FUNDAMENTALS - 1

BASIC PHYSICS THEORY

This is a simplified discussion of the basic Physics which underlies our theory of Electricity
and Magnetism.

LESSON 1
1.1 THE ELECTRIC CURRENT

Classical theory holds that Matter can be divided into smaller and yet smaller particles until a
point is reached where further division causes a change in the properties of that matter. This
smallest unchanged particle is called a Molecule. Most molecules can be sub-divided but often
the resulting Matter has properties which are unlike those of the parent molecules; these smaller
particles of different form are known as Atoms. Any attempt to sub-divide atoms results in their
destruction and so atoms represent Nature's ultimate building-blocks. Atoms are known in more
than 90 different kinds and, because they are the fundamental building-blocks, they are known
as elemental atoms and the different kinds of Matter which they form are called Elements.
Matter in which the molecules are formed by bonding together different atoms is referred to as
a Compound. (A mixture differs in that it contains different kinds of molecules which are not
bonded together and so they can be separated from the mixture as Elements.)

Elemental atoms bond with each other to give us the pure substances such as Copper,
Oxygen and Iron (not Steel which is a compound of Iron). Compounds of different atoms are
represented by such things as Water (Hydrogen and Oxygen), Salt (Sodium and Chlorine) and
ultimately Life itself.

** Research into the nature of atoms gave rise to puzzling phenomena which, under
mathematical analysis, has led to the conclusion that the atoms are themselves constructed from
a large variety of very-small particles. For our purpose here it is necessary to consider only the
first stage of that analysis which says that an atom contains a relatively-heavy Nucleus around
which is gathered a number of extremely-small Electrons. The different Elemental Atoms owe
their differing characteristics to the differing numbers of electrons which they possess.

** It is found by experiment that, when electrons are placed together, they actively repel each
other; similarly when two nuclei are placed together they too actively repel each other.
However, when a nucleus and an electron come together, they exert a powerful mutual
attraction.

** This behaviour is explained in terms of two differing Electric Charges which, by tradition,
are labeled as positive (+) and negative (-). Also by tradition the positive charges are attributed
to nuclei and the negative charges to electrons. The argument then put forward is that two like
charges (either two nuclei or two electrons) mutually repel while two unlike charges (a nucleus
and an electron) mutually attract.

** A complete atom does not show a tendency either to attract or to repel other atoms and so it
must be concluded that overall they have zero electric charge; this must indicate that there are
an equal number of positive and negative charges. To cope with this the heavy nucleus is sub-
divided into an appropriate number of Protons each of which carries a unit positive charge;
there is one proton to match each electron in every atom.

Some electrons, held only weekly by their parent atom, can be dislodged; atoms which lose
electrons in this manner are left with positively-charged holes which then exert a mutually-
attracting force with any passing electron and so tend to re-fill themselves. However, if the
disturbing force is more powerful than the recovery force of the atom, the dislodged electrons
can drift away and form a moving stream of "free" electrons which is believed to constitute
what we know as an Electric Current.

1.2 THE WATER ANALOGY

Just before 1800 interest began in a new Science called Galvanism following the
experiments of an Italian scientist named Galvani. In an effort to make sense of the phenomena
they were uncovering the learned men of the day drew an analogy between the electrical effects
they were discovering and the water systems which they used and understood.

They knew that a water pump, by raising a pressure of water, could drive a current of
water through a system of pipes. Similarly they imagined that their generators of electricity
behaved as an electricity pump, that they raised an electrical pressure which then drove
a current of electricitythrough a system of electrically-conducting paths.

To a large extent that analogy is still accepted today. As described above a generator of
electricity is considered to "pump" electrons continuously from a point of low-potential to a
point of high-potential (it creates an electrical pressure); the point from which negatively-
charged electrons are removed is left with a residual positive charge while the point where the
electrons are dumped acquires a surplus of negative charge.

A circuit Designer provides a suitable conducting path external to the generator through
which the surplus electrons, piled-up on the negative terminal stream back to the positive
terminal where they are gathered up and pumped again and again to the negative terminal This
is why there must always be a complete circuit before electricity will flow.

This analogy with a water system is useful in explaining the ideas behind electrical theory
and it is used often in the following text.

1.3 ELECTRIC SOURCES

In practice it is not that easy to set electrons drifting and, as a general principle, they have to
be pulled from one direction with a positive charge and pushed from the opposite direction with
a negative charge. The prime-mover in such an operation is known as a Generator of which the
two major types are the Chemical Cell and the Mechanical Generator.

1.3.1 The Chemical Cell

 ** There are many designs of Cell and they have many intended purposes but their general
requirements are a mixture of chemicals in which are immersed two plates
called Electrodes. These are usually made from different materials; see Fig.1. The different
materials of the electrodes react differently with the chemical mix and this results in electrons
being stripped from one electrode with an equal number deposited on the other.

** The basic construction of electric cells is shown in Fig. 1. The electrode which loses
electrons :is left with a non-neutralised positive charge while the other, with too many electrons,
acquires a negative charge.

Experimenters quickly discover also that it is necessary to mechanically-restrain the

electrodes because the unlike charges which they carry set up a mutually-attractive force which
pulls the electrodes together.

** As already discussed the holes on a

positively-charged electrode will attract
electrons while the surplus electrons on a
negatively-charged electrode will repel
electrons; it follows therefore that, as the
electrodes acquire their charges, so those
charges try to force electrons back across the
cell.

** Ultimately a state of equilibrium is

reached where the chemical action of the Cell is
exactly balanced by the reverse action of the
charged electrodes and the electro-chemical
activity of the Cell is halted.

** Between the two electrodes there now

exists a distinct difference in their state of
charge and this is referred to as a potential
difference but is usually referred to simply as
a p.d. The essential character of a p.d. is that it
is called into being by a transfer of electrons (a
transfer of charge); a p.d. can be generated
across other electrical components where it can
be used as a "secondary generator" to start
another transfer of charge - another electric
current.

** A potential difference is an electric force

which is equivalent to pressure in the water
analogy and it is measured in units
named volts in honour of the Italian scientist
Volta.

An electric current, which flows under the

 influence of electric pressure (volts), is
measured in terms of the number of electrons
which drift past a given point in a period of one
second; of course we cannot count them and
current has to be assessed through the various
effects which it produces.

(The standard torch battery delivers approx.

1.2 to 1.5 volts: this experimental cell delivers
0.6 volts approx.)

** The number of electrons involved is measured in coulombs (after Coulomb) which

therefore are units of quantity (charge or no. of electrons); the flow-rate in coulombs-per-second
([Link] electrons per second) indicates the strength of the current and this rate is expressed
as amps. Properly the unit of current-flow is the ampère named after the French scientist
Ampère but it is invariably shortened to the more convenient amp.

1.3.2 Work and Power

To deliver water from a reservoir positioned at some height above the delivery point requires
a pipe; at the lower end of that pipe there is a pressure caused by the weight of the water; this is
measured in terms of the height of the water-surface above the delivery point and is known as
the Head of Water.

The pipe allows water to be lowered from the reservoir or, given a pump that can more than
oppose the pressure, it allows water to be raised to the reservoir. We have all moved objects
either horizontally over the ground or vertically and we complain that we have had to work;
fuel/food consumed in doing that work manifests itself as heat and we have to take steps to cool
down again or suffer heat-stroke. The faster the rate at which we work so the hotter we get.

In such a mechanical world work is measured by multiplying the weight that is moved by
the distance through which it is moved and the result can be expressed in units such as foot-
pounds or centimetre-grams.

In the water system referred to above there is a choice in that a small-diameter pipe can be
used to deliver a given quantity of water over a long period or a larger diameter pipe can be used
to deliver the same amount in an instant deluge. Should you be sprayed by the smaller pipe you
may get wet but otherwise would be unharmed; a similar acquaintance with the larger pipe
however could have serious consequences. The ability of the large pipe to cause damage is
described by saying that it delivers a powerful jet of water.

** Because the larger pipe works (delivers water) at the greater rate we say that it is a system of
greater Power. Power expresses the rate at which work is done and is measured here by
multiplying the pressure by the amount of water delivered per second. (Note that a river
delivers (at low pressure) - and therefore in innocuous fashion - much more water-per-second
than a dangerous water hose.)

** To transfer this concept into the electrical world:

The amount of water delivered becomes the amount (the number)of electrons moved (or the
amount of charge which has been transferred); the amount is measured in coulombs.
The rate at which the water is delivered becomes the rate at which electrons drift and is
measured in amps.
The head of water (the height through which the water falls) becomes the electrical pressure or
p.d. and is measured in volts.
The power of the water jet becomes the electrical power dissipated as heat and is measured as
before by multiplying pressure (volts) by rate (amps).

** The unit of power in an electrical circuit (volts x amps) is the watt named in honour of the
Engineer James Watt. Power appears as heat in the component through which the current flows
and across which the voltage is developed and, of course, it causes a rise in the temperature of
that component. This temperature rise continues until the rate at which heat is lost to the
environment equals the rate at which heat is being generated. The heating effect is inescapable
and so components must be designed so that they dissipate enough heat to keep the temperature
within designed limits; care must be exercised in placing hot components.

When, as in Fig.2, an electrical path is connected between the terminals of a cell it is to be

expected that an electric current will flow as electrons are propelled from the overcrowded
negative terminal to the positive terminal. The process deserves a closer look however because
it conceals a very important principle.

As soon as the external connection is made the electrons begin to slide, snakes-and-ladders
fashion, down the potential gradient from the negative to the positive terminal. In doing so they
upset the equilibrium that had been established between the charged electrodes and the chemical
mix and so the chemical generation of electric-charge re-commences. This causes the escaped
electrons to be gathered up again and replaced on the negative terminal to go around the circuit
again and again; the cell is seen to be acting as a pump as envisaged when drawing the water
analogy.

** The important action, which applies to every generator, is the destruction of the
equilibrium; while that remains intact the chemical action is halted and the call cannot maintain
a current flow. It is a fact of Life that, when any generator is called on to supply current, its
terminal-voltage falls. The greater the demand for current so the greater must be that fall; the
greater must be the failure to achieve that equilibrium.

This natural phenomenon is an annoying inconvenience for the designer-engineer who, to

calculate the current which will flow in the circuit he designs, needs to know the terminal
voltage; the problem is that he cannot know the terminal voltage until the value of the current is
known! There is a simple and elegant solution to this problem which involves thinking in terms
of "perfect" theoretical generators which could not possibly have a practical existence; the
matter is dealt with under 1.15: Internal Resistance.
 ** From the above description it is clear that
current which is supplied by a chemical cell
flows always in the same direction namely from
the negative to the positive terminal. Confusion
arises because, in early theorising, it was
decided arbitrarily that current would be taken
as flowing from positive to negative.
Enshrined in many of the classical "Laws", and
still to be found in old textbooks, the
convention survives and so it is important,
where confusion can arise, to state which
convention is being used. It is customary to
refer to old-fashioned pos-to-neg flow as
classical current and to the now-accepted neg-
to-pos flow either as electron-flow or as
electron current.

Current does not always flow in this one-

way fashion and so, as described here, it is
called direct current or simply dc. Reference is
often made to a "dc voltage"; this seldom
causes confusion but it is best avoided by using
Fig.2 Cell with an External Path
the correct term direct voltage.

The word battery is commonly used wrongly

also; it is a contraction of "battery of cells" and
implies a package which contains more than
one cell. A 1.4-volt torch battery should be
referred to as a cell; a (nominal) 9-volt battery
contains six such cells.

1.3.4 Majority-carriers & Minority-carriers

The discussion above proposes that an electric current is in fact a flow of "free" electrons.
When dealing with transistors and other semiconductor devices there is a great deal of talk about
n-type and p-type semiconductors and about majority and minority carriers. The meaning of
these terms is not particularly important for the RAE but, for those who wish to know, the
following explains the basic theory.

The basis of most semiconductor devices today is the Element Silicon. However pure
Silicon does not conduct an electric current and it has to be contaminated or doped. Some
doping materials increase the number of free electrons (carriers) which are available to create an
electric current and the result is known as semiconducting material with majority carriers ;
others cause a decrease in the number of free electrons and these form semiconducting material
with minority carriers.

But how can a flow of electrons take place in a material which is deficient in free electrons
? Imagine a pea-shooter which is filled with peas ; if an extra pea is pushed into the left-hand
end of the tube a pea is expelled from the other end as all the peas move along the tube from left
to right. This is likened to the action in a majority carrier.

Now consider the same pea-shooter filled with peas but, instead of pushing-in one extra pea,
a pea is sucked-out from the right-hand end so leaving an empty space or "hole". Under the
force of suction the next pea in line jumps from position 2 into the end position once occupied
by the extracted pea so causing an impression that the hole has jumped to the left from position
1 to position 2. The pea in position 3 jumps into the hole vacated by pea no.2 thus causing the
hole to move apparently one more step to the left. In this manner the peas all move to the right
as before but the hole appears to move in the opposite direction until a space is created at the
left-hand end of the tube for the injection of another pea. This is the mechanism suggested for
current-flow by minority carriers.

The two types of material produce transistor devices which have different properties.

END OF LESSON 1

* * * * * * * * * * * *

QUESTIONS

1. The very smallest particle of Matter is the Atom. How is this related to the Molecule?

2. What are the constituent particles of the Atom as far as Radio Theory is concerned? How do
we allocate electric charges to them?

3. Although atoms bond together chemically to form large quantities of either Elements or
Compounds they do not show any electrical attraction (or repulsion) to each other. How do
we account for this electrical neutrality?

4. What does Standard Theory propose as an explanation of the flow of electric current? What
is thought to be the action of a Cell by which it causes current to flow around an electrically-
conducting path?

5. In constructing electric cells how do we ensure that the two electrodes offer different
potentials?

6. Unless it is provided with a complete electrically-conducting path between the electrodes a

cell ceases to produce current; what sets-up the equilibrium that makes current-production
impossible?

7. What are the Units that are used to express the magnitude of::

(a) current-flow

(b) electrical pressure or p.d.

(c) power that is dissipated when current is driven through an electrically-conducting

path?

Back to Top of Page

 LESSON 2
1.3.5 Mechanical Generators

** These generators too come in many different forms and serve a variety of purposes but they
have a common principle of operation. Mostly they are rotary machines which combine a rotary
mechanical force (a torque) with a magnetic force and, in some manner which is well
documented but totally unexplained, cause a flow of electric current.

The traditional argument is that an electric current flows under the influence of a p.d. and
indeed this is the explanation usually accepted for mechanical generators whose action is
thought to generate a voltage. However, as electrical theory is developed, problems can arise if
this line of thought is taken too literally. In as much as we are unable to explain the operation of
these machines there is justification for accepting the reverse order namely that the machine
causes current to flow and that this current transfers electrons from one of its terminals to the
other. To accept that positive and negative potentials arise in this manner does not affect the
above argument in the creation of equilibrium conditions and so does not invalidate the
explanation of the fall in terminal voltage when current is permitted to flow.

** Whatever the truth may be it is usual to regard all generators as possessing an internal force
that encourages the migration of electrons; this motivating force is called the electromotive
force of the generator and is invariably used in its contracted form emf. The emf represents the
voltage that appears between the terminals of a generator when the current-flow is zero; i.e. it is
the equilibrium voltage referred to above. The lesser voltage which results when current is
flowing is usually called the terminal voltage.

** The difference between emf and terminal-voltage is important particularly in practical

calculations but the subject is dealt with in greater detail under 1.15: Internal Resistance.

** There are occasions when the concept of emf is abandoned and a generator is considered as
delivering a current; this matter too is dealt with under Internal Resistance.

** Because of the construction of mechanical generators and their mode of operation they do
not produce direct current but one which changes cyclically in both strength and direction. The
output from these machines is called alternating current but it is usually referred to by the
initials ac; again it is not uncommon to see reference to an "ac voltage" but the correct term
is alternating voltage. Machines which produce ac are called alternators

There is a rotary machine called a dynamo which allegedly produces dc; in fact it is an
alternator but has a built-in rotary switch (a commutator) that, literally, reverses the connections
each time the current reverses direction. The "alternators" fitted to modern cars are in fact
alternators but they have built-in semiconductor devices (see below) which imitate the action of
the dynamo's rotary switch.

** In general it is easier and cheaper to generate in ac form than in dc and, where direct current
is required, a means must be provided to convert between the two forms. This is dealt with fully
in Part 5: Power Suppliers. It is sufficient to note here that conversion from ac to dc is a
process known as rectification and the devices used are called rectifiers. Conversion from dc to
ac is termed inversion and is achieved in equipment known as an inverter.
1.3.6 Polarity

** This term is used to describe the direction in which an emf or p.d. acts. For example it
might be said that the polarity is such that Terminal-A is more positive than Terminal-B. At the
output terminals of an alternator the polarity constantly reverses. Perhaps from this has sprung
the word pole which is used to indicate a point of concentration; e.g. the terminals of a cell or
a generator can be described as its positive and negative poles; magnets, including the Earth,
have North and South poles.

1.4 CONDUCTORS AND INSULATORS

** The discussion of electron-drift began with the words: "… Some electrons, held only weakly
by their parent atom …" . There are of course atoms which hold on very tightly to all their
electrons and consequently it is difficult to cause electron drift. Materials in which electrons are
not easily dislodged and which therefore do not willingly conduct an electric current are called
either non-conductors or insulators.

** Materials in which "free" electrons are easily dislodged are called conductors and, usually,
they are also good conductors of heat. Most good conductors are metals but a notable exception
is the element Carbon; this appears in many forms and is a most useful material in electrical
engineering.

** None of the insulating materials are perfect in their action and they all allow passage to
small leakage currents. Leakages tend to remain small even when the applied voltage is
increased considerably but, as the leakage current increases, so too does the amount of heat
generated. Because most insulators are poor conductors of heat excess leakage can cause a rapid
local rise in temperature and so insulating materials tend to fail suddenly and self-destruct.

** There is a class of materials midway between conductors and insulators which are known
as semiconductors; these can be switched electrically between states of conduction and non-
conduction without suffering damage and they form the basis of transistor and chip technology.

1.5 RESISTANCE AND RESISTORS

The argument above has been that some kind of electromotive force is required to dislodge
free electrons if an electric current is to be set in motion. Without that emf any current-flow in
existence ceases and so the only logical conclusion is that the material fights back; it does not
approve of its electrons being harried and moved-on . In fact such action has been mooted
already by pointing out that a dislodged electron leaves behind it a positively-charged hole with
a distinct come-hither to any nearby electron.

** In short, even though a material may be classed as a conductor, it resists a flow of electric
current. This action is referred to as the material's resistance; it is measured in ohms in honour
of the German physicist Ohm.
 ** For most materials resistance is a property which is determined only by the length and
cross-section area of the conducting path. Considered as a standard unit, such as a 1-inch cube,
the resistance does not change. In terms of a standard unit however each material has its
own specific resistance. Metals such as Copper and Silver have very low resistances but glass
offers a very high resistance.

** Some specially-made materials have resistances which vary according to some other
parameter; for example; their resistance may either rise or fall when they get hot. See also
under 1.12.1 Ohms Law.

** In one sense resistance is a nuisance to the electrical engineer and yet, without resistance, it
would not be possible to control a flow of current. The behaviour of electrical and electronic
circuits depends on the values of resistance that are present in different parts of those circuits.
Components are manufactured therefore to provide known specified amounts of resistance and
these are known as Resistors.

** Apart from possessing different values in ohms resistors are manufactured in different
physical sizes; large components provide a larger surface area through which to lose heat and so
they can handle greater powers than small components. Modern materials can operate at higher
temperatures than the older carbon-type resistors and so the physical sizes have tended to
decrease; additionally in the low-power circuits of modern semiconductors power-ratings tend
to be much lower than in valve circuits. However it is necessary to be aware of the heat which
resistors may generate and to place them accordingly in a piece of equipment.

** Resistors are manufactured also in forms which enable their value to be adjusted. One such
has been mentioned above which is made from material which is temperature sensitive. Other
types can be varied by mechanical adjustment, by varying the voltage applied to them, by
illuminating them with light or some other radiation, by exerting on them a mechanical pressure.

** Valves and transistors in their various functions perform as variable resistors; see Part
3: Electronic Amplifiers.

In some branches of engineering resistance itself is the end product. There can be few homes
which do not possess a physically-large resistor designed deliberately to get hot without self-
destruction and known variously as an electric fire, a kettle element, an immersion heater, a
light bulb … ?

1.6 ELECTROMAGNETISM

** Under 1.3.4 Mechanical Generators it was stated that an emf drives an electric current
through a conducting material but there are also other effects of applying an emf. Fig. 3shows a
length of conductor connected across a cell via a switch. It was probably discovered by accident
but, when a small compass needle is placed beside the conductor, the needle is seen to move
whenever the switch is closed and to return to its original position when the switch is opened
once more.
 ** A compass needle possesses the property of magnetism which means that it exerts
a magnetic force on certain materials the most common of which is the element Iron. The exact
nature and origin of the force are not known.

A compass needle has at one end a North Pole and at the other end a South Pole; the North
Pole is named because it points toward the North Pole of the Earth and this indicates that the
Earth too is a magnet. There is a natural rule that unlike forces attract while like forces mutually
repel (as described with electric charges) and so the North Pole of a compass needle should not
point to the North Pole of the Earth.

The confusion arises because the Earth's "North Pole" should properly be called the "North
Geographic Pole" and the north pole of the compass needle is properly called the "North-seeking
Pole"; the Earth's north magnetic pole is close to its South Geographic pole.

** To return to Fig. 3 that the compass needle moves when the switch is closed, shows that an
electric current calls into being a magnetic force and that this dies away again when the current
ceases to flow. A magnetic force which is sustained by an electric current is called
an electromagnetic force; apart from its dependence on a current such a force is
indistinguishable from the permanent magnetic force exerted by magnets such as the Earth or a
compass needle. See also under 1.8: Fields of Force and under 1.9: Induction, Inductance and
Inductors.

1.7 CAPACITANCE and CAPACITORS

1.7.1 Electron-flow in a Capacitor

** Capacitance is the ability to store electric charge; sometimes it is a desirable property and
sometimes it is a downright disadvantage. A component manufactured to provide a known or
controllable amount of storage capacity is called a capacitor.

** A capacitor consists of two parallel plates made of conducting material which are separated
by an insulating material referred to as the dielectric; sometimes the dielectric is just air. Such
an arrangement is shown in Fig. 4 where it is connected to an electric cell.

** As chemical action removes electrons from the positive plate of the cell so electrons
belonging to the attached capacitor plate are attracted and, for a short period, an electric current
flows in the connecting wire. Equally, as the cell deposits electrons on its negative plate, so they
are repelled and move to the other plate of the capacitor; for a similar short period a similar
current flows in this connecting wire also.
 Fig. 4 Charging a Capacitor

** The surplus of electrons accumulated on the negative plate of the capacitor cannot transfer to
the positive plate and so return to the cell because they lack a conducting path and so +ve and -
ve charges build-up on the capacitor plates. When the p d. between the capacitor plates equals
the p.d. between the cell terminals then electrons can no longer be forced either into or out-of
the capacitor and so the action ceases. This process is referred to as charging the capacitor.

** In the equilibrium condition, with the current in each connecting wire at zero, the capacitor
is said to be fully-charged but it is a purely relative term. Should a larger-voltage battery be
substituted for the single cell shown then the equilibrium would be destroyed and so the
charging process must recommence.

An instrument built to record the passage of an electric current (an amp meter) is known as
an ammeter; such an instrument set in either of the connecting wires would, during the charging
process, record the passage of a direct current which started at some value and decayed to zero.

** The real usefulness of a capacitor is revealed however when the cell or battery is removed
and an alternator substituted. While the capacitor does its best to acquire a charge it is
continually frustrated because the alternator constantly reverses the polarity; electrons are first
pushed into the capacitor, then withdrawn only to be pushed back again and again with each
successive cycle of the alternator. The ammeter perceives only a stream of electrons which
moves first one way and then the other and so it dutifully records the passage of an alternating
current. Note that electrons do not circulate as with a d.c.

** A capacitor therefore has a very-important property in that it blocks the passage of dc (after
an initial charging current) but apparently permits free passage to ac.

The oppositely-charged plates of a capacitor exhibit a strong mutual attraction as do the plates
of a cell; an instrument for measuring voltage can be made by allowing one plate to move
against the return-action of a spring.
1.7.2 Dimensions of a Capacitor

When the plates of a capacitor are allowed to approach each other the opposing charges
which they carry tend more and more to cancel; the resulting destruction of equilibrium means
that the cell will re-commence charging the capacitor. The effect of reducing the gap between
the plates of a capacitor therefore is that it can accept a greater charge; i.e. reducing the gap
increases the capacitance.

More electrons can be stored on the surface of a large plate than is possible on a small plate
and so capacitance (the ability to store charge) is dependent also on the surface area of the
plates; the greater that area the greater is the capacitance. The shape of the plates is not relevant.

That capacitance depends only on the area of the plates and on their separation means that
capacitance can exist between any surfaces which have sufficient conduction to allow electrons
to spread over their total area.

** Capacitance is measured in units named after Faraday and called Farads; however a
capacitor that offers but a single Farad is about the size of a normal room and practical
capacitors are measured in microfarads which are one-millionth of a Farad (mathematically 10-
6
Farads). The practical range of capacitors goes down yet again by a million to what were once
called micro-microfarads but which are now named picofarads (10-12 Farads); pF's are
normally referred to as "puffs".

1.7.3 Stray Capacitance

** It was stated in the previous Section that capacitance exists between any conducting
surfaces; such all-pervading and unavoidable capacitances are referred to as stray
capacitances or simply as strays. The alternating currents which these allow to leak in all
directions are a curse. Experience and careful design can keep their values to a few picofarads
and, in general, they do not become a serious nuisance until the frequency range approaches a
few MHz.

** There are four ways of dealing with stray capacitance once it has been reduced to a
minimum:

(a) make use of it by incorporating it into the design (see Distributed Amplifiers)

(b) arrange the circuitry so that currents introduced through stray couplings cancel; see under
1.10: Transformers,

(c) screening (shielding) techniques which are described in the next Section

(d) deliberately introduce a similar current which cancels the flow through the stray
coupling; this technique is referred to as neutralising and is described in Part
4.2: Transmitters.

1.7.4 Static Electricity

In Lesson-1 it was proposed that an electric current consisted of a drift of free electrons
which are propelled away from a negatively-charged electrode (or terminal) toward one that is
positively-charged . In this Lesson it was proposed also that such a drift of electrons could
flow "into" a capacitor ; i.e. that free electrons would accumulate on one plate of the capacitor
and that a dearth of electrons (a collection of "holes") would accumulate on the other plate.

When the source of e.m.f. is disconnected from such a charged capacitor then clearly the
electrons must tend to flow from the negatively-charged plate back to the positively-charged
plate but this is not possible because they lack a conducting path along which to create that flow.

The result of such actions is a capacitor which has:

(a) a surplus of electrons on one plate

(b) a dearth of electrons on the other plate
(c) an electric field in existence between the two plates (there is an electric force between the
plates)
(d) a mechanic force which tends to pull the two plates together.

Clearly the capacitor is involved with electricity but , even though the electrons are free
from their parent atoms, they are not in the form defined in that they are not drifting.

A stream of drifting electrons are mobile electrons or dynamic electrons and the electricity
which is ascribed to their action can be described as dynamic electricity. Electrons which are
sitting on one plate of a capacitor - electrons which are being stored in a capacitor - form a
standing/stationary charge and are referred to as static electricity.

Perhaps the most noticeable effect of static electricity is the mechanic force which it
manifests. A useful party trick to amuse children while the jelly and cream is awaited is to rub a
balloon on a woolen Jersey or some suit materials and then "stick" it to the ceiling. Provided
the room is dry then friction between the cloth and balloon will transfer electrons; one of them
acquires a positive charge and the other a negative charge. As the balloon approaches the (dry)
ceiling so it induces an opposite charge on the surface and the mechanic force between the two
charges is capable of supporting the weight of the balloon.

Why the emphasis on the word "dry" ? Water is not a very good conductor of electricity but
it is sufficiently good to allow small static charges to leak away. Hence, in a damp atmosphere,
the balloon's surface is slightly moistened and any charge induced on the balloon may well
return to the cloth as fast as it is formed. Alternatively it may leak away via the performer's
hand and finally it must leak away across the ceiling so that the balloon remains in place for a
disappointingly short period.

In a warm and dry room friction effects between clothing, furnishings, shoes, carpets and the
air as we move about generate static charges of surprisingly large values; they can be as high as
1,500 volts. Such high voltages can be lethal - indeed under the right circumstances 50-volts
can kill - but normally we are totally unaware of these static charges. Most people however
have experienced the crackle as they remove clothing or comb their hair which is generally
described as "static"?

The capacitor which stores these static charges is in fact our own bodies. Perhaps, to be more
accurate, it is the capacitance between our bodies and surrounding objects. This stray
capacitance has a very small value and so it requires but few electrons to charge it to high
values. The transfer of such small amounts of charge represents but a very-small electric current
and so we remain unaware of their presence - mostly !

One area in which these small charge-transfers become important even to the point of
disaster is when handling semiconductor devices of the type described in later Lessons
as CMOS (complementary metal oxide semiconductor). These employ insulating layers of glass
a few nanometres thick (virtually a few molecules in thickness) and this insulation is easily
punctured by the static discharge. CMOS devices are destroyed simply by grasping them in the
hand unless precautions are taken to discharge the static electricity. Briefly this involves
ensuring that the human operator, tools, bench surface, CMOS devices and the intended
destination board are all at the same potential (most conveniently Earth potential).

It is a dangerous practice however to connect yourself to Earth because an inadvertent

contact with a point of high-potential could result in electrocution. So-called anti-static
precautions involve connecting everything to earth via a high-resistance which, while allowing
static charges to leak away, limits accidental currents to safe values.

Why do static charges produce visible flashes and audible crackles when clothing is
removed or a comb is passed through the hair? The ability of a capacitor to store charge (its
capacitance) is proportional to the area of its plates . As those plates are moved closer together,
so the opposite charges tend to cancel and so the potential across that capacitor decreases. By
the same argument, as the plates are pulled apart, so the cancellation effect is decreased and the
potential across the capacitor increases.

In removing an item of clothing the gap between cloth and body is increased and so the
potential-difference between them is increased ; the same effect occurs as a comb is withdrawn
from the hair. The p.d. rises until the dielectric - the air - breaks down (it dissociates into
positive arid negative ions) and the capacitor is discharged by a spark. The spark causes the
flash; its supersonic passage

through the air generates a shock-wave which is the crackle. This is traditional thunder-and-
lightning on a microscopic scale.

These effects are most noticeable in homes which use central-heating which causes air to be
re-circulated and so rendered dry. Where rooms are heated by open coal fires air is constantly
vented through a chimney so that fresh air is drawn in from outside; this raises the humidity and
lessens both the static charges that accumulate and unpleasant bronchial effects.

Static electricity has some intriguing habits but the subject is not examined for the R.A.E.
Here are three curios for the thoughtful:

1. Static charges always reside on the outer surface of a hollow body. Hence it is possible to
generate very large static voltages between two hollow metal spheres, with thunderous
discharges occurring between them, while machinery and operators sit safely within.

2. An early capacitor, the Leyden Jar, consisted of two metal cups one inside the other separated
by a glass cup (or jar). Once charged the Jar can be disassembled when tests will fail to reveal
any sign of charge. Reassembled (with care) the jar is found to be still charged
3. The handrails/bars on public transport vehicles are bound with cord where passengers are
likely to grip them. This is because friction between the moving vehicle and the air charges the
bodywork and provides a nasty surprise for indignant would-be boarders. Similarly friction
between clothing and leather/plastic upholstery can charge passengers and drivers of motor cars
who receive similar unpleasant surprises when a foot touches the ground. To a large extent these
effects are now nullified by including an amount of graphite in the tyre composition to provide a
static leak.

In connection with (1) above it should be noted that dangling leather straps and chains can
do little to relieve the car-sickness of passengers which is often attributed to these static charges.
Such charges - if present - will all be found on the outside surface of the car and so cannot affect
the passengers. A leak from car-body to ground will not discharge the electric field between
passenger and seat; the cure is to change the upholstery material

END OF LESSON 2

* * * * * * * * * * * *

QUESTIONS

1. Describe the basic difference in electron behaviour when

(a) a direct current flows through a resistor

(b) an alternating current flows through a capacitor.

In what particular way is a capacitor useful in electronic circuits ?

2. What is the basic difference between conductors, insulators and semi-conductors?

3. Insulators are given to sudden failure under excessive voltage; what is the reason for the
destruction which usually follows ?

4. Why are Resistors manufactured in different sizes ?

5. What is an electromagnetic force

6. Given two capacitors of different physical sizes and apparently of similar construction which
would you expect to exhibit the larger capacitance ? Explain why.

7. What is the difference between a microfarad and a "puff". Why do we not encounter
capacitors that offer 1 Farad ?

8. What do you understand by the term stray capacitances ?

What is the principle that lies behind methods of eliminating their effects?

Back to Top of Page

LESSON 3
1.7.4 Time Constant
 ** To charge or to discharge a capacitor involves either depositing electrons or removing
electrons from it and this takes time. The amount of time taken is determined by the current
which flows because this is a measure of the rate at which the electrons move. In turn the
magnitude of the current is determined by the resistance of the complete circuit.

** Fig.4(A) shows a capacitor connected via a

resistor to a source of emf. At the instant the
emf is connected the capacitor is without charge
and so its terminal voltage is zero; this means
that all the emf must appear across the resistor
and so the current which flows is determined
only by the values of the emf (E) and the
resistance (R). The current I is given by
E/R(Ohm's Law).

For the capacitor/resistor arrangement the

time required at the initial rate is given by the
product RC and so, as a practical rule-of-thumb,
a full charge or discharge takes about 5RC
seconds where R is in Ohms and C is in Farads.
The quantity RC is known as the circuit Time
Constant and it appears in any circuit where
elapsed time is important.

Back to Top of Page

1.7.5. Screening and Earthing

** Stray capacitive couplings (and the stray magnetic couplings described under 1.9) which
occur between a signal circuit and any other piece of electrical equipment lead to the
introduction of unwanted noise signals and also to the creation of unwanted feedback loops (see
under 1.11: Amplification). Stray couplings are a fundamental part of Nature and cannot be
avoided but the signal currents which they allow to pass can be diverted and rendered harmless.

** The great mass of the Earth can be regarded as an "electron sink"; to pass unwanted
electrons to the Earth or to acquire needed electrons from the Earth does not change its potential
in practical terms. Thus any conductor which is connected electrically to the Earth is held at a
constant potential.
 Fig. 5 Mechanism of Screening Against Capacitive Couplings

Screening consists of placing an earthed piece of metal (or metal mesh) between the
circuit to be protected and the source of interference. As shown in Fig. 5 the stray coupling is
divided into two parts: (a) stray capacitance between the circuit and the screen and (b) stray
capacitance between the screen and the interfering source. Of itself the screen provides little or
no protection because currents from the interfering source produce noise potentials on the screen
and these potentials on the screen then act as a noise-source in their own right to drive noise
currents through the circuit.

** Earthing the screen provides a path to earth for noise currents which flow toward the signal
circuit and for signal currents which flow from it. Because the screen is now held at a constant
potential it can no longer act as a source to relay unwanted signals either into or out-of the signal
circuit.

** As a general precaution against stray capacitive couplings it is usual to earth all signal
circuits at some point thus ensuring that they carry only the wanted signal potentials. When one
of the circuit connections (one "leg") is earthed, which means that signal information can be
carried only on the other leg, the circuit is said to be unbalanced; greater protection from stray
couplings is provided by balanced circuits which are described under 1.10.3: Transformers.

** Where voltages in excess of 50 volts are present earthing is employed as a matter of safety;
equipment is enclosed either in insulated cases or in metal cases and these last must be
efficiently earthed if they are to provide protection. Earth connections must not be removed
even when fault-finding or carrying-out maintenance routines unless the Operator thoroughly
understands the procedure; an unearthed metal case is coupled by stray capacitance to the
equipment inside and is only too willing to use the unwary as a path to Earth.

1.7.6 Capacitive Reactance

 ** Although a capacitor does not provide a conducting path between its two connecting wires
those two connections always carry identical currents. A capacitor thus provides a means of
passing information in the form of an alternating current between two points which are not
electrically connected.

** The magnitude (the size) of this current is a matter of the number of electrons which move
into and out of the capacitor and is determined therefore by (a) the magnitude of the charging
emf and (b) the capacitance (the storage capacity) of the capacitor. The capacitance-value thus
controls the magnitude of an alternating current in the same manner as does a resistor.

** However there is a significant difference between the alternating current which flows
through a capacitor and that which flows through a resistor. When a capacitor is fully charged
(when the p.d. between its terminals equals the charging emf) then the current falls to zero; i. e.
the current reaches a zero value when the p.d. reaches its maximum value.

** When a capacitor is in the middle of the operation of discharging and then recharging in the
opposite direction then the current (the flow of electrons) is at its maximum value but, at that
moment, the p.d. across the capacitor is zero as it reverses polarity; i.e. the current reaches a
maximum value when the p.d. is zero.

** Where the ac applied to the capacitor is of sine-waveform (see Part 7: Mathematics) the
waveforms of voltage and current are always displaced by 900 and are said to be in quadrature;
the effect of this is that the power-flow (the product of voltage and current) is zero.

** In fact a capacitor can pass fairly-large currents without getting warm while a resistor
handling the same current could be destroyed by the heat generated. For this reason ac which
flows in a purely capacitive circuit is often referred to as wattless current (power-less current).
To distinguish this wattless current-controlling action of a capacitor from the power-dissipating
action of a resistor it is given the name Reactance but it is still measured in ohms; with certain
provisos it obeys Ohm's Law as does resistance.

Back to Top of Page

** Because a resistive current and a capacitive current have different phase relative to the emf
which drives them it is not possible to add the two values directly and this is dealt with
under 1.12: Resistance, Reactance and Impedance.

Similar behaviour is exhibited by inductors (see under 1.9) but they produce a different
quadrature relation between current and voltage; to distinguish between them the two reactances
are described by the terms capacitive reactance and inductive reactance.

1.8 FIELDS OF FORCE

In the domestic world a field is an area (2-dimensions) characterised by grass although

sometimes it may be specified as (for example) "a field of corn" or "a cornfield".

In Physics a field is a volume (3-dimensions namely length, breadth and height) throughout
which a force acts; it may be specified as an electric field (which surrounds an electric charge
or fills the space between electric charges), as a magnetic field (which fills the space between
and around magnetic poles) or perhaps as a gravitation field (about which we know very little).

Fields of force are represented graphically by drawing lines-of-force; these have the
following purposes

(a) as with the contour lines which are drawn on maps (they join points of equal height) lines-of-
force join points of equal field-strength.

(b) at each point on a line-of-force the line indicates the direction in which the force acts; for
example, an electron placed in an electric field will be attracted to the positive pole and repelled
by the negative pole and a line-of-force which passes through that electron shows the direction
of the resultant force (the overall effect of these two forces when they act simultaneously).

(c) many lines-of-force drawn close together are used to indicate a powerful field; this arises
from their role as contour lines in that they indicate the gradient of the field (the way in which
the field-strength changes over a distance) although it is seldom that lines-of-force are allocated
specific values.

** In an electric field the lines-of-force are defined as showing the path that would be traveled
by an isolated positive charge; it follows therefore that electric force is deemed to act from the
positive pole to the negative pole.

** In a magnetic field the lines-of-force are defined as showing the path that would be traveled
by an isolated North Pole; in fact we cannot achieve an isolated magnetic Pole but the concept
gives rise to the convention that magnetic force acts from the North Pole to the South Pole. (An
isolated Pole can be approximated by using a very-long bar or rod magnet.)

Back to Top of Page

There is one very important difference between electric lines-of-force and magnetic lines-of-
force; see Fig. 6. Electric forces always form open lines which terminate at one end on a
positive pole and at the other end on a negative pole; magnetic lines always form closed loops
which enter the magnetic source at the South Pole and continue through the source to exit at the
North Pole.
Any attempt to close an electric line-of-force into a loop, or any attempt to open a magnetic
loop, invariably decreases the magnitude of the force concerned.

Lines-of-force often show graphically the behaviour of a system. For example Fig.7 shows in
diagram (a) the electric field between a negative pole and a positive pole while, in diagram (b),
it shows the field between two opposing positive poles. Diagram (a) is an ordered pleasing
pattern which is believably trying to contract while diagram (b) is angular and the field patterns
associated with each pole do appear to be pushing against each other.

END OF LESSON 3

* * * * * * * * * * * *

QUESTIONS

1. Potential difference (p.d.) and electromotive force (emf) are both measured in volts. In
what manner do they differ?
2. What is the property known as Reactance?

What do you understand by the term "wattless current ?

3. Describe the principle used in screening electrical circuits against intruding electric
fields.

4. Why is it that resistive and reactive currents cannot be added directly using simple
arithmetic ?

5. What is an electric field ?

6. What would you deduce from a diagram that showed many lines of electric force close
together ?

7. Why is it that the terminal voltage of a generator must fall before the generator is able to
supply current ?
(Not needed for R .A. E.)

8. What is the difference between a battery and a cell ?

9. What is the effect of connecting a conducting body to Earth ?

** Paragraphs marked with a double asterisk contain material which is relevant to the
RAE;
the remainder form a continuous explanation.

Back to Top of Page

LESSON 4
1.9 INDUCTION, INDUCTANCE AND INDUCTORS

1.9.1 Induction

** Under 1.6: Electromagnetism with reference to Fig.3 it was explained that an electric
current generates a magnetic force. By exploring with a small compass needle it can be
established that the force acts in a circular path that surrounds, and is concentric with, the
conductor; the magnetic field has the form of a cylinder pierced from end to end by the
conductor.

** The direction of the circular lines-of-force is related to the direction of current-flow and this
is the subject of a classical Law which states that, when an Observer looks along the conductor
in the direction of classical current-flow (i.e. from positive to negative), then the magnetic force
is always seen to act clockwise.
** Fig.8 shows a development of Fig. 3 in which a second conductor has been introduced close
to the first (or primary) conductor so that it lies within the magnetic field; this secondary
conductor is connected to an ammeter.

** It is an experimental fact which we cannot explain that, at the moment the switch is closed, a
current flows for a short period in the secondary conductor. The secondary conductor then
remains quiescent until the switch is opened when again a current flows but this time in the
opposite direction.

** That the secondary current-flow occurs whenever the primary current changes its value can
be demonstrated by substituting an alternator for the cell; the primary current now changes
value continuously and a constantly-changing current of similar form is found to flow in the
secondary conductor.

This somewhat mysterious transfer of electrical power from the primary circuit to the
secondary circuit is termed electro-magnetic induction; it is the basis of an important
component called a transformer which is described in the next Section, but the mechanism of
the power-transfer is not known.

The current that is induced in the secondary circuit is exactly the same as any other current
and so it too brings into being an attendant magnetic field; like the primary field this magnetic
field too surrounds both conductors and the two are said to be magnetically coupled. A
question arises as to whether the secondary field aids or opposes the primary magnetic field?

To the ammeter the secondary conductor is a generator and so there must be a fall in
terminal voltage (the p.d. across the ammeter) before current can flow and this fall must be
dependent on the strength of the current-flow; see under 1.3.3. Because the secondary
conductor derives it power from the primary by means of the magnetic field such a fall in
terminal voltage must be caused by a fall in the magnetic-field strength. It follows that the
magnetic field generated by the secondary current must oppose the magnetic field generated by
the primary current. Therefore, as drawn inFig. 8 the secondary current must flow in a direction
opposite to that of the primary current.

1.9.2 Self-Inductance

The above argument can be pursued however without introducing the secondary conductor.
When the primary current creates a magnetic field the primary conductor itself is encompassed
by that field; the primary conductor therefore is in exactly the same situation as any secondary
conductor. Hence, when the primary current varies, an opposing secondary current should
appear in that primary conductor.

** Pursued to the limit this argument implies that any conductor will carry a somewhat
smaller ac than would be expected from its performance with dc ?

By definition a direct current does not change its direction of flow but it does often change its
value especially when either building from zero at switch-on or collapsing to zero at switch-off.
If the above arguments are valid it is to be expected that, even with dc, these switch-on and
switch-off periods are modified by the inter-action between a conductor and its surrounding
primary magnetic field.

** In practice all these effects are observed; in any given conductor, for a given emf, the value
of alternating current is less than that of direct current. Because of this reduced-current effect it
takes an appreciable period of time for current to reach its final value at switch-on and to fall to
zero after switch-off.

** This phenomenon is given the name Inductance. It is not difficult to believe that
inductance can be a nuisance but, as described in a later Lesson, it can also be of use and so
components are manufactured to provide a specified amount of inductance. These are
called Inductors and the amount of inductance provided is measured in units called Henrys (in
honour of Henry); note the spelling.

**The reduced value of alternating currents with respect to direct currents is interpreted as an
increase in resistance and, quite reasonably, it is measured in ohms. However, as with
capacitors, it is found that the alternating current does not vary exactly in time with the applied
emf and so the apparent "extra resistance" of an inductor is referred to as its inductive
reactance (see also under 1.7.5: Capacitive Reactance).

** Again, because of the quadrature relation between current and voltage, an inductor draws a
wattless current and provides an efficient means of reducing a

current-flow. In this application inductors are less efficient than capacitors but they are often
used where it is required to preserve a dc-connection between terminals.

** The inductive reactance of a simple conductor is fairly small and is not noticed in practice
until the frequency of the alternating current rises well above 1 MHz (1-million cycles per
second). To achieve appreciable values of inductance it is necessary to use long lengths of
conductor and, for convenience, such lengths are coiled. Fortunately the effect of coiling is to
increase the inductance still further; as shown in Fig. 9, the magnetic field which surrounds each
turn of the coil reinforces the magnetic fields of all the other turns and the effective increase in
this field enhances the self-inductance.
 **Note that by winding half the turns in the opposite direction the fields can be made to cancel;
if, by careful construction, the magnetic field is reduced to zero, then all that remains to control
current-flow is the conductor's resistance. A resistor constructed in this manner is referred to as
a non-inductive resistor. These are found for example in test meters (multi-meters) which are
required to work equally with ac as with dc.

Inductors are often referred to as "coils" (for obvious reasons). Because they are used to
oppose or "choke-off" a flow of alternating current, they are also called chokes.

** Note that the standard textbook approach to Inductors (and the Transformers described
later) is that it is an emf which is induced into the secondary conductor and that this then drives
an alternating current through any connected load. Investigation into this theory shows that such
an emf is proportional to the rate at which the primary current changes its value; in particular
that the more rapidly the primary current changes its value the greater is the value of the induced
secondary emf.

**Clearly if the number of current-reversals per second increases (the frequency rises) then the
current must change at a greater rate; thus the secondary emf increases with increasing
frequency. Therefore, as far as Inductors are concerned, (inductive-) reactance increases with
increasing frequency; see under 1.12: Resistance, Reactance, Impedance, Q-factor.

** Note that the physical property of Inductance, the degree to which the conductor reacts with
its own surrounding magnetic field, does not change with frequency; this is a function of the
number of turns, the shape and spacing of those turns and the magnetic properties of the core
material (if any); see below.

**The important characteristic of an inductor is its resistance to change of current-flow. Any

attempt to increase the current calls into being a counter-current which mitigates that increase;
any attempt to reduce the current calls into being a counter-current which mitigates that
reduction. The inductor is the electrical equivalent of the mechanical flywheel in that it tries to
maintain current-flow at a constant (the present) value. It is in this role that it appears in
the ripple filters which are described in Part 5 as a necessary part of Power Suppliers; it plays a
flywheel-like part also in the action of Transmission Lines which are described in Part 2 (under
2.3).

1.9.3 Magnetic Cores

 ** The magnetic field which surrounds an inductor, and hence the inductance, can be increased
still further by introducing a magnetic material into the magnetic field (the path of the lines-of-
force). A magnetic material is one which, under the influence of a magnetising force, will itself
become magnetised and, of course, it adds its resulting magnetic field to the surrounding
magnetic field. In turn this must increase the induced secondary current.

So-called soft magnetic materials lose most of their induced magnetism when the magnetising
force is removed; hard magnetic materials retain most of the induced magnetism and so
become permanent magnets. When soft magnetic materials are placed in the varying field of an
alternating current their magnetic field-strength varies with the waveform of the alternating
current and so they add an alternating effect which increases the inductance. This argument
applies equally to a dc that is varying in value; the inductor offers no extra "resistance" to the dc
but the presence of a magnetic material greatly increases its "flywheel" opposition to the current
variations.

** The usual material used to enhance inductance in this manner is iron which results in the
name iron-cored chokes. There are other materials however some of which contain iron in
finely-powdered form but others are specially prepared ceramics.

** Air-cored inductors are those without additional magnetic materials but others may be
wound directly on to the magnetic core, on a bobbin which is then fitted on to a magnetic core
or wound on a special former into which a specially-formed core can be screwed a variable
inductor).

** Magnetic cores have a limitation however in that they magnetically saturate. Although the
magnetic field which surrounds a current-path is always proportional to the current-flow,
magnetic materials perform only up to a critical value after which their magnetism ceases to
increase further. Saturation does not take place abruptly but, as the magnetic field fails to follow
the variations in current, so the performance of any inductive device suffers.

The chokes used in ripple filters (see Part 5: Power Suppliers) carry the direct current which
is supplied to the associated equipment and this adds its quota of magnetising force; as a result
there is a risk that the iron core will saturate at the peaks of the ripple current. This is avoided by
creating a small air-gap in the magnetic path through the core which increases the magnetising
force required to cause saturation; it also reduces the effective inductance but usually this is not
a problem. Such components are referred to as gapped chokes.

** A more serious problem with iron-cored inductors arises from the fact that the core is made
from conducting material which is magnetically coupled to the current-carrying coil; as a
consequence the core acts like a secondary winding and eddy currents appear in it. These
currents can have very large values because the core is both a secondary winding and its own
short-circuiting connection. Eddy currents are destructive because they reduce the effective
inductance and also dissipate signal energy in heating the core.

The cure is to raise the resistance of the core material so reducing the value of the eddy
currents. This is achieved by dividing the core into thin slices or laminations (see under 1.12.1:
Ohm's-Law). Laminations are assembled in various ways but the finished device needs to be
clamped firmly to eliminate air-gaps and to prevent them vibrating in sympathy with the
alternating magnetic field. Core material which uses finely-divided iron uses a non-conducting
binder material to reduce eddy-current losses.
1.9.4 Skin Effect

** The magnetic field which surrounds the path of an electric current is strongest at its centre
and diminishes with increasing distance from that centre. Because the generation of a secondary
current depends on the strength of the magnetic field it follows that the strength of the
(opposing) secondary current must be strongest at the centre of the field; i.e. it will be strongest
at the centre of the conductor.

** From this it follows that alternating-current does not flow evenly throughout the cross-
section of a conductor; it is weakest in the centre and has its greatest value in the surface layer
of the conductor.

** This tendency for ac to flow only through the surface layers of a conductor is termed skin
effect; it does not occur at all with dc but, as frequency rises, it causes the current to crowd
more and more into an ever thinner layer at the surface. Thus, quite apart from inductive effects,
it is found that the resistance of a conductor increases as frequency rises; in practice this means
that a conductor becomes increasingly lossy.

Skin effect becomes appreciable above 20 kHz (i.e. just above the audio range). Up to about
1 MHz its effects are minimised by using Litz wire which consists of a number of fine insulated
wires interwoven so that small individual currents are forced to flow in all parts of the
conductor's cross-section. At higher frequencies the technique is rendered useless by stray
capacitance between the fibres which, as it were, allows current to flow radially and so negates
the effect of the Litz wire.

For high-frequency working thick conductors are used so as to obtain the maximum possible
surface area and often they are silver-plated because silver does not so readily oxidise in air.
Unplated copper conductors become coated with a high-resistance layer of oxide into which skin
effect crowds the current.

1.9.5 Core Material for HF Inductors

Increasing losses limit the use of soft-iron cores to the audio band. From around 20 kHz to
a few MHz dust-iron cores are used; as implied by the name these consist of finely-divided iron
particles held in an insulating binder and molded to required shapes (this is an extreme form of
lamination). "Ferrite", an iron-carbon alloy, is used in a similar manner. Such moldings may
take the form of pot-cores which, when assembled, completely enclose the component thus
acting as both magnetic core and magnetic screen (see below).

Ferrite is also molded into small rings or annuli (singular annulus) which can be slipped over
a conductor. They are made in various grades; some are intended to increase the inductance but
are useful also in making small current-transformers

see next Section) while others, by increasing the losses, are intended to dissipate the power of
unwanted (spurious) signals.

At very-high frequencies, and also for high-power working, core material has to be
abandoned. Coils are wound either on insulating formers or are made self-supporting by using
thick wire; they are usually referred to as air-cored coils.
1.9.6 Magnetic Screening

Under 1.7.4 it was shown that screening against electric fields required the interposition of a
metallic sheet or screen between the circuit to be protected and any source of interference and
also that the screen must be earthed.

** Magnetic screening also requires the interposition of a conducting sheet but, although it
does not need to be earthed, it must be made from a material with good conductivity, it must
completely surround the circuit to be protected and it must be continuous (i.e. either seamless or
with all seams soldered). The purpose is to ensure that eddy-currents can flow freely in the
screen and produce magnetic fields which cancel any ambient fields. As with electric screening
this works in both directions; magnetic fields set up by currents in the protected circuit are
cancelled outside the screen and, simultaneously, intruding fields from outside the screen are
cancelled within.

** It is usual however to earth a screening can because it then acts as both a magnetic and an
electric screen. The canceling action of the eddy-currents' magnetic fields reduces the
inductance of an enclosed circuit and so circuit adjustments must always be made with the
screen in place.

A mesh screen provides very good electric screening but it allows magnetic fields to
penetrate. Such a Faraday Screen enables the effects of stray capacitance to be removed while
still permitting magnetic coupling between two circuits. Such a device is to be found in the
output stage of some transmitters where a balanced tank-circuit coil is coupled to an unbalanced
output circuit; (see next Section on Transformers and in Part 4.2: Transmitters).

1.9.7 Variable Inductors

The inductance of air-cored coils is adjusted by adding or removing turns but, for fine
trimming, the coils are either stretched or compressed; this is because the inductance of a coil
depends not only on the number and size of the turns but also on the "form-factor" which is the
ratio between the length and the diameter of the coil.

Iron-cored inductors are not so easy to adjust but, where the core is fitted with a gap, the
inductance can be adjusted laboriously by fitting shims of differing thickness in the gap and then
re-clamping. Sometimes a sliding "slug" of magnetic material is fitted over a too-big gap where
it forms an adjustable magnetic short-circuit that partially removes the effect of the gap.

Small-signal high-frequency inductors are wound on special formers with hollow centres
within which a screw-thread is formed. Ferrite or dust-iron cores are molded in the form of a
threaded rod so that, as the core is screwed into the coil-former, so the inductance of the coil
increases. (It must be remembered that, as the core passes the centre of the coil, so the
inductance falls again; this peak is often mistaken for a tuning point.)

By substituting a silver-plated copper core the inductance of a coil can be reduced as the
core is screwed-in; this technique is another example of eddy-currents being used to cancel a
magnetic field.

END OF LESSON 4
 * * * * * * * * * * * *

QUESTIONS

1. What is the property known as reactance ?

What do you understand by the term "wattless current" ?

2. How does electromagnetism differ from other forms of magnetism?

A conductor is connected to a source of electricity and is in close proximity to a second
conductor. When would you expect to draw current from the second conductor ?

3. What is "skin effect" ?

4. Describe the principle used to protect electrical circuits from the effects of intruding magnetic
fields.
In what ways does this resemble and differ from the technique used to protect against
intruding electric fields ?

5. Why are coils often wound on magnetic cores ? When would you use a copper core? How
does such a core work ?

6. How does (a) inductance and (b) inductive reactance change with frequency?

Paragraphs marked with a double asterisk contain material which is relevant to the RAE;
the remainder form a continuous explanation.

Back to Top of Page

LESSON 5
1.10 THE TRANSFORMER

1.10.1 Turns Ratio

In the previous Section it was shown that, when a primary current varies, secondary
conductors to which it is magnetically-coupled carry alternating currents whose waveforms copy
the variations. Each of these secondary currents flows in a direction which causes their
associated magnetic fields to oppose the changes in the primary field.

** On the assumption that each secondary conductor is coupled to the primary conductor with
equal efficiency they should all carry identical currents and so it is possible to connect all
secondary conductors in series. Fig. 10 shows this arrangement in the diagrammatic form which
is normally used although here the diagram is being taken more literally than is normal.
 Each primary and secondary "conductor" referred to above is shown as a single turn of wire
wound around a common iron core which is indicated by the solid line drawn between the
primary and secondary coils. The purpose of the core is to concentrate the magnetic field to
ensure that all fields pass through all turns of wire a condition which is termed maximum
coupling between the conductors.

The designations primary and secondary are ours and are introduced to identify the different
coils; as far as the transformer is concerned they are all just lengths of wire wound around the
core and so it is to be expected that, irrespective of which turn actually carries the primary
current, ALL TURNS carry equal secondary currents.

** If the turns are equal in every way and they carry equal currents then it follows that equal
p.d.s must be developed across them. Thus, if the primary consists of a single turn and a
secondary consists say of three turns, it is to be expected that the p.d. developed across the
secondary is three-times that which appears across the primary?

** There is a general rule therefore that the ratio

secondary p.d./primary p.d.

must equal the ratio

no. of turns on secondary winding/no. of turns on primary winding.

This is always referred to as the turns-ratio of a transformer and it is usual to quote it as

Output/Input as given above; sometimes, either for clarity or convenience, transformers are
referred to as step-up or step-down transformers which is intended to show the manner in which
the voltage changes as the signal progresses from primary to secondary.

The above is a very simplified approach to the transformer and full treatments in every
degree of detail are given in many textbooks. The discussion here is sufficient for the purpose
of the R.A.E.

As shown already the magnetic fields of secondary currents tend to cancel the magnetic field
of the primary current. In a transformer where the secondary winding has more turns than the
primary winding there is an obvious difficulty in that the sum of the secondary fields is likely to
either obliterate the primary field (transformer action would cease) or even reverse it. In that a
transformer works it has to be concluded that, as the turns ratio is increased (and the voltage
ratio also is increased), so the current ratio between secondary and primary must decrease.

** In Lesson-1 under 1.3.2: Work and Power it is stated that the power dissipated in an
electrical circuit is obtained by multiplying the voltage by the current. A transformer is an
inactive device and does not have any means of generating additional power; it does not get
particularly hot and so it does not dissipate any appreciable power. If the power-output is to be
equal to the power-input then the product of volts and amps in the primary circuit must equal the
product of volts and amps in the secondary circuit.

** A most important aspect of transformer behaviour therefore is that, as it steps the circuit
voltage either up or down, so it must step the circuit current either down or up in an exactly
inverse ratio. This is written in mathematical terms as follows:

Let the turns-ratio be n ; let the primary voltage be Vp ; let the secondary voltage be Vs ;
let the primary current be Ip ; let the secondary current be Is

THEN n = Vs/Vp = Ip/Is

This matter is referred to again when dealing with Impedance Matching

1.10.2 Balanced Circuits

** Although alternating or varying signals can pass between the windings of a transformer the
lack of a conducting path between those windings means that dc signals are blocked. As a
consequence it is possible for the different windings of a transformer to be at different
potentials. Such a potential difference is not confined to direct potentials and it may be either a
varying direct voltage or an unwanted alternating signal.

Unwanted or noise signals appear in circuits through stray capacitive or magnetic couplings
(see under 1.7.3, 1.7.4, 1.9.6) and it is usual to minimise these couplings by earthing a circuit at
some convenient point.

** The transformer offers a useful technique in the form of the balanced circuit. Fig.11 shows
the two conductors of an electric circuit (perhaps a telephone cable) connected between two
transformers; the input to the first transformer and the output from the second each have one
leg earthed and so, in those parts of the circuit, the signal information is carried only on the one
(non-earthed) wire. Such circuits are said to be unbalanced.
 Fig. 11 Use of Screened Centre-tapped Transformers to Create a Balanced Circuit

The remaining two transformer windings are each centre-tapped and those centre
connections are taken to earth. Signal currents in this part of the circuit are carried
simultaneously on both wires but, at any instant in time, the two currents flow in opposite
directions. Such an arrangement is sometimes said to be operating in push-pull; one leg is
"pushing" the signal while the other is "pulling". Each wire carries the same value of current
and the same voltage but, with respect to earth, the voltages are of opposite polarity and the
circuit is described as balanced.

** There are many reasons for requiring a balanced circuit but, in terms of noise suppression,
it results in stray signals being injected equally into both legs of the circuit. The two noise
signals travel along the balanced cable in the same direction and are sometimes described as
being in push-push. At the delivery end of the balanced circuit the wanted signal (in push-pull)
is transferred from the primary to the secondary winding but the noise signals (in push-push)
flow in opposite directions through the halves of the primary winding and so cancel.

In telephone cables such transformers are known as repeating-coils; they are carefully
balanced and their purpose is to prevent cross-coupling between the many pairs of wires in the
telephone cable.

** Transformers of various kinds which are used to couple balanced circuits to unbalanced
circuits are known as bal to unbal transformers and have acquired the

composite name balun. They incorporate a Faraday Screen (see under 1.9.6) between the two
windings to eliminate stray capacitive couplings that otherwise would destroy their balance; this
must be earthed.

1.10.3 High-frequency Transformers

** In the above discussion transformer windings were said to be wound on an iron core to
ensure efficient coupling between them. However, to magnetise an iron core and to constantly
reverse that magnetisation, consumes power and the consequent hysteresis losses increase as the
frequency increases. Coupled with eddy-current losses this restricts the usefulness of iron-cored
transformers roughly to the audio band; i.e. 50 Hz to 20 kHz.

** The comments on iron-cored Inductors under 1.9.3 apply equally to iron-cored

transformers. For high-frequency working it is usual for one or both transformer windings to be
resonated as described later under Resonance and Bandwidth. As with variable inductors the
inductance of each winding may be adjusted by use of screw-in ferrite or copper cores.
 ** For very-high frequencies and for high-power working air-cored self-supporting windings
are used which can be fine-tuned by either stretching or squeezing the coils; the distance
between windings may also be adjusted to control the degree of coupling between primary and
secondary; see also under Transmitters.

** Note that the coils are self-supporting because they are made from large-diameter wire to
reduce skin effect (see under 1.9.4.).

1.10.4 The Autotransformer

The discussion of Inductance

under 1.9.1 was extended to Self-inductance
under 1.9.2 by eliminating the secondary
winding. The same argument can be applied to
transformer action in that a single winding can
serve both as primary and secondary.

** Fig. 12 shows at (a) a step-down

transformer in which one leg of each winding
has been " commoned". In diagram (b) the
secondary winding has been omitted and the
output is taken from a tapping point at y on the
primary. This arrangement is known as
an autotransformer; diagram (C) shows a
balanced form which is often met in
transmitters.

Because the primary turns carry a current which is common to them all it can be deduced
that the applied emf is divided equally across them. Thus the ratio output-voltage/input-
voltage is given by:

number of turns yz / number of turns xz.

** The autotransformer is suitable only for small step-up or small step-down ratios but it does
provide a saving in the amount of copper required and in the copper-losses in comparison with
an equivalent double-wound transformer. It is necessary however to be aware of its
shortcomings especially when used with mains power supplies:

1. There is a direct connection between input and output circuits.

2. Should a break occur in the common part of the winding then the input emf is applied
directly to the output (another reason for limiting the step-down ratio).

3. Part of the winding must be wound with wire that is thick enough to carry the larger current
while the whole device must be insulated to withstand the larger voltage.
4. A very dangerous situation arises in mains applications if the output is inadvertently taken
from across xy .

** The Autotransformer is used a great deal in aerial-matching circuits and in the final (P.A.)
stages of transmitters. It is also the basis of the "Variac" transformer which supplies a variable-
voltage supply from the mains by making the tapping-point into a sliding contact. When using a
Variac always check its coil for continuity, make sure that the output connections are taken
from the correct points and do not exceed the recommended current loadings.

1.10.5 The Toroidal Transformer

This consists of a small-diameter helix of wire which is wrapped (like a collar) around a
conductor; the conductor forms the primary winding of a transformer and the helical coil forms
the secondary. The magnetic field of the primary follows the same path around the conductor as
does the helix and so they are linked.

The device forms a useful way of measuring ac without interrupting the primary circuit and
is the basis of the many instruments often described as claup meters which are simply clamped
around any conductor in which the current-flow is to be measured. In these meters the coil is
wound in standard fashion on a two-part soft-iron core which forms the jaws of the clamp.

The toroidal transformer, often wound on a toroidal (ring) core, is also the basis of the so-
called bridge arrangements used in standing-wave-ratio meters which are described in Part 7 of
this Course.

A version of the toroidal transformer is sometimes created in which both primary and
secondary windings consist of a single conductor; such a current transformer is achieved by
laying the two wires side by side and threading a suitable ferrite bead over them. It can be a
useful means of checking alternating current-flow where space is more important than absolute
accuracy.

Paragraphs marked with a double asterisk contain material which is relevant to the RAE;
the remainder form a continuous explanation.

1.11 AMPLIFICATION

** Amplification is the technical term for deliberately increasing the strength

(the amplitude) of a signal. The reverse operation, that of starting with a too large signal and
cutting it down to a practical value, is called attenuation.

There are two ways of increasing size:

(i) a snowball set rolling downhill gathers snow and steadily increases its bulk

(ii) a marine architect may first build a scale model of his ship and then, with the problems
under control, set about building the full-scale version; to do this he takes a supply of new
materials and ends up with both the model and the finished ship.

In engineering the process of amplification uses the second method.

 The first steam engines were built in Cornwall to raise flood water from the mines; these
engines did not operate continuously but required constant driving. Basically steam was raised
at high pressure in a boiler and was then allowed to enter a cylinder in which it pushed a piston
to and fro; ultimately the piston movement, with the power of the steam boiler behind it, was
used to raise water to the surface.

Steam was guided through the engine by a man who used the relatively-puny power of his
hands to operate steam-control valves. Thus the result of simple low-power hand movements
was that tons of water were raised through several hundred feet. Such power-controlling work
systems are known as servo systems. It could be said that the power of the steam boiler through
the agency of the steam engine amplified the power of the man; in addition the man provided a
controlling intelligence that, in turn, was guided by signals received from the engine through his
eyes and ears. Modern engines move their own valves but the principle remains.

** An electronic amplifier performs the same function. It requires an auxiliary source of

power from which it fashions a new higher-power output signal; the form of the output signal is
determined by copying the form of a low-power input signal (the control signal). The device
on which the input signal operates to control the power-flow is called an active device because it
actively does something; in the steam-engine system described above the active device was the
steam valve; in the overall water-raising system the active-device was the steam engine. In
electronic amplifiers there are a variety of active devices but the most notable are the thermionic
valve and the transistor which are described in detail in the Lesson on Electronic Amplifiers .

** All amplifying systems, from the engine through horse-and-cart to the domestic Hi-Fi
system, incorporate the basic requirements:

1. A source of power sufficient to meet the required demand

2. An active device to control the flow of power from the auxiliary source

3. A control signal - an input signal -which decides the form of the power-flow by controlling
the action of the active device

4. A source of information which originates the control signal (a generator) ..

** Fig.13 shows the basic layout of an amplifier and Fig. 14 shows a very important
variation. The variation was mooted above when it was stated that modern engines move their
own valves. Some of the work output is tapped-off and fed back to the control-signal to provide
a secondary control of the power-flow. Such feedback control takes many forms and often goes
unrecognised.

For example: at the end of the piston's stroke the steam must be shut-off and an exhaust
valve opened to allow the spent steam to escape as the piston returns for the next power stroke.
A lever system which links the valves and the piston can do this job and so dispense with the
bored engineer and his twiddling hands; properly engineered against mechanical failure this
feedback lever-system will never get bored and so will never make mistakes.

Yet again, without the engineer to watch and control it, the engine might overspeed and so
destroy itself. As shown in in Fig. 14 a sensor can be fitted which measures the speed of the
engine (against a reference which is provided for the purpose) and generates an error signal;
this low-power signal is fed back to control the steam input to the engine in such a manner that it
corrects any tendency to depart from the required speed. This "correct" speed is set by the
reference signal which now becomes the speed control or throttle.

** The kind of feedback system described above, which operates so as to maintain control and
order, is referred to as negative feedback because the error-signal is used to cancel or
to negate any errors. Negative-feedback (n.f.b.) systems at their most simple are exemplified
by the mechanical flywheel which

(a) is difficult to start spinning

(b) once spinning is difficult to stop
(c) has a preferred spinning rate namely the one it already possesses.

In short, flywheels like to preserve the status quo - they dislike change and oppose
acceleration. Indeed n.f.b. systems not only behave like flywheels but, in electromechanical
systems such as video-recorders, they are actually used as flywheels because they can be
switched-off when their conservative nature becomes inconvenient. They are also more compact
and less lethal than mechanical types.

** In the example given above it was the engine speed that was measured and the result was
used to control the engine speed. This circular action gives rise to the technical term "feedback
loop". Note that, whatever is sampled in the measuring process, it is the sampled parameter that
is controlled.

** In electronic equipment the output may take the form of either a voltage waveform (with
negligible ability to supply current) or a current waveform (with negligible ability to develop
voltage); if both voltage and current waveforms are required then it is necessary to use a
power-amplifier. If a feedback sample is derived from the voltage waveform then the voltage
waveform is maintained accurately at the expense if need be of the current waveform. If a
feedback sample is derived from the current waveform then the current waveform is maintained
accurately at the expense if need be of the voltage waveform.
 ** To talk of negative-feedback implies that there is a positive-feedback (p.f.b.) system and it
is so; where the feedback loop is connected so that the error-signal helps an error instead of
negating it then chaos results. Hard perhaps to believe but there are conditions under which
p.f.b. can be helpful; for example it is a most effective way of speeding-up some operations but
the Designer must be careful to keep matters under control. Positive feedback is the essence of
systems which continuously oscillate (such as a clock pendulum or an electronic oscillator );
control is maintained by a system that restricts operation to a narrow band of frequencies (see
Lesson on Electronic Oscillators). In particular most p.f.b. systems are controlled by a n.f.b.
loop.

END OF LESSON 5

************

QUESTIONS

1. What is meant by the "Turns Ratio" of a transformer ? How does this ratio relate: to

(a) the input and output voltages

(b) the primary and secondary currents ?

2. What is the thinking behind a balanced circuit?

3. What is a Balun?

4. In what manner do Audio Transformers and HF Transformers differ?

5. Why would you not use an Autotransformer in the power supply circuits of a piece of
equipment?

6. An amplifier increases the level of a signal and a transformer can be used to step-up the
voltage of a signal. Why is the transformer never referred to as an amplifier?

7. In feedback systems what is it that is fed-back? What is the purpose of such a technique?

8. Can you state briefly the different effects of positive and negative feedback?

Back to Top of Page

LESSON 6
1.12 Resistance, Reactance, Impedance, Q-factor

1.12.1 Resistance and Ohm's Law

In the previous Sections it was shown that conductors have a natural resistance to the flow of
electric current and that, when a current varies, the natural resistance apparently increases
because of inductive reactance.

Although all conductors must exhibit inductive effects there are so-called non-inductive
resistors. One popular method of constructing these is to take a length of wire which has the
required resistance, fold it at the centre and then wind the doubled wire on to a former taking
care that the twinned conductors are kept in close proximity. The idea is that current flows in
opposite directions along the two parts of the wire and so their magnetic fields cancel. Apart
from their awkward bulk such components can be very effective and are often found for
example in multimeters (test-meters) which function over a range of frequencies.

This Section is concerned with true (non-inductive)resistance and so it is applicable only to

circuits which handle dc, very-low-frequency ac or to non-inductive components.

Consider several conductors which are alike in every detail; they are made from the same
material, they have identical widths (W), identical thickness (T) and identical length (L). When
such identical conductors are all connected across the same source of emf it is to be expected
that they will all draw the same current I amps.

Fig.14(a) shows one of these conductors connected across a source of emf; the current I sets-
up a p.d. across the Resistor and that p.d. must be equal to the value of the emf (else the current
increases or decreases until equilibrium is attained).

