and C-l are again in opposition, and the output once moreapproximates to the ideal.

What happens can be put like this. Without the compensating network, thevoltage across the coupling capacitor changes; and this change, subtracted from thesteady anode voltage, leaves a distorted voltage across R-2. With the compensatingnetwork, the voltage across the coupling capacitor still changes; but these changes

subtract, not from a steady anode voltage, but from an anode voltage which isitself changing in the opposite direction, because of the charge and discharge of C-3.

In a properly designed circuit, the voltage across R-2 should be an almost perfectsquare wave.

Low -FrequencyCompensating Network

3.8[§

Improving Frequency Response—Negative FeedbackLet us now take another form of compensating circuit, and see how that affects

distortion appearing in the output.

Suppose that a perfect square wave (waveform 1) were to be connected to thegrid of the amplifier illustrated on page 3.4, but that the output (the voltage acrossR-2) came out distorted as shown in waveform 2.

Let us see what would happen if we tried to distort the input waveform in such away as to compensate for the distortion introduced by the amplifier. We could dothis by adding a suitable signal, such as that shown in waveform 3, to the original

input signal. The modified input signal would then look like waveform 4.

If the added signal were exactly right, the effect of the amplifier distortion on

waveform 4 would be to produce exactly the output signal we want (waveform 5).

But the added signal cannot be exactly right, because (as you will see on the next

page) it is derived from the output signal. The distortion can, however, be reducedby feeding in this signal from the output; so that waveform 2 might, for example, be

transformed into waveform 6. The distortion has not been eliminated, but it has

been greatly reduced.

Note that the amplitude has also been reduced. This is a misfortune which

cannot be avoided if the unwanted distortion is to be reduced by this "negative

feedback" technique.

§1] 3.9

Improving Frequency Response—Negative Feedback (continued)The negative-feedback method of decreasing amplifier distortion described on the

last page is also known as "degeneration." One of the simplest and most widelyused ways to obtain it is to have a cathode bias resistor (R-4, below) un-bypassed.

When this is done, the cathode voltage will not be steady d.c, but will vary asthe current varies. When the grid goes positive (or less negative), the cathodecurrent increases and the cathode voltage goes positive. This cathode voltage

decreases the grid-to-cathode signal, and so lowers the gain of the stage.

But, if distortion is present in the circuit, the cathode current will not be of the

same wave shape as the grid voltage. The cathode voltage will contain this dis-tortion; and the difference between the grid and the cathode voltages will contain

just the opposite distortion. The result will be a reduction of distortion in the out-put, accompanied by a reduction in gain.

FEEDBACK THROUGHo UNDECOUP1ZD CATHODE RESISTOR

::::!:::i::::::::«:i;::iiii:R:

fjjjjjjijpPiiiigij

:H»:::::::g"H:H:H:S»:n:HH:n&:H!U:»SH£&^

Another way of obtaining negative feedback is to use a voltage divider consistingof R-6, the grid resistor, and R-5. This voltage divider is connected across the

output so that part of the output voltage appears across R-6. In addition, the input

signal appears across R-6.

Since the output and input signals are 180 degrees out of phase, the resultant input

signal to the grid will be the difference between the input to the stage and that part

of the output signal which is fed back. This, again, will counteract part of the

distortion in the output (and also reduce the gain).

In either of these two circuits, it is possible to adjust the amount of signal fed back.If the un-bypassed cathode resistor, R-4, is increased, or (in the other circuit) if R-5 is

decreased, the feedback will be increased. This will still further lower the gain of

the stage, but will make the output signal more closely resemble the input signal.

3.10 [§l

Improving Frequency Response—Negative Feedback (continued)In the illustrations shown on page 3.8, you saw how negative feedback improves

the response of the amplifier to the flat portion of the square wave. Below, you seethe six waveforms which explain how it improves the steepness of the steep portionof the square wave.

As before, waveform 1 is the original input to the grid; waveform 2 is the signalwhich would appear in the output if no negative feedback were used; waveform 3is the part of the output signal which is fed back, and which appears between gridand cathode; waveform 4 is the resultant grid-to-cathode signal; waveform 5 is theoutput waveform desired; and waveform 6 is the actual output voltage—reduced inheight, but with much less distortion than there was in waveform 2.

Original

input

Original

output

Feed-back

Resultant

input

Ideal

output

signal

Outputsignal

achieved

The important thing to remember is that the signal which is fed back containsthe distortion which exists in the output. When this signal is correctly combinedwith the original grid signal, the resulting anode signal has less distortion than it hadoriginally.

In this respect, negative feedback does, almost the same thing as a compensatingnetwork. The important differences are these:

1. Negative feedback will reduce most types of distortion, while a compensatingnetwork will work only for the type of distortion it is designed to eliminate.

2. Negative feedback will always result in decreased gain.

Because negative feedback will reduce most types of distortion, it is widely usedin audio amplifiers to ensure good frequency response with comparatively un-distorted output.

In addition to its function as a distortion-reducer, negative feedback has the

further important advantage of providing means whereby an amplifier can be maderelatively immune from the undesirable effects of changes in supply-voltages, incomponent values, and in valve characteristics.

§1]

REVIEW of Video Amplifiers

Video Amplifier amplifies pulses, triangu-

lar or square waves, without distortion;

whereas an audio amplifier distorts these

waveforms because of its poor high- and

low-frequency response.

3.11

hj^jVideo

Amplifier

*$Low-frequency Response can be improved

by increasing the capacity of the coupling

capacitor, and by adding a low-frequency

compensating network.

Low-frequency Compensating Network

develops a varying voltage which, when

added to the square wave input voltage,

counteracts the distortion of the flat portion

of the square wave caused by the coupling

capacitor.

High-frequency Response can be improved

by reducing the value of the anode load

resistor, and by adding a compensating coil

in the anode lead.

Compensating Coil counteracts the effect

of stray capacitance, which tends to round

off the leading edge of the square wave.

OUTPUT

Lowfrequencyresponse

POOR GOOD

wvwv-*-—+-***+S OUTPUT

INPUT

POOR

OUTPUT

Highfrequencyresponse

GOOD

INPUT 1 VO/

p.II

r-y0ojJv-

3.12

REVIEW of Video Amplifiers (continued)[§'

Negative Feedback is a methodof overcoming any type of square

wave distortion by returning part ofthe output as a grid signal. Resul-

tant output contains very little

distortion.

>riginal Original Feed- Resultant Idealinput output back input output

signal

Outputsignal

achieved

Un-Bypassed Cathode Resistor

provides negative feedback by for-

cing the cathode voltage to vary as

the current varies, introducing dis-

tortion opposite to that already

present, and so reducing it.

Voltage Divider Feedback intro-

duces part of the output voltage

across the grid, but 180 degrees out

of phase; so that distortion is intro-

duced "in reverse", and thus

reduced.

NEGATIVE FEEDBACK

Video Amplifier Circuit maycontain all or several of these

methods of reducing distortion.

§2: R.F. AMPLIFIERS 3.13

Introduction

The best way to begin learning about r.f. amplifiers is to review what you know

about amplifiers and amplification generally.

Suppose you need 10 volts to drive a pair of headphones or a loudspeaker, and

that the signal voltage available is only 01 volts, which is too small to be used.

This signal is fed into the grid of an amplifier valve which builds it up to 1 volt.

Then you feed the 1 volt into another amplifier valve and it in turn raises the output

to 10 volts.

An alternative, of course, would be to use one amplifier valve with a gain of 100,

and so build up the voltage to 10 volts in one step.

When two valves are used to do the job, it is called a two-stage amplifier. When

one is used, it is a single-stage amplifier. Some amplifiers use as many as five

stages to build up a voltage large enough to drive a piece of equipment.

Input

Signal

01 volt

Gains 10

Signal _

Gains 10

OutputAmplifierstage

Amplifierstage1 volt 10 volts

sssssssssss^

SINGLE STAGE Amplifa*

Input

Signal

Gains 100

Amplifierstage

Output

0-1 volt 10 volts

When an amplifier builds up the voltage 10 times, it is said to have a voltage gainof 10. The voltage gain is the number of times a stage, or a group of stages,

amplifies the signal.

3.14[§2

What an R.F. Amplifier DoesWhat is it, then, that makes an r.f. amplifier different from other types of

amplifiers?

You remember that:

1. Audio amplifiers amplify all frequencies from about 15 to 15,000 cycles persecond.

2. Video amplifiers amplify all frequencies from about 30 to 6,000,000 cycles persecond.

FREQUENCY RANGESOF AMPLIFIERS

RF Amplifiers

Video Amplifiers

llllllllllllllllllllllllllllllll

Audio Amplifiers j !

Illlllll 1iiiil 1 I1

'! ' |

i : : 1 i ;j

i 1 i i

1 100 10,000 1 1,000,000 1 100,000,000 1 10,000,000, 000 1

R.F. amplifiers are designed to amplify a narrow band of frequencies in the rangefrom about 30 kc/s to 30,000 Mc/s.The amplifier does not, however, amplify this entire frequency range at once. It

selects one small portion—the portion occupied by the radio signal sent out by onetransmitter—and amplifies that. For instance, the B.B.C. Home Service broadcastsat a frequency of 908 kilocycles, and is allotted a band whose limits are 4-5 kilo-cycles either side of 908. (Most standard broadcast stations are allotted a band4-5 kc/s either side of a centre frequency.) When you tune a broadcast receiver tothe B.B.C, you are in fact adjusting the r.f. amplifier so as to pick up signals in theband of frequencies 903-5 to 912-4 kc/s.

Exactly the same principle applies to short wave and to television stations. Forexample, a station at 10 Mc/s might have a bandwidth from 9-8 to 10-2 Mc/s. Aperfect r.f. amplifier would select that range of frequencies and reject all others.

Television channel 1 occupies the band from 41 to 46 Mc/s. When you tunea television set to channel 1, you are adjusting the r.f. amplifier to handle signals inthat band and to reject all others.

J 21

What an R.F. Amplifier Does (continued)

You already know that a signal sent out by a radio transmittertravels through

the air to reach your radio, sometimes for thousands ofmiles. Thus, though the

'transmitter itself may be putting out thousands of watts of power, when thesignal

reaches your receiver it may be very weak.

In practice, the signal coming into your receiver is usually inthe order of a few

millionths of a volt only, and it must be amplified many times (and have agood few

other things done to it as well!) before it will drive a loudspeakeror a set of ear-

phones.

AMPLIFICATIONWeakRF

AA/WW YSignal \/ Aerial

1—WAAAH RFAmplifierAmplified

To nextstage

RF Signal

It is possible to amplify a radio signal in several different ways. You may amplifyit at radio frequency as it comes from the aerial; or you may convert it to lower

radio frequencies, or even to audio frequency, and then amplify it. All these various

methods will be discussed later on, in Part 5 under "Radio Receivers."

The important thing to remember at the moment is that amplification is not the

only function of an r.f . amplifier. Another important thing it does is to pick out

the required signal from the multitude of other signals reaching the receiver aerial.

This process is called "tuning." When you "tune" a receiver or a transmitter,you are in fact changing the band of frequencies which the r.f. amplifier can handle.

\\ tSeveral RF signals TUNING

^ One AmplifiedTo nextstage

RF Signal

3.16 „ 2This Is What You Will Learn about R.F. Amplifiers

Before you learn more about r.f. amplifiers, you will need to have a brief reviewof resonance. Then, in addition to what you learnt about coils and capacitors inBasic Electricity, you will be shown how the resonant effect is used to tune an r.f.amplifier stage. The selectivity of tuned circuits, and what is known as "Q," willbe explained; and you will be shown the construction of aerial and r.f. coils actuallyin use.

After the section on resonant circuits, you will see why pentodes are nearly alwaysused as r.f. amplifiers. Then you will be shown actual r.f. amplifier circuits, illus-trating how you may connect the tuned circuits to the amplifier valve, and how thecorrect voltages are applied to the valve. You will learn about the parts used in atypical r.f. amplifier stage in a broadcast receiver; and you will find out how theuse of more than one r.f. amplifier stage affects selectivity.

Wl" IEARN

§3: TUNED CIRCUITS 3.17

What a Tuned Circuit Is

You will have gathered from this introduction that all r.f. amplifiershave two

important functions to perform:

1. They amplify the signal at radio frequencies.

2. They select one narrow band of frequencies and reject all others.

Amplification of the signal is accomplished by a valve, just as in audioand video

amplifiers. The job of selecting one narrow band of frequencies to be amplifiedis

performed in the r.f. amplifier by what is called its "tuned circuit."

The tuned circuit consists of a coil and one or more capacitors so connectedas to

form an LC circuit which is made to resonate, or is "tuned" to the desired frequency.

On the following pages you will see how tuned circuits in r.f. amplifiers work.

S3Revision of Series LC Circuits

In a radio receiver there are many signals of different frequencies coming into theaerial. The listener tunes the radio by adjusting the tuning capacitor This makes theaerial coil and capacitor resonate at the frequency of the desired station and soenables them to give optimum response to an incoming signal from that station

In order to understand exactly what this resonant effect is, let us briefly review seriesLC circuits, and then parallel LC circuits.You remember that a coil offers less opposition to low frequencies than it does to

high frequencies. A capacitor, on the other hand, offers less opposition to highfrequencies than to low ones. In either case, the opposition offered is called "re-actance".

High

REACTANCE

LowFREQUENCY

If, in the series LC circuit shown below, the signal generator is delivering a verylow frequency, the coil will offer little opposition to this, but the capacitor will offervery high opposition. Therefore very little current will flow, because the totalreactance of the circuit is high. If, on the other hand, the signal generator deliversa very high frequency, the coil will offer very high opposition; and the current willstill be low.

At some intermediate frequency, however, the reactance of the coil will equal thereactance of the capacitor. At this frequency (the resonant frequency), the im-pedance of the circuit will be minimum, and the current therefore maximum.

Below ResonantFrequency

Highreactance

Above ResonantFrequency

Lowreactance

At ResonantFrequency

Both reactances equal,cancel each other

Ov Signal aVVGenerator

Lowreactance

^=Hh

JvQftfl/- vOOQ *-

Highreactance

Current

You will remember from page 4.34 of Basic Electricity (Part 4) that the frequencyat which the reactance of coil and capacitor are equal is called the resonant frequency(/r) and that/r = l/ln^LC c/s, L being expressed in henrys and C in farads.