Fig.14(b & c) shows two graphs; the upper one shows the variation of the applied emf with
time and the lower shows the variation of the p.d. At the moment the switch is closed the
applied emf leaps instantaneously from zero to the value of the cell's emf but the p.d. builds
slowly to this final value. This is because the current is prevented from changing
instantaneously by the delaying effects of both inductance and stray capacitance

In a similar manner, when the switch is opened, the applied emf falls instantaneously to zero
but the current - and therefore the p.d. across the resistor - falls slowly. There is a philosophical
problem here in that, after the switch has been opened, the current has nowhere to flow and so it
should also fall to zero instantaneously. Consider that, for the current to fall at all, a mechanism
is required to discharge both the stray capacitance and the magnetic field. This is usually
explained in terms of the spark that jumps the switch contacts as an inductive circuit is broken.
 ** Fig.15(a.)shows two of the identical
conductors connected across the cell in a
manner that requires the current to flow first
through one and then through the other. The
conductors are said to be connected in series or
to be series-connected. Because the one
current flows through the identical resistors it
must develop the same p.d. across each and so
equilibrium is established when each p.d. is
equal to half the applied emf. It follows that the
current which flows through two identical
resistors in series must be half that which flows
through one of them when connected to the
same emf (Fig.14).

The exercise is repeated in diagram (b) with

three series-connected resistors to show that the
current has to fall to a value I/3 so as to
produce a p.d. across each resistor of emf/3.
From this it is possible to deduce a general rule
that, when n conductors are connected in
series, the current falls to I/n so that the
individual p.d.'s fall to emf/n.

** This simple relationship is the subject

of Ohm's Law which states that in association
with any given conductor, the ratio of voltage-
to-current is always constant; that constant is
known as the resistance of that conductor.

The Law is expressed in symbols as:

E/I = R

where E is the voltage in volts developed across

the conductor, I is the current in amps which
flows through the conductor and R is the
resistance of the conductor which is expressed
in ohms.

** Given any two of these quantities the third

can be calculated by manipulating the
expression (see Lesson on Algebra).

E = I.R or I = E/R

Again it is best to remember only one of these

relations and to obtain the others by cross-
Fig.16 shows two identical conductors
connected to a source so that a current flows
through each of them independently;
connected in this way they are said to be
connected in parallel or to be parallel-
connected Once again the p.d.'s developed
across each conductor are identical (equal to
the source emf) but the source is called
upon to supply twice as much current.
When three conductors are connected in
parallel the current in each stabilises at a
value I, the p.d. across each conductor
stabilises at the value of the emf and the
total load current drawn from the source
becomes 3.I.

** Because the current demanded increases with the number of parallel-connected conductors
it would appear to the source that the load resistance presented by n conductors each of
resistance R looks like R/n.

** In practice however it is fairly unusual to be connecting identical resistors in parallel or in

series. With series connection the resulting total resistance is straightforward. Although they are
of different values the resistors carry the same current and so they develop different values of
p.d.; these individual p.d.,s sum to the value of the emf. Draw a diagram similar to Fig.15 and
you will see that the total effective resistance Reff is given by R1 + R2 + R3 + . . . .

** Different resistors connected in parallel produce a more complicated calculation because it

is now the currents which sum instead of the p.d.'s and current (I) is on the bottom line of the
equation E/I = R. For the RAE you are not expected to derive these formulae only to be able to
quote them and use them. For resistors in parallel the total effective resistance is given by

1/Reff = 1/R1 + 1/R2 1/R3 + .....

For two resistors in parallel this is usually remembered as product-over-sum

Reff = R1.R2
R1+ R2

** There is sometimes a requirement to find the equivalent resistance of a network which

combines parallel-connected resistors and series-connected resistors. The rule is:

(i) reduce all parallel-connected groups to single equivalent resistors

(ii) collect all series combinations to single equivalent resistors
(iii) repeat steps (i) & (ii) until only one equivalent resistor remains.

Under 1.3.2 it was shown that the power dissipated in an electrical circuit is obtained by
multiplying together the values of electrical pressure (volts) and electrical flow-rate (amps); for
a purely-resistive (i.e. non-reactive) circuit this is expressed in symbols as

P= ExI
where P is the power in watts, E is the voltage applied across the component (or the p.d.
developed across that component) and I in amps is the current which flows through the
component.

** However the two parameters E and I form part of the expression for Ohm's Law and so it
is possible to express power in terms of the circuit resistance; in the above expression for
power the product I.R can be substituted for E or the ratio E/R can be substituted for I which
yields either

P = I.R x I = I2 .R or P = E x E/R = E2/R

(These two variations should be remembered.)

1.12.2 Inductive Reactance

Under 1.9 it was stated that, when the current which flows in an electric circuit varies in
value, then there is interaction between the electrical path and the surrounding magnetic field.
This interaction causes the current to fall to a value below that predicted by Ohm's Law. This
result is usually interpreted as an increase in the apparent resistance of the conducting path but
matters are not at all so delightfully simple.

A basic experiment in the early study of electricity and magnetism shows that current can be
induced in a coil of wire by moving a permanent magnet in the vicinity of the coil. It is easily
established that, although the strength of the magnet is important, the major factor which
controls the rate of current-flow is the speed of the magnet. relative to the coil. The greater the
rate of relative-movement (the greater the rate at which the magnetic field changes) so the
greater is the induced electric current.

** In a circuit which carries ac the rate at which the associated magnetic field varies is
controlled by the frequency of the ac; it follows therefore that, as frequency is increased, so also
does the induced (the opposing) current. Thus the apparent increase in resistance to an
alternating current becomes more pronounced as frequency rises.

** Note that the "proper" resistance of the conductor remains unchanged and this is usually
referred to as the dc resistance. It is only the apparent increase in resistance - the inductive
reactance - which is proportional to frequency; this reactance must be added to the dc-
resistance to obtain a prediction of the overall behaviour of the electric circuit ( see
below under 1.12.4: Impedance)

** The inductive reactance of a conductor (or of an Inductor) is given the symbol XL and is
expressed in ohms. Where the inductance has a value L (Henrys) and the frequency is f
(Hertz or cycles-per-second) the inductive reactance is given by

XL = 2‫ח‬fL

The constant 2‫ ח‬arises because frequency is measured as the number of revolutions per
second of a vector and each complete revolution is expressed as 2.‫ ח‬radians (see under 7.2:
Trigonometry). The combination 2‫ח‬f recurs frequently in ac theory and it is convenient to
replace it with the Greek symbol omega (lower case) ω . Thus the above expression is more
usually written as
 XL = ωL

(ω is known as the angular velocity in that it defines the number of radians traced out every
second - it defines the rate of change).

The above expression for inductive reactance is shown in Fig.17 as a plot of reactance
against frequency (see under 7.3: Graphs). Because all terms are in linear measure (i.e. the
expression does not include squared or higher-order terms) the graph is a straight line although
we still refer to it as a curve. We say that reactance, inductance and frequency are linearly
related. The full line shows the variation of XL with frequency; note that at zero frequency
(i.e. dc) the reactance is zero. The broken line shows the practical result when dc-resistance is
added.

** The difference between dc-resistance and inductive reactance lies in the relative timing of
current and voltage waveforms.. In a truly resistive circuit the current. is always related to the
voltage by Ohm's Law. If the current and voltage values are sampled at any instant of time the
ratio always yields the same figure. In a reactive circuit, either inductive or capacitive, this is
not so.

The flywheel-like action of inductance (see under 1.9.2) causes a circuit to resist changes in
current flow and this means that a period of time is required for current either to increase or to
decrease in value. In an ac circuit the applied emf constantly cycles and the resulting current,
although it varies in value with the same waveform, is delayed with respect to the voltage
waveform.

In Lesson-2 of Trigonometry (under Mathematics) a sinewave is derived as a result of

rotating a vector; in this waveform the amplitude is proportional to the sine of the angular
displacement. Such a sinewave is shown in Fig.18 where, in diagram (a), it represents the
alternating emf applied to an inductor.

Diagram (b) shows the waveform of the current which flows through the Inductor; it is
shown as delayed by a ¼-cycle or 900. In fact this is the delay (or the relative phase-angle) that
would be caused by a theoretically-perfect Inductor which would have zero dc-resistance. In a
practical Inductor the presence of resistance reduces the relative phase-angle; as the resistance
is increased so the phase-difference moves toward zero as the component changes from an
effective inductor to an effective resistor.

Note that, because inductive reactance is proportional to frequency so the relative phase-
angle moves towards 900 as frequency rises. However the general result of increasing the
frequency is that an Inductor becomes less efficient (or more "lossy") and this reduces the phase
angle. (It is shown below that power cannot be dissipated in pure reactances and so losses must
be represented as the presence of resistance.)

In a pure inductor the 900 relative-phase between current and voltage is often described by
saying that the two parameters are in time quadrature; the expression simply means that they
are displaced in time by 900. . (The electric and magnetic components of the field radiated from
an aerial are mutually at right-angles and are said to be in space-quadrature; see under 2.4:
Aerials.) It is usual to drop the adjective and simply refer to them as being "in quadrature".

Note that current maxima occur where the voltage is zero and is undergoing its maximum
rate-of-change; current minima occur where the voltage is at maximum value with zero rate-of-
change.

** Under 1.3.2 it was shown that the power dissipated in a circuit is obtained by multiplying
together the values of voltage and current. When using alternating current this can be done
only on a point-by-point basis; for each value of angle (along the horizontal axis) the
instantaneous values of current and voltage are selected and multiplied with the result shown
in Fig 18(c). (See also in Mathematics under RMS Va1ues). When two positive values or
two negative values are multiplied the result has a positive sign but, when either the voltage or
the current (but not both) are negative, the product has a negative sign.

** Fig.18 (c) appears to indicate that power is first drawn from the source for a ¼-cycle and
then returned to the source during the following ¼-cycle. However this is a mathematical result
and requires interpretation; clearly the average power during each whole cycle is zero. For this
reason a purely reactive component - one in which the voltage and current are in quadrature - is
said to draw a wattless current.

** All practical components possess dc-resistance and this causes the voltage and current
waveforms to depart from exact quadrature; current which flows through the dc-resistance
causes the dissipation of power. This problem is dealt with in calculation by treating each
practical component as two separate "pure" components namely a perfect inductor in series with
a perfect resistor; the voltage waveform developed across the pure resistor is in phase with the
current and so power is dissipated; the voltage waveform developed across the pure Inductor is
in quadrature and the theoretical component does not dissipate power.

** Clearly the two voltages cannot be added directly although they must sum to the value of
the applied emf; this matter is dealt with under 1.12.4 Impedance.

END OF LESSON 6

* * * * * * * * * * * *

QUESTIONS

1. What is a non-inductive resistor ?

2. , Three resistors each of value 300 ohms are connected (a) in series and (b) in parallel; what
is their effective value In each arrangement ?
3. Two of the resistors are connected in parallel and the third in series with the combination.
What now is their effective resistance?

4. What is the voltage developed across the 50-ohm [Link] of an aerial which is accepting 50
watts ?

5. What is the difference between a purely-resistive circuit and a practical circuit (one that
contains reactance) when they carry ac ?

6. Given that you have a normal multimeter (measures amps, volts or ohms either ac or dc) what
can you measure about an inductor ?

7. What effects would you expect to observe when an inductor is connected to a dc source ?

8. An emf is applied to a resistor; if the emf is doubled in value what do you expect to happen
to the current flowing through the resistor ?

Which Law of Physics enables you to solve this question simply ?

9. CATCH QUESTION If the emf applied to resistor is doubled in value what do you expect
to happen to the power that is dissipated by that resistor ?

Back to Top of Page

LESSON 7
1.12.3 Capacitive Reactance

** A capacitor of course does not pass a direct current and its ability to pass alternating
current is more a matter of effect than reality. However, as with an inductor, the current and
voltage waveforms do not vary in step; in a loss-free capacitor they are in quadrature. There is
a significant difference however in that capacitive current is advanced with respect to the
voltage as distinct from inductive current which lags behind the voltage.

It is reasonable to ask how the current is able to lead on the voltage; how could the current
know what the voltage is going to do in the immediate future ? The answer is that this
relationship is possible only with recurrent waveforms; information which controls current flow
comes from the preceding cycles. In effect the current is not so much a quarter-cycle in advance
as three-quarter cycle lagging behind and this can be deduced as follows:

** Alternating current flows in the connections to a capacitor and alternating p.d.s build-up
between the capacitor plates. The current in the connecting leads falls to zero as the capacitor
p.d. reaches its peak value; current in the connecting leads reaches its maximum value as the
capacitor p.d. falls to zero (as the p.d. reverses polarity) and the capacitor begins to re-charge
with the opposite polarity.

** As with inductors this quadrature relation between current and voltage means that the
term capacitive-reactance must be substituted for resistance.
 ** Alternating current which flows in the connections to a capacitor represents the charge (the
number of electrons) which is being moved in and out of the capacitor ; thus the larger the
capacitance the greater the amount of charge to be moved and so the greater must be the
alternating current. Such an increase of current (as a result of an increase of capacitance) is
interpreted through measurement as a decrease of reactance.

As the frequency of an ac is increased so the capacitor is allowed less and less time in which
to charge and discharge and so the alternating p.d. which develops across it is reduced - given
that the value of the ac remains constant. According to Ohm's Law the p.d. is the result of
multiplying current and resistance or, in this instance, current and reactance. If the value of the
ac remains constant as the p.d. decreases then the fall of p.d. with increasing frequency has to be
interpreted as another fall in reactance.

** Thus capacitive reactance (Xc) falls when either the capacitance or the frequency is
increased and this is expressed in mathematical symbols as

Xc = 1/(2πfC)

where the capacitive reactance (Xc) is in ohms, the frequency is expressed in circular measure
as 2πf radians per second and the capacitance (C) is in Farads Again it is usual to insert the
Greek symbol w for the angular frequency and the above formula becomes

Xc = 1/ωC

The expression for Xc. is plotted in Fig.19. Although all the terms are linear the result is not
a straight line because of the inverse relationship. As frequency falls toward zero (dc) so the
capacitive reactance increases toward an infinitely-large value (an open-circuit) ; at the other
end of the curve, where frequency is rising toward an infinitely-large value, so the value of the
reactance is falling toward zero (a short-circuit).
 Note that at each end of the curve there is a gradual approach to the relevant axis although,
in theory, contact never occurs. We say that the curve is asymptotic* to (in this instance) both
axes. In practice the actual point of contact is of no importance because, as the interval becomes
smaller, so the curve gets lost in the vagaries of measurement - we say that it falls below
the noise level

As with inductors the quadrature relation between current and voltage causes capacitors to
draw a wattless current. Capacitors too have dc-resistance, both in their connecting leads and in
the plates themselves, and there are other losses associated with the dielectric ; all these
combine to reduce the relative phase-angle below 900 degrees. In terms of Fig.19 this causes
the right-hand end of the curve to be asymptotic to a line slightly above zero ohms and this line
represents the effective dc-resistance (the losses) of the capacitor.

** Because the current that flows through a capacitor leads the applied voltage by
approximately 900 while that which flows through an inductor lags the applied voltage by
approximately 90 then, in any circuit which contains both components, the two currents must be
nearly 180-degrees out of phase; in other words they must tend to cancel. The opposing nature
of capacitive and inductive reactances
____________________________________________________________________
* Pronounced a - simp - totic
are shown mathematically by placing a minus(negative) sign before the symbols or figures for
capacitive reactance. Hence the expression given above for capacitive reactance should be
written as

Xc = - 1/wC

The negative sign ensures that, when an expression is written to describe the overall
behaviour of an ac circuit, the inductive and capacitive reactances cancel each other
automatically; for example, a circuit which contains resistance, inductive-reactance and
capacitive-reactance is described as R + wL - 1/wC

** Although resistance and both forms of reactance inevitably occur in the same circuit their
overall effect cannot be expressed simply by adding together their individual values in ohms.
For the series resistive circuit considered under 1.12.1 (Lesson-6) the overall resistance was
found by adding the individual p.d.'s and dividing by the common current ; for the parallel
circuit the overall resistance was found by adding together the various currents and then
dividing the total into the common p.d. For an ac circuit neither voltages nor currents can be
added directly unless they vary in sympathy; i.e. either the circuit is free of all reactance or the
inductive and capacitive reactances exactly cancel (see later under 1.13: Resonance)

The solution to this problem is geometric using vector addition. A vector quantity is
simply one which indicates direction as well as size (magnitude). For example it is correct to
say that a car travels at a speed of 30 m.p.h. but, to be credited with a velocity, it must be given a
direction such as 30 m.p.h. due east. Speed is a scaler quantity - it indicates only scale or size.
Velocity is a vector quantity. In this example the motion of the car could be depicted
graphically by an arrow (called a vector) of length 3-inches pointing to the right where the scale
chosen for the diagram is 1-inch represents 10 m.p.h.

** The overall effective "a-resistance" of a circuit which contains both dc-resistance and either
or both forms of reactance is given the name impedance.

The effective resistance or reactance which is present in any circuit is measured by the
amount of current which it permits to flow for a given applied emf. Fig.20 (a) shows three
vectors:

(i) an arrow AS whose length represents the dc-resistance (magnitude of the direct current)
which, by choice, is drawn horizontally

(ii) a second arrow BC whose length represents the inductive reactance (magnitude of the
alternating current); current which flows through this reactance leads by 900 and so the vector is
drawn at 900 to the resistance vector but rotated anticlockwise because this is the standard
agreed in mathematics for positive-rotation (inductive-reactance is given a +ve sign).

(iii) a third arrow AC which shows the overall effect of the first two and which is therefore a
vector that indicates both the magnitude of the resulting impedance (Z) and the phase-
angle between the applied voltage and the resulting alternating current.

It is perhaps easier to justify this vector-addition if the scenario is changed; consider AB as

the velocity of an aeroplane flying due east while BC represents the motion of the air (the wind)
which is blowing toward the north. The aeroplane may travel from A to C either by:
(a) traveling along AB in still air for a given time and then remaining stationary with respect to
the air while the wind blows for the same period of time

(b) traveling directly from A to C by flying vector AB at the same time as the wind blows vector
BC; the arrow AC represents the track over the ground as the aeroplane progresses
simultaneously along both AB and BC. Thus AC is a vector whose length represents the true
speed over the ground and whose direction shows the true direction of travel.

This technique can be used to add resistance to reactance, reactance to reactance or to find
the resultant waveform when two out-of-phase waveforms are added; it can be extended to sum
any number of such quantities. Fig.20(b) shows the vector diagram which is used to add
resistance to capacitive-reactance and diagram (c) shows the construction to add resistance to a
combination of both kinds of reactance. Note that the reactances and the resistances (all "like"
quantities) are summed first before vector addition of unlike quantities is attempted.

The above method using vector diagrams is a valid way of adding vector quantities and,
where it is required to add several such quantities, it may well prove to be the best method. It
can also be the best method when the quantities involved are not in exact quadrature; see
under Mathematics: Vector Diagrams. For calculations involved in electrical work, where
practical problems are usually

concerned with essentially pure components, all that is required is the solution to a right-angled
triangle. As shown in Mathematics under Lesson-1 of Trigonometry this can be achieved by
use either of Pythagoras Theorem or by using Trigonometry Tables.
 ** Pythagoras Theorem states that, in any right-angle triangle, the square on the hypotenuse
(the longest side) equals the sum of the squares on the other two sides; with reference
to Fig.20 this relation is expressed as

AC2 = AB2 + BC2

which translates as

Z2 = R2 + X2

Thus the square of the Impedance can be calculated by adding together the squares of resistance
and reactance. Of the three quantities resistance, reactance and impedance, if any two are
known then the third can be calculated.

** A vector diagram also shows the relative phase-angle between quadrature components. As
shown in Fig.21(a) a perfect inductor connected in series with a perfect resistor (which
represents a practical inductor) has a common-current I, a voltage across the inductor VL, a
voltage across the resistor VR and a voltage across the whole component VZ ; this last must
equal the applied emf.

In diagram (b) the current-vector I is shown as a reference for direction; the voltage VR is
drawn in phase with I and the voltage VL is shown leading the common-current by 900.
Because the current is common the lengths of the two voltage vectors represents also the
magnitudes of the resistance and reactance; thus their vector sum VZ represents both the
applied emf and the magnitude of the impedance. The angle φ (Greek letter phi) shows the
angle by which the emf leads the current which flows through the practical inductor (if
preferred, it shows the angle by which the current lags behind the applied voltage.

**From Fig.21 it can be deduced that the relative-phase angle φ is given by

cos φ = VR/Vz = R/Z

tan φ = VL/VR = XL/R

(see Lesson on Trigonometry)

** The quantity cosφ is known as the Power Factor of that particular component.
Remember that the perfect inductor draws a wattless current ; any power that the practical
inductor dissipates must be consumed in the perfect resistor. The Power Factor expresses the
ratio between apparent power (the applied voltage x the current ....known as volt-amps) and the
true power (p.d.-across-R x current). To obtain the actual power dissipated in a reactive circuit
therefore it is necessary to multiply the apparent power in volt-amps by the Power Factor ; i.e.

Power (Watts) = V x I x cosφ

** The quantity tan φ is the ratio between perfect reactance and perfect resistance and so
expresses the "goodness" of the component or the degree by which it departs from being a
perfect inductor. The ratio is known as the Q-factor ; for example a Q of 1 indicates equality
between the reactive and resistive components while a Q of 1,000 shows that the inductor is
close to being a perfect component.

END OF LESSON 7

* * * * * * * * * * * *

QUESTIONS

1. What is the basic difference between the ac which flows through an inductor and the ac which
flows through a capacitor with respect to the applied emf?

2. What other difference is there between these two currents in terms of frequency ?

3. In what way does resistance modify inductive and capacitive currents ?

4. Why is capacitive reactance written mathematically with a negative sign ? What is

impedance ?
A circuit contains 3-ohms resistance and 4-ohms inductive-reactance; what is its impedance
?

6. What is the significance of the Power Factor?

** Paragraphs marked with a double asterisk contain material which is relevant to the RAE ;
the remainder form a continuous explanation.

Back to Top of Page

LESSON 8
1.12.5 Notation (not essential for RAE)

It is perhaps easiest to write " R + X = Z " and to remember (hopefully) that the addition
must be made vectorially as discussed in Lesson 7 ; i.e. first square R and X, then add the
results and then find the square-root to obtain Z.
 This is not too difficult when looking at the symbols R and X but, once numbers are inserted
in place of those symbols, then the likelihood of error becomes very great. As a reminder it is
possible (and indeed sometimes done) to add some kind of symbol to indicate that the quantities
are vectors: for example

R+ X= Z

Such a notation however is clumsy and does not readily lend itself to algebraic
manipulations (where the requirement is to solve a problem in general terms rather than to
obtain an arithmetical answer). The mathematical tool used to handle these expressions is the j-
notation which is described in detail in the Mathematics Lesson.

At its most simple a letter j is placed before each item which is reactive thus for example

Z = 52 + j67 - j15 + 12 + ,j6

indicates that a series circuit consists of a 52-ohm resistance followed by a 67-ohm inductive-
reactance followed by a 15-ohm capacitive-reactance followed by a 12-ohm resistance followed
by a 6-ohm inductive-reactance.

Had the frequency been given these figures could be interpreted in terms of inductors and
capacitors but, without such information, all that can be derived from the expression is the
overall impedance (at whatever the frequency may be). The first step in this process is to collect
together all the similar terms; to make life even easier we put all the reactive terms together
inside a set of brackets and use a single j in front of the brackets

Z = 52 + 12 + j(67 - 15 + 6)

= 64 + j58

From this can be calculated:

(a) the impedance Z = √(642 + 582)

(b) the phase-angle φ between current and voltage = tan-1 (58/64)

(The notation tan-1 is mathematicians' shorthand which simply means " ... the angle whose
tangent is ... ")

A full treatment is given in Mathematics but here it is noted only that j is more than just an
indicator of a quadrature relation; it can be used very effectively in manipulating equations and
is known as an Operator. A quantity

expressed as R + JX is called a complex quantity. These are sometimes expressed in what are
known as polar co-ordinates which consist of a figure that gives the magnitude (Z) followed by
an angle that represents the relative phase angle; for example the above expression 64 + j58
becomes

√ (642 + 582), tan-1 (58/64)

 The above discussion deals with a set of components connected in series. When components
are connected in parallel the formulae derived from Ohm's Law deal in inverse quantities (see
under 1.12.1 in Lesson-6). For ease in handling these formulae a set of names has been given to
the inverse of resistance and reactances namely

1/resistance (1 /R) is called conductance (G)

1/reactance (1/X) is called susceptance (B)
1/impedance (1/Z) is called admittance (Y)

Thus in place of l/Z = 1/R + 1/jX we can write Y = G + JB.

The benefits of this alternative notation may not be immediately evident until it is realised
that Instruments for measuring resistance/reactance can be built to give their results in either the
normal or inverted form. An Impedance Bridge gives its results in terms of R and jX (a series-
connected circuit) but an Admittance Bridge gives results in terms of G and jB (a parallel-
connected circuit). Where measurements are made with an Impedance Bridge the circuit is
adjusted by adding extra components in series (often difficult); where measurements are made
with an Admittance Bridge the circuit is adjusted by adding extra components in parallel.

1.12.6 Impedance Matching

** Consider two circuits both of which carry a signal at 10 watts. Circuit A draws a current of
1-amp and so has 10-volts developed across it (power = volts x amps) while circuit B draws a
current of 0.1 amps (100 mA) and so has 100-volts developed across it. The application of
Ohm's Law shows that circuit A has an impedance of 10 ohms but that circuit B has an
impedance of 1,000 ohms.

A problem arises when it is required to transfer the 10-watt signal from circuit-A into
circuit-B. Because circuit-A can supply only 10 volts from which circuit-B will accept only a
maximum of 10/1,000 A (10 mA) it follows that the maximum signal power that can be
transferred from A into B is 100 mW or 0.1 W . Additionally circuit-A is left with a surplus of
990 mA equivalent to 9.9 W which it has to dump somewhere.

** This inability to transfer power between two circuits because they operate at different
impedances is known as an impedance mismatch ; to transfer the maximum possible power
between two circuits it is essential that their impedances match. Such a transfer can be achieved
between two un-matched circuits however by use of a suitable coupling arrangement and the
process is referred to as Impedance Matching ; basically the trick is to find a device whose
input impedance matches the impedance of the source-circuit but whose output impedance
matches that of the receiving-circuit (the load).

** One type of matching device, the Transformer, has been discussed already in Lesson-5.
Note that the transformer cannot amplify signals because it lacks any form of auxiliary power.
Thus, when a transformer steps-up the circuit voltage, it must at the same time step-down the
circuit current ; an increase (or fall) in the circuit voltage accompanied by a decrease (or rise) in
the circuit current corresponds to a change of impedance as the signal is transferred across the
transformer from primary winding to secondary winding.

** In the example above a matching transformer required to transform the 10-volts of circuit-
A to the 100-volts of circuit-B would require a turns-ratio (n) of 1:10. However the
transformer is matching between a 10-ohm circuit and a 1,000-ohm circuit; thus a turns-ratio of
10 produces an impedance-ratio of 100 and, in general terms, we can write that

impedance-ratio = n2

** Matching can be achieved also by using networks of inductors and capacitors which are
referred to generally as matching filters, matching sections or simply as networks. It can be
achieved also by using networks of resistors although these produce a reduction in the signal
level, known as attenuation, that may sometimes be a disadvantage. These resistive networks
are variously called attenuators or pads and they may also be introduced into a system to
provide only attenuation or to isolate circuits that otherwise would mutually interfere. There are
occasions when all three functions are indistinguishable.

Detailed knowledge of filters and attenuators is not required for the RAE but their use to
cure or to avoid interference is dealt with under 6.4: Measurements and EMC. They are also
introduced under Transmitters.

In general terms (i.e. not usually used in Amateur Radio) impedance matching can be
achieved also by using a feedback-amplifier whose input and output impedances can be tailored
to the required values; these circuits are described in later Sections of this Course.

There is a Maximum Power-transfer Theorem which states that the maximum possible
power is transferred between circuits when their impedances are matched. As described
under 1.15: Internal Impedance this means that a maximum 50% of the signal power can be
transferred; the remainder is dissipated in the source circuit.

1.12.7 Component Q-factor

** Although a "pure" inductor as defined draws a wattless current a practical inductor is

wound with wire that has dc-resistance; together with eddy-currents and stray couplings these
cause losses.

Given the experimental technique it is possible to measure each of these power-losses but, in
general, such information is of little use. Interest centres on the total power consumed by the
inductor and so all losses are treated as though caused by dissipation as heat in a single
theoretical resistor.

One way of determining this total loss is to measure the relative phase- angle between the
current and voltage waveforms. A more accurate method is to use an ac instrument called
a Bridge which, as already discussed, will show a reactance value in series with a resistance
value (an Impedance Bridge) or as a reactance value in parallel with a resistance value (an
Admittance Bridge). For a high-quality inductor the values of reactance will not differ
appreciably between the two methods of measurement but that is not so for types with
appreciable losses.

** An efficient (low-loss) inductor is indicated by either a low-value series resistor or by a

high-value parallel resistor but, with either method of measurement, the ratio between resistance
and reactance is the same. This ratio is in effect a quality rating and is known as the Q-
value. As discussed above the ratio also describes the relative phase between voltage and
current waveforms.
 ** So far the Q-factor has been defined in terms of an inductor but there is no reason why the
idea should not be extended to a capacitor. Although a capacitor does not provide a dc path
between its plates it does permit alternating currents to flow in its connections and also to flow
across the plate surfaces; where current flows there is a heating effect which represents loss of
signal power. Additionally the dielectric dissipates a small amount of power. All these losses
are usually amalgamated into a single (theoretical) resistor which again can be calculated as
either a series resistor or a parallel resistor. Once again an efficient component is characterised
by a high Q-value.

Q-factors are involved also in resonant circuits ; see next Section.

1.13 RESONANCE

1.13.1 Introduction

** Resonance occurs in electric circuits which contain both inductive and capacitive
reactances ; inevitably resistance must be present also but, although it modifies the behaviour
of the circuit, it does not itself play a significant part in the basic phenomenon of resonance.

Fig. 22 shows a practical inductor and a

practical capacitor connected together.
Assume that in some manner, perhaps by
electromagnetic coupling, a current has been
induced to flow in the inductor.

As discussed under 1.9.2 Self-Inductance the

inductor has a flywheel-like nature that tries to
keep a current flowing and which resists any
changes in the rate-of-flow. As a consequence
the capacitor collects all the available charge
(all the free electrons) and so eventually forces
a cessation of current flow. Whatever energy
was induced into the circuit is now collected
Fig.22 Practical Inductor Connected to a in the electric field between the plates of the
Practical Capacitor capacitor while the cessation of current-flow
has terminated the flywheel action of the
Inductor.

Thus the charged capacitor is now free to discharge through the Inductor; in doing so, it
causes a current to flow in the reverse direction and so sets-up a new flywheel action to maintain
that reversed flow.

Once again, with the reverse current established, the inductor remains in control until all the
circuit energy is again stored in the capacitor but this time with the polarity reversed.

This cycle must repeat and would do so endlessly but for the practical reality of circuit
resistance (and other small losses) which slowly convert the stored energy into heat which, in
the form of infra-red radiation, is then lost from the circuit.
 ** This constant interchange of energy between two different storage media is the mechanism
of resonance; for electrical resonance it is the interchange between:

(a) static or stationary charge held in the electric field between the plates of a capacitor and
(b) dynamic or moving charge (an electric current) where the energy is held in the
accompanying magnetic field.

The rate at which this interchange takes place is controlled by the amount of storage
available and clearly must be a function of both the capacitive storage and the inductive storage.
The larger the amount of storage so the longer is the period required to discharge one of them in
the process of charging the other; in other words, if either the capacitance or the inductance is
increased, the circuit must operate at a slower speed and there are fewer cycles of operation per
second.

** For any combination of inductance and capacitance there is a preferred (or natural) rate
of oscillation at which the interchanging action takes place and this, reckoned as the number of
complete cycles per second, is known as the natural frequency or as the resonance
frequency of the circuit.

(It is common to see this frequency referred to as the "resonant" frequency which implies
that it is the frequency which is resonating. The correct term is derived from the full expression
" the frequency of resonance ". )

** Resonance frequencies like all frequencies are expressed in Hertz (Hz) where 1-Hz is
defined as 1 cycle per second. In an electrical circuit the frequency at which a combination of
inductance and capacitance resonates is given by the expression

f = 1/(2π.√LC)

where f is in Hz (cycles per second), L is in Henrys and C is in Farads.

NOTE This formula is required for the R.A.E.

** The resonance frequency is found to be that frequency at which the inductive reactance is
numerically equal to the capacitive reactance ; if XL is equal to -XC then the overall reactance
in a resonant circuit is zero and so, at resonance, the current that circulates between the two
components rises to a maximum value which is controlled only by the circuit resistance RL +
RC .

** Thus the resonance frequency is that frequency at which it is most easy to cause current to
circulate between the inductor and capacitor.

It is worth remembering that resonance is not purely an electrical phenomenon but occurs in
almost every aspect of everyday life. The well-known leg is a pendulum hanging from the
pelvic girdle and it is moved with minimum effort by swinging it at its natural frequency (see
next Lesson). Hence, in developing the technique of locomotion, life forms varied their speed
by changing the length of each stride rather than trying to change the striking frequency of the
leg(s).
 When walking, each leg remains in contact with the ground for the majority of the time ;
for two-legged locomotion, such as is used by humans, one leg supports the body while the
other moves through to catch the body as it falls off the supporting leg. In the trot only one leg
remains in contact with the ground in a series of hops from leg to leg. In the run both legs lose
contact with the ground over most of the running period as the runner leaps from leg to leg to
gain the maximum possible length of stride.

Quite simple tests will convince you that changes in the striking rate of the legs are brought
about by bending the knees so shortening leg-length.

END OF LESSON 8

* * * * * * * * * * * *

QUESTIONS

1. What is the purpose of using the letter j in calculations ?

What is the approximate impedance of a circuit comprised of 7 + j7 ?

2. Why do we want to match one circuit to another ?

Given one circuit of 50 ohms and another of 200 ohms what must be the turns-ratio of a
transformer that will produce a satisfactory match ?

3. What is meant by the term Q-factor as applied to an inductor ?

4. Describe the basic mechanism of Resonance

5. What is the natural resonance frequency of a circuit which contains on inductance of 1 mH

and a capacitance of 0.001 microfarad (i.e. 1/1000 Henrys and 1/1,000,000 000 Farad)

Back to Top of Page

LESSON 9
1.13.1 Introduction (continued)

Examination of the mechanism responsible for resonance in a pendulum shows that its
behaviour is very similar to that of the resonant electric circuit. Energy-transfer in the two
systems is compared in Fig.23.
 To start a pendulum swinging it is first
displaced to one side as a result of which the
suspension raises the Bob above its starting
position. Because the Bob is raised against the
force of Gravity (which attempts to return it to
its original position) the Bob possesses the
ability (or the potential) to do work (perhaps to
cause damage ?) at some future time when it is
released and allowed to fall.

Thus the Bob may be considered as storing

the ability to do work as distinct from actually
doing work. For this reason it is said to
possess potential energy and this corresponds
to the static energy-storage in the electric field
of a capacitor (see diagrams a & b).

When the Bob is released its vertical fall is

converted by the suspension into a horizontal
motion ; thus diagram (d) shows the Bob at the
bottom of its fall where it no longer possesses
potential energy but has acquired a different
ability to do work because of its horizontal
motion. Energy is now stored in the moving
mass of the Bob and its suspension and, in this
incarnation, is referred to as dynamic (moving)
or kinetic energy. If you doubt the ability of
kinetic energy to do work then - with great care
- try standing in front of a child's swing.

The condition shown in diagram (d) Fig.23 Comparison of Pendulum and

corresponds to the condition in an electrical Resonant Circuit
resonant circuit where the static energy stored
in the capacitor has become zero (with zero p.d.
)

d the current-flow has created a magnetic field ; see diagram (c). In diagram (f) the pendulum
is shown to have swung beyond the starting position and to have climbed to a second position of
potential energy-storage. By virtue of its mass and velocity it has, like the child's swing, been
reluctant to stop at the bottom of its fall and has imitated the flywheel-like action of the
inductor. This reluctance to stop is referred to as momentum.

Diagram (e) shows the equivalent condition in the electrical circuit where the capacitor is fully
charged with reversed polarity. Diagrams (g and (h) complete the cycle of action.

The similarity between the electrical circuit and the mechanical pendulum is very real and it
is common practice to solve mechanical design problems by drawing an equivalent electrical
circuit in which inductors, capacitors and resistors represent mechanical parameters such as
mass, compliance and loss. The problems involved in the design say of a loudspeaker can thus
be solved using standard ac theory.
 ** That the resonance frequency is the
frequency at which things happen most easily
can be deduced by inspecting the impedance
expressions derived in the previous Section. It
was said that inductive reactance is written as
ωL and capacitive reactance as -1/ωC ; in any
circuit which involves both kinds of reactance
there results the combination

ωL - 1/ωC (or XL - Xc).

Figs. 17 & 19 show that X L increases with

frequency while X C decreases; there has to be a
point where the two curves intersect thus
showing that, at that (resonance) frequency, the
reactances have the same value and so cancel.
At this frequency therefore the value of the
current is determined only by the circuit
resistance.

Fig. 22 showed a practical inductor and a

practical capacitor connected together. The two
resistors R L and Rc represent the losses of the
two components which consist mainly of so-
called ohmic losses that arise from the passage
of current through their dc- resistance. In
practice the losses associated with capacitors
are low relative to those of inductors (it is said
that the capacitor-losses are swamped by the
inductor- losses) and it is legitimate either to
ignore Rc or to combine it with RL. This leads
to the more practical representation of Fig.
24(a).

** Impedance is measured with an instrument called a Bridge which, being without intelligence,
has to be instructed as to whether the load presented to it is a series combination of components
or a parallel combination. Fig.24(b) shows the arrangement of diagram (a) broken open so that
measurements can be made as a series circuit ; note particularly that a common-current flows
through all three components and also through the Bridge.

** In Fig.24(c) the LCR circuit is left closed and the Bridge has been connected across the
parallel combination of inductor and capacitor; in fact the load presented to the Bridge is the
combination of a capacitor in parallel with a series combination of pure inductor and pure
resistance. Note that, in this arrangement, there is a common-voltage developed across the
capacitor, across the practical inductor and across the Bridge; the currents that flow through the
capacitor, the inductor and the Bridge are all different.

There is a difficulty here; in diagram (b) the Bridge is clearly presented with a series
combination of L, C and R but in diagram (c) it is presented with a mixture of series and parallel
components. If told that arrangement (c) was a parallel circuit the Bridge would interpret its
measurement results in the form shown in diagram (d).

The fact that both diagram (b) and diagram (d) can represent the same combination of
components is a bit startling but it is a most useful result. To convert from one to the other form
is a straightforward mathematical process known as a series-parallel transformation but it is not
required for the R.A.E. A point to remember however is that such transformations are valid
only for the particular frequency for which they were calculated.

Both Rs in the series arrangement and Rp in the parallel arrangement represent the circuit losses
which means that, whatever their calculated or measured values, they must dissipate equal
powers. They differ in that Rs carries the common-current - the circulating current - that
circulates between inductor and capacitor while Rp has its own private current which is driven
by an emf that is common to itself, the inductor and the capacitor.

** The circulating current is controlled by the small-value Rs and so a large current can be
expected. This large circulating current must set-up large p.d.'s across the reactors but, because
the reactances cancel (i.e. the voltage waveforms cancel), it is only the applied emf that appears
across the series resistor.

** When the circuit is measured as a parallel combination the very large p.d. across the L and C
becomes apparent but the overall current (the make-up current) is small because cancellation
now occurs between the currents which flow through the inductor and the capacitor.

** Thus a series measurement shows the losses as a large resistive current with a small voltage
which requires a low-value resistor. A parallel measurement shows the losses as a small resistive
current with a large voltage which requires a high-value resistor.

When measurements are made as shown in diagram (b) an LCR circuit is referred to as a series-
resonant circuit; when measurements are made as shown in diagram (d) an LCR circuit is
referred to as a parallel-resonant circuit. The two "arrangements" are essentially the same
circuit but in their applications they are very different.

1.13.2 The Series-resonant Circuit

When an LC resonant circuit is used in the series mode the alternating circulating current is
determined according to Ohm's Law by the applied emf and the overall impedance at that
particular frequency. In the general form this is given by

I = E/Z = E/(R + XL - XC)

At the resonance frequency, where the inductive reactance equals the capacitive reactance,
this reduces to E/R. When the frequency is reduced below the resonance frequency the
capacitive reactance Xc is increased but the inductive reactance XL is reduced and so the two
reactances no longer cancel. Similarly if the frequency is increased above the resonance
frequency the capacitive reactance decreases but the inductive reactance increases; it is
only at the frequency of resonance that complete cancellation is possible.

** Fig.25 is a graph on which is re-plotted the inductive and capacitive reactance curves
of Figs. 17 & 19 together with a curve that represents the difference (XL - Xc:). This shows
that the impedance is reduced to Rs alone at resonance with the curve rising toward the
capacitive-reactance curve at lower frequencies and toward the inductive-reactance curve at
higher frequencies.

** Because, at resonance, the series

impedance falls to its lowest value then the
circulating current must reach its highest
value; as already discussed the p.d. developed
across each reactor must also reach its
maximum value. Note that the impedance
curve is not symmetrical because the two
reactance curves have different forms.

Although at resonance the series circuit

appears to be purely resistive and to present its
lowest impedance it is important to remember
that nevertheless it does contain reactors and
that there is a relatively large voltage across
each of them. If the given inductor and
capacitor are removed from the circuit and
replaced by components of the same value but
of higher quality (i.e. with lower losses) then
the value of the equivalent resistor Rs. is
reduced and so the circulating current is
increased ; in turn this leads to an even higher
p.d. developing across the reactors. With
components of even reasonable quality these
voltages can be 10-times the value of the
applied emf and are often much greater.

WARNING When dealing with transmitters these voltages can be lethal.

** This voltage-multiplication action is one of the most useful properties of resonant circuits.
If an applied Source delivers a mixture of signals at different frequencies (as does a receiving
aerial) then that signal whose frequency is closest to the resonance frequency develops a much
larger voltage across both the inductor and the capacitor than do the other signals. This provides
a means of selecting a wanted signal from amongst many. The action of adjusting a circuit so
that its resonance frequency equals the frequency of a wanted signal is known as tuning the
circuit to that signal and so these resonant circuits have acquired the name tuned circuits.

** A practical arrangement which enables a wanted signal to be selectively enhanced before

being passed to a receiver is shown in Fig.26(a) ; the arrow drawn through the capacitor
symbol indicates that the component can be adjusted mechanically to provide a means of tuning
the circuit. Similar tuning arrangements can be provided by varying the inductor but, as a
general rule, it is less easy and more expensive to achieve than the variable capacitor.
 Fig. 26 Use of Resonant Circuits In Aerial Arrangements

(a) series-tuned Aerial Transformer

(b) parallel-tuned Aerial Transformer
(c) Rejector and Acceptor Traps or Notch Filters

Note that, although the wanted signal is magnified relative to unwanted signals, this is not an
example of amplification; that term implies an increase in signal power and requires the use of
an auxiliary power source.

The degree of magnification obtained is determined by the current-flow at resonance

compared with the current-flow at off-resonance frequencies. The current-flow at resonance is
controlled by the circuit resistance, the current-flow below the resonance frequency is
controlled by the capacitive reactance and the current-flow above the resonance frequency is
controlled by the inductive reactance.

** Signal magnification therefore depends on the ratio of reactance to resistance and this was
the criterion used to define the quality (the Q-value) of a reactor. Thus the performance of tuned
circuits is also given in terms of Q.

In general the losses connected with capacitors are negligibly small and so the magnification
of a resonant circuit is usually taken as the Q-value of the inductor. For example: a resonant
circuit which contains an inductor with a Q of 10 would be expected to provide a signal
magnification of 10 times at the resonance frequency. In most practical circuits a Q of 10 is
about the lowest likely to be acceptable although, in the final stages of a transmitter, the output
circuit may have a Q as low as 2 because it is required to be "lossy" - it is required to lose its
power to the aerial

A very important point which has implications in receiver design and also in oscillator
circuits arises from the tuning arrangements outlined above. Suppose for example that it is
required to raise the resonance frequency of a tuned circuit; the value of the capacitor is
decreased so that its reactance at the new frequency will equal the higher reactance of the
inductor at that new frequency and this adjustment will not appreciably alter the value of the
circuit losses Rs. However, because the reactances have increased while the value of Rs is
unchanged, the circuit Q-factor has been increased.
 ** This means that the magnification of a variable tuned-circuit varies as its resonance-
frequency is adjusted.

** An effect of much greater importance

however is shown in Fig.27 where the
magnification of resonant circuits is shown as it
varies with changes of frequency. At low values
of Q E the magnification is almost independent
of frequency and we say that the selectivity of
the circuit is low. At high Q-values the
magnification peaks significantly at the
resonance frequency and such circuits are said
to offer high selectivity.

** If the Q-value of adjustable circuits varies

with frequency then clearly the magnification
also varies with frequency and this is a serious
handicap for equipment which is required to
operate over a band of frequencies. For this
reason the simple arrangement of Fig.26(a) is
seldom found by itself ; see
under Bandwidth in the next Section and also
in the Lessons dealing with Receivers.

** Series-tuned circuits are characterised by a large flow of current at the resonance frequency
and they are regarded as being willing to accept signals at that frequency while rejecting all
others. From this they have acquired the name Acceptor Circuits . Their stock-in-trade is a
reduction of impedance at resonance.

END OF LESSON 9

* * * * * * * * * * * *

QUESTIONS

1. What particular characteristic of a reactive circuit indicates that it is operating at its resonance
frequency ?

2. Explain in simple terms the difference between a so-called series-resonant circuit (or
Acceptor circuit) and a so-called parallel-resonant circuit (or Rejector circuit).

3. At what frequency will a 2 μH choke resonate with stray capacitances measured as 10 pF ?

See Lesson-8. (This question tests your ability to work with powers of 10 and is the type that
can earn easy marks in the exam. If you have problems then please ask for help.)

4. Why are very large voltages present in tuned circuits which are handling power (such as the
final stages of a transmitter) ?

5. What is the basic purpose of an aerial transformer as used in a receiver?

6. Why is a simple tuned-circuit arrangement not very satisfactory in equipment which is
required to operate over a fairly large band of
frequencies ?

Back to Top of Page

LESSON 10
1.13.3 The Parallel-resonant Circuit

An alternative way of measuring a resonant circuit is simply to connect a Bridge across the
combination of components as shown in Fig.24 (c). From the measurement point of view the
Inductor, the Capacitor and the (theoretical) Resistor are now seen as being connected in parallel
as shown in Fig.22(d).

** When a bridge is connected in this manner it is no longer possible to determine the

common circulating current and only the common voltage across the components can be
measured. To the Bridge the capacitive and inductive currents now flow in parallel and, as
already discussed, they cancel (exactly at the resonance frequency). Thus the bridge measures a
relatively small current which is known as the make-up current ; this represents the losses in the
circuit.

** In the series circuit discussed in the previous Section there is a common current and the
two reactive voltages differ; in this parallel circuit there is a common voltage and it is the two
reactive currents that differ. As the frequency of resonance is approached so the current-
canceling action becomes total and the make-up current consists of that which flows through the
resistor alone. Hence, in this circuit, resonance is indicated by a minimum value of current
which is interpreted as a maximum value of impedance.

** Where the series circuit accepts signals at or close to the resonance frequency, this parallel
circuit rejects such signals and is accordingly known as a Rejector Circuit.

** Remember however that, although the bridge cannot measure the current which circulates
between the Inductor and Capacitor, that current nevertheless still flows and it still reaches it
maximum value (controlled by the effective series resistance) at the resonance frequency. Thus
the parallel circuit rejects current at resonance but develops a magnified voltage at that
frequency. The use of such a circuit in selecting a wanted signal from an aerial is shown
in Fig.26 (b).

** The impedance presented by an aerial may be relatively-low or extremely reactive so that it

either reduces the Q of the tuned-circuit or de-tunes it. This effect is off-set by using the tuned
inductor as an autotransformer (see Lesson-5) and so the two arrangements of Fig.26
(a & b) are perhaps less different than it may appear.

** Fig.26 (C) shows another use for both acceptor and rejector circuits ; here they are used as
traps to reduce the passage of powerful unwanted signals. A rejector connected in series with the
aerial feed reduces the level of signals at or close to the frequency to which it is tuned ; an
acceptor connected across the terminals which feed the receiver tends to short-circuit signals at
or close to the frequency to which it is tuned.
 Because these circuit arrangements result in a "hole" in the range of receivable frequencies
they are also called notch filters.

** The two different methods of measuring resonant circuits in fact result in slightly different
resonance frequencies and reference books will be found to give different formulae. In practice
however when dealing with circuits of

reasonable Q-value the difference is negligible and for all practical purposes the standard
formula, as already given, is

f = 1/(2π.√(LC)

where f is in Hertz, C is in Farads, L is in Henrys.

1.13.4 Dynamic Impedance

Where say a rejector circuit is to be used to develop a signal voltage at a preferred frequency
it is helpful to the Designer to know exactly what impedance it will present ; off-resonance that
impedance will be a mixture of resistance with either inductive or capacitive reactance
depending on the direction in which the circuit is off-tune.

The impedance thus presented by a rejector circuit at resonance is known as its dynamic
impedance; at resonance the impedance should be exactly resistive and so is sometimes
referred to as the dynamic resistance.

With the application of Ohm's Law the dynamic impedance is obtained by dividing the
applied voltage by the make-up current which, in turn, depends on the Q-factor. A low Q
implies that the components are lossy and that perfect cancellation between the reactive currents
is not possible; in turn this means that the make-up current is not small and so the dynamic
impedance is low.

With similar reasoning a high-Q circuit presents a large dynamic impedance.

As already discussed the Q of a resonant circuit varies with

(a) frequency - because the ratio of resistance to reactance changes

(b) the value chosen for the inductor/capacitor (L/C) ratio combination because this too
changes the ratio of resistance to reactance ; the resonance frequency is determined by the
product L x C.

Thus the value of the dynamic resistance can be controlled by the Designer because he is free to
choose the combination of L and C to suit the value of resistance that is imposed by the circuit
losses. He is also faced with a compromise decision because the ratio X/R determines the
circuit Q which controls the band of frequencies over which the tuned circuit operates.

This matter is usually dealt with in a somewhat rule-of-thumb manner by considering the
ratio of inductance to capacitance (i.e. L/C) which is called the L-C Ratio ; a high L/C ratio
results in a high dynamic resistance.
See also the later Sections on Bandwidth, Bandpass Circuits, Amplifiers, Oscillators, Receivers
and Transmitters.

1.14 BANDWIDTH

1.14.1 Definitions

** The term bandwidth refers to the band of frequencies over which a particular piece of
equipment will operate satisfactorily. Bandwidth is limited by a number of factors:

(1.) At the lower end of the frequency range performance deteriorates because of the increasing
attenuation of coupling circuits; see Fig. 28.
** The coupling capacitor in Fig. 28(a) is
inserted to separate an ac signal from dc with
which it is mixed but at low frequencies the
rising reactance of the capacitor reduces the
wanted signal also.

** In Fig.28(b) a transformer is used to

separate the wanted signal from a dc. However
the transformer windings appear as Inductances
connected across the signal path and, as the
frequency decreases, so the falling reactance
diverts signal current away from the intended
destination.

** (ii) Another cause of loss at low

frequencies is the increase of feedback through
the failure of by-pass capacitors (see Part 2
Electronic Amplifiers.

(iii) At the high end of the frequency range

stray capacitances, shown in Fig. 28(a) in
broken line, provide increasingly-effective
alternative paths for signal currents which
therefore fail to reach their intended destination
as the frequency increases.

** (iv) In designing transformers it is very

difficult to ensure that all the magnetic field
from one winding interacts with the other
winding and this "lost" magnetic coupling is
referred to as leakage inductance. As shown
in the equivalent circuit (C) leakage inductance
appears as an inductive reactance in series with
the signal path and this limits the high-
frequency response.

** (v) In tuned (radio-frequency) amplifiers

the bandwidth is determined solely by the
 characteristics of their resonant circuits ; see
under 2.6 and under 1. 13 in the previous
Lesson.

Bandwidth however, expressed as a lower-to-upper frequency range, can be a deceptive

quantity. For example:
an audio amplifier ranges from 50 to 15,000 Hz
a video (television) amplifier ranges from 0 to 5MHz
a radio-frequency amplifier might range from 100 to 120 MHz

In the order shown the bandwidths may appear to increase from 15 kHz to 20 MHz yet the
difficult tasks are to extend the bandwidths of the audio amplifier and to restrict the bandwidth
of the rf amplifier.

** Another way of expressing bandwidth is to give the upper and lower limits as variations on
the centre frequency. Thus the rf amplifier, which has a centre frequency of 110-MHz and
extends both upward and downward by 10 MHz, can be described as having a bandwidth of
110 ± 10 MHz. Alternatively this amplifier is said to have a relative bandwidth of
(approximately) 9% which shows the frequency range as a fraction of the centre (or average)
frequency.

** Again there is a difficulty in that at 1 MHz a relative bandwidth of 10% is only 100 kHz
which is at variance with the same relative bandwidth at 100 MHz. Clearly, as the centre
frequency is increased, so it becomes ever more difficult to restrict the relative bandwidth.

As a description of performance in audio and video amplifiers the term relative

bandwidth becomes something of a joke. For example, the video amplifier would be said to
cover a range 2.5 ± 2.5 MHz ; expressed as relative bandwidth this becomes +100% - ??% .

Where, as in video amplifiers, the bandwidth approaches zero frequency (or dc) then
bandwidth is described in terms of the number of octaves covered. Mathematically an octave
describes a frequency interval in which the upper limit is twice the lower frequency-limit; to the
Human ear an octave is not exactly 2:1. For the above audio amplifier the bandwidth is
approximately 8 octaves; for the video amplifier the practical bandwidth is about 17 octaves;
for the rf amplifier the bandwidth is about 0.2 octaves. This method of expressing bandwidth
makes clear the difficulty of extending the range of audio and video equipment and of
compressing the range of rf amplifiers.

Apart from the limitations described above all three methods are valid and the choice is
governed by the particular requirements of the moment.

It must always be remembered however that a given bandwidth is purely a matter of

definitions because electronic equipment does not suddenly cut-off at the edges of the band.
For amplifiers it is usual to define the band as the range between those frequencies where the
gain has dropped to 0.7 of the maximum gain (known as the 3-dB points). For aerials the
bandwidth is usually defined in terms of the standing-wave ratio (see under 2. 3: Transmission
Lines) which is a measure of the accuracy to which an aerial system is tuned and matched.

1.14.2 The Significance of Bandwidth

** The sinewave (see Trigonometry in Part 7 : Mathematics) is the simplest, indeed the purist,
of repetitive waveforms. All waveforms which repeat can be broken down into a series of
sinewaves which are harmonically-related to the repetition frequency (or frequencies). The
repetition frequency is called the first harmonic or the Fundamental, the frequency which is
twice that of the Fundamental is called the second harmonic, the frequency which is three times
that of the Fundamental is called the third harmonic and so on. For any one specific repetitive
waveform each harmonic has a specific amplitude and a specific phase relative to the
Fundamental.

The process of deriving these component sinewaves from a given repeating waveform is
known as Fourier Analysis after the French mathematician who first derived it but knowledge
of this is not required for the R.A.E.

** It is the presence of these extra frequencies that makes it necessary to consider bandwidth
in the performance of any equipment. In general the ear is not fussy about phase relationships
and so, in audio-amplifier design, phase-distortion (disturbance of phase relations) is not
directly of concern; the important thing is to preserve the relative amplitudes if the original
sound is to be reproduced faithfully. However, when feedback is used in an amplifier, phase-
distortion must be controlled else negative-feedback turns into positive-feedback with
unfortunate results. (Note that the amplifier must be well designed before feedback can be
applied.

The eye is not so accommodating and, in video amplifiers, it is essential to preserve both the
amplitude and phase relations.

** In general rapid phase shifts are associated with rapid changes of gain as frequency is
varied and a design which eliminates too-rapid changes of gain at the edges of the pass-band
should not suffer problems from phase-distortion,

** In communications engineering, and especially applicable to Radio Amateur working,

bandwidth is an important consideration in limiting interference which a transmitter may cause
to other installations. A signal which varies its amplitude slowly and only by small amounts
generates few harmonics and they have relatively small amplitudes; a signal which varies
rapidly and by an appreciable amount generates a large number of large-amplitude harmonics.

** In particular those signals which change amplitude abruptly contain high-order harmonics
which can have very large amplitudes and intrude into a large slice of the frequency spectrum.
Thus to keep interference to a minimum - to restrict the frequency-band over which the signal
spreads - the basic requirement is to ensure that signals do riot change too rapidly.

** Of course, when transmitting in Morse code, it is not possible to either close or open a key
slowly; when speaking into a microphone it is not possible to avoid those sharp explosive
sounds (expletives) which cause transients. These matters are dealt with electronically by
"restricting the bandwidth"; when high-order harmonics are removed then sharp-edged
waveforms become softened.

** For W/T working it is mandatory to fit so-called key-click filters which prevent a too-rapid
change between transmitter-on and transmitter- off. For R/T working the audio amplifier must
be suitably restricted in its bandwidth to ensure that it neither passes nor produces those high-
order harmonics. See also under 4.2.5 Low-pass Filters.

1.14.3 Bandpass Arrangements

In radio-frequency amplifiers tuned circuits are used as impedances and these have a
bandwidth which is determined by the circuit Q-factor ; see under 1.13, 4.2 and 4.6 . High-Q
circuits are relatively loss-free and produce high dynamic impedances but they restrict the
bandwidth. Low-Q

circuits are lossy and produce low dynamic impedances but they offer wide bandwidths.

Apart from the obvious compromise required there is a further complication mentioned
under 1. 13. 2 that the circuit Q - and therefore the bandwidth - varies as tuning adjustments are
made. One expensive cure for this problem is to arrange that the tuning control varies
simultaneously both the capacitor and the inductor. If the trick is made to work then the LC
product is varied to change the frequency while the L/C ratio is varied to correct the bandwidth
and/or the dynamic impedance.

A cheaper method is to use so-called bandpass circuits; see under 2.7.3 . Basically there are
two types;

In the first of these the signal is passed through a succession of resonant circuits whose
resonance frequencies are staggered across the required band. The alignment of such circuits is
not easy but they can offer a substantially constant response and they are relatively independent
of Q-values. Because most of the tuned circuits are not operating at their resonance
frequencies, and therefore not providing their maximum magnification, the signal suffers
attenuation as it passes through. These are known as stagger-tuned circuits.

The second type is the bandpass-coupled circuit in which two or more resonant circuits, each
tuned to the centre frequency of the required band, are coupled together. The coupling may be
magnetic (transformer) or via a common impedance which can be either inductive, capacitive or
a combination of both. A common-resistor coupling is possible but this lowers the Q.
In these circuits not only is energy circulated between the capacitor and inductor of each
resonant circuit but it is also interchanged between the coupled circuits. The effect is to produce
a constant response over the required bandwidth but, with each circuit providing maximum
magnification, the signal suffers less attenuation. The Q-values of the resonant circuits must be
well matched.

Circuit Q-values are dependent on circuit losses and these are not easily predicted; it is
usual to design the resonant circuits with too-high a Q and to add damping resistors as
necessary.

Although bandpass circuits produce a substantially-constant response over a band of

frequencies the price exacted is that the response falls rapidly at the edges of the band. The
resultant phase-distortion could cause problems but these steep skirts in the response curve is a
desirable feature in receivers where they effectively remove unwanted signals.

** The nature of bandpass circuits means that they are only really practical for fixed-
frequency working; to make them tunable over any useful range requires constant control of Q,
dynamic impedance, bandwidth and coupling factors. The problem is overcome in radio
receivers by using thesuperheterodyne principle; in such receivers the majority of the circuit
operates at a fixed frequency (known as the Intermediate Frequency or IF) and any wanted
incoming signal is converted to that frequency in a frequency-changer stage. See also under 2.
7 and 4.6.

END OF LESSON 10

* * * * * * * * * * * *

QUESTIONS

1. An Inductor and a Capacitor are connected together. Say briefly why this arrangement is often
considered as two different circuits ?/p>

2. Why and when would you beware of dangerously high voltages in such an arrangement?

3. There is a requirement for a trap to attenuate a signal at a frequency of 1-MHz in a circuit

which operates at an impedance of 1,000 ohms. Two combinations of L and C are given below
both of which resonate at (approx) 1 MHz ; state which you would choose to form a high-Q
trap.

(a) 0.003 μF and 8 μH each offer 50 ohms at 1-MHz

(b) 5 pF and 5 mH each offer 30,000 ohms at 1-MHz.
Give the Q-factor you would expect to achieve.

(Think carefully about this one - you are unlikely to meet this in the R.A.E.)

4. What is the purpose of a notch filter ? 5. What problems arise from attempts to make a
resonant circuit adjustable to a range of resonance frequencies ?

6. A transmitter and receiver are required to operate at the Amateur frequency of 3.66 MHz
using Morse code. Why must both pieces of equipment have a bandwidth that will accommodate
more than just the keyed carrier on 3.66 MHz ? 7. Why must the bandwidth of equipment be
limited to just that range which is required for its specific purpose

8. Constant performance over a given bandwidth is achieved by use of bandpass circuits. What
are the major advantages and disadvantages of using either stagger-tuning or coupled circuits ?
(Not for R. A. E.)

9. Variable-tuning circuits are a particular disadvantage in receiver design. How is this problem
overcome?

Back to Top of Page

LESSON 11
1.15 INTERNAL RESISTANCE

1.15.1 Introduction

Under 1.3.3. in Lesson-1 it was shown that a generator cannot deliver current until its
terminal voltage has fallen because current cannot flow between points of equal potential; the
terminal voltage falls according to the current-demand. This phenomenon introduces a problem
because the current that will flow through a given load depends on the value of the terminal
voltage while that voltage itself depends on the current-flow.

** The solution to this problem is elegant and yet absurd. A circuit is invented which behaves
in exactly the same manner as the practical generator but which uses only a theoretically-perfect
generator. Such a circuit (which cannot possibly exist in practice) is known as an equivalent
circuit. The generator used in this equivalent circuit is theoretically untroubled by problems of
internal equilibrium and so its terminal voltage does not fall when current is demanded. To
form the equivalent circuit the theoretical generator is degraded (i.e. made normal) by adding
theoretical losses until the performance of the combination matches that of the practical
generator.

The idea of the perfect generator causes the problem to split into two separate solutions
which, in fact, are but two ways of looking at the same phenomenon. Either the perfect
generator can be made to supply any required current without its terminal voltage changing or it
can be made to supply any required voltage without the current changing. It cannot be asked to
keep both parameters constant because that would severely limit the usefulness of the generator.

Perfect generators of the first type are called constant-voltage generators (or cv-generators)
and the second type are called constant-current generators (or cc-generators)

1.15.2 The Constant-voltage Generator

 Fig.29 shows a cv-generator (note the
symbol) connected through a resistor labeled
r int to the output terminals of a practical
generator; a load resistor RL is shown
connected externally. The resistor r int which is
shown "inside" the generator is always known
as the internal resistance of the practical
generator.

Without the connected load the supplied

current I must be zero and so the p.d.
developed across r int must be zero also.
Under no-load conditions therefore the
voltage at the terminals of the practical
generator is that of the constant-voltage
generator. This voltage is referred to as
the emf (electromotive force) of the practical
generator and is given the symbol E. The
symbol V is reserved for the terminal voltage
of a generator when current is flowing and
also for the p.d. developed across a load
(caused by the transfer of electrons).

The purpose of the resistor r int in the equivalent circuit is to ensure that, as current is
demanded, so the terminal voltage (of the equivalent circuit) falls in imitation of the terminal
voltage of the practical generator which it represents.

** The only way to assign a value to r int is to measure it experimentally; the values of V are
noted for different values of current-flow when the value of r int can be found using Ohm's Law
as

r int = change of V
change of I

** The value of E can be measured at the terminals with zero load current thus, once the value
of r int has been measured, it is possible to calculate the current for any given value of load
resistor (RL) from the equation

E = I x (RL + r int )

(the current multiplied by the total resistance around the circuit).

** Once the value of I has been determined it is possible to calculate the p.d. across the
internal-resistance; the terminal voltage is then found by subtracting that p.d. from the emf E.
Alternatively the value of V can be found directly by multiplying RL x I .

** Yet again the two resistors RL and r int form a potential divider across the emf; part of the
total voltage appears across r int and the remainder appears across RL. From this it is an
exercise in fractions to deduce the voltage across the load
 V = E x RL/(RL + r int )

** For generators which produce ac the internal resistance becomes the internal impedance
which has the complex form (r int + jX int ); see Lesson-8. It is important to remember that,
although the effect of internal resistance is real enough, its existence is pure fiction - a
mathematical convenience. For active systems such as amplifiers r int may be referred to as
the output impedance or as the source impedance of the system.

(The same idea in reverse leads to the concept of input impedance which describes the
current which flows into a device as a result of applying a given input voltage.)

In separating the cv-generator from the failings of the real generator it has been stipulated
that the cv-generator shall have an internal-resistance of zero; this is the reason its terminal
voltage does not fall under load. It is this aspect of the equivalent circuit which is absurd. The
Maximum Power-transfer Theorem states that a generator delivers maximum power into a load
which exactly matches its internal impedance; this cv-generator therefore must

deliver its maximum power into a short-circuit (zero ohms) and zero output power does not
make it an ideal power source. Nevertheless the trick provides a legitimate way to calculate the
performance of circuits and that is a prerequisite for design work.

Consider again the equality above in which the e.m.f. is equated with the product I x (RL +
R int ) ; in mathematicians' language we say that, if the value of r int is small compared with the
value of the load, then r int can be neglected . In that event the terminal voltage does not fall
when current flows. We say that r int is swamped by the load; it has negligible effect on the
current.

In practice therefore, as long as a load is considerably greater than the internal resistance of
the generator, then the generator tends to behave as though it were a perfect constant-voltage
generator (zero internal impedance). This is an important consideration in circuits where it is
required to maintain an undistorted voltage-waveform ; before pursuing the matter further
however it is best to consider the alternative equivalent circuit which arises when the internal-
resistance is the greater and the load becomes negligible.

1.15.3 The Constant-current Generator

Fig. 30 shows a cc-generator (again note the symbol) which is connected directly to the
output terminals of the practical generator which it represents. Be reminded that this theoretical
generator produces an unvarying current without the intervention of an emf.

For the internal resistance to be much larger than any load it must have an infinitely-large
value. But an infinitely-large value means that r int is not there and the generator would be
required to produce an output when it is not connected to the load.
 ** The solution (in this theoretical circuit) is
to place the internal resistor across the
generator ; its function now is to absorb, or to
divert, any of the constant-current that the load
cannot accept. This perhaps odd arrangement is
perfectly legitimate because we are trying to
invent an equivalent circuit that will exactly
imitate the behaviour of the real circuit.

In this equivalent circuit the distinction

between E and V is meaningless because this
theoretical generator produces a current (as
distinct from producing an emf) . The current
from the generator (I) divides according to the
relative impedances of the two available paths.

Using the parallel version of Ohm's Law the

load current IL is this total current I multiplied
by r int / RL + r int ) . If RL is negligible then all
the current is flowing through the load.

The voltage that is developed across the Load is

the product

I x RL

In practice as long as the internal resistance is much larger than the load then the generator
tends to behave as though it were a perfect constant-current device with infinite internal
impedance. This is an important consideration in circuits where it is required to maintain an
undistorted current waveform.

The concept of a cc-generator is in fact as absurd as the cv-generator as can be seen by again
referring to the Maximum Power-transfer Theorem. This generator must deliver its maximum
power into a load that is not there but again it provides a means of performing circuit
calculations.

1.15.4 Voltage and Current Sources

Any practical generator behaves either as a cv device or as a cc device determined by the

circuit into which it is driving. A signal source is referred to as a voltage source if its operation
tends to that of a cv- generator because its output impedance is low in comparison with the load
when variations in the load impedance do not affect the voltage waveform.

The same signal source would be referred to as a current source if its operation tended to that
of a cc-generator because its output impedance was high in comparison with the load ;
variations in the load impedance would not affect the current waveform.
 A practical application of this is shown in Fig. 31 where a variable resistor is connected as
a pot'meter to vary the drive to an amplifier stage.

Behaviour of a valve is determined by the voltage-waveform which is impressed between its

grid and cathode but the similar behaviour of a (bipolar) transistor is determined by the current
which is injected into its base-emitter path.

Diagram (a) shows that the input to a valve has the nature of a capacitor and the reactance of this
reduces as the frequency is increased. To avoid distorting the voltage waveform at the grid of
the valve it is imperative that it be driven from a source whose internal impedance is low
relative to the overall impedance between grid and cathode. The gain control must therefore be a
potential divider which taps-off a portion of the applied voltage.

Diagram (b) shows that the input to a transistor offers a dc path and so the device can be
expected to draw current from the source. What the diagram fails to show is that the resistance
of the base-emitter path depends on the magnitude of the applied signal-current and so the input
resistance varies in sympathy with the signal waveform. To avoid distorting the current
waveform between base and emitter it is imperative that the transistor is driven from a source
whose internal impedance is relatively high. The gain control for this simple amplifier therefore
must be a current divider.

As shown the signal is split between two alternative paths and the pot'meter is connected as
a differential control which shunts unwanted signal away from the transistor. The series
resistor R int is inserted as necessary to prevent the pot'meter short-circuiting the source at
minimum-gain setting.

The arrangement of (a) is sometimes seen controlling a transistor stage and the error has two
consequences:

(a) operation of the control to increase the gain causes the gain first to rise and then to fall

(b) the transistor is driven from a source-impedance represented by the lower half of the
pot'meter ; consequently at low-gain settings the drive is converted to cv operation and
distortion ensues.

Note however that the application of feedback to the transistor stage may convert it
effectively to voltage-operation and so use of arrangement (a) would be correct.

1.15.5 Practical Significance of Internal Impedance

 The concept of internal impedance was introduced to make it possible to perform circuit
calculations. To demonstrate its impact in the practical world consider a unit intended for test
purposes which is known as a Signal Generator (or S.G.) . This produces a signal at a frequency
which can be adjusted and also at a level which can be adjusted accurately over a range of
values.

From what has been said above it is clear that the actual signal which the test equipment
injects into a load must depend on both the input-impedance of that load and the output-
impedance of the S.G. and this could make nonsense of any attempt to calibrate the output level
of the generator. In practical work all generators are considered as voltage generators unless
otherwise specified and so manufacturers state S.G. outputs in terms of an emf and internal
resistance.

For example it might be given as an emf of 1 volt at 50 ohms. Because an emf' is stated we
know that, connected to a 50-ohm resistor, this generator will provide an output of 0.5 volts
when all the output attenuators are set for zero attenuation (maximum output). It is a bit
inconvenient to work relative to 0.5 volts and so most S.G.'s are built with an emf of 2 volts.

Sometimes a generator may be quoted as "delivering 1-volt across 50 ohms"; this is a

statement of terminal voltage when the Instrument is correctly matched and so we can deduce
that it affords an emf of 2-volts with an Internal resistance of 50 ohms.

However not all equipment offers an input impedance of 50 ohms and it is an expensive
hobby to keep a flock of S.G.'s to provide a range of output impedances. It is perfectly
legitimate to connect a 50-ohm generator to say the 75-ohm input of a receiver - the Designer
will have ensured that this does not affect the operation of the device - but the Operator must re-
calculate the voltages that will appear across the receiver input-terminals. This is a simple
matter ; go back to the value of the emf, note the output impedance (r int ) and call up the
equation given above wherein

V = emf x RL/(RL + r int )

Once this value has been established for any convenient setting of the output-attenuator(s) all
other outputs are changed in the same ratio.

An alternative is to connect the S.G. to the receiver via a matching pad which is a network
of resistors whose input resistance matches r int and whose output resistance matches the input
resistance of the receiver. Of course allowance must be made for the attenuation caused by the
pad which is most easily done by directly measuring the voltage across 50-ohms without the pad
and the voltage developed across (say) 75-ohms with the pad.