§ 3]3.19

Revision of Parallel LC CircuitsMost radio receivers employ parallel-resonant, rather than series-resonant, circuits

for tuning to different frequencies.

You saw on the last page that the reactance of coils and capacitors varies with thefrequency applied to them. But coils and capacitors have also another property which

is important in resonant circuits. A coil causes the current to lag behind the appliedvoltage by 90 degrees. A capacitor causes the current to lead the applied voltage by90 degrees.

VOLTAGE AND CURRENTWAVE FORMS ACROSS A

COIL

VOLTAGE AND CURRENTWAVE FORMS ACROSS A

CAPACITOR

Coil causescurrent lag — 90*

mCapacitor causescurrent lead — 90*

\LS

If you connect a coil in parallel with a capacitor, the current in the coil is 90 degrees

behind the applied voltage, and the current in the capacitor is 90 degrees ahead

of the applied voltage. In other words, the two currents are 180 degrees out of

phase. You remember from Basic Electricity that currents which are 180 degreesout of phase tend to cancel one another out. So, if the capacitor alone draws 3

amps and the coil alone draws 2 amps, the combination of the two will draw only

3 minus 2, or 1 amp.

CAPACITOR AND COILIN PARALLEL

3.20 [§3

Revision of Parallel LC Circuits (continued)Since the coil and capacitor are in parallel, the voltage across them is the same.

If you apply a voltage of the frequency at which the reactance of the coil equals thereactance of the capacitor, the current in the coil will be equal and opposite to thecurrent in the capacitor; and no current will flow in the external circuit—i.e., fromthe voltage source.

The frequency at which this occurs is called the resonant frequency, and it is at thisfrequency that the tuned circuit's impedance is greatest. (Remember again theformula for the resonant frequency:/r= l/27ryXC.)

*?tt *76eai

§3] 3.21

How the Resonant Circuit Selects StationsSo far you have learnt that a parallel tuned circuit has a very high impedance at

the resonant frequency, and a lower impedance at all other frequencies. If you

remember this, it will be easy to see how a parallel LC circuit selects stations.

PARALLEL TUNEDCIRCUIT—HIGHIMPEDANCE ONLYAT RESONANT

FREQUENCY

High

Impedanceof tunedcircuit

Resonantfrequency

Frequency High

In the circuit shown below, signals of different frequencies strike the aerial. Each

of them starts a current flowing in the primary of the aerial coil. Each of these

currents in the primary induces a voltage in the secondary.

A variable capacitor is in parallel with the secondary of the aerial coil. Aparallel LC circuit has a high impedance at its resonant frequency, and a lowerimpedance at all other frequencies. Therefore the induced voltage will tend to be

largest at the resonant frequency.

I

TUNING

'Tn /T\ >T\' * / \ ' 1/ 1 \ / 1 \ / 1 \/ \ / 1 \ / \1 ' v/ 1 \/ N

' 1 X X ' \

HILVERSUM MUNICH BBC^746 800 908

Frequency High

Changing Capacitor Sotting— Changes

One particular coil and one particular capacitor will resonate to one frequency

only. If you vary either the inductance or the capacitance of the tuned circuit, you

will change the resonant frequency. In the process of tuning, you change the

capacitance of the tuned circuit by using a variable capacitor.

When the resonant frequency of the LC circuit coincides with the frequency ofsome signal, you have tuned the r.f. amplifier to that signal. But no tuned circuit

is perfect. Signals whose frequencies are very close to the frequency of the wanted

signal will also evoke a response from the tuned circuit. This unwanted response

will give rise to what is termed "adjacent-channel interference."

[§33.22

"Q" and SelectivityYou will remember that, in both audio and video amplifiers, it is desirable to

have the amplification stage pass a wide range of frequencies. In r.f. amplifiers,on the other hand, we want the amplification stage to select only a narrow band offrequencies, and to reject the rest.

The narrower the band of frequencies passed by an amplifier, the greater is itsselectivity. Thus selectivity can be defined as the ability of an amplifier to dis-criminate between signals which lie close together in frequency.

POOR SELECTIVITY

^/f&&&>*!\!

ft

59>

/l 1/ 1 1/ 1 !

J \1 \

/ 746 800f ! 1

908 \1 \

Frequency High

746 800 908^—l I

Frequency High

The selectivity of an r.f. amplifier is determined by its tuned circuits. Themeasure of a tuned circuit's selectivity is proportional to what is known as its "Q."The Q of the tuned circuit can in practice be considered to be the Q of its coil.

This is equal to the reactance of the coil divided by its resistance, and is a measureof the efficiency (or "goodness") of the tuned circuit.

Q REACTANCE OF COILRESISTANCE OF COILHigh Q

Low Q

Sharp tuning ^^I^V^^^fl^^^^^H

Broad tuning H^BHtf

^g^^jl^^^^^jg^j^^i^^gj

§3] 3.23

How Tuning Capacitors Are ConstructedIn Basic Electricity you were shown the construction of the two types of capacitors,

fixed and variable. Variable capacitors are used in tuned circuits so that you can

vary their capacitance, and thus change the resonant frequency of the circuit.

Variable capacitors have one set of plates called rotors, which can be rotated in

and out of another set of fixed plates called stators. The dielectric is usually air.

As the rotor plates are rotated farther and farther out of the stators, so the capacity

of the unit decreases.

Most radio receivers with r.f. amplifiers employ more than one tuned circuit.

But each tuned circuit needs a variable capacitor. If you mounted each variable

capacitor separately, you would have to tune each one separately—which would beinconvenient.

Instead, you can mount the rotors of several identical variable capacitors on a

single shaft. This is called "ganging" them. When one rotor is turned, the othersall turn the same amount.

Ganged capacitors involve one big difficulty, however. Although each of the

ganged units is of the same size and has the same spacing, there are small

differences in capacity between the units. This is because it is impractical to manu-

facture any two things which are of exactly the same size.

To compensate for the differences in capacity, a small variable capacitor is con-

nected in parallel with each variable capacitor unit. These small compensating

capacitors can be adjusted separately until all the units have the same capacity at

the minimum-capacitance setting of the gang (which is when the rotor plates of the

capacitors are fully "out").

These compensating capacitors are called "trimmers."

£K £*TRIMMERS

3.24

How Tuning Coils Are ConstructedMany tuned r.f. coils are really transformers

and the secondary. The coils are wound onceramics or other suitable insulating material;

of powdered-iron or of a suitable non-ferrous

In order to prevent stray electric fields fromscreens are generally placed round the coils,

as they are called) alter the inductance of the

ments, such as the alignment process which

performed with the screening cans in position.

[§3

;, and have two windings—the primaryformers made of cardboard, bakelite,and have cores either of air, or mademetal.

affecting the action of r.f. coils, metal

These screens (or "screening cans,"coil. Therefore any receiver adjust-

will be described shortly, should be

§3]

REVIEW of Tuned Circuits

Tuning. Means selecting a signal at one

frequency and rejecting signals at all other

frequencies.