1.15.6 The Constant-resistance Network

In the above discussion the internal impedance of the S.G. has been assumed to be a pure
resistance but the argument is not in any way affected if that internal impedance is complex.
Most definitely however that is not true for a practical situation ; if the internal impedance is
not resistive it is imperative to know its exact value for all possible output frequencies so that
the true terminal voltage can be calculated. To prepare a full calibration is not practicable.
 Fortunately it is possible to make the output impedance substantially resistive for all
frequencies by adding corrective components which turn the output circuit into a constant-
resistance network. The underlying mathematical derivation is not required here but the circuit
arrangement is shown in Fig. 32. For example, if the output is measured as an inductor in
series with a resistor, then that output circuit is shunted with a capacitor in series with a resistor
of the same value such that

L/C = R2

where L is in Henrys, C is in Farads and R is in Ohms. When this is done correctly then the
output impedance is resistive at all frequencies.

Such a correcting network is built into the instrument which is why its r int can be quoted as
(say) 50-ohms. (Note that the equation does not include a frequency term; i.e. neither w nor 2πf
appear

This same circuit trick is called upon to obviate the effects of reactances where they become
a nuisance. For example electrical equipment causes radio interference because sparks are
produced when an inductive circuit is interrupted. When the device is incorporated into a
suitableconstant-resistance network the circuit inductance is neutralised and interruptions of the
current no longer cause sparking and consequent radiation.

1.15.7 The Maximum Power-transfer Theorem

This theorem states that a source delivers the maximum possible power to a load which has
exactly the same value as the source's internal impedance; i.e. to a properly-matched load.

With reference to the cv equivalent circuit of Fig. 29 the current which flows through a
voltage generator and its load is given by E/ (RL + r int ) which is mathematical talk for emf
divided by the total circuit resistance. The power dissipated in the equivalent circuit comes in
two packages:

(a) power in the load = [Link]

(b) power in the internal resistance = I2.r int

When the load has a very low value the voltage developed across it becomes very small and the
power it dissipates becomes negligible. When the load has a very high value the current which
flows through it becomes small and so once again the power dissipated by the load becomes
negligible.

Somewhere between these two extremes the power dissipation rises to a maximum value.
For those who enjoy algebra it is possible to show that this maximum occurs when RL =
r int . If load and internal-resistance have the same value (they are matched) then they
dissipate equal powers and so it follows that a generator can deliver only half the. theoretical
generated-power and only then when the output is properly matched.

END OF LESSON 11

* * * * * * * * * * * *

QUESTIONS

1. To make calculations on electric circuits it is necessary to overcome the problem of the

variable output-voltage from a generator. The solution to this problem is to substitute a
theoretical circuit for the real circuit. What is the basis of this theoretical circuit ?

3. Constant-voltage and constant-current generators are both theoretically perfect generators. In

practice the concept gives rise to both voltage-sources and current-sources. In what way - if
any - do these two sources differ. (Not a question for R.A.E.)

4. A battery is to be used in the design of a circuit but its internal resistance is not known. On
test it yields the following figures:

5. The internal impedance of a generator is given as 2 ohms and its open- circuit terminal
voltage is 10 volts.
When a lamp is connected to the generator the terminal voltage falls to 8 volts. What is the
power dissipated by the lamp?

6. The input impedance of an amplifier is 30 ohms. The amplifier is driven from a signal
generator with a source impedance quoted by the manufacturer as 75 ohms. Given that the S.G.
produces 1-volt across 75-ohms what is the voltage it will produce at the input to the amplifier ?

Back to Top of Page

E
 FUNDAMENTALS PART 2
BASIC RADIO THEORY

LESSON 1
2.1 GENERAL

Fundamentals-1 dealt with basic Physics: this second Part of Fundamentals deals with
basic aspects of Radio Engineering. Radio is a word which originated in America and which
now replaces the original English word "wireless"; it refers to the transmission of information
between two points without the use of connecting wires.

There is of course nothing new in transmitting messages over long distances; some of the
early mechanisms include runners, horse-riders, drums, pigeons, semaphore machines, fire,
smoke and the heliograph but the discovery of electricity brought the electric Telegraph. Two
or more stations were connected by a pair of continuous conductors and, by means of
a Telegraph Key, current was switched on and off according to an agreed code. With each
change of current a Telegraph Sounder emitted a click.

The problems of the Telegraph were mainly that messages could be passed only between
those stations which were connected by wires and those wires were vulnerable both to the
weather and to accidents or sabotage. Marconi's famous experiments with electromagnetic (e-
m) waves opened the way to world-wide communication because connecting wires were no
longer necessary.

Marconi did not invent radio but built on the theoretical work of the physicist James Clerk
Maxwell who first predicted the possibility of e-m waves. That such waves can exist is one
thing; to use them for practical communication required a whole new technology to create such
waves and then to utilise them.

** A radio communication chain requires four separate parts (see Fig. 1)

 >>>>>>>>>>>>>>>>>>>> PAGE 2 <<<<<<<<<<<<<<<<<<<<

(iii) a propagation path over which the e-m wave travels from the sending position to the intended
reception point. This is the least reliable part of a radio-communication chain and it is an
unfortunate fact that radio techniques do not automatically provide communication between two
selected points

(iv) a Receiver whose function is to select the required signal from the many which an aerial
system picks up, to amplify that selected signal and then to recover from it the information.

2.2 TRANSMITTERS

2.2.1 Introduction

A Transmitter is an electrical ac-generator which is designed specifically to excite an aerial

system into radiating an e-m wave. It is possible to connect a microphone-amplifier combination
directly to an aerial but the resulting range of communication is limited to a few feet only. One
exception to this is the Induction-loop which provides communication at audio frequencies with
any point which is enclosed by the loop. This type of communication however is limited to one-
channel operation because it is not possible to separate two or more signals.

** For e-m waves to propagate over reasonable distances it is necessary to use frequencies
which are above the audio range ; however, efficient aerial systems are impossibly large until
the frequency rises above several hundred kilohertz. Each communication channel is allocated a
specific frequency and so it becomes possible to separate and to identify them (see under
1.13: Resonance).
 ** The information to be passed through each communication channel is impressed on that
channel's individual radiation which, because it "carries" the information, is known as a carrier
wave or simply as the carrier . The information signal is referred to as the modulation.

Modulation of a carrier-wave can take many forms and, in setting up a modulation system, it
is of course necessary to consider the means of recovering the information - of demodulating
the carrier . This process is also referred to as detection.

** A transmitter therefore is required to generate a carrier-wave at a specified frequency and to

provide facilities for modulating that carrier with information. There are two important
restrictions:

(a) it must maintain that frequency at a constant value else it will be lost to the receiver and will
intrude on other transmissions.

(b) it must not produce signals at frequencies other than its allotted carrier frequency (known
as spurious signals) because these too would interfere with other transmissions.

** Every transmitter has five main parts which are illustrated in Fig.2.

(i) The Master Oscillator or M.O.

>>>>>>>>>>>>>>>>>>>> PAGE 3 <<<<<<<<<<<<<<<<<<<<

(ii) A Modulator

(iii) The Power Amplifier or P.A.

(iv) Filter and Aerial-matching Units

(v) The Power Supplier

2.2.2 Master Oscillator

The first transmitters were called "spark transmitters" because they shocked a resonant LC
circuit into trains of damped oscillations by discharging a series of sparks across a high-voltage
spark gap; oscillations are dealt with in detail in Part 4: Electronic Oscillators. Sparks however
are noisy things, electrically speaking, and the immense disturbance caused over a wide range of
frequencies severely limited the number of transmitters able to operate simultaneously.

** The invention of the triode valve with its ability to amplify signals made it possible to
maintain a resonant circuit in continuous oscillation. In an arrangement which utilises both
positive-feedback and negative-feedback (see under 1.11 : Amplification in Lesson 5) the valve
makes it possible to exercise tight control over the oscillatory process and so produce a stable
single-frequency continuous sine-waveform; see Fig. 3. As a result the spark transmitter was
outlawed and continuous-wave (c.w.) operation is the only accepted standard.

** The most important function of the Master Oscillator is to determine the desired frequency
and, once set, to maintain that frequency constant. Stability of frequency is best achieved by
utilising the very-rapid change of phase which occurs between voltage and current in a high-Q
resonant circuit as it passes through resonance but, in practice, it is not all that easy to maintain
the constant Q-value which is required. The greatest enemy is changes of temperature which
cause:

>>>>>>>>>>>>>>>>>>>> PAGE 4 <<<<<<<<<<<<<<<<<<<<

(a) changes in the physical size of components which thus change their values of L and C

(b) changes in the properties of the materials used in constructing those components; this
affects the values of L and C and also the leakage currents in semiconductor devices
(c) changes in the inter-electrode spacings of valves which change the circuit stray
capacitances and so move the resonance frequency

(d) changes in the characteristics of valves and semiconductor devices which, by changing the
circuit amplification, effectively changes the equivalent-circuit losses arid so affects the values
of Q

** Variations in the power-supply voltage(s) also cause changes in device characteristics and,
in semiconductors, changes in leakage currents and these too add their quota to the overall
frequency drift.

>>>>>>>>>>>>>>>>>>>> PAGE 5 <<<<<<<<<<<<<<<<<<<<

** The Q-value of an LC circuit is dependent also on the effective load which is presented to
it (the damping) and this can be changed either directly by variations in the load itself or through
changes in stray couplings to local bodies.
 ** A M.O. therefore requires care in both its design and construction:

1. The resonant circuit is kept away from heat generating components

2. The resonant circuit is enclosed in a ventilated screening-box which ensures that stray
couplings remain constant

3. The resonant circuit is coupled only loosely to the maintaining device (either valve or
semiconductor) so that changes in its characteristics have minimum effect; this also results in
the circuit being maintained at a minimum level of oscillation thus eliminating harmonic
generation which also causes drift problems

4. The resonant circuit is coupled to the load by means of a buffer amplifier whose purpose is to
provide isolation from load variations

5. All components associated with the resonant circuit are rigidly mounted to prevent changes of
form being caused by mechanical shock

6. Power supplies for the maintaining device are carefully stabilised and are filtered and
screened to prevent their providing an access path for intruding signals

7. Metering arrangements, which are necessary both for maintenance work and for the proper
adjustment of the M.O., are also carefully filtered and screened.

** The majority of transmitters are required to work over a range of frequencies and so the
M.O. is usually designed to be adjustable as discussed above ; this can be achieved by arranging
for either the inductor or the capacitor (or both) to be adjustable. In most instances the capacitor
is made variable because that is the easiest and cheapest method ; with reference to Section
1.13 however this method leads to variation in the circuit Q-values and so the oscillator stability
varies across the tuning range if that range is excessive. In such transmitters the M.O. is referred
to as the variable-frequency oscillator or the V.F.O.

** Transmitters which are required to operate on a single fixed frequency can achieve very
high orders of stability and waveform-purity because the design can be optimised. Generally the
LC resonant circuit is replaced by a mechanical resonator in the form of a thin slice of quartz
crystal ; other forms of resonator are available today but the crystal probably provides the
greatest accuracy and stability. These devices have the useful property of deforming
mechanically when subjected to an electric voltage and, in reverse, of generating an electric
voltage when they are mechanically deformed. Crystals can offer effective Q-values up to
30,000 but, with special mountings and careful circuit design, they can go as high as a half-
million.

** When loosely-coupled to a maintaining circuit and lightly driven crystal-oscillators offer

great immunity from frequency drift and this can be improved further by mounting the crystal in
a temperature-controlled oven. Transmitters

>>>>>>>>>>>>>>>>>>>> PAGE 6 <<<<<<<<<<<<<<<<<<<<

which use a crystal in their M.O. are said to be crystal controlled and the M.O. is referred to as
a Xtal. Osc.
 ** The range over which crystals can operate is limited. At the lowest limit around 2 kHz the
crystal slice becomes a relatively massive bar and difficulty can be experienced in persuading it
to oscillate. At the highest frequencies the slice becomes thin and fragile and is easily destroyed
by over-driving. By cutting crystals to operate in so-called overtone modes (i.e. harmonic modes)
it is possible to use them up to around 25-MHz. This range can be extended by using frequency-
multiplying techniques and/or frequency-translating techniques which will be described fully
under 2.2.4 and 2.2.5

** A continuous sinewave can also be generated by a technique known as frequency

synthesis which does not require a resonant LC circuit or other device; however a crystal
oscillator usually forms part of the arrangement because it provides the necessary frequency
stability . In use as a transmitter drive these circuits are simply referred to as a V.F.O. They
offer ease of adjustment, good frequency stability, good waveform and compactness but they do
suffer from background noise. Also their frequency-adjustment is not continuous as with an LC
arrangement ; the frequency changes in discrete intervals although the size of the interval is a
matter of expense.

** The oscillator section of a transmitter plus any follow-on buffers, amplifiers and multipliers
is often referred to as the . When a transmitter is used in conjunction with a separate receiver it
is necessary to fit a drive-switch which, in the Receive position, shuts off the drive section.
Without this facility radio- frequency (rf) signals leaking from the transmitter block the receiver
to signals on the operating frequency. In Transceivers the transmitter and receiver are combined
in a single unit ; the drive section is frequency-shifted and incorporated into the receiver when
not transmitting and so the problem does not arise.

2.2.3 Frequency and Wavelength

** The above Section on the Master-oscillator has described the Carrier-wave in terms of its
frequency ; i.e. the number of complete cycles of oscillation that take place each second. Carrier
signals are often referred to in terms of wavelength and the purpose here is to explore the
relation and meaning of these two terms.

Light is an electromagnetic radiation which differs from radio waves only in that it has a
much higher frequency. Early experiments which attempted to measure the speed at which light
propagates produced the surprising result that the velocity of light appeared to be a constant ;
i.e. the speed at which a light source moves or the speed of any observer receiving that light
does not affect in any way the speed with which the light propagates from the Source to the
Observer.

The meaning of this strange result was the basis of a theoretical discussion by one Albert
Einstein who finished up with his oft-quoted Theory of Relativity which, fortunately, does not
have any place in the R.A.E. This constant speed at which all electromagnetic radiation
seemingly propagates is given the symbol e and has the value 3 x 108 metres/second (or roughly
180,000 m.p.h.).

>>>>>>>>>>>>>>>>>>>> PAGE 7 <<<<<<<<<<<<<<<<<<<<

Electromagnetic waves at radio frequencies are generated by an aerial when it is excited by

an alternating-current . The closely coupled electric and magnetic fields around the aerial
constantly build to maximum values, fall to zero and then build again in the reverse direction
and this repeating pattern propagates away from the aerial.

** Given that the carrier is being generated at a constant frequency (f) then successive peaks
are generated at constant time intervals and this must mean that the peaks of the propagating
field are separated by equal and constant distances. The distance between successive cycles -
the distance traveled by an e-m wave during the time of one cycle - is known as
the wavelength (λ).

** As the frequency is increased so the time interval for each cycle is decreased and so the
wavelength must also decrease. This results in a simple relationship between frequency and
wavelength linked by the constant velocity of propagation:

f x λ = e

Thus given either the frequency in cycles/second (Hertz) or the wavelength in metres the
alternative form can be calculated by dividing the known value into 3 x 108 .

** Questions involving this calculation are very likely to arise in the R.A.E. and the most
likely cause of failure to earn easy marks is to forget that the basic units are cycles and metres.
kHz should be expressed as f x 103, MHz as f x 106 and wavelengths expressed in metres.

END OF LESSON 1

*******************

QUESTIONS

1. Fig.1 shows the basic communication chain. What are the functions of the switches ?

2. The propagation path was said to be the least reliable link in the communication chain. What
particular facility must it offer that is most likely to cause a failure in communication ?

3. Detail the parts of a transmitter and state briefly their function.

4. What is the difference between a VFO and a [Link]. 7 ?

5. On what equipment would you expect to find a "Drive" switch and for what purpose would
you use it ?

6. Confronted with a working transmitter how would you determine whether it was fitted with a
standard VFO or
a Synthesiser without going inside ?

7. What is the frequency of a carrier operating on 20.2 metres ?

What is the wavelength when the carrier frequency is (a) 805 kHz & (b) 15.15 MHz 7

>>>>>>>>>>>>>>>>>>>> PAGE 7(again) <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 LESSON 2
2.2.3 Modulation

** The process of loading a carrier wave with information is called modulation and the
circuit arrangement which carries out the task is called a modulator. The purpose is to vary one
of the characteristics (one of the parameters) of the carrier-wave in accordance with the form of
the information signal.

** The simplest modulation system switches the carrier on and off in a series of long and short
bursts to form the internationally-accepted Morse Code. The message is spelled out one character
at a time ; although laborious the code can be both sent and read by experienced operators as
easily as they read the human voice. Because of the distinctive rhythms of the characters a
message can be read against a background of noise that completely obliterates all other systems.

An early version of this type of modulation was known as interrupted continuous-wave

(I.C.W.) in which the carrier was regularly interrupted by a buzzer or a tonewheel mechanism. If
the interruptions occurred at audio frequency then the presence of the carrier appeared as an
audible tone at the output of a simple receiver. A cw carrier produces only dc at the receiver
output and additional components are necessary to make the incoming signal audible. (See
under 2.6 Receivers.)

CW-working is often referred to as carrier-wave working but the initials stand for continuous-
wave as distinct from the discontinuous wave-trains of the early spark transmitters. To the
pedant the unmodulated wave cannot be a "carrier" but it can be argued that morse is indeed a
form of modulation because it does produce sidebands (described below). In practice confusion
does not result and the exact meaning of the nomenclature is not important.

** There are four basic ways in which a carrier can be modulated:

(a) by variation of the carrier amplitude - amplitude modulation(A.M.)

(b) by variation of the carrier frequency - frequency modulation(F.M.)
(c) by variation of the carrier phase - phase modulation (Ph.M,)
(d) by pulsing the carrier - pulse modulation(P.M.)

(a) Amplitude Modulation is a system in which the amplitude of a carrier-signal is varied

according to the form of a modulating signal; see Fig.4. Maximum possible depth of
modulation (100% modulation) occurs when the carrier amplitude varies from twice the
unmodulated value down to zero; modulation is normally limited to around 80% to avoid
distortions which arise from the limitations of circuitry and not from the modulation process
itself.

** Where the modulating signal consists of a single frequency (or tone) the resulting
modulated carrier is termed a modulated continuous-wave (M.C.W.) and this can be used for
Morse-code transmissions either by keying the carrier or by keying the modulation. The
advantage of keying the modulation is that the carrier remains at a constant amplitude during the
"spaces" and serves to quell the noise that otherwise appears at the receiver output but the
method is seldom used today except occasionally with F.M. ICW operation mentioned above is
a special instance of M.C.W. used with A.M.
 ** To vary the amplitude of a carrier-wave it is necessary to cause a small amount of
distortion from cycle to cycle and so a modulated carrier cannot be

>>>>>>>>>>>>>>>>>>>> PAGE 8 <<<<<<<<<<<<<<<<<<<<

>>>>>>>>>>>>>>>>>>>> PAGE 9 <<<<<<<<<<<<<<<<<<<<

regarded as a pure single-frequency carrier. In fact, for every frequency in the

modulating waveform, two extra near-carrier frequencies are produced one on each side of the
carrier proper ; if fc is the carrier frequency and fm, is the modulating frequency then the
additional two frequencies produced by A.M. are fc + fm and fc - fm.

** These two extra frequencies are properly referred to as side frequencies ; a group of side
frequencies either above the carrier frequency or below it are called a sideband and they are
designated either as the upper sideband or the lower sideband. In common parlance however they
are usually referred to collectively as the sideband frequencies or simply as the "sidebands"

** In a 100% modulated A.M. carrier one-third of the total emitted power resides in the
sidebands and each sideband contains all the modulated information . It follows that the
full double-sideband amplitude-modulated signal wastes 5/6 of its total power and that it occupies
a band of frequencies which is twice that strictly necessary.

** Modulator circuit techniques make it possible to remove the carrier and leave only the two
sidebands; in such a suppressed-carrier system the total emitted power is divided equally
between the sidebands but still 50% of the power is wasted in duplication and the bandwidth is
still twice that necessary.

** A suitable filter can be placed after the Modulator to remove either the upper or the lower
sideband and the result is known as a single-sideband (ssb) emission. In such a signal all the
emitted power is utilised to carry information and the bandwidth is half that of the parent A.M.
signal.

** In Radio-telephony (R/T , voice or 'phone) ssb is now the only form of A.M. which is
tolerated . (Note that, on the Amateur Bands, it is accepted that the lower sideband is used for
160, 80 and 40 metres (1,8, 3.5 & 7.0 MHz) and the upper sideband for all other transmissions.

** Sidebands are produced only when information is modulated on to a carrier; when the
modulating signal is zero then only the carrier signal is present. In ssb operation that carrier is
removed and so it is characteristic of ssb that the transmitter output remains at zero until the
Operator speaks into the microphone. Thus a power meter or an s.w.r. meter, which usually are
connected between transmitter and aerial system, remains quiescent until the Operator begins to
speak.

** (b) Frequency Modulation is a system in which the carrier amplitude is maintained at a

constant value but the frequency is varied according to the form of a modulating signal. There is
a clear limitation to this system in that the carrier frequency cannot be allowed to approach zero
; in fact it cannot be taken very far at all expressed as a percentage of the normal (unmodulated)
frequency. Thus F.M. tends to be used only with high carrier frequencies.

** The depth of modulation is expressed as a modulation index which indicates the relation
between the frequency of the modulating signal (i.e. the rate at which the carrier frequency is
being shifted) and the extent of the frequency shift at the peaks of the modulating waveform.
The range over which the carrier frequency is swung is known as the frequency deviation
; see Fig. 4.

** Although the amplitude of a F.M. carrier is kept constant it is still true that distortion
occurs from cycle to cycle as the frequency is continuously varied and so this modulation
system too produces sidebands. Any modulation system must
 >>>>>>>>>>>>>>>>>>>> PAGE 10 <<<<<<<<<<<<<<<<<<<<

produce sidebands because that is where the information lies. The modulated carrier is no
longer a pure single-frequency waveform ; for every frequency in the modulating signal pairs of
sideband frequencies appear either side of the carrier. When the modulation-index is less than
0.5 the F.M. signal is similar to an A.M. signal but, as the deviation is increased, so more
sideband frequencies appear spaced at intervals which are equal to the modulation frequency fm
. The bandwidth of the modulated signal is approximately the deviation plus fm.

** With F.M. it is not possible to remove the carrier and so this system is not as efficient in its
use of power as a ssb system ; where large values of deviation are used the F.M. signal spreads
over a large portion of the frequency spectrum which is another reason why its use is generally
confined to the higher carrier frequencies (see under 1.14: Relative Bandwidth).

** Practical modulators are not as co-operative as one might wish and they invariably produce
A.M. as well as F.M. This usually is dealt with by using a limiter stage after the Modulator
which removes the A.M. (With A.M. systems a poorly-designed modulator is equally likely to
produce F.M. as well as A.M.).

** Despite their wide bandwidth and inefficient use of power F.M. systems play an important
part in communications because of their relative immunity to noise. Most forms of interference
cause changes of amplitude; because such changes are not significant when using F.M. the
receiver can pass an incoming signal through a limiter stage thus obtaining a carrier with a
constant (noise-free) amplitude with only F.M. remaining.

** There are two minor drawbacks to F.M. working which arise when the received signal level
becomes too small:

(a) noise takes over very abruptly as the limiter fails

(b) an unwanted signal on the same or close-by frequency suppresses the wanted modulation
and substitutes its own in a process known as Capture Effect

** The only practical way to use Morse with an F.M. system is to use M.C.W. and key the
modulation. Frequency-shift keying can be used in which the carrier frequency is shifted when
the key is closed ; this could be looked on as a form of F.M.

(c) Phase Modulation is a system in which the phase of the carrier is changed (its timing is
changed) according to the form of a modulating signal ; the frequency is held constant but the
carrier is shifted in Time. However it can be argued that the difference between phase
modulation arid frequency modulation is more a matter of mathematical definition and, in
practice, the greatest difference lies in the techniques used to achieve the modulation. A detailed
knowledge of Ph.M. is not required f or the R.A.E.

Suffice it to say that, to shift a carrier-wave in Time, it is necessary momentarily to either

stretch the waveform or to compress it and this involves a momentary change of frequency ; for
a continuously-varying modulation signal clearly the frequency change is continuous also. The
argument operates in reverse in that, to change the frequency of a carrier-wave, its timing (or
phase) must be changed. It is not proposed to go further than this simplified approach.
(d) Pulse Modulation is a system in which the carrier is pulsed ; i.e. it is generated in short bursts
very similar to an I.C.W. system except that the pulses occur at a rate well above the audio
range.

>>>>>>>>>>>>>>>>>>>> PAGE 11 <<<<<<<<<<<<<<<<<<<<

 >>>>>>>>>>>>>>>>>>>> PAGE 12 <<<<<<<<<<<<<<<<<<<<

Fig. 5(a)shows a pulse as it would be generated in an electronic circuit (note that the term
"impulse" is now obsolete) and diagram (b) shows the rf pulse or rf burst which results when a
pulse is used to modulate a carrier. Diagram (c) shows a train of such rf-pulses in the form of
a pulse-modulated carrier.

A pulse-modulated carrier can be used in a variety of ways and for a variety of purposes

(a) Pulse-amplitude Modulation (P.A.M.) is a system in which the pulse-amplitude is varied

according to the form of a modulating signal.

(b) Pulse-duration Modulation (P.D.M.) is a system in which the pulse duration (its length in
Time) is varied according to the form of a modulating signal.

(c) Pulse-position Modulation P.P.M.) is a system in which the positions (in time) of the pulses
are varied according to the form of a modulating signal.

(d) Pulse-frequency Modulation (P.F.M.) is a system in which the pulse repetition rate (p.r.f.)
is varied according to the form of a modulating signal.

(e) Pulse-code Modulation (P.C.M.) refers in general to any system in which the pulses are
modulated according to a predetermined pulse-code.

Pulse modulation is particularly useful in systems which seek to send several different
information signals simultaneously over a single channel ; such systems are known as multiplex
systems. In a time-division multiplex equipment the channel is shared between different
information signals on a time-rota basis and, using P.C.M., this can be done by modulating each
signal on to selected pulses in the train. For example, to carry three signals, the first information
signal would be modulated on to pulses 1, 4, 7, 10 etc.; the second information signal would be
modulated on to pulses 2, 5, 8, 11 etc.; the third information signal would be modulated on to
pulses 3, 6, 9, 12 etc. The pulses are sorted at the receiving point by use of a gate circuit which
directs pulses to the appropriate destination channels.

** Pulse-modulation techniques are permitted by the Amateur Licence with certain restrictions
that are laid down in the Licence document itself as well as in the Licence Regulations. The
matter is not examined for the R.A.E.

END OF LESSON 2

***************

QUESTIONS

1. What is meant by the term "Modulation" ?

2. Why does a modulated carrier-wave no longer behave as though it were a single-frequency

wave ?
3. A carrier on 1.8 MHz is modulated simultaneously by signals at 100 Hz (hum), 1 kHz and its
third-harmonic at 3 kHz. What are the various frequencies that emerge from the modulator
circuit ?

4. State the major advantage and the major disadvantage of an F.M. channel

5. When a transmitter is switched on it fails to produce an output signal. What is the probable
reason for this and how would you check the diagnosis ?

>>>>>>>>>>>>>>>>>>>> PAGE 13 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 3
MASTER OSCILLATOR

INTRODUCTION

** It is difficult to build high-frequency oscillators which are stable in frequency and so, in
general, master oscillators are designed to operate at low frequencies. To achieve good stability
and accuracy in setting they need also to be designed so that they cover only a small band of
frequencies. The output from a M.O. is then converted to the frequency range at which the
output is required

** Crystals have limited ranges of operation and so frequency-conversion is necessary also if

crystal oscillators are to be used above 25 MHz.

** The frequency-conversion process can be carried out in two different ways

namely frequency-multiplication or frequency-translation.

2.2.4 Frequency Multiplication

** Frequency-multiplication involves deliberately distorting the waveform of the M.O. output

so as to generate harmonic frequencies ; the required harmonic is then selected by a suitably-
tuned resonant circuit and then amplified in a tuned rf- amplifier. Such a multiplier stage is
driven by the constant-frequency signal of the M.O. and so, despite the waveform distortion, the
frequency stability is maintained. Note that the waveform within the M.O. must not be distorted
else frequency drifting must occur (effective reduction of Q).

** One limitation of this technique is obvious in that the required final frequency has to be a
harmonic of the M.O. frequency but this is not a problem when the M.O. is variable and so the
necessary fundamental frequency is obtainable.

** A second limitation is that the generated harmonics have to be kept within the frequency—
multiplying stage and not allowed either to appear in the final output signal or to radiate from
the equipment.
A ** A third limitation is that the frequency—multiplying action operates also on any side
frequencies that may be present which means that the modulating frequencies are multiplied also
; the actual effect depends on the modulating system in use :

(a) Effect of Multiplication on A.M. Signals

Consider a carrier of frequency 1 MHz which is amplitude-modulated at I kHz: the side

frequencies therefore are 999 kHz and 1,001 kHz . When such a modulated carrier is subjected
to a frequency multiplication of 10 the result is a carrier at 10 MHz flanked by side frequencies
at 9.99 MHz and 10.01 MHz . This represents a modulating frequency of 10 kHz and
demonstrates that the modulation also has been multiplied by 10.

** With A.M. therefore it is necessary to carry out all multiplying operations before applying
the modulation.

The same argument applies to any frequency-drift of the M.O. ; a change of 1% from 1
MHz to 1.01 MHz (admittedly a poor oscillator) would appear at the multiplier-output as a
change from 10-MHz to 10-1 MHz. This still represents a drift of 1% but the increased change
of frequency, as measured in cycles-per-second might well take the carrier outside the
bandwidth of the 10-MHz circuits.

>>>>>>>>>>>>>>>>>>>> PAGE 14 <<<<<<<<<<<<<<<<<<<<

(b) Effect of Multiplication on F.M. Signals

Consider a carrier of frequency 1 MHz which is frequency-modulated by a signal at fm hertz

with a deviation of I kHz (i.e. the carrier swings from 999 kHz to 1.001 MHz). Frequency
multiplication by a factor of 10 will again result in a carrier at 10 MHz but with a deviation from
9.99 MHz to 10.01 MHz. The rate at which the carrier is swung between 9.99 MHz and 10.01
MHz will not be affected and so the modulating frequency remains unchanged.

** With F.M .therefore modulation can be effected either before or after frequency-multiplying
operations are carried out. However, the modulation-index is increased because the deviation
has been increased while the modulating- frequency remains unchanged. This implies that more
side-frequencies have been generated on either side of the carrier and these are spaced at
intervals equal to the modulation frequency fm.

** Although frequency-multiplication does not of itself increase the relative-

bandwidth(bandwidth relative to the centre frequency) the change of modulation-index increases
the spread of the signal. When designing or aligning a F.M. transmitter the deviation must be set
at the output of the modulator stage so as to achieve the desired modulation-index at the output
of the multipliers.

The problem of frequency-drift in the M.0. remains the same as in A.M.

2.2.5 Frequency Translation

To translate means to move to another place or, in this instance, to move a signal "sideways"
through the frequency spectrum to another position. Thus a carrier on 1 MHz might be
translated to 5.4 MHz which is an operation that cannot be carried out with the frequency-
multiplication technique. Frequency translation is also the only means of shifting a signal down
the spectrum to a lower frequency.<

** Frequency translation is achieved by mixing the basic carrier with another sinusoidal
waveform. As shown under 2.2.3 and in Fig. 6 when two sinewaves are mixed then, either in a
linear circuit or in a nonlinear circuit, two extra sinewaves are generated with frequencies that
are the sum and difference of the two parent frequencies.

Thus, with reference to the 1-MHz carrier mentioned above, to translate it to 5.4 MHz it must be
mixed with either a 4.4-MHz signal or with a 6.4-MHz signal.

The carrier might have been derived from a VFO which ranged from 0.5 MHz to 0.8 MHz;
the signals available at the output of the frequency-translater would be:

with 4.4-MHz 3.9 to 3.6 MHz (f 1 — f 2 )

4.9 to 5.2 MHz (f 1 + f 2 )

with 6.4-MHz 5.9 to 5.6 MHz (f 1 — f 2 )

6.9 to 7.2 MHz (f 1 + f 2 )

Such a combination of VFO, two oscillators and a frequency-mixer could, when equipped with
suitable filters and switching, cover four ranges of frequencies

3.6 to 3.9 MHz 5.6 to 5.9 MHz

4.9 to 5.2 MHz 6.9 to 7.2 MHz

>>>>>>>>>>>>>>>>>>>> PAGE 15 <<<<<<<<<<<<<<<<<<<<

Fig. 6 Result of Mixing Two Sinewaves

The secondary signal is not required to vary and so it can be derived from a fixed-frequency
oscillator of optimised design which is crystal controlled. Crystal oscillators are capable of
frequency stabilities of 1 part in 10,000 (1 cycle at 10-kHz) without special precautions ; with
specially-designed crystals mounted in temperature-controlled ovens stabilities up to 1 part in
106 or even greater can be achieved and this has a beneficial effect on the overall performance of
the M.0. system.

The 1-MHz oscillator mooted above was given a poor stability of 1% (1 part in 100); when
mixed with a secondary-oscillator output with a stability of 1 in 104 the stability of the
frequency-translated output can be calculated as follows:

Carrier at 1 MHz mixed with waveform at 4.4 MHz results in a translated carrier at 5.4 MHz.
When the carrier drifts to 1.01 MHz the translated carrier drifts to 5.41 MHz. Maximum drift
of the secondary oscillator is from 4.4 MHz to 4.4004 MHz and this will move the translated
carrier further out to 5.41044 MHz if both oscillators drift in the same direction.

This represents an overall drift (maximum drift) of approximately 1,044 parts in 540,000
which is approximately 1 : 540 or 0.2 % Thus, in terms of the translated output, there has been
a 5-fold improvement in the frequency-stability. Note too that, where the carrier is derived
from a VFO, that Master Oscillator works always over the same range ; were a simple oscillator
used with switching to cover the two ranges 0.5 to 0.8 MHz and 10.5 to 10.8 MHz it would be
necessary to cope with relative ranges of 50% for the first mode and 3% for the second.

>>>>>>>>>>>>>>>>>>>> PAGE 16 <<<<<<<<<<<<<<<<<<<<

The effect of frequency-translation on different modulation systems is important too:

(i) Amplitude Modulation

** Consider a carrier at 1 MHz modulated at 1 kHz and which therefore has side-frequencies
at 999 kHz and 1,001 kHz. When mixed with the 4.4-MHz secondary signal the carrier is
translated to 5.4 MHz and the side-frequencies to 5.399 MHz and 5.401 MHz . This still
represents a modulating frequency at 1-kHz ud so translation can be used either before or after
modulating a carrier; one instance of such is found in the superheterodyne receiver described
later in the Course.

** The relative bandwidth is decreased when a carrier is translated to a higher frequency and
correspondingly increased when a carrier is translated to a lower frequency.

(ii) Frequency Modulation

** Consider a carrier at 1 MHz modulated by a signal at fm with a deviation of 1-kHz. When

mixed with a secondary signal at 4.4 MHz the carrier is again translated to 5.4 MHz and the
frequency-swing (twice the deviation) becomes 5.399 to 5.40 1 MHz . The rate at which the
carrier-frequency is swung remains unchanged and so the modulating frequency remains
unchanged. Thus frequency-translation too can be used either before or after modulating a
carrier.

However the deviation has remained at ±1 kHz and so the modulation-index has not been
changed ; in turn this means that the range of side-frequencies has not been changed with the
increase of carrier frequency. Thus the relative bandwidth (see under 1.14.1 on p.64
of Fundamentals—1) has been reduced as a percentage of the carrier frequency.
2.2.6 Block Diagram of M.0. System

** From the above it is to be concluded that the best M.0. systems are based on the general
arrangement shown in Fig. 7and require:

>>>>>>>>>>>>>>>>>>>> PAGE 17 <<<<<<<<<<<<<<<<<<<<

(i) a low-frequency VFO which is capable of covering the greatest required range of frequency
as measured in cycles-per-second.

(ii) a frequency-translating system which uses well-designed fixed-frequency oscillators

(preferably crystal-controlled oscillators) with best-possible frequency stability to convert the
output of the VFO to the required ranges. By providing a choice of such oscillators, with
frequency-multipliers as necessary, the M.O. can be switched to cover different ranges of
operation while preserving the frequency stability of the fixed-range low-frequency VFO.

Transmitters which are intended for operation on a fixed frequency may use frequency-
multipliers to arrive at the final output frequency provided that the oscillator is sufficiently
stable for the final output-signal drift to be within tolerance. In such transmitters however care is
necessary to ensure that the many harmonics which are generated neither appear in the final
output signal nor radiate directly from the transmitter casing. Thus an important part of a
transmitter drive-section is adequate filtering, screening and buffering.

END OF LESSON 3

* * * * * * * * * * * *

QUESTIONS
1. In the Master-oscillator section of a transmitter why is it desirable to indulge in frequency-
changing ?

2. What is the difference between frequency-multiplication and frequency-translation ? Each

technique has its merits ; what are they ?

3. Why is it not possible to amplitude-modulate before frequency-multiplication is undertaken ?

4. Describe the basic technique used in frequency-translation.

>>>>>>>>>>>>>>>>>>>> PAGE 18 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 4
2.2.7 Frequency Synthesisers

Although frequency synthesisers produce more background noise then standard oscillators
they are useful in that they are compact and can produce a variable frequency with a stability
close to that of a crystal oscillator . They have a minor disadvantage in that the frequency can be
adjusted only in discrete steps but the size of those steps is a matter of cost rather than technical
difficulty.

** The output signal from a synthesiser is locked in frequency to that of a reference signal
which may be either

(a) an internal crystal-controlled oscillator

(b) an externally-supplied signal.

For many uses, including amateur radio, the stability of a well-designed crystal oscillator
without a crystal oven is sufficient and so a synthesiser which uses an internal crystal-oscillator
as a reference can be used as a VFO. Provision for alternative use of an external reference
allows the VFO to be locked to such things as a local frequency-standard or to the mains
frequency; this last can be useful in avoiding the adverse effects of mains-induced hum.

The inference therefore is that frequency-synthesisers translate either upward or downward

from the frequency standard but they differ from the translating techniques already described in
that they retain the stability of the standard. This is achieved by use of a Phase-locked Loop
(PLL). Despite its intimidating name this device is simply an implementation of the basic
servo-system which was described in Lesson-5 of Fundamentals-1under 1.11: Amplification.

The two diagrams given in Fig. 13of that Lesson are based on signals that vary continuously
and which are described as an analogue system. The PLL makes use of digital techniques in which
signals are handled as a series of numbers.

2.2.8 The Phase-locked Loop

** The basic PLL is shown in Fig. 8 and it consists of four parts:

(a) a voltage-controlled oscillator (VCO) usually tuned by a varactor diode or a varicap
(b) a phase-detector or phase-comparator
(c) a low-pass filter (LPF)
(d) an error-signal amplifier

>>>>>>>>>>>>>>>>>>>> PAGE 19 <<<<<<<<<<<<<<<<<<<<

** The phase-detector compares the outputs from the VCO and from the reference source; it
produces an error signal which is proportional in magnitude and in polarity to the manner in
which the two signals differ in phase. Ideally, with the system in a stable condition, this error-
signal is either zero or a small-value dc and the purpose of the LPF which follows the detector is
to ensure that all other "noise" signals are excluded.

** The error-signal is applied to the voltage-controlled oscillator to adjust its frequency in

such a way that any phase-error is reduced. With the system properly adjusted the frequency of
the VFO is locked to that of the reference source ; any residual phase-error is indicated by the
magnitude of the error-signal.

Note that, if the two frequencies are not identical, the two signals must constantly change
their relative phase and the error-signal becomes a small- amplitude sinewave. In some
applications of the PLL this sinusoidal signal is the required output signal but a simple use for
the basic PLL as just described is that the output from the VCO is a "cleaned-up" version of a
noisy reference signal.

As described so far the basic PLL does not differ from the system illustrated by Fig.
13 of Fundamentals-I; it could be purely analogue in its operation. In use as a frequency-
synthesiser it requires also another type of unit known as a frequency-divider. As already
described the frequency of a signal can be multiplied by distorting the waveform and selecting
from the harmonics thus generated ; to divide the frequency of a signal so as to produce a sub-
harmonc it is necessary to use digital techniques.

2.2.9 The Binary Counter

To compare phases or to count cycles it is necessary to identify accurately the same part of a
waveform in each and every cycle. The best way with a sinewave, and usually with most other
waveforms, is to use the zero-crossings; i.e. the moment that the signal passes through zero. To
do this the sinewave is first greatly amplified in a limiting-amplifier so that it is turned into a
square wave; in this way the zero-crossings become almost instantaneous as the signal level
changes abruptly between two fixed levels. This is illustrated in Fig. 9.

The squarewave is then passed through a differentiating circuit (to be described later) which
produces short pulses that are derived from the zero-crossing "edges" of the squarewave;
positive-going transitions produce positive-going pulses and the negative-going transitions
produce negative-going pulses. Either the positive-going pulses or the negative-going pulses are
removed by use of a one-way device (a diode) so that the final output is a string of
unidirectional pulses, one for each cycle of the input waveform . Clearly these pulses have a
repetition rate which is equal to the input frequency and they are rigidly locked to the input (the
driving) waveform.

A squarewave generator (as distinct from the limiting amplifier described above) is a circuit
arrangement which changes its output between two fixed levels and it can be arranged that it
executes these changes each time it receives an input pulse. When a string of pulses, such as that
derived above, is applied to a squarewave generator the output-waveform that it produces is a
squarewave at half the frequency of that applied to the differentiating circuit.

>>>>>>>>>>>>>>>>>>>> PAGE 20 <<<<<<<<<<<<<<<<<<<<

A comparison of the input and output frequencies of the system thus leads to the conclusion
that the circuit has divided the input frequency by a factor of 2 and so the system is referred to
as a binary counter or a binary divider.

** By cascading several of these binary stages it is possible to produce circuits which divide
by 2, 4, 8, 16, 32, etc.; the output from each successive stage represents division by factors of
2, 22, 23, 24, etc. Compare this "scale-of-two" counting system with our more normal
"decimal" or scale-of-ten system in which each successive counter registers a power of 10;
writing 137 means add together (102) + (3 x 10) + ( 7 x 1) .
 In modern electronics it is not necessary to design squarewave generators, pulse-forming
circuits and all the necessary ancillaries to make frequency dividers; they come ready made in
the form of "chips".

The type of binary counter described above is much used in computers. When a string of
pulses is entered the first pulse affects only the first binary stage; the second pulse causes the
first stage to reset to its original condition and, at the same time, to "hand-on" a pulse to the
second stage; the fourth input pulse causes the second stage to reset and hand-on a pulse to the
third stage; the eighth pulse causes the third stage to reset and hand-on a pulse to the fourth
stage and so on. Thus, if the string of dividers is examined to determine the

>>>>>>>>>>>>>>>>>>>> PAGE 21 <<<<<<<<<<<<<<<<<<<<

state of each stage, it is possible to deduce the number of pulses that were applied to the input.

A string of binary counters used in this way is known as a shift-register because successive
stimulation at the input causes the change-of-state to shift along the string from stage to stage. In
computing circles it is usual to refer to the arrangement simply as a Register, meaning a device
which will hold a (binary) number; however, it can be much easier to understand a computer if
it is remembered that a register has the function of shifting numbers and that they can be made
to shift in either direction (distinguished as "counting-up" and "counting- down"). This is a most
important concept in getting computers to calculate.
 ** The final trick with a binary string is to turn it into a decimal counter by employing
feedback techniques; pulses are extracted from appropriate stages and fed-back to earlier stages
so as to cause "false" triggering. For example, if the output pulse from a divide-by-four is fed
back to its input, then every third pulse will generate a feedback-pulse which takes the place of a
would-be fourth input pulse; the one-time divide-by-four has thus become a divide-by-three.
Other examples are shown in Fig. 9. Note that feedback reduces the count.

Again with modern electronics the whole process is contained within a single chip in which
the division-ratio can be set by making appropriate connections externally between specified
pins. A single chip may contain more than one counter-chain. Such a device is no longer a
binary-counter and it is usual to refer to it as a frequency-counter or perhaps more usually as
a frequency divider.

** To make a frequency-synthesiser a PLL-oscillator is combined with a frequency-divider

and there are two possible positions for it as shown in Fig. 10.

** With the divider between the VCO and the phase-detector the synthesiser can operate with
the frequency of the VCO greater than the frequency of the reference signal; with the divider
between the reference and the phase-detector the synthesiser operates with the frequency of the
VCO less than that of the reference.

A combination of both techniques makes the device extremely flexible. By dividing both signals
to a common factor it becomes possible to produce almost any frequency from the VCO. To
produce a frequency-stable but variable oscillator

>>>>>>>>>>>>>>>>>>>> PAGE 22 <<<<<<<<<<<<<<<<<<<<

unit all that is necessary is to provide a means of switching the feedback connections which
control the operation of the two dividers.

However, the operation of the phase-comparator requires that the dividers produce a
common factor of the two frequencies; this means that the output from a synthesiser can be
adjusted only in discrete steps. The size of those steps depends on the number of decimal places
available in the division-ratio and so determines the cost of the arrangement. See Fig.11.

(You should be familiar with this diagram for the R.A.E.)

>>>>>>>>>>>>>>>>>>>> PAGE 23 <<<<<<<<<<<<<<<<<<<<

END OF LESSON 4

* * * * * * * * * * * *

QUESTIONS

1. Why does a frequency-synthesiser use a reference signal ?

2. Drew in block form a basic Phase-locked Loop arrangement and describe its action. (You
might well receive a question of this nature in the R.A.E.)

3. Given that the error-signal in a PLL is a dc why can you be sure that the two oscillators are
locked to the same frequency ?

4. Drew a chain of binary counters with feedback connections to give a division ratio of (a) 17
(b) 33 . (Not required for R.A.E.)

5. Draw in block form a frequency synthesiser and describe its action.

(Questions do appear on this subject in the R.A.E. although you would not be required to draw
the arrangement - multiple choice is used,)

6. At the bottom of Page-18 in this Lesson there is reference to a Varicap. What is this device ?

7. An LC oscillator is required to work over the range 1.0 to 1.7 MHz . A tuning capacitor is
available which can be varied from 100 pF to 300 pF. Calculate a suitable value for the
inductance required.
 >>>>>>>>>>>>>>>>>>>> PAGE 23(again) <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 5
2.2.10 The Modulator

The Modulator is that part of a transmitter which accepts both a carrier-signal and a modulating-
signal and modulates the one with the other. It is a circuit arrangement in which one of the
carrier parameters is altered according to the waveform of the modulation.

(a) Amplitude Modu1ation

The simplest form of amplitude modulator consists of a radio-frequency amplifier in which

the output power depends on the value of the modulating-signal waveform. This is achieved
either by using the modulating signal to vary the auxiliary power supplier or by varying the
characteristics of the amplifying device (either a valve or a transistor).

A fully-modulated A.M. carrier has one-third of its total power in the sidebands ; this is another
way of saying that modulating a carrier increases its total power by 50% and this additional
power represents the information which has been added to the carrier. This extra power is
drawn from the auxiliary power-supplier under the control of the modulating signal.

** A simple way to do this is to amplify the modulating signal. to high power and then to
connect this in series with the auxiliary power-supplier in an arrangement similar to that shown
inFig.12(a). This is called anode modulation in valve

(a) Power Modulation in Class-C Rf-amplifier

(b) Low-power Modulation by Variation of the Bias circuits or collector modulation in
transistor circuits. The diagram shows the power-signal coupled into the auxiliary-power feed
by means of a transformer. There are a few points to note:

(i) The modulation transformer must be able to handle the power of the modulating signal and
also be able to carry the direct current drawn by the rf-amplifier.

>>>>>>>>>>>>>>>>>>>> PAGE 24 <<<<<<<<<<<<<<<<<<<<

(ii) The modulation transformer must be insulated well enough to withstand the high voltages
encountered in a valve rf-amplifier.

(iii) The audio amplifier which is supplying the modulating signal must be able to supply audio
power (i.e. current as well as voltage) to at least 50% of the carrier power.

** For high-power transmitters (and especially so in those that use valves) anode modulation
requires the use of expensive high-voltage units but it allows the rf-amplifier to be operated in
the most efficient mode known as Class-C operation(see Part 3 of this Course). Anode
modulation of a Class-C valve amplifier can result in very-low distortion levels and provides the
greatest power output for a given size of valve . The circuit is also much easier to adjust than the
alternative methods of modulation. The problems that arise from high-power working however
are not a consideration for relatively low-power amateur transmitters and so this arrangement is
often used.

** The type of modulator circuit shown in diagram (b) uses the modulating signal to vary the
stage-gain of the rf-amplifier thus causing the output power to vary according to the modulation
waveform. Usually this is done at low power-levels and the resulting A.M. signal must then be
amplified to the final output power by means of one or more non-distorting rf-amplifiers. For
distortion not to occur the output power must always be proportional to the value of the input
waveform: plotted as a graph of output versus input the result is a straight line and so this type
of amplifier is referred to as a linear amplifier or more simply as a "linear".

** The first stage in achieving ssb-operation is to produce an amplitude-modulated carrier but

with the carrier itself suppressed. The signal which results from this is illustrated in Fig. 13(a);
with each successive half-cycle of the modulating waveform the carrier suffers on abrupt
reversal of phase. One circuit capable of this is the balanced modulator shown in Fig. 13(b).
 >>>>>>>>>>>>>>>>>>>> PAGE 25 <<<<<<<<<<<<<<<<<<<<

** The arrangement offers two signal paths into which the rf-carrier signal is fed in push-push
so that, at the output, the two carrier signals cancel. The modulating signal is fed in push-pull so
that it upsets the balance between the two carrier paths and allows carrier to appear at the output
first through one side and then through the other. When combined in the output transformer the
result shows the phase-reversals which characterise the double-sideband suppressed-carrier
signal.

** Finally, to arrive at a single-sideband suppressed-carrier signal, the output from a balanced

modulator is passed through a filter; depending on the design of this filter it eliminates either
the upper sideband or the lower sideband.

** The advent of small semiconductor diodes has made possible the Ring Modulator. This uses
four diodes connected in a ring formation in such a way that they act as a balanced modulator.
This circuit arrangement has a variety of uses including the demodulation of ssb signals. These
circuits are described in Part 5 of this Course but a detailed knowledge of their working is not
required for the R.A.E.

(c) Modulated Oscillator

Any of the methods of amplitude-modulating an rf-amplifier can be applied directly to an

oscillator thus producing a relatively simple transmitter. Skilful design, construction, alignment
and operation can result in successful use of such but the practice is to be deprecated.

The need for sufficient bandwidth to handle speech signals limits the Q-value of the resonant
circuit; on the other hand, to achieve good frequency stability and a good waveform (i.e. a low
harmonic content), a high Q-value is necessary.
 Single-stage transmitters are found most often in CW-working particularly in so-
called QRP working (output less than 5 watts); in these transmitters there is no alternative to
keying the master oscillator. Frequency drift tends to occur whenever the key is either closed or
opened mainly because the voltage of the auxiliary power-supply changes as the load-current
varies with the keying. The "squeal" which often accompanies the start and finish of each dash
or dot is known as chirp which, apart from being an illegal emission, is most irritating to read.

(d) Frequency Modulation

To vary the frequency of an LC oscillator it is necessary to vary the value of either the
inductor or the capacitor. Variation of the inductor is possible by using a "saturating reactor";
this is a cored inductor which is provided with an extra winding through which the modulating
signal is passed.

The magnetic field which accompanies the

modulating signal increases and decreases the
magnetic saturation of the core and so changes
the inductance value. Magnetic devices
however are not very linear and it is difficult
to design good audio modulators in this
manner. A typical circuit arrangement is
shown in Fig.14; because the modulation
signal is an ac it is effectively turned into
a varying dc by adding dc-bias.

>>>>>>>>>>>>>>>>>>>> PAGE 26 <<<<<<<<<<<<<<<<<<<<

Before semiconductors, F.M. was achieved by use of the Reactance Valve. This is an
amplifying stage with negative-feedback applied directly from output to input via a capacitor.
The 900 phase shift which the capacitor introduces into the feedback path results in the input of
the stage appearing as a large capacitor whose value is the feedback-capacitance multiplied by
the stage gain. Thus, if the modulating signal is applied so as to vary the stage gain, the input-
capacitance of the stage varies accordingly. Such a circuit, when connected across the tuned
circuit of an LC oscillator, results in frequency-modulation. See Fig. 15(a).
 ** The method works less well with transistors because of the large capacitances inherent in
transistor construction. The more-usual F.M. modulator uses the variable-capacitance
diode or Varicap; these are manufactured to exploit the fact that reverse-biased semiconductor
diodes exhibit a capacitance which varies with the applied voltage. Connected across a resonant
circuit these diodes are used both for F.M. modulating, as shown in Fig.15(b), and for non-
mechanical tuning.

RC oscillators produce the necessary phase-shift effects by using networks of resistors and
capacitors only . These can be modulated in similar fashion either by use of a variable-
capacitance diode or by using a valve or transistor as a variable resistance.

(e) Phase Modulation

Resonant LC circuits are characterised by a rapid change of phase between applied voltage
and circulating current as an applied signal passes through the resonance frequency of that
circuit. This in fact is the mechanism which controls the frequency of a maintained oscillator.

The argument can be applied in reverse; if the frequency is maintained constant but the
resonance-frequency of the circuit is varied there will be phase changes that follow the form of
the tuning-changes.

This is the basic mechanism used to achieve phase-modulation and it is illustrated

in Fig. 16. Here the resonant circuit is de-tuned by the variable capacitance but the circuit is
driven by the constant frequency from a M.O.

>>>>>>>>>>>>>>>>>>>> PAGE 27 <<<<<<<<<<<<<<<<<<<<

(f) Mixed Modulation

** Varying the tuning of a resonant circuit as described in the modulating arrangements above
also changes the circuit impedances and so causes a certain amount of amplitude modulation in
addition to the F.M. or Ph.M. Similarly the circuits described to achieve A.M. often produce
some F.M. (a bad example is the chirp from a keyed oscillator).

To overcome this effect F.M. and Ph.M. circuits are usually followed by some kind of
limiting stage which removes the A.M. The only way to deal with unintentional F.M. in an
amplitude-modulator is to take care with the design and construction so as to avoid the problem.

2.2.11 The Power Amplifier (P.A.)

** The P.A. of a transmitter is the final stage of amplification in which the (perhaps
modulated) radio signal is raised to the required power-level before it is passed to the aerial
system where it will be converted to an electromagnetic wave. Note that the Amateur Licence
restricts the maximum power level which may be radiated and that this level varies according to
the Band in use; these limitations are laid out both in the licence document and in other
literature and you are required to be familiar with them.

Because of the P.A.'s position "at the end" of a transmitter it is often known as the Final.

The detailed characteristics of tuned power-amplifiers are dealt with later under Electronic
Amplifiers and in particular under the heading Class-A, Class-B and Class-C Amplifiers. The
process known as "tuning-up" a transmitter consists of first setting the M.O. for the required
frequency of operation and then adjusting the P.A. for maximum efficiency in its operation; the
P.A. handles power and, unless it delivers that power efficiently to the aerial system, it will be
destroyed.

Many transmitters have a Mode switch on which one position is marked "Tune". In this
position the P.A. is specifically restricted in its ability to draw power and so it becomes possible
to make the necessary adjustments without fear of damage. Once these adjustments are
completed the switch is returned to the "Normal" position and then, carefully, the final
adjustments are made on full power.

>>>>>>>>>>>>>>>>>>>> PAGE 28 <<<<<<<<<<<<<<<<<<<<

** While making such adjustments considerable interference can be caused to others already
working around the chosen frequency and so setting-up procedures should be carried out with
the transmitter connected to a dummy load. This is simply a screened and ventilated resistor,
usually of 50-ohms, that is capable of handling the transmitter output without overheating. For
setting up of course it need handle only the reduced power but full-power tests should not
be made using an aerial - unless the object is to test the aerial.

There is another reason for using a dummy load which is of especial importance when using
a transmitter fitted with a transistor P.A. Part of the necessary adjustments are concerned with
achieving a proper match between the P.A. and the aerial system (see under 1.12.5: Impedance
Matching). Failure to match properly makes it impossible for the P.A. to transfer power to the
aerial system and the large voltages and/or current-demands that result can easily destroy a
transistor P.A. Valves are fairly robust and, within reason, will withstand a mismatch but it is
not good engineering practice to use transistors in a P.A. for other than low powers.

A further complication arises in that an aerial is unlikely to provide a good match to its
feeder except at a few spot frequencies and this results in the appearance on the feeder of a
standing wave which is explained in the next Section. Standing-wave problems can be
overcome by use of a so-called Aerial Tuning Unit (ATU) but, until this unit has been properly
adjusted, there is again a risk of destroying a P.A. Note: the name Aerial-tuning Unit is a
misnomer because the one thing which the unit does NOT do is tune the aerial! (The initials
really stand for Aerial Transducer Unit where a transducer is a device which changes the form of
energy; e.g. an electric motor is an electro-mechanic transducer.)

** A P.A. should unload its power into an aerial in the form of the intended radio signal but it
is possible for part of that power to appear in the form of spurious signals . These can be
harmonics of the carrier and of the side-frequencies caused perhaps by over-modulation or they
might be unintended oscillations which can appear on unexpected frequencies. Additionally
some of these frequencies may beat together to form sum-and-difference frequencies or they
may even beat with transmissions from other stations. It is a requirement of the Amateur
Licence that transmitting equipment is checked regularly to ensure that such emissions do not
occur and all such tests should be recorded in your Log. See also under Filters.

WARNING

The P.A. in particular is the part of a transmitter where danger lurks for the unwary . With
valve P.A.'s the auxiliary supply may operate up to 1,000 volts and be capable of supplying a
half-amp which most definitely IS LETHAL. Transistor P.A. s seldom require more than 50-
volts but, under some conditions, even 50 volts can kill.

Never wear a wristwatch with a metal strap when going inside a P.A. or any other
semiconductor equipment; the low-voltage supply will be capable of producing many amps
which will make that wrist-strap red-hot in a contact of seconds.
 Large rf voltages and circulating currents can be encountered, especially in ATU's, which
can be just as lethal. Additionally rf burns cause serious damage to skin structure and so take a
long time to heal as well as being extremely painful.

>>>>>>>>>>>>>>>>>>>> PAGE 29 <<<<<<<<<<<<<<<<<<<<

2.2.12 Filters

** At the output from a transmitter the P.A. should produce only the carrier signal and those
side-frequencies which arise as a consequence of the intended modulation. In practice however
there is likely to be also harmonics of the carrier, of the side-frequencies arid other so-called
spurious signals. There could also be cross-modulation products which are the sum-and-difference
frequencies which result when all these signals beat with each other.

** Harmonics of both carrier and modulation frequencies arise naturally from non-linearities
(imperfections) of the various active devices used in the amplifying stages but harmonics of the
side-frequencies arise specifically through over-modulation. All signals whose frequencies are
greater than the highest legitimate side-frequency can be eliminated by use of a low-pass
filter (L.P.F.) but unintended side-frequencies which are below the carrier frequency cannot.

** A low-pass filter is a network of inductors and capacitors which has negligible effect on the
passage of signal currents up to a given frequency; above that frequency (the cut-off
frequency) signals are increasingly attenuated. Such a filter, placed at the very output of a
transmitter between the P.A. and the feeder, eliminates the risk of inadvertently radiating
spurious signals which are above the legitimate transmitting range. Harmonics of the modulating
signal which have been produced either before the modulator stage or during the modulation
process produce their own side-frequencies both above the carrier and below it; these lower-
sideband frequencies cannot be stopped by the L.P.F.

** Protection from emissions below the carrier requires the use of more expensive bandpass
filters (B.P.F.) which, as the name implies, have a lower as well as an upper cut-off frequency.
Good engineering practice - and common sense - dictate that equipment should be properly
adjusted so that modulation-signal harmonics are not generated.

** Probably the major cause of this type of distortion is abuse of the so-called speech
processor with which most modern transmitters are equipped. The idea is to amplify the speech
signal so that the quietest sounds fully modulate the transmitter; all signals of greater strength
are then reduced so that over-modulation does not occur and the transmitter always produces
maximum modulation which results in maximum information-power.

** The best way to achieve this aim is by use of a Limiter which is a variable-gain amplifier
whose gain is adjusted by the signal level (a feedback arrangement). A cheaper and so favoured
method is to amplify the signal but then to clip the waveform at a pre-determined level by means
of a diode ; this of course causes serious waveform distortion which is then removed by use of
an audio low-pass filter. Such processors or compressors work well provided that the amount of
speech compression is limited ; beyond a certain point they render the speech characterless and
emphasise breathing noises. Eventually speech is degraded into a series of thumps and becomes
all but unintelligible.
 ** Filters at the output of a transmitter come in a variety of guises. One obvious LC circuit is
the resonant circuit which forms the load of the final amplifier ; see under Electronic
Amplifiers. In fact this is a bandpass circuit but its performance as such may not give much
protection. It has been mentioned already that this circuit exists to lose power to the aerial
system and so operates with a relatively-low Q-factor. Nevertheless it can be arranged so that it
acts also as a L.P.F. the most famous form of which is probably the P1 Tank Circuit. This is
discussed in detail later and it is sufficient to note here that the

>>>>>>>>>>>>>>>>>>>> PAGE 30 <<<<<<<<<<<<<<<<<<<<

inductor is placed in series with the signal path to the aerial while the capacitor, divided into two
discrete parts, is arranged to shunt the signal path to ground. Because the circuit is resonated at
the carrier frequency it does not attenuate the carrier and its sidebands.

** The P1 circuit illustrates another function often fulfilled by an output filter namely
impedance matching. Under 1.10.4 it was shown that a tapped inductor acts as a single-wound
transformer (an autotransformer). In a resonant circuit a similar facility can be obtained also
by tapping the capacitive leg; i.e. the tuning capacitor is split into two or more discrete
capacitors as is seen in the PI circuit.

**Another filter arrangement increasingly seen at the output is the misnamed ATU (aerial-
tuning unit) . An aerial can be constructed and adjusted so that it matches the impedance of its
feeder but only at one frequency. It is possible to obtain a good match at several spot
frequencies so producing nominally "wide-band aerials" but effective matching over a
continuous band of frequencies is not possible. Amateur aerials are required to operate with
bandwidths up to 10%. of centre frequency over several harmonically-related Bands.

** The result of a mismatch between aerial and feeder is that, when in use for transmitting,
a standing-wave appears on the feeder (see next Section). This means that, at the other
(transmitter) end of that feeder, the impedance presented to the transmitter varies with
frequency. The purpose of the ATU is to provide continuously-variable correction so that,
whatever operating frequency is chosen, the feeder impedance can be matched to the P.A.

** An important point not generally realised is that an ATU does NOT remove the standing
wave on the feeder; to remedy that either the aerial must be made to match the feeder or the
ATU must be placed between the aerial and the feeder. This is not a practical arrangement
because it requires servo-systems to adjust the ATU controls and a weatherproof casing.

** Few ATU circuits provide attenuation of harmonics and a separate L.P.F. should always be
used.

END OF LESSON 5

** ** ** ** ** **

QUESTIONS

1. Impressing information on to a carrier wave by amplitude-modulation increases the total

carrier power. Why is this and from where is the additional power drawn ?

2. What is the advantage of single-sideband operation ?

3. Give a brief description of the manner in which frequency modulation is achieved usually in
modem equipment. Why is such a circuit usually followed by a limiter stage ?

4. Describe the purpose of the "Tune" mode in a transmitter

5. Why is a dummy load an essential piece of transmitting equipment ?

6. Why are filters used at the output of transmitters ?

7. An Inductor of 0.5 Henrys has a dc-resistance of 12 ohms and is connected in parallel with a
resistor of 6,000 ohms. What is the impedance of the combination at 1 kHz ? Make reasonable
practical approximations.

>>>>>>>>>>>>>>>>>>>> PAGE 31 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 6
(Only paragraphs marked ** are required for the R.A.E.)

2.3 TRANSMISSION LINES

2.3.1 General

** A transmission line is a system of conductors which is used to transfer electrical energy

from one point to another. An everyday example is the overhead Grid distribution system of the
Electricity Boards but we are concerned here with the transfer of rf-power from a transmitter to
an aerial system or from that aerial system to the input of a receiver. Such transmission lines are
often called feeders.

** Usually a feeder consists of a pair of conductors which are arranged:

(a) straight and parallel held apart by "spreaders" (an open-wire feeder),
(b) twisted together (a twisted pair)
(c) placed co-axially one within the other.

Feeders which handle very high powers may consists of four conductors in open-wire form with
the diagonally-opposite conductors connected in parallel.

The energy is actually carried in the electro-magnetic field (e-m) field which travels between
the conductors; it is supported by the voltages between the conductors and by the currents
which flow on the surfaces of those conductors. In a coaxial line this e-m field is completely
enclosed by the outer conductor and so is less likely to cause unintentional radiation or to be
influenced by incident e-m radiations (noise).

For technical theoretical reasons waveguides, used at extremely high frequencies, are not
classified as transmission lines but they can be viewed as coaxial lines with the centre conductor
removed which does demonstrate that the flow of energy is contained in the traveling e-m field.
It is possible too at these frequencies to make a waveguide from a tube or rod of dielectric
material without conductors or from a single conductor which is thinly coated with a dielectric
material.

2.3.2 Behaviour a Transmission Line

Under 1.7.2 in Lesson-2 of Fundamentals-1 the capacitance of a capacitor was stated to be a

function of plate area and of plate separation and not to depend in any way on the shape of those
plates. Thus a transmission line can be regarded as a capacitor which has been constructed from
two very long and thin plates. Fig. 17(a) shows such a capacitor connected to a cell via a switch.

When the switch is closed the capacitor can be expected to charge as discussed previously.
However, the capacitor discussed in Lesson-2 of Fundamentals-1consisted of two square plates
with the connections taken to their centres; here, with the connections made at one end,
electrons have to travel the length of the narrow plates as they either enter the capacitor or leave
it. Such a movement of electrons constitutes a flow of current ; that current takes a real
(or finite-length of) time to travel from one end of the transmission line to the other.

>>>>>>>>>>>>>>>>>>>> PAGE 32 <<<<<<<<<<<<<<<<<<<<

Under 1.9 in Lesson 4 of Fundamentals-1: Induction, Inductors and Inductance it was shown that
conductors which carry current interact with a surrounding magnetic field to
produce Inductance whereby changes of current-flow are opposed. The transmission line
therefore must be more than just a long thin capacitor; electrons cannot move freely along the
line but are impeded by the inductance of the conductors.

(a) Path of Electrons During Charging

(b) Representation of Electron-path Inductance

Fig. 17(b)shows an attempt to represent this by drawing the conductors as long Inductors
which are connected at regular intervals by a series of small capacitors. However the diagram
is not correct because the real line capacitance is not divided into a number of discrete
components but is distributed evenly along the line as are the inductances. For this reason a
transmission line is said to possess distributed inductance and distributed capacitance as distinct
from the lumped inductance and lumped capacitance of discrete manufactured components.

Nevertheless the distributed inductance does have the effect of isolating small sections of
distributed capacitance in that it delays the passage of electrons from section to section as the
line attempts to charge from the cell. In fact an artificial line can be constructed from discrete
inductors and capacitors as shown in Fig.18 and the arrangement does indeed behave in the
manner of a transmission line. Although such a device is not employed to convey power it can
be used for example to delay the passage of a signal between two points and then it is known as
a delay line.

2.3.3 Transmission Line Connected to a dc Source

When the switch is closed the terminal voltage of the cell and the terminal voltage of the
input-capacitor C 1 become identical; the capacitor however is not charged and so the cell-
terminal voltage has to fall to zero.

With zero voltage across the line current cannot flow into it via inductors L 1a and L 1b and so
the entire current from the cell must flow into C 1 . As the capacitor charges so a p.d. begins to
build-up between its terminals which

>>>>>>>>>>>>>>>>>>>> PAGE 33 <<<<<<<<<<<<<<<<<<<<

(a) opposes the flow of current into C 1 and

(b) begins to drive a current into the line through distributed inductances L 1a and L lb

However, as discussed in Fundamentals-1, the nature of inductors is to oppose any change

in the rate of current-flow and so the introduction of charge to distributed-capacitance C 2 is
delayed.

As a repeat of the last two paragraphs - current cannot flow into the transmission line
beyond C 2 until that capacitor has acquired an accumulation of charge and the inductive effects
of L 2 have been overcome.

Fig. 18 Artificial Transmission Line

There are several points of interest in this line-charging action:

(i) Current does not flow instantly along the length of the transmission line when the switch is
closed. Time is required for the current to steeplechase its way from section to section
constantly delayed by the distributed inductance (which resists the build-up of current) and by
the distributed capacitance (which delays the build-up of voltage).
 This is expressed in technical language by saying that a wavefront propagates along the line;
a wavefront in which current begins to flow in the line and charge begins to accumulate in its
distributed capacitance. This is shown graphically in Fig. 19 where the right-hand side of the
curve shows the changing value of current as the wavefront passes along the line. Because this is
the first wavefront to pass down the line it is referred to as the leading wavefront.

>>>>>>>>>>>>>>>>>>>> PAGE 34 <<<<<<<<<<<<<<<<<<<<

(ii) When the wavefront reaches the far end of the line what happens depends on what is
connected at that end. If the line is open-circuited then the final capacitive section charges to the
value of the emf and the line current falls to zero.

If the line is short-circuited at the far end then the final capacitive section fails to charge and
the line current reaches a value equal to the emf of the cell divided by the internal-resistance of
the cell.

Fig. 19 Charge and Discharge Wavefronts with DC Excitation

If a load is connected then the final values of line current and line voltage must settle in the
ratio defined by Ohm's Law for that load.

(iii) At the moment the wavefront is launched into the line the p.d. across the line begins to rise
and so too does the line current. This can continue only for a short time because the line
current, flowing through the cell's internal resistance, must limit the terminal voltage to a value
less than the cell's emf.

We can conclude therefore that, if the transmission line is very long, the ratio of line-voltage
to line current at the input will stabilise before the last sections of the line have begun to charge
(before the leading wavefront arrives at the far end). This combination of line-voltage and line-
current has nothing whatever to do with any load which may be connected at the far end ; it is
determined only by the distributed-inductance and distributed-capacitance of the line.
 >>>>>>>>>>>>>>>>>>>> PAGE 35 <<<<<<<<<<<<<<<<<<<<

(iv) There is a mischievous thought "What happens if, when the leading wavefront is only
half-way along the line, the switched is opened ?

Although the cell is no longer available to supply the line current the distributed inductance
resists any fall in the current value ; this continuing current-flow can be obtained only by
discharging C 1 and the current in L 1 will fall gradually to zero as the capacitance is discharged.

This action is then repeated discharging first C 2 , then C 3 and so on along the line. Thus a second
discharging (or trailing) wavefront propagates down the line.

Between the leading and trailing wavefronts there now exists the strange phenomenon of a short
pulse of direct-current which is traveling along the line toward the load end. Note that this
voltage/current relationship is not influenced in any way by a connected load (which, as yet, is
not aware that a generator has been connected to the line) nor by the cell which, after causing all
the commotion, has now been disconnected from the line.

Of particular interest is the fact that the dc-pulse continues to travel toward the load when it
might be expected to spread evenly along the line. Had the cell been connected to the centre of
the line, and then disconnected again, the input current would have divided and flowed in both
directions but the end result would have been two pulses traveling in opposite directions.

It is the presence of distributed-inductance and distributed-capacitance which keeps current

flowing once launched into the line and prevents the accumulated charge from spreading
evenly throughout the distributed capacitance.

(v) The last point of interest arises from the different possibilities of a connected load:

(a) Not Terminated

Without a connected load the final capacitive section of the transmission line in Fig. 18 is denied
a discharge path. Thus, because the distributed inductance maintains the current-flow, so this
final capacitive section must acquire the entire charge that was delivered by the cell. It can be
shown mathematically and practically that, at the moment when current finally ceases, the final
capacitive section is charged to a value twice that of the line-voltage between the two
wavefronts. (This final value is greater than the emf of the cell; it can be shown that equal
power is carried in the electric and magnetic field).