3.25

Frequency

R.F. Coils. Generally have primary and

secondary windings. The secondary is usu-

ally tuned. Most of them have screening

cans.

Tuning Capacitors. These are variable air

capacitors. Several air capacitors may beganged into one unit.

Trimmers. Are small capacitors placed in

parallel with each unit of a ganged capacitor.

Their function is to compensate for small

differences in capacity between the units.

Parallel Resonant Circuit. The tuning

capacitor is connected in parallel with the coil.

The combination has high impedance at the

resonant frequency, and low impedance at all

other frequencies. It builds up a high voltage

at the resonant frequency and a lower voltage

at all other frequencies. This type of circuit is

the one most often used to tune radio receivers.

3.26 §4: THE SINGLE-STAGE R.F. AMPLIFIERRemember that there are two requirements for an r.f. amplifier:

1. a means of selecting the signals to be amplified; and2. a means of amplifying these signals when selected.

We have already seen that tuned circuits can be used to select the signals foramplification. Let us now think what sort of a valve we shall need to- use in theamplifier.

The three valves which might in theory be used are the triode, the tetrode and

the pentode. Of these, the pentode is the most suitable in practice, because:

(a) it has a much higher amplification factor than has either the triode or tetrode;

(b) it suffers less from the effect of inter-electrode capacitance; and

(c) it does not suffer from the effect of secondary emission.

An r.f. amplifier generally has tuned circuits both in its input and its outputcircuits. The anode-to-grid capacitance of a triode provides a feedback path from

output to input which might cause the amplifier to generate its own signal, or to

"oscillate." This effect will be explained more thoroughly later on. At the moment

all you need know is that the triode is liable to oscillate, instead of amplifying; and

that it is therefore unsuitable as an r.f. amplifier unless certain precautions are taken

(see Part 4, page 28, "Neutralization").

The tetrode is less likely to oscillate, because of its reduced inter-electrode

capacitance. But it has a lower amplification factor than has the pentode. More-

over, it is always liable to be affected by the distortion caused by the effect of

secondary emission. So it, too, is generally ruled out.

§4]3.27

How the Coils and Capacitors are Connected to the Valve

You can connect coils and capacitors in two possible ways to make them tune to

a given frequency—in series or in parallel. Almost all r.f. amplifiers employ paralleltuned circuits. The most common method of arranging tuned circuits in an r.f.

amplifier, and the one you will use, is shown below.

:::::::!::::::::

!.!

USUAL COUPLING METHOD-RF AMPLIFIER

YYAerial /

Coil /RF

AmplifierValve

To nextstage

*»•-

pill!

However, this is not the only type of circuit which can be used to tune an r.f.

amplifier. You may occasionally find other circuit arrangements. Either the inputcircuit or the output circuit may be altered, and still produce a practical tuningarrangement. Here are some other ways.

DIFFERENT WAYS OFCONNECTING THE INPUT

LowImpedancePrimary

DIFFERENT WAYS OFCONNECTING THE OUTPUT

TunedPrimary andSecondary

+-

NoPrimary

CapacitiveCoupling

-I n

No InductiveCoupling

NoPrimary

»

5-ih

CapacitiveCoupling

Resistor orRF Choke

3.28 [§4

The Development of the R.F. Amplifier Circuit

You remember from the sections on valve theory how the valve of an amplifierworks. The signal to be amplified is fed into the grid, and the output of the valve istaken from a load impedance in the anode circuit.

You know how pentodes operate, and you know how tuned circuits operate.Now put them together, and this is what you get.

tf

THE BASICRF AMPLIFIER

/ 2GANG -7/ TOANOOe

/ VOLTAGEl SUPPLY ^•NEXTSTAGE

T-TRIMMER

There are, in fact, only a few differences between the above circuit and circuits

which are actually used in radio receivers.

First, separate power supplies are seldom used for the anode and screen voltages.

The anode voltage supply is generally lowered by a dropping resistor or a voltage

divider to supply voltage to the screen (see R-2 in diagram below). Then the screen

will need its own decoupling capacitor (C-6), because the voltage on the screen mustnot be allowed to vary up and down with changes in the signal.

Secondly, grid bias voltage is seldom taken from an outside source. Instead it

is generated by the valve itself, using a method which is familiar to you—cathodebias (R-l). The cathode bias resistor is by-passed so as to prevent negative

feedback (C-5).

Very often in multi-stage r.f. amplifiers a decoupling filter (C-7 and R-3) is

connected in the anode lead, in order to prevent one stage from interacting with

another. Indeed, it is good policy to use a decoupling filter even in a single-stage

r.f. amplifier.

ANACTUALCIRCUIT

§4]3.29

The Variable-Mu Pentode

If you look up the 6SK7 in a valve manual, you will find it is called a "variable-Mu"

pentode. In other places it is called a "remote cut-off" pentode. The two titles

describe an important feature of this valve.

To help you to understand this feature, let us compare an 7a- Vs characteristic ofthis yariable-Mu pentode, the 6SK7 (on page 3.30), with an 7a-Kg characteristic of

another pentode, the 6SJ7 (below). You will find that with 250 volts on the anode

and 100 volts on the screen, minus 9 volts on the grid cuts off the 6SJ7. Under

the same conditions, however, minus 35 volts are needed to cut off the 6SK7. This

is why the 6SK7 is sometimes called a "remote cut-off" valve.

Varying the bias of the 6SJ7 pentode has very little effect on the amplification of

the valve except near the cut-off point. This is because the Ia-Vg curve is a straight

line except near the cut-off point.

jjA

//

'

1

i{/

/^/

Vg

\Cut

i ;

off

r < i I \ * \\ i

i

i

Output with-IV bias

Output with

-3V bias

Input signal

la -VgCURVEFOR

6SJ7 VALVE

Va= 250

Vg2= 100

:iiiiiiiiii::iu3

:!j::::j:::::::H5

::i:ii:::iii::H§

iiiilnlllll

::::H:::::::H:H::::::::::::::::a

i

:::liyi::::ll:?H

iiiiiiiiiiiiilHs

|:::::):[:::j::fH

llllllllllllllllil

:':::::::::::::::3

::::::::::::::::3

:::::|::::::::::3

HI::i:ii!!:!l:IH

::::;::::::::):::;

:::::::::::::::H>

::::::::::::::::U

i:i::i:i:ii::i:in

2*

3.30 r§ 4

The Variable-Mu Pentode {continued)

Varying the bias of the 6SK7 pentode varies the amplification. To understand this,look at the 7a-Kg characteristic of this valve. The graph shows a signal appliedto the grid of a 6SK7 under two different conditions : (1) when the bias is - 1 volt and(2) when the bias is -9 volts. Note that the anode current variation is greaterin the first case than it is in the second. Therefore the amplification varies with thebias. This is why the 6SK7 is called a "variable-Mu pentode."

iallllllllllll

16

""A"14 / \ Output with\ / -1 volt bias

12 V/ llllllllllllio ill!

8 ill

6

4 ii-innni:

Vg -16 -14 -12 -10 -8

Jla -Vg CURVE

FOR6SK7 VALVE

ANODE VOLTAGE * 250 VSCREEN VOLTAGE « 100 V

Input Signal

In a variable-Mu pentode, the turns of the control grid are spaced closely at the edges

and farther apart at the centre. This means that the electron flow through the edgesof the grid is cut off by very little negative bias ; but the electron flow through the centreof the grid can only be cut off by a high negative bias.