The final cessation of line current signals the end of the inductive effects that kept the current
flowing toward the load end of the line. Thus the final capacitive section is now free to
discharge back into the line so causing a flow of current toward the cell. As this section becomes
discharged however it simulates the original action of opening the switch and so a new dc-pulse
appears traveling toward the cell end of the line.

>>>>>>>>>>>>>>>>>>>> PAGE 36 <<<<<<<<<<<<<<<<<<<<

Although this return pulse is a new pulse re-constituted from the stored charge it is usual to
regard it as the original pulse reflected from the open-circuited end of the line. Note that, when
reflected from an open-circuit, the pulse retains the same polarity but the direction of current
flow on the line (and the direction of propagation for the pulse) has been reversed.

The reflected pulse travels to the input end of the line where the switch is open and so suffers a
second reflection from an open-circuit. In fact it must continue to bounce up and down the line
but it slowly decreases in amplitude because its energy is dissipated as heat in the resistance of
the conductors.

(b) Terminated by a Short-circuit

Referring again to Fig. 18 a short-circuit termination means that the final capacitive-section
represented as Cn cannot charge. When the leading wavefront arrives at the end of the line the
inductive effect causes current to flow through the short-circuit from (as shown) the lower
conductor to the upper conductor. As this current-flow continues so the penultimate capacitive-
section C n-1 becomes first discharged and then recharged with opposite polarity.

Once again a "reflected pulse" appears propagating toward the input end of the line but with a
difference. This pulse, reflected from a short-circuit, has reversed polarity but retains the same
direction of current-flow. Again this corresponds with the reversed direction of propagation for
the pulse.

When this pulse suffers re-reflection at the open-circuited input end of the line it retains its
polarity but undergoes a reversal of current as before and conditions on the line become
complicated until the pulses have been dissipated. If required this can be averted by arranging
the switch so that, when it disconnects the cell, it shorts the transmission line ; can you draw
such an arrangement ?

(c) Resistive Termination

A resistor connected across the line allows current to flow between the two conductors but
restricts its value ; at the same time it develops a p.d. across the final capacitive section. This
modifies any reflected pulse mainly in its amplitude with respect to the incident pulse ; it also
dissipates quickly the energy of reflected pulses.

There is a value of resistance however which will support the ratio of current to voltage which
exists on the line between the two wavefronts. If such a resistance value is used as a termination
then the energy of the incident pulse is utilised completely in heating the resistor without
disturbing the voltage/current relation ; current-flow does not cease until the distributed
capacitance is fully discharged and so reflected pulses do not appear.

The value of resistance which supports the same voltage/current ratio as does the transmission
line is known as a matching load; a line which does not produce reflections is said to
be matched. This ratio of voltage/current which exists within the dc-pulse, irrespective of input
or load conditions, is referred to as the characteristic resistance of the

>>>>>>>>>>>>>>>>>>>> PAGE 37 <<<<<<<<<<<<<<<<<<<<

transmission line; thus a line is said to be matched and fails to produce reflections when it is
terminated in a resistive load equal to its characteristic resistance.
There are two practical points worth noting. All transmission lines have resistance and so
dissipate energy as heat; thus, as a pulse propagates along the line, it slowly decreases in
amplitude. If a line is sufficiently long it will totally absorb a pulse before it reaches the load
end. Under these conditions a reflected pulse cannot appear whatever the load may be. Even for
a shorter line it may be that the return length (to the far end and back to the input end) is
sufficiently long to totally absorb a pulse ; thus, although a reflected pulse may be generated, it
will not appear at the input.

** In both instances an observer might conclude that the load was absorbing the power and so
the line must be matched. Thus an infinitely (or sufficiently) long line will behave as a
resistance equal to its characteristic resistance independent of its terminations.

** The second point is that the characteristic resistance is determined by the ratio of
distributed inductance to distributed capacitance and, in turn, these are determined by the
physical dimensions of the line . Thus the characteristic resistance of a line is controlled by
selecting the diameter of the conductors and the spacing between them.

END OF LESSON 6

* * * * * * * * * * * *

QUESTIONS

1. What is the reason for the somewhat strange behaviour of electric charges on a transmission
line ?

2. A pulse is traveling along a transmission line ; why does it continue to travel as a pulse rather
than disperse its charge evenly throughout the length of the line ?

3. The characteristic resistance of a transmission line is not at all connected with the dc resistance
of its conductors ; what determines this characteristic resistance ?

4. A positive-going pulse is reflected from the open-circuited end of a transmission line.

Describe the reflected pulse.

5. What are the two conditions on a transmission line each of which would result in the non-
return of a pulse to the input end ?

>>>>>>>>>>>>>>>>>>>> PAGE 37(again) <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 7
2.3.4 Transmission Line Connected to an AC Source

When a transmission line is connected to an ac source the result, at first, is the same in that a
leading wavefront is launched into the line in which current rises, and charge accumulates, in
the distributed capacitance.
 The intervention of a switch is not required here however because the ac-source reduces the
applied voltage to zero and then reverses the polarity. Hence the second wavefront becomes one
in which the distributed capacitance is first discharged to zero but then re-charged with opposite
polarity ; at the same time the line current also falls to zero and builds-up again but flowing in
the reverse direction. This is illustrated in Fig. 20.

Fig. 20 Charge and Discharge Wavefronts With AC Excitation

Note that, in the second wavefront, both current and voltage reverse direction and the
wavefronts continue to propagate toward the load end of the line.

There are two major differences in the line behaviour when the source is changed from dc to
ac:

(i) with a dc-source (reference Fig. 18), charge is transferred continuously along the line from
C 1 to C 2 to C 3 ... With an ac-source, because the line current constantly reverses, the charge
flows back and forth along the same section of line ; e.g. from C 1 to C 2 and then back to C 1 .

>>>>>>>>>>>>>>>>>>> PAGE 38 <<<<<<<<<<<<<<<<<<<<

(ii) the leading wavefront and the following wavefront now define a section of line in which
current-flow occurs in both directions toward a point of zero current. This point of zero current
is associated with a point of zero voltage where the waveform changes polarity.

This entire configuration repeats as the source cycles and it travels along the transmission
line away from the source ; this moving pattern is referred to as a traveling wave.
 In the transmission line driven by an ac-source electrons oscillate back and forth along a
short path but the magnetic and electric patterns (fields) which they create travel continually
toward the load. It is the combination of these two traveling fields that carries the power along
the line (power requires both voltage and current).

At the end of the transmission line, if a matching load is connected, the ratio of voltage-to-
current (as represented by the electric and magnetic fields) can be maintained and so the power
is totally absorbed by the load and dissipated as heat. Note that current minima are coincident
with voltage minima ; i.e. the current and voltage waveforms are in phase which means that the
line is behaving like a resistance - its characteristic resistance.

If the connected load is anything other that the characteristic resistance then, as with the dc-
source, the load cannot absorb all the power and there is a residual alternating-current or
alternating-voltage which is re-launched back down the line. The two traveling alternating-
waves, one moving toward the load and the other moving back toward the generator, are
otherwise identical and they interfere to form a distinctive cyclic pattern which does not change
provided the conditions remain constant.

This fixed pattern is known as a standing wave and it differs from the traveling wave in that it
is not truly a wave pattern at all ; it has the nature of a graph which illustrates the amplitude (the
size) of alternating current-flow or alternating voltage at successive points along the line . The
electric-field graph is not in phase with the magnetic-field graph . This implies that the line is
behaving in the manner of a reactance and is carrying a certain amount of wattless current ; this
is not surprising when the load is failing to absorb all the power that (apparently) is being
injected by the source.

** It is not necessary to analyse the line behaviour in detail for the RAE but the following
points are important to grasp. There are places at regular intervals along a mismatched line
where the line-voltage is high but the line current is small; at such points the line is presenting a
high impedance. Mid-way between these places there are others where the line-voltage is low
but the line-current is large; at such points the line is presenting a low impedance.

** Should one of these points of high voltage fall at the input terminals of a receiver then the
voltage-waveform of the incoming signal will be enhanced: should one of the points of low
voltage fall at the input terminals of a receiver then the voltage-waveform of the incoming signal
will be degraded. However, should a point of high voltage or high current-demand fall at the
output terminals of a transmitter then there is considerable risk of damage occurring to the P.A.
stage. For this reason it is most important that a transmitter be connected only to a matched line.

>>>>>>>>>>>>>>>>>>>> PAGE 39 <<<<<<<<<<<<<<<<<<<<

There is advantage when receiving in having a ms-matched aerial system such that a point of
high-impedance occurs at the receiver input but successful use of this ploy requires two
independent aerials. This is not a practical objective when using a Transceiver(a piece of
equipment that is switched to serve alternately as a transmitter and then as a receiver) ; a single-
aerial system as required by a transceiver must be properly-matched.

For those not happy with the above explanation of a standing-wave the following may help . The
concept of power being reflected from a mismatching load is a powerful way of explaining to
students the queer things that happen on a transmission line . Indeed , if power is applied to a
line as a train of rf-pulses, then the forward and reflected pulses can be viewed on an
oscilloscope.

Do not take this explanation too far ; it does not work for continuous-wave signals. For
example consider power derived from a transmitter and reflected from a mismatched
load ; where does that power go when it arrives back at the transmitter ? It cannot be absorbed
by the transmitter . If it is re-reflected then the transmitter must reduced its output by the
amount of that reflected power else even funnier things must happen ?

The voltage/current ratio dictated on the line by the distributed L & C is not supported by a
so-called mismatched load ; the ratio must change . The situation is analogous to that discussed
in connection with real and apparent power involving a power factor ; where a product of
voltage x current does not correspond to the amount of power being dissipated there must be a
power-factor involved which means that the current and voltage must depart from being in-
phase (see Fundamentals-1, Lesson-7) The lack of synchronism between voltage and current
coupled with the distributed L & C gives rise to the amplitude-variations that constitute a
standing-wave.

2.3.5 Transmission Line as a Reactor

It was stated above that the patterns of voltage and current on a mis-matched line are not in
phase and so such a line behaves like a reactance; by contrast a matched line behaves as a
resistance.

** Along the length of a mis-matched line the impedance constantly changes cycling between
capacitive reactance and inductive reactance ; at points approximately a quarter-wavelength
apart the reactance changes sign and here the line becomes purely resistive. Such points of
resistance alternate between being greater than the characteristic resistance and less than the
characteristic resistance. Thus a length of transmission line with well-chosen characteristic
resistance can be used as a high-frequency transformer ; it is cut to be a quarter-wavelength at
the required frequency and so supports at one end the lower of the two impedances to be
matched and, at the other end, the higher of the two impedances. A more convenient way of
achieving this is given in the next paragraph.

** Short lengths of transmission line are used also to correct the driving-point impedances of
aerials; these are known as stubs. Cut either longer or shorter than a quarter-wavelength they
provide an inductive reactance or a capacitive reactance that is connected either in series or in
parallel with an aerial to make it appear purely resistive. A quarter-wavelength long
transmission-line (a quarterwave transformer) can then be used to transform this resistive value to
that required for matching to a feeder.

>>>>>>>>>>>>>>>>>>>> PAGE 40 <<<<<<<<<<<<<<<<<<<<

** A standing-wave pattern is related to the frequency of the rf-signal in that it repeats with
each successive cycle. However the wavelength of the pattern in a feeder is shorter than that
calculated for a radiating e-m wave because the distributed reactances of a transmission line
delay the propagation. The ratio of the two rates of propagation is expressed as a velocity
ratio or velocity factor whose value is obtained from data issued by the manufacturer and this has
to be taken into the calculation when cutting the required length.
 Short lengths of transmission line are used also as reactors and as resonant circuits in rf
amplifiers, oscillators and insulators. They are particularly useful at frequencies above 100 MHz
where normal lumped reactors become impossibly small.

2.16 Standing Waves and SWR

** The appearance of a standing-wave on a transmission line which is fed with ac indicates

that a mismatch exists at the receiving end of that feeder. For example a standing-wave on a
feeder that is carrying power to an aerial shows that the driving-point impedance of the aerial
does not match the characteristic impedance of the feeder. A standing-wave on the same feeder
which is connecting the same aerial to a receiver shows that the receiver input-impedance does
not match the feeder impedance.

Although it is true that, in both instances, re-reflection must occur at the input-end mismatch
it is a greater truth that a standing-wave cannot be generated at all unless there is a mis-match at
the delivery point. This is the reason for saying that an A.T.U. is mis-named ; while it may be
able to produce a good match between feeder and P.A. output it cannot remove the standing-
wave on the feeder because this is caused by the mis-match between feeder and aerial. Note too
that an ATU cannot be adjusted by tuning for maximum incoming signal .

** A standing-wave is an undesirable feature for several reasons:

(a) its cause is the failure of the load to absorb all the power that could be presented by the
source - the system is acting inefficiently

(b) at points separated by a half-wavelength the line is subjected to excessive build-up of voltage
which could damage the feeder

(c) at these points of high voltage the dielectric losses in the insulating material, which are
proportional to the square of the voltage, can become excessive causing localised heating with
possible failure of the feeder

(d) midway between points of voltage maxima there are points of current maxima where again
localised heat-generation, which is proportional to the square of the current, could cause the
feeder to fail

(e) failure to present the P.A. with the correct value of resistive load causes a loss of efficiency
in that P.A. which, at the best, limits the output power which the stage can generate without self-
destruction

>>>>>>>>>>>>>>>>>>>> PAGE 41 <<<<<<<<<<<<<<<<<<<<

(f) when a point of maximum voltage or of maximum current-demand appears at the P.A.
terminals then damage is likely to result in the stage ; this is of particular importance where a
P.A. employs semiconductor devices.

[Items (e) & (f) are the main reasons for employing an ATU in Amateur working ; in high-
power systems which transmit several kilowatts items (b), (c)& (d) become significant also.]
 ** In practice it is not possible to eradicate a standing-wave entirely and a criterion is required
for its assessment. A perfect match is indicated by the non-appearance of a standing-wave
whereas a total mismatch (i.e. either an open-circuit or a short-circuit both of which produce a
100% reflection) results in both current and voltage waveforms varying between zero (complete
cancellation) and twice normal amplitude.

** As the condition of perfect-matching is approached so the standing wave tends to disappear

or, as it might be interpreted, the ratio of maximum to minimum values approaches 1:1 . With a
total mis-match so the ratio of maximum to minimum values tends toward an infinite value.

** This standing-wave ratio or swr is used as an indicator of the quality of the matching. The
ideal perfect match is often referred as a swr of 1:1 which is a bit of a puzzle to the uninitiated.
In practice however it is not attainable. A swr of 2:1 is beginning to get serious on low-power
installations such as Amateur rigs and really is serious with transistor P.A. stages. A ratio of 1.1
: 1 (or about 10%) is about as good a match as is possible if costs are to be kept within reason
both in the equipment itself and in the measuring gear used to achieve that degree of matching.
However the problem usually lies not so much in achieving a good match as in ensuring that the
good match remains.

>>>>>>>>>>>>>>>>>>>> PAGE 42 <<<<<<<<<<<<<<<<<<<<

In a transmitting system
the ideal is 1: 1 because it
indicates maximum
efficiency and that all the
power is being transferred to
the aerial for conversion to a
radiating e-m wave. In a
receiving system however, as
already mentioned, the
perfect match is not
necessarily the best
objective. The feeder is
made to support a standing-
wave by ensuring that the
input impedance of the
receiver is greater than the
characteristic impedance of
the feeder ; thus a voltage
maximum occurs at the
receiver input. In this
manner the feeder is made to
provide a selective "voltage
gain" (but it does not provide
Fig. 20(A) Receiver System Gain amplification).
(a) Load Matches Feeder-Z
(b) Load Greater Than Feeder-Z This neat little trick
(c) Load Less Than Feeder-Z applies only to fixed-
frequency working or to
where the receiving
 equipment is required to
cover only a small range of
frequencies.

Receiving installations
operate only at microwatts
and so the disadvantages of
permitting a standing-wave,
as listed above, do not apply.
As a purely practical point
however it should be noted
that a swr of 1:1 cannot be
achieved in practice and so
every effort should be made
to ensure that error in the
input impedance of any
receiver makes it greater
than the characteristic
impedance of the feeder to
which it is connected .
Should the input impedance
be less than the characteristic
impedance then the feeder
will provide a voltage-loss.

Although swr is loosely defined as the ratio of maximum to minimum values this is only part
of the story because, expressed in this way, it does not reveal whether the mismatch is too high
or too low. Where the information is significant it is important to express the ratio as terminal-
voltage to next maximum or minimum back down the line. For example a receiver with an
input impedance of 82 ohms connected to a feeder of 75 ohms would produce a swr of 1.1: 1
while a receiver with an input impedance of 68 ohms would produce a swr of 0.9:1 . In practice
the result is an apparent difference in sensitivity of 20% (2 dB).

Equipment used to measure swr is described later in this Course in Part 6: Measurements &
E.M.C.

>>>>>>>>>>>>>>>>>>>> PAGE 42a <<<<<<<<<<<<<<<<<<<<

END OF LESSON 7

* * * * * * * * * * * *

QUESTIONS

1. A voltage-probe moved along two identical transmission lines shows that one is carrying a
substantial standing wave while, on the other, a standing-wave does not appear to be present.
What conclusion can be drawn ?
2. What is the major difference between a traveling-wave and a standing-wave?

3. A transmission line carries an appreciable standing-wave. Why should not this line be
connected to a transmitter without some kind of correcting equipment ?

4. Transmission lines can be used as either capacitors or inductors ; what condition must be
fulfilled ?

5. What is a standing-wave ratio ?

6. A network of 8 resistors consists of three 300-ohm resistors in parallel and four 200-ohm
resistors in parallel; these two groups are connected in series and the whole is shunted by a 150-
ohm resistor. What is the overall resistance ?

7. A meter offers a full-scale-deflection (f.s.d.) of 100 microamps but it is desired to measure a

maximum of 1 milliamp. The solution is to shunt away all the current in excess of 100
microamps; this requires that a "shunt" resistor be connected in parallel with the meter thus by-
passing the excess current. If the resistance of the meter is 180 ohms calculate the value of the
required shunt.

>>>>>>>>>>>>>>>>>>>> PAGE 43 >>>>>>>>>>>>>>>>>>>>

Back to Top of Page

LESSON 8
2.3.7 The Coaxial Transmission Line

The transmission line considered above consisted of two conductors whereon the currents
and voltages on each leg differed only in that they were 1800 out of phase; such a line does not
have to be balanced about earth of course but normally, both for safety and to equalise the
effects of stray capacitances, it would be operated as a balanced feeder see under 1.1O.2:
Transformers.

One disadvantage of such a feeder has been mentioned in that, should it become unbalanced,
then an overall external field appears and the feeder radiates.

** The coaxial transmission line overcomes the problem of radiation because one of the
conductors is formed into a tube which completely surrounds the other. Between the conductors
the traveling e-m fields which carry the power are now formed into a totally enclosed system.
Given that the outer conductor has sufficient thickness then the signal current is confined to its
inner surface by skin effect. Thus the outer surface of the outer conductor can be earthed so
forming it into both a capacitive and a magnetic screen (see under 1.7.4 and 1.9.6).

** The resulting unbalanced feeder often causes some confusion. It is important to remember
that, as described under 1.9.4: Skin Effect, rf currents flow only in a thin layer on the inner
surface of the outer conductor. The outer of a coaxial feeder is indeed earthed on the outside and
it should be free of signal currents; on the inside however it can -and does-provide a conducting
path for the signal currents and there are rf voltages between it and the inner conductor.
 In turning concepts into engineering realities however it is usually necessary to adopt
compromises between conflicting requirements. Large-power installations use very large and
robust feeders which are permanently installed and designed for a specific job; the outer
conductor will be manufactured from sheet metal whose thickness ensures that the rf-skin does
not penetrate to the outside.

** The more usual demand is for a smaller feeder which is bought in bulk and installed in a
manner which owes more to convenience than to technical requirements. Such a feeder must
offer a degree of flexibility so that it can be laid around bends or corners and change between
horizontal and vertical runs. To make the inner conductor flexible is not difficult but the larger-
diameter outer presents problems; these are usually overcome by constructing it from fine
copper threads braided together, from copper tape which is overlapped in a spiral lay or
sometimes both constructions may be used together in a single feeder.

** These flexible outer conductors possess neither sufficient thickness nor sufficient
continuity to prevent the currents on the inner surface penetrating to the outside surface and so
there is a penalty in that the feeder is not so efficiently screened. It is the quality of this outer-
construction which is largely responsible for the differing costs of coaxial cables. Heavy braid
laid-up over overlapping tape produces the best screening, the least flexibility and the greatest
cost.

Balanced so-called open-wire feeders require insulators only at relatively long intervals and
their function is mainly to keep the two conductors at the required distance apart. Inside a
coaxial feeder however there is a requirement for more robust support to keep the thin inner
conductor accurately in place even when the

>>>>>>>>>>>>>>>>>>>> PAGE 44 <<<<<<<<<<<<<<<<<<<<

cable is bent. This may take the form of discrete insulators, a thin thread wrapped spirally
around the inner or some kind of expanded/molded plastic filler; although its presence is
unavoidable such insulating material reduces the efficiency of the feeder and the choice of
material is another major item in performance and cost.

As with balanced feeders the characteristic impedance of the line is determined by the
spacing of the conductors; the closer the conductors the lower is the line impedance. However
this means that the lower-impedance cables are more susceptible to flash-over problems as the
power is raised although of course the lower impedance means that lower voltages are
developed for the same power.

For many years the standard impedance for small coaxial cables was 75-ohms but, with the
advent of suitable insulating materials, 50-ohm cables are now readily available. Very thin
coaxial cables of 75-ohm impedance are regularly used for receiver installations where they
match the centre-point impedance of a dipole that is not required to handle power.

2.3.8 Stubs

** In the discussion of a resistive termination with ac excitation it was mentioned that a

mismatched line can present a series of points, separated by a quarter-wavelength, where the
impedance is resistive; these values are alternately greater than and less than the characteristic
impedance of the line. These resistive-impedance values are related to each other and to the line
characteristic-impedance by the expression:

[Link] = Zo2

where Z 1 is the lower value, Zh is the higher value and Zo is the characteristic impedance of the
transmission line.

The phenomenon is easiest to understand when the comparison is made between points of
purely-resistive behaviour but, in fact, the relation holds for the complex-impedances
encountered at any pair of points which are separated by a quarter-wavelength. The discussion
can be extended by noting that between other points on the line which are not separated by a
quarter-wavelength almost any complex relation can be found by suitable choice of measuring
points.

The phenomenon is very useful in several ways:

2.3.9 Impedance Matching: The Quarter-wave Transformer

The constantly-changing impedance along a length of transmission line provides a means of

converting a given impedance to another value. For example, if the driving-point impedance of
an aerial is measured as 38 + j29 (38 ohms of resistance in series with 29 ohms of inductive
reactance) and it has to be connected to a 50-ohm coaxial feeder, then some means must be
provided to transform the aerial impedance to 50 ohms to avoid a serious mismatch.

** Consider a length of transmission line of characteristic impedance Zo (which is to be

determined) on which a standing-wave has been set up. Travel down the line until a point is
found where the resistive value (either greater than or less than Z 0 ) equals 50 ohms; cut the line
at this point and

>>>>>>>>>>>>>>>>>>>> PAGE 45 <<<<<<<<<<<<<<<<<<<<

substitute the feeder so producing a match between the feeder and the piece of line. Travel
further down the line until a point is found where the line impedance is 38 + j29; cut the line at
this point arid substitute the aerial so producing a match between the aerial and the piece of line.

** The result of this exercise is a 3-part system - aerial/line/feeder - wherein each section is
matched to its neighbouring section thus producing an overall match. Note that the only part of
the system which supports a standing-wave is the piece of line - the Matching Section- which is
being used therefore as an impedance transformer.

** There remains two practical problems:

(a) to find the appropriate value for Zo

(b) to find the required length of the piece of line.

Usually it is easier and more practical to solve two questions with two separate answers. At the
beginning of this Section there is a formula which relates two resistive impedances when they
are connected by a quarter-wavelength of line which is therefore known as a quarter-wavelength
transformer; thus the required two solutions are:
(i) correct the reactive part of the impedance by adding the same number of ohms of opposite
sign (in this example 29 ohms of capacitive reactance)

(ii) use a quarter-wave transformer to transform from 38 to 50 ohms.

The length of line required is determined of course by the frequency of the signal ; the required
line impedance is given by:

Zo = √(38 x 50) = .√1900 = 44 ohms (approx)

2.3.10 Impedance Matching: Stubs

** Stubs are short lengths of transmission line which are used as reactive loads in correcting
complex impedances as proposed in (i) above. In the quarter-wave- line equation the value of a
characteristic impedance is a constant and so, if the termination at one end of the transformer
increases (say), it follows that the termination at the other end must decrease if the result of
multiplying them together is to remain constant.

** Between any two resistive points however there are regions where the relation between
voltage and current is always reactive and this provides a series of inductors and capacitors with
very useful properties. Aerial circuits tend to be mainly low-impedance around 50 - 75 ohms
which means that the transmission of any reasonable amount of power involves considerable
current-flow; equally the presence of a standing-wave means that, in places, considerable
voltage develops. Thus any component used to correct a mismatch must be able to withstand
high voltages and at the same time to conduct large currents.

** A piece of transmission line is obviously ideal for the job because it performs those
functions anyway. To set-up the necessary standing-wave the stub must be mismatched; if the
stub is not to dissipate the signal power then only two terminations can be allowed namely an
open-circuit or a short-circuit.

>>>>>>>>>>>>>>>>>>>> PAGE 46 <<<<<<<<<<<<<<<<<<<<

When a stub is cut to a length less than a quarter-wave then a short-circuited termination
produces an inductive reactance at the input terminals; as might perhaps be expected an open-
circuited termination (i.e. no termination at all) produces a capacitive reactance at the input
terminals.

When a stub is cut to a length greater than a quarter-wave then the effect is reversed; a short-
circuit termination produces a capacitive reactance at the input and an open-circuit termination
produces an inductive reactance.

At first these various options may seem confusing but, in practice, they are very useful.
Consider a short-circuited stub which is cut a little too short for it to resonate at a particular
frequency; this means that the signal frequency is too low for the stub to resonate? As stated
above such a stub presents an inductive reactance.

As the frequency is steadily raised so the stub comes closer to resonance eventually to pass
through the resonant condition and then to present a capacitive input impedance. Had the stub
initially been open-circuited then, as the frequency was increased, so the input impedance would
have passed from a capacitive value to an inductive value.

** These differing effects make it important to select the correct type of stub when carrying
out step (i) above. The aerial input-impedance was given as 38 + j29 which means that it is
inductive and requires correcting with a capacitive stub. There is a choice between an open-
circuited under-length stub or a short-circuited over-length stub and, to make this choice, it is
necessary to test the aerial to see which way the impedance changes as the frequency is either
increased or decreased.

** An aerial which has been corrected with a "lumped" capacitor (i.e. a component) is
matched only at the frequency at which the initial measurement was made. When the correction
is made using a stub then the reactance presented by the stub changes also with the frequency;
by choosing a stub whose reactance changes in the opposite way to that of the aerial it is
possible to maintain the correction (nearly) over a range of frequencies.

** In all branches of engineering perfection is unattainable and it is necessary to decide limits

within which any given parameter must be kept; such tolerances can be defined either as absolute
quantities or as a percentage of the given value and the choice is determined most often by the
circumstances. In impedance- matching exercises the tolerance can be given in two ways:

Percentage: for example 50 + 10% means that the impedance must remain resistive and
within the range 55 to 45 ohms.

S.W.R. The amplitude of a standing-wave is determined by the degree to which the system
fails to achieve a perfect match; hence the swr can be used to indicate the degree of mismatch
irrespective of the resistance/reactance ratio which is causing the standing-wave.

For example the 10% tolerance specified above would yield a swr of 1.1 or 0.9; strictly speaking
this should be written as 1.1:1 but omission of the ratio does not cause ambiguity.

(Other ways of expressing a mismatch are not required for the R.A.E.)

>>>>>>>>>>>>>>>>>>>> PAGE 47 <<<<<<<<<<<<<<<<<<<<

In the example given above the aerial was stated to have an impedance of 38 + j29 which
means that it was measured as a resistance of 38 ohms in series with an inductive-reactance of
29 ohms; to correct this the capacitive stub must be connected in series between the aerial and
the 50-ohm feeder. It depends of course on the particular problem but it is usually inconvenient
to connect in series and certainly it is usually easier to connect transmission-line stubs in parallel
across a feeder.

For this the measured aerial impedance must either be re-calculated as a resistor in parallel
with a reactance (a so-called series-parallel transformation not required for the R.A.E.) or it
must be measured on an Admittance Bridge. Note however that a result of 38 + j29 from an
Admittance Bridge represents a capacitive reactance and so the stub must be organised as an
inductive element (which changes with frequency in the correct manner!) which is then
connected across the aerial terminals in parallel with the feeder.
 It is not unusual for an aerial to be designed so that its feed-point requires correction because
the (reverse) action of the stub spreads out the range of frequencies over which the matching
remains within tolerance; in technical language the Bandwidth of the aerial is increased. (This
increase of course is with respect to the single-frequency correction that would be achieved by a
simple component.)

Detailed knowledge of transmission-line transformers and stubs is not required for the
R.A.E. but you should understand the general principles of their usage.

2.3.11 Transmission Lines as Resonant Circuits

In the previous Section it was shown that a stub can be used as a reactance; moreover this
reactance showed resonance properties in that, as the signal- frequency passed through the
frequency at which the stub was exactly a quarter-wavelength, so the stub-reactance reversed
sign. It would be reasonable to propose therefore that a stub could be used as the tuned-load for
an amplifying arrangement and indeed this proves to be correct.

A stub cannot be used in this way except at very-high frequencies because of the practical
difficulties in stowing a long length of line; long lines tend to be lossy also. However, as
frequency is increased, so the values of capacitance and inductance tend to decrease and
eventually components become impossibly small; it is at this point that transmission-line
circuits come into their own.

At 100 MHz for example a quarter-wave line is about 10 to12 inches long; this can be
varied by cutting off the end of the line and substituting a lumped component either in the form
of a capacitor or an inductor. Use of a variable capacitor makes it possible to tune the line. Such
resonant-line circuits can be used in the same way as LC resonant circuits to build rf amplifiers,
oscillators and filters; for high-power vhf transmitter installations they provide the only
practical way to build the necessary filters.

Yet another advantage of resonant-line circuits is their size; as conventional components

become too small so difficulties arise in construction. Resonant lines spread-out the circuit.

>>>>>>>>>>>>>>>>>>>> PAGE 48 <<<<<<<<<<<<<<<<<<<<

Under 2.3.1 (Fig.18) the idea was introduced of an "artificial line" which, although
constructed from lumped components, behaves exactly as a transmission line. At high
frequencies artificial lines are employed to build so-called distributed amplifiers with the aim
of eliminating the ill-effects of shunt stray capacitances introduced mainly by valves. The idea is
to use the valve capacitances, augmented if necessary with lumped capacitances, as the shunt
elements in an artificial transmission line; it is a good engineering concept to make use of
something which you cannot eliminate. In this application the line is not used in a resonant
mode and so must be correctly terminated. ( See also PART-3: Electronic Amplifiers.)

Transmission-line circuits - or distributed circuits - offer advantages in stability over lumped

circuits and very stable oscillators can be built.

END OF LESSON 8

* * * * * * * * * * * *
 QUESTIONS

1. Why do coaxial feeders often radiate even though the outer conductor is properly earthed ?

2. How is a quarter-wave transformer used to match a load ?

3. Why would a stub be connected across the driving-point of an aerial?

4. An aerial used for transmitting sets up a swr on its feeder of 1.4:1 What information is NOT
given by this statement?

5. A 4-ohm loudspeaker removed from a transistor amplifier is to be used with a valve

amplifier. Given that the valve needs to be presented with a 3,500-ohm load give a practical
value for turns-ratio in the necessary matching transformer.

>>>>>>>>>>>>>>>>>>>> PAGE 49 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 9
2.4 AERIALS

2.4.1 Introduction

In Section 2.3 it was shown that, in a long transmission line, the electron-flow injected by a
signal source at one end may not actually reach the load at the other end; in particular with an
alternating signal electrons merely oscillate up and down one small portion of the line. The
power which is transferred from the source to the load is conveyed by the traveling pattern of
closely-associated electric and magnetic fields.

Fig.21(a) shows a transmission line which, because it is un-terminated, must be supporting a

standing-wave; current cannot flow at the unterminated end and so that becomes a point of
maximum electric force (minimum current). This alternating electric force exists between the
two conductors but, beyond the end of the transmission line, there must be a short distance over
which electric force can
 (a) Leakage of Electric Field (b) End View from Point 0
(c) Leakage of Both Fields (d) E-M Field Propagates Through O

be experienced even though the conductors do not extend that far. In other words the electric
field "leaks" from the open end of the line; diagram (b) shows an end view of the transmission
line and its associated electric field.

Such leakage can be encouraged by fanning-out the end of the line as shown in diagram (c);
this exposes the line where current is flowing and so causes

(a) the magnetic fields which surround each leg of the transmission line to become tilted and so
the magnetic force leaks also.

>>>>>>>>>>>>>>>>>>>> PAGE 50 <<<<<<<<<<<<<<<<<<<<

(b) the leakage of both electric and magnetic fields implies that power is being lost from the end
of the line; diagram (d) shows that the electric and magnetic forces are mutually perpendicular
(at right-angles)

(c) the increasing separation between the conductors progressively reduces the distributed-
capacitance of the line and so the fanned-out end section must have a steadily-modified
characteristic impedance (see under 2.3.3). This implies that the standing-wave pattern on the
transmission line is modified also.
(d) if the characteristic-impedance of the end section changes along its length then, with
reference to its ends, it is acting as an impedance-changing device (see The Transformer under
1.10 and the Quarterwave Transformer in Lesson 8).

These effects combine happily in a single explanation. If power is leaking from the open
end of the line then there must effectively be some kind of load connected which is dissipating
that power. If there is a standing-wave present then that load must be non-matching.

The question which arises concerns the destination of that lost power; what is the nature of
this apparently non-existent load?

The answer, at its most simple, is that the traveling electric and magnetic fields (mutually at
right angles) simply journey on into space launched as it were by the horn-shaped end of the
line. The resulting electromagnetic radiation (e--m radiation) was originally predicted
mathematically by the 19th-century physicist James Clerk Maxwell.

2.4.2 Vector Quantities

A quantity is specified by its magnitude (its size) but that is not its only significant
parameter. To specify an electric force as say 10 volts is not sufficient because we need to know
its polarity. Quantities which have both magnitude and direction are termed Vector Quantities and
they are represented graphically by drawing Vectors; these are arrows which obviously indicate
direction but whose length is proportional to the magnitude of the vector quantity.

A radiating e-m field is specified by three vectors: electric and magnetic forces and the
velocity with which the field propagates (velocity is defined as speed in a given direction). For
an e-m field the velocity in a given medium is constant and so the velocity vector may indicate
direction only.

In a propagating e-m field it is found that the electric and magnetic vectors are always
mutually-perpendicular and that both are perpendicular to the direction of propagation. The
relation between the three vectors is fixed; reversal of any one requires a reversal of one other.
Thus reversal of either the electric or the magnetic vectors forces a reversal in the direction of
propagation; reversal of both vectors leaves the direction of propagation unchanged.

In any given radiating e-m field it is also found that the relation between the magnitudes of
the electric and magnetic forces is constant and that they are always in phase; this can be
interpreted by saying that free-space has a constant resistance as far as a propagating wave is
concerned (see 1.12.1: Ohm's Law). That a propagating e-m wave does indeed represent a
power-flow is demonstrated daily by the ability of "receiving aerials" to extract signal power
from such a wave.

>>>>>>>>>>>>>>>>>>>> PAGE 51 <<<<<<<<<<<<<<<<<<<<

2.4.3 The Dipole

** To send information efficiently to a distant point it is desirable to radiate from the end of the
line all the power that is supplied by the Source at the input end; i.e. the line must be
matched (Maximum Power-transfer Theorem) and this should be indicated by the non-appearance
of a standing-wave.
 To obtain this maximum-possible radiation it is necessary to further encourage the leakage
of electric and magnetic fields and to arrange for the effective "termination" to be a resistive
match to the (resistive) characteristic impedance of the transmission line.

The resistive match is easy because it has been noted already that the electric and magnetic forces
in a propagating e-m field are in phase. All that remains is to provide an impedance-matching
device (see under 1.10: The Transformer) that can transform the effective termination to
the characteristic-impedance of the line.

The required device is obtained by further developing the impedance-changing end-section

of a feeder (discussed above) and the result is known of course as an Aerial. The "effective
termination" is known as the Impedance of Space.

As shown in Fig.22 to obtain maximum leakage from the line requires that the magnetic
field be tilted through 900. However the end of an open-circuited transmission line is a point
of minimum current-f low (see under 2.3.3); if maximum current is to flow in this aerial the
arrangement must ensure that the fan-out includes a point where the current has a maximum
value and, for each limb, this is a quarter-wavelength from the very end of the line.

>>>>>>>>>>>>>>>>>>>> PAGE 52 <<<<<<<<<<<<<<<<<<<<

The radiating section which results is an Aerial (antenna in American and Amateur circles)
and measures a half-wavelength overall; because it has two (opposing) electric poles it is
known as a Dipole or Doublet.
 Wavelength is given the symbol λ (greek lambda) and a half-wave aerial is referred to as a
λ/2 radiator (spoken as "lamda-by-two" or more simply as "lambda-two" or even as "half-
lambda"). Similarly a quarter-wave section is designated as a λ/4 or "lambda-four" section.

Detailed aerial theory is given in many textbooks but the following points are worth noting
here:

(a) current cannot flow at the ends of the aerial and so a standing-wave pattern is to be
expected on the two "limbs" of the dipole; however, if a match is being achieved between the
dipole and the transmission line, then this standing-wave pattern will not extend back along the
line (or Feeder) beyond the start of the fan-out

(b) Signal current flows on the feeder in push-pull; i.e. at any given instant current flows from
one leg of the feeder into the aerial but from the aerial into the other leg. As shown in the
diagram the current in all parts of the aerial flows in the same direction. (This is not true of all
aerials,)

(c) The quadrature (900) relation between voltage and current on the aerial seems to be in
conflict with the statement above that, in an e-m radiation, the electric and magnetic vectors are
always in phase. In fact Maxwell's equations indicate that an e-m field, as it immediately
surrounds an aerial, has two components:

(i) an induction field in which the electric and magnetic vectors are in quadrature but which
extends only a few wavelengths from the aerial. This field has considerable magnitude and is the
cause of short-range interference.

(ii) a radiation field in which the electric and magnetic vectors are in phase and which is
capable of extending to great distances.

(Note that the concept of two fields is but an interpretation of mathematics; in practice the field
changes from near-quadrature to being in-phase over a few wavelengths as it departs from the
vicinity of the aerial.)

2.4.4 The Impedance of a Dipole

** The impedance which a dipole presents at its centre where the feeder is connected is usually
in the range 50 to 100 ohms. The exact value depends on the details of the design.

** It is shown in Fig.22 that the current flows in the same direction throughout the length of
the dipole; it flows into one leg of the feeder but out of the other. Assuming that the feeder is
properly balanced the currents in each leg are equal and so the currents on each limb at the
centre of the aerial must be equal. It follows that it does not greatly affect the operation of the
device if the two limbs of the aerial are constructed from a single conductor as shown
in Fig.23(a).

** It may look like a short-circuit but the standing-wave decrees that the short length of wire at
the centre of the dipole has a p.d. across it and that it carries current at the maximum value.

>>>>>>>>>>>>>>>>>>>> PAGE 53 <<<<<<<<<<<<<<<<<<<<

 There is a range of possibilities for making the connection between feeder and dipole of
which the two extremes are:

(a) feeder connected to the outer extremities of the dipole which effectively terminates the
feeder in an open-circuit

(b) feeder connections both taken to the centre of the dipole which terminates the feeder in a
short-circuit.

** Between these two extremes any desired impedance

can be obtained as a termination for the feeder; as shown
in Fig.23(b) the end of the feeder is physically fanned-out
to be attached at the appropriate points.

2.4.5 The End-fed Dipole

The simple introduction above to the simplest of the

aerials resulted in a dipole which had the feeder
connected at its centre. Because of the low-voltage high-
current conditions at the feedpoint this centre-fed dipole
is often said to be current fed.

** The basic parameter of the Dipole is that it has two

poles or two limbs; these are driven against each other
and the most often-used form is a half-wavelength overall
at the required frequency of radiation.

** If not too far off frequency a dipole will still radiate

well but its centre-point impedance will not be resistive
and so a standing-wave will appear on the feeder. In
turn this means that the feeder will not deliver to the
dipole the maximum possible power.

It has been shown already that a dipole need not

a) Continuous Dipole (b) The necessarily be broken in the middle; where it consists of
Delta Match a single length of wire it is possible to drive it at either
end instead of at the centre. If this appears to be wrong or
somewhat implausible you are to be congratulated;

Question: How can power be fed to an aerial through a

point where current does not flow?

** The answer is that we cheat. Each half of a centre-fed dipole is driven against the other;
that means that limb-a is first driven positive with respect to limb-b and then, during the next
half-cycle, limb-a is driven negative with respect to limb-b. The voltage of each limb is
measured with respect to the other (their relative voltage); the absolute voltage (with respect to
Earth) is not important as far as driving the dipole is concerned.

** To end-feed a dipole one end only is driven first positive and then negative with respect to
Earth. The aerial DOES radiate power as a consequence and so it
 >>>>>>>>>>>>>>>>>>>> PAGE 54 < < < <<<<<<<<<<<<<<<<<

must be accepted that current flows of necessity from the feeder into the end of the dipole. At
the far end of the dipole it is still true to say that current has nowhere to flow and so this is a
point of maximum voltage and zero current but, at the driven end (the input end), conditions are
not the same.

The dipole supports a standing wave and so the two ends, a half-wavelength apart, could be
expected to have similar conditions; they differ by a small amount because the driven end has
been modified by connecting a feeder to it. Thus, while the free end is theoretically an open-
circuit, the driven end must settle for no more than a high impedance; i.e. it is a point of high
voltage and low current rather than zero current.

** The basic difference between the two dipole configurations therefore is that the centre-fed
one requires a low-impedance drive (a current drive) and the end-fed requires a high-impedance
drive (a voltage drive).
.

But how is it achieved? Fig.24 (a) shows the ba

have been devised wherein a generator is connect
and ground. This indeed provides all that is neces
arrangement with practical hardware, would requi
or a very-low aerial. Clearly a transmission line m
this matter that confusion usually arises.

** When a transmission line is taken to the end o

has to be connected to the end of the radiating con
ground (against which the dipole is to be driven).
. the dipole needs to be elevated above the ground.

** The solution is to provide some kind of artific

for the ground but which is in fact well above the
arrangements have been devised but the most-ofte
leaves the other feeder connection up there next to
connected to anything.

** Yes, it is unhygienic and yes the practice does

arrangement, as illustrated in Fig.24 (b), is known
and it can give good results.

>>>>>>>>>>>>>>>>>>>> PAGE 55 <<<<<<<<<<<<<<<<<<<<

** Six paragraphs above it was said that the feeder modifies the conditions on the aerial at the
driven end. It is equally true to say that the feeder-conditions are modified when an aerial is
connected to it. By drawing current from one leg of the feeder only, the aerial upsets the relation
between the standing-wave patterns of the two legs. Which perhaps is a long-winded way of
saying that the end-fed dipole produces an unbalanced mismatch at the feeder-connection.
 ** On a properly-balanced feeder, whether or not it is connected to a matching load, the
current at any given point flows in opposite directions on the two legs. Thus, external to the
feeder, little if any field is encountered because the contributions from each leg cancel. When
that balance is upset, as it is with a simple end-feed, this cancellation fails and an external field
appears around the feeder. In other words the feeder adds its own quota to the overall radiated
e-m field. This may or may not be harmful or desirable and it may cause interference.

** The various developments of the end-feed system all aim to correct this imbalance and so
prevent radiation from the feeder; they are described in later Sections which deal with aerial-
matching. It might be thought that the Coaxial Feeder, which is described in Section 2.3.5,
would overcome this difficulty but it is found that, when aerial and feeder are improperly
matched, current flows on the outside of a coaxial feeder and so causes unintentional radiation.

END OF LESSON 9

* * * * * * * * * * * *

QUESTIONS

1. How is power carried by a transmission line ? What is the basic principle employed in
transforming this power into an e-m wave.

2. An aerial can be viewed as an impedance-matching device which couples a feeder to free

space. Standard feeders come in 50-ohm and 75-ohm types; how can both these be matched by
a single dipole aerial?

3. Your "shack" of the future might be equipped with both a centre-fed dipole and an end-fed
dipole. How would you expect the feed arrangements to differ?

4. In broad terms only would you expect to find a series-resonant circuit or a parallel-resonant
circuit driving an end-fed dipole?

5. A network of resistors consists of 20-ohms in series with 10-ohms shunted overall with 30-
ohms. A 9-volt battery, connected across the combination, has a measured internal resistance of
3-ohms. What is the terminal voltage of the battery when the network is connected?

>>>>>>>>>>>>>>>>>>>> PAGE 56 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 10
2.4.5 Parasitic Radiators

In the previous Section it was argued that current flowed on a radiating conductor as a result
of its being driven from the end of an energised feeder. However, it cannot be supposed that the
radiator "knows" how the current came into being and so it follows that, as long as an alternating
current is induced to flow, then the conductor will launch an e-m wave.

Such a current can be induced by placing a conductor in a radiated e-m field. The simplest
to understand, and the most effective combination, is when the conductor is a half-wavelength in
length and matches the radiator which launched the e-m wave; it also matches the physical
dimensions - in terms of wavelength of the wave. The electric component of the field produces
a voltage between the two ends of the conductor and so current flows.

The argument thus comes full-circle in that the new conductor with its induced alternating
current must in its turn radiate an e-m wave. The phenomenon whereby a conductor first absorbs
power from a passing wave and then radiates that power back into space is known as re-
radiation. A conductor which is behaving in this manner is referred to as a parasitic radiator -
one which is stealing its power.

Parasitic radiators can be very useful devices and are frequently used in constructing aerial
arrays in which the purpose is to control the direction in which radiation takes place. (See later
under 2.4.9.) It is important to remember however that any piece of conducting material
behaves in this manner and, when close to a transmitting aerial, rusty joints or intermittent
contacts are responsible for interference (see under E.M.C.) and noisy reception.

2.4.7 Vertical and Horizontal Polarisation

The simple development of the dipole aerial appears in Fig.22 to refer to a vertical aerial but
in fact the argument is not altered by turning the dipole/feeder assembly through 90-degrees thus
forming a horizontal dipole. When considered as a radiating element in free space the
terms vertical and horizontal are meaningless.

While the attitude of any particular conductor is not important to launching an e-m wave it
matters a great deal when trying to recover the wave by absorbing it into another conductor. The
electric component of an e-m field will not induce an emf into a parasitic conductor unless that
conductor is aligned with the electric component.

The direction of the electric component is known as the polarisation of the radiated wave
and clearly it is defined by the radiating conductor. Thus a horizontal dipole is said to radiate
a horizontally-polarised wave and a vertical dipole radiates a vertically-polarised wave.

The use of a dipole in the wrong attitude does not necessarily preclude the establishment of a
communication channel but it very much weakens the received signal. Cross-polarised
working is possible because the plane of polarisation is disturbed by re-radiation and also during
propagation (see later).

>>>>>>>>>>>>>>>>>>>> PAGE 57 <<<<<<<<<<<<<<<<<<<<

A practical point (which is not very important for the R.A.E.) is that a horizontal dipole is
balanced with respect to Earth while a vertical dipole has its lower limb closer to Earth and so is
subject to increased stray capacitance on that limb. Unless the vertical dipole is mounted several
wavelengths above the Earth's surface an out-of-balance current flows on the outside of a
coaxial feeder and causes difficulty with matching.

2.4.8 The Unipole or Quarter-wave Aerial

Most practical aerials are required to operate within a few wavelengths of the Earth's surface
(or some other surface such as the skin of an aeroplane). Any part of the e-m field which is
radiated below the horizon line causes rf currents to flow in the adjacent surface which therefore
acts as a parasitic radiator and re-radiates a secondary signal.

As illustrated in Fig.25 this re-radiated signal can be regarded as a reflected signal and so
the surface acts as a mirror. The re-radiated signal appears to come from a secondary aerial
below the surface.

This is an extremely useful phenomenon where long wavelengths mean that a λ/2 dipole will
be physically too large for a practical construction; the conductor is cut to a length λ/4 and
mounted over a reflecting (re-radiating) surface so that it appears effectively to be λ/2. The
resulting half-dipole has only a single "pole" and so it is often referred to as a unipole. Very often
the supporting mast becomes the vertical aerial; where the mast lacks the height to resonate at
λ/4 it is loaded with a coil (in effect the extra height required is coiled up) and such a device is
known as a Marconi Aerial.

>>>>>>>>>>>>>>>>>>>> PAGE 58 <<<<<<<<<<<<<<<<<<<<

The unipole aerial is useful at high frequencies in solving the problem of providing a virtual
earth; the quarter-wave radiator is mounted above a reflecting surface which is connected to the
external shield of a coaxial feeder.

The reflecting surface is known as a counterpoise or as a ground plane and it should be at least
λ/4 in radius. A difficulty with such a ground plane is that it offers considerable wind resistance
and so the reflecting surface may be replaced with a wire mesh or by a system of λ/4 rods. Such
rods are usually four in number radiating at 900 intervals; occasionally aerials are seen with
only two such rods but, in practice, it is found that their number and direction is of less
importance after the first one has been connected.

A unipole is essentially one-half of a centre-fed dipole and so its driving-point impedance

(dipole-end to groundplane) is around 35 ohms.

2.4.9 Directional Aerials (Beam Aerials)

 A directional aerial is one which radiates the majority of its power in a given direction.
However the same aerial when used for reception concentrates its receiving performance in that
same direction. Such aerials can therefore provide a gain with respect to a simple dipole; note
however that they do not amplify the signal but merely concentrate in one direction the power
that otherwise would be broadcast. The gain is available at both ends of a communications
channel.

Directivity results from radiating two (or more) signals from points which are separated by a
critical distance; the separation means that in some directions the two signals are in phase and
so add to form a large signal while, in other directions, the two signals are out-of-phase and
mutually cancel. This is illustrated in Fig. 26.

>>>>>>>>>>>>>>>>>>>> PAGE 59 <<<<<<<<<<<<<<<<<<<<

The two positions A and B represent two radiators which are separated by a half-wavelength
but which are fed with identical signals . The overall field which they lay-down depends on the
position of the reception point; along their common line either in direction AB or BA one signal
arrives a half-cycle later than the other and so cancellation results; at 900 to the common line
the signals arrive simultaneously and so add.

As an alternative the two radiators could be fed with signals that are 180 out of phase; in
such circumstances the maximum radiation occurs along the line AB and minimum radiation at
900 to the common line.

The interfering signals are derived either by sharing the available power between separate
radiators or by making one radiator parasitic on the other. The first technique results in so-
called arrays; the various forms of curtain array, end-fire array, broadside array, steerable array
etc. are described in the literature. Such aerial systems require a lot of room with substantial
masting and rigging and are not of interest in Amateur circles.
 The second technique using parasitic elements comes into its own at those frequencies where
the wavelength makes it practical to use self-supporting rods as the radiators. The ultimate in
such designs is the greatly-favoured Yagi aerial.

The first step is to place a parasitic element (or a reflecting plate/mesh) "behind" the radiator
to act as a reflector. The resulting "H" aerial was a common domestic feature during the reign of
the Band-I Television Service. The reflector and radiator both are resonant (or tuned) elements
and so their length is critical; if cut over or under the correct length their impedance exhibits
either inductance or capacitance which causes phase-shift effects. Such phase-errors can be
corrected by altering the spacing of the two conductors and this facility is exploited to design
aerials which are smaller than the basic λ/2 dipole

The next step is to add other parasitic resonators (directors) "in front of" the driven radiator;
there is little point in placing them behind the reflector. These are usually cut progressively
shorter toward the front of the aerial and again correction is applied by varying the spacing. Not
only does such trickery reduce the physical size of the aerial but it allows the bandwidth and
matching to be adjusted.

Bandwidth as applied to an aerial is the range over which its driving-point impedance
matches a given feeder but, in practice, it is very much a matter of how it is defined. A yagi
aerial is essentially a set of resonators coupled together electromagnetically. Coupling means
that an adjustment to any one radiator causes a change in the impedance coupled into all the
others and so they become detuned. As a result it is all but impossible to set-up such an aerial
with accuracy so that it gives a constant performance over a required range of frequencies.

The best that can be required is that it should remain within a certain swr but even that
proves to be difficult except at a few spaced frequencies, This is why television aerials are
specific to areas and there is seldom any point in taking the aerial with you when you move
house.

The above description shows that power is concentrated in a given direction but is does not
make clear that this effect takes place both around the horizon (in azimuth) and also in the
vertical plane (in altitude). Thus a multi-element yagi is

>>>>>>>>>>>>>>>>>>>> PAGE 60 <<<<<<<<<<<<<<<<<<<<

capable of producing a pencil-like beam which has some very distinct disadvantages. Such an
aerial must be aimed very accurately in both azimuth and altitude or it may cause signal-
attenuation instead of the intended signal-gain.

To complicate matters still further the addition and cancellation of signals is seldom that
complete and the pencil beam is found to be accompanied by a whole family of side lobes;
some of these side-lobes may be in the rearward direction where the reflector is supposed to
have eliminated radiation. The gaps between lobes (null points) are usually fairly sharply
defined and can be useful in eliminating powerful interfering signals during reception but it is all
too easy to align a yagi aerial so that one of the side-lobes is in operation rather than the main
lobe.

There are two important parameters of directional aerials about which there is often
confusion:
(a) Front-to-back Ratio This is the ratio of signal strength radiated (or received) in the
forward direction to that radiated in the rearward direction. (For those aerials in which a
rearward lobe does not exist then the nearest side-lobe is used,)

(b) Aerial Gain This is the actual forward gain which the aerial affords relative to a simple
dipole because it diverts all the power into a given direction. (Strictly speaking it is measured
with respect to a theoretical perfect radiator, known as an isotropic radiator, which radiates
equally in all directions throughout a sphere; in practice it is generally of more interest to
compare with a simple dipole.)

The difference is easiest to grasp by considering that an inefficient aerial may have a large
back-to-front ratio (i.e. may be very directional) yet produce a signal which is poor in
comparison with that from a dipole.

2.4.10 The Folded Dipole

In multi-element aerials such as the yagi described above each parasitic element absorbs
power from, and therefore loads, the driven element. The effect of this is to lower the driving-
point impedance of the main radiator and so make it difficult to achieve an impedance-match to
the feeder.

The folded dipole illustrated

in Fig.27 can be regarded as a "squashed"
loop and, as with the Loop
Aerial described in the following Section,
it offers a driving-point impedance of 300
ohms. This is of course reduced by the
proximity of parasitic elements but the
final value is closer to the 50-ohms or 75-
ohms of standard coaxial feeders.

>>>>>>>>>>>>>>>>>>>> PAGE 61 <<<<<<<<<<<<<<<<<<<<

2.4.11 The Loop or Frame Aerial

This aerial is often referred to (mistakenly) as a magnetic loop. To achieve communication

over any distance an aerial must generate e-m waves. The magnetic or induction loop is used at
low (even audio) frequencies and is characterised by its very limited range outside the area of
the loop.

The impedance presented by a radiating loop is around 300 ohms and it does not support a
standing wave.
 Because the loop-current is uniform (i.e. a standing-wave is not present) the external field is
the direct sum of the e-rn field radiated by the opposing limbs. In Fig.28 the current in AB
opposes that in CD when viewed from a point on the loop's axis and so cancellation takes place
in this direction. When viewed from any direction in the plane of the loop such cancellation
cannot be total because of the phase difference produced by the different distances from the
observer.

Thus a loop aerial shows directional properties but it cannot be classed as a beam aerial
because it is not diverting power into any given direction and is not providing a gain.

The radiation pattern of a loop is a single lobe which resembles a large doughnut
surrounding the loop. When mounted vertically a loop provides a sharp null in azimuth; this
can be useful in reception for eliminating an unwanted signal and is also used for direction
finding. For transmitting purposes however a loop is best used horizontally in which position it
covers 3600 in azimuth and most of the useful angles in altitude (see
later Skywave under Propagation).

An interesting property of loop aerials is their ability to work at any height above ground
from zero upward. They also pick-up very little background noise.

>>>>>>>>>>>>>>>>>>>> PAGE 62 <<<<<<<<<<<<<<<<<<<<

2.4.12 Other Aerials

Aerials for both transmitting and receiving (or both) are available in a multiplicity of forms
which are well documented in the literature. Their study here would not add anything to the
general understanding of the subject and this Course does not deal further with them. However I
am always willing to accept questions or requests for help.

END OF LESSON 10

* * * * * * * * * * * *

QUESTIONS

1. What is the purpose of a parasitic radiator?

2. What is meant when a radiation is said to be vertically polarised?

3. What is the essential component of a unipole aerial?

4. Why is it that a loop aerial cannot be regarded as a beam aerial?

5. Why is a folded dipole commonly seen in multi-element aerial arrays?

6. A circuit used to couple a transmitter to a feeder consists of a series inductor followed by a

capacitor connected to Earth. In terms of frequency only how would you expect such a circuit to
behave?

7. A tuned parallel-circuit with a measured Q of 20 draws 20 mA from a generator by which it is

being driven at its resonance frequency. What is the circulating current within this tuned circuit?

>>>>>>>>>>>>>>>>>>>> PAGE 62 again <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 11
2.5 PROPAGATION

2.5.1 General

The term propagation refers to the manner in which radio signals travel from a radiating
aerial to a receiving site. The behaviour of a propagating e-m wave depends on its frequency (or
wavelength), the time of day, the time of year, the nature of the Earth's surface-covering, the
condition of the Earth's atmosphere and on the sunspot cycle.

In Lesson 9 it was stated that a transmitting aerial sets-up a radiation field that can be
regarded as two separate components namely an induction field and a radiation field. The
induction field does not extend beyond a few wavelengths from the aerial.

The radiation field can itself be divided Into

two separate components as shown in Fig.
29. One of these is dependent on currents which
flow in the Earth's surface and so it remains
close to the ground and is known as the ground
wave The launch of such a wave is possible
only by a vertical aerial which is at a height
comparable with the wavelength. (The electric
component of any ground wave launched by a
horizontal aerial would be short-circuited by
the Earth.)

The propagation of a ground-wave depends

on the conductivity of the earth and its energy
is dissipated in earth currents. Ground-losses
increase with frequency and so range decreases
 as frequency rises.

The other component, independent of the Earth, is known as the space (or sky) wave. This
component is itself divided into a direct wave and a ground-reflected wave. The Earth's upper
atmosphere is powerfully irradiated by the Sun and this causes it to ionise; i.e. to break up into
positively-charged and negatively-charged particles (ions) which are capable of carrying electric
currents. These form in several layers at different heights and, by absorbing and re-radiating rf
energy, they act as "reflecting" layers. The Sun's activity varies in an 11-year cycle and this is
reflected in the number of sunspots visible on its disc. Thus sky-wave propagation phenomena
are linked to the Sunspot Number.

In general signals which are radiated at high angles of elevation tend to pass straight through
these layers and are lost to Space. Radiation below certain critical angles is returned to Earth
with varying effects. The distance between the transmitting site and the point where the signal
returns to Earth is known as the skip distance.

2.5.2 Long Wave Band

At frequencies up to about 300 kHz the ground wave is very strong, propagates over very-
long distances and may penetrate to a significant depth; frequencies around 15 to 20 kHz are
used to communicate with submerged submarines.

>>>>>>>>>>>>>>>>>>>> PAGE 63 <<<<<<<<<<<<<<<<<<<<

In the frequency range 20-kH to 100-kHz both ground-wave and sky-wave absorption is very
low and most of the transmitted energy can reach receiving points up to 1,000 km by way of the
ground wave. Received signals show very little daily, seasonal or yearly variation.

At very great distances where the ground-wave is completely absorbed the received signals
are entirely sky-wave and show diurnal (day & night) and seasonal variations; the strongest
signals occur at night and in the Winter. Transmission is also characterised by a sudden drop in
signal strength when the sunset line falls across the transmission path but, otherwise, fading
does not occur at these very low frequencies.

In the frequency range 100-kHz to 550-kHz the ground-wave attenuation increases thus
raising dependence on the sky-wave for distance communication. Sky-wave attenuation
however tends to be much greater in the daytime and in the summer and, in general, long
distance communication at these frequencies is not possible during the day particularly in
summer.

2.5.3 Medium Wave Band

This Band extends from 300 kHz to 3 MHz ( λ/4 = 250 to 25 metres) and quarter-wave
resonant conductors become practicable; at the lower end of the Band the Marconi-type λ/4
vertical radiators are typical. The ground wave is less extensive but extends over sufficient
distance to permit domestic broadcast services. For this purpose the signal needs to be of
sufficient strength to override interference and be as free as possible from fading and distortions.
The area where such conditions prevail is known as the primary service area which typically
extends to around 50 - 100 miles.
 Reflection at the lowest of the ionised layers (the D layer) gives the shortest distance
between transmitter and the return point but, because sky-waves at broadcast frequencies are
completely absorbed in the daytime, broadcast signals rely entirely on ground-wave
propagation.

After dark, as the ionisation decreases, the D layer disappears and signals are reflected from
the E layer above it (and sometimes by the F-layer also). Close to the transmitter the reflected
sky-wave is weak and so the ground-wave predominates. As the distance from the transmitter
is increased so the ground-wave becomes attenuated but the sky-wave increases in strength and
so the primary service area becomes surrounded by one in which the two signals are of similar
strength.

The reflecting layers are not rigidly bounded and so the effective height from which
reflection takes place is variable; this means that the relative phase between ground and sky
waves is constantly changing and so the two signals vary between mutually aiding and mutually
canceling. The result is the after-dark fading which is familiar to all who listen to medium-
wave broadcasts at some distance from a transmitter.

Another feature of this area is selective fading. Whether ground-wave and sky-wave cancel
or aid is determined by the difference in path lengths as measured in wavelengths. The
wavelength differs for each side-frequency and so the relation between carrier and sidebands is
greatly disturbed. This results in severe distortion which varies in much the same way as does
fading.

Beyond the effective area of the ground-wave the sky-wave remains useful for long-distance
communication but with limitations.

>>>>>>>>>>>>>>>>>>>> PAGE 64 <<<<<<<<<<<<<<<<<<<<

Signals sometimes propagate over more than one path between transmitting and receiving
aerials which means that fading and distortion can occur even in areas where the ground-wave
does not exist. These effects occur to a much lesser extent that when a ground-wave is present.

Ground-wave signals in the Long and Medium Wavebands decrease in direct proportion to
the distance.

2.5.4. Short Wave Band

The short-wave Band, which ranges from 3 MHz to 30 MHz, includes all the hf bands
permitted for Amateur use except the so-called Top Band (1.8 to 2.0 MHz). At these
frequencies the ground wave is rapidly dissipated and does not offer any real facilities. Amateur
operating is permitted only in specified ranges of which the higher-frequency bands are of real
use only during periods around sunspot maxima.

In this Band a quarter-wave is between 25 & 2.5 metres and so efficient aerials can be built
in many forms including multi-element directional units.

Aerials may be erected either vertically or horizontally but it is not practical for the Amateur
to set them very far (in terms of wavelengths) above the Earth's surface. Their radiation must
therefore set-up ground currents and so ground conductivity plays an important part in
performance. Over-sea paths perform best but it is common practice to use one or more ground-
wires or counterpoises to improve conductivity at the launch site.

Between the outer edge of the ground wave and the point where a sky-wave returns to Earth
there is an area where signals cannot be received. At the return point however a signal with
sufficient strength can set-up earth currents and so be re-radiated (or re-reflected); such re-
radiation follows the optic rule that the angle of reflection is equal to the angle of incidence and
so the sky-wave is re-launched toward the Ionosphere to be reflected once again. Thus a series of
skips (or hops) make it possible to communicate beyond the horizon and indeed all around the
globe.

However things are seldom perfect and this world-wide progress makes its own demands.
To communicate by sky-wave it is necessary to ensure that the target area does not lie within
one of the skip zones. This requires that the sky-wave be launched at a certain angle and
it must suffer reflection as against passing straight through the reflecting layers. To ensure
reflection means choosing a frequency that will be reflected by the appropriate layer at the given
time of day and at the given angle of incidence on that layer.

Needless to say the prediction of "correct" frequencies is a science in its own right and the
work is based on daily measurements. These predictions are published in table form for the
benefit of the communications world; one source for the Amateur is the monthly publication of
the Radio Society of Great Britain Radio Communication or RadCom. Most amateur aerials
radiate over a fairly-large range of elevations and so the only requirement is to use the correct
frequency-band for the chosen destination and time of day.

Depending on the frequency and angle of launch a reflection of the sky-wave may take place
at any of the ionospheric layers and so several different paths may be available. Over long-
distance routes propagation differs markedly between east-west paths and north-south paths.
This is because of the different distributions of sunlight and different seasonal variations along
those paths.

>>>>>>>>>>>>>>>>>>>> PAGE 65 <<<<<<<<<<<<<<<<<<<<

There is benefit to be obtained from using aerials with directional properties both in azimuth
and elevation. Reflection from the Ionosphere is best when the propagating wave enters at low
angles but, if the wave is too close to the horizontal, it is absorbed by the ground. Vertical λ/4-
type aerials are often used because they give good low-angle radiation.

Vertically-polarised and horizontally-polarised aerials are equally successful for transmitting

because, on reflection from the Ionosphere, signals are found to possess both vertically-
polarised and horizontally-polarised components which bear no apparent relation to the
polarisation of the transmitting aerial. These two components fade independently and are not in
phase. These effects are mainly caused by the Earth's magnetic field.

Communication channels in this band are sometimes completely cut for periods of 15 to 60
minutes by what are known as sudden ionospheric disturbances or S.I.D's. This is believed to be
caused by absorption in the D-layer caused by bursts of ultraviolet light from the Sun.

2.5.5 VHF/UHF Band

 The VHF Band (very-high frequencies) extends from 30 MHz to 300 MHz and the UHF
Band (ultra-high frequencies) from 300 MHz to 3,000 MHz. At these frequencies the ground
wave is non-existent; usually the aerials are mounted several wavelengths above the Earth's
surface and consist either of balanced dipole-types or ground-plane quarter-wave types.

At these frequencies skyward radiations pass straight

through the ionised layers and so all that is left is the
direct-wave which propagates along the line-of-
sight path. As shown in Fig.30 the distance over
which communication can be achieved is governed
by the height of both the transmitting and receiving
aerials.

The diagram shows a true geometric construction

but, in fact, the path-length is greater than indicated.
This is because the e-m wave tends to be refracted by
the Atmosphere and follows a curved path which
effectively lowers the intervening horizon. A
reasonable prediction of path length can be made by
assuming that the Earth's radius is 4/3 its true radius;
i.e. approx 5,200 miles.

It follows that the range over which communication can be achieved at vhf and uhf is
dependent on the height of the transmitting and receiving aerials.

This distance can be drastically shortened however where a tall building or a hill breaks the
transmission path. Although structures such as these are opaque to very-high frequencies weak
signals can be obtained in their radio-shadow due to a phenomenon known as diffraction.
Because of interference effects at an edge a portion of the field is bent down into the shadow
area. Additionally the signal may arrive in the shadow region by reflection from objects on
other bearings so effectively directing a signal around the obstruction. In television-
engineering such multipath propagation causes echoes known as "ghosting".

>>>>>>>>>>>>>>>>>>>> PAGE 66 <<<<<<<<<<<<<<<<<<<<

An elevated vhf or uhf aerial launches a signal directly toward or over any receiving aerial
but it is not prevented from radiating below the direct line-of-sight path. Such signals must
strike the Earth's surface, they induce current-flow and so suffer reflection. As shown
in Fig.31 this reflected signal travels over a longer path than the direct signal; the path
difference (in wavelengths) will depend on the overall separation of transmitting and receiving
aerials, on the height of each above the ground and on the frequency.

On reflection at these low angles the

signal suffers a phase-change of 1800 with
the result that the direct and reflected
waves tend to cancel and so reduce the
strength of the received signal. As one or
 other of the aerials is raised so there must
be a steady increase in the path difference,
Fig. 31 Signal-reduction Caused by the cancellation is reduced and the received
Ground-reflected Component signal increases in strength. This
phenomenon is referred to as height-gain.

Although there is a real bonus available by mounting vhf and uhf aerials as high as possible
(apart from line-of-sight considerations and avoiding the screening action of buildings etc.) the
height-gain effect does not continue endlessly. At some point the path difference reaches 1800
and further increase in height will result in a height-loss. This is a consideration only in
commercial fields where masts of 500-ft or more may be worth considering but the planners
need to be aware that the additional cost of a higher mast might result in a signal loss. Height-
gain is taken for granted in Amateur circles although a few who live on the edge of high ground
may find that their ground-reflection point is on a plain several hundred feet lower and could
result in a height-loss.

Where propagation involves interaction between direct and ground-reflected components of

a space-wave the signal is attenuated inversely as the square of the distance; compare this with
the attenuation of a

Anomalous Propagation

This term is applied to signals which arrive unexpectedly and apparently by breaking the
accepted rules. It is usually the result of the signal encountering a propagation path wherein the
attenuation is less than normal.

Apart from the regular ionised layers there is an irregular and unpredictable type of
ionisation which occurs at about the level of the E-layer. This takes the form of "clouds" which
may be many kilometres across and which may last from a few minutes to a few hours. These
clouds have been known to reflect signals up to 80 MHz and such propagation is known
as sporadic-E. Sporadic-E does not occur when weather systems keep the Atmosphere well
mixed.

>>>>>>>>>>>>>>>>>>>> PAGE 67 <<<<<<<<<<<<<<<<<<<<

In general the properties of the Atmosphere follow a standard pattern with height above the
Earth's surface but irregularities in temperature-gradients and moisture content often bring about
changes in the manner in which the Atmosphere refracts (bends) e-m waves. Instead of
producing the effect already mentioned whereby the Earth can be considered as having its radius
increased by 4/3 the refraction may be either increased or decreased.

Abnormal refraction can result in the path of a propagating wave following the Earth's
curvature thus traveling distances much greater than normal. At high frequencies it is possible
for the wave to be trapped between two layers thus drastically reducing attenuation. Such effects
are known as ducting.

The path can also be bent away from the Earth's surface so reducing the range.
 Other areas of ionisation produce visual effects known as the Aurora and these too can be
used for communicating over paths not otherwise available. Yet another path often explored by
Amateurs is to "bounce" signals off the surface of the Moon.

END OF LESSON 11

************

QUESTIONS

1. What are the causes of fading?

2. Above 30 MHz what determines the maximum range of reliable transmission?

3. Why is it that in the Amateur Bands we can use either vertically-polarised or horizontally-
polarised aerials as convenient?

4. What is the difference between a counterpoise as it might be used under an amateur short-
wave aerial, and a ground-plane?

5. Why is the choice of frequency important to establishing radio contact on short-waves?

6. Why are vertical λ/4—-type aerials useful in long-distance contacts?

7. Why do tall buildings or other radio-opaque features disrupt communications at vhf but not
on short-wave bands?

>>>>>>>>>>>>>>>>>>>> PAGE 68 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 12
2.6 RECEIVERS

2.6.1 The Basics of Radio Receivers

** A receiver has 10 basic requirements:

1. To accept from an aerial/feeder system the energy which has been extracted from passing e-m
waves and select from the mass of signals the one that is required.

2. To maximise the wanted signal and to reject as much as possible the remaining signals and
extraneous noise.

3. To add to the required signal as little noise of its own creation as possible.

4. To amplify that signal to a level at which it can be processed without degradation.

5. To extract from the signal the information which it carries as some form of modulation.
6. To amplify the information signal to a specified power level so that it may drive the requisite
termination gear: e.g. loudspeaker, printer, servo-motor.