The grid of a normal pentode, on the other hand, is uniformly spaced. The entireanode current is equally affected by any one value of bias, and cut-off is quicklyreached.

CONTROL GRID OF SHARP «CUT-OFF PENTODE !

iil

CONTROL GRID OF REMOTECUT-OFF PENTODE

You will learn later on how the varying gain which can be achieved by using avariable-Mu pentode makes this valve useful for certain special applications.

§4]

REVIEW of the Single-stage R.F. Amplifier

3.31

Pentode. The value used almost uni-

versally in r.f. amplifiers, because it has

the highest gain, least tendency to oscil-

late, and least distortion.

Variable-Mu Pentode. A pentodewhose h-VB graph curves continuously.For this reason its gain varies with the

bias on the grid. Also called "remote

cut-off" pentode.

R.F. Coil. The primary is connected

between the anode of the r.f. amplifier and

H.T. The secondary is connected be-tween the grid of the next stage and earth.

Selectivity. This is the ability of an

r.f. amplifier to separate signals whose

frequencies are close together. Thenarrower the band of frequencies passed

by an r.f. amplifier, the more selective it is.

Selectivity Curve. This is a nameused to describe the graph ofthe frequency

response of an r.f. amplifier. It shows

you the gain of the amplifier over a par-

ticular portion of the range of frequencies

to which the amplifier is designed to tune.

wm (|> fIK"fijfe -^^ w *****& &r;

3.32 §5: THE TWO-STAGE R.F. AMPLIFIERWhy More Than One R.F. Stage Is UsedRemember, once again, the two functions which an r.f. amplifier performs:

1. It amplifies an r.f. signal.

2. It selects signals of a particular frequency for amplification.

When a receiver is situated quite close to a transmitter, the signal picked up bythe receiver is strong. Such receivers probably need only one r.f. amplifier stage.

On the other hand, some receivers are designed to pick up signals from transmittersseveral thousand miles away, and by the time the signal reaches the receiver it is

very weak. The receiver therefore needs extra amplification to boost it.It is receivers such as this which require more than one r.f . amplifier stage.

^eeo.8WO**888

There is a second, and less obvious, reason for using more than one r.f. stage.

This is that more than one r.f. amplifier stage gives greater selectivity. More selec-

tivity permits a receiver to separate stations whose frequencies are very close together.

Many radio signals are transmitted at frequencies which are close together in thesame r.f. band; and receivers designed to cover these bands need more selectivity

than one tuned r.f. amplifier can give them.

????QSY-604Mc/s \

s GSA-605Mcfs /*»•

.

"

TAGES ARE NEEDED WHERE SIGNAL EREQUENCIES ARE CLOSE TOGETHER

§ 5] 3.33

Why More Than One R.F. Stage Is Used (continued)

In order to understand how more r.f. stages give better selectivity, suppose you

examine a typical selectivity curve more carefully.

Figure 1 below is a curve for a single-stage r.f. amplifier tuned to 500 kc/s. 10

microvolts of r.f. are fed into the amplifier.

You will notice that the output of the amplifier at 500 kc/s is 100 microvolts.OUTPUT 100

Therefore the gain at 500 kc/s is: TNPttT =-tq-= 10.

The output at 465 kc/s and 535 kc/s is 50 microvolts. Therefore the gain at these

frequencies is 50/10=5.

In other words, this single-stage amplifier amplifies at the resonant frequency twice as

. ., ,.. _ . ,.. , GAIN at 500 kc/s 10much as it amplifies at frequencies 35 kc/s away: GAIN ^^ qt^ kc/s =T =2-Suppose, now, that you take another r.f. stage, identical with the first one, and

connect it up to amplify the output of the first stage. Figure 2 below shows the

overall selectivity curve for both stages.

Each stage has a gain of ten at 500 kc/s. Therefore the gain of both stages together

at this frequency is 10 x 10= 100. But each stage also amplifies five times at 465 kc/sand 535 kc/s. Therefore the gain at these frequencies is 5 x 5=25. You see, there-fore, that this two-stage r.f. amplifier amplifies at the resonant frequencyfour times as

much as it amplifies at frequencies 35 kc/s away. This is a much greater selectivitythan one tuned r.f. amplifier will give.

Adding a third tuned r.f. stage will give you still more selectivity than two tuned r.f.

stages; though it may not always be desirable to have as much selectivity as that.

Selectivity (Zonae*,

100

Voltagegain = 10

1000

750

500

250

Voltagegain = 100

450 ! 475 500 525 i 550465 535

450465

475 500 525! 550535

FREQUENCIES

INPUT VOLTAGE - 10 MICROVOLTS

Untuned r.f. stages have little or no effect on the selectivity.

3.34 [§ 5

Selectivity and Bandwidth

It is the job of tuned r.f . stages to amplify over a narrow band of frequencies, and

to reject the rest. The ideal r.f. selectivity, or "response," curve is therefore as

follows:

VOLTSor

GAIN

rBAND_WIDTH

ftFREQUENCY

An amplifier with such a response curve, when tuned to a transmitting station,would readily accept the intelligence transmitted, and would reject all other signals.

The ideal response curve, however, is impossible to produce in practice; and

although by very careful design it is possible to make something very near to the

ideal, you will normally meet r.f. amplifiers with the sort of response shown on the

diagram below:

PRACTICAL

fr.

FREQUENCY

Whereas the gain of an amplifier with an ideal response would be constant over

the full range of frequencies passed, in practice the gain varies over what is called

the "bandwidth."

The bandwidth is the name given to the range of frequencies which a particularr.f. amplifier will satisfactorily select and amplify. The gain of every amplifier youare ever likely to meet will be a good deal better at some of the frequencies lying

within its bandwidth than it will at the others.

§5]

Selectivity and Bandwidth (continued)

Look again at the response curve for a single-stage r.f . amplifier.

3.35

IOO-

75

> 50

25

Ideal response curve.

Response curvefor single stageamplifier.

Response curve

for more selectiveamplifier.

436 452 465 473 SOO

Kc/s

527 535 548 564

The bandwidth for a minimum output of 75 \iV is 527 kc/s to 473 kc/s = 54 kc/s.The bandwidth for a minimum output of 50 nV is 535 kc/s to 465 kc/s = 70 kc/s.The bandwidth for a minimum output of 25 \iV is 548 kc/s to 452 kc/s = 96 kc/s.

These bandwidths indicate how far the response of the r.f. amplifier varies fromthe ideal.

You can see that if the amplifier response curve was more "peaky" (as shownby the dotted line)—that is, if its tuned circuits were more selective—the bandwidthwould be narrower at all levels of output. The greater the selectivity, the narrowerthe bandwidth.