7. To provide automatic and/or manual control of signal levels at all stages of the receiver.

8. To remain "tuned" to the frequency of the required signal without constant attention.

9. Not to cause radiation of signals via the aerial system to which it is connected.

10. To remain unaffected in its operation by powerful signals which may, for example, emanate
from nearby transmitting equipment.

** Although a receiver operates as an entire system it can be broken down into a number of
distinct units as illustrated in the block diagram of Fig. 32.

(a )**Aerial Attenuator

The input circuits of receivers, and especially those utilising semiconductors, are easily
overloaded by strong signals. When this happens the input circuits lose their ability to select a
wanted signal — the receiver becomes untunable. This effect can be caused either by the
wanted signal or by an unwanted signal against which the selective circuits should discriminate.

The aerial attenuator is usually a set of resistive pads which can be switched-in by the
Operator so as to reduce the signal level which reaches the receiver input. In some sophisticated
receivers this process is automatically controlled.

(b)**Aeria1-coupling and Tuning Circuits

In this section receivers fall roughly into two categories:

(i) Domestic receivers intended to be used with reasonably-strong signals (that will swamp any
ambient noise) and with any kind of aerial including a simple length of wire;

>>>>>>>>>>>>>>>>>>> PAGE 69 <<<<<<<<<<<<<<<<<<<<

in these the input-tuning (selective) circuits also perform the task of matching and tuning the
aerial. Mostly such a description is but a euphemism; the best that can be achieved is to "hold
off" the aerial so that it does not destroy the performance of the tuned circuits.

** Ferrite—rod aerials, in which a rod of magnetic material is used to concentrate the e-m fields,
can largely overcome the problem.
** (ii) Communication Receivers

As the name implies these receivers are used in communication channels and the prime
requirement is reliability in performance and the ability to extract weak
signals (sensitivity). Thus they are intended for use with properly-engineered aerial systems.

The aerial-tuning circuits therefore are required only to select the required frequency; they
are presented with a feeder that requires a resistive termination usually either 75 or 50 ohms.

(c)** Radio-frequency (rf) Amplifier

This part of a receiver, plus the input-tuning circuits, is often referred to as the front end. Its
purpose is to raise the selected signal to a higher level where it can be processed without being
degraded by the addition of noise.

The rf-amplifier uses tuned circuits in place of resistive loads and so provides further
selectivity. Noise generated in the receiver is only significant at the input to the rf amplifier
where the incoming signal is at it weakest; it is here that the noise performance of a receiver is
established.

To cope with the very-wide range of signals incoming from the aerial the gain of the rf-
amp is made adjustable; this control is turned down when receiving strong signals thus
suppressing the noise (and weak signals!) heard at the output of the receiver. Usually such gain
adjustment is incorporated in an automatic gain-control (a.g.c.) circuit but, in communication
receivers, it is not uncommon for an additional manual control to be fitted together with a
facility for disconnecting the a.g.c. A.G.C. has a primary function in combating fading.

>>>>>>>>>>>>>>>>>>> PAGE 70 <<<<<<<<<<<<<<<<<<<<

(d) ** Demodulator

In early receivers this section was known as the detector stage and its function is to extract
the information content from the incoming signal. Now, with the use of a.m., f.m. and ssb in a
single receiver, the more comprehensive term is used. Details are given in later Sections.
 The demodulator stage is also used to derive a dc signal whose amplitude is proportional to
the signal strength incoming to the demodulator. This is used to operate the a.g.c. circuit; it is
a negative-feedback arrangement that tends to maintain the signal level at a constant value at the
point of demodulation (i.e. the point at which the agc-control signal was extracted).

For c.w. working (Morse code) a somewhat different method of "demodulation" is required
in the form of a local oscillator (L.O.), known as a beat-frequency oscillator (B.F.O.) that can be set
to beat with the incoming carrier thus forming an audible signal at the receiver output. The
frequency of this audible tone can be adjusted by the Operator through use of a L.O. or BFO
control which affords a fine adjustment of audio frequency.

An LO On/off switch must be provided also else this oscillator will render the receiver
useless for other types of reception. However, for those receivers not equipped to demodulate
ssb signals, it is possible to use the L.O. as a carrier-injection oscillator; this requires
considerable skill on the part of the Operator however as, apart from accurate adjustment of the
L.O. frequency for either upper or lower sideband as appropriate, it is necessary to achieve
correct balance between the levels of ssb and L.O. signals. A manual rf-gain is a great help in
such activities and, of course, it is preferable to switch-off the a.g.c.

Reasonable frequency stability is desirable from the L.O. if the output-tone is not to require
constant re-adjustment; this is particularly important where narrow-band audio filters are in use
to reduce noise. Also to use the L.O. successfully for ssb reception it is essential that it be very
stable in frequency.

(e)** Audio Amplifier

The general requirement here is to accept the low-power audio signal delivered by the
demodulator and raise it to a power level suitable to drive the final equipment. This might be a
loudspeaker, a pair of headphones, a recording milliameter, a servo-system as in model-control
or any combination of a variety of purposes.

In domestic receivers the requirement is quality which means a full frequency bandwidth,
low distortion levels, low hum and other noise levels and sufficient power to drive a
loudspeaker or perhaps more than one loudspeaker.

In communication receivers the requirements are almost the opposite. To reduce noise
levels bandwidth is restricted to "telephone quality" - i.e. roughly 300 Hz to 3.5 kHz. For c.w.
working the audio bandwidth may be restricted, at the Operator's discretion, to as little as 100
Hz provided that the requisite frequency-stability is available. A certain amount of distortion is
acceptable if it is necessary to enable information to be extracted from weak signals.

Output-power requirements are less for a communications receiver. Usually they are fitted
with a switched output, suitable to drive a loudspeaker, and one or more outputs to drive
headphones.

>>>>>>>>>>>>>>>>>>>> PAGE 71 <<<<<<<<<<<<<<<<<<<<

** Manual control of audio gain is usual but communications receivers are built in which the
a.g.c. is derived from the audio circuits thus ensuring that the audio level remains constant.
These receivers also have circuits to restrict the audio bandwidth as already mentioned. Such
circuits may be audio filter-networks incorporated in the audio amplifier stages or in the feed to
the loudspeaker; alternatively they may be incorporated in the rf-amplifier circuits where they
restrict the bandwidth and so eliminate the appropriate side-frequencies.

(f) ** A Power Supplier is necessary where a receiver is to operate from the ac-mains supply
(see Part 5 of this Course). Receivers are powered from batteries usually for portable operation
but battery-operation is also used where reception/measurement of very-weak signals requires a
low noise level. The most common requirement for portability is in equipment intended for
mobile use in vehicles; the normal 12-volt battery system is utilised.

Many equipments are fitted with dual mains/battery facilities.

2.6.2 The T.R.F. Receiver

** The simplest of all receivers is the tuned radio-frequency receiver which is restricted to a.m.
working. This is a straightforward receiver which follows almost exactly the outline given in
the previous Section. The price paid for the simplicity is a lack of selectivity which means that it
can be difficult to separate a wanted signal from a nearby powerful signal.

** The TRF receiver does have advantages however. Within the primary service area of a
domestic broadcast transmitter (where a strong signal is guaranteed) the lack of selectivity
means that a large range of side-frequencies is admitted which provides very good performance
in terms of audio bandwidth. The receiver is also immune to the self-generated whistles which
are a feature of the more complex circuits described below.

The simplest TRF receiver is the one-valve Leaky-grid Detector (or simply grid-detector). This
can be built using a transistor but its operation is easiest to follow in the valve form. It is
illustrated in Fig.33.
A valve responds to the
voltage which is impressed
between its grid (dotted-line)
and cathode. As long as the grid
remains negative with respect
to the cathode it cannot draw
current from the heated
cathode. It is usual to add a dc
negative bias voltage to any
signal to ensure that the
positive swings of the signal do
not take the grid into
conduction. In the grid
detector this bias is omitted and
so the grid is driven into
conduction by the positive
swings of the modulated
carrier. The grid current
charges the grid capacitor and
so generates a bias voltage
which at all times is
 proportional to the carrier
amplitude.

>>>>>>>>>>>>>>>>>>> PAGE 72 <<<<<<<<<<<<<<<<<<<<

Finally the current which flows from cathode to anode is determined by the grid voltage.
Headphones connected as shown respond to the audio component of the anode current as it
varies in sympathy with the varying grid-cathode voltage.

Allowing the grid to conduct reduces the input resistance between grid and cathode and this
loads the tuned aerial circuit; in turn this reduces the circuit Q-value and so destroys both its
selectivity and its magnification.

The circuit Q can be restored and/or improved by using positive-feedback; the anode
current contains a component at the carrier frequency and this is fed back in-phase to reinforce
the current which circulates in the aerial tuned circuit. The arrangement, known
as regeneration or as reaction makes

Its disadvantages however are serious:

(a) If the Reaction control is advanced too far the p.f.b. becomes excessive and the circuit bursts
into self-oscillation at the resonance frequency to which the aerial circuit is tuned. Efficient
transfer between aerial and tuned-circuit is a necessary part of the receiver action and so, when
oscillation commences, the circuit turns into a transmitter — to the annoyance of everyone who
is trying to operate around that frequency. -

(b) The receiver is at its most sensitive just before the onset of oscillation when the circuit Q is
at its greatest usable value. However the high Q-value means that the bandwidth is at its
narrowest.

(c) It has been shown under 1.13.2, 1.14.3 (Lessons 9+10 in Fundamentals—i) that the Q,
impedance and magnification of a tuned circuit vary as the circuit tuning is adjusted. Thus this
simple receiver either becomes insensitive or oscillates when any attempt is made to re-adjust
the tuning.

** The next logical step in developing the TRF receiver is to isolate the tuned circuit from the
aerial so that, should it oscillate, it does not create a nuisance. This requires the addition of an
RF Amplifier; this circuit accepts the input from the aerial, selects the required signal, amplifies
it and then passes it to the tuned circuit that feeds the grid detector. If properly designed the rf-
amp will not pass signals from its anode circuit to its grid circuit and so the aerial is
protected (or buffered) from the detector stage.

** To drive a loudspeaker rather than the headphones an audio power-amplifier stage can be
added thus producing the 3-stage TRF receiver shown in block form in Flg.34. With such an
arrangement it is possible (and permissible) to adjust the detector for a weak oscillation to
enable reception of cw signals.
 >>>>>>>>>>>>>>>>>>>> PAGE 73 <<<<<<<<<<<<<<<<<<<<

** TRF receivers have been built using more than one rf-amp and often an audio-amp is added
partly to increase the overall sensitivity but mainly to provide the necessary signal level to drive
the power amplifier. Placing several rf-amps in cascade leads to a circuit arrangement wherein
amplifiers have tuned circuits at both their inputs and outputs ; such a circuit is described
in Part 4: Electronic Oscillators as the tuned-anode-tuned-grid (T.A.T.G.) oscillator.

** The problem of oscillation in the multistage rf-amp is further complicated by the need to
tune each of the resonant circuits simultaneously if the receiver is to operate over a band of
frequencies. See also under 3.13.13: Decoupling .

** Most of the problems with TRF receivers stem from the need to tune them over a band of
frequencies but it is a problem with all receivers that bandwidth, selectivity and sensitivity
change with the tuning.

** Clearly the difficulties can be designed away for a receiver that is to operate at a fixed
frequency. All modern receivers endeavour to get the best of both worlds by using a receiver in
which the main body operates at one frequency only in combination with a front-end that can be
tuned over any desired band but whose output is always at the frequency of the fixed receiver.
(See Superhet receivers in Lesson 13.)

END OF LESSON 12

* * * * * * * * * * * *

QUESTIONS

1. Why are some communications receivers fitted with an aerial attenuator?

2. Why do most domestic receivers not employ an rf-amplifier?

3. What tasks does the Demodulator perform?

4. In what ways do domestic and communications receivers differ?

5. If a receiver is unable to resolve a weak signal when other receivers can cope where would
you suspect the trouble to lie?
6. State briefly the problems encountered with a TRF receiver and outline the steps proposed for
their solution,

7. A capacitance of 0.1 µF is connected in series with one of 0.033 pF. What is the effective
capacitance of the combination?

8. The above series combination of 0.1 and 0.033 µF is connected across a choke of 1 H. What
is the resonance frequency of the resulting circuit?

9. How would you express the bandwidth of a circuit which passes frequencies from 460.5 kHz
to 469.5 kHz ?

>>>>>>>>>>>>>>>>>>>> PAGE 74 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 13
2.7 THE SUPERHETERODYNE RECEIVER

2.7.1 General

The previous Lesson ended with the observation that many of the ailments of a simple
receiver can be overcome with fixed-frequency working. The superhet receiver uses this
principle together with a tuneable front-end.

The name is a contraction of supersonic heterodyne which means that a supersonic (above
audible range) beat-note is obtained by heterodyning (beating) the incoming carrier with another
signal. The carrier-frequency of each required incoming signal is translated to a common carrier-
frequency . (See Lesson-3, Funds-2.)

The bulk of the receiver, that which sets the gain and bandwidth of the overall system, is
constructed as a fixed-frequency unit. The front end contains a variable oscillator (the local
oscillator) which is tuned so that its frequency always differs from that of the desired incoming
carrier by a constant amount. The output from the front end thus always contains a component
— a beat note — with a frequency that is constant whatever carrier the front end is set to
receive.

In domestic receivers the front end seldom consists of more than the frequency-changer stage
together with the aerial-tuning coil and the variable local oscillator. Its output is usually in the
region of 465 kHz. Communications receivers invariably incorporate one or more rf-
amplifierstages before the frequency-changer stage and a variety of output frequencies are used
depending on the purpose of the receiver.

The fixed-frequency output from a frequency-changer is referred to as the intermediate

frequency or the I.F. because it is intermediate between the rf carrier-frequency and the final
audio frequencies.

The fixed-frequency part of a superhet receiver operates at the intermediate frequency and so
is usually known as the i.f. unit, i.f. amplifier, i.f. stages or simply as the i.f. It is too often
forgotten however that the i.f. is, in fact, the receiver proper. Fig. 35 shows the basic block
diagram of a typical superhet receiver.

>>>>>>>>>>>>>>>>>>>> PAGE 75 <<<<<<<<<<<<<<<<<<<<

2.7.2 The Frequency-changer

As shown in Lesson-3 of Part-2 frequency-translation requires that a carrier is mixed with

another signal of suitable frequency in a non-linear circuit. Such a non-linear circuit can be a
diode (either thermionic or semi-conductor) which is a two-terminal device that conducts in one
direction but not in the other. Although at uhf and higher frequencies there is often no
alternative to a small diode the technique is seldom found in domestic or communications
receivers because other methods provide a larger i.f. output. The ratio of if-signal output to rf-
signal input is known as the conversion gain.

It is usual to use a valve or transistor-device as the non-linear mixer and there are many such
devices and as many more circuit arrangements. As indicated in Lesson-3, frequency-changers
(or mixers) come in two basic forms namely additive mixers and multiplicative mixers. As a
guide the rf and L.O. signals are added to the same input of an additive mixer but to different
inputs of a multiplicative mixer.

A disadvantage of additive mixing is that, unless the added complication, and cost, of a
buffer stage is used after the local-oscillator, the L.O. has a tendency to "pull" into step with the
carrier signal; this tendency is resisted at the higher frequencies by raising the i.f. frequency
which increases the separation.

There is a problem fundamental to all mixer stages in that the local-oscillator tuning must be
kept at all times a fixed interval either above or below that of the required incoming carrier-
frequency. At a first glance this may appear a simple matter but unfortunately it is not so.

Consider a tuneable receiver that is to work over a band from 1 to 2 MHz with an
intermediate frequency of 100 kHz. The local oscillator might be either above or below the
signal frequency and so we have:
Signal Range: 1 to 2 MHz or 1.5 MHz ± 0.5 MHz

Local Oscillator high 1.1 MHz to 2.1 MHz or 1.6 MHz ± 0.5 MHz

Local Oscillator low 0.9 MHz to 1.9 MHz or 1.4 MHz ± 0.5 MHz

which seems straightforward enough ?

However the tuning range of a variable capacitor is from its minimum to its maximum
capacitance irrespective of the centre frequency which is chosen by selecting the inductance.
The parameter of interest here is the amount that the frequency has to change as the capacitor is
moved between minimum and maximum settings. In the Table above it is not the range of ± 0.5
MHz that matters but the comparison between that figure and the centre frequency. Thus:

at 1.5 MHz a range of ± 0.5 MHz represents a change of ± 33.33%

at 1.6 MHz a range of ± 0.5 MHz represents a change of ± 31.25%

at 1.4 MHz a range of ± 0.5 MHz represents a change of ± 35.7%

This shows that, with the local oscillator operating above the frequency range of the
incoming signal, it must tune over a lesser range while, with the local oscillator operating below
the range of the incoming signal, it must tune over a greater range.

>>>>>>>>>>>>>>>>>>>> P76 <<<<<<<<<<<<<<<<<<<<

Tuning several resonant circuits simultaneously with the one knob is achieved by use of
"ganged" variable capacitors; i.e. several of these units mounted on a single rotating shaft. The
immediate requirement for a superhet receiver therefore is a multi-gang capacitor in which one
section has either a reduced or an enhanced range of capacitance.

Indeed this solution has been used by many manufacturers but it is a costly solution and one
that is still subject to pitfalls; e.g. the actual range achieved by a variable capacitor is determined
by the stray capacitances which shunt it and so, to use a specially-made component, it becomes
necessary to control the strays.

Inevitably some kind of trimming is required to bring all the resonant circuits into tune with
each other and the usual solution adopted is to use a ganged capacitor in which all the units are
identical and to "trim" the range electrically.

One method of reducing the range is simply to place a fixed capacitor across the variable
thus reducing its relative range. However this involves an increase in the total capacitance and
so the value of the inductance must be changed. The end result of such a simple arrangement is
lack of tracking the accepted term for keeping the rf-circuits and the L.O. circuit tuned in. step.
 The usual technique is illustrated
in Fig.36. A small variable trimmer capacitor is
connected in parallel with the L.O. section of
the ganged capacitor and a second much-larger
variable padder capacitor is connected in series.

At high-frequency settings, where the main

capacitor is set to its minimum value, the
trimmer increases the minimum
capacitance and so restricts the upper
frequency-limit; the (relatively-large) series-
connected padder makes very little difference to
the combined capacitance. At the low-
frequency settings, where the main capacitor is
set to its maximum value, the series-connected
padder reduces the

maximum capacitance and so restricts the lower frequency-limit; the (relatively-small) parallel-
connected trimmer makes very little difference to the combined capacitance.

** There are two important points which are likely to be

(i) because this technique restricts the tuning range at both ends it is necessary to set the L.O. to
work above the incoming carrier frequency.

(ii) because the trimmer and padder perform their tasks under different conditions they make it
possible to tailor the relation between capacitor-setting and the resulting frequency.

Although the trick can be made to work very well it nevertheless is not a perfect solution.
The actual received frequency which is passed to the i.f. unit is determined by the L.O. setting;
the failure to achieve perfect tracking is not important provided that, as a result, the rf-signal
circuits are not required to operate outside their bandwidths. This is a matter of design
compromise.

>>>>>>>>>>>>>>>>>>>> PAGE 77 <<<<<<<<<<<<<<<<<<<<

The tracking is usually adjusted at points roughly 20% from each end of the range after which it
drifts back and forth as the tuning is varied across the band.

2.7.3 Problems Inherent in Frequency-changing

(a) Image or Second-channel Interference

As given in the previous Section the frequency of the local oscillator in a superhet receiver is
always higher than the required signal frequency by an amount equal to the required i.f.
frequency. However there is another frequency, higher than the L.0. signal by the same amount,
that is able to beat with the L.0. to produce the designed i.f.; this frequency is separated from
the required frequency by twice the intermediate frequency and is known as the image frequency.
 Should a signal be present on such a frequency then, provided it is strong enough to
overcome the discrimination of the r.f. stages, it too will send an if-signal to the I.F. section of
the receiver and the two signals cannot be separated thereafter. The interference which this
causes is known as second-channel interference or sometimes as image interference.

The only way to eliminate this nuisance is to prevent signals on the image channel reaching
the frequency-changer. This is one of the reasons (a major reason) for placing one or more rf-
stages in front of the mixer stage. But once again the solution is not so simple as it may seem.

An amateur receiver operating on 14 MHz with an I,F. of 100 kHz may have an interfering
station putting in a powerful signal on its image channel at 14.2 MHz; i.e. 200-kHz higher. This
is a very good separation as far as the likely bandwidth of either signal is concerned. However,
in terms of asking the rf-amplifiers to eliminate 2nd-channel interference, the separation is 200-
kHz in 14-MHz which is about 1.4% -a very tall order especially for a variable-tuned circuit.

The answer here is to raise the intermediate frequency. An i.f. of 1-MHz would put the
image channel at 16-MHz and increase the separation for the rf-amps to 2 MHz in 14 MHz or
14%.

(b) Adjacent-channel Interference

As the name implies this is interference caused by stations on frequencies so close to the
required frequency that the tuned circuits of the receiver cannot separate them. Narrow-
bandwidth circuits are not very difficult to design but to make them narrow and also tuneable is
all but impossible within reasonable cost. It is here that the superhet principle really shines.

The I.F. stages of the receiver operate at a fixed frequency and so the design parameters for
this part of the equipment can be chosen for optimum performance. To get the steep edges (or
skirts) to the passband which is required if adjacent-channel signals are to be eliminated requires
use of the bandpass circuits which were introduced in Fundamentals-1 Lesson-10.

Once again the separation of two signals depends on there being a significant difference in
their frequencies as a percentage of their centre-frequency. It follows therefore that, to remove
signals that are separated from the wanted carrier by 3-kHz (or even less), requires the i.f. to be
kept low and a value of 100 kHz is not unusual.

>>>>>>>>>>>>>>>>>>>> PAGE 78 <<<<<<<<<<<<<<<<<<<<

Section 2.7.3 shows that the successful construction of a superhet receiver has two
conflicting requirements; a high i.f. to eliminate image-channel problems and a low i.f. to
eliminate adjacent-channel problems. Fortunately the superhet principle itself provides an ideal
answer.
 As shown in Fig.37 the double-superhet performs the frequency-changing act twice and uses
two intermediate-frequencies in two cascaded if-amplifiers (they follow each other). The first
i.f. is set as high as practicable so as to remove the image-channel from consideration. The
second i.f., now protected from image interference, is made as low as practicable so that its
bandwidth can be made very small to eliminate adjacent-channels.

For technical reasons beyond the scope of these Lessons the principle is sometimes extended
further to a triple-superhet. Note however that the Racal receivers, which use the Wadley Loop
and which are often described as triple- superhets, are in fact double-superhets; these receivers
are much misunderstood. (See later in this Section.)

The frequency-changer stage is often referred to as the first detector which reflects its non-
linear task in extracting the intermediate frequency. In a double-superhet the second frequency-
changer stage becomes the second-detector. The final detector stage, that which extracts the
modulating signal, is known as either the second-detector or third-detector as appropriate.

In the double superhet receiver the second local-oscillator operates in conjunction with the
fixed frequency of the first-i.f. and consequently it does not need to be tuneable. It is usually a
crystal-controlled oscillator.

2.7.5 Other Problems

Problems with superhet receivers are unique to the technique and consist mainly of a series
of whistles hums and noises which, very often, are extremely difficult to trace and eliminate.

>>>>>>>>>>>>>>>>>>>> PAGE 79 <<<<<<<<<<<<<<<<<<<<

A good starting point is to note that any signal with a frequency within the i.f. passband,
which reaches the input to the i.f. unit, will appear at the receiver output. Such a spoof i.f.
signal can be derived from:

(a) image-channel and adjacent-channel signals as described above

(b) two signals beating together whose frequency-separation is equal to the i.f. frequency or
where they produce harmonics with that separation; this problem occurs mainly with additive
mixers.

(c) any signal whose carrier-frequency is separated from a local-oscillator harmonic by an

amount equal to the i.f. frequency

(d) any harmonic of a carrier-signal which is separated from the local-oscillator frequency by an
amount equal to the i.f. frequency

(e) by combining c and d above wherein a beat at if-frequency is set up between a harmonic of a
carrier-signal and a harmonic of the local-oscillator

(f) a harmonic of the i.f-signal which manages to arrive back in the rf-circuits and beats with a
harmonic of the local-oscillator.

(g) an rf-carrier at or close to the i.f. frequency which is permitted to reach the input of the
frequency-changer.

That is enough to demonstrate the facility with which a superhet receiver can be difficult.
The problem of whistles is greatly exacerbated by the fact that the strength of the resulting beat
note is dependent on the power of EITHER of the carriers involved. Thus only one of the inter-
acting signals needs to be strong.

The sources of the harmonics mentioned above are not difficult to identify although, in
practice, they can be very difficult to trace; they can be generated in any non-linear circuit
within the receiver. This can result from a straightforward overload by a too powerful signal or
because a powerful signal has developed a large a.g.c. bias which is over-enthusiastic in
reducing the receiver gain. Hence the need for an aerial attenuator. Many problems are caused
by insufficient isolation between the aerial circuits and the i.f. amplifier and by unwanted
signals invading the local oscillator or escaping from the local oscillator.

2.7.6 Amateur Receivers

In the discussion above the point was made that the local oscillator in the frequency-changer
stage operates of necessity at a frequency above that of the incoming signal. This point is
stressed in the R.A.E.

It is worth noting however that there are two circumstances in which the rule may be
waived:

(i) In a double superhet the second L.O. can be placed either above or below the first-i.f.
frequency because the L.O, operates at a fixed frequency and so tracking problems do not arise.

(ii) The rule holds for all-band (General Coverage) receivers but not for receivers which cover
only limited bands. Amateur receivers for example, especially those which are incorporated
into Transceivers, cover only the frequency ranges permitted by the amateur- licence;

>>>>>>>>>>>>>>>>>>>> PAGE 80 <<<<<<<<<<<<<<<<<<<<

the greatest relative bandwidth here is that from 3.5 MHz to 3.8 MHz (about 8 %). Over such
a limited range a loss of tracking would not be important and so the L.O. could be placed on
either side of the incoming carrier as dictated by other considerations. This can be beneficial
where the first i.f. is set very high; for example, should it be decided to place the 1st-if at 10.7
MHz then, to cover the 3.5-MHz Band, the L.O. must cover from 14.2 MHz to 14.5 MHz or
from 7.2 MHz to 6.9 MHz. It is much easier to deal with an oscillator operating on 7 MHz than
with one operating on 14 MHz.

Modern "black-box" transceivers are fitted with an IF-Shift control. This can be very useful
in cw working to eliminate a powerful adjacent-channel signal by pushing it "over-the-edge" of
the if bandpass characteristic. The BFO is set to whichever side gives the unwanted signal the
higher pitch and then the IF is shifted until the unwanted signal is attenuated. In fact, to make a
bandpass if amplifier with a tuneable characteristic is very expensive and most unlikely to be
satisfactory as a long-term strategy. With a double superhet however all that is necessary is to
drift the frequency of the (nominally-fixed frequency) 2nd L.O.; in other words it is not the if-
amp which is shifted relative to the if carrier but the carrier which is shifted relative to the if-
amplifier.

This technique is most successful in double superhet receivers because of the relatively-
narrow bandwidth of the 2nd-if amplifier.

END OF LESSON 13

************

QUESTIONS

1. What is the basic difficulty which is overcome by the complication of using the
superheterodyne principle?

2. What is the basic difficulty which justifies the use of a double-superheterodyne design?

3. Why do you not find tuneable superhet receivers in which the local-oscillator output is set
below the carrier frequencies?

4. What design features would you expect to find in an i.f, unit?

5. In a double-superhet receiver where would you expect to find the beat-frequency oscillator for
c.w. reception?

6. For a good technical reason a receiver is to have an i.f. of 10.7 MHz but it is required to tune
over the continuous range 10 to 12 MHz. Clearly, at the lower end of this working range, the rf-
input will gain direct access to the frequency-changer input. In broad terms only, can you
suggest a technique for preventing trouble?

7. Explain the terms frequency-translation and frequency-multiplication.

>>>>>>>>>>>>>>>>>>>> PAGE 81 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 LESSON 14
2.7.7 RF Amplifiers

** An rf-amplifier provides gain to raise signal-level from the very small value delivered by
an aerial to one more suitable to drive the following stage be it a frequency-changer or a
demodulator. However, gain of itself is not the main reason for employing one or more rf
stages; low signal-level can often be overcome by other arrangements such as the leaky-grid
detector described above.

** The real benefit of rf-amplification is that each stage introduces an extra tuned circuit and
so increases the selectivity. As described in the previous Lesson good selectivity before the
frequency-changer is necessary to eliminate 2nd-channel interference.

** Of course, any form of enhanced selectivity will also provide a measure of protection from
adjacent-channel signals but such is not part of any design intention because the rf-circuits must
be tuneable and so they will exhibit variable bandwidth.

** A typical rf-amplifier is shown in block

form in Fig. 38. The input, via the coupling
capacitor, may come from an aerial-coupling
coil or from the output circuit of a previous rf-
amplifier. The output load, through which the
amplifier drives its output current, is a tuned-
circuit. This presents its highest impedance at
the frequency to which it is resonated and so
develops the maximum output voltage at that
frequency. In this way an rf-amplifier
discriminates against unwanted carriers.

** A well-designed and constructed amplifier

allows the passage of rf signals from its input to
its output (amplifying them on the way) but
does not transmit such signals in the opposite
direction. An rf- amplifier thus prevents output
from the local-oscillator reaching the aerial.

An important function of the first rf-

amplifier (the very first stage of a receiver) is
that it controls the noise performance of the
receiver. incoming from the aerial the ratio of
signal-amplitude to background noise can be
adjusted only in the design of the aerial
system; it cannot be altered in the following
receiver. However, it is a fundamental concept
that any receiver must inevitably add its own
quota of noise and so it unavoidably degrades
the incoming signal-to-noise ratio.
 The point most vulnerable to noise in a
receiver is the aerial input where the signal is at
its lowest value; from there on the receiver
amplifies equally both signal and noise. Thus it
is important to design the first stage so that it
adds as little noise as possible

It is normal to make the input impedance of a receiver 75-ohms so that it matches the low-
power coaxial cables usually used to connect a receiving aerial.

>>>>>>>>>>>>>>>>>>>> PAGE 82 <<<<<<<<<<<<<<<<<<<<

** Where the receiver is part of a combined transmitter/receiver equipment (a transceiver) then

the receiver input-impedance is the same as the transmitter output-impedance which is normally
50-ohms. In Fundamentals- 1, Lesson 11 it was pointed out that there is advantage in making
the input-impedance of a receiver higher than that of the connecting feeder. Circuit details of rf-
amplifiers and aerial connections are given in Part-3 of this Course.

2.7.8 I.F. Amplifiers

** Essentially if-amplifiers and rf-amplifiers do not differ except that the if-amplifier operates
at a fixed frequency. The i.f. section of a receiver has two principal functions:

(a) it provides the major part of the receiver amplification

(b) it uses band-pass circuits to eliminate adjacent-channel interference while preserving the
balance of the side-frequencies.

The passband requirements depend on the use for which the receiver is intended. For
domestic "music" a.m. reception the passband ought to be adequate to accommodate the side-
frequencies which result from a full audio bandwidth; i.e. 50 Hz up to 15 kHz. However the
great demand for channels means limiting the channel separation to around 9 kHz and so the
bandwidth of the if-amps is usually around ± 7.5 kHz, (Remember that bandwidth is specified at
the 6-dB points where the signal strength has been reduced by one-half.).

** Communications receivers however are not concerned with faithful sound reproduction but
require that the signal should be read as easily as possible. The noise which a receiver both
picks-up and generates is directly proportional to the bandwidth and so the design of
communications receivers aims to restrict the bandwidth as much as possible.

** For phone operation audio bandwidth is restricted to "telephone" quality about 300 Hz to
3.5 kHz. For a full double-sideband a.m. signal this would require an i.f. band of frequencies
extending to 3.5 kHz either side of the carrier frequency. For single-sideband (ssb) operation,
the bandwidth required in the i.f. is just 3.5 kHz starting from the carrier frequency and
extending upward (for upper-sideband working) or downward (for lower-sideband working).

** For c.w. working however, where only the carrier frequency is involved, the bandwidth can
be drastically reduced and this eliminates much of the background noise together with most of
the adjacent-channel interference. There is a lower limit to i.f. bandwidth which is set by the
stability of the local oscillator(s). Drift in these causes drift in the frequency of the i.f. signal and
clearly this signal must remain within the passband of the i.f. amplifier.
 It is not unusual for communications receivers to offer a (selectable) bandwidth as low as
300 Hz but use of this setting depends not only on the stability of the oscillators concerned but
also on the skill of the Operator in setting-up the receiver. The front-end of Racal receivers
(mentioned earlier) make use of a circuit configuration known as the Wadley Loop which
eliminates the effects of oscillator drift; consequently these receivers offer an additional
bandwidth of 100 Hz to those with the operational skills to adjust the receiver.

** Variable bandwidth or, to be more accurate, switch-selectable bandwidth is usually

provided in the i.f. unit. There are two basic methods:

>>>>>>>>>>>>>>>>>>>> PAGE 83 <<<<<<<<<<<<<<<<<<<<

(a) set the i.f. bandwidth to at least the greatest value desired and switch in filter networks to
reduce it as required

(b) vary the bandwidth of the bandpass circuits by changing the damping (see Lesson 10 under
1.14.3); this cannot be done when using stagger-tuned circuits.

** The design and construction of variable bandwidth circuits is not dealt with in this Course.
The use of crystal filters may be touched on in the R.A.E. however. All that is required is that
bandpass-circuits can be constructed with very steep skirts and also with very narrow
bandwidths by using crystal (quartz) resonators in place of, or in conjunction with, the normal
i.f. tuned circuits. Some filters use mechanical resonators in place of the quartz crystals.
Normally such filters are bought ready made. Some early receivers used a single crystal together
with a phasing control which effectively re-balanced the circuit to compensate for the
introduction of the crystal.

Note that when using frequency modulation the bandwidth required is determined by the
deviation and the modulation index and not necessarily by the bandwidth of the modulating
signal; see Lesson-2 of this Circuit details of i.f. amplifiers are given in Part 3.

2.7.9 The Demodulator

** The function of the demodulator stage is to accept the modulated signal from the last i.f.
amplifier stage and to extract from it the modulated information. In a modern multi-purpose
receiver this stage may have to cope with double-sideband am signals (domestic broadcast
programmes), ssb, fm and cw.

(a) C.W. Signals

** C.W. transmissions use only a keyed carrier. To make this audible at the receiver output
the i.f. signal is heterodyned with a local oscillator (the Beat-frequency-oscillator). It is usual to
set this B.F.O. nominally to the centre i.f. frequency and to provide a control that shifts it either
up or down by about 5 kHz.

** The signal presented to the detector varies in amplitude at the difference (beat) frequency
(see Lesson 3 of this Part) which, after detection, is heard as an audible beat-note whose pitch
can be adjusted by the B.F.O. control.
 ** That the B.F.O. can be shifted either up or down in frequency is an important facility.
Where an interfering signal is on the same frequency as the desired signal it is not possible to
separate them and only the morse-operator's skill can save the day. More usually such an
interfering signal is not exactly on the same frequency and its separation can be increased by
shifting the B.F.0. to the "other side" of the wanted carrier. Separation is also aided by adjusting
the B.F.0. so that the desired signal produces the more insistent sound or by arranging that the
beat with the unwanted signal is outside the audio bandwidth. Selectable or adjustable audio
filters are a great help in such a situation.

(b) A.M. Signals

** A normal carrier is a continuous sine-waveform which makes equal excursions above and
below zero ; its average value over a period of several cycles is zero. When such a wave is
modulated by varying its amplitude the average value remains at zero as the crest values, both
positive-going and negative-going, are varied.

>>>>>>>>>>>>>>>>>>>> PAGE 84 <<<<<<<<<<<<<<<<<<<<

** Such a modulated wave is shown

in Fig.39(a) with the base (zero) line as the
average value. In diagram (b) the lower half of
the modulated waveform has been removed; as
shown the average value of the discontinuous
waveform which is left follows the modulation
waveform as indicated by the carrier envelope.

This form of demodulation requires

a rectifier or detector in the form of a diode
which permits current-flow in one direction but
not in the other. The direction of flow is a
matter of which way round the device is
connected into the circuit and that is decided by
the requirement either for a positive-going or a
negative-going output.

** The output from such a detector is a

varying direct-current which, as shown in
diagram (c), can be analysed into an ac audio
signal, an ac carrier signal and a dc bias. These
are separated, usually by use of a capacitor, and
the audio ac-signal is passed to the audio
amplifier; the residual carrier is filtered out.

** The value of the dc depends on the amplitude of the modulated carrier before the detection
process; this dc-signal is used to operate the automatic gain-control (a.g.c.) system which is
described later.

(c) SSB Signals

 ** As described in Fundamentals-1 a suppressed-carrier signal is without a carrier component
and the waveform of the combined side-frequencies constantly reverses phase. To recover an
undistorted modulation signal it is necessary to re-insert the carrier and this has to be done with
considerable precision both in frequency and in relative amplitude. In theory the reconstituted
signal is then processed as a double-sideband signal.

** A single-sideband signal must undergo essentially the same process but of course one set of
side-frequencies is absent. The process is usually carried-out in one stage using one of the forms
of ring-modulator or balanced modulator operated "in reverse". In most transceivers the same
circuit is pressed into operation both as a modulator while transmitting and as a demodulator
when receiving. Details of these circuits are described in Part 3 of this Course.

It is possible to demodulate an ssb signal using a communications receiver which is not

suitably equipped by setting-up the c.w. local-oscillator (the beat-frequency oscillator or B.F.O.)
as a carrier-insertion oscillator. This is not an easy option however; it requires good frequency
stability from the oscillator and a slow-motion drive to allow accurate adjustment of its
frequency. Additionally it is necessary to juggle the rf-gain and aerial-attenuator controls to
obtain the correct relative amplitudes of ssb signal and local oscillator. Of necessity the local-
oscillator must be adjusted to the correct side of the ssb signal, dependent on it being an upper-
sideband or lower-sideband signal, and the receiver must be set for a.m. reception.

>>>>>>>>>>>>>>>>>>>> PAGE 85 <<<<<<<<<<<<<<<<<<<<

(d) F.M. Signals

** The modulation component is extracted from a frequency-modulated signal by means of a

circuit called a discriminator. According to its method of operation it may be referred to as
a slope detector, ratio detector, Foster-Sealey discriminator and others.

** Essentially these circuits produce two i.f. outputs which are combined so that the overall
output signal is proportional to the frequency of an applied carrier. (We say that the circuit is
made frequency conscious). Thus, as the carrier frequency is varied by the modulation, the signal
is converted to amplitude + frequency modulation. The f.m. is then ignored and the modulation
detected as with an a.m. signal; these circuits are described in detail in Part 3 of this Course.

** It is usual to over-drive the active device which feeds the discriminator so that this stage
acts also as the Limiter. Most of the noise, both incoming and self-generated within the
receiver, is in the form of amplitude-modulation and the limiting action removes all amplitude
variations thus producing the extremely-quiet background which is a feature of f.m.
transmissions.

It is possible to receive f.m. transmissions with a communications receiver that is not fitted
with a discriminator. With the B.F.O. switched off and the a.m. mode selected the receiver is
detuned so as to bring the required signal on to the skirts of the i.f. bandpass circuits; this makes
the i.f. amplifier frequency-conscious so that the i.f. signal, which is presented to the detector,
changes amplitude with the variations in frequency.

Once again however the trick is not easy to achieve. The bandpass circuits have steep skirts
which may not facilitate either the full swing of the modulated carrier or drift in any of the
superheterodyne oscillators. This can be offset by selecting a wide bandwidth setting - with all
the disadvantages of wide bandwidth. The conversion of f.m. to a.m. is not likely to be very
linear and so considerable distortion is to be expected. Above all the a.m. receiver is not fitted
with a Limiter stage and so a considerable amount of noise is to be expected also. Possibly the
greatest limitation is the available accuracy in tuning so as to place the nominal carrier-
frequency in the centre of the i.f. skirt.

When attempting to receive f.m. signals in this manner the automatic gain-control (a.g.c.)
should be switched off. Much depends on the filtering in the a.g.c. line but, with the receiver
de-tuned, the signal at the detector stage is likely to be fairly weak, the r.f. and i.f. stages will be
operating around their full gain and any audio which gets into the a.g.c.

2.7.10 A.F. Amplifiers

The A.F. Amplifier performs three functions:

(i) Amplifies the low-level signal from the demodulator

(ii) Provides the power necessary where it is desired to operate with a loudspeaker

(iii) Provides an Audio Gain control which is used to set the post-detector gain as determined by
the noise output from the detector with weak signals

>>>>>>>>>>>>>>>>>>>> PAGE 86 <<<<<<<<<<<<<<<<<<<<

It is not generally realised that full-band audio amplifiers are difficult to design arid can be a
source of considerable trouble; as shown in Lesson-1O of Fundamentals-1 an audio amplifier
has by far the greatest relative bandwidth. However, as far as communications receivers are
concerned, the audio amplifier is restricted in bandwidth and in gain and it is not required to
produce "high-fidelity" sound.

** It is not uncommon nowadays to use audio filters although these are often built-in to a
separate loudspeaker unit. Use of a narrow bandpass filter can greatly reduce the post-detector
noise and those persistent adjacent-channel transmissions which often seem determined to have
their say. Alternatively a so-called "notch" filter, which can be tuned across the audio band, can
be used to attenuate such an unwanted signal. There is no reason of course why both should not
be used simultaneously.

Such audio filters were possible at one time only by use of heavy and bulky iron-cored
inductors. The advent of semiconductor operational amplifiers has made it possible to construct
compact and easily tunable active filters; these are described in Part 3 of this Course.

END OF LESSON 14

***********

QUESTIONS

1. In what manner do the if-amplifiers differ in domestic and communications receivers?

2. In cw-working, where side frequencies are generated only at the start and finish of each
Morse character, there is benefit in restricting the bandwidth of a receiver. What is the benefit
obtained, where can the bandwidth restriction take place and what limits the usable bandwidth?

3. (a) How is so-called variable bandwidth achieved?

(b) How are very narrow bandwidths achieved?

4. How does the recovery of the modulation from an f.m. carrier differ from other forms
normally found in a communications receiver?

5. What is most important in demodulating a s.s.b signal?

6. What is a PLL ?

7. While working cw interference is coming in from two separate stations which are sitting
nearly on top of your contact's signal. Detuning the receiver improves matters with one of them
but always seems to make matters worse with the other. Is there some other technique which
you could try?

8. Why should you not connect a dipole to a transmitter via a coaxial cable ? Many amateurs do
this and seem to get good results; why is this?

>>>>>>>>>>>>>>>>>>>> Page 87 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 15
2.7.11 Crystal Filters

Because i.f. amplifiers operate at a fixed frequency it is possible to use bandpass circuits in
their construction. However, to achieve narrow bandwidths with very steep skirts, it is
necessary to use several such i.f. stages in cascade. The number of stages required (ignoring
considerations of gain) can be reduced by making use of the high-Q properties of crystal
resonators. Many amateurs use old valve-operated receivers in which the single-crystal crystal-
gate circuit is often found.

(a) Single Crystal or Crystal-gate Circuit

Pig.40 shows an i.f. transformer with a centre-tapped secondary which supplies, with respect
to earth, two outputs of opposite phase. In some circuits it is not the inductor that is centre-
tapped but (effectively) the capacitor by using two variable capacitors as shown in diagram (b).
 Both outputs are connected to the following stage one via a crystal and the other via a small
variable capacitor. At a frequency well-removed from the resonance frequency of the crystal
(where the crystal is ineffective) a signal is injected and the variable capacitor (the phasing
capacitor) is adjusted so that its capacitance equals that of the crystal holder plus strays; a sharp
null in the output shows that the two now-equal and opposite signals cancel.

At the series-resonance frequency of the

crystal this circuit balance is disturbed by the
crystal's low impedance, cancellation fails and a
signal appears at the output. Under these
conditions the circuit produces a high-Q
passband similar to that of the crystal alone.

A useful variation is obtained by altering

the setting of the phasing-capacitor. Not only
does this upset the balance but it introduces a
small phase-error with the result that a point of
complete cancellation (a notch) appears on one
side of the passband as shown in Fig.41. This
notch is moved to the other side of the passband
when the phasing capacitor is shifted in the
opposite direction.

>>>>>>>>>>>>>>>>>>>> PAGE 88 <<<<<<<<<<<<<<<<<<<<

This notch can be very useful in removing an interfering adjacent-channel signal but there is
a penalty in that the selectivity is degraded on the non-notch side.

A similar effect can be obtained by leaving the phasing capacitor undisturbed and varying
the position of the tap. This is not practicable with a tapped inductor but can be arranged in Fig.
40(b) by varying the two capacitors differentially, i.e. by increasing one and decreasing the other
so that the overall value remains constant.

The selectivity of the crystal gate can be varied by adjusting the tuning of the if transformer
which drives it; the effect arises from the change in the dynamic
(b) Double Crystal Filter

The double-crystal circuit uses essentially the same arrangement but there is a crystal in each
output from the i.f. transformer. The two crystals have different resonance frequencies which are
separated by the required bandwidth. Each crystal behaves exactly as in the single-crystal circuit
but, in the frequency range between their resonances, the two outputs add and so produce a
bandpass effect.

The two crystals produce much steeper cut-off slopes with the result that out-of-band signals
suffer some 20-dB more attenuation (a factor of 10) than with the single-crystal circuit.

In the two-crystal arrangement of Fig.42 variation of the phasing capacitor produces a notch
on each side of the passband simultaneously. To vary the bandwidth it is necessary to switch
between filters.

2.8 OTHER RECEIVERS

2.8.1 Autodyne Reception

In superheterodyne reception an incoming carrier-signal is changed to a different frequency

by beating it with the output from a local oscillator and detecting the result. In beat reception
(or heterodyne reception) an unmodulated carrier is made audible by beating it against a local
oscillator

>>>>>>>>>>>>>>>>>>>> PAGE 89 <<<<<<<<<<<<<<<<<<<<

whose frequency is close to that of the carrier frequency and again detecting the
result. Autodyne reception is a form of beat reception in which the three functions of local-
oscillator, mixer and detector are combined in a single stage. In some ingenious one-valve
forms the resulting audio signal was reflexed (i.e. re-circulated) through the stage so that it acted
also as an audio amplifier.

2.8.2 Homodyne Reception

The homodyne (or synchrodyne) receiver takes the autodyne technique a step further in that
the local-oscillation takes place at the frequency of the incoming carrier. The output from the
circuit is the modulation waveform. The circuit can be viewed as a superheterodyne in which the
i.f. carrier-frequency is zero. Once again reflex techniques were used to amplify the audio in a
one-valve design.

To use this circuit in double-sideband reception however it is necessary to ensure correct

phase-relation between carrier and the self-oscillation of the receiver. Needless to say, in both
the above techniques, care is needed

2.8.3 D.C. Receivers

** The dc receiver is most unfortunately named in that the initials stand for Direct
Conversion. It is a natural development of the synchrodyne technique but, strictly speaking, it is
not an autodyne circuit because it does not combine all the necessary functions in one circuit.

** The dc-receiver uses one of the ring-modulator or balanced-mixer type of circuits in a

mode that can be regarded as a product detector (see later Lessons). In essence it reverses the
modulation process; instead of applying a carrier and modulation waveforms and obtaining a
modulated-carrier the dc-receiver applies a modulated carrier and an unmodulated carrier and
derives the modulation waveform.

** Usually the detector is preceded by a simple rf stage, often untuned (i.e. broad-band) and
followed by an audio filter and audio amplifier; these last two would most probably be
combined in an active filter. The result is a very simple but often effective receiver. When
incorporated into a transceiver the Transmit mode uses the demodulator in reverse as a
modulator.

The lack of pre-detector selectivity and the susceptibility of the diode ring to overload
renders the receiver useless in the presence of strong interfering signals in that it takes on the
role of a broadband amplifier

2.8.4 Super-regenerative Reception

This receiver technique is developed from the leaky-grid circuit with reaction that was
described in Lesson 12. As discussed the sensitivity of the detector increases as the reaction is
increased until self-oscillation sets in; maximum usable sensitivity occurs at that point.

>>>>>>>>>>>>>>>>>>>> PAGE 90 <<<<<<<<<<<<<<<<<<<<

The super-regen. circuit operates with the stage set for self-oscillation but the oscillation
is quenched periodically by an imposed waveform that momentarily destroys the circuit loop-
gain. The frequency of this quench waveform must be above the audio range (i.e. not less than
10 kHz) else it appears in the audio output as a high-pitched whistle. Equally the quenching
frequency must not be too high else insufficient time elapses between quench-cycles for the
stage to cease oscillation completely. Usually the range lies between 10 KHz and 100 kHz.

It is possible to arrange the circuit so that it squeggs (see later in Part 4: Electronic
Oscillators). In this mode oscillation repeatedly stops and re-starts and so the circuit becomes
self-quenching. A super-regenerative receiver offers poor selectivity and, without an incoming
carrier, a high background noise; it also provides very-high sensitivity and excellent a.g.c.
action. In general its performance all round improves as the carrier frequency is increased and it
is at its most useful above 30 MHz.

Before the advent of semiconductors it was a popular device for radio-controlled models
although its temperamental behaviour was responsible for spectacular failures. When a signal is
detected the quench frequency changes and there is a significant increase of current drawn from
the supply; this can be utilised to operate a relay or solenoid.

During WWII the circuit was developed to a high standard of reliability; if my memory is
not at fault it was the basis of the airborne IFF Transponder (Identification Friend or Foe) which
became necessary with the introduction of night-fighter aircraft. The self-quenching circuit can
be adjusted so that it remains dormant until an interrogating signal appears on the frequency to
which it is tuned; this triggers the quench circuit and the resulting bursts of oscillation radiate
from the aerial as a response.

2.9 AUTOMATIC GAIN CONTROL (A.G.C.)

** The main purpose of any form of gain control is to adjust the output level for the comfort
of the 0perator; normally this is the function of the AF Gain control. There is a secondary use in
that strong signals fed into a high-gain receiver are likely to cause overloading in the high-level
stages unless the gain is reduced in the early part of the receiver; this is the prime function of
any RF Gain control. Note too that reducing gain in the early stages reduces the background
noise level. It is unusual to provide a manual control of gain for an i.f. amplifier.

** The automatic control of receiver gain is normally applied only to rf and if amplifiers. Its
two-fold purpose is to enable a receiver to be tuned between strong and weak signals without the
need for constant gain adjustment and to reduce the unpleasant effects of signal fading.

** In Lesson-4 it was shown that one of the output components from an amplitude detector is
a dc which is proportional to the carrier amplitude; for strong signals this dc component is
greater than it is for weak signals and, when fading occurs, this component varies in step with
the changes in signal strength. This dc-component is used as the output sample in a negative
feedback-loop (see Lesson-5 in Fundamentals-1) whose function is to maintain a constant level
of signal at the detector.

** The dc-output from the detector (the a.g.c. line) is connected to (some of) the rf and if
stages, and often to the frequency-changers as well, as a bias which reduces the gain of these
stages thus causing the pre-detector gain to fall as the signal strength increases.

>>>>>>>>>>>>>>>>>>>> PAGE 91 <<<<<<<<<<<<<<<<<<<<

It is necessary to grade the amount of feedback applied to each stage and also to leave some
stages without a.g.c.; in the presence of a very strong signal the large a.g.c. bias which is
developed could disturb the dc-working conditions of the active devices (see Part 3) and so
cause serious distortion in those later stages where the signal has been amplified to large
amplitude.
 ** A simple a.g.c. system, as shown in Fig.43 has a distinct disadvantage in that a feedback
control-signal is always present and so even weak stations cause some reduction in gain. It
would be better if the a.g.c. loop were to wait until the incoming signal reached a level that was
approaching the maximum permitted at the detector and then began to restrict the gain so as to
avoid overload effects. Such a control loop is referred to as Delayed A.G.C. In simple terms the
dc component from the detector is inhibited by a bias voltage which keeps the a.g.c. line at
ground potential until the dc-component exceeds the bias level. One method of adjusting the rf-
gain is to vary this delay bias.

Fig.43 Block Diagrams of A.G.C. Systems

** The final if stage which drives the detector can be relieved of the necessity to supply a
large output for a.g.c. purposes by interposing a dc-amplifier between the detector and the a.g.c.
line. The extra gain introduced into the feedback loop makes the resulting Delayed Amplified
A.G.C. very effective.

** In general, fm receivers are not fitted with an a.g.c. system because they use a Limiter and
the full gain of the rf/if stages is utilised to ensure that this device is well and truly limiting.
Nevertheless, where high gain is provided to enable the reception of very weak signals, it could
be that a very-strong signal would swamp the limiter stage. Such receivers may tap the bias-
potential developed in the limiter and feed this back to some of the earlier stages as an
elementary a.g.c.

>>>>>>>>>>>>>>>>>>>> PAGE 92 <<<<<<<<<<<<<<<<<<<<

Any manual rf (or if) gain controls which fall within the a.g.c. feedback loop will be
rendered inoperative by the a.g.c. action; thus, where manual gain-control is provided as an
option, it is essential to provide an a.g.c. On/Off switch. This switch is also used in cw-working
where the receiver uses a diode-type detector because this type of detector requires a high-level
of b.f.o. output and this causes a large a.g.c. voltage which reduces the receiver sensitivity.
 ** In some receivers an a.g.c., signal may be derived from the audio output signal as a means
of ensuring that the audio remains constant; this a.g.c. must be derived before the AF-
gain control however if this control is to remain functional. Note that this technique maintains a
constant-level of audio and so cannot be used for broadcast working where there is a
requirement for receiving music.

[A large orchestra can produce a very-large dynamic range from very loud passages to very
quiet. One of the main reasons that broadcast or recorded programmes never sound exactly like
a "live" performance is that the dynamic range has to be limited because of problems with
recording media and with transmitters. Despite constant manipulation of the sound level to
maximise the effects of the musical dynamics a sudden crash of sound can overload a powerful
transmitter and cause its protection circuits to shut-down the equipment. The necessary
safeguards are provided by "Limiters" which utilise a form of a.g.c.]

** It is usual to find an A.G.C. Tine Constant (see Lesson 3, Funds.-1)) switch which, usually,
is incorporated into a 3-position switch with the a.g.c. on/off. This varies the capacitance
connected across the a.g.c. line so varying the time taken for the a.g.c. voltage either to rise to its
final value or to die away again. In general the a.g.c. should respond quickly but not die away
until it is certain that the signal has gone; usual attack and decay times are about 2 mS and 200-
300 mS (milliseconds).

** When working cw there is the possibility of the a.g.c. circuit "remembering" the signals that
have passed and adding them together so as to exaggerate the signal strength; with continuing
morse the receiver gain would slowly decrease. In mathematical language we say that the a.g.c.
circuit would integrate the morse symbols. This is avoided by decreasing the a.g.c. time-
constant; hence the "fast" a.g.c. setting is used for morse and the "slow" setting for other voice
working.

2.10 TUNING INDICATORS

2.10.1 Tuning Meter

** The potential on the a.g.c. line is a direct indication of the incoming signal strength and the
simplest form of tuning indicator therefore is a voltmeter which indicates this potential; tuning
is adjusted for maximum indication.

In practice however the a.g.c. line is usually at high impedance; i.e. it cannot supply
sufficient current to drive a meter without seriously affecting the a.g.c. voltage (see Internal
Resistance in Lesson-11 of Fundamentals-1). One solution is to supply a (current) amplifier
whose high input-impedance does not affect the a.g.c. voltage but whose low output impedance
can supply the meter. See Op-amps.

>>>>>>>>>>>>>>>>>>>> PAGE 93 <<<<<<<<<<<<<<<<<<<<

Where this solution is adopted, particularly in old-style valve receivers, it is most likely that the
blocking capacitor at the output of the detector is omitted so that both the audio and the dc-
component are applied to the audio amplifier; the meter is then connected in the output dc-
circuit of the stage.
 ** Communications receivers are usually fitted with meters and the scale is calibrated in S-
points. It is common practice in amateur circles to report the strengths of received signals in
these units. However, because of the enormous range of possible signal strengths the metering
is tailored to a logarithmic law which means that the meter indication is proportional not to
changes in the received signal strength but to the logarithm of those changes.

As shown in Part-7: Mathematics this should mean that the meter reads in decibels but
unfortunately there seems to be a lack of standardisation in exactly what it means. The S-meter
scale ranges from S1 to S9 and then it is usually calibrated in steps of "10-dB over S-9" which
implies that each S-point represents 10-dB. It is a bit difficult to arrange the a.g.c. characteristic
to be even approximately logarithmic but, most importantly, it is an expensive exercise. As a
result not only does the meaning of an S-point vary from receiver to receiver but it also varies
within each receiver; the change in signal level as represented by S1 to S2 is almost certainly
different to the change indicated by S7 to S8.

** For fm reception there is a different requirement for accurate tuning. An indication of

maximum is not sufficient because the if frequency-characteristic is flat-topped and, for an fm
discriminator, it is essential that the tuning be adjusted to the exact centre of the transfer-
characteristic (see Part 3 of this Course).

** The dc component produced in a discriminator can be either positive or negative depending

on the direction in which the tuning is displaced; at the correct tuning point the component is
zero. Hence the requirement is for a centre-zero meter which deflects in either direction
according to the error. In some arrangements a normal left-hand zero meter is used with a dc-
bias applied to make the meter read at centre scale when the tuning is correct; the dc-component
from the

2.10.2 Magic Eye Tuning Indicators

A meter which gives a quantitative indication of level is a desirable feature in a

communications receiver but, for domestic broadcast receivers, all that is required is some
indication of the correct tuning point. For this purpose cathode-ray valve indicators were often
used in which a gap in a fluorescent display varied in width with the a.g.c. voltage.

2.10.3 LED Arrays

** A modern tendency is to use an array of light-emitting diodes (L.E.D.'s) for this purpose.
The display may take one of two forms; either an extra diode is illuminated as the voltage rises
(a bar display) or the illumination moves up through the stack of diodes with all the rest dark
(a moving-dot display).

>>>>>>>>>>>>>>>>>>>> PAGE 94 <<<<<<<<<<<<<<<<<<<<

** This type of display requires that the a.g.c. voltage is processed to turn the continuously-
varying (analogue) signal into one which advances in discrete steps that can be designated by
numbers (a digital signal). This process is referred to as analogue-to-digital conversion (ADC).
(The reverse process is DAC.) It is possible to design and construct your own circuit with a
system on stepped biases but the whole process is nowadays supplied in one single and very
cheap chip known as a bargraph driver; by simply changing the pin connections the chip will
deliver either of the two displays given above.
 END OF LESSON 15

* * * * * * * * * * * *

Questions

1. What is a D.C. Receiver?

2. Why do we use a,g.c.? Why is it provided with an On/Off switch.

3. What is a b.f.o. and for what purposes is it used?

4. In what way does delayed-a.g.c. differ from the basic form?

5. (a) A valve amplifier with an output impedance of 12,000 ohms is driving a current of 5 mA
into a tuned circuit with an impedance at resonance of 2,000 ohms. What voltage will be
developed across the tuned circuit?
(b) the valve is replaced by a transistor amplifier with an output impedance of a few hundred
ohms. What difference would you expect
in performance?

6. What is the difference between a Master Oscillator, a Beat-frequency Oscillator and a

heterodyne or Local Oscillator? Can you offer an explanation of the
description Audio Oscillator?

END OF FUNDAMENTALS PART 2

Back to Top of Page

>>>>>>>>>>>>>>>>>>>> PAGE 95 <<<<<<<<<<<<<<<<<<<<

 PART 3: ELECTRONIC AMPLIFIERS
LESSON 1
3.1 GENERAL

In this Part of the Course the discussion moves on to the technical details of radio and
electronics circuits and by far the greater part of these are amplifiers. There are many types of
amplifiers and many ways of designing and constructing each of those types and so, before
delving into their internal workings, it is a good idea to define what each type is and its intended
purpose.

While reading the following paragraphs however do bear in mind that many specific
amplifiers fall into several different categories either simultaneously or because, without
particular alteration, they can perform several different tasks. Some of the differences may also
be little more than a matter of definition.

3.2 VOLTAGE, CURRENT and POWER AMPLIFIERS

An electrical signal can be measured by either of its two parameters namely voltage or
current; where both are present with reasonable amplitude then the circuit is handling power.

A voltage amplifier is one which increases the amplitude of the voltage component but takes
little, if any, account of the current; by reference to Ohm's Law such an amplifier operates at
high impedance.

A current amplifier is one which increases the amplitude of the current component but takes
little, if any, account of the voltage; by reference to Ohm's Law such an amplifier operates at
low impedance.

A power amplifier is required to supply a signal to some power-consuming device such as an

aerial or a loudspeaker and so it must be able to increase the amplitudes of both the voltage and
current components simultaneously. An amplifier which deals in power must itself dissipate
power (see Maximum Power-transfer Theorem end of Lesson 11 in Fundamentals-1) and so is
in danger of overheating.

(There is an unfortunate mix-up in terminology in that a constant-voltage amplifier (see

Lesson-11 in Fundamentals-1) may sometimes be referred to as a voltage source meaning that
that particular amplifier maintains an accurate waveform for the voltage component irrespective
of any current distortion. A constant-voltage amplifier is of course a voltage-amplifier also but it
is not necessarily true that one designed to amplify the voltage component maintains a constant-
voltage output. As explained in the Lesson on Internal Resistance it is a matter of the relation
between amplifier output-impedance and the Load.]

As a general rule - though not by any means a strict rule - valves operate at high impedances
and so amplify the voltage component while transistors operate at low impedances and so
amplify the current component. By various circuit tricks and the use of the different kinds of
feedback techniques it is possible to make valve current-amplifiers and transistor voltage-
amplifiers.

>>>>>>>>>>>>>>>>>>>> PAGE 2 <<<<<<<<<<<<<<<<<<<<

Power amplifiers can be constructed from either valves or transistors but valves are much
more rugged than transistors (or other semiconductor devices) and will survive considerable
abuses where semiconductors are often destroyed instantly by the heat generated.

3.3 DC and AC AMPLIFIERS

Both valves and transistors require that their various connections be held at specific direct
potentials and that they draw specific direct currents; these supplies are purely to maintain the
device in the condition required for it to function as the required amplifier. Often a change in dc-
conditions will convert the one device into several different classes of amplifier.

The vast majority of signals handled by electronic equipment have alternating waveforms
and these are superimposed on the direct potentials and currents; the required alternating output
signals are recovered by separating dc and ac through the agency of devices which will pass
only alternating currents ; e.g. capacitors and transformers.

However, requirements do arise to:

(a) amplify small changes in direct potentials or currents (for example in making measurements
such as temperature and pressure)

(b) amplify signals whose frequency range goes all the way down to zero-frequency - or dc.
Perhaps the most well-known of these is a video signal which is the total electronic signal that
describes a picture.

(Note: a vision signal is an rf-carrier which is modulated by a video signal: a television

signal is the combination of a vision signal with the corresponding sound-carrier modulated
with the sound-signal.)

The various ac components of a video signal describe the changes in relative brightness (the
picture details) along each line of which the picture is composed; a direct component describes
the amount of white light present apart from any detail information. This dc-component defines
the difference between a high-key (bright) picture and a low-key (dark) picture.

Clearly an amplifier which relies on capacitive or transformer couplings cannot pass on such
dc information and it is a purely ac-amplifier.

A dc-amplifier however is a little more subtle. As used in thermometry to amplify the

temperature-dependent output from a thermocouple the amplifier would amplify only dc and, in
particular, a slowly-varying dc; it would be described as having a narrow bandwidth (perhaps
only a few Hz starting from zero) because detailed information is not present.
 A video amplifier which handles a picture signal as it comes perhaps from a television
camera, may have a bandwidth that extends from zero (dc) up to 5 MHz or even greater. It
would be described as having a wide bandwidth.

Both the above amplifiers are described as dc-amplifiers not because they amplify direct
signals but because they use dc-couplings (direct-couplings) as distinct from capacitive or
transformer ac-couplings. The term means that each stage of amplification is connected directly
to the next so that the dc-conditions

>>>>>>>>>>>>>>>>>>>> PAGE 3 < <<<<<<<<<<<<<<<<<<<

are passed on from stage to stage as well as the ac information. Needless to say the technique
brings some problems particularly with valves which require fairly high-voltages.

The greatest problem with dc-amplifiers is drift mainly caused by temperature changes.
Reasonably small changes in dc potentials and currents do not affect the performance of an
amplifier as such but clearly, where the dc-potential is itself the information signal, then drift is
a matter for concern. A common solution to this problem is to chop the direct signal with a
square-wave (a gate-circuit is opened and closed regularly by a square-wave signal) thus turning
the direct signal into an alternating one whose amplitude represents the magnitude of the dc.
This ac-signal is then amplified free of drift problems in an ac-amplifier and finally turned back
into a direct signal. Such an amplifier is referred to as a chopper-amplifier.

A wide-band so-called dc-amplifier can be used to amplify ac signals. A good example of

this perhaps is in an oscilloscope. The spot on the screen of an oscilloscope is deflected by
applying a voltage waveform between the appropriate pins of the tube. Direct potentials are
required to position the spot but an ac-amplifier is required to raise the signal voltage to the
several hundred volts required to form the waveform picture. A dc-amp is often used for this
job; it acts as a straightforward amplifier as far as the input signal is concerned but, by juggling
the dc conditions within the amplifier, the position of the picture on the screen can be controlled.
There is a bonus in that, because the amplifier is dc-coupled, the oscilloscope can measure direct
as well as alternating waveforms.

3.4 NARROW-BAND and WIDE-BAND AMPLIFIERS

The definition of these two classes is obvious but deceptively so. As explained
under 1.14.2 in Lesson 10 of Funds.-1 there is a relation between bandwidth and the speed at
which a waveform changes. For example, when a cw-transmitter is keyed, the resulting Morse
characters have sharp edges in which the carrier-amplitude first rises and then falls abruptly.
Such edges generate high-order harmonics with considerable amplitude and so the signal would
spread interference throughout a large part of the radio spectrum. The introduction of a (key-
click) filter to remove these harmonics results in the amplitude-changes taking place more
slowly.

Thus a wide-band amplifier can handle signals which change amplitude rapidly while a
narrow-band amplifier can handle only those signals which change amplitude slowly.
 The difference is important in control systems where feedback loops abound. For example,
in an a.g.c. system the a.g.c. line is heavily filtered (or decoupled) and this produces a slow
response; this is useful in that it prevents the a.g.c. from arguing with the stages whose gain it
controls which process could produce an effect of fading where signal-fading is not occurring.

(A good example of the difference between wide-band and narrow-band systems is found in
the human body. The semi-circular canals of the inner ear are usually represented as organs of
balance; experiments show that they have a bandwidth of several octaves and so must have a
fast response. In fact the canals do not tell us that we are falling but how fast we are falling.
They constitute an accelerometer.

>>>>>>>>>>>>>>>>>>>> PAGE 4 <<<<<<<<<<<<<<<<<<<<

By contrast the bandwidth of the eye is a small fraction of an octave and so its response must
be slow. Your eye will certainly tell you that you are falling but the information arrives too late
to be of much use; the function of the eye is to tell you exactly where you are. This slow-speed
response is exploited hilariously by some boxers and wrestlers who, by slowly swaying their
body, cause an opponent's attack to miss them completely.

3.5 AMPLIFIERS WITH FEEDBACK and AMPLIFIERS WITHOUT FEEDBACK

** As described briefly in Fundamentals-1 the feedback technique takes a sample of the

output from a system and feeds it back to the input where it is mixed with the normal drive
signal. Positive-feedback loops are to be found in amplifier designs but the subject is not a
matter for this Course.

In a negative-feedback loop the feedback sample is mixed with the input signal in such a
way that it tends to cancel that input signal. One of its greatest assets is that it also tends to
cancel any distortion that is introduced by the amplifier. For this reason a n.f.b. audio-amplifier
tends to produce a "cleaner" sound than one without feedback but all too often the technique is
abused. It is a strict rule that n.f.b. can be used to make a good amplifier better but it cannot
be used to make a bad amplifier good. In fact, the application of n.f.b. to a poorly designed
amplifier can lead to disastrous results; you have every right to be suspicious when you see the
phrase " ... cleaned-up by the application of feedback..."

The discussion in Lesson 11 of Fundamentals-1 shows that the relation between a source and
a load determines the operation of electronic circuits. For example; in driving a bipolar-
transistor amplifier, where input-resistance changes with the signal waveform, it is essential to
preserve the current waveform. This requires that the source of the driving waveform should be
a constant-current device; the source impedance must greatly exceed the transistor input-
impedance. This is arranged by the judicious use of n.f.b. either on the transistor in question, on
the source circuit or on both circuits. (Note that the base current that flows to the transistor must
be controlled by the source-resistance and NOT by the varying base-input resistance.)

BUT, if an amplifier has a natural bandwidth which extends only to 5 kHz and this has been
"extended" to 7 kHz by the application of feedback, trouble is bound to ensue. The apparent
extension of the bandwidth has arisen because the n.f.b. has reduced the in-band gain; as the
gain falls away from 5 kHz to 7 kHz so too does the n.f.b. It follows therefore that, from 5 kHz
to 7 kHz, the output impedance too changes and the amplifier performance becomes
progressively inferior.

** Negative feedback is applied to amplifier systems to:

1. Reduce distortion

2. Stabilise the gain (ac feedback)

3. Stabilise the dc-conditions (dc-feedback)

4. Stabilise the circuit against changes of active device (valve or transistor) which may occur
through ageing, replacement following failure or changes in ambient temperature.

>>>>>>>>>>>>>>>>>>>> PAGE 5 <<<<<<<<<<<<<<<<<<<<

5. Stabilise the circuit against the ageing/temperature changes of other components such as
resistors, capacitors and diodes

6. Control the phase response

7. Adjust the input impedance

8. Adjust the output impedance

9. Adjust the frequency response by weighting the feedback loop; this term refers to placing
filters in the feedback loop to make its effects variable with frequency. (Note that this practice
affects also all the items above.) A low-pass filter placed in an amplifier will remove the high
frequencies; the same filter placed in the feedback loop removes the n.f.b. at high frequencies
and so increases the gain at high frequencies.

The fact that a single feedback loop operates simultaneously all nine of the above factors
indicates the complexity that can arise.

3.6 AMPLIFIERS FOR VERY-LOW, AUDIO, VIDEO and RADIO

FREQUENCIES

The term very-low frequencies refers here to the range below normal hearing say from zero
to around 30 Hz. In this range inductive and capacitive elements are not practical arid so
amplifiers required to operate here would almost certainly be dc-coupled as discussed under 3.4

** An audio amplifier covers the range of frequencies normally distinguished by the human
ear say 50 Hz to a maximum of 16 kHz. Voice frequencies are defined as those essential for the
transmission of speech of commercial quality, often referred to as telephone quality and
covering the range 330 - 3,400 Hz. The full bandwidth is sometimes described as "music
quality".

The full range from 50 Hz to 16 kHz is enjoyed only by the young. In noisy environments
the upper range falls away with increasing age and often the upper limit is around 6 or 7 kHz.
 Amplifiers which cover the full range are known colloquially as high-fidelity (or hi-fi)
amplifiers and, usually, are capacitively or inductively (transformer) coupled. They use n.f.b. to
reduce distortion and to provide an output impedance that damps any mechanical resonances in
the loudspeaker.

A video amplifier is used to handle picture signals either for television purposes or other
displays such as computer outputs. Their bandwidth extends from zero frequency up to around
10 MHz maximum depending on the definition required in the displayed picture.

Such amplifiers may be dc-coupled, may use an approximate method of replacing the dc-
component after amplification known as dc-restoration or use an accurate method of replacing
the dc-component known as clamping; this matter is not required for the RAE. In valve-
operated video amplifiers the high-impedance and shunt capacitances of the valves make it very
difficult to obtain these very-wide bandwidths; transistors on the other hand, although
intrinsically offering greater capacitances, are low-impedance devices and the usual problem is
to restrict the available bandwidth to that required; frequency-weighted n.f.b. loops are the
usual means of achieving this limitation.