3.36 [§5

Selectivity and Bandwidth (continued)

You learnt in Part 2 of Basic Electronics that the frequency response of an a.f.amplifier is normally indicated by selecting a reference frequency of 1,000 c/s, and

by expressing the power output at all other frequencies as a gain or loss in decibels

(dB) compared with the output at this frequency.

In the r.f . amplifier, however, variations of gain can be related to any convenient

reference frequency.

Call the gain at such a reference frequency Pi, and the gain at any other frequency

you care to choose Pi. Then

PiRelative Gain in dB = 10 log

PiBut, as you know, Power can be expressed as E2/R. Therefore,

Relative Gain in dB = 10 log *j„,J .L Z2IR2Now R\ and R2 are identical, because Pi and P2 are developed across the same

resistance. Ri and R2 therefore cancel out, and the equation becomesl2

Relative Gain in dB = >0.os(|)!

The logarithm of the square of a number is twice the logarithm of the number

itself. So

EiRelative Gain in dB = 20 log -rr-.

OdB KDOwV-

-3dB

-6dB

TOKV-

-20dB

-40dB

50uV-

500K

§ 5] 3.37

How R.F. Stages Are CoupledThere are several possible ways of coupling two r.f. stages, as you have already

seen on page 3.27.

The most commonly used of these methods is shown in the illustration below.

USUAL CIRCUITFOR COUPLING RFAMPLIFIER STAGES

Notice that the anode load for the first stage is a coil. This coil is the primarywinding of an r.f. transformer, and has a high impedance at radio frequencies.The r.f. signal current flowing through the coil induces a voltage in the secondary

winding. The secondary winding is tuned with a variable capacitor which is gangedto the capacitor that tunes the first r.f. stage.

The action of this r.f. transformer is essentially the same as the action of an audiotransformer in a transformer-coupled audio amplifier, except that the secondary ofthe r.f. transformer is tuned.

AnodeCurrent

RF COIL ACTSLIKE ANY OTHERTRANSFORMER

3.38 [§ 5

A Typical Two-stage AmplifierThe circuit of a typical two-stage r.f. amplifier using two pentode valves is

illustrated below.

The signal received at the aerial is fed to the grid of the first valve by a trans-

former (T-l), whose secondary is tunable. The anode load of this valve is the

primary of a second transformer (T-2). The amplified signal induces a voltage in the

secondary of T-2, which is also tunable, the tuning capacitors being ganged.

The valve used in the first stage is a variable-Mu pentode, whose cathode bias can

be varied by adjusting the variable resistor in its cathode lead. This in turn varies

the gain of the stage. The H.T. supply is applied to the anode load of the valve

through a decoupling filter consisting of a 47-KQ resistor and a 001-(xF capacitor.The voltage induced into the secondary of T-2 is applied to the grid of the second

valve, and amplified again.

The second pentode valve is operating under fixed cathode bias conditions.

§5]

REVIEW of the Two-stage R.F. Amplifier

Two-stage R.F. Amplifier. When cor-rectly adjusted, it gives more amplification

and more selectivity than does a single-stage

r.f. amplifier.

3.39

Bandwidth. The range of frequenciesselected and passed by an r.f. amplifier.

-to* an

Alignment. The process of adjustinga group of tuned amplifier stages so that all

the stages tune to the same frequency.When a multiple-stage r.f. amplifier iscorrectly aligned, it will give more selectivity

than will a single-stage amplifier. Whenit is not correctly aligned, it may give farless selectivity—and it may not be able toseparate a wanted signal from an unwanted

adjacent one.

Output

0bcilUdcM>

3.42 §6: OSCILLATORSIntroduction to Oscillators

You have seen how amplifiers function in electronic circuits. No less importantare oscillator circuits, or "oscillators."

Most modern radio receivers which you have used in your home contain oscil-lators. Every transmitter that sends intelligence through the air employs anoscillator to produce these signals. This is not only true of "ground stations" likeLuxembourg or the B.B.C. Home Service; it applies to every transmitter in a tank,on a ship, or in a plane.

Radio communications of every kind would be greatly handicapped and limitedif oscillator circuits were not available.

3&feM*£&&:'£&

Nor are oscillators used exclusively in communications equipment. Most of thetest equipment you use—signal generators, frequency meters, and the like—containsoscillator circuits. You will find oscillators in radar and line equipment, and incertain types of guided missiles and torpedoes.

§ 6] 3.43

What an Oscillator Does

Basically, an oscillator does nothing more than generate an a.c. voltage at a

desired frequency.

The audio signal generator used in your work on audio amplifiers is an audio

oscillator. The audio frequency oscillator generates an a.c. voltage at any frequency

from to 15,000 cycles per second. The r.f. signal generator used when working

with r.f. amplifiers is a radio frequency oscillator. Both these oscillators supply a

test signal which enables you to check and find faults in amplifiers.

A radio transmitter generates a high frequency a.c. signal, amplifies it, and thenradiates this amplified signal by means of an aerial.

Where does this high frequency a.c. signal come from? From an oscillator.Indeed, a radio transmitter is really nothing more than an oscillator with some

high-power r.f. amplifiers to step up the oscillator signal so that it can be radiated

long distances by the aerial. (You will learn more fully about transmitters in Part 4

of Basic Electronics)

Again, the most advanced type of radio receiver, the superheterodyne receiver,

also contains oscillator circuits. (And the "superhet" in turn is the subject-matter,

of the second half of Part 5.)

THE OsC 1 1latOf GENERATES AN ACVOLTAGE AT A DESIRED FREQUENCY

Radio FrequencyOscillator

3.44 [§6

What Oscillations Are

Anything which swings back and forth in a uniform way is said to be "oscillating."A violin string "oscillates" when a bow is drawn over it. A swing moving back andforth "oscillates.' A pendulum swinging on a clock "oscillates."

X^̂

A CHILDS

Consider the pendulum. When itreaches the extreme left-hand side of its

swing, it comes to rest momentarily;

and all its energy is stored as "potential

energy." Half way through its swing,it is moving at its greatest speed; and

all its energy has been converted into

"kinetic energy," or "energy of motion."

When it completes one swing, arriv-ing at the extreme right-hand position,

it again comes momentarily to rest; and

its energy is all "potential" once again.

Now, it is possible to represent this motion by one-half of a sine wave, plotting

velocity against time—velocity from the left towards the right being always reckonedas positive.

§6]

What Oscillations Are (continued)

Since the return swing from

right to left is a reversal of

direction, the second half of

the sine curve is shown below

the line. Thus one complete

cycle of oscillation of the pen-

dulum may be represented by

one complete cycle of the sine

wave.

..ONE COMPLETECYCLE OF OSCILLATION

3.45

Did you ever notice that any one complete trip on a swing takes the same time

as any other trip? Indeed, you can perfectly well represent any three cycles of a

swing in motion like this (see below): the time from h to h is always the same as

that from h to t5 , or that from ts to h. And the time required for the various half

cycles (fi to h, h to t3, etc.) is also always the same.

Clocks and watches keep accurate time because the time taken by any one swing

of the pendulum or balance wheel is always the same as that for any other swing.

This is as true of the seventh swing as it is of the first.

Two conditions must exist before anything can be said to oscillate: (1) there must

be back-and-forth motion (vibration), and (2) the period of time taken by each

complete cycle of oscillation must be the same.

3-46B 4

What Oscillations Are {continued)You know that the motion of a swing will eventually run down. You know, too,

that this loss of energy is due to friction; and that to compensate for this loss,additional outside energy must be supplied in a uniform way. What happens whenoutside energy is not supplied can be shown by this curve:

additional energysupplied

This is called a "damped" wave. It is like a sine wave, but with the height(amplitude) of successive cycles diminishing gradually. The time interval remainsthe same.

How would you supply the necessary energy to prevent "damping?"If you were pushing a child on a swing, you would not make the next push until

the swing had just completed its arc and was about to reverse its direction. Thisapplication of energy at the proper point or with the proper timing is said to be"in phase" with the original motion.

To supply energy to an oscillator in order to support its natural period of oscil-lations, the outside source of energy must be in phase with the natural period ofthe oscillator.

You know now that to support a stable oscillator, two conditions must be fulfilled:1

.

Energy must be supplied to compensate for loss of energy in the oscillator.2. When supplied, the outside source of energy must be in phase with the natural

period of the oscillator.

§6]3.47

The Electronic Oscillator

An electronic oscillator is a simple circuit, consisting of a capacitor and of a coil

connected in parallel.

To understand how such a circuit can be made to oscillate, consider what happens

when a capacitor is charged and discharged.

An uncharged capacitor has an equal number of positive and negative electrons

on each of its plates. But when this capacitor is connected across a source of d.c.

voltage, one plate will be charged negatively and the other will be charged positively.

What has happened is that there are now more than the original number of electrons

on the negative plate, and fewer than the original number of electrons on the positive

plate. Moreover, the excess of electrons on the negative plate is exactly equal to

the loss of electrons on the positive plate.

UNCHARGED CAPACITOR CHARGED CAPACITOR

S3 S3Si G© £32.

S3 S3 S3©© G© fi)

When a short circuit it put across the charged capacitor, the excess electrons are

attracted back through the shorting wire on to the positive plate. Each plate once

more has an equal number of positive and negative charges, and the capacitor is

uncharged.

CAPACITOR DISCHARGING CAPACITOR DISCHARGED

S3 S3 Sc© S3 &_

-0-

K-) S3 S3G© £©

3-48g 6

The Electronic Oscillator {continued)

On the previous page you saw what happens when a short circuit is connectedacross the charged capacitor. If an inductor is connected across the chargedcapacitor instead, however, the results are quite different.You will remember from your work in Basic Electricity that an inductor has a

peculiar electrical characteristic—it tends to resist any change of current throughitself. You remember also that when current flows through a coil, a magneticfield is generated round the coil. Any change in the current causes the magneticfield to expand or contract. This expansion or contraction of the magnetic fieldcauses the magnetic lines to cut across the turns of the coil—resulting in thegeneration of a voltage which opposes the change in current.When, therefore, a charged capacitor is switched across a coil at a moment when

the voltage across the circuit is at its maximum negative value (see top illustrationbelow), the electrons stored on the negative plate of the capacitor cannot immediatelyrush through the coil to its positive plate. For as soon as current starts to flow intothe coil, a magnetic field starts to build up, inducing a voltage across the coil whichopposes the flow of electrons from the negative plate.

In consequence, the charged capacitor cannot discharge immediately through thecoil; and the larger the coil, the longer it takes for the capacitor to discharge.As the capacitor discharges, so the voltage across the circuit decreases; and the

energy stored in the capacitor is transferred to the magnetic field which builds upabout the coil (see illustration (2) below).

MAGNETIC FIELD INCREASESAS CAPACITOR CHARGE DECREASES

Switch Thrown /'o

\ no current flow

i O E

Increasing *~

current fm

Capacitor J.dischargesT

Magnetic fieldbuilds up

§6]3.49

The Electronic Oscillator (continued)

By the time the capacitor has completely discharged, all of its electricalenergy

has been transformed into magnetic-field energy around the coil.

But as soon as this condition has been reached, the magnetic fieldbegins to col-

lapse around the coil—see (3) below. The collapsing magnetic lines cut acrossthe

turns of the coil, and induce a voltage across the coil. This induced voltageis

opposite in polarity to the original voltage across the capacitor,and increases in

magnitude as the magnetic field collapses.

Because of this induced voltage, electrons are forced to flow through thecoil in

the same direction. They are stripped off the upper plate of the capacitorand

forced through the coil on to the lower plate.

.

All the energy of the collapsing magnetic field goes into forcing a negativecharge

on to the lower capacitor plate. By the time the field has completely collapsed, all

the magnetic energy has been returned to the capacitor as an electric charge;and the

voltage across the capacitor is exactly opposite in polarity to the originalcharge (see

(4) below).

WHEN THE CAPACITOR STOPS DISCHARGING,

COIL RECHARGES THE CAPACITOR

DecreasingCurrent

Magnetic fieldstarts to collapse

Capacitorbegins to charge J.with opposite

polarity

\F-

-f

Magnetic fieldcompletely collapsed

No current

Capacitorcompletely +recharged

with

opposite polarity

i ©

3.50[§6

The Electronic Oscillator {continued)

Now that the electrons are all stored on the lower plate of the capacitor, thecharge is exactly the opposite of what it was originally.

So the electrons are now attracted to the upper positive plate through the coil.As the capacitor discharges, a magnetic field builds up around the coil (see (5)below). The collapse of this magnetic field forces additional electrons off the lowerplate on to the upper plate. By the time the magnetic field has completely collapsed(see (6) below), all the electrons are back on the upper plate, and the situation isexactly the same as it was when the capacitor was first charged.The entire cycle then repeats itself over and over again. Electrical energy is

alternately stored as a charge on a capacitor, and as a magnetic field around a coil.This is what is meant by electronic oscillation.

RECHARGING CAPACITOR TO ORIGINAL CONDITION

Magnetic fieldIncreasing builds upCurrent

Magnetic fieldcollapses

Capacitorbegins to J„+

discharge in ^p»_opposite

direction

Capacitor J.recharges +T

% »

*

Decreasing CurrentIf an oscilloscope were connected in parallel across the coil and capacitor, and if

there were no resistance in any part of the circuit, the rise and fall of voltage wouldappear as a sine wave.

Moreover, the oscillations would continue indefinitely, if there were no resistancein the circuit.

But resistance cannot in practice be completely eliminated from any circuit, andsome of the electrical energy of oscillation is dissipated by the resistance as heat.Because of this loss of electrical energy, the voltage becomes lower and lower on eachswing, and the oscillation eventually disappears.

§6]3 '51

The Electronic Oscillator (continued)

In order to make the oscillations continue indefinitely, enough electrical energy

must be put back into the LC circuit (called a "tuned circuit") to overcome these

losses caused by resistance.

Moreover, this electrical energy must be put back into the circuit at just the right

moment, so that it will give its little extra "push" or "kick" at the proper time Gust

as if it were the little push you give to a swing at the end of its arc).

One way of supplying this electrical push to the LC circuit is to switch a source

of voltage across the capacitor just at the moment when the capacitor is reaching

its full charge. In this manner, oscillations can be made to continue indefinitely.

t Loss of voltage in onecycle due to resistance

Voltage across LC circuitwhen no additional energy isapplied after initial charge.