>>>>>>>>>>>>>>>>>>>> PAGE 6 <<<<<<<<<<<<<<<<<<<<

It is necessary to restrict bandwidth because this reduces the generated noise but also because it
removes the possibility of spurious oscillations occurring.

** Radio-frequency Amplifiers, as used here, indicates all amplifiers which handle signals
above the audio range. This definition excludes video amplifiers which extend downward to
zero frequency. They are mostly tuned amplifiers which use either simple circuits or bandpass
circuits.

** Wideband rf-amplifiers is a term which usually refers to an untuned rf-amplifier and this
uses a simple resistive load; because the stray capacitances are not eliminated by incorporation
into a tuned circuit they shunt the resistive load and so tend to restrict the bandwidth. The only
cure is to choose a low value for the load resistor and this means that such amplifiers usually
offer low gain; mostly they are used to buffer one circuit from another.

3.7 CLASS-A, CLASS-B and CLASS-C AMPLIFIERS

** These definitions refer to the mode in which amplifiers operate as determined by the dc-
conditions.

** Class-A Amplifiers are biased to the centre of their normal operating range so that the
direct current drawn from the auxiliary supply can be either increased or decreased by the same
amount. This mode gives the least distortion and is the one used for most voltage or current
amplifiers. It is used frequently too for small power amplifiers but results in the least efficiency
- such an amplifier gets very hot.

** Class-C Amplifiers are biased so that, without an input signal, they do not draw current
from the auxiliary supply - they are said to be normally cut-off. In fact they are biased well
beyond the cut-off point. An input signal drives a Class-C active-device into conduction during
a small part of one half-cycle only and so greatly distorts the waveform; as a consequence it can
be used only in some form of (un-modulated) rf-amplifier where a tuned circuit can be used
(like a flywheel) to restore the output waveform. The benefit however is greatly-increased
efficiency.

** Class-B Amplifiers are biased half-way between the Class-A and Class-C modes and
effectively amplify one half of the input waveform without too much distortion. As a result it is
necessary to construct Class-B amplifiers using two active devices such that one amplifies one
half of the waveform and the other device amplifies the other half of the waveform. Such an
arrangement is referred to as a push-pull amplifier.

** The two halves of the waveform are usually reconstituted by use of a special transformer.
This technique is used for efficient high-power audio amplifiers where a single-ended Class-A
amplifier would cause heating problems. With semiconductor devices it is possible to
build transformerless push-pull output stages.

** There is an arrangement which falls between the Class-A and Class-B circuits known
as Class AB. As might be expected such an amplifying stage is not biased for Class-A operation
but it is not so heavily biased as to take it into Class-B; various options are available between
the two extremes and, depending on the degree of biasing, they are designated ABl, AB2, etc.

>>>>>>>>>>>>>>>>>>>> PAGE 7 <<<<<<<<<<<<<<<<<<<<

** All Class-AB and Class-B amplifiers cause some distortion and negative feedback is used to
restore the waveform after everything possible has been done, by design and adjustment, to
minimise that distortion.

3.8 TRANSFORMER-COUPLED, RC-COUPLED and DIRECT-COUPLED

AMPLIFIERS

** These classes of amplifiers have been covered already in the above Sections. Transformer-
coupled designs use a transformer to couple the output of one amplifier stage to the input of the
next thus preventing the dc in the output from upsetting the dc-conditions at the input of the
following stage. Designing these transformers for full-bandwidth amplifiers is a job for the
expert.

** The RC-coupled amplifier uses capacitors to block the dc between stages but it is necessary
to supply a high-resistance path to enable the capacitor to both charge and discharge with the
signal waveform. The combination of resistance and capacitance in the coupling has given its
name.

The coupling circuits however begin to fail with falling frequency as their time-
constant (see Lesson 3) becomes comparable with the time of one cycle (the periodicity of the
signal). With several RC circuits in cascade throughout the amplifier the resulting fall-off
(or roll-off) in frequency can become so rapid that (with feedback applied) phase-shift troubles
can lead to instability. This has to be countered by staggering the time-constants; i.e. using a
different combination of capacitor and resistor for each coupling circuit.
 As already described direct-coupled (or dc) circuits do not use coupling networks but
connect the output of one stage directly to the input of the following stage. Usually, to maintain
the necessary dc-operating conditions, this involves the use of some kind of potential-divider
network together with extra biasing lines or the inclusion of semiconductor diodes to adjust the
dc potentials.

END OF LESSON 1

*******

QUESTIONS

1. What is the difference between a voltage amplifier and a constant-voltage amplifier ?

2. What is the difference between a Class-A amplifier and a Class-C amplifier ? Where would
you expect to find a Class-AB1 amplifier ?

3. How is distortion overcome in a Class-C amplifier?

4. What is self-inductance?

5. Given three components namely a capacitor, an inductor and a resistor they can be connected
either as a series circuit or as a parallel circuit. Given also that, for practical purposes, the two
circuits exhibit the same resonance frequency what do you see as the difference between them ?

6. A video amplifier will handle dc and frequencies from 25 Hz up to 10 MHz. What is the best
method for expressing its bandwidth ?

7. What is meant by the term ground-plane aerial ?

>>>>>>>>>>>>>>>>>>>> PAGE 8 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 2
3.9 PULSE AMPLIFIERS

In the wake of Radar and Electronic Navigation Aids just before, and during, the second
World War a new discipline arose known as Special Circuits. Post-war this became known
as Pulse Circuits. It was concerned with the generation, shaping and grouping of short-
duration [Link] a carrier is modulated by a train of pulses the result is a train of RF
Pulses. (At first pulses were often called impulses but that term is now deprecated.)

Pulses can be generated in passive circuits or in active circuits; passive circuits are
combinations of reactive and resistive components (CR or LR circuits) and active circuits utilise
either valves or transistors. Once generated the pulses have to be shaped to the desired form and
this almost always involves use of a Pulse Amplifier.

This class of amplifier operates in a mode that might well be compared with Class-C
operation in that it is normally biased either so that it is well cut-off or well and truly driven into
conduction. Large-amplitude pulses applied to the amplifier either turn it abruptly ON or
abruptly OFF thus producing an output which is timed by the input pulses but which has very
fast rise-times and fall-times.

Today the art of generating pulses and/or pulse-trains is fast becoming lost except to the
designers of chips. The low cost of units capable of generating all types of pulses and pulse
trains makes it both unnecessary and uneconomic to generate them with discrete components.

The subject is not covered for the R.A.E.

3.10 DIFFERENTIAL AMPLIFIERS

A differential amplifier essentially is two amplifiers which share a common output load but
one of these amplifiers inverts the signal. Thus, if the same signal is applied simultaneously to
both amplifiers the output is zero because the two output signals cancel. Any output signal
represents the difference between the two input signals.

In early logic circuits the arrangement was sometimes used as an OR-gate which means that
its output changed only if one OR other of the inputs changed between a logic-level 0 and a
logic-level 1; if both inputs changed then the output remained unchanged.

Probably the major use of this amplifier is in making waveform measurements. The input-
signal applied to a system under test is fed to one input of a differential amplifier ; the output
from the system is fed via an attenuator to the other input. When the attenuator is adjusted to
make the two signals of equal amplitude then the differential-output (the difference) represents
the distortions introduced into the signal path by that system, This difference signal, when
amplified, is a powerful indication of any form of distortion.

This type of amplifier is found also in the stabilised power-supplier with a balanced shunt-
amplifier which is described in Part 5 of this Course.

The subject is not covered for the R.A.E.

>>>>>>>>>>>>>>>>>>>> PAGE 9 <<<<<<<<<<<<<<<<<<<<

3.11 OPERATIONAL AMPLIFIERS (Op-amps)

The op-amp first appeared in valve circuits where it consisted of a single-stage amplifier
with a feedback component connected between output and input terminals (between anode and
grid). Where the feedback path is purely resistive the device offers a purely-resistive input
impedance with low but very stable gain; it might be used, for example, to invert a logic signal
from a positive-going pulse to a negative-going pulse. Where the feedback path consists of a
capacitor then the device offers an input impedance equivalent roughly to the feedback-
capacitance multiplied by the stage-gain (measured with the feedback path disconnected).

Modern semiconductor op-amps are, by comparison, formidable devices. The chip

incorporates several transistor amplifiers which produce an open-loop gain up to 106 .
Additionally the device has two separate (differential) inputs; one of these produces a signal at
the output which is in-phase with the input signal and the other produces an output signal which
is inverted. These are always labeled "+" and "-" respectively. These devices are not
examined for the R.A.E. but, for those with an interest, some details are given under 3.13.

3.12 VALVES and TRANSISTORS in AMPLIFIER CIRCUITS

3.12.1 Introduction

All amplifier circuits require an auxiliary power-supplier and an active device to control the
flow of that power to the output circuit. In small-signal circuits the valve has been entirely
displaced by the transistor on the grounds of size, cost, weight, power-consumption and
bandwidth. In high-power circuits however the valve still reigns supreme because it can absorb
overloads and various forms of abuses which destroy transistors almost instantaneously.

This situation however is being reversed by the modern semiconductor "chip" which
incorporates a complex and sophisticated circuit on to one small piece of silicon. Many of these
now offer built-in current-limiting circuits which prevent damage by overload or by short-circuit
and also incorporate heat-sensing elements which shut-down the device if it begins to overheat.
Under fault conditions the device ceases to function, of course, but at least it is not destroyed.
Needless to say this act of self-preservation produces some interesting fault-finding problems ;
when, for any reason at all, the chip cools then the "fault" disappears for a short period.

Designing with valves and designing with transistors are almost completely opposite
disciplines:

(1) Valves use lethal voltages between 50 and 350 volts; transistors seldom exceed 50 volts.
The P.A. stage in valve-output amateur transmitters run usually around 1,000 volts - BEWARE.

Note that, although transistor circuits are normally quite safe to touch and handle while live,
a mains-operated unit will have 240-volts ac within it somewhere - BEWARE.

Note also that television receivers and oscilloscopes provide high voltages to run the display
tube whether they are transistor-operated or valve-operated - BEWARE.

>>>>>>>>>>>>>>>>>>>> PAGE 10 <<<<<<<<<<<<<<<<<<<<

(ii) Valves draw currents between 1 mA and 350 mA ; transistors draw currents up to 5 amps or
more. These high currents demand low-resistance paths, low-resistance contacts in plugs and
sockets and cause problems in designing power-suppliers that do not radiate interference either
directly or via the mains (see Part 5).
(iii) Valves are high-impedance devices and so stray capacitances limit their high-frequency
performance.

Transistors are low-impedance devices and normal strays do not cause problems ; the problem
with transistors is too much bandwidth and it is usual to weight any feedback loop so as to
restrict the bandwidth to that required.

(iv) Unwanted feedback, often causing self-oscillation, occurs with valves mainly because of the
high signal-voltages that are generated at high impedance.

Unwanted feedback, often causing self-oscillation, occurs with transistors mainly because of the
low-impedance paths that exist internally between transistor terminals but mainly because of the
very-small separation between input and output connections. This problem is particularly
serious with the integrated-circuit "chips" where gains in excess of 106 can exist between
connecting pins separated by less than ½-inch.

(v) The current which flows from the auxiliary power-supplier through a valve is controlled by
the voltage which is impressed between its grid and cathode connections. The signal waveform
is easy to observe and measure with a normal oscilloscope.

The current which flows from the auxiliary power-supplier through a (bipolar) transistor is
controlled by the current which is injected between its base and emitter connections. This is
both a low-impedance and a variable-impedance path and the signal waveform, as seen on a
normal oscilloscope with difficulty, is not very meaningful. It is necessary to use some kind of
current probe.

3.12.2 The Thermionic Valve as an Active Device

The simplest valve is the diode; the name

derives from the prefix di meaning two. As
illustrated in Fig.1 it has two electrodes within
an evacuated envelope. One of the these,
the cathode, is specially treated so that, when
heated, if gives off a cloud of free electrons.
These form a space charge around the cathode
within the evacuated envelope. In early valves
the cathode was a simple heated filament
similar to an electric lamp but so-
called indirectly-heated valves now use a
cylindrical cathode with a filament heater
Fig. 1 The Diode Valve mounted inside it. You may encounter a valve
in which he cathode is simply a filament where
that valve is intended for low-voltage working
in small dry-battery-operated equipments.

>>>>>>>>>>>>>>>>>>>> PAGE 11 <<<<<<<<<<<<<<<<<<<<

The second electrode or anode is a metal plate which surrounds the cathode. When charged
negatively with respect to the cathode it repels the space-charge back toward the cathode. When
charged positively with respect to the cathode it attracts the electrons of the space-charge and so
a flow of electrons (an electric current) occurs from the cathode to the anode.

As described in Part-2 (Detectors) and in Part-5 (Rectifiers) such "one-way" diodes can be
used to convert ac supplies to dc.

Triode valves, as the name implies, have a third electrode and this takes the form of a metal
mesh or a spiral winding mounted between the cathode and anode. In all valves except the diode
the anode is always maintained at a positive potential (with respect to the cathode) so that anode
current flows continuously. When the grid is made negative with respect to (w.r.t.) the cathode
then it repels electrons back toward the cathode and so reduces the anode current. Over a range
of grid-cathode (negative) voltages the change of anode current is reasonably proportional to the
change of grid voltage - we say that they are linearly related.

When the grid is made positive w.r.t. the

cathode it accelerates the electrons and so
increases the anode current BUT it also collects
some of those electrons causing a flow of grid
current; as a result the increase in anode
current does not accurately reflect the increase
in grid voltage. Except for Class-C operation
the grid is not allowed to draw current because
of the ensuing distortion and also because few
valves can withstand the heat generated at the
grid.

Fig. 2 The Triode Valve As long as the grid is maintained negative

w.r.t. the cathode the input (grid-cathode) is
open-circuited and so only the voltage
waveform is of importance. Fig. 3 shows the
basic action of a valve as a voltage-amplifier.
The anode current (Ia) is drawn from
the auxiliary supplier
(the high-tension or H.T. source) through
an anode-load resistor Ra (Note - this is
a capital R) and, as a consequence, there
is a p.d. across the resistor given by
Ohm's Law as

Ia x Ra

Because of this p.d. the anode

potential is less than that produced by the
auxiliary source. At any moment in time
the anode-current, and consequently the
anode potential, is determined by the
voltage on the grid w.r.t. the cathode.
 >>>>>>>>>>>>>>>>>>>> PAGE 12 <<<<<<<<<<<<<<<<<<<<

Without a signal input the value of Ia is determined by a standing bias-voltage on the grid
which is set so that the anode potential is about half-way between the H.T. value and ground. An
alternating signal-voltage applied to the grid, and mixed with the bias, thus causes the anode
potential to rise and fall in sympathy with the grid-input voltage. A simple circuit of this nature
is shown in diagram (a). The capacitor Cin isolates the signal source from the bias.

Diagram (b) shows the usual arrangement. The biasing requirement is that the grid should be
negative w.r.t. the cathode. This is equally served by making the cathode positive w.r.t. the grid.
Bias for the cathode can be developed by inserting a cathode-bias resistor (Rk) between cathode
and ground line. The anode-current must flow through this resistor to return to the cathode and,
in doing so, it raises the cathode positive w.r.t. that ground line by a p.d. equal to I x Rk.

At this stage we are interested only in developing a cathode bias; i.e. a direct potential.
With an alternating signal present at the grid the bias across Rk must vary in sympathy with the
signal waveform. This is avoided by connecting a large-value capacitor Ck across the cathode-
resistor as shown. This capacitor is an electrolytic capacitor which must be connected correctly
; note the symbol. There are two ways of considering the action of Ck:

(a) the cathode-bypass capacitor must have an impedance to the ac which is much lower than
Rk; as a result the alternating component of the current flows only through the capacitor while
the direct component flows only through the resistor

(b) the time-constant of [Link] (see Funds-1, Lesson-3) must be large enough that the cathode-
capacitor cannot charge (or discharge) in the time permitted by each cycle of the signal
waveform.

The grid has a tendency to intercept electrons as they pass through and, as these accumulate,
so they produce a negative potential on the grid. A grid leak resistor Rg is necessary to return
these electrons to the cathode,

Tetrode and pentode valves are made which have respectively two and three grids. The
second grid (the screen-grid) is held at a constant positive voltage, as indicated in the
manufacturers' data, by means of a dropping-resistor; this functions in the same way as the
cathode resistor and needs to be decoupled with a large-value screen-decoupling
capacitor. This screen grid reduces the internal stray capacitances of the valve (a Faraday
Screen) and also enables it to produce higher voltage-gain.

The third suppressor grid is connected either to the cathode or to ground; its main function
is to prevent a "negative-resistance" effect caused by electrons ejected from an overheating
anode flowing to the screen grid. (The result is that the anode current falls as the anode voltage
is raised.)

Valves circuits are not examined for the R.A.E. except for transmitter valve-operated P.A.
stages

>>>>>>>>>>>>>>>>>>>> PAGE 13 <<<<<<<<<<<<<<<<<<<<

3.12.3 The Transistor as an Active Device

 As with valve devices the simplest semiconductor device is the diode. These come in many
different constructions but the basic operation is as follows. A semiconducting material is doped
at one end as n-type and at the other end as p-type; see Fig.4. This means that one end of the
material is rich in electrons and the other end is rich in holes. This is known as a p-n Junction.

Such a construction results in the formation

of a potential "barrier" at the junction of the
two types and the electrons and holes are driven
away from this. When a p.d. is applied across
the ends of this device, such that the n-type is
made more positive than the p-type, the
tendency to migrate toward the ends is
encouraged; because electrons and holes may
not come into contact it is not possible for an
Fig.4 Forward-biased P-N Junction Diode
electron current to flow.

When the n-type material is made more negative than the p-type material, as shown
in Fig.4 , electrons are repelled toward the p-type and drawn across the barrier toward the more
positive potential and so an electron-current flows from negative-to-positive. This is equivalent
to a flow of classical-currrent from positive-to-negative (i.e. from p-type to n-type material).

The two most common types of diode are made from Germanium and Silicon. Germanium-
based diodes offer forward conduction voltages around 0.2-volt and are ideal for small-signal
applications such as detectors. However they are not very robust in terms of voltage and power-
dissipation. Silicon diodes offer forward-conduction voltages around 0.6-volt which is less
efficient for very-small signal rectification but they can handle relatively large powers because
of their ability to withstand higher temperatures.

The simplest transistors have three connections and may be likened to the triode valve. The
earliest were the bipolar transistors which get their name because conduction takes place
through the agency of both majority and minority carriers (see Lesson-1 in Fundamentals-1).

As shown in Fig.5, current from the auxiliary

supplier flows through these transistors
between an emitter and a collector and its value
is determined by the current injected between
the emitter and a base connection. The signal-
current which flows between collector and
emitter is greater than the input signal-current
Fig.5 Bipolar Transistor
by a factor known as the current-gain and
given the symbol h FE .

An improved transistor for many purposes is the Field-effect transistor or FET which has an
open-circuited input in much the same manner as does a valve. Indeed, except for the lower
direct potentials required, it can often be used in exactly the same manner as a valve.
 >>>>>>>>>>>>>>>>>>>> PAGE 14 <<<<<<<<<<<<<<<<<<<<

In a FET the input connection is

a gate which is the equivalent of the base in a
bipolar transistor or of the grid in a valve (the
analogy should not be taken too literally) ; see
Fig.6. The gate is insulated from the main
structure (the substrate) . The current from the
auxiliary supplier flows from a source to
a drain and passes between the gate and
substrate (body of the device). An appropriate
voltage applied between gate and substrate
"pinches-off" this current thus acting as a
current regulator.

END OF LESSON 2

*******

QUESTIONS

1. Why is a differential-amplifier useful as a measurement tool ?

2. Briefly what is a modern Op-amp ?

3. Give the main differences between valve-operated circuits and transistor circuits.

4. Why does a valve need a filament supply ?

5. What is the function of the grid in a triode valve ?

6. Re-draw Fig. 3(b) using a bipolar transistor.

7. The emitter-bias resistor has a value of 1 k-ohms. Give a suitable value for the bypass
capacitor if the stage has to handle a signal of 100 Hz.

>>>>>>>>>>>>>>>>>>>> PAGE 15 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 3
3.12.4 Amplifier Circuits - General

** The basic amplifying device has three connections ; current from the auxiliary power-
supplier flows between two of these under the control of an input-signal which is applied to the
third.
 In the vast majority of circuits one of these electrodes is either earthed or it is maintained at
earth potential for alternating signals. This is convenient in terms of safety, the elimination of
noise signals and the control of stray couplings.

Fig. 7 shows the basic arrangement in what appears to be two forms. In diagram (a) the
signal source is returned to the lower of the two other electrodes and, in diagram (b), it is
returned to the upper of the two electrodes. In terms of signal currents the two forms are in fact
identical.

** The auxiliary supplier is required to provide dc and so it contains large-value smoothing

capacitors which are connected directly between the two supply lines ; see Part-5. The low-
impedance of these ensures that alternating potentials cannot develop across the dc supply thus,
as far as ac-signals are concerned, the two supply lines (one positive and the other negative) are
connected together.

** Where equipment is operated from batteries it is usually necessary to connect a "by-pass

capacitor" across the supply lines to ensure that there is a low-impedance path around the battery
for signal currents. Without this the battery acts as an impedance which is common to all
amplifier stages and the resulting inter-stage coupling can result in a low-frequency oscillation
known as "motor-boating"

Provided that the dc-supplier is properly decoupled the input signal-source may be returned
to either line as may be necessary to suit dc-requirements. Precautions are necessary of course
to ensure that the signal-source does not draw direct-current from the supplier.

Note that an Earth connection is not shown in either of these two diagrams because it is not
relevant to the action of the circuit. This earth can be connected wherever it is most convenient
and this gives rise to three possibilities:

>>>>>>>>>>>>>>>>>>>> PAGE 16 <<<<<<<<<<<<<<<<<<<

3.12.5 The Grounded-collector (Grounded-anode) Amplifier

** Fig. 8 shows the basic configuration with the collector connected directly to the supply line
; as explained above this is exactly the same signal-wise as connecting the collector to the
grounded side of the supply. The signal current injected between base and emitter controls the
flow of collector-emitter current.

** It is convenient here to introduce one of the

complications of semiconductor devices which
arises from their construction using both p-type
and n-type materials (see Lesson-
1 of Fundamentals-1).

** Bipolar transistors are built as sandwiches

of p-type and n-type material and, according to
the structure, are known as
either pnp or npntransistors. For the R.A.E.
the only significant detail is that pnp transistors
require a negative supply-line while npn types
require a positive supply-line. The difference is
shown in circuit diagrams by the direction of
the small arrowhead which delineates the
emitter (the collector is not marked). An
outward-pointing arrow indicates an npn
transistor.

** Diagram (a) shows the input-signal connected between base and emitter and this is simply
the basic diagram of Fig. 6 with the earth connection taken to the collector. A load resistor is
connected between the emitter and ground so that the collector-emitter current develops an
output-voltage.

** A more convenient, and more usual, connection for the driving source is shown in diagram
(b) where the source is returned to ground rather than to the emitter (if you prefer it the signal-
source has now been returned to the collector). As a result the input circuit between base and
emitter now contains the output-voltage in series with the source-voltage. This is an example of
negative-feedback in which ALL the output voltage is fed back to the input and it has two
important consequences:

(i) If (say) the base is taken more positive with respect to the emitter then the collector-emitter
current increases, the p.d. across the emitter resistor increases and so the emitter too goes more
positive. For this reason the arrangement is referred to as an emitter-follower circuit. Whatever
happens to the base is imitated by the emitter.

>>>>>>>>>>>>>>>>>>>> PAGE 17 <<<<<<<<<<<<<<<<<<<<

(ii) For the base-bias to remain positive w.r.t. the emitter the voltage-excursion at the emitter
must be smaller than the excursion at the base. This means that the signal-gain, from input to
output, must be less than 1 - there is always a slight voltage-loss.

** This may appear to be a poor amplifier and, in terms of voltage gain, so it is. The analysis
of the circuit is not required for the R.A.E. and it is necessary to note only that the circuit
provides a high input impedance and a very-low output impedance ; in other words it is
a current-amplifier. It is a most useful device for driving low-impedance or capacitive
(frequency-conscious) loads where it is required to buffer (i.e. to protect) the circuit from the
effects of any load. Note that the output voltage is in phase with the input voltage ; i.e. if the
input signal goes positive then the output also goes positive.

Where the circuit is built using a valve it is called a cathode-fol1ower.

3. 12. 6 The Grounded-grid (Grounded-base) Amplifier

** Fig. 9 again shows the basic amplifier circuit but this time it uses a valve with the input-
connection (the grid) connected to ground. The flow of anode-cathode current is controlled by
the voltage which is impressed between the grid and cathode connections thus, if the grid is to
be earthed, the stage must be driven by varying the voltage of the cathode.

** The circuit has much in common with the

cathode-follower described above except that this
time only part of the output-voltage is developed
across the cathode resistor ; a much larger output
voltage appears across the anode load. The source
and fed-back voltages are now in parallel. The great
advantage of this arrangement is that the earthed grid
acts as a Faraday Screen between anode and
cathode and so eliminates the effect of the anode-grid
capacitance. In rf-amplifiers this capacitance can
provide unwanted positive-feedback..

If, say, the cathode is driven positive then the grid

becomes more negative w.r.t. the cathode and so the
valve draws less current. Less anode-current means
that the voltage developed across the anode-load
resistance is decreased and so the anode potential
moves toward the positive supply line. Thus once
again the input and output voltages are in phase.

>>>>>>>>>>>>>>>>>>>> PAGE 18 <<<<<<<<<<<<<<<<<<<<

3. 12. 7 The Grounded-emitter (Grounded-cathode) Amplifier

** This version of the basic circuit, shown in Fig. 10, is the most frequently occurring form.
In this it is the emitter which is earthed.

** However, there are a few extra complications which arise from the deficiencies of the
bipolar transistor. These transistors suffer from leakage current which flows from the collector
to the base and this current drives the transistor into conduction. The resulting power-dissipation
causes the device to generate heat which further increases the leakage-current. The result is a
phenomenon known as thermal-runaway which quickly destroys the transistor
 ** The cure for this problem is dc negative-
feedback and this is the purpose of the two
resistors in the base circuit and the third resistor
in the emitter circuit which are designated R 1 ,
R 2 and R 3 . The base-resistors form a potential-
divider across the supply ; they have relatively
low-values so that they form a constant-voltage
source to which the base is connected. If (say)
the collector-emitter current tends to increase
this causes an increase in the p.d. developed
across R3 and this, by raising the potential of
the emitter (against the constant potential of the
base), reduces the base-emitter voltage and so
tends to oppose the increase of collector-emitter
current.

** A large-value electrolytic capacitor is also

connected between emitter and ground to
provide a low-impedance path for the ac signal-
component; this prevents a signal voltage
developing across R 3 which, as ac negative-
feedback, would reduce the overall stage-gain.

In a valve version of this amplifier, and also in the FET version described below, the
cathode-resistor Rk serves to generate automatic bias for the amplifier. As described earlier this
sets the correct operating point for the valve by raising the cathode-voltage with respect to the
grid. Thermal runaway is not a problem because valves do not suffer from leakage currents and
so the low-impedance potential-divider circuit is unnecessary. However a grid leak resistor is
required as described above to ensure that the grid does not become negatively-charged.

It is not unusual to see circuit diagrams in which the envelopes of valves or transistors are
omitted; in practical situations the omission is a matter of convenience and speed but there is
not a rule.

>>>>>>>>>>>>>>>>>>>> PAGE 19 <<<<<<<<<<<<<<<<<<<<

3. 12.8 The FET as an Amplifier

** The field-effect transistor (FET) is a transistor which uses a different method of current-
control as described in Lesson-2 of this Part. The emitter and collector of the bipolar transistor
are replaced by a source and a drain between which current flows in a channel . The effective
"size" of this channel is controlled by an electric field which is set-up by applying a voltage to
a gate (which replaces the base of a bipolar transistor).
 ** The great difference between these two
types of transistors is that a gate electrode is
insulated from the main structure and
therefore it does not draw current. Only the
electric field is concerned and so the passage
of auxiliary current between source and drain
is determined by the voltage applied to the
gate. Thus an FET is, in effect, a solid-state
valve.

Note that the gate symbol indicates which

connection is the source. It indicates also
Fig.11 Comparison of Bipolar Transistor and whether the channel material is p-type or n-
FET type by means of the small arrow and it is here
that complications begin to accumulate.

The FET device shown in Fig. 11 has an inwards-pointing arrow which indicates an n-
channel and thereby that the device requires a positive supply. Alas ; thereafter it gets ever
more confusing but, fortunately, you are not required to sort them out for the R.A.E.

** FET devices come with two kinds of gate construction:

1. semiconductor junction (JFET)

2. metal-oxide-semiconductor insulated (MOSFET) sometimes known as insulated-gate

FETs or IGFETs.

In the first kind the gate electrode is formed as a PN-diode junction exactly as in the bipolar
types ; as a result, if it is allowed to become forward-biased, the gate will draw current.

In MOSFET devices the gate is insulated by a very-thin layer of glass and so it can be biased
in either direction without drawing current. When the gate potential is carried toward the supply
potential conduction takes place between source and drain.

** The glass insulating-layer is extremely thin and so vulnerable to damage by static charges
such as may exist on a human body ; thus, unless precautions are taken to discharge all static, a
MOSFET will be destroyed simply by the act of picking it up. However, once soldered into
position on a circuit board, the risk is much reduced.

>>>>>>>>>>>>>>>>>>>> PAGE 20 <<<<<<<<<<<<<<<<<<<<

 ** Fig. 12 shows a basic voltage-amplifier
circuit using a MOSFET. Compare this with
the bipolar transistor amplifier in Fig.10.

** In both voltage-amplifier circuits there is a

limit to the amplitude of the input signal. At
one extremity the transistor or valve becomes
completely cut-off, the p.d. across the load
becomes zero and so the drain (anode) voltage
rises to the value of the positive-supply line.
The output cannot rise beyond this value.

** At the other extremity the device is driven

heavily into conduction with the result that the
p.d. developed across the load drives the drain
(anode) voltage to a minimum value. For a
valve this is around 50 volts but for a silicon
transistor it is around 0.6 volts. In this condition
the device is said to be saturated or bottomed.

** For the output signal to be a [Link] of the input waveform there needs to be a linear
relation between input voltage and source-drain current (cathode-anode current). Such a relation
only holds however over a limited part of the excursion from cut-off to saturation ; to amplify
the positive half-cycles and negative half-cycles equally it is necessary to ensure that:

(i) the zero-level of the input signal coincides with the centre of the linear range ; this is the
purpose of the bias voltage that is developed across the source (cathode)resistor.

(ii) the amplitude of the input signal does not exceed the linear part of the range

This is a Class-A amplifier circuit.

3. 12. 9 The Application of Feedback

** The simplest method of applying feedback to an amplifier stage is to omit the bypass
capacitor which normally is connected across the source (cathode) resistor. The varying dc
which flows through the resistor now generates a varying direct-voltage instead of the steady
direct-voltage which is required for biasing. Note that this feedback voltage is developed from
the output-signal current and so represents a sample of the output current. One effect of this
arrangement therefore must be to stabilise the output current ; i.e. it raises the output (or source)
impedance of the amplifier stage.

This feedback voltage appears in series with the input voltage between gate (base or grid)
and source (emitter or cathode) and so tends to cancel the input-signal voltage . The result is to
reduce the stage-gain but it can be expressed as a need to raise the input voltage to maintain the
same level of output voltage.

>>>>>>>>>>>>>>>>>>>> PAGE 21 <<<<<<<<<<<<<<<<<<<<

The need to raise the voltage implies that the input-impedance of the stage has been increased.

** Fig.13 shows two methods of applying voltage-derived feedback. These arrangements,

because they sample the output voltage, tend to stabilise that output voltage which implies that
the output impedance has been reduced.

In diagram (a), where the feedback voltage is applied to the emitter, then the input
impedance is again increased. In diagram (b), where the feedback voltage is applied to the grid,
the input and feedback signals appear at the grid in parallel. When (say) the input-signal source
is trying to supply current so the negative-feedback source is abstracting current and so the
source experiences a demand for more current ; it interprets this as a reduced input impedance.

(For the R.A.E. feedback circuit arrangements are not examined in great detail. Familiarise
yourself with this Section but do not put too much emphasis on trying to learn it.)

In connection with the cathode by-pass capacitor (emitter by-pass capacitor) shown in the
above diagrams the reactance of the component increases as the frequency is decreased. Thus, at
low frequencies, the component fails in its task of decoupling and the circuit gain falls as
negative-feedback develops.

3. 12. 10 Neutralising

Although not generally regarded as a feedback-amplifier arrangement nevertheless the

process of neutralising rf-amplifiers does use feedback techniques.

A problem arises from the natural (stray) capacitance which exists between the anode and
grid of a valve (or the similar electrodes of semiconductor devices) and the natural stray
capacitances which exist between parts of the input and output circuit components. At radio
frequencies such small-value strays offer a low-impedance path to signal currents and so part of
the output signal from an amplifier stage is fed back to its input circuit.

>>>>>>>>>>>>>>>>>>>> PAGE 22 <<<<<<<<<<<<<<<<<<<<

 Depending on the tuning of the various circuits these inherent feedback paths may result in
either negative-feedback or positive-feedback or a variety of conditions between the two
extremes. The result is either an uncontrollable oscillation or obscure forms of misbehaviour
from the would- be amplifier stage.

The strays themselves can be reduced by

careful design but they cannot be eliminated
and so circuit designers have dreamed up a
variety of tricks to eliminate their unwanted
effects . Fig. 13A illustrates the basic idea in
which a small sample of the output signal is
derived with "opposite phase" (i.e. 1800 phase
shift) and this is fed back also to the input
circuit.

The trick lies in adjusting the level of this

fed-back signal so that it exactly cancels
(neutralises) the unwanted signal.

One technique is to disable the stage (by disconnecting the dc power-supply) and, with the
rf-drive in operation, the output-circuit tuning is then varied. As the output circuit passes
through resonance at the drive frequency so it absorbs power from the drive via the stray
coupling ; the consequent load imposed on the drive-source causes a fall in its output level and
this is reflected by a fall (a dip) in the input current. The level of the neutralising-signal is
adjusted until this dip is minimised thus showing that cancellation is taking place.

END OF LESSON 3

**************

QUESTIONS

1. From the signal point of view why is it that the source may be connected either between the
input and earth or between the input and supply line of an amplifier ?

2. Why do we use emitter-followers with their voltage-loss ?

3. What is the main objective of a grounded-grid amplifier ?

4. Why is a large-value capacitor connected between emitter (or cathode) and ground in a
grounded-emitter amplifier ?
5. What is the advantage of using a FET transistor against a bipolar type?

>>>>>>>>>>>>>>>>>>>> PAGE 22 (again) < <<< < <<<<<<<<<<<<<<<

Back to Top of Page

LESSON 4
3.13 CIRCUIT DETAILS

3.13.1 AF Amplifiers

The audio amplifier is often considered to be a simple straightforward arrangement because

it handles low frequencies. In fact it has a very-large relative bandwidth covering some eight
octaves.

The design of a high-quality audio amplifier can be a tricky business. I have lost two
workshop loudspeakers because a "hi-fi" amplifier of considerable power intermittently burst
into a supersonic (above audible range) oscillation. Any audio amplifier with a hissy
background noise should be checked for performance as an rf-oscillator but even this is
complicated by today's habit of designing power-suppliers without smoothing chokes. The
resulting large current-peaks drawn from the mains often make an otherwise well-behaved
amplifier sound as though it were oscillating.

** The R.A.E. is not concerned with high-quality reproduction of sound but only with the
audio stages of receivers. There is however a requirement for good design in microphone
circuits, modulator circuits and auxiliary circuits such as speech processors.

** A basic single-stage audio amplifier

circuit is shown in Fig.12 of the previous
Lesson and is reproduced again here as a
2-stage amplifier (Fig. 14). The FET
device shown can be directly replaced
with a valve provided of course that
suitable adjustments are made to the
supply voltages and component values.

** The object of a 2-stage amplifier is

to increase the overall gain
but cascading two stages in this manner
introduces a few complications.

** The signal applied to the gate of the second stage is much larger than the non-amplified
signal applied to the first stage and so it may be necessary to use a different bias for the second
stage by changing the value of the source-resistor R6. A change of bias means that the standing
current (the current without signal input) is changed and so it may be necessary to adjust the
drain-resistor R5 also. It may also prove necessary to use a transistor of different signal-handling
capability in the second stage.
 ** The coupling capacitorC2 must be introduced to ensure that the direct-potential at the
drain of transistor TR1 does not bias the gate of transistor TR2. The resistor R4 must then be
introduced as a leak to ensure that C2 does not acquire a charge which again would bias the gate
of TR2. The combination C2 and R4 is known as an RC-coupling.

** An RC-coupling introduces a new complication in that it is frequency-conscious. At low

frequencies the reactance of the capacitor increases and a proportion of the (alternating) signal
voltage appears across it so reducing the amplitude of the signal applied to TR2.

>>>>>>>>>>>>>>>>>>>> PAGE 23 <<<<<<<<<<<<<<<<<<<<

Source decoupling-capacitors suffer the same frequency-related problem and so again, as

frequency falls, the steady bias turns into bias+signal. The resulting negative-feedback reduces
the stage gain at low frequencies.

To reduce the effects of small distortions negative feedback may be applied either overall,
to the individual stages or in both arrangements simultaneously. As already mentioned in earlier
Lessons the rule is that n.f.b. can be used to improve a good amplifier - but not to make a bad
design work. The frequency dependent attenuation introduced by circuit components such as
RC-couplings and bypass capacitors is accompanied by phase-shifts and, if these become too
large, the n.f.b. is converted to p.f.b. As a result the amplifier develops bad habits of which
perhaps the most obvious is a continuous oscillation. Depending on the severity of the phase-
shift, and the amount of gain available, an amplifier may burst into oscillation only on the
positive peaks of a signal waveform.

Excessive phase-shifts occur in multi-stage amplifiers because the effects of RC-circuits are
additive. A simple cure for this is to arrange that all the RC combinations have different time-
constants(the product of resistance and capacitance) which ensures that phase-shift effects from
each combination do not overlap.

** Audio amplifiers which provide either voltage or current gain invariably operate in Class-A
while output (power) stages may be found in Class-A, Class-AB or Class-B . As discussed in
previous Lessons power amplifiers in any mode other than Class-A must be push-pull
arrangements and would normally incorporate n.f.b. loops to reduce the last traces of distortion.
Such distortions are particularly noticeable at zero crossover points where the two halves of the
waveform are assembled.

** A Class-A 2-stage circuit using bipolar transistors is shown in Fig.15 and it differs from
the FET circuit of Fig.14 in one important way. The FET is voltage-operated and would
therefore be used to amplify a voltage waveform. Thus the value of each drain resistor would be
adjusted so that, without an input signal, the drain potential lies midway between that of the
supply line and the bottoming voltage of the transistor.
 ** The bipolar transistors are current
operated. To provide amplification it must
be arranged that, when TR1 demands an
increase in collector current, the extra
current is drawn from the base of TR2.
Similarly, when TR1 demands less
collector current, the surplus current is
passed to the base of TR2. To this end the
value of TRI collector-resistor is chosen so
that, without input signal, transistor TRI is
bottomed. The resistor cannot pass more
current and so any extra must be drawn
from the base of TR2; when TRI collector-
current decreases then the collector-
potential tries to rise which increases the
base-current of TR2.

>>>>>>>>>>>>>>>>>>>> PAGE 24 <<<<<<<<<<<<<<<<<<<<

A proper understanding of this is important when fault finding. An oscilloscope (which is

voltage-operated) attached to the drain of Tr1 in Fig.14 would display the amplified signal-
voltage waveform. An oscilloscope attached to the collector of TR1 in Fig.15would show a
puzzling small-amplitude voltage waveform with considerable distortion.

The value of the resistor in TR2 collector depends on the nature of the following stage. If it
is to drive a third amplifier then TR2 would be bottomed also when in the quiescent state. If the
output is required as a voltage then TR2 collector-potential is set midway between the supply
line and the bottomed potential.

** It is possible to use a transformer coupling either between stages or as an output-coupling

device. For the FET circuit (arid also for a valve circuit) this would be a step-up transformer
from drain to high-impedance gate thus providing additional voltage-gain. For the bipolar
arrangement it would need to be a step-down transformer from collector to low-impedance base
thus providing additional current-gain.

** In single-ended power amplifiers (not push-pull) a transformer coupling is essential

between the final output stage and a loudspeaker but push-pull arrangements permit one of the
so-called transformerless circuits.
 ** A basic form of this is illustrated
in Fig.16 With transistors such a circuit
can be designed to provide an output-
impedance at the centre point which is
suitable for the direct connection of a
loudspeaker. There are a variety of
practical circuits each of which seeks to
overcome one or more of the problems
inherent in the transformerless design.
The circuit is not examined for the
R.A.E. but details are to be found in the
literature.

** Note that, with an inductive load, the quiescent voltage(without an input signal) at the
collector (or drain) is that of the supply line because of the low-resistance of the transformer
primary winding. Passage of the alternating signal current induces signal voltage across the
primary winding and so the collector voltage swings both above and below the supply-potential.
The sum of the supply voltage and peak signal-voltage must not exceed the maximum rated
voltage of the transistor and a diode catching circuit may be included to ensure that the
maximum voltage rating of the device is not exceeded.

3.13.2 RF Amplifiers

** The description RF-amplifier is a little loose in that it covers several different types of
amplifier although all are used to amplify radio-frequency signals ; i.e. frequencies above the
audio range. They include:

(a) the "front-end" stages of receivers either the pre-detector stages of a TRF receiver or those
which precede the frequency-changer in a superhet receiver

>>>>>>>>>>>>>>>>>>>> PAGE 25 <<<<<<<<<<<<<<<<<<<<

(b) amplifier, buffer and multiplier stases in v.f.o., crystal oscillator and synthesiser circuits

(c) the various amplifier stages in transmitters and signal-generators which handle rf signals,
whether cw or modulated, including the final P.A. stage and its power drivers and any
external Linear Amplifiers which are interposed between a transmitter and the aerial.

Strictly the term applies also to IF amplifiers in that they handle radio-frequency signals. In
general it can be assumed that the term rf-amplifier refers to a (variable) single-tuned stage or
stages and that if-amplifier is used for the fixed-frequency band-pass stage or stages that follow
a frequency-changer.

** A radio-frequency amplifier uses tuned circuits as impedances (compare this with the
resistors used in af-amplifiers) because

(i) they provide the narrow bandwidth required for a selective amplifier
(ii) unless the circuits are resonated the stage gain would be destroyed by the low reactances
presented by stray capacitances at radio frequencies ; the strays are incorporated in the tuning
capacitance thus making use of capacitance that, otherwise, would be a disastrous nuisance

(iii) tuned circuits provide high (dynamic) impedances to maintain stage-gain; if resistive loads
were used they would have to be of low value to prevent them being bypassed by stray
capacitances.

** A 2-stage rf-amplifier circuit is shown

in Fig. 17. The tuned circuits make
couplings easy because the coils can be
constructed as transformers.

** It is necessary to tune each set of coils

simultaneously if variable tuning is to be
used. It is usual to do this with ganged
capacitors so that all tuning adjustments
are carried out with a single knob. These
are variable capacitors mounted on a
common frame with their adjustable vanes
mounted on a common (earthed) shaft.

** Accurate tuning with such an arrangement however requires accurately- matched inductors
and it is easier (and cheaper) to arrange extra variable capacitance known as trimmers; these are
shown in Fig. 17 with a hammer-head arrow which indicates that they are variable but pre-set
and then left. The tuning capacitors are indicated by the arrows drawn through them and the
dotted line which joins the arrows indicates that they are ganged on a single shaft.

Strictly of course trimmers are not required on all stages but the capacitors are manufactured
in standard form with the trimmers built-in.

At frequencies above about 100 MHz variable-tuned circuits as described become difficult
to handle because of the decreasing sizes of the inductors and capacitors required. As a
consequence circuits are encountered which are tuned by:

>>>>>>>>>>>>>>>>>>>> PAGE 26 < <<<<<<<<<<<<<<<<<<<

(a) variable-capacitance diodes whose ganging is adjusted by means of resistive trimmers

(b) variable inductors which are tuned by mechanical adjustment of their ferrite or copper cores
(see Lesson 4, Section 1.9.7 in Part 1). Trimming is carried out using small metal vanes which
are bent to adjust their capacitance to the chassis and/or by stretching the coils

(c) The tuned circuit may be replaced by a resonant cavity (a metal box) in which it is not at all
easy to vary the tuning.
(Circuits of types b & c should not be entered except in dire emergencies and when the
necessary test equipment is available.)

** The prime function of a frequency-changer stage in a receiver is to translate the frequency

of a signal to that of the IF amplifiers. Nevertheless it is an rf-amplifier and its signal input is
variable-tuned. In some receivers the aerial input is taken directly to a frequency-changer stage
and so it constitutes the only rf-amplifier in the receiver.

** In a receiver the input (base, gate or grid) of an rf-amplifier is usually connected to the
automatic gain-control (a.g.c.) line thus closing the n.f.b-loop that ensures a fairly-constant
signal level at the detector. All stages which operate within this loop have fairly-constant signal
levels presented to them except when the input-level from the aerial falls below a minimum and
the a.g.c. action fails.

The one exception to this rule is the first rf-amplifier, the stage which receives its input
directly from the aerial. The problem is that, with a strong incoming signal, the stage is trying to
cope with an unusually-large input signal at a time when the e.g.c. voltage is reducing the ability
of the stage to handle it. On the other hand this first stage is the most effective point at which
a.g.c. can operate. The very first stage of a receiver is often left without a.g.c. or only a portion
of the a.g.c. voltage is applied.

The a.g.c. voltage changes not only the gain of a stage but it also modifies the characteristics
of the active device. The input and output inter-electrode capacitances of the device form part of
the resonant circuits and, when these change value with changing a.g.c., so the resonant circuits
become de-tuned. This can be corrected by resorting once again to negative feedback. In valve
amplifiers a resistor as small as 10-ohms (un bypassed) in the cathode lead gives sufficient n.f.b.
to stabilise the tuning over the full range of input signals. No doubt the same technique works
with semiconductor circuits but my experience in this technique is limited to valves.

** With rf-amplifiers it is necessary to provide shielding between input and output circuits of
each stage. Although stray capacitances may amount to no more than a few picofarads their
reactances at radio frequencies can be sufficiently low to provide good feedback paths and so
cause instability. The most obvious cause of this problem can be seen in multi-stage rf-
amplifiers where the output and input circuits of a stage are both formed by resonant circuits
tuned (nominally) to the same frequency ; in Part 4 such a circuit arrangement is offered as a
(tuned-anode-tuned-grid) T.A.T.G. Oscillator.

>>>>>>>>>>>>>>>>>>>> PAGE 27 <<<<<<<<<<<<<<<<<<<<

** The varying voltage on the a.g.c. line causes the standing current in each rf and if stage to
vary with the signal strength. This forms the basis of the signal-strength meter or S-meter which
is found on the front panel of most commercial receivers or the various forms of tuning-
indicator which appear on domestic receivers. In fact, as signal strength increases, so the a.g.c.
action causes the standing current to decrease. A meter which indicates an increasing quantity
by deflecting to the left is unfamiliar and confusing and so circuit tricks are used to reverse this
action. Alternatively a meter can be constructed with a right-hand zero which simply means
that it deflects to the left with increasing current.

Occasionally circuits are encountered which use a wide-band rf-amplifier. Generally such
terminology indicates an untuned stage which uses a low-value resistor or low-Q choke as the
output load. It is essential that this load has a low-impedance otherwise it will be bypassed by
stray capacitances and so the stage-gain must be limited perhaps even to a value of 1. Such
amplifiers function mostly as buffers or impedance changers.

Linears or linear-amplifiers are rf-amplifiers whose output level is proportional to the input
level ; i.e. they do not cause distortion. They are used at the output of a transmitter to raise its
power level. As discussed in Lesson-2 of Part-2 a modulated wave must be amplified in linear
fashion else other side frequencies and harmonics are generated. The design of a non-distorting
Class-A amplifier does not pose many problems but, at the output of a transmitter, there is a
requirement for power. The lack of efficiency in a Class-A stage poses a problem in heat
dissipation and so Class-B or Class-AB is a more likely choice, The subject is not examined for
the R.A.E.

Before leaving the subject of rf-amplifiers mention should be made of the arrangements at
the output of a transmitter P.A. stage for feeding the developed rf-power to the aerial. Especially
in semiconductor equipments the standard output today is a 50-ohm coaxial connector of some
sort which, of course, is intended to house the end of a coaxial aerial-feeder.

Coax however is expensive, relatively lossy and short-lived and a popular feeder used to be
the 300-ohm balanced open-wire which brings problems of its own. Unbalanced coax outputs
often use a simple RC coupling, especially where power levels are low, but for high-power and
balanced circuits a transformer is used.

It is possible to find balanced (push-pull) amplifiers driving an unbalanced feeder or

alternatively an unbalanced amplifier driving a balanced feeder and it is necessary to find some
means of isolating the circuits if the balance is to be preserved. In Fundamentals-1 Lesson-4
mention was made of the Faraday screen which provides electrical isolation while permitting
magnetic coupling. In a P.A. stage such a screen takes the form of an earthed wire-grid placed
between the primary and secondary windings of the output (air-cored) transformer. (In a.f.
amplifiers it takes the form of a non-closed turn of copper foil wound over the inner winding
before the outer winding is added.)

>>>>>>>>>>>>>>>>>>>> PAGE 28 <<<<<<<<<<<<<<<<<<<<

END OF LESSON 4

*******

QUESTIONS

1. Why is good quality sound-amplification required in transmitter circuits ?

2. What is an RC-coupling? Why does it cause distortion ?

3. What is the difference between small-signal amplifiers and large-signal or power amplifiers
?
4. Why arid how is a Class-A amplifier biased ?

5. In what ways do AF and RF amplifiers differ ?

6. Why is screening important in rf-amplifiers ?

>>>>>>>>>>>>>>>>>>>> PAGE 28 (again) <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 5
3.13.3 IF Amplifiers

As already mentioned an if-amplifier is essentially an rf-amplifier but with a specific

purpose. It amplifies equally over a specified band of frequencies the centre frequency of which
does not vary. It achieves its bandpass characteristic by using multiple tuned circuits (or other
resonators), coupled together, in place of the single variable-tuned circuits found in normal rf-
amplifiers. The term "if-amplifier" is used in reference to both a single stage of amplification
and to an entire amplifying unit of several if stages.

** Fig. 18 shows three versions of a basic single-stage if amplifier which uses double-tuned
rf-transformers as impedances. The stage uses automatic emitter (source or cathode) bias and the
input is returned to ground via the dc resistance of the driving coil.

In diagram (a) the bipolar transistor offers a very-low input impedance which would damp
the transformer secondary and destroy its Q-value. This is corrected by tapping down the coil
which is thus doubling as an autotransformer (see Fundamentals 1, Lesson5). In diagram (b) the
FET offers a very-high input impedance and so the tapped coil is not necessary at the gate
connection.

Diagram (c) shows the same circuit but using a valve. This too offers a high input-
impedance but the valve itself is also a high-impedance device and so the performance is
susceptible to the anode-grid capacitance which forms a feedback path between anode and grid.
This is overcome by the addition of a screen-grid between the two electrodes which is earthed
to signal currents so that it acts as a Faraday screen (see Funds.-1, Lesson-3). Were this screen
to be earthed literally it would repel the electron stream which flows from the cathode to the
anode ; to maintain the flow of cathode-anode current, the screen is raised to a high potential by
connecting it through a dropping resistor to the positive supply line.

The dropping resistor and the decoupling capacitor (which earths the screen-grid to ac-
signals) form an RC network ; to prevent a signal-voltage developing on the screen the time-
constant RC must be large compared to the period of the signal waveform;

>>>>>>>>>>>>>>>>>>>> PAGE 29 <<<<<<<<<<<<<<<<<<<<

i.e. RC = 5 x 1/f where R is in ohms, C is in Farads and f is in Hertz.

Most small-signal rf valves are fitted also with a third grid making them into pentodes(5
electrodes). The main purpose of this grid is to intercept electrons that may be ejected from the
anode ; should these return to the screen the valve may oscillate. This suppressor grid
is normally connected directly to the cathode.

** As with rf-amplifiers, the if-amplifiers of a receiver are connected to the a.g.c. line to
maintain a constant audio level. However, the final if-stage receives a large signal-input, it must
not distort that signal and it is required to supply a certain amount of power to drive the
demodulator stage. It is usual either to leave this stage without a.g.c. or to reduce the a.g.c.
input.

** For f.m. receivers a.g.c. is not used. The final if-stage is operated as a limiter so that its
output drives the demodulator-circuit (the discriminator) with a signal which is free from
amplitude modulation. Generally the limiter is driven hard into its limiting mode which action
suppresses most of the (a.m.) noise and so provides the high-quality audio signal which is the
major asset of f.m. working.

(For those with a flair for such things the limiter produces a bias voltage as part of its action and,
provided the if-unit affords sufficient gain, this can be fed-back as an a.g.c. signal.]

3.13.4 Bandpass Circuits

The theory of coupled circuits is not required for the R.A.E. A simple explanation is based
on the fact that a capacitive reactance at the input of a transformer is coupled into the secondary
circuit as an inductive reactance and, similarly, an inductive reactance appears in the secondary
as a capacitive reactance.

When the applied frequency is below the resonance frequency both tuned circuits present a
capacitive reactance ; thus each circuit reflects an inductive reactance into the other and this
tends to preserve the resonant condition. Similarly, when the applied frequency is higher than
the resonance frequency, both circuits present an inductive reactance which is reflected as a
capacitive reactance into the other.

** To achieve the best result, in which the gain remains constant over the desired band of
frequencies, it is necessary to obtain an exact degree of coupling between the two circuits. This
particular value is known as critical coupling and its value depends on the inductance of each
coil and on the Q of each tuned circuit of the pair.
 The initial alignment of such a transformer is reasonably simple. First the two coils are
pushed as far apart as possible to minimise the coupling i.e. to ensure that neither detunes the
other through reflected reactances. Each circuit is then tuned to resonate at the required centre
frequency and the coils slowly pushed together until the required "flat-topped" characteristic is
obtained. The coils are then cemented into place.

>>>>>>>>>>>>>>>>>>>> PAGE 30 <<<<<<<<<<<<<<<<<<<<

Sometimes it is not feasible to move the coils (a commercially-supplied unit may have the
coils already cemented). Any attempt to resonate the second coil then detunes the first one
already set-up. The procedure is to connect a low-value damping resistor across one of the coils
thus destroying its Q-value and so rendering it almost non-reactive. The other circuit is then
correctly tuned. Next the damping resistor is transferred to the second coil thus reducing its
reactive component which enables the remaining circuit to be correctly resonated. Finally the
removal of the damping resistor will leave the circuit properly aligned assuming that the
coupling was correctly set.

These alignment procedures are not examined for the R.A.E. but the above brief study
may help to understand the bandpass arrangement.

The double-tuned transformer is not the only bandpass circuit that can be used but it is the
simplest, the cheapest and the most compact. There are occasions however when its
compactness can be a problem in that, with the requirement at high frequencies to keep
connections short, the designer runs out of room. On such occasions it can be helpful to use two
entirely separate coils in separate screening cans and to couple them via a small adjustable
capacitor ; this enables the circuit to be spread out without incurring problems with stray
inductances.

There is a direct relation between bandwidth, the degree of coupling and the Q of each tuned
circuit. Often the active devices do not provide the amount of loss required to achieve a desired
bandwidth and so extra loss must be provided In the form of damping resistors. These may be
connected across each bandpass coupled circuit either as a single resistor of appropriate value or
divided into two separate resistors one of which is connected across each coil of the pair.

There is a purely practical pont here that involves the difficulty of constructing
coils/transformers of exactly the required parameters. Under manufacturing conditions where
many hundreds of units are to be constructed this can be dealt with but, for a one-off job, it is
easier to aim for a too-high Q and add damping resistors on a suck-it-and-see basis.
 Critical-coupling results in a flat response
as shown in Fig. 19. Coupled circuits are
usually adjusted with slightly more than critical
coupling because it gives a slightly increased
passband and slightly steeper skirts at the band-
edges. As shown in Fig. 19it also produces a
double-humped characteristic in which the size
of the bumps is easily controlled by adjusting
the coupling factor.

Note that a single tuned circuit produces a

single-peaked response, a double over-coupled
circuit produces two peaks and a triple over-
coupled circuit produces three peaks.

It is possible to produce bandpass filters which consist of many LC circuits magnetically-

coupled as described above; such filters are to be found for example in the Racal RA17
receivers. Unless you have experience in adjusting such circuits and the necessary equipment it
is foolish to interfere with these because any alteration to one of the LC combinations detunes
all the others. This type of filter is an exact analogy of the Yagi "beam" aerial which consists of
a string of resonant rods spaced at critical distances (for distances read coupling-factors). The
above discussion of the multi-unit bandpass filter will perhaps make clear why the claims made
for Yagi aerials should be taken with a pinch of salt.

>>>>>>>>>>>>>>>>>>>> PAGE 31 <<<<<<<<<<<<<<<<<<<<

The criteria by which such circuits are designed are very simple but not required for the
R.A.E. If you have the necessary equipment and wish to construct your own if-unit I will
supply the information on request.

Warning Many domestic receivers use stagger-tuned circuits in the if-amplifier section
because of the easier alignment. Before attempting to "peak-up" an if. unit make sure that it is
using coupled circuits and not stagger-tuning. Adjustment of trimmer capacitors in the latter
results in a rapid increase of gain and the onset of self-oscillation as each stage tries its hand as a
T.A.T.G. oscillator (see Part 4). A good designer will ensure that, when a coupled circuit is
adjusted for critical coupling, the gain per stage is limited to a safe value.

** Communications receivers have a requirement for variable bandwidth and this will be
incorporated into the if unit because it is that section of a receiver which provides most of the
selectivity and gain and it operates at a fixed frequency. The range varies from some 3 kHz for
ssb down to 300 Hz for cw working ; a receiver intended to reproduce broadcast channels also
may need to extend to ±6k Hz or more.

In elderly valve-receivers the if transformers were multi-wound and bandwidths were

changed by switching between tapings and coils. Narrow bands for cw were obtained by using
various forms of crystal filters or gates (see Lesson-15 of Part-2).
 ** Modern technology however has produced a range of crystal and mechanical filters which
are often superior in performance to coupled circuits especially when price and stability are
taken into account. An if-amplifier can be designed therefore as a straightforward rf-amplifier
and proprietary filter units switched-in to achieve the desired bandwidth.

3. 13.5 Video Amplifiers

Video amplifiers are not examined for the R.A.E. and are of interest only to the amateur who
is intending to take up amateur television or computer work but a short discussion is included
here for completeness and general information.

A video amplifier handles picture signals and must cope with a frequency range which
extends from zero (d.c.) up to several MHz where the upper limit is determined by the picture
definition ; this term refers to the size of the detail which is to be displayed. The old 405-line
system required an upper limit around 3 MHz while the modern 625-line system requires at least
5 MHz. Greater definition, usually referred to by the number of lines into which a picture is
analysed, requires a still-higher upper limit while the so-called low-definition and slow-scan
(SSTV) systems use less spectrum.

Stray capacitances become a positive menace because tuned-circuits cannot be used in place
of impedances. When the valve was the only practical amplifying device many complex circuit-
structures were devised to extend the range of a straightforward amplifier and these techniques
were made necessary because of the high impedances imposed by valves.

>>>>>>>>>>>>>>>>>>>> PAGE 32 <<<<<<<<<<<<<<<<<<<<

With the advent of transistors the problem was reversed. Circuit impedances are very low,
often no more than a few hundred ohms, and so stray capacitances could be neglected. The
problem became one of limiting bandwidth to the upper limit required. Out-of-band gain is
reduced to remove unnecessary noise. Such bandwidth restriction is usually achieved by
employing negative-feedback loops whose frequency-response is weighted so that amplification
in the main signal-path is reduced outside the required frequency range.

A large part of the difficulties with video amplifiers stems from the need to encompass zero
frequency. Actually the lowest frequency component occurs at half the picture-repetition rate
which, in the UK, means 25 Hz (one-half the frequency of mains-induced hum to which the
picture-rate is locked). Amplifier design can be greatly simplified by limiting the passband at
25-Hz and then making provision to restore the dc-component. The two techniques used are
known as dc—restoration and (much superior but more expensive) clamping. These
arrangements are not dealt with in this Course but information can be supplied to any who
enquire.

3.13.6 Demodulators

These fall into three general classes ; (a) a.m. (b) f.m. (c) ssb. There are other circuit
techniques used but they are not of interest in this Course.

A. M Demodulators or Detectors
 ** As discussed in Lesson-14 of Part-2 an a.m. signal is demodulated by using a rectifier to
remove one-half of the rf-waveform ; the unidirectional amplitude-modulated pulses which are
left have an average value which varies with the modulation.

** One of the components which result from detection is a direct voltage whose amplitude is
proportional to the carrier amplitude (without modulation) and this can be used to operate a
simple a.g.c. system. However a simple a.g.c. system biases-back the receiver gain whenever a
signal is detected and this is a serious handicap when trying to read a weak signal. The problem
is solved using a delayed-a.g.c. in which the detector diode is biased so that it does not conduct
until the signal level exceeds the bias voltage. Note that, if the delay voltage is too large, the
a.g.c. action will not cope with weak-signal fading.

** In terms of audio output a biased diode is a disaster ; until the signal level reaches the bias
value there cannot be any audio output at all. Thus it is necessary to use two separate diode
detectors. It is usual to feed the a.g.c. detector from the primary winding of the final if-
transformer because, at this point, the bandwidth is greater than at the output from the secondary
winding. This ensures that, when "tuning-through", the a.g.c-action holds down the receiver gain
until any signal has substantially disappeared thus avoiding an unpleasant tuning effect known
as sideband shriek>

>>>>>>>>>>>>>>>>>>>> PAGE 33 <<<<<<<<<<<<<<<<<<<<

In valve circuits these two diodes were incorporated into a single envelope and the valve was
known as a double-diode. In a further development the two diodes were combined with the first
audio amplifier to form either a double-diode-triode or a double-diode-pentode. The audio
amplifier was then utilised, by making it direct-coupled to the detector, to achieve amplified-
delayed-a.g.c Valve circuits are not recognised for the R.A.E. and they will not be dealt with
further ; many examples of these circuits can be found in the literature however.

(It is perhaps worth noting that a European directive requires that the R.A.E. should examine
valve circuitry because a large proportion of amateur rigs still employ "old" technology.)

** The output from a diode detector is a string of amplitude-modulated pulses and these have
to be smoothed into a varying dc in much the same manner as described for dc power-suppliers
in Part-5. The diode load consists of a parallel combination of capacitance and resistance
(Rl.C1) as shown in Fig.20. During that carrier half-cycle when the diode conducts the
capacitor is charged to the peak carrier voltage ; during the following half-cycle the capacitor
discharges slowly through the resistor (or leak). This ensures that the direct-voltage across the
capacitor follows the carrier amplitude as it increases and decreases with the modulation
waveform.

** This varying voltage is resolved into a

direct component, representing the amplitude
of the unmodulated carrier, and an alternating
modulation voltage. Capacitor C2 blocks the
dc but passes the AF to the Volume control
VR2. Resistor R3 and capacitor C3 form a
low-pass filter that permits only the direct-
voltage to pass to the a.g.c. line.
 ** As always the values of the different components has to be a compromise between
conflicting requirements. If the leak resistor Ri is made too small the capacitor will discharge
too much and so cause an if-frequency ripple to appear and the audio output may not follow the
increases in carrier amplitude. If the leak is made too large the capacitor will discharge too
slowly for the output voltage to follow the decreases in carrier amplitude. Capacitor C2 has to be
large enough to pass the lowest audio frequency but not so large that it passes also the variations
in direct-voltage caused, for example, by signal fading. The time-constant of C3,R3 is less
critical but it must remove both the rf and audio frequencies from the a,g.c. line without
reducing the slower variations caused by fading.

** Diode detectors produce distortion of the modulation waveform because the load presented
to them varies with frequency. For the dc component of the diode output the diode load consists
of the resistor R1. For the audio component the load consists of R 1 in parallel with VR2 for
those frequencies which pass C2. At lower frequencies the coupling capacitor introduces phase
shift also. This difference in loading leads to waveform distortion when the modulation index
approaches 100% and this is the reason why a.m. transmitters are adjusted for a maximum
modulation depth around 80%.

A detector accepts an rf-signal input but produces an output at audio frequencies. The
manner in which it converts one to the other is often

>>>>>>>>>>>>>>>>>>>> PAGE 34 <<<<<<<<<<<<<<<<<<<<

expressed as a transfer characteristic. For distortion-free detection this transfer-characteristic

should be linear (i.e. a straight line which indicates that the output is directly proportional to the
input). For very-low input signals, which are only just able to drive the diode into conduction,
this characteristic tends to a square law ; i.e. the audio-output varies as the square of the if
input. In this mode the arrangement is referred to as a square-law detector. To ensure linearity
the diode should produce about 1-volt which, apart from selectivity, is the main reason for pre-
detector amplification.

** For a communications receiver there is one other facility required and that is an A.G.C.
On/Off switch. It can be useful to remove the a.g.c. when a signal, although weak, nevertheless
exceeds the delay voltage and reduces the gain. The main purpose of this switch however is that,
in cw mode, the B.F.O. signal generates a substantial a.g.c. voltage which obviously precludes
the reception of any but the strongest signals.

END OF LESSON 5

************

QUESTIONS

1. What are the differences between an rf-amplifier and an if-amplifier ?

2. In modern circuits why are the connections to active devices taken from (and to) tapings on
the coils ?
3. What is the purpose of a.g.c. ? In which particular circuits would you not expect to find such
a facility ?

4. In the text (page 31) it is suggested that if-transformers are tuned by means of trimmer
capacitors. How else could tuning be achieved and can you point out a problem that could arise
?

5. What is critical-coupling and why is it not necessarily the ideal ?

6. A bandpass coupled-pair is constructed from two separate coils which are coupled through a
small variable capacitor. At each end the capacitor connects to a tapping one-quarter of the way
up each coil from the earthy end. What would be the effect on the capacitor if both tapping-
points were moved to the centres of the coils ? (NOT for R.A.E.]

7. Why do communications receivers require an A.G.C. On/Off switch ?

8. What is the purpose of "delayed a.g.c." ?

9. In a diode detector circuit why would you suspect that the load-resistor had become open-
circuited ? How could you check this using a meter ?

>>>>>>>>>>>>>>>>>>>> PAGE 35 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 6
3.13.7 FM Demodulators (Discriminators)

** As already described a discriminator first puts the frequency-modulated signal through a

frequency-conscious network which transforms it into a carrier that is simultaneously frequency-
modulated and amplitude-modulated. The audio waveform is then extracted with a diode as is
done with an a.m. signal.

Perhaps the simplest way to achieve this is to detune an rf-amplifier until the carrier
frequency lies about half-way down the skirts of the frequency characteristic (see lower diagram
in Fig. 19 of Lesson-5). With the receiver adjusted in this manner the rf output increases as the
carrier-frequency moves toward the tuned-circuit resonance but decreases as the carrier-
frequency swings in the opposite direction.

An fm transmission can be read in this fashion if all else fails but it is seldom very
satisfactory. (Switch-off the a.g.c.!) The obvious problem is the lack of linearity ; i.e. the lack of
a direct relation between the carrier excursion (or deviation) and the amplitude of the output
signal because the characteristic is curved.
 The next simplest consists of
two separate tuned circuits one
tuned to the desired upper
frequency-limit and the other
tuned to the desired lower
frequency-limit. As shown
in Fig.21 each circuit has its own
diode detector but they are
connected, in opposite sense, to a
common load.

The outputs from the diodes cancel and so, when the carrier frequency passes through its un-
modulated value fc , the output voltage developed across the load-resistor R is zero. When the
carrier-frequency swings toward the resonance-frequency of the upper tuned-circuit then the
positive output from diode Dl is greater than the negative output from diode D2 and so the
output is positive. Conversely, as the carrier-frequency swings toward the resonance-frequency
of the lower tuned-circuit, then the negative output from diode D2 predominates and so the
output is negative. The overall effect is illustrated in the graph alongside.

This type of circuit does not produce a dc-component which represents the carrier amplitude
because, at the centre frequency, the diode outputs sum to zero.

This simple arrangement produces a much better linearity than the de-tuned rf amplifier
described above but it is not easy to get it to align so that it gives both good linearity (low
distortion) and also a reasonable sensitivity. However the arrangement has been used in refined
form to demodulate f.m. signals in the uhf band and above and it was once used as a cheap
sensing circuit to drive motorised tuning devices.

>>>>>>>>>>>>>>>>>>>> PAGE 36 <<<<<<<<<<<<<<<<<<<<

** Discriminator circuits such as the Foster-Seeley and the Ratio Detector provide much
better linearity . The ratio-detector is the more common these days because it provides
discrimination against carrier-amplitude variations (a.m.) ; for this reason it is not unusual to
see a ratio-detector used without a limiter stage.

** The ratio-detector is shown in Fig.22

; it is easily distinguished from the Foster-
Seeley by the reverse-connected diodes. Both
discriminators derive quadrature samples of
the incoming carrier and produce signals
which are the sum and the difference of these
two.

** The final if-transformer is double-tuned

in the normal bandpass arrangement but the
secondary is centre-tapped. A portion of the
primary voltage is tapped-off by a tertiary
winding ( a capacitor could be used) and fed
 to the centre of the secondary. Either
coupling component provides a 900 phase-
shift. Thus the two diodes are fed (a) with
primary-signal plus secondary-signal and (b)
with primary-signal minus secondary-signal.

** At the centre-frequency, where both tuned circuits are resonant, the outputs from the two
diodes are equal, the load capacitors Cl and C2 are charged to equal voltages and a direct
potential is developed across the two resistors R1 and R2. In operation this potential changes
very little and so a large-value capacitor (2 to10 µF) can be connected across the circuit and it is
this that provides the immunity to carrier-amplitude changes.

** As the carrier-frequency swings about the resonance-frequency so the phase of the

secondary signal shifts. This causes the voltage applied to one diode to increase as the voltage
applied to the other decreases. As a result the charges on Cl and C2 vary and the potential of
their centre point varies with the modulation waveform. The Ratio Detector derives it name
from the fact that the audio signal depends on the ratio of the voltages across the two load
capacitors.

** The audio signal can be obtained across either Cl or C2 but such an arrangement would
upset the rf balance of the circuit. It is better to tap the resistive load at its centre (hence R1 =
R2) to provide the earth-connection. (Note that the circuit as drawn in Benbow's RAE Manual is
not wrong but is drawn deceptively ; the upper of his two audio connections is earthy.]

N.B. You should understand the ratio-detector but you will not receive any searching questions
about the details of its operation.

>>>>>>>>>>>>>>>>>>>> PAGE 37 <<<<<<<<<<<<<<<<<<<<

3.13.8 Detection of SSB Signals

** The term ssb is the accepted form for suppressed-carrier single-sideband transmissions and
the first step in recovering the modulation information is to re-insert the carrier. This must be
achieved with an accuracy of at least 25 Hz.

** In an ssb receiver the required carrier signal is generated in a carrier-insertion oscillator

(c.i.o) but, as already discussed, it is possible in a communications receiver to use the b.f.o. for
this purpose. The carrier level required to demodulate an ssb signal however is fairly large while
that required from a b.f.o. for cw reception is small. To demodulate ssb successfully with a
normal communications receiver it is necessary to juggle with the RF-Gain, Aerial-Attenuator
and AF-Gain controls; where they are fitted the Bandwidth and IF-Shift can be useful too.

As discussed in Lesson-3 of Part-2 there are two methods of heterodyning signals namely
add them or multiply them. Simply adding two signals (or mixing them) results in a beat which
has to be extracted (or detected) by use of a rectifying diode. This is the method used when the
bfo is employed in a normal communications receiver. Multiplying one signal by another yields
the sum and difference frequencies directly and this is the method used in the so-called Product
Detector.
 ** A basic circuit for a product detector
is shown in Fig. 23(a). Essentially this is a
simple common-source amplifier stage
with separate inputs to the gate and source.
The current which flows between source
and drain is controlled by both input
signals with multiplicative effect. The
audio output is taken from the drain via an
rf-filter.

** The Ring Modulator shown

in Fig.23is an example of additive mixing.
The four diodes are connected in a
continuous ring formation (compare this
with the bridge-formation in Part-5). The
circuit is exactly the same as the Ring
Modulator used in transmitters to achieve
ssb but here it is used in a "reverse" sense.
Instead of putting-in an audio and a carrier
and recovering a ssb signal, here we put-in
a ssb signal and a carrier and recover the
audio.

Analysis of these circuits is not

required for the R.A.E. but, for those who
are curious, the easiest way to understand
them is to regard the diodes as switches
which are operated in pairs by the inserted
carrier. The ssb signal is directed through
the audio load in a series of pulses at if-rate
which are smoothed out by an rf-filter into
the audio waveform.

>>>>>>>>>>>>>>>>>>>> PAGE 38 <<<<<<<<<<<<<<<<<<<<

** The fact that a ring circuit can be used for both modulation and demodulation means that it
is often found in Transceivers where it is switched between the two modes for either
transmitting or receiving purposes.

** The product detector too has its advantages because it can demodulate both ssb and cw
signals with the same low-level input from a local oscillator.

3.13.9 Beat-frequency Oscillator

** The b.f.o. operates over a small range of frequencies centred on the i.f. Frequency-
adjustment is by means of a front-panel control which enables the operator to set the audio
output frequency either to a preferred note or so that he can the more easily discriminate
between a wanted signal and an unwanted signal.
 ** This second reason is the main purpose of the b.f.o. control. Rocking the receiver main-
tuning control affects the frequency of the carrier delivered to the i.f. section (from the
frequency-changer) and so, in general terms at least, a fixed-frequency b.f o. is sufficient as far
as the audio note is concerned.

** The b.f.o. control normally gives some 3-kHz adjustment both above and below the
intermediate frequency. This provides two possible settings for the b.f.o. to produce a given
audio output but, while one of these settings produces a high-frequency audio beat with an
interfering signal, the other setting must produce a low-frequency beat with that interfering
signal. Of course, if the interfering signal is exactly on the frequency of the required signal, it is
not possible to separate them.

Skilful use of the b.f.o.-control, of the bandwidth-control, of the if-shift control and of the
main tuning (in combination) can often eliminate an unwanted cw signal.

** Any of the oscillator circuits described in Part-4 of this Course can be used for the b.f.o.
Generally it is desirable to keep its output at as low a level as possible else it generates a large
a.g.c. signal and so reduces the receiver gain ; an a.g.c. switch is an alternative here. The
oscillator should be well screened, decoupled and filtered to prevent it adding its quota to the
traditional family of howls and squeals that can beset superhet receivers.

3.13.10 Noise Limiters

You are not required to produce circuit details of noise-limiting circuits for the R.A.E. which
is just as well because most of them seem to be little more than decorative knobs for the front
panel.

** Most intrusive noises consist of large-amplitude and wide-spectrum spikes caused by such
things as lightning, car-ignition circuits, switching surges and "splashes" or the continuous noise
from electric brush-motors.

>>>>>>>>>>>>>>>>>>>> PAGE 39 <<<<<<<<<<<<<<<<<<<<

** Spikes are dealt with by using biased diodes to clip-off that part of the spike which exceeds
the amplitude of the wanted signal ; these are known as clipper circuits. The idea is that, when
the signal+noise exceeds the bias voltage, the diode is driven into conduction and reduces the
circuit gain. That part of the spike which is left embedded in the signal cannot be dealt with.
The clipping-level can be adjusted by making the diode bias variable.

To describe the action of these clippers is often a matter of preference ; sometimes the
conducting diode provides a shunt path that diverts current from the input to a transistor,
sometimes it connects a secondary resistor in parallel with a load resistor and so reduces the
circuit gain, sometimes it connects the signal circuit to a low-impedance (a constant-voltage)
direct-voltage source so that, for the duration of the pulse, the signal amplitude is held constant
; this last circuit-arrangement is often referred to as a clamp.

There was a pulsed form of interference known as "woodpecker" which seemed to stem
from some sort of Russian radar and which presumably will not be heard again now that Russia
has undergone a change of policy. I have not discovered how the anti-woodpecker noise-
limiters are supposed to work which probably explains why they have never worked for me.

Continuous noise such as that caused by arcing at the brushes of rotating machinery cannot
be dealt with by noise limiters in radio receivers. However, if the interference is severe in that its
peaks exceed the amplitude of the signal, then of course its effect can be mitigated by a diode-
clipper circuit. Motor noise must be dealt with at source as described in Part-6 of this Course.
Often a cure (at source) can be effected, where arcing is occurring at switching points, by use of
the constant-resistance network which is described in Part 8: Information Sheets; No.10.

3.13.11 Squelch Circuits

** Again circuit details are not required for the R.A.E. Especially in f.m. receivers the output
consists of a large-amplitude hiss when a carrier is not present at the input. This is because an
a.g.c. signal is not being generated or the limiter stage is not in operation and so the receiver
operates at full gain.

** A squelch circuit is intended to obviate the nuisance of such noise by using a diode with
variable bias (or some similar "threshold device") to shut-off the receiver audio circuits until a
carrier appears. (Note that the squelch cannot operate on the rf-circuits.) The squelch bias is set
so that the receiver "opens-up" on receipt of a signal.

Some people seem to think that a squelch can be adjusted so as to reduce the background
noise level when working with a very weak signal but this is not true. The squelch needs to have
an adjustable threshold to prevent a weak unwanted signal continually opening the receiver but a
wanted signal has to be either strong enough to operate the squelch or too weak to do so. A
weak borderline fading signal may in fact trigger the squelch intermittently and then the only
cure is to switch-off the circuit and rely on the ear to sort the signal from the noise.

>>>>>>>>>>>>>>>>>>>> PAGE 40 <<<<<<<<<<<<<<<<<<<<

3.13.12 Signal-strength Meters

** Nearly all communications receivers have a front-panel S-meter whose primary function is
to indicate the received signal-strength. As a secondary function it serves to indicate the point
of accurate tuning, when the signal-strength has reached a maximum value, but this is often a
snare and delusion because the signal is fading. In transceivers the meter doubles also as a
monitoring device which aids in transmitter adjustments.

For fm reception a tuning meter needs to operate with a centre-zero to indicate the correct
tuning point so that, when the carrier is at its unmodulated frequency, the meter does not deflect;
see Fig. 21. The meter must be switched from the discriminator output to measure the bias
developed in the limiter stage if an indication of signal-strength is required.

The range of input signal to a receiver can be enormous ranging from a few microvolts up to
even a volt if a powerful transmission is arriving from a nearby location. To compress such a
range within the scale of a meter it is necessary to use a logarithmic law (see Part-7:
Mathematics). This means that the meter-deflection is not proportional to the signal strength but
to the logarithm of the signal-level change and so the scale is calibrated in terms of decibels.

However it is not necessary to use expensive meters whose magnetic fields have been
tailored to a logarithmic law because the receiver itself supplies the necessary characteristic
when a.g.c. is in use. Note that an S-meter does not function when the a.g.c. switch is set to
OFF..

The a.g.c. signal causes the current drain of the receiver to fall and, if this current is passed
through the meter, an indication of signal strength is obtained. Unfortunately the meter reduces
its deflection as signal strength rises. There are many circuit arrangements which can be used to
reverse the meter action but an alternative is to arrange for the meter to indicate the magnitude
of the a.g.c. signal.

The a.g.c. characteristic seldom shows a good logarithmic law except for a part of the signal
range and, for this reason, an S-meter is not a reliable indicator of incoming signal strength.
There is no reason of course, if you have a suitable signal generator, why an S-meter cannot be
calibrated.

In high-quality equipment however you may find circuits which tailor the meter response to
improve the law. That part of a diode characteristic where the current is low (just after
conduction begins) is usually a very good logarithmic law ; an improved characteristic can be
obtained just by shunting a meter with a diode and, probably, by adding a suitable resistor in
series either with the meter or with the diode.

The circuit which drives the meter from the a.g.c. line may also be loaded with a series of diodes
with graded biases so that, as the signal-strength increases, the diodes progressively come into
conduction and modify the meter response.

>>>>>>>>>>>>>>>>>>>> PAGE 41 <<<<<<<<<<<<<<<<<<<<

Sometimes a.g.c. may be taken from the audio output. In such an arrangement a bridge
rectifier (see Part-5) is used to obtain a dc-signal from the af-signal. This does not materially
affect the action of any S-meter.

3.13.13 Decoupling

** When two circuits share a common impedance, such as the internal-resistance of a power-
supplier or battery, then they are coupled together by that common-impedance and unfortunate
feedback effects can occur. For example, a large-signal stage which is drawing a large signal-
current can develop a signal-voltage across the power-supplier. When that supplier is also the
auxiliary supply for an earlier smell-signal amplifier then an amplified signal is fed-back from
the output-stage to the small-signal input stage and oscillation can result.
 ** Fig. 24 shows the above arrangement in
block form with a stage of decoupling inserted
to isolate them against this kind of feedback.
The time-constant C1R1 seconds (C in Farads,
R in ohms) must be large with respect to the
period(time for one cycle) of the lowest
frequency which can pass through the
amplifier.

More decoupling is shown also in the a.g.c.

line to prevent unwanted feedback via this
route.

** Briefly the capacitor provides a-low-

impedance path to ground for any feedback
signals while the resistor raises the impedance
of the alternative path.

3.13.14 IF-shift and IF-notch Circuits

Many modern receivers are fitted with a notch filter whose rejection frequency can be
adjusted by means of a front-panel control. Its use in reducing an unwanted signal in cw-
working is obvious. The IF-shift control is perhaps a little more mysterious. It affords a means
of sliding the if bandpass-characteristic along the spectrum so that, for example, an interfering
signal can be "pushed over the edge" of the pass band and so attenuated without affecting the
wanted signal. It can be very effective when used in combination with the notch filter.

In practice the if-passband is not itself moved. The control drifts the local-oscillator
frequency so that all the signals that emerge from the frequency-changer within the if band are
moved with respect to the passband. It matters not whether the group of signals or the passband
is moved; the effect is the same.

>>>>>>>>>>>>>>>>>>>> PAGE 42 <<<<<<<<<<<<<<<<<<<<

3.13.15 Active Filters

An Active Filter is one which includes an active device because this enables it to perform in
much the same manner (or even better) than an LC filter but without the bulky (and lossy)
presence of inductors. While it is convenient to talk about "artificial" inductors there is in some
instances a subtle difference. Operational-amplifiers (Op-amps) are used in conjunction with
capacitors to produce inductor-like effects in that the current lags the applied voltage ; such a
circuit can resonate with an external capacitor to produce the many different forms of LCR
filters.

A circuit arrangement known as a Gyrator converts a capacitive reactance to a genuine

inductive reactance ; alternatively the negative-impedance converter changes a capacitive
reactance to a negative-inductive reactance which means that, although the current lags the
voltage as required of an inductor, the reactance of the circuit falls as frequency increases in the
manner of capacitive reactance.

Active filters can be made in low-pass, high-pass, bandpass, band-stop and tunable forms
and are particularly useful in audio circuits. In addition an "all-pass" filter can be made ; this
offers a flat frequency-response but controls phase-shift (also known as a delay-equaliser).

3.13.16 Modern Operational Amplifiers

The Op-amp is manufactured as a chip, smaller in size than a postage stamp, with an
enormous open-loop gain (i.e. gain before the application of feedback). Op-amps come also
in dual and quad forms in which a single chip offers either two or four completely separate and
often matched op-amps. With gains up to one-million-times between two points less than a half-
inch apart stray couplings can spell trouble and so circuit lay-out can be as important as the
electronic design.

They are offered as "general-purpose" types and also designed for specific purposes such as
high-quality audio amplifiers. Their construction may use standard bipolar-transistor technology
or the more fragile MOSFET technology which features very-low power consumption ideal for
battery-operated equipment.

They are usually operated from a split supply( i.e. between a positive-rail and a negative-
rail) but some types are capable of operating from single-ended supplies and some obliging
types can cope with either.

They all have three features in common:

(a) they are dc-coupled

(b) they are intended to be used with large amounts of negative-feedback

(c) they have two differential input terminals labeled " + " and " — " but this does not indicate
electrical polarity. An input-signal applied to the " + " terminal appears at the output in the
same non-inverted form while the same signal applied to the " — " terminal appears at the
output inverted.

>>>>>>>>>>>>>>>>>>>> PAGE 43 <<<<<<<<<<<<<<<<<<<<

Note that, if a signal is applied simultaneously to both input terminals, then cancellation occurs
and the output is zero; this is exploited in some audio op-amps to provide a control of signal
amplitude.

The internal circuitry of these op-amps is very complex and a "gain" control as such is not
available except by variation of the external feedback circuit.

Detailed knowledge of op-amp operation is not required for the R.A.E. However the
devices feature in many of the projects which appear in amateur-radio publications and a short
description of their basic operation follows.
3.13.17 The Effects of Feedback on Op-amp Behaviour

The internal circuitry of op-amps employs 20 or more transistors so that the overall voltage
gain is normally in the range 105 to 106 . This means that it requires only a few microvolts
(applied between the differential inputs) to swing the output terminal between the extreme limits
of the power supply.

When negative-feedback is applied to an op-amp these enormous gains mean that the
behaviour of the circuit is controlled almost entirely by the characteristics of the feedback loop.

Fig. 25 shows an op-amp connected as

an inverting amplifier which means only that,
whatever else it does to the signal, it also turns
it "upside down". Note that the standard
symbol does not show the connections from the
chip to the positive and negative rails of the
power-supplier.

The resistor R 2 is the feedback path through

which the output voltage drives a current which
cancels the input current arriving through R 1 .

The input-voltage developed between the

two input-connections is only a few microvolts
and is negligible when compared with the
voltage developed across R 1 . For practical
purposes therefore the input-voltage appears
across R 1 and so this resistor defines the input-
impedance of the amplifier.

For those with a bent for algebra it is not difficult to show that the voltage-gain of this circuit
is given by the ratio R 2 /R 1

In stating that the input voltage between the " + and " — " terminals is negligibly small we
are saying that the feedback action (in practical terms) ties the two inputs together. With the
" + " terminal taken to earth it follows that the"-" terminal is also very close to earth potential
and this is described as a virtual earth.

When such an amplifier is required to provide a fairly high gain the value of R 1 becomes
small and consequently the input-impedance is small.

>>>>>>>>>>>>>>>>>>>> PAGE 44 <<<<<<<<<<<<<<<<<<<<

 Input impedances up to 1012 ohms or higher
are possible with the non-inverting amplifier
shown in Fig. 26.

For both arrangements the output impedance

is a fraction of an ohm.

The above can be summarised as two rules:

1. The operation of the feedback loop ensures that the output settles at a value which brings the
two input terminals to (nearly) the same potential.

2. The small pd developed between the two input terminals means that the input current (to the
op-amp proper) is (almost) zero.

Op-amps, op-amp circuits and their applications are multitudinous and beyond the scope of
these Lessons. However, should any reader be presented with a problem, I will be pleased to
advise.

END OF LESSON 6

*******

QUESTIONS

1. Describe briefly how a ratio-detector demodulates an fm signal.

2. What would you expect a product-detector to do

3. What are the various uses to which a b.f.o. may be put ?

4. An f.m. receiver (e.g. a C.B. rig) remains obstinately quiet. Suggest a reason and appropriate
remedial action.

5. What is the first essential in detecting an ssb signal ?

Why would you expect an ssb receiver to contain two crystals ? (This has not been explored in
the above text ; if you cannot work out an answer I will deal with it in my reply.)

6. A superheterodyne receiver operates over the range 14 to 15 MHz. If the i.f. is 10.7 MHz
what range will the local oscillator have to cover ?

7. In an amplifier stage what components would be associated with a bipolar transistor which
would not be present with a FET ?

8. What is automatic bias? Where would you find the appropriate components in an amplifier
circuit ?
 END OF PART 3

Back to Top of Page

>>>>>>>>>>>>>>>>>>>> PAGE 45 <<<<<<<<<<<<<<<<<<<<

 PART 4: ELECTRONIC OSCILLATORS
LESSON 1
4.1 THE DAMPED OSCILLATION

As described in Lesson 9 of Part-1 resonance results when two energy stores are coupled
together. In electrical circuits the two components concerned are a capacitor and an inductor and
the energy is interchanged between the electrical field of the capacitor and the magnetic field of
the inductor.

Practical circuits however also possess resistance to the flow of current and this results in
heat-generation which represents a loss of energy from the circuit. There are other losses too
such as heat which develops in the dielectric of the capacitor and energy lost through radiation
(other than heat) but these are generally small compared to the "ohmic" losses.

** Without such losses a perfect circuit, which would be constructed from a "perfect" inductor
and a "perfect" capacitor, would continue to oscillate for ever once it had been energised. The
practical circuit however loses an amount of energy each time there is a flow of current. It
follows that each cycle of oscillation must be smaller than the previous cycle and this is shown
in Fig. 1 . A resonant circuit which loses energy in this manner is said to be damped (the term is
an exact analogy with the similar act of "damping" a bell) and so the series of waveforms as
illustrated is referred to as a damped-train.

In each cycle the current which flows

through the inductor is a function of the
voltage which appears across the
capacitor ; at the end of each cycle the
capacitor acquires a slightly lower voltage
because of the ohmic losses. The losses
themselves are a function of the
magnitude of the current. This
argument shows that the losses in each
cycle are dependent on the magnitude of
the previous cycle and also determine the
magnitude of the cycle which is to follow
?

This phenomenon, in which what happens next is determined by what has just happened, is
an example of a logarithmic law (see 7.2 in Part-7 Mathematics). The envelope of the decaying
wave follows a logarithmic curve and so we say that the waveform suffers an exponential
decay. As with all logarithmic functions you may take any two consecutive cycles and the ratio
of their peak values will always be the same. This is another way of saying that the decay from
one cycle to the next is always the same percentage decrease. (With other mathematical laws
the decrease might be (say) a constant percentage of the starting value.)
 >>>>>>>>>>>>>>>>>>>> PAGE 2 <<<<<<<<<<<<<<<<<<<<

The Spark Transmitter mentioned in Lesson-1 of Part-2 radiated a continuous train of these
damped oscillations because the means were not then available to overcome the damping effect
of the circuit losses. Thus it was necessary to constantly re-charge the resonant circuit so as to
generate a waveform which could be radiated from an aerial.

4.2 MAINTAINED OSCILLATION

4.2.1 Use of Feedback

As mentioned above a continuous oscillation is possible if the circuit losses can be

eliminated. In fact this is possible using so-called cryogenic techniques to take the conducting
material down to a very - low temperature (close to Absolute Zero) when it
becomes superconductive - i.e. virtually electrically lossless. The exact meaning of this is made
clear perhaps by considering a superconducting store (memory). When a loop of wire
is/rendered superconducting and a current pulse is induced into it that pulse propagates round
and round the loop endlessly without dissipating its energy. Powerful superconducting
electromagnets remain cool because the electric coils do not dissipate energy.

However, maintaining cryogenic equipment is not a practical way to make an oscillator. At

normal temperatures the circuit losses cannot be eliminated but it is possible to eliminate their
effect. To do this it is necessary to introduce a small amount of energy to replace that dissipated
in the circuit resistance. In Part 1 it was mentioned also that this first became possible with the
invention of the triode valve.

Consider a resonant circuit which loses 10% of its contained energy with each cycle of
oscillation. A signal, taken from that resonant circuit, is passed through an amplifier with a
power-gain of 1.1 (i.e. 10%) and then returned to the resonant circuit ; each cycle of oscillation
now has the same magnitude as the previous cycle. The output from the amplifier has been used
to boost the input signal and so this is an example of positive feedback.

Should the gain of the amplifier be too small, then the oscillation will die away; it will take
longer to decay than would the same circuit without the amplifier but it would decay
nonetheless. If the gain of the amplifier stage is too large then each cycle will be of greater
magnitude than its predecessor ; i.e. the oscillation will build-up.

The last effect will not necessarily lead to disaster because, as the oscillation amplitude
increases, so too will the 10% losses and eventually a condition could be reached in which
oscillation is once more maintained at a constant (but larger) amplitude.

Clearly a control mechanism is required that is capable of adjusting the amplifier gain. If the
oscillation amplitude begins to increase then the amplifier gain must be decreased slightly ; if
the oscillation amplitude begins to decrease then the amplifier gain must be increased slightly.
Given this automatic control of amplitude then the circuit can be maintained in continuous and
steady oscillation.

>>>>>>>>>>>>>>>>>>>> PAGE 3 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

There are several devices which can perform this control function and they all operate within
a negative-feedback loop. The oscillation-amplitude is monitored and an error-signal is
generated which describes the manner in which the output deviates from the required amplitude
; this error-signal is then used to adjust the gain of the maintaining amplifier. The most common
form of control loop is an automatic-bias circuit. The maintaining device is allowed to pass a
current which is proportional to the amplitude of the oscillatory signal and this charges a
capacitor; the voltage across this capacitor is then used as a bias to reduce the amplifier gain.
When the oscillation amplitude increases the capacitor is charged to a higher voltage ; when the
oscillation amplitude decreases the capacitor voltage falls as the capacitor

In devices such as precision signal-generators, where the Designer has tried to avoid the
waveform distortion which is inherent in automatic bias circuits, you may find more expensive
solutions. Some of the output power from the oscillatory circuit may be fed to
a thermocouple which generates a small voltage that is proportional to the temperature of a two-
wire junction. Alternatively a small heater may be used to adjust the resistance of a heat-
sensitive device or, today, the resistance of a semiconductor device may be adjusted by means of
a small light-emitting diode which is fed with the oscillator output. Needless to say these circuits
are slower in their reaction to amplitude disturbances.

4.2.2 Frequency Stability

To understand the mechanism which keeps an electronic oscillator to a single frequency it is

necessary to examine the resonant circuit more closely. Fig. 2 is a graph which shows the
changes of impedance presented to a generator by a parallel combination of inductor and
capacitor.

At low frequencies the capacitor

is virtually open-circuited and all the
signal current flows through the
inductor. The reactance therefore is
a very-low positive value. As the
frequency rises toward the right-
hand end of the graph so the
reactance increases as shown. (See
Lesson-6 of Part-1.)

At very high frequencies the

inductor is virtually open-circuited
and all the signal current flows
through the capacitor. The reactance
therefore is a very-low negative
value. As the frequency falls toward
the left-hand end of the graph so the
reactance increases toward a high
negative value. (See Lesson-7 of
Part-1.)
 >>>>>>>>>>>>>>>>>>>> PAGE 4 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

At the centre of the graph we have to create a continuous curve by joining together the high
positive reactance of the inductor and the high negative reactance of the capacitor ; as shown
this results in a very abrupt transition. Note that, if we choose a large inductor and small
capacitor , this transition will be much more abrupt than if we use a small inductor and large
capacitor. However the maximum values of impedance will be limited by the circuit losses
which appear as a parallel-connected resistor. Thus while the sharpness of this transition
depends on the values chosen for the L and C it is critically dependent on the circuit Q-factor
(see Lessons 8 & 9 in Part-1).

** The important point is that, when the circuit oscillates exactly at its resonance frequency,
then the overall reactance is zero (wL -1/wC) which corresponds to the point where the abrupt
part of the curve crosses the zero line. Should the frequency rise then the reactance very-rapidly
becomes capacitive ; should the frequency fall then the reactance very-rapidly becomes
inductive. It is the resultant rapid shift in the phase-relation between applied voltage and
resulting current which controls the behaviour of the positive-feedback circuit.

** Very-high-Q circuits (low-loss circuits) present transitions that are very steep indeed. Thus
oscillators which use high-Q circuits are constrained to keep within very-narrow frequency
limits. (See Fundamentals-2, Lesson-1, Page 4)

4.2.3 RC and Crystal Oscillator Circuits

** As an introduction to oscillator circuits the LC oscillator is probably the easiest to follow.

However, all that is required is a positive- feedback circuit to make good the natural circuit
losses (and losses due to signal-extraction), a negative-feedback circuit to control the amplitude
of oscillation and a phase-sensitive circuit to act as a frequency-controlling device.

A series or parallel combination of inductor and capacitor is the most obvious choice but
phase-sensitive networks can be created from either a combination of resistance and inductance
or from a combination of resistance and capacitance.

These two families each have two forms (a) the reactor in series and the resistor in shunt
(b) the reactor in shunt and the resistor in series. As a rule-of-thumb series-capacitor circuits are
used where the phase-shift network doubles also as a RC-coupling between amplifier stages.
 Shunt-capacitor and also inductive networks
find their uses in current-driven stages
(e.g. bipolar-transistor maintaining amplifiers).

** Fig. 3 shows the equivalent electrical

circuit for a slice of quartz crystal which has
been cut and mounted for use in controlling an
oscillator. The crystal itself behaves like a
series combination of Land C and this is
shunted by a small capacitance which results
from the crystal
mounting. Typical values are given in the
diagram.

>>>>>>>>>>>>>>>>>>>> PAGE 5 <<<<<<<<<<<<<<<<<<<<

Above the natural resonance frequency of the crystal the controlling (series) reactance is
inductive and this resonates with the parallel stray capacitance to produce a secondary
resonance. It is important to recognise three factors:

(a) The lower of the two resonance frequencies is a series mode and therefore the crystal is
working at low impedance

(b) The higher of the two resonance frequencies is a parallel mode and therefore the crystal is
working at high impedance

(c) Stray capacitances form part of the parallel-mode circuit and so a parallel crystal is ground to
work with a specified capacitance.

I have met many instances of crystal oscillators producing the "wrong" frequency and there
are many people who believe that a crystal-oscillator can only operate on its "correct"
frequency. Investigation has shown that either the wrong crystal has been inserted, a crystal
calibrated for series operation is being used in a parallel-mode circuit, a crystal calibrated for
parallel operation is being used in a series-mode circuit or that a parallel-crystal is being used
with shunt stray-capacitance other than that originally specified.

** In a circuit operating with a parallel crystal it is possible to connect a small variable

trimmer-capacitance across the crystal to provide a fine adjustment to frequency. It should be
remembered that this trimmer belongs to the crystal and not to the circuit, that it should be used
to set the crystal on the specified frequency and that the crystal cannot give of its best if any
other frequency is used. Such a trimmer can operate over a limited range only and will not be
effective in a series-mode circuit.

** Oscillators should always be driven as lightly as possible; i.e. just hard enough to ensure
reliable operation. A maintaining amplifier which is overdriven will distort the waveform and
thus destroy the frequency stability. A distorted waveform indicates the presence of harmonics
which, in turn, requires the resonant circuit to increase its bandwidth i.e. the circuit Q has been
lowered. Alternatively you may prefer to consider that an over-driven amplifier is likely to draw
excessive input current, which corresponds to a fall in input resistance which adds excessive
damping to the resonant circuit ; once again the circuit Q is degraded.
 ** Note that, if the bias circuit time-constant is made too large, then the charge on the capacitor
cannot leak away fast enough and oscillation ceases. Oscillation restarts once the bias has
decayed sufficiently. The result is a series of short bursts of oscillation — a performance known
as squegging.

4.3 OSCILLATOR CIRCUITS

4.3.1 The T.A.T.G. Oscillator

The tuned-anode-tuned-grid oscillator gets its name obviously from the early days of its use
with valve maintaining-devices. Not only is it possibly the simplest of the oscillator circuits to
understand but it is probably the most commonly-encountered circuit — usually without intent.

Back to Top of Page

>>>>>>>>>>>>>>>>>>>> PAGE 6 <<<<<<<<<<<<<<<<<<<<

Fig. 4 shows the circuit in its valve form

using a triode. It consists of two tuned LC
circuits, one in the anode and the other in the
grid, which are coupled via the anode-to-grid
stray capacitance. Sometimes, and especially at
low frequencies, it is necessary to increase this
feedback capacitance by adding an external
component.

The grid circuit is the frequency

determining element ; the anode circuit needs to
be tuned to a slightly lower frequency so that
the anode load appears to be inductive. This
offsets the phase-shift caused by the capacitive
feedback path and so ensures that, at the desired
resonance frequency, the feedback is entirely
positive.

The simplicity of the circuit comes at a price. The need to tune the anode circuit to a
different frequency means that the tuning elements cannot be ganged. Additionally ganging
would not be easy because both sides of the anode capacitor are at 250 volts. It is possible to
connect the anode capacitor between anode and the negative line, thus allowing the tuning shaft
to be at ground potential, but this means that the capacitor has to withstand the sum of the h.t.
supply plus the peak oscillatory voltage (up to twice the h.t. potential).

The grid frequency-determining circuit can be replaced by a crystal operating in its parallel
mode and the result is a very reliable and flexible oscillator. This t.a.t.g. valve circuit was
frequently used in a pentode arrangement in which the screen-grid acted as the oscillator anode
and this allowed the anode-circuit to be tuned to the second or third harmonic of the crystal.
Such an arrangement provides excellent buffering for the oscillator ; i.e. the oscillator is
protected from loading by the output circuit. This and similar circuits are known as electron-
coupled oscillators.
 The circuit, though seldom used today, is of importance in that it is often created
unintentionally. For example, a multi-stage rf or if amplifier consists of several amplifier-stages
separated by tuned (or double-tuned) circuits. If one of these circuits is set to a too-low
frequency and there is sufficient gain available, the entire "amplifier" will turn abruptly into a
t.a.t.g. oscillator (see Lesson-5 in Part-3),

Note the grid-coupling capacitor and grid-leak resistor shown in Fig. 4. The valve is not
provided with automatic cathode bias and so the positive-going half-cycles of input voltage
drive the grid into drawing grid current. This current charges the coupling capacitor and so
drives the grid negative which lowers the stage gain. This charge continually leaks from the
capacitor via the resistor and is replenished every half-cycle. Thus any increase in oscillation-
amplitude increases the bias while any decrease in oscillation-amplitude allows the bias voltage
to decrease. If this grid combination is removed and replaced by an automatic cathode-bias
circuit the oscillator will not function.

>>>>>>>>>>>>>>>>>>>> PAGE 7 <<<<<<<<<<<<<<<<<<<<

END OF LESSON 1

* * * * * * * * * * * *

QUESTIONS

1. What is the prominent characteristic of a damped oscillation ?

2. How does an active device turn a damped oscillation into continuously-maintained

oscillation?
What else is required to ensure continuous oscillation?

3. Why does such an arrangement prefer to operate at a single frequency ?

[Link] reasons why a crystal oscillator may NOT operate at the frequency which is printed on
the crystal casing.

5. What adjustment must you make before a T.A.T.G. oscillator will burst into song ?
What is probably wrong if the oscillator produces bursts of oscillation instead of a continuously-
maintained waveform ?

6. Given the crystal constants shown in Fig. 3 calculate the (series) resonance frequency of the
crystal. Use reasonable approximations to avoid complicated arithmetic.

Can you indicate the method of calculating the parallel-resonance frequency ? You need an
algebraic solution; if your maths do not run to this exercise can you at least show the way ? (Not
for the R.A.E.)

>>>>>>>>>>>>>>>>>>>> PAGE 8 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 LESSON 2
4.3.2 The Transformer-coupled Oscillator

The simplest way to feed the output of an amplifier-stage back to its input is by means of a
transformer. Where the oscillation is maintained by a valve or by a field-effect transistor the
transformer must feed back a voltage-sample of the output ; where the oscillation is maintained
by a bipolar transistor then the transformer must feed back a current-sample of the output.

The difference is significant. The input-impedance of a valve or of a FET is very high and so
a step-up (voltage-gain) transformer is required (see Lesson-5 in Part-1). The input-impedance
of a bipolar transistor is very low and so a step-down (current-gain) transformer is required.

As in the t.a.t.g. circuit the control of

oscillation-amplitude can be left to an
automatic grid-bias (base-bias) circuit . If the
transformer is resonated it will also provide the
necessary phase-sensitive network to control
frequency-stability. Note that, to achieve stable
single-frequency operation, the transformer
must be single-tuned i.e. not bandpass-tuned.

** The result is shown in Fig. 5 where a

bipolar npn transistor is used. The collector
winding has been resonated because it is the
high-
impedance winding and so requires the smaller
capacitance. The choice of time-constant for the
base CR combination is dependent on the
frequency of oscillation. Increasing the time-
constant (with frequency unchanged) increases
the base bias and so reduces the amplitude of
oscillation. (Note that the value of R is obtained
from the parallel-combination of R1 and R2. )

As before it is necessary not to drive the amplifier too hard and, where appropriate, it is
important not to saturate the iron core of the transformer.

In a valve maintained version it is the grid which is resonated and the result is known as
a Meissner Oscillator.

It is not so easy to use a MOSFET device as an oscillator because its insulated gate does not
draw current and so an automatic bias circuit, as described above, is not possible. Other methods
are available but these are not examined for the R.A.E.

>>>>>>>>>>>>>>>>>>>> PAGE 9 <<<<<<<<<<<<<<<<<<<<

 The Hartley Oscillator (and the Colpitts described next) is a natural development of the
transformer oscillator in that it uses a single tapped coil as an autotransformer (see Lesson-5 in
Part-1). The basic circuit is shown in Fig. 6 ; diagram (a) gives the straightforward form and

diagram (b) a so-called inverted form. The

advantage of the inverted Hartley is that the
various components can be isolated from the
supply voltage which is an important
consideration in high-voltage valve circuits.

Note the use of a resistor-capacitor

decoupling circuit to prevent feedback from the
buffer amplifier to the oscillator.

The buffer amplifier is shown with an

untuned load thus avoiding the need for a twin-
ganged tuning capacitor. Greater isolation of
the oscillator may be obtained by operating the
buffer stage at either the second or third
harmonic of the oscillator frequency but this is
practical only with fixed-frequency operation
unless two tuning-controls are acceptable.

In diagram (a) the dc-supply is fed to the

transistor through the windings of the inductor
and this arrangement is referred to as a series-
fed oscillator. In the inverted form of diagram
(b) the dc-path has been diverted from the coil
which is capacitively-coupled to the valve; the
choke provides a dc-path in parallel with the
oscillatory circuit and so this arrangement is
referred to as an inverted shunt-fed or parallel-
fed Hartley oscillator.

By careful design a Hartley circuit can be electrically-balanced so that the frequency of

oscillation is substantially unaffected by the parameters of the active device and this can greatly
increase its frequency stability.

Clearly this circuit cannot be adapted for crystal-control.

Back to Top of Page

>>>>>>>>>>>>>>>>>>>> PAGE 10 >>>>>>>>>>>>>>>>>>>>

4.3.4 The Colpitts Oscillator

 One version of this circuit is illustrated
in Fig.7. It differs from the Hartley only in
that, instead of using a connection to the
inductor, the tuned-circuit is tapped in its
capacitive leg.

This circuit too can be constructed in both

normal and inverted forms and can be either
series-fed or shunt-fed.

Inspection of the circuit shows that the

internal capacitances of the active device are in
parallel with the two parts of the tuning
capacitance ; variations in the active device can
thus cause frequency drift and it is essential that
Fig. 7 A Colpitts Oscillator
the tuning capacitors greatly exceed the values
of the strays.
Fig.8 shows the Colpitts adapted in
its inverted shunt-fed form for parallel-
mode crystal operation. There are
hidden traps waiting for the unwary. The
coupling capacitors Cl and C2 provide a
tapping across the crystal but, acting
together, they also form extra
capacitance across the crystal ; this extra
capacitance must be accounted for when
specifying the crystal parallel resonance.

The values of Cl and C2 would

likely be in the region of 100 pF ; if
these are increased to values around
1,000 pF the circuit turns into a
Clapp/Gouriet oscillator and will jump
Fig. 8 A Crystal-controlled Colpitts
to the series-resonance mode of the
or Gouriet Oscillator
crystal (see next Section). Unless well
designed this arrangement is not a very
good bet therefore unless you have a
need for an oscillator that might jump
erratically between modes.

>>>>>>>>>>>>>>>>>>>> PAGE 11 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

4.3.5 The Clapp or Gouriet Oscillator

 This circuit was developed specifically as a
very-stable drive for MW-transmitters. It was
developed in this country during the second
world-war by G.G. Gouriet who could not
publish his results for security reasons. It was
also developed by Clapp in America.

The circuit seeks to eliminate the drifting

effects of the maintaining device by using very-
large value capacitors to tap into the resonant
circuit. To do this the LC circuit is used in its
series-resonant mode.

Particularly in its crystal-oscillator form the

circuit regularly appears with small coupling
capacitors ; this overdrives the crystal and,
apart from destroying its frequency-stability
Fig. 9 The Clapp/Gouriet Oscillator and accuracy, can cause destruction of the
crystal. Another common fault is the failure to
realise that, properly-designed in this
Clapp/Gouriet arrangement, the crystal will
oscillate in its series-resonant mode.

4.3.6 The Pierce Oscillator

This circuit uses the phase-sensitive

arrangement in series with the positive-
feedback path and hence it too uses a series-
resonant mode. In Fig. 10 a Pierce Oscillator is
shown using a crystal and a JFET. In this
circuit the crystal capacitance combined with
the resistor to ground, form the biasing CR. The
capacitor Cl serves to protect the crystal from
the supply voltage. The radio-frequency choke
(R.F.C.) could be replaced by a resonant circuit.

4.3.7 Resistance-capacitance Oscillators

In the previously-described oscillators the positive-feedback path was maintained via an LC

resonant circuit which produces an exact p. f. b. only at the resonance frequency. A similar
frequency-sensitive circuit can be obtained however by a network of resistors and capacitors ;
these fall roughly into two classes:

(a) circuits which produce 1800 phase-shift

(b) circuits which produce zero phase-shift.
 >>>>>>>>>>>>>>>>>>>> PAGE 12 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

Fig. 11 shows examples of each circuit and
both depend
on the fact that the current which flows through
a capacitor leads the voltage across that
capacitor by 900.

In diagram (a) a resistor is connected in

series with the capacitor. With respect to the
common current the voltage across the
capacitor leads by 90° while the voltage across
the resistor is in phase with the current.

The value of the capacitor is chosen so that,

at the required frequency, the reactance is equal
to the resistance ; draw a vector diagram and
you will see that this results in the output-
voltage being phase-shifted by 450 with respect
to the input-voltage.

In theory two such R-C sections should give an overall phase-shift of 900 but, of course,
practical components have losses and so the phase-shift must be slightly less than 450. For this
reason it is usual to use three sections and operate at 600 per section thus giving a margin for
error. To make an oscillator with this arrangement an amplifier stage is required that both
inverts the signal (1800 phase-shift) and makes good the attenuation through the three-stage RC
network.

In diagram (b) a series RC network is employed together with a parallel RC network in

a Wien Bridge circuit. The current which flows through the series circuit is divided between the
two arms of the parallel circuit which produces a 450 voltage phase-lag so that the overall phase-
shift is zero. To make an oscillator with this arrangement an amplifier is required that does NOT
invert the signal but which makes good the attenuation through the R-C network. To do this with
normal discrete amplifying stages a 2-stage amplifier is necessary but Fig. 12(b) shows a neat
way of achieving an oscillator using the differential inputs of an op-amp.
 >>>>>>>>>>>>>>>>>>>> PAGE 13 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

As explained under 3.11 & 3.13 an op-amp has two differential inputs of which one
produces an in-phase output signal while the other produces an anti-phase output signal.
Feedback via the Wien-bridge circuit is applied to the in-phase (+) input thus achieving a p.f.b.
loop. Feedback to the other (-) input produces a n. f. b. loop which controls the gain through an
automatic-bias circuit. This utilises a filament lamp as a temperature-variable resistor whose
resistance increases as the current through it increases and so, by increasing the bias, decreases
the stage-gain.

4.3.8 Relaxation Oscillators

These are oscillators which, essentially, have too much uncontrolled gain around the
positive-feedback loop and this causes them to generate excessive bias which shuts-down the
amplifier. The bias, stored as a charge in a capacitor, then leaks away through a resistor (the
circuit is said to relax) until the amplifier gain is sufficient to support oscillation when the cycle
repeats.

The first of these circuits is the Blocking Oscillator which, essentially, is a transformer-
coupled oscillator as illustrated in Fig. 5. The automatic base-bias (grid-bias) time-constant is
made too large for continuous oscillation with the result that the circuit performs a one-half-
cycle oscillation and then remains quiescent for a period determined by the time-constant. The
output from such an oscillator consists of a series of large-amplitude pulses separated by
relatively-long periods of time. They often find use in driving time-bases for picture-monitor
tubes perhaps because they are easy to synchronise to an external waveform.

The second type of relaxation oscillator is the multivibrator. This uses a 2-stage amplifier in
which the output of each stage is coupled to the input of the other thus creating a positive-
feedback loop with considerable gain. Again an automatic-bias circuit is employed with a large
time-constant that rapidly generates a bias but only allows that bias to leak away slowly. The
output from these oscillators is a rectangular wave with very rapid transitions between the full
supply value and near-ground. This lower level is approximately 50-volts in valve-operated
circuits.

Multivibrators are found in an almost endless variety of forms but they all fall into the three
categories namely monostable, bistable and astable. Monostable forms have one direct-coupling
and one RC-coupling and, as a result, they have one state in which they remain stable ; e.g. with
amplifier-1 cut-off and amplifier-2 bottomed (fully-on). When triggered these m.v.'s switch into
the alternative unstable state (amplifier-1 bottomed, amplifier-2 cut-off) and remain there for a
period determined by the RC time-constant before reverting to the stable state. They are used
mainly for generating time-delays.

Bistable m.v.'s are dc-coupled and have two stable states ; when triggered they switch to the
alternative state and remain there until triggered again. These are the basis of the binary
counting-circuits described in Part-2, Lesson-4.

The astable forms are entirely RC-coupled and so are without a stable state ; when powered,
they constantly change between the two possible states ; these are often referred to as free-
running multivibrators. Some examples are given in Fig. 13.

>>>>>>>>>>>>>>>>>>>> PAGE 14 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

4.3.9 The Timer Chip 555

The 555 is a very versatile - even famous - chip that, although basically a timer, can be used
for an almost endless variety of tasks. Indeed its possibilities form the subject of several books.
The 8-pin chip contains a monostable multivibrator, two voltage-comparators, an output
amplifier and a couple of other inputs and outputs. By varying its connections and changing
external components it can be used as an oscillator, a timer, a pulse generator, a pulse-position
modulator — to name but a few. It has a sister chip the 556 which, with 14 pins, offers two
separate 555 units.

The basic chip costs around 35p and is an ideal device on which the newcomer to electronics
can try his hand at practical construction.

4.3.10 Harmonic Oscillators

This may appear to be a somewhat obscure title but it refers to oscillators which produce an
output at a frequency which is greater than the natural resonance frequency of the frequency-
determining circuit. In general such oscillators are suitable only for fixed-frequency working.

There are two basic techniques employed:

1. The oscillator-unit is a multi-stage device in which the first stage is an oscillator and the
following stages act as frequency-multipliers thus producing an output which is a harmonic of
the oscillation frequency.

2. An oscillator circuit is arranged to have considerable gain at the required harmonic of the
frequency-determining circuit so that oscillation takes place at that harmonic frequency and so
suppresses the fundamental mode of oscillation.

>>>>>>>>>>>>>>>>>>>> PAGE 15 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

The first type is referred to as an oscillator and multiplier chain and indeed the oscillator
stage may employ a circuit of the second type. Each stage must be resonated at the frequency at
which it is to operate and it is this requirement which makes it difficult to construct the device as
a VFO. Note that the frequency-multiplication also multiplies any frequency drift of the
fundamental oscillator.

In valve circuits the technique was often used in the so-called electron-coupled oscillators
which provide isolation between the fundamental oscillator and the following stages. The
oscillator stage uses a pentode valve in which the screen-grid is pressed into service as the anode
of a triode; this part of the valve forms the fundamental oscillator. Its
operation "modulates" the electron stream that passes through the screen-grid to arrive at the
anode. The anode circuit is resonated at the desired harmonic frequency - usually 2nd or 3rd
harmonic — and it is driven by the modulated anode-current. The excellent isolation between
output and oscillator arises from the difference in their frequencies of operation.

** The second type of harmonic oscillator is most usually a crystal-controlled type in which
the crystal is specially cut to make it active at the desired harmonic frequency. Such a crystal is
labeled with the frequency at which it is intended to operate but it is easy to spot because that
frequency is above the normal range of crystals. But beware ! The harmonic crystal is capable
of sustaining oscillation at its fundamental frequency and confusion results.

I have been confronted with a frequency synthesiser which refused to produce a variable
output so that the output remained fixed no matter how the frequency-control was set. Both
oscillators were operating powerfully and, at first, I took it for granted that the crystal oscillator
was on the correct frequency. When, in desperation, I checked it with a wavemeter I found it to
be having a Ball at the fundamental; i.e. one-third of the frequency marked on its case. The
cause was low current-gain (about 1/3rd) in the maintaining transistor. [A likely explanation is
that there was insufficient gain to maintain the 3rd-harmonic mode of oscillation, that this
resulted in low-bias on the transistor which therefore exhibited a too-large input capacitance; in
turn this modified the oscillator circuit so that there was sufficient gain for the fundamental
mode to take over. If this is true it indicates that the design of the oscillator was faulty.] The
problem was cured by simply exchanging the transistor.

The above is not an exhaustive list of oscillator arrangements but contains most of those
likely to be met in amateur radio. The only ones likely to be examined for the R.A.E. are the
Colpitts and the Hartley but even then not in any detail. However, if you buy elderly valve-
operated gear, you might well meet any of these except the 555.

END OF LESSON 2

END OF PART 4

Back to Top of Page

 PART 5: POWER SUPPLIERS
LESSON 1
5.1 RECTIFICATION

5.1.1 Introduction

** Electrical power is generated and distributed in alternating form for convenience, cheapness
and safety but often it is required in direct form. The process of deriving dc from an alternating
supply is called rectification and the device which is used to accomplish rectification is called
a rectifier. The reverse process, that of obtaining ac from a dc-supply, is known
as inversion and the electronic equipment used to achieve it is called an inverter.

** The majority of electronic equipment requires dc-supplies and so, to operate from the
normal 240-volt ac "mains supply", a unit must be employed to carry out the operation of
rectification, to provide the correct voltage and to provide the necessary smoothing the purpose
of which is to remove all traces of ac (see later). This overall equipment is referred to as a Power
Supplier.

** There are two approaches to the problem of rectification:

(i) Allow the alternating current to flow in the required direction but prevent it from flowing the
other way

(ii) Provide a switching mechanism that will reverse the circuit connections each time the
current-flow reverses ; this method has been mentioned already in the mechanical switch of the
so-called dynamo (see Fundamentals-1: Mechanical Generators. However, with the mains-
supply reversing 50-times every second, a mechanical switch cannot be used and electronic
switching is required.

** The first method results in a series of current pulses which occur during alternate half-
cycles of the applied emf and the current remains at zero during the intervening periods ; this is
known as half-wave rectification and is illustrated in Fig.1.

** The second method produces current pulses during each and every half-cycle of the
applied emf and is therefore known as full-wave rectification ; this is illustrated in Fig.3.

** Rectifiers are made in both thermionic and semiconductor form ; they have only two
electrodes and so are known as diodes. In effect they are switches that are operated
automatically according to the direction of current-flow. In one direction they offer between the
two electrodes a low resistance to the passage of current but, in the opposite direction, they offer
high resistance.

Thermionic diodes, like all valves, contain a heated cathode which emits electrons into an
evacuated envelope ; when the second electrode or anode is charged positively with respect to
the cathode these electrons are attracted to the anode so forming a flow of electron-current from
cathode to anode.
 When the anode is charged negatively with respect to the cathode the electrons are repelled
back toward the cathode ; the unheated anode does not emit electrons and so a current-flow is
not possible. In general valve diodes are useful for rectification above 50 volts.

>>>>>>>>>>>>>>>>>>>> PAGE 2 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** Semiconductor diodes do not use a heated cathode but the physics of their layered internal
construction ensures that conduction takes place in one direction only. These diodes are useful
for low-voltage rectification but there are some types that are capable of working at several
hundred volts. Semiconductor diodes are generally less robust than valve types and must not be
allowed to overheat.

** There is one important difference in that, when a semiconductor diode is subjected to

excessive voltage in the reverse direction, it will abruptly switch to a second conduction mode ;
the voltage at which this takes place is remarkably constant irrespective of the current which
flows and this is known as the zener voltage of the diode. So-called zener diodes are
manufactured especially for use in this mode ; they have several uses but the most important is
as a reference-voltage source as discussed later.

5. 1.2 The Half-wave Rectifier

** Fig. 1(a) shows a diode (general symbol) connected between an ac source and a load
represented as a resistor RL. The arrow-head symbol represents the direction of classical-current
flow (from positive to negative) and therefore the upper end of RL has a positive potential with
respect to the lower end ; semiconductor diodes are marked with a red spot or a line to indicate
the positive-output (cathode) connection.

** Fig. 1(b) shows the waveform of a sinusoidal input voltage and the rectified half-wave
current that flows through RL. These current pulses spaced at half-cycle intervals are useless as a
source of dc and means must be found to keep current flowing during the gaps. The first step in
this process is shown in Fig. 2(a) where a large-value capacitor is connected across RL,

** During the half-cycle when the diode conducts there is a flow of current through RL but
current flows also into the capacitor and charges it to the peak value of the alternating supply
voltage. During the following half-cycle, when the diode is non-conducting, the capacitor
discharges through RL so maintaining the load current ; during this process the p.d. across the
capacitor falls steadily, as shown in diagram (b), until the emf of the ac-source rises once more
and exceeds the residual voltage across the capacitor. Because this capacitor stores charge
against future needs it is known as the reservoir capacitor.

>>>>>>>>>>>>>>>>>>>> PAGE 3 <<<<<<<<<<<<<<<<<<<<

Back to top of Page

** The voltage developed across RL is still not a true direct voltage ; it has a ripple
component which will interfere with the operation of electronic circuits. The ripple is removed
by filtering techniques described later under Ripple Filters For low-voltage semiconductor
circuits this is usually dealt with in stabilising-circuits built into a single chip.

5.1.3 The Full-wave Rectifier

** A full-wave rectifier circuit is shown in Fig. 3 and two obvious requirements are an extra
rectifier and a transformer that is wound with a centre-tapped secondary winding. The
transformer performs the task of "reversing the connections" ; in fact it provides two alternative
feeds which, by virtue of the centre-tap, have opposite polarities. Each diode has its positive-
output connection (or cathode) taken to the same end of RL and so, on alternate half-cycles,
each drives a pulse of current through the load in the same direction.

>>>>>>>>>>>>>>>>>>>> PAGE 4 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** A full-wave rectifier circuit produces twice as many current pulses per cycle and the half-
cycle gaps are eliminated ; thus, compared with the half-cycle arrangement, the task of the
reservoir capacitor is greatly eased. Full-wave rectification offers several advantages:

(a) Improved regulation. A rectifying dc-power supplier has the same limitations as any other
source in that, when current is demanded, the terminal voltage falls. The term regulation refers
to the amount of that fall with regard to the corresponding current-demand. With reference to
Ohm's Law the voltage drop per unit current may be expressed as a resistance in ohms ;
alternatively the drop is expressed as a percentage of the off-load voltage when the current is
increased from zero to the full rated-value.

A full-wave rectifier delivers twice the number of current pulses and leaves the reservoir
capacitor only half the time in which to discharge. Thus its regulation is considerably better than
that of a half-wave circuit.

(b) Ripple from a full-wave circuit is much reduced in amplitude because the capacitor suffers
less discharge. The frequency of that ripple is twice that from a half-wave circuit. Both factors
reduce the demand on the following ripple-filter arrangements which therefore can be
constructed either more cheaply or with greater effect.

(c)Because the reservoir capacitor is re-charged twice each cycle and is discharged for a shorter
period the pulses of re-charging current which flow through each diode are greatly reduced as
compared to the diode-current in a half-wave circuit ; this reduces the cost of the diodes and
lengthens their life-expectancy.
 Alternatively it allows the value of the reservoir capacitor to be reduced with a consequent
saving in cost. However, a reduction of the capacitance must increase the value of the diode
current to compensate for the increased rate of discharge.

Yet again reducing the value of the capacitance reduces the initial surge at switch-on as the
capacitor charges from zero ; such surges (known as in-rush current) can blow fuses
particularly if the moment of switch-on coincides with the ac-input reaching a peak value.

Clearly this is an area where a Designer has to compromise between conflicting factors.

5.1.4 The Bridge Rectifier

** A disadvantage of the full-wave circuit described above is that it requires a balanced feed ;
i. e. it needs two equal supplies of opposite polarity. The centre-tapped transformer which
supplies these feeds is specified as (say) 100-0-100 volts and is expensive. The bridge rectifier
requires only a single feed and so uses a smaller and cheaper transformer; it can operate also
without the need for a transformer. The bridge circuit is also useful where perhaps a centre-tap is
not available on an existing transformer.

The arrangement requires four diodes instead of two but this is a reasonable trade. The
bridge-rectifier is less suitable for valves because of the need for four separate heater feeds or
for well-insulated heaters.

>>>>>>>>>>>>>>>>>>>> PAGE 5 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** The circuit is shown in Fig. 4 where the

heavy lines indicate current flow during one
half-cycle of input; the diodes, acting in
pairs, interchange the connections each time
the alternating supply reverses polarity.

** A disadvantage of the bridge rectifier is

that it lacks a common connection between
the ac-input to the Bridge and the dc-output
and so both cannot be earthed simultaneously.

5.1.5 Voltage-multiplier Rectifiers

These circuits provide full-wave rectification but they produce direct-voltage outputs that are
greater than the peak value of the ac-input. They offer also the advantage of a common
connection between input and output and so the entire arrangement can be securely earthed. A
transformer feed is not necessary.

The disadvantage of voltage-multipliers is their poor regulation. The circuit action moves
stored charges between a multiplicity of capacitors and so only small currents can be supplied.
Probably their greatest use is in instruments which use small cathode-ray tubes where they
supply the e.h.t. (extra-high tension) at around 600 to 1,000 volts with only a few milliamps
loading.

Fig.5 shows a voltage-multiplier circuit. In diagrams (i) to (iv) the heavy lines show the
circuit action in its first four stages after switch-on; diagram (v) shows the usual presentation
which, perhaps more graphically, shows the manner in which charge is boosted step-by-step up
the ladder of capacitors switched by a series of diodes.

The principle of operation is that a charge-transfer capacitor is first charged directly from the
input ac supply; during the following half-cycle this capacitor is then switched in series with the
supply to charge a reservoir capacitor to the sum of the input peak-voltage and the similar p.d.
across the charged capacitor. This basic action is extended by adding successive stages of
charge-transfer capacitor and voltage-adding reservoir capacitor. The high-voltage output is
taken from across all the reservoir capacitors connected in series.

Diagram (i) shows the first charge-transfer capacitor being charged to the peak positive
voltage of the input waveform because its right-hand side has been connected to the common
earthed line by diode Dl. In diagram (ii) the supply voltage has reversed and diode Dl, now
reverse-biased, has become non-conducting. Diode D2 has now connected the input supply in
series with Cl across the first reservoir capacitor C4 and so (after several cycles) C4 becomes
charged to twice the peak voltage (Vpk) of the supply.

In diagram (iii) both Dl and D2 have become reverse-biased as the supply again reverses
polarity. Diode D3 has now connected the charge-transfer capacitors C1 and C2 in series across
the supply but, instead of returning

>>>>>>>>>>>>>>>>>>>> PAGE 6 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 >>>>>>>>>>>>>>>>>>>> PAGE 7 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

the connection to the common earthed line, it has taken it to the -[Link] potential at the top of
C4. The total voltage from which Cl and C2 in series are charging is therefore [Link] . On the
fourth half-cycle the two reservoir capacitors C4 and C5 are charged, from the series
combination of Cl + C2 + supply, to a total of [Link] .

This process can be continued with each section adding [Link] to its associated reservoir
capacitor so increasing the output voltage in steps twice the peak voltage of the supply. Note
however that this value [Link] per section is not achieved if charge is lost ; i.e. if current is
being drawn from the final output terminal.

A positive output can be obtained from the same circuit by reversing the connections to all
diodes.

The arrangement is known as a cascade multiplier. The Designer must pay attention to the
voltage rating of the capacitors and diodes. Each reservoir capacitor is subjected to a maximum
of twice the input peak voltage and the charge-transfer capacitors should have a similar rating.

There is another form of voltage-doubler

circuit which is shown in Fig.6. This uses two
diodes to select in turn the alternate half-cycles
of an alternating supply; each diode charges its
own reservoir capacitor and again the output is
taken from across the two capacitors in series.
This circuit has the disadvantage of not
providing a common earthed connection from
input to output and it cannot be extended
beyond a multiplication factor of 2; it must be
transformer fed.

Voltage-doubling circuits are often found in

detector stages of [Link]. receivers where they
provide a modulating signal of greater
amplitude than is obtained with simple diode
detectors.

END OF LESSON 1

>>>>>>>>>>>>>>>>>>>> PAGE 8 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

QUESTIONS

1. Why are Power Suppliers necessary ?

2. Explain the differences between half-wave and full-wave rectifier circuits.

3. What is the basic difference in the purposes of a reservoir capacitor, which is placed at the
input to a Ripple Filter, and the smoothing capacitor which is placed at the output of that filter.?

4. What is meant by the term "Regulation" as applied to a Power Supplier?

5. What are the advantages of the full-wave rectifier circuit over the half-wave ?

Where might you expect to use the half-wave arrangement ?

6. Suggest a reason why a bridge-rectifier device is useful in metering circuits such as you might
find monitoring the performance of a receiver.

>>>>>>>>>>>>>>>>>>>> PAGE 9 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 2
5.2 STABILISED POWER SUPPLIERS

5.2. 1 Introduction

** A stabilised power-supplier provides a direct-voltage output with a value that is maintained

at a pre-set or adjustable value irrespective of changes in the load-current or of changes in the
supply voltage.

That definition sounds as though stabilised suppliers disobey the "internal resistance" rule
that terminal voltage must fall as current-demand is increased but, although they can get very
close to doing so, they do not in fact disobey the rule. These stabilisers employ negative-
feedback techniques (see under 1.11 Amplification in Fundamentals-1) in which changes in
output voltage are measured, amplified and fed-back to a control system that adjusts the circuit
to correct those changes. Thus there is a requirement that the output must change if there is to be
any control but the magnitude of that change is reduced according to the amount of
amplification within the feedback loop.

Depending on the loop-gain and the characteristics of the feedback-loop it is possible for the
output voltage to increase with an increase in current-demand and such a circuit could be
described as having a "negative output-impedance". This can be a useful facility where, for
example, the supply is transmitted over a lengthy run of cable and it is desired to keep the
voltage as constant as possible at the delivery point. Such design needs care however if the
arrangement is to remain stable.

** Power supplied by Electricity Boards is not supposed to vary but the unpredictable nature
of the demand means that some variation is inevitable. Varying demands on a supplier arise
also from the natural use of the load ; for example, a transmitter draws a heavy current when
keyed but only a small current in stand-by mode
.

5.2.2 Simple Stabiliser Using a Shunt Device

 For voltages greater than about 80 volts a shunt stabiliser device would be a gas-discharge
valve ; for lower voltages semiconductor diodes are operated in the zener mode which was
described in Lesson-1 of this Part.

A gas-discharge device is a diode which does not use a heated cathode but its envelope is
filled with a gas at low pressure. A commonly-used gas is Neon and so these valves have
acquired the generic name Neons. Electron-flow through the valve is achieved, not by boiling
electrons from a heated cathode (thermionic conduction), but by breaking-down the gas into
charged components or ions which process is known therefore as ionisation. When the gas
ionises and a current flows the gas emits a coloured glow and the valve is said to strike.

Ionisation of the gas occurs at a potential which depends on the gas and on its pressure.
However, once ionisation has been effected, the p.d. between the electrodes falls and remains at
a (reasonably) constant value irrespective of the current which flows. Circuit design must then
ensure that the current remains within the limits set by the Manufacturer. Outside this range the
p.d. changes in small but discrete steps.

>>>>>>>>>>>>>>>>>>>> PAGE 10 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

Gas-discharge devices are manufactured to provide a range of voltages which remain

constant within the life of the valve so long as the current-rating is observed. Thus it is
necessary to select both for the required voltage and for the current-range.

** Fig. 7 shows a circuit arrangement in which

either a gas discharge valve or a semiconductor
diode can be used to stabilise the voltage of a
small-current supply. The current through the
resistor has two components one of which flows
to the load and the other through the stabiliser.
Because the stabiliser maintains a constant
voltage between its electrodes any change in load
current is compensated by an opposite change in
stabiliser current. The
value of the resistor is chosen so that the current
through it equals the maximum load current plus
the minimum stabiliser current ; should the load
current fall to zero it is essential that the stabiliser
is able to carry the total current.

** This simple arrangement is frequently used to

supply small currents in local parts of large
equipments but its capabilities are limited. Should
the load-current rise too far then the stabiliser
becomes starved of current, ionisation is
extinguished and the stabilising action ceases ;
should the load-current fall too far then the
stabiliser could be damaged.
 A second disadvantage is that it is not possible to vary the voltage at which the output is
maintained.

** The output (or source) impedance of the circuit may not be sufficiently low with the result
that alternating voltage is superimposed on the dc as a result of an alternating component in the
load current. This is usually dealt with by placing a suitable capacitor across the dc output.

However, the parallel combination of a capacitor and a gas-discharge valve can produce an
efficient (relaxation) oscillator. An excessive demand f or current extinguishes the valve but the
output voltage is unable to rise until the capacitor has been charged. When the output voltage
finally reaches the striking voltage of the valve it re-strikes thus increasing the total demand for
current which is supplied temporarily by the capacitor. As the capacitor discharges it finally
reaches the extinguishing potential of the valve and the cycle repeats. This kind of fault in
equipment which has already seen service usually indicates that the gas-discharge device is
reaching the end of its useful life.

** The action of this circuit arrangement, in which the stabiliser shunts current away from the
load, has given it the name shunt-stabiliser.

>>>>>>>>>>>>>>>>>>>> PAGE 11 <<<<<<<<<<<<<<<<<<<<

** Although resistors R 2 and R3 reduce the output voltage as required they also reduce in the
same ratio the error-voltage and this degrades the performance as a stabiliser (it degrades the
regulation); it is also the reason for requiring amplification in the shunt amplifier. The
stabiliser's response to rapid variations is improved by adding capacitor C 2 which, by bypassing
R 2 , raises the ac-gain around the feedback loop.

** One likely variation in the output voltage is a residual ripple component from the rectifier
circuit which has survived the action of the ripple filter; with the bypass capacitor in place the
feedback means that a series-stabiliser can provide a useful improvement in smoothing.

(A considerable improvement in the action of this arrangement can be achieved by re-

designing the shunt amplifier as a cascade amplifier but I have not included this design here.)

** Because the object of this arrangement is to deliver a direct supply it is of necessity a dc-
coupled circuit and prone to drift problems. The accuracy of the output voltage is dependent on
the gain-stability of the series and shunt amplifiers. With a transistor-operated circuit the main
cause of gain changes is likely to be changes in the ambient temperature or heat developed in the
devices themselves. With valves a major problem is that variation of the ac-supply must alter
the voltage of the filament supply.

** Fig. 9 shows a detailed transistor circuit in which such drifts are largely eliminated by use
of a balanced amplifier in the shunt circuit; (see Differential Amplifier in Part-3, Lesson-2.)
The sample-voltage and the reference voltage are compared in the differential-amplifier Tr1 and
Tr 2 and the amplified difference is developed across R6. The intention is that both transistors
will drift in identical manner and so the effects of such drifts will cancel.
 >>>>>>>>>>>>>>>>>>>> PAGE 13 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** In more detail, the reference voltage is developed across the zener diode ZD and transferred
by emitter-follower action (see under 3.12) in TR 1 across the common emitter-resistor R 5 ; the
sample output-voltage is applied to the base of TR 2 so that the input to this transistor (between
base and emitter) is the difference between the two voltages. As an example, suppose that a
temperature change raised the base-emitter voltage of the transistors ; the reference voltage
across R5 would fall but this would take-up the error caused by the increase in base-emitter
voltage of TR 2 .

Note that the two transistors in the shunt amplifier should be a matched pair and should be
mounted in the same way as close to each other as possible.

** Changes in the output voltage at the top end of R 2 , from whatever cause, are fed only to
the base of TR 2 . and so upset the comparison of reference and sample voltages. Such changes
can be deliberately introduced by adding the variable-resistor VR 4 which thus enables the
voltage at which the output is stabilised to be adjusted.

NOTE Although bipolar transistors are current-operated it is legitimate to describe the

circuit operation in terms of voltages because the common emitter-resistor provides negative-
feedback which makes both devices voltage-operated.

5.2. 4 Integrated-circuit Suppliers

 ** In essence these are not different to the series-stabiliser with shunt-amplifier arrangements
described above. The integrated-circuit (IC) device embodies the reference-voltage source, the
series-stabiliser and the shunt-amplifier transistors in a single small package. They require only
the external connection of the rectified ac-supply, the large-value smoothing/reservoir capacitors
and (where necessary) a set-volts control. There is a great deal of gain incorporated within the
chip and it is often necessary to connect a capacitor of around 330 nF (0.33 μF) directly between
the input and common pins to ensure stability.

They differ however in that a separate ripple filter is not usually required and also in that
they are far more versatile than the basic circuit using discrete components. They must not be
allowed to overheat which means that they must be properly mounted on an adequate heat
sink which is a large metal component which will efficiently conduct away the heat and
dissipate it into the surrounding air.

(I have twice encountered a vice wherein, for no apparent reason, the device explodes and
spreads itself into the far corners of the workshop. I believe this to be caused either by a failure
of the heatsinking within the encapsulation of the chip or (most likely) by instability
(oscillation) causing local heating: the instability results from omission of the stabilising
capacitor .

** Voltage-stabiliser ICs come basically in two kinds namely with fixed or variable output.
The variable types offer a pin to which a voltage-control circuit can be connected according to
the Maker's specification. However, the fixed types can provide output voltages greater than
that specified by use of the circuit shown in Fig. 10(b). (See also 5. 2. 7. Switch-mode. )

>>>>>>>>>>>>>>>>>>>> PAGE 14 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

A resistor R 1 is connected in the common lead and extra current is bled through this via
resistor R 2 ; this current should be about 5-times that normally drawn by the IC through its
common connection so as to swamp that current. Normal operation of the stabiliser is to
maintain a constant voltage between its Output pin and its Common pin and this constant-
voltage maintains a constant-current through R 2 . This constant-current develops a substantially-
constant p.d. across R 1 . The overall output voltage is thus the rated output voltage plus the p.d.
developed across R 1 .
 These ICs can also be used to provide a
constant-current source as shown in Fig.11
. The simplest explanation is once again to
note that the IC maintains a constant voltage
across the resistor and so enforces a constant-
current.

5.2.5 Ripple Filters

** The basic Ripple Filter is shown in Fig.

12 ; it is a potential-divider network in which
the lower element is a capacitor. There are
different ways of explaining its action but,
ultimately, all explanations are the same.

(a) at the ripple frequency the low impedance of

the capacitor provides a shunt path for the
alternating ripple-component and so separates it
from the dc which can flow only through the
load.

(b) the capacitor has a large value so that it

needs an appreciable time to charge or to
discharge: partial charging and discharging
results in only a small proportion of the ripple
voltage appearing across the capacitor (which is
across the load).

>>>>>>>>>>>>>>>>>>>> PAGE 15 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

(c) relative to the resistor the capacitor presents a high-impedance to the dc but a low-impedance
to the ripple ; this means that the majority of the direct voltage appears across the capacitor
(across the load) and the majority of the ripple-voltage appears across the resistor.

** A resistor is not the best component to use in a ripple filter because the dc develops a p.d.
across it and that p.d. is dependent in value on the current-demand ; we say that the
resistor degrades the regulation. Nevertheless the circuit is frequently used in this form when
low current-requirement limits both the voltage drop and the power-dissipation to an acceptable
value and the regulation is not of great importance. It provides a very cheap ripple filter.

** For large load-currents, or in those circumstances where the current varies widely, it is usual
to replace the resistor with an inductor whose function is to offer a high impedance to the ripple
current but low resistance to the dc. Ripple has a frequency of 50-Hz for a half-wave rectifier
or 100-Hz for a full-wave or bridge type and so the inductor requires an iron core. To prevent
the core from becoming magnetically-saturated by the dc it is necessary to introduce an air gap
into the magnetic path.

** When, as in transmitters, a large and widely-varying current is experienced and good

regulation is desired it is advantageous to omit the input (reservoir) capacitor (this capacitor is
usually regarded as part of a capacitive-input filter). Without the reservoir capacitor the filter is
known as a choke-input filter or, because the rectifier end of the choke is not restricted in its
voltage excursions, as a swinging-choke filter. See Fig. 13(a).

Strictly speaking the improved

regulation of the swinging-choke filter is
illusory in that it results from the fact that the
output-voltage under light load does not
reach the peak value of the input waveform.
In practice however the important factor is
the variation of the output voltage and so,
provided the reduced (average) output-
voltage is acceptable, then the improved
regulation is real enough.

** For proper operation of a choke-input

filter it is essential that the load current
should not fall below a critical value else the
regulation suffers seriously. To this end
a bleeder resistor (bleed resistor or
bleeder) is sometimes connected across the
filter output.

>>>>>>>>>>>>>>>>>>>> PAGE 16 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** Those rectifier-IC arrangements which dispense with the series inductor have been the
cause of recent legislation. The very-large input capacitor (often as much as 10,000 μF)
required certainly reduces the ripple voltage but, as a consequence, it must draw a very large-
amplitude narrow pulse of current during the short time which is available to replenish its
charge. These large current-pulses load the mains circuit from which the unit is drawing its
primary power and cause serious interference in other equipment over a very large radius. In
future it will be a legal requirement to fit mains-input filters to prevent this current-spike
entering the mains distribution system.

5.2.6 Inverters

Inversion is the process of obtaining ac from a dc-source and is a necessary first step when,
working from battery sources, there is a requirement for supply-voltages in excess of the battery
voltage. The dc is first converted to ac, the voltage is stepped up, or down, by means of a
transformer and then applied to a rectifier/filter system as described above.
 Three methods of dc-to-ac conversion are available:

(a) chopper

dc is interrupted by either electronic or mechanical means thus converting it to a squarewave in

which the amplitude represents the value of the direct voltage. After voltage-transformation this
is rectified as described above.

Today such interruption is accomplished by some form of gate circuit which is driven by a
locally-generated gate-drive waveform.

This system is the basis of the chopper-type dc-amplifiers (see Part-3) for which a synchronous
detector is essential to preserve the polarity of the dc-signal. The synchronous detector is
basically another gate circuit which is driven by the same gate-drive waveform that serves the
chopper circuit.

In early equipments a mechanical vibrator was used to chop the dc and a secondary set of
contacts on the same vibrator served as the synchronous detector.

(b) oscillator

with the advent of low-voltage power transistors it is possible to construct an oscillator to work
directly from the low-voltage dc source. A transformer-coupled oscillator is used (see Part 4)
and the transformer is provided with a high-voltage winding which is then used as the source for
a rectifier unit. In such a circuit the ac-waveform is a sinewave.

This type of circuit is often found in units which use cathode-ray tubes.

(c) motor-generators

are seldom seen nowadays. Basically the battery source is used to drive an electric motor which
is mechanically-coupled to a generator; the generator provides either ac or dc supplies as
required. A motor-generator is a machine in which the armature is double-wound ; one
winding is used to drive the machine as a motor while the other acts as a generator. A motor-
generator set consists of a single motor mechanically coupled to several different generators on
the same shaft.
Back to Top of Page

>>>>>>>>>>>>>>>>>>>> PAGE 17 <<<<<<<<<<<<<<<<<<<<

5.2.7 Switch-mode Power Suppliers

These units, usually referred to as "switchers", provide very compact and reasonably-light
power-suppliers but they are most complex in their design and construction and hazardous for
the unskilled or unwary serviceman.