+ K

Extra voltage beingapplied to LC circuit.

Sine wave with additionalelectrical energy being supplied

at the proper time.

—7ZZZ Voltage supplied to'f"" LC circuit by battery.

Notice that the only kick the oscillator circuit receives is the small fraction ot a

volt necessary to overcome the voltage drop caused by the resistance in the circuit.

Yet the LC circuit is able to generate a sine wave voltage even though the kick it

receives does not resemble a sine wave in any manner, and even though the kick

lasts for only a very small part of the cycle.

Before you begin to feel surprised at this, however, remember that the flywheel on

a one-cylinder engine is able to make one complete turn even though it receives

only a very brief push from the piston on each revolution. This resemblance between

the action of an LC circuit and the flywheel of a one-cylinder engine has led to the

use of the term "flywheel effect" to describe the oscillations in an LC circuit.

[§63.52

The Feedback Circuit

The method of supplying extra energy to the LC circuit described on the previouspage would work very well if there were some switching arrangement which couldoperate at the frequencies required. But some oscillators must be able to workat frequencies well over 100 million cycles per second, and it is quite obvious thatno mechanical switch could work at this speed.The answer to this problem of supplying electrical energy at the proper instant

(however rapidly this instant recurs) is to use a valve circuit.By connecting the LC circuit to the grid of a valve, the oscillating voltage can be

amplified. If a small portion of this amplified voltage can be fed back in the properphase, enough electrical energy will be put back into the LC circuit to overcome theresistance losses in the LC circuit.Note that this valve used in an oscillator does not itself do any oscillating. It

is the LC circuit that oscillates, and it is the valve that gives the kick.

THE AMPLIFIER VALVEIS USED TO KEEP THE

4(2 gincuit OtciUatuty

§ 6]3 '53

Frequency Stability of Oscillators

One of the important characteristics of oscillators which you are going to learn

about is their frequency stability.

Although you have not yet had much of a chance to learn about oscillators, it

ought to be obvious to you already that one of the things an oscillator must do is

to maintain the frequency to which it is set. Unfortunately, however, all oscillators

(though some less than others) tend to drift in frequency unless steps are taken to

stop them doing so.

Imagine what would happen if frequency drift were not prevented. If one wireless

station was transmitting to another, and the oscillator in its transmitter drifted off

frequency, the message would not be received. If the oscillator in a police car

receiver drifted off frequency, that car would receive no messages at all. If the

oscillator in a ship's submarine detecting unit drifted off frequency, that ship could

not detect submarines and would be easy meat for the first enemy submarine that

came along.A great proportion of all electronic equipment contains oscillators. If these

oscillators were allowed to drift off frequency, all this equipment would be useless

until the oscillators were reset to the correct frequency.

Drift is caused by several factors. Vibration, varying loads and varying supply

voltages will cause an oscillator to drift, as will changes in temperature. And since

much electronic equipment is subject to all of these factors, some compensation is

usually included in all equipment which contains an oscillator.

You will learn about these methods of compensation in the next few pages.

3.54 §7: THE ARMSTRONG, HARTLEY, ANDCOLPITTS OSCILLATORS

The Armstrong Oscillator

Most oscillators operate in the same way. Energy is coupled back from theoutput of an amplifying valve to a tuned oscillating circuit connected to its grid.The object is to compensate for the inevitable heat losses in the tuned circuit causedby electrical resistance. If these losses can be compensated, oscillations will continue.

If you understand how oscillations are maintained in the Armstrong oscillator,you will understand the basic principles underlying most oscillators.

ASMSTOWS

Coupling

Coil

Oscillator

Coil

HT>

Jiff-dociSSSu

mM&mThe Armstrong oscillator is like an r.f . amplifier with one modification—a coil is

introduced into the anode circuit.This coil is called the feedback, or coupling, coil. It is wound adjacent to the

oscillator coil (usually both coils are wound on the same coil former), so that whenanode current flows through this coupling coil, L c , an e.m.f. will be induced in theoscillator coil L. (Actually it is not the d.c. anode current itself, but variationsin this d.c. current, which produce the changing magnetic fields responsible forgenerating the induced voltages across the oscillator coil.)

It is this induced voltage which acts as the feedback voltage whose function is tosustain oscillation.

§ 7]155

The Armstrong Oscillator (continued)—Grid Leak Bias

There is nothing complicated about the operation of the Armstrong oscillator.

When the power supply is turned on, a flow of electrons surges from the cathode to

the anode, and through the coupling coil to H.T. This surge of current causes

the rapid build-up of a magnetic field round the coupling coil; and this expanding

magnetic field suddenly induces a voltage in the coil of the LC circuit.

This voltage surge in the LC circuit is sufficient to begin oscillations. All that

the valve and coupling coil have to do from now on is give a voltage "kick" to the

LC circuit at the proper time during the cycle of oscillations.Notice that no cathode resistor or battery bias is used in this Armstrong oscillator

circuit. The proper negative bias on the grid is obtained from the resistor and

capacitor in the grid circuit. This method of obtaining bias is called "grid leak"

bias, and is used in many other oscillator circuits.

As you will shortly see, the valve must be biased well below cut-off for most of

its cycle of operation—which means that the valve is operating as a Class C amplifier.

This high negative bias is maintained by means of the resistor and the capacitor in

the grid circuit. When anode current first begins to flow, there is no negative bias on

the grid. This means that a very large anode current will flow through the coupling

coil (causing oscillations to begin in the LC circuit); and in addition there will be a

sudden pulse of electron current in the grid circuit. This flow of electrons causes

a voltage to be developed across the resistor (and across the capacitor inparallel

with it), and this voltage is such that the grid becomes strongly negative with respect

to the cathode.

The grid capacitor stores up enough electrons to keep the grid negative for nearly

all the cycle of oscillation. The charge on the negative side of the grid capacitor is

large enough to counterbalance a positive charge on the top plate of the LC circuit

capacitor. Only when the positive charge on the top plate of the LC circuit

capacitor reaches its maximum will it counterbalance the grid capacitor, and cause

anode current to flow.

fcdfafiP***

3.56[§7

The Armstrong Oscillator {continued)—How Oscillations Are MaintainedNow let us analyse what happens through an entire cycle of oscillation.Begin at the time when the electrons have arrived on the top plate of the tuning

capacitor, after travelling through the coil from the bottom plate. At this stage noanode current flows.

After the upper plate of the tuning capacitor has reached its maximum negativecharge, the electrons begin to flow back through the coil to the bottom plate. Asthe electrons begin to accumulate on the bottom plate, the top plate becomes moreand more positive with respect to the bottom plate. The only thing that preventsanode current from flowing is the fact that not all of the electrons on the chargedgrid capacitor have leaked off through the grid resistor, and the grid charge can stillcounteract the positive voltage at the top of the LC circuit.

Finally, however, the voltage at the top of the LC circuit becomes so positive thatit overcomes the negative bias maintained by the grid capacitor. At this pointanode current begins to flow; and it continues to flow for the brief space of timeduring which the grid voltage is above the cut-off value.

OPE