The incoming ac-mains supply is applied directly to a bridge-rectifier thus eliminating the
bulky mains-transformer. The resulting full-wave-rectified dc is roughly smoothed with a
capacitor-filter in which neither side is earthed. BEWARE.
 This dc is then switched across the input to a transformer which has multiple windings that,
via individual rectifier circuits, provide various supplies as required. The switching (or
chopping) is carried out by means of a squarewave-driven power transistor.

One of the dc outputs is sampled to provide an error-signal which is fed-back to the

oscillator circuit to maintain control over the output level. This is achieved (for example) by
varying the mark-space ratio of the squarewave (pulse-width modulation).

A large part of these power suppliers consist of safety circuits which (for example) shut-
down the unit in the event of excessive load current. One odd feature is the so-called kick-
start mechanism. The squarewave generator and its associated control circuits require dc
supplies which are not available until the oscillator has started to oscillate. At switch-on the
rising dc from the rectifier is made to charge a capacitor and this, with a trip circuit, starts the
oscillator and then effectively removes itself from circuit.

5.2. 8 Switching Regulator IC

A similar technique is used in the integrated-circuit switching-regulators which increasingly

are replacing the linear regulators described under 5.2.4. In use these require considerably more
external components but they have the advantage that the output voltage can be greater than or
less than the input voltage provided that the total device power-dissipation is not exceeded.

Their mode of operation is similar to that of the coil-ignition circuits used in motor cars .
The input direct-current flows through an inductor and is switched at a frequency around 100
kHz by means of a square-wave oscillator; when the current is interrupted the voltage
developed across the inductor charges a capacitor via a blocking diode . A variable resistor
varies the mark-space ratio of the switching squarewave and this sets the output voltage.

A disadvantage is that the switching-oscillator can cause problems with e.m.c. and care is
required in construction.

END OF LESSON 2

END OF PART 5

Back to Top of Page

 PART 6: MEASUREMENTS AND E.M.C.
LESSON 1
6.1 WAVEMETERS

6.1.1 General

A wavemeter is a device for measuring frequency but it derives its name from the world of
physics where, historically, it was easier to measure e-m waves in terms of wavelength.

Suitable frequency-checking equipment is mandatory in an Amateur-radio Station and

details can be found on the last pages (Appendix-C) of the DTI document How to Become a
Radio Amateur. A general-coverage receiver which incorporates, or which can be used in
conjunction with, a crystal calibration-oscillator will suffice but it should be remembered that
the frequency-range of measuring equipment must be such as to cover the second and,
preferably the third, harmonic of any carrier which the station may radiate.

The following sub-sections describe the different techniques used to measure radio
frequencies. Basically the signal under test may be injected into a pre-calibrated variable
resonant circuit, heterodyned against a variable calibrated signal-source or the number of cycles
executed in one second can be counted.

6.1.2 Absorption Wavemeters

This instrument consists of an L-C resonant circuit which can be tuned over a band of
frequencies. The resonant circuit is coupled to some form of indicator to show the presence of
current in the circuit. A simple form of absorption wavemeter is shown in Fig. 1. A diode
detector, tapped down the inductor to reduce damping, has as its load a capacitor and a dc-meter.

The signal under test is coupled usually to the inductor

and the tuning capacitor is rotated until, as the circuit
resonance approaches the frequency of the signal, power is
absorbed from the test signal. The resulting circulating
current reaches a maximum value when the circuit
resonance is adjusted to the frequency of the signal.

This condition is indicated by a maximum deflection of the meter. A sensitivity control may
be included for convenience, either by shunting the meter with a variable resistor or by
including such a resistor in series with the meter, but it is not strictly necessary because the
operator can control the degree of coupling to the circuit under test. Alternatively a diode
shunted across the meter can be used to create a logarithmic scale.

>>>>>>>>>>>>>>>>>>>> PAGE 2 <<<<<<<<<<<<<<<<<<<<

Go to Top of Page
 A simple arrangement of this type has a lower limit to its sensitivity because there is a
minimum voltage at which the detector diode conducts. This could be overcome by interposing
an rf-amplifier between the wavemeter and the diode (but see below) or it is possible with
modern op-amps to build detectors which do not suffer these threshold effects.

In practice sensitive wavemeters can be built

by simply interposing a dc- amplifier between
the detector and the meter. Fig. 2 shows such
an arrangement as built by me some 50-years
ago and still giving very good service not only
as a wavemeter but also as a probe with which
to go hunting for spurious oscillations. A range
of inductors are plugged into the hand-held unit
to form a probe. (I have no doubt that its
performance could today be enhanced still
further by replacing the transistor with a FET-
input op-amp.)

** Note that a resonant circuit responds only to that frequency to which it is tuned. The
Absorption Wavemeter therefore does not respond in either a harmonic or a sub-harmonic mode
and so its primary function is the positive identification of a fundamental frequency.

** The disadvantage of the above simple Absorption Wavemeter is the inability of the diode
to conduct below a threshold value (about 0.2-volts for a germanium diode and about 0.6-volts
for a silicon diode). It is useless for example for identifying the frequency of a carrier incoming
on an aerial (unless the transmitter is very local). The solution suggested above, that a rf-
amplifier be interposed between source and wavemeter, is self-defeating because such an
amplifier could certainly generate harmonics at a power-level sufficient to register on the meter.

** The accuracy of absorption wavemeters is limited by the ability to read the calibration
marks on the dial.

6.1.3 Heterodyne Wavemeters

These wavemeters use a calibrated oscillator that feeds into a mixer circuit together with the
signal under test. The two signals beat as described under 2.2.5 and 2.7 and the result is passed
to a detector-circuit and thence to either headphones or loudspeaker. In principle these
wavemeters are but carefully constructed and calibrated superhet receivers.

** The audio output from a frequency-translator and detector combination is proportional to

the amplitude of both the input and the (strong) local-oscillator signals. Thus these wavemeters
can be very sensitive indeed and easily respond to signals incoming on an aerial. They also offer
greater accuracy than the absorption types in that built-in crystal oscillators with harmonic-
generators enable the calibration scale to be checked at intervals ; generally a moveable cursor
is incorporated so that the scale can be adjusted over the frequency-range of interest.

>>>>>>>>>>>>>>>>>>>> PAGE 3 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 ** The disadvantage of the Heterodyne Wavemeter is that it is extremely sensitive also to
harmonics of the signal under test and also produces responses to spurious signals that arise
from (a) beats between the test-signal and the L.O.-harmonics and (b) beats that arise between
harmonics of the signal and of the local oscillator. These often appear to be sub-harmonics.

** It follows therefore that, to measure the frequency of a signal, an absorption wavemeter is

required to identify the fundamental (unless it is already known) and then a heterodyne
wavemeter can be used to obtain a degree of accuracy. A block diagram of a typical wavemeter
is given in Fig.3.

6. 1. 4 Frequency Counters

** These instruments use digital-circuit techniques to actually count the number of cycles
which occur in a defined period. A typical Counter is shown in block-diagram form in Fig.
4. The essential component of these instruments is a gate circuit which blocks the passage of a
signal until the gate is "Opened" by a suitable pulse. The gate remains open until it is closed by
a second pulse or perhaps by termination of the pulse that was holding it open.

** The signal to be measured is applied to the gate-input ; if the gate is held open for exactly
1-second then the number of cycles which are passed to its output represents the frequency in
cycles-per-second (Hz).

The number of cycles to be counted can be varied by adjusting the gate width (sometimes
referred to as the "time-base"). For example a 1-MHz carrier would produce a count of 106 with
the gate set to 1-second . A 100-MHz carrier would produce the same count of 106 with the gate
set to 10-milliseconds (10 ms). Thus the control which sets the gate-width becomes
a multiplier to be used in conjunction with the display. The 10-ms position would be labeled " x
100 " .

>>>>>>>>>>>>>>>>>>>> PAGE 4 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 The output from the gate circuit is applied to a shift register as described in Part-2, Lesson-
4 under 2.2.9 The Binary Counter. When the gate is closed the condition of the various stages
of the Counter indicate, in binary form, the number of cycles which passed through the gate. The
count is held for a short period, the Counter is then reset to zero and the gate re-opened so that
the frequency is continually monitored for any change.

During the period that the count is held in the Register the result is sampled by opening
another gate - in fact a set of gates one for each stage of the Counter; the count is transferred
(a parallel transfer) to a second "holding" register for processing while the counter is recycled.
The connections from this second register are arranged so that the binary number is converted to
a decimal number before being passed to the display.

The display consists basically of a number of panels each of which contains a number of
light-emitting diodes (L.E.D.s) arranged in a pattern such that, by energising an appropriate
combination of diodes, any required number can be simulated. There is one of these panels of
diodes for each digit to be displayed. ( Alpha-numeric, or "starburst", panels are available which
can display letters of the alphabet as well as numerals.]

The complex process of selecting the requisite L.E.D.s is undertaken by specially designed
chips. The display is constantly "strobed" at high-frequency so that each panel of the display is
illuminated in sequence: to increase the display brightness the strobe frequency is increased. At
each pass the panels are corrected (up-dated) according to the latest count from the counting
register.
 >>>>>>>>>>>>>>>>>>>> PAGE 5 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

6.1.5 Standard Frequency Transmissions

Special transmitters have been set up whose purpose is to radiate a signal the carrier-
frequency of which is controlled to a very high degree. Their intended purpose is frequency-
measurement by heterodyne methods.

6.1.6 The Grid-dip Oscillator (G.D.O.)

The grid-dip oscillator (or grid-dipper) is a simple tool with two basic functions which, in
today's all-semiconductor world, is seemingly misnamed. It is an absorption wavemeter but, by
switching-in an active device, it can be used also "in reverse" to determine the resonance
frequency of an LC circuit.

** A typical circuit in its original form is shown in Fig. 5. To achieve a wide frequency-
coverage the instrument is provided with a range of plug-in coils ; thus, to simplify the coil
construction, the circuit is shown as an inverted Colpitts oscillator. A milliammeter is connected
in series with the grid leak where it measures the grid current.

** As an oscillator its action is controlled

by the RC bias-combination in the grid
circuit. With the coil close to the circuit under
test the capacitor is varied to adjust the
frequency of oscillation ; as this passes
through the resonance frequency of the test
circuit so that circuit absorbs power from the
oscillator. This power-loss lowers the
amplitude of the oscillation and this decreases
the bias which appears on the meter as a drop
in grid current. This is the "grid dip". The
Meissner circuit is often used in G.D.O.
construction.

** When the switch in the anode circuit is opened the oscillation ceases. The instrument can
then be used as an absorption wavemeter with the grid-cathode acting as a diode detector and
the meter as the indicator. With a pair of headphones or a loudspeaker plugged into the anode-
circuit jack and the switch closed the device also serves as an inaccurate heterodyne wavemeter.
With the switch open it serves as a signal monitor/tracer.

** A disadvantage of this device is the requirement for a power supplier but this is less serious
when the active device is a transistor or op-amp and an internal battery can be utilised.
However, it is a good idea to fit a L.E.D. indicator to warn that the battery-switch is closed.

>>>>>>>>>>>>>>>>>>>> PAGE 6 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

The G.D.O. comes into its own on the constructor's bench where it can be used to determine
the number of turns required in setting-up LC circuits and also in chasing spurious oscillations.
It is particularly useful in that it can provide measurements on circuits yet to be completed and
which therefore cannot be powered.

END OF LESSON 1

* * * * * * * * * * * *

QUESTIONS

1. Outline the D.T.I. amateur radio station requirements for measuring facilities in an amateur-
radio station.

2. Give the advantages and disadvantages of an absorption wavemeter.

In place of an indicating meter an absorption wavemeter is fitted with a small lamp. What is the
likely intended use for such an instrument?

3. What is the major purpose of a heterodyne wavemeter?

4. Give the basic functional details of a frequency—counter instrument. (You are not required to
describe the internal workings .)

5. A Counter is displaying 157.00582 kHz with the Multiplier set to x10, What is the frequency
being measured ?

6. You are presented with a brand-new superhet receiver which has a poor response. How would
you use a G.D.O. to check a suspicion that the aerial-input circuit was in fact operating on the
wrong ranges?

>>>>>>>>>>>>>>>>>>>>>>>>>>>> PAGE 7 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 2
6.2 ANALOGUE METERS

6.2.1 Moving-coil Meters

An analogue meter is one which measures a quantity by analogy ; most analogue meters
represent the magnitude of a parameter as an angle of rotation but some show it as a linear
(straight line) motion.

The most common of meters is the moving coil meter in which the electric current to be
measured is passed through a lightweight coil. The coil is suspended on pivots in the magnetic
field of a permanent magnet. As shown in Fundamentals-1 Lesson-4 coiling a conductor
collects the magnetic field into the same form as that of a bar magnet. The two magnetic fields
interact and a torque (a rotational force) turns the moving coil so that the fields are aligned. A
pointer attached to the moving coil indicates the degree of rotation.

As described so far the meter is useless because any current of whatever value would cause
the two magnetic fields to align end so produce the maximum possible deflection. The coil is
restrained by a small spring and the rotation is checked when the tension in the spring exactly
balances the torque set up by the test current. An increase or decrease in the current causes an
increase or decrease in the torque and so the coil deflects through a greater or lesser angle until
there is sufficient tension in the spring to balance the system once more.

[Note that the function of the spring is not to return the pointer to zero when the current is
removed ; it is an essential part of the meter- action.]

The resistance of the coil is very low and so care must be taken to limit the applied voltage.
The magnitude of the full-scale-deflection (f.s.d.) current can be controlled by selecting the
tension of the spring but this is limited by the load imposed on the pivots.

The f.s.d. current is increased by connecting a meter shunt in parallel with the meter so that
excess current flows around the meter rather than through it. The value of the shunt resistance
(Rs) is related to the coil resistance (R m ) in the same (but inverse) ratio as the current through
the meter (Im) is related to the current (Is) which flows through the shunt

R s /R m = I m /I s

where of course the shunt-current equals the total current less the meter current.

These meters are adapted for use as voltmeters by connecting a multiplier resistor in series.
This resistor is chosen so that the voltage dropped across the multiplier equals the full applied
voltage less that dropped across the meter coil. In general the voltage across the meter is so
small compared with that across the multiplier that the meter resistance can be ignored ;
however, for accurate measurements (a Grade-I Meter) the meter resistance must be taken into
account.

>>>>>>>>>>>>>>>>>>>> PAGE 8 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

These meters are adapted also for use as ohmmeters by including a small battery in circuit.
The meter indicates the current which is driven through the resistor to be measured ; because
the terminal voltage of the battery is known the meter scale can be calibrated in ohms instead
of amps.

There is a problem here in that the terminal voltage of a practical battery is not constant
throughout its life. This is overcome by including a small variable resistor whose function is to
reduce the voltage from the battery. Before making a measurement the meter connecting-leads
are first connected together (zero ohms) and the variable adjusted for f.s.d. — this point is Zero
on the ohms-scale.
 These three basic functions (and others too) are incorporated into a single instrument known
as a multimeter in which the different modes (amps, volts, ohms) are selected by switches and
also a number of ranges are offered by switching different shunts and multipliers.

The basic instrument can measure only direct quantities but it can be adapted for ac
measurements by including a small rectifier unit (usually a bridge rectifier — see Part-5).
There are three things to remember when using such an instrument on ac:

(a) the instrument can be calibrated to indicate either the peak waveform voltage, the average
voltage or the rms voltage (see Part-7)

(b) the instrument will function satisfactorily only up to a frequency which depends on the
effects of stray capacitance and stray inductance within the instrument

(c) the reading given by the instrument will be affected by the waveform of the signal ; i.e.
distortion will change the rms, average and peak values. Normally meters are calibrated in terms
of rms values for a sinewave.

Rough measurements of capacitance can be made by measuring the ac-current which a

capacitor passes for a known applied voltage. Power is measured (within the frequency range of
the Instrument) by dissipating the applied energy in a known resistive load and using the meter
to indicate the voltage developed across that resistor.

Switches are both expensive and troublesome and sometimes the mode switch is dispensed
with — or simplified — by providing alternative connections to a multimeter.

Multimeters require care in use because it is too easy to attempt a voltage measurement with
the instrument set for current measurement. The disadvantage of the versatile moving-coil (rn-c)
meter is its lack of robustness.

Delicate rn-c meters (say up to 1-mA f.s.d.) should always be kept with their terminals
connected together when not in use ; good quality electronic rnultimeters (see next section)
arrange for such a short-circuit connection to be made when the instrument is switched off. The
short circuit allows current to flow when the meter-movement is disturbed ; the coil, moving
within the permanent magnetic field, acts as a generator. The magnetic field which results reacts
with the permanent field and forms a braking action on the movement which protects it from
damage.

>>>>>>>>>>>>>>>>>>>> PAGE 9 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

A serious disadvantage of the moving-coil meter is that it demands a current from the circuit
under test. Connection of such a meter therefore changes the conditions in the test circuit and so
produces a false reading. This is demonstrated in Fig. 6.
 ** Diagram (a) shows a simple potential-
divider network with a voltmeter connected
across the lower half. There are two ways of
considering this circuit:
(a) the current drawn by the meter is added to
that drawn through the upper resistor and so it
lowers the voltage at the junction.

(b) the meter resistance is in parallel with the

lower resistor and so the lower resistor is
effectively reduced in value : because the
potential divider has been modified the
potential at the junction is lowered.

** Diagram (b) shows the error caused by attempting to measure the current. The resistance of
the meter is added to that of the resistor and so the circuit current is reduced. If the resistor has a
value at least 10-times that of the meter the error is not likely to be significant in practice but,
where errors below 10% are cause for concern, it is necessary to calculate the correct result from
the meter reading.

Clearly the addition of multipliers changes the overall resistance of a meter and, in a multi-
range multimeter, some criterion is required to describe the overall resistance. The value of each
multiplier depends on the f.s.d. current of the movement and on the intended f.s.d. voltage for a
particular range. For example, if the instrument uses a rn-c meter with a f.s.d. of 100 μA then, by
Ohms Law, on the 10-volt range it will present a resistance of 10/0,0001 ohms or 100 kilohms.

On the 100-volt range its resistance will be 100/0.0001 or 1 megohm. This linear relation
between f.s.d-voltage and overall resistance is expressed as 10-kilohrns per volt. Had the
multimeter been constructed using a rn-c meter of 1-mA f.s.d. then the multimeter would have
been quoted as 1-kilohm per volt.

In complex circuits the effect of connecting a low-resistance meter can have far-reaching
results and this is particularly so in circuits which employ dc-feedback loops which are very
common in modern electronics.

>>>>>>>>>>>>>>>>>>>> PAGE 10 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

6.2.2 Electronic Meters

This description today usually means a digital meter (described later) but the operation of
those is much more sophisticated.

The electronic multimeter drives the analogue device through a dc-amplifier and it is not too
difficult to arrange that the amplifier presents a very-high impedance at the input terminals.
Range-changing can be effected both by use of multiplier resistors and by changing the
amplifier gain usually by switching the feedback circuit. For current measurement a set of
shunts must be provided across the meter input but this is a straightforward matter because the
instrument does not draw current from the test path. The difficulty with these instruments
however lies in designing the dc-amplifier so that it does not suffer from drift problems with
changes in ambient temperature or changes in the power-supply. In addition high-impedance
circuits are very prone to pick-up hum.

6.2.3 Meter Accuracy

Analogue meters in all forms offer the great advantage of giving a pictorial display of circuit
conditions. It is immediately obvious that the current is either rising, falling or reaching a
steady-state. The greatest disadvantage lies in the inability to read them very accurately.

Consider for example a meter which is calibrated from 0 to 10 mA. It will have 10 major
divisions which indicate steps of 1 mA and each of these will most probably be sub-divided into
five smaller divisions of 200 μA each. (A physically-large instrument may well be sub-divided
into 10 divisions of 100 μA each.) . The eye can distinguish between a half-division and a
whole division and so the meter can be read to an accuracy of 100 μA. But that is only part of
the story.

When the meter is operated at f.s.d. the error is 100 μA in 10 mA ... or 1%. When the meter
is indicating 1 mA the error becomes 100 μA in 1 mA which is 10%. Thus it is important to
choose an analogue meter such that it operates at or close to its f.s.d.

6.2.4 Linear Meters

These give their indication in the form of an extending line and are most commonly found in
LED displays known as bargraphs. Variations of the parameter to be measured are converted to
a set of numbers (using an analogue-to digital-convertor or ADC) and each number is
represented by an LED in the display. As the parameter rises through each pre-determined level
so the appropriate LED becomes illuminated. The display is seldom used for quantitative
measurement but only to present a general picture such as signal-level monitoring in a tape-
recorder.

The display may take two alternative forms either a bargraph, in which the LED's are
illuminated to form an extending line, or a moving-dot display in which only one LED
(representing the largest number) is illuminated at any one time.

>>>>>>>>>>>>>>>>>>>> PAGE 11 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

6. 3 DIGITAL METERS

Essentially a digital meter is one which displays its measured result as a number; i.e. a
string of digits with a decimal point. However, in its internal workings, it is if anything more
analogue than a m-c meter. The nomenclature of these instruments causes puzzlement when
reference is made to (say) a 3½-digit display ; the ½-digit is the decimal point.

The first stage in a digital meter is an analogue-to-digital converter whose function is to

translate the variable input (analogue) signal into a number. It is this process which gives rise to
the meter's drawback. The sampling and ADC process has to be constantly repeated to derive a
new number whenever the input value changes.

As explained in the previous Lesson a digital display is illuminated one digit at a time when
it is up-dated from the display memory appropriate to the digit. The ADC must keep the display
memory in step with the input-signal variations.

When the input signal changes rapidly it may require several cycles of this up-dating process
before the meter catches-up with the input signal and so digital displays tend to be confusing in
that there is always uncertainty as to whether the result is increasing or decreasing. Some meters
incorporate a bargraph display also which gives this information. Often the need to wait until the
display settles makes the meter less useful than an m-c instrument. In tuning operations, where
the operator is looking for a maximum or a minimum meter reading, a digital meter is fairly
useless.

There are several techniques available for digital measurements but a basic idea is given in
the following paragraphs for those interested; this is not required for the R.A.E.

A ramp voltage or ramp-waveform is one which steadily increases its value up to a fixed
level ; it then rapidly reverts to zero ready to start again. It may be generated by passing a
current of strictly-controlled value into a good-quality capacitor. In such a waveform there is a
fixed relation between the elapsed time since the waveform commenced and the value to which
the voltage has risen. (Further information will be given under Pulse Techniques.)

It is easy to measure time accurately by counting the number of cycles of a crystal-controlled

oscillator (see Frequency Counters in previous Lesson). The oscillator waveform is passed
through a gate-circuit which is opened as the ramp commences ; counting then continues until
the gate is closed after which the display indicates the time (in cycles) for which the gate
remained open.

The ramp waveform is applied to a voltage-comparator circuit where it is compared to the

voltage of the applied test-signal. When the two voltages are equal the gate is closed. Thus the
display (a number of cycles representing elapsed time) also represents the time taken by the
ramp to reach the voltage of the test signal. The relation between elapsed time and ramp-voltage
is known (in fact it is designed) and so the display can be calibrated in terms of volts.

>>>>>>>>>>>>>>>>>>>> PAGE 12 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

A digital meter can give an accurate reading to several decimal places which is little affected
by considerations of f.s.d. as outlined above for the analogue meter. However, its habit of
constantly re-evaluating the signal by counting-up from zero makes the read-out somewhat
baffling with rapidly-changing signals.

Digital systems are used in the construction of multimeters. They have two great advantages ;
(a) their very-high input impedance prevents loading of the circuit under test (b) they can be
constructed to limit the voltage they may apply to the circuit under test thus avoiding damage to
integrated circuits of the CMOS type.
6. 4 PROBLEMS of HIGH-IMPEDANCE METERS

Electronic meters offer input impedances up to many hundreds of megohms and are a
blessing indeed in low-current easily-damaged electronic circuits. At the other end of the scale
however, in high-current circuits, they can lead the unwary astray.

Consider a motor-car which refuses to start. On turning the ignition key all the electrics
come on but, as soon as the starter switch is engaged, everything dies. Flat battery ? Yes,
except that, when the starter switch is released, all the electrics come back to life.

End-of-battery-life is a reasonable diagnosis except that the car behaved perfectly yesterday.
Clearly there is a bad connection somewhere which can be sorted quickly with the aid of a
meter. Of course, the most convenient meter to use is that handy little electronic digital meter.

Car electrics handle anything up to 35 amps and, in the starter-motor circuit, anything up to
100 amps ; at 35 amps a corroding connection offering only 0.3 ohms would reduce the
available battery voltage to 2 volts. After 10 minutes work with the digital meter it is clear that
here is another of those electrical nightmares in which there is nothing wrong with the device
except that it will not work.

The cure is to put the electronic meter away and do the circuit checking with a spare
headlamp bulb and a couple of lengths of wire. Given a dry-terminal resistance of 50 ohms the
electronic meter will indicate a good connection because it draws probably 1μA or less — the
drop in terminal voltage at 50 ohms amounts to 50 μV. The headlamp bulb requires around 4
amps and it will not function with 50-ohms in series.

In high-current circuits always use a high-current indicator but, above all, keep all connections
clean and lightly smeared with vaseline (chassis grease is a good second choice). Finally
remember that, in mobile amateur equipment, a current in excess of 20 amps will be drawn by a
rig which is transmitting on full power.

>>>>>>>>>>>>>>>>>>>> PAGE 13 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

6.5 POWER and SWR METERS

6.5.1 Dummy Load and Detector

One method of measuring power has been indicated already under 6.2.1 wherein the power
is dissipated in a resistor of known value and the voltage which is developed across that resistor
is measured. Reference to Ohms Law however reminds us that power is proportional to the
square of the voltage and so this simple method of measuring power produces a non-linear meter
scale which is cramped toward the high-power end.

Nonetheless this system is often used ; I own a RF Power Meter which is simply a resistive
dummy-load to which is attached a diode detector and m-c meter. A switchable meter shunt
reduces the meter sensitivity to give two ranges 0-50 watts and 0-100 watts.
 RF-power has a nasty habit of leaking and, if the power does not reach the dummy load,
then it cannot be included in the final power reading. The entire apparatus must be enclosed in a
metal screening-box fitted with a coaxial connector so that a coax feeder can be used to connect
it directly to the transmitter output. The case must be ventilated and be large enough to transfer
the heat to the atmosphere at a rate sufficient to limit the temperature rise.

Additionally the resistive load and its case must be designed so that, over the desired
frequency range, it looks like 50-ohms to the feeder else it will set up a standing-wave that will
destroy both the transmitter and all possibility of accurate measurement.

The diode detector poses a new problem in that, depending on the frequency, it may act as a
peak detector or as a something-less-than-peak detector ; to measure power it is necessary to
use rms values (see Part-7). One simple way to calibrate the instrument is to drive it with a
direct current because this can be measured, either in voltage or current, with normal workshop
meters. For each power setting of the wattmeter the true power can be calculated. This only
works of course within the passband set by the stray reactances and the time-constant of the
diode detector.

A second method is to sample the voltage or current. A toroidal transformer (see

Fundamentals-1 Lesson-5) can be used to extract a small proportion of the rf-current.
Alternatively a simple voltage-divider, either resistive or capacitive, can be used to sample the
rf-voltage. Provided the resistance of the load is known accurately the power can be calculated.
This method has the advantage that detection is carried out at low level.

A more usual method is to use either a directional coupler or a so-called bridge arrangement
which forms part of an swr meter. These are described in the next Lesson.

END OF LESSON 2

* * * * * * * * * * * *

QUESTIONS

1. A milliammeter with f.s.d. 1-mA and resistance 120-ohms is to be used to make a simple
multimeter with voltage ranges up to 10 volts and 100 volts and current ranges up to 100 mA
and 1 amp. Calculate the values of the multiplier and shunt resistors that will be required.

2. The diagram shows a valve with an anode-load resistor of 10,000 ohms. The anode voltage is
measured with a multimeter offering 1,000 ohms/volt. On the 200 volt range the meter shows
the anode voltage as 140 volts. Is the correct answer 200 volts, 144 volts or 100 volts ?

3. A cell of 2 volts is connected across a 2000-ohm resistor. If the above 120-ohm milliammeter
is connected in series what would you expect it to read ? What voltage would a high-impedance
voltmeter record across the resistor with the milliammeter in circuit?

4. What is an electronic meter as distinct from a digital meter ?

5. Why would an LED array make a suitable tuning indicator for an A.M. receiver but not be of
use in an f.m receiver or as a means of checking the performance of a transmitter ?

>>>>>>>>>>>>>>>>>>>> PAGE 14 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 3
6.5.2 Dummy Load and RF Ammeter

The method given under 6.5.1 in the previous Lesson results in a power calculation made
from the expression

power = E2/R

The method described here also relies on a calculation but uses the formula

power = I2R

RF current can be measured directly by means of either a hot-wire ammeter or a

thermocouple meter both of which rely on the heating effect of an electric current.. There is a
frequency limitation however because, as frequency rises, so stray capacitances tend to bypass
the heat element.

The Hot-wire Meter A short length of wire is stretched between two anchor points and
tensioned at its centre by a cord which, after passing around a pulley, is attached to a spring. The
pulley carries the meter's pointer. Any current which flows through the wire causes it to heat and
so expand. This expansion is taken-up by the spring which, in drawing the cord over the pulley,
moves the pointer. This is a fairly robust meter.

The Thermocouple Meter A thermocouple, in simple terms, consists of two dissimilar

wires with their junction attached to (but insulated from) a small heater. The current to be
measured is fed through the heater which thus heats the junction. Because the junction contains
two dissimilar metals an emf is generated whose magnitude depends on the temperature of that
junction. The output from this thermo-electric transducer is fed to a sensitive microammeter.

Because these meters depend on the heating effect of a current they register power whether
the current is dc or ac. As a consequence they can be calibrated by passing a direct current
whose value is easily measured by means of a standard dc meter. They give a true indication of
power irrespective of waveform ; (see under RMS Values for a Sinewave in Part-7).

To measure power a dummy load is used as before but the rf-ammeter is connected in series.
The power is calculated as I2.R where R is the value of the dummy load. The square law results
in a non-linear scale but this is often corrected in the mechanical and/or electrical design.

>>>>>>>>>>>>>>>>>>>> PAGE 15 <<<<<<<<<<<<<<<<<<<<

6.5.3 The Directional Coupler

 Fig. 7 shows one form of directional coupler. It consists of two pieces of coax feeder joined
side by side which are coupled together in two places by allowing the electric and magnetic
fields to leak through holes. All four ends are terminated in coax couplers.

The diagram shows power injected into the left-hand end of the top cable and removed from
the right-hand end. In fact the coupler can be used either way round with the main power-flow
in either cable.

Power which leaks through the two holes propagates in both directions along the second
cable. Consider first the power which travels from left to right along the second cable. That
which leaks through the first hole and travels to the second hole is in phase with that which
leaks through the second hole. Thus the two signals travel together to the right-hand end of the
lower cable and add together in whatever load is connected there.

Power which leaks through the second hole but travels toward the left-hand end of the lower
cable arrives back at the first hole but is delayed in phase by twice the time taken for the signal
to travel the distance from one hole to the other (i.e. there and back). At the left-hand end of the
lower cable these two signals tend to cancel ; the degree of cancellation depends on the relation
between frequency (i.e. wavelength) and the distance which separates the holes. The device
therefore is frequency conscious ; also, if there is to be any appreciable cancellation, the hole-
spacing must be comparable in size to the wavelength and this limits use of the directional-
coupler roughly to frequencies above 30 MHz.

It is found that, for any given frequency or small range of frequencies, the outputs from the
lower cable are a fixed percentage of the power which flows on the main cable. Should the main
feeder path be mismatched then the current and voltage relation becomes reactive and this
affects the addition and subtraction of signals in the second cable.

The power which emerges at the right-hand end of the lower cable is always a fixed
percentage of the power traveling from left to right on the main feeder — mismatched or
matched. The power which emerges at the left-hand end of the lower cable is always a fixed
percentage of the power traveling from right to left ; i.e. power which is reflected from a
mismatched load. Hence a simple subtraction gives the power being absorbed by the load and a
comparison of the two outputs enables a calculation of the s.w.r.

>>>>>>>>>>>>>>>>>>>> PAGE 16 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

 A directional-coupler has to be calibrated before it is of much use. To do this power is
injected at (say) the left-hand end of the cable which is to be the main-feed line ; the other three
connections are terminated in matching dummy loads. A sensitive detector (such as a receiver
with a calibrated attenuator) is now used to measure the signal levels at the two right-hand end
connectors ; this gives the ratio of signal-level on the main line to signal-level coupled-out of
the "forward" side-connector. A similar measurement at the "reverse" side-connector should
show a very large attenuation of the signal.

The power is now fed along the main feeder in the opposite direction and the process
repeated. The directional-coupler is now ready to indicate the forward and reverse power-flows
on the main feeder as a proportion of that main flow (this is the Coupler's attenuation factor).

By itself the directional-coupler can now be used for measuring s.w.r. Additionally, when
the output level at a side-connector is measured (by use of a receiver and a signal-generator) this
can be used with the attenuation-factor to calculate the power-flow on the main feeder. Note that
the attenuation- factor must be measured for every (small) range of frequencies

Some directional couplers are made by threading a small length of wire down a section of
feeder between the outer and inner conductors. This forms a loop, short with respect to the
wavelength, which is inductively-coupled to the main feeder. The voltage/current ratios which
are induced along the loop differ along its length and its operation can be argued in a manner
similar to that above.

Yet another method is to use printed-wiring techniques to lay down a small length of
transmission line which, in appearance, is a balanced feeder. It is used as two separate feeders
coupled in the manner of the loop described above.

Note that all forms of directional-coupler are frequency (more accurately wavelength)
conscious and they must be calibrated for each frequency of use.

This device is not subject to examination for the R.A.E.

6.5.4 The S.W.R. Bridge

This device works on a similar principle but it derives the two signals (to be added and
subtracted) in a different manner.

Firstly it measures power by sampling both the feeder current and the feeder voltage and
produces a meter indication which is the product of the two values (see Ohms Law in which
power equals volts x amps). One such circuit is shown in Fig.8.

>>>>>>>>>>>>>>>>>>>> PAGE 17 <<<<<<<<<<<<<<<<<<<<

 The current is sampled by means of
a toroidal transformer as described under
1.10.5 in Lesson-5 of Fundamentals-1 . The
resultant current is fed into two identical
resistors ; this provides a centre-tap from
which, at opposite ends of the transformer, two
1800 signals are available.

The line voltage is sampled by simply

tapping-off a proportion of the signal through a
small capacitor. (A resistor can be used here
but, not only does it dissipate some of the line-
power, it must have considerable physical size
to enable it to get rid of the heat generated
within. ) Note however that, for a high-power
meter, the capacitor must also be capable of
carrying the current which results and also of
withstanding the line voltage that will result
should the feeder not be properly terminated
when full power is applied. Two outputs are
provided each one being a product of current
and voltage but the two currents used are
1800 displaced.

To set-up the circuit power is fed into the matched feeder, first in one direction and then in
the other, and the appropriate variable-resistor is adjusted to achieve zero power indication. The
device is often arranged with two meter-movements one of which has a right-hand zero. The
scale is calibrated so that the pointer which moves to the right indicates the forward power and
the pointer which moves from right to left indicates the reverse power. Where the two pointers
cross a special scale indicates the standing-wave ratio.

In theory the device works equally in either direction. In practice however the "forward"
circuit is desensitised so that the two meters give roughly equal indications over the normal
operating range which makes possible the cross-pointer indication of swr. Thus a wrong
connection of these instruments could result in damage.

In these devices for use below 30 MHz the toroidal transformer usually takes the form of a
ferrite ring through which the centre conductor of a coax feeder is threaded. The secondary
winding is a few turns threaded around the ferrite. Obviously the ferrite ring must be suitable for
the frequencies to be used and it can operate only up to that power level where magnetic-
saturation begins to occur.

Properly designed these instruments are not frequency-conscious and so they can be
calibrated in terms of Absolute Power provided of course that bandwidth limitations are
observed.
** For the R.A.E. you should be familiar with the formula which relates the degree of
mismatch with the s.w.r. i.e.

s.w.r. = Vmax/Vmin = Imax/Imin = Zo/R

>>>>>>>>>>>>>>>>>>>> PAGE 18 <<<<<<<<<<<<<<<<<<<<

where Z 0 is the characteristic-impedance of the feeder and R is the resistance of the load (the
expression assumes some complication when the load is reactive). See also under 2.3.6
Standing Waves and SWR in Lesson-7 of Fundamentals-2.

6.6 THE OSCILLOSCOPE

6.6.1 General

The above methods of measuring power are suitable for continuous-wave emissions and,
theoretically, for some forms of modulated waves, For example, the division of power between
the carrier and the sidebands is known and so it is possible to make allowance for any
modulation. There is a very real problem however in that the sideband power increases with the
degree of modulation and so determination of the modulation index is necessary.

Power measurements with ssb transmissions cannot be made with these simple techniques
because the output from a ssb transmitter is zero until modulation is applied. While the peak
power could be measured by applying a continuous audio tone (a single audio frequency) few
P.A. stages would survive such treatment for more than a few seconds. As with a.m. signals
there is still the problem of determining the degree of modulation.

The oscilloscope gives a pictorial display of waveform and enables these difficulties to be
resolved. It can also be used for straightforward measurements in place of a voltmeter (from dc
up to the maximum usable frequency of the oscilloscope), to measure waveform distortion and
to determine the modulation index.

6.6.2 The Cathode-ray Tube

The heart of an oscilloscope is the cathode-ray display tube which, in reality, is an

overgrown valve. Its two main parts are the electron gun and the phosphor-coated screen which
is sprayed on to the inner surface of an evacuated glass envelope.

The electron-gun is named because its purpose is to shoot a pencil-beam of electrons at the
screen which glows at the point of impact. By moving the beam it is possible to trace diagrams
on the screen. The glow lasts for only a short time after the beam has moved on and, by
constantly rewriting the diagram, a moving display is formed. Use of long-persistence screens
makes it possible for slow-moving or short-duration signals to be displayed but, of course, such
tubes blur signals which change rapidly.

The gun consists of a heated cathode and a grid which, as in the triode valve described under
3.12.2 (Lesson-2 of Electronic Amplifiers), produces a controlled flow of electrons to an anode.
In the electron gun however the anode has several parts often referred to as anode-1, anode-2
and anode-3 ; by varying the relative potentials on these components it is possible to focus the
electrons into a fine beam.

>>>>>>>>>>>>>>>>>>>> PAGE 19 <<<<<<<<<<<<<<<<<<<<

The gun also possesses two pairs of plates, one pair positioned horizontally and one pair
vertically between which the beam passes on its way to the screen. Voltages impressed between
the plates of each pair cause the beam to bend and so change the position of the glowing spot
seen on the screen. In line with mathematical convention horizontal movement of the spot is
referred to as the X-direction and vertical movements as the Y-direction.

In most general usages the horizontal movement is used to indicate time and so the
waveform which causes this deflection is referred to as the Timebase. Any signal which causes
vertical deflection of the spot, in conjunction with the Timebase, forms a waveform trace across
the screen which becomes a plot of voltage against time.

The screen has a transparent cover on which is inscribed a graticule - a grid of squares -
which facilitates measurements on displayed waveforms. This is used in conjunction with
calibrated controls.

6.6.3 Circuit of an Oscilloscope

Fig.9 shows the main features of an oscilloscope in block-diagram form. For the display tube
the power-supplier produces a negative extra-high tension (e.h.t.) supply around 2-kV - which
can be very unpleasant if not dangerous. The beam-deflection plates must be held at or close to
the final anode potential if they are not to repel the electron-beam and so the vertical-deflection
amplifier (the Y-anplifier or Y-amp) needs to deliver its output waveform relative to this
potential. For both safety and convenience the input to the Y-amp must have one leg grounded
or, at the very least, close to ground potential and so the tube is operated with the final anode at
zero and the cathode taken to -2 kilovolts.
 >>>>>>>>>>>>>>>>>>>> PAGE 20 <<<<<<<<<<<<<<<<<<<<

The filament (or heater) of the tube must be supplied from a separate well-insulated winding
on the mains-transformer.

The e.h.t. is applied across a resistive potential-divider to obtain the various potentials
required for the tube operation. Generally anode-1 & anode-3 are taken both to the full supply
voltage but anode-2 to a variable resistor and this provides the Focus control. Anodes 1&3 are
discs with a hole through which the electron beam passes while anode 2 is cylindrical. The
lowermost resistor provides cathode bias for the gun and a variable-potential applied to the grid
is the Brilliance control ; it sets the beam-current.

The Timebase Generator can be switched to be free-running (i.e. it oscillates) or to produce

an X-deflection only when prodded or triggered. In either mode it may be fed with a
Synchronising or triggering signal which ensures that it keeps in step with the applied waveform
so as to produce a steady display. This trigger waveform is derived, either from the input
waveform or from a separate External Trigger input, in a Trigger or Synchronising Amplifier.

The timebase waveform has a sawtooth shape that sweeps the spot steadily across the screen
from left to right and then rapidly returns it to the left-hand starting position ready to start the
next sweep. This waveform is fed to the X-amplifier which provides a facility to control the
amplitude (the X-Gain Control) and also mixes it with a variable dc-signal which adjusts the
position of the trace horizontally across the screen.

The waveform under test is applied to a Y-amplifier which provides a Y- Gain facility and
also mixes in a second variable dc-signal which adjusts the position of the trace vertically across
the screen. There is often another optional input to this amplifier in the form of a square-wave
generator with a fixed amplitude which is used to calibrate the oscilloscope in the Y-direction
thus facilitating accurate measurements.

There are many variations in oscilloscope design which depend largely on the major use for
which the instrument is intended but, essentially, they all follow this basic layout.

The uses of an oscilloscope for measurements likely to be needed in an Amateur Station are
given in the next Lesson.

END OF LESSON 3

>>>>>>>>>>>>>>>>>>>> PAGE 21 <<<<<<<<<<<<<<<<<<<<

QUESTIONS

1. State the the two methods of measuring c.w. power using a dummy load and rf-meter.

2. What problem arises in trying to measure power with this technique when the carrier is
modulated ?

Why is this technique useless for ssb transmissions ?

3. In your own words describe the basic action of a reflectometer ; i. e. a directional-coupler or

a s.w.r. bridge arrangement. (The R.A.E. does not require a detailed knowledge of their
operation.)

4. In broad outline only what facility does an oscilloscope provide ?

Given that you have available only an oscilloscope with a 1-volt peak-to-peak squarewave
calibration-waveform can you put together an experiment to check the emf of a battery believed
to be 6-volts but encapsulated and with all labeling destroyed ?

6. RF power is to be measured using a dummy load and an oscilloscope as a voltmeter. So far in

this Course we have not dealt with the difference between peak, average and R.M.S. values of a
sinewave but can you suggest why measuring the peak-voltage of the carrier will not yield the
correct power ?

>>>>>>>>>>>>>>>>>>>> PAGE 22 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 4
6.6.4 Use of the Oscilloscope

An oscilloscope can be used to make most of the measurements carried out by standard
multimeters plus a few more which are difficult without the pictorial display. For some jobs the
'scope is perhaps a bit clumsy but it is capable of great accuracy.

The display tube is normally electrostatically deflected, which means that the trace responds
to a voltage input, and so current measurements are not easy without introducing an impedance
into the current path. Elderly valve-operated 'scopes may be found which are fitted with
deflection coils to facilitate current measurements by magnetic deflection ; although it may be
thought that such devices would be useful, with modern current-operated semiconductor devices
the need to open the circuit for series-connection means that it is usually easier to make voltage
measurements and to calculate the current.

Measurement of Direct Voltages

This is a simple matter with a 'scope. Although the timebase is not required it is best to allow
it to run so that a straight line appears on the tube. A stationary spot left for some time is likely
to damage the screen-phosphor.

With the 'scope input shorted the Y-shift control is adjusted to bring the line on to one of the
graticule lines. The direct voltage to be measured is then connected and the new position of the
line noted ; if the line goes off-screen or fails to move sufficiently then the Input Range selector
must be set appropriately.

The screen graticule may now be calibrated by using the internal calibration waveform or by
applying a signal from a known calibrated source (e.g. a Standard Cell as described in
Fundamentals-1).

Measurement of Alternating Voltages

The basic technique for measuring the amplitude of an alternating signal is just the same but
there are a couple of pitfalls. The timebase should be set so that several cycles are displayed
which makes it much easier to set the waveform against the graticule lines.

The easiest parameter to measure is the peak-to-peak voltage ; i.e. from the most negative
part of the waveform to the most positive part of the waveform. Indeed for irregular (non-
sinusoidal) waveforms this is the only practical measurement. As shown in Part 7 however the
average value for a sinewave between the two peaks (or crests) is zero and so this measurement
is only a step on the way.

At the peaks of a sinewave the maximum value of the voltage will drive a maximum value
of current through a resistive load and so produce a maximum value for the power which is
being dissipated. Between the peaks the voltage is zero and power is not dissipated at all. We
need to find a value, somewhere between these two, which represents the true power being
dissipated: we are looking for that value which gives the average power that is dissipated over a
complete cycle, the value that makes the heating effect of an alternating current the same as the
heating effect for a direct current.

>>>>>>>>>>>>>>>>>>>> page 23 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

This value is known as the r.m.s.(root-mean-square) value of the waveform and it is derived
for you in Part 7 of this Course ; for the R.A.E. you do not need to be able to derive this value
but you must understand its meaning.
 The peak (or crest) value is found by halving the peak-to-peak value; this is a more
accurate method than trying to estimate the position of the zero line. From this measurement the
RMS Value is obtained by dividing by root-2 (rough rule-of-thumb either multiply by 0.7 or
divide by 1.4).

Peak-to-peak values are measured with an oscilloscope by comparing the waveform with the
calibrated graticule or a calibrated Shift control; from this the R.M.S. value is found by
calculation.

Measurement of Phase Difference

Because an oscilloscope draws a picture of the waveform it is not difficult to observe phase
shifts provided that both waveforms can be displayed simultaneously. There are three ways to
achieve this facility:

(a) The best and most expensive way is to equip the display tube with two separate guns.

(b) A single gun may be fitted with a beam-splitter plate which, after the beam has passed
through the X-deflector plates, divides it into two separate parts each of which passes through its
own Y-deflector plates.

(c) A single-gun tube is employed but an electronic switch alternates the display between the
two input amplifiers and 1 simultaneously, switches the shift potential to give the impression of a
2-gun display.

Switching can take place either during the flyback period between traces or by using a
chopper technique in which the switching takes place at a relatively high frequency. Provided
the switch operates with truly 'fast' transitions the interruption of the displayed waveform is not
a problem.

Chopper techniques are particularly useful when observing slowly-changing waveforms

because they eliminate the flicker caused by displaying each "trace" alternately.

With two traces displayed on the screen it is easy to observe any phase-difference but not so
easy to make an accurate measurement. The

>>>>>>>>>>>>>>>>>>>> PAGE 24 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

technique used is illustrated in Fig. 10 . The time-base is adjusted so that alternate sweeps are
displaced by 1800 and each positive-going crest is accompanied by a negative-going crest. The
cross-over points give an accurate position for the zero line and also are accurately separated by
1800.
 The distance between crossovers, measured
against the graticule (or with a small ruler),
gives a value for 1800 and the displacement
between cross-overs is measured in the same
units. A direct comparison of these two figures
enables the phase- shift to be calculated in
degrees.

For such measurements to be successful it is

necessary of course that the two input and
amplifier circuits have very similar
performances with regard to both frequency and
phase ; i.e. they must produce identical signal-
delays.

Measurement of Distortion

Small amounts of waveform distortion are not very noticeable on an oscilloscope trace
below about 5% . In general there are three types of distortion which it is informative to
recognise:

(a) Second-harmonic Distortion is characterised by a lop-sided waveform in which the "half-

cycles" on one side of the zero-line have longer durations than the half-cycles on the other side

(b) Third-harmonic distortion produces a balanced waveform with the successive half-cycles
identical but characterised by flat or double-humped crests with too-steep skirts between

(c) Distortion produced by overloading a negative-feedback circuit is very distinctive because

the rounded shapes are replaced by sharp spiky waveforms. Negative feedback does its best to
cancel out distortions but, once the circuit has reached its limit, the failure of feedback control is
spectacular and the onset of distortion is very sudden.

It is not far from true to say that there is little point in measuring distortions which are easily
visible to the eye. Quantitive measurements are most useful when the distortion is small (the
harmonic-amplitudes are relatively low) and when, of course, they are most difficult to detect.

For such measurements an oscilloscope is put into differential mode (often designated as
Y1-Y2). The (hopefully-pure) waveform from the signal generator is fed also to one input of the
'scope ; the other input to the 'scope is connected to the output of the device under test. The two
amplifiers are connected so that their outputs are 1800 out of phase and so tend to cancel. With
suitable adjustment of gains those components which are present at both the input and output of
the equipment under test

>>>>>>>>>>>>>>>>>>>> PAGE 25 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

are cancelled and so the only signal which is displayed is those small— amplitude components
which are generated within the equipment namely the products of distortion. Because the main
signal components have been eliminated it is now possible to raise the gain of the oscilloscope
amplifier and so render the distortion more visible,

Once again however this trick is only successful if the two oscilloscope amplifiers have similar
performances in both frequency and phase.

** Measurement of Modulation Depth

Few Amateurs work straightforward a.m. these days but you should be familiar at least with
methods used to check modulation depth both with double-sideband signals and ssb signals.
Should you under-modulate then the only problem is that your signal will be less successful in
making contacts; over-modulation however produces spurious and illegal transmissions which
can spread over a surprising amount of the spectrum.

With double-sideband amplitude-modulation the carrier amplitude is made to vary with the
value of the modulating waveform ; over-modulation results in the carrier being interrupted for
short periods at the (-ve) peaks of the modulating waveform.

Fig. 11 shows in diagram (a) the normal representation of an amplitude modulated carrier with
low-level modulation, in diagram (b) the same carrier with 100% modulation and, in diagram
(c) the result of over-modulation. To measure the depth of modulation the maximum and
minimum amplitudes of the carrier are required as described in Fig. 4, Lesson 2 of
Fundamentals 2.

The most accurate way to make this measurement is shown in diagrams (d, e, f). Here the
timebase has been switched-off and the X-amplifier is driven with the modulating waveform. As
the modulation increases from (say) negative-peak to positive-peak so the trace moves from left
to right across the face of the tube ; at the same time the modulated carrier increases from a
minimum value to a maximum value.
 >>>>>>>>>>>>>>>>>>>> PAGE 26 <<<<<<<<<<<<<<<<<<<<

This action is reversed during the following half-cycle of the modulating waveform. The
display should be straight-edged and trapezium shaped as shown ; variations from this form are
evidence of distortion.

** Measurement of Power with S.S.B.

A single-sideband transmission consists of side-frequencies only and so

(a) a signal is present only when modulation is applied
(b) signal level is determined by modulation amplitude.

The term modulation index has little meaning.

The only practical way to measure a s.s.b. signal is to determine the maximum power-level
known as the peak envelope power or p.e.p. Such a measurement has two possible purposes:

(i) to determine the performance of a P.A. stage

(ii) to control the signal-level that is being radiated.

To measure the first of these it is necessary (in theory) only to increase the modulation
amplitude until a limiting condition is achieved. However, in use, a P.A. produces maximum
output intermittently mainly as a result of speech explosives (such as the letters t,d,k,p) and so it
is uneconomical to build a P.A. stage which is capable of delivering its maximum output
continuously. Thus the above power-measuring technique would destroy the equipment.

Short-duration peaks of power can be measured by use of a slugged meter; this is arranged
with a shunt capacitor and a diode to have a fast charging-time but a slow discharging-time .
Although such a meter tends to dwell on the peaks it is not a very accurate method for very short
pulses of power such as are encountered in speech waveforms.

To obtain a modulation waveform which will deliver short-duration peaks in a controlled

fashion a two-tone oscillator is used. The outputs from two separate audio oscillators are
combined to produce a beat ; when their amplitudes are adjusted to the same level the beating
signal has the appearance of a 100%-modulated carrier in that the amplitude varies from zero to
twice the amplitude of either.

Fed into an s.s.b. transmitter this produces a sideband signal which beats in the form shown
in Fig. 12(see also Lesson-5 in Fundamentals-2)

This rf waveform can be displayed on an

oscilloscope connected across a dummy load
and the peak-to-peak voltage determined. From
this the p.e.p can be calculated using the value
of the dummy load ; the maximum value of the
peak-to-peak voltage (V in the diagram) is first
halved to obtain the peak voltage, multiplied by
0.707 to obtain the R.M.S. value, squared and
then divided by the resistance of the dummy
load (E2/R).
 The average (or mean) power is less than the
peak power; for a

>>>>>>>>>>>>>>>>>>>> PAGE 27 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

reasonably symmetrical waveform it could be expected to be around 50% of the peak power but
it depends entirely on the waveform of the modulating signal. It is the mean power for which
the transmitter is designed; although the design permits this figure to be exceeded for brief
periods any prolonged output at peak power must cause overheating and destruction of the P.A.

The mean power can be measured by inserting an r. f. ammeter in series with the dummy
load ; the meter must of course be usable at the carrier frequency in use and must give its
indication in terms of R.M.S. values. Such a meter is often incorporated into dummy loads and it
is calibrated in terms of power.

These measurements cannot be undertaken when the modulation is a speech waveform and it
is for this reason that the Licence conditions specify a maximum permitted p. e. p. ; if the limits
of the 2-tone envelope are noted on the oscilloscope display when the output is set to 400 watts
p.e.p. then the modulation controls of the transmitter can be set so that, when speaking into the
microphone, the transmitted peaks do not exceed the permitted maximum level as marked on the
oscilloscope screen.

Lissajous Figures

These provide a simple but useful technique for measuring the parameters of a sinewave.
The figures are formed by switching-off the timebase and feeding the horizontal (X-deflection)
amplifier with a known sinewave derived, for example, from a signal generator ; the Y-amplifier
is fed with the signal under test.

To start, suppose that the X-deflection and the Y-deflection are exactly similar waveforms.
As the waveform voltage shifts from max-negative to max-positive so the spot on the
oscilloscope tube will travel from left to right and also from bottom to top. During the following
half-cycle the spot will retrace its steps exactly. The result is a straight line sloping upwards and
to the right. If the two waveforms are of equal amplitude then this line will lie exactly at 45o.

If the Y-deflection waveform has a frequency twice that of the X-deflection waveform then,
while the spot is moving from left to right, it will move also from bottom to top and then back to
the bottom again. The result is a bent line. During the next half-cycle of the X-waveform the
spot will repeat its up-and-down motion to draw another bent line on top of the first trace. Any
phase-shift present will cause the bent line to open out into a loop and the display appears as two
"sinewaves' forming a ring.

If the Y-deflection waveform has a frequency three times that of the Xdeflection the ring
will show three complete cycles. If the Y-waveform is not an exact multiple of the X-waveform
then the displayed figure will constantly rotate like a circle of children playing ring-of-roses.
 Thus, to measure frequency accurately (or, at least, as accurately as the calibration of your
signal generator), the timebase is replaced by the signal generator and the frequency is altered
until the Lissajous figure becomes stationary. The number of complete cycles displayed on the
figure shows the harmonic relation between the signal under test and the signal-generator
output.

>>>>>>>>>>>>>>>>>>>> PAGE 28 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

Component Testing

An oscilloscope can provide a very useful way to test components without removing them
from circuit. Bear in mind however that all measurements may need to be interpreted arid not
taken at face value. For example, a resistor may give a test result that claims it is a diode. This
may mean that the resistor has become open-circuit and the equipment is measuring a diode via
a shunt path ; OR it may mean that the resistor has a very high value compared with the reverse
resistance of the diode and so the equipment is "seeing" only the diode. The answer lies in the
circuit diagram and it may eventually be necessary to remove the component from the circuit for
testing.

This test circuit is a development from the Lissajous-figure technique described above.
Similar waveforms are fed to the X and Y amplifiers but that to the Y-amp goes via the test-
prods. When a resistor is connected the result is a straight but angled line as described above.
When the test prods are open-circuit the Y-deflection becomes zero and so a high-resistance
appears as a horizontal line. Similarly a very low resistance appears as a vertical line.

As the applied voltage is increased so a diode at first pretends to be an insulator but then
abruptly changes its mind and becomes a low-resistance. Consequently the display switches
abruptly from horizontal to vertical (assuming that the diode is still a diode!).

The exact form of the display depends on the test circuit, on the setting of the gain controls
and on the nature of the item under test.

6.7 THE BRIDGE

The basic Bridge circuit, shown in Fig. 13,

is the Wheatstone Bridge. This consists of four
resistors connected as two series pairs in
parallel across a source. The source can be
either dc or ac but, for accurate measurement,
dc is the more usual because it eliminates the
effects of reactances.

In its simplest form the Bridge consists of

four equal resistors each of which must draw
equal currents ; it follows that the p.d. across
each is half the source voltage. Under these
conditions the potential at point A is exactly the
 same as the potential at point C and so a meter
connected between A and C shows zero
deflection.

The value of an unknown resistor can be determined from this arrangement ; resistor BC is
made variable with a calibrated scale while resistor CD is replaced by a pair of terminals to
which the unknown resistor is connected.

>>>>>>>>>>>>>>>>>>>> PAGE 29 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

Resistor BC is then varied until the meter-deflection is zero when, as described above, the value
of BC (read from its calibrated scale) is the same as that of the unknown.

The range of the Bridge can be increased by using different ratios between resistors BA and
AD ; in general the equality BA/AD = BC/CD will always result in zero deflection of the
meter.

The basic Wheatstone Bridge can be re-arranged in a variety of ways using inductors and/or
capacitors with resistors, together with an ac source, to measure reactances. Sometimes the ratio
arms BA and AD take the form of a tapped secondary winding on a transformer through which
the ac-source voltage is fed. Dependent on the arrangement of the reactances the bridge may be
balanced by either a series combination of R+X or by a parallel arrangement of R & X thus
yielding the phase conditions as either an impedance or an admittance (see Lessons 7, 8 & 9 in
Fundamentals-1).

Knowledge of bridges and bridge measurements is not required for the R.A.E. but you may
be puzzled by references in the literature to Noise Bridges. It is an unfortunate name because it
does not refer to a bridge arrangement intended to measure noise ; it is a bridge arrangement
meant to measure L, C & R but which uses a noise-source as the ac source mentioned above.
Instead of using a meter to indicate the balance point the Operator listens to "white noise" on
headphones and determines balance by a dip in that noise level.

However, there is a point that I would raise here and that is the importance of the quality of
signal generator which is used to drive the bridge. When a Bridge has been balanced, and the
output (either on a meter or on headphones) is at its minimum, the balanced condition applies
only to the fundamental of the Bridge-oscillator; any harmonics present in the waveform will
break-through the Bridge because, at that frequency(ies) the Bridge is NOT balanced. Hence
when starting to set-up a measurement the Bridge-oscillator must be set to its lowest possible
output and the Bridge-detector (which feeds either a meter or headphones) should be set to
maximum gain. As the balanced condition is approached so the detector gain is turned down
and the oscillator output increased. A low-quality signal source with a rich complement of
harmonics is useless for Bridge work.

END OF LESSON 4
 ***********

QUESTIONS

1. A sinewave is displayed on an oscilloscope screen with a peak-to-peak waveform that

occupies the space between four 1-cm graticule squares. Given that the Y-gain controls are set
to "5OO mV per cm" what is the R.M.S. value of the waveform ?

2. The above waveform is being developed across a 50-ohm dummy load. What power is being
delivered by the source ?

3. Why is power measurement difficult in s.s.b. equipment ?

How would you use an oscilloscope display to set the limiting circuits(probably a so-called
speech processor) so that you did not exceed the Licensed limit of 400 watts p.e.p. ?

4. A transmitter output is fed to a dummy load across which an oscilloscope is connected. The
transmitter is operating in the a.m. mode with a modulation waveform which is fed also to the
X-deflection of the oscilloscope.

In the trapezium-shaped display the maximum height is 2½ cm and the minimum height is ½-
cm. What is the depth of modulation ?

>>>>>>>>>>>>>>>>>>>> PAGE 30 <<<<<<<<<<<<<<<<<<<

Back to Top of Page

LESSON 5
6.8 ELECTROMAGNETIC COMPATIBILITY

This rather curious title is usually shortened to the letters E.M.C. and it refers to the twin
matters of causing interference to other users of the radio spectrum and to receiving interference
from other users.

The basic idea seems to be that all electrical/electronic equipment should be compatible with
the electrical environment ; I cannot say that it enlightens me however.

6.8.1 Causes of Interference

Many of the most common and obvious causes of being a nuisance to other users have been
mentioned already in previous Sections of this Course:

(i) Harmonic generation caused by overdriving, wrong bias or other faults in transmitter
adjustment

(ii) Harmonic generation caused by bad keying technique - see key-click filter.
(iii) Excessive side-frequency generation through over-modulation or because the microphone
audio circuits are generating harmonics of the modulating waveform ; see speech processor.

(iv) Harmonic radiation from the Drive

(v) Spurious oscillations caused by wrong adjustment of the transmitter or because the apparatus
lacks sufficient margin of stability and it bursts into oscillation on modulation peaks

(vi) Intermodulation products which are the result of beats between two signals which may be
either legitimate signals or unintentional products as listed above.

(vii) RF radiation from computer-type equipment

(viii) Waveform distortion imposed on the mains from the modern habit of generating dc
supplies without the use of a smoothing choke. The very-large values of capacitance involved
result in very-large but short-duration spikes of current being drawn from the mains at the peak
of each cycle. The high-order harmonics generated are radiated from the mains wiring . It is now
illegal to design or to supply such power units.

In general the "cure" for all the transmitter problems is a suitable low-pass filter but it must
be emphasised that such filters are placed at the output of transmitters as (a) a precaution (b) as a
final clean-up of an otherwise good signal. It is very bad practice to be sloppy about adjusting a
transmitter and then (hopefully) to rely on a l.p.f. to reduce the resulting mess.

>>>>>>>>>>>>>>>>>>>> PAGE 31 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

** As a cure for problems of the type in (v) above you may often see a small-value resistor
(around 100 to 200 ohms) connected in series with the collector of a transistor (or the anode of a
valve) P.A. stage. The resistor or perhaps a small choke is mounted as close to the active device
as possible. Such a resistor is referred to as a stopper (anode-stopper, grid-stopper, collector-
stopper) and is a likely subject for a question in the R.A.E. Each electrode of a valve or
transistor has a (stray) capacitance to the others and the stopper resistor is in series with that
capacitance and so forms a low-pass filter.

** The stopper must have an impedance

which is low in comparison with the load at the
operating frequency but which is greater than
the inter-electrode reactance at the parasitic
frequency. It is not uncommon to see a stopper
in the form of a small choke wound over a
resistor thus forming a lossy (or low-Q) choke
whose function is to both stop and dissipate
spurious signals - compare this with the use of a
ferrite bead as mentioned in 6.6.2 below. [With
reference to the Lesson on oscillators it should
be noted that the addition of resistance (loss) to
 a resonant circuit can ensure that the Q is too
low for oscillation to take place.]

As mentioned in Part-2 a low-pass filter

cannot cure all ailments. Distortion in the
modulating-signal chain will produce unwanted
side-frequencies both above and below the
carrier frequency. The final l.p.f. will remove
the out-of-band upper side-frequencies but it
will ignore the lower side-frequencies.

A bandpass filter will, of course, take care of this situation but, apart from being an expensive
and unnecessary cure, it represents a sloppy approach that is likely to land an Operator in more
trouble than he reckoned

Unfortunately it is possible to cause severe interference even when the transmitted signal is
immaculate. Where this is the fault of the Amateur Operator he must, of course, take steps to
eliminate the nuisance ; where it is not his fault he is not liable but, if you wish to live at peace
with your neighbours, a courteous and helpful attitude is to be recommended.

Many receivers use an intermediate-frequency around 10.7 MHz which is the third-harmonic
of carriers in use on the amateur 3.5 — 3.8 MHz band. When close to a transmitting aerial the
strong inductive-field is very likely to find its way into the if-amplifier and so give rise to a
complaint that " ... you are coming in all round the dial ... " ; in fact, of course, the interference
is not coming through the tunable front-end at all. The carrier may be immaculate, the keying
waveform perfect and the transmission is not in any way spreading itself across the spectrum.
The fault lies in using an unscreened receiver too close to a transmitting aerial. That one is very
difficult indeed to explain to a non-technical and irate neighbour.

Harmonics can be generated in a parasitic circuit which contains a nonlinearity — i.e. a

partial or low-grade rectifier. This could take the form of a rusty joint or bolt in local metalwork,
a corroded joint in plumbing, a badly made or corroded connection in household wiring. The
offending "circuit" is excited by radiation from the transmitter aerial, generates harmonics
through its non-linear element and then re-radiates the degraded signal. Such a parasitic circuit
may not be at all easy to locate.

>>>>>>>>>>>>>>>>>>>> PAGE 32 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

Sometimes household wiring or plumbing can form a closed loop which just happens to
resonate within one of the amateur bands ; the result can be a large circulating current with any
sort of interesting effect. The problem here is to first locate that loop and then to add a reactance
to detune it.

Non-linear elements are to be found in all household electronic equipments in the form of
transistors or similar semiconductor devices. If the signal radiated from a transmitter aerial gets
into a television receiver, a video-recorder or any form of audio equipment then it is likely to be
demodulated and the result becomes a permanent part of the signal which the equipment is
supposed to be processing. It can even appear in the telephone
 The only practical cure for this type of problem, apart from reducing power, is to re-site the
aerial and/or investigate its design to ensure that the signal is radiated into space and is not
irradiating the neighbour's (or your own) household.

The field-strength (signal strength) laid down by an aerial is proportional to the square-root
of the radiated power and so it might appear that power levels do not have much effect on
interference. In practice however it is found that power reductions can have a marked effect and
it is good practice never to use more power than is required for the Contact of the moment.

** Marks can be picked up in the R.A.E. by calculating field-strength from a simplified

formula:

e = 7√erp/d

where e is the signal strength in Volts-per-metre (V/m), erp is in watts (estimated radiated
power) and d (in metres) is the distance from the aerial to the point in question. However it is
difficult to see much practical use for the calculation.

It is said that the formula can be used to give an idea of the probable field strength at the
point where interference is being caused by an amateur transmission. At the close range at
which such incidents take place it is likely that the close-in inductive field has more to do with
the problem than the long-distance radiation field with which the formula deals. In any event it
is difficult to see the benefit of the calculation when trying to suppress such interference.

6.8.2 Eliminating Interference

There is not a real set of rules for dealing with incoming interference. The important thing is
to know the tools that are available and the techniques for which they have been formulated. In
general the idea is to divert unwanted interference-currents away from the signal path and to
divert unwanted voltages to ground. Mostly the techniques used to prevent unwanted radiations
can be adapted to deal with unwanted incoming signals (irradiations).

>>>>>>>>>>>>>>>>>>>> PAGE 33 <<<<<<<<<<<<<<<<<<<<

By way of illustration Fig. 16 shows a small brush motor such as may be found driving an
electric typewriter or a sewing machine. The circle represents the rotating armature and the two
sloping lines represent the "brushes" which carry power to that armature. Sparks which occur at
the brushes may cause radiation directly but they also cause noise signals to flow into the supply
cable which then acts both as a transmission line and an aerial.
 As shown these are of two types:

(a) push-pull currents which flow in opposite

directions on the supply lines
(b) push-push currents which flow along both
legs of the supply cable in the same direction.

The first of these can be confined to the motor by connecting a suitable small capacitor
across the brushes (C 1 in Fig. 16). The second type are diverted to ground or to the metal casing
of the motor (C2 & C 3 ) from which they complete their circuit by capacitive coupling to the
armature. The three capacitors can be obtained built into a single unit with three connections
which is known as a delta capacitor because of its resemblance to the Greek (triangular) symbol
delta.

That a low-impedance capacitive path has been provided does not necessarily mean that the
intended currents will take that path. Current will divide between the cable and capacitive paths
according to their relative impedances thus, to encourage most of the current to take the
capacitive paths, it is necessary to increase the impedance of the cable at the frequencies of the
interference signals.

As shown in Fig. 16 small chokes are added in series with each leg of the cable. These may
be bought off-the-shelf or hand-wound on small ferrite cores ; their small inductance presents
little opposition to the flow of 50-Hz mains current but virtually open-circuits the cable to
radiating frequencies.

As an alternative small ferrite beads are available which can be slipped over the wire. They
come in different grades but, basically, these increase the line inductance ; they also add some
loss to the circuit and so dissipate the unwanted energy.

As shown the choke/capacitor combinations keep unwanted interference currents within the
device which is generating them. However, to protect that device against intruding currents
driven into the mains by some other piece of badly-protected equipment, the chokes must be
placed on the other side of the capacitors; i.e. between the capacitors and the device. A general-
purpose filter unit can be made, which will keep rf currents both within and without the device,
by placing a capacitor on each side of the chokes. Such a filter may be referred to as a pi-filter
because of its resemblance to the Greek letter π.

>>>>>>>>>>>>>>>>>>>> PAGE 34 <<<<<<<<<<<<<<<<<<<<

6.8.3 Filters

With reference to Fundaznentals-2, Lesson-5 the above pi-filter is a low-pass filter . At high
frequencies the choke open-circuits the signal path while the capacitors short-circuit that path to
ground. At low frequencies neither the choke nor the capacitors are effective.
 This filter is converted to a high-pass configuration by interchanging the positions of its
components ; i.e. the series path is interrupted by a capacitor and the choke(s) is put in shunt to
ground.

Both configurations can be developed by

resonating the elements as shown in Fig. 17to
form either bandpass or bandstop (trap) filters.
The bandpass circuit can be useful in
preventing the escape of unwanted signals
while the bandstop can be useful in permitting
the passage of wanted signals (such as your
neighbour' s television service) while stopping
the intrusion of your amateur transmissions.

For the R.A.E. you are not required to design such filters but it is perhaps worth
understanding their limitations. In the lpf, for example, the series choke is shunted by stray
capacitances which set an upper limit to its effectiveness as an inductance. In a similar manner
the shunt capacitors exhibit a small amount of inductance which sets an upper limit to their
effectiveness as capacitors.

Indeed this is the situation illustrated in Fig. 17 where the reactors have been deliberately
resonated. It follows that all chokes and all capacitors have limited capability set by their own
natural resonance frequencies. Above their individual resonance-frequencies an inductor
becomes a capacitor and a capacitor becomes an inductor. As frequency is increased the only
cure for this is to reduce the value of the component but, again, there is a physical limit.
Fortunately the reduced value of the component is compensated, in terms of reactance, by the
increased frequency.

6.8.4 The Braid-breaker

The coaxial feeder which connects a television receiver to its chimney-top aerial is often
some 30 to 50 feet in length (say 10 to 15 metres) which means that it can act as a reasonably
efficient receiving aerial which is λ/2 in length for the amateur 20 and 30 metre bands or λ/4 in
length for the lower-frequency amateur bands. Earthing the braid of the coax at the television
receiver merely turns this into a more efficient hf aerial .

>>>>>>>>>>>>>>>>>>>> PAGE 35 <<<<<<<<<<<<<<<<<<<<

For a nearby transmitter putting out some 100 watts or more there is likely to be a considerable
current flowing on the tv feeder and it is not surprising if this appears in the internal circuitry
and finally on the picture and/or the sound channels.

[As a matter of interest this was a serious problem when the first television transmissions
took place from Alexandra Palace just before WW II. From ground to the top of the television-
transmitter mast was almost exactly a quarter-wave at the frequencies of the medium-wave
broadcast service from nearby Brookmans Park. As a result the broadcast services insisted on
their rightful place in every camera and every piece of equipment used in the studio and their
exclusion was a major headache. ]

An obvious cure is to disconnect the braid from the receiver but this has two unfortunate
effects

(i) the television signal is attenuated as well as the hf interference

(ii) the inner, no longer screened, takes over as the vertical receiving aerial with possibly worse
results.

One trick might be to break the braid and then bridge the gap with a small capacitor which will
act as a somewhat crude high-pass filter. There is a possible danger however in that the amateur
transmission could induce high voltages on the now unearthed coax feeder ; apart from
tickling-up the family this voltage would then be capacitively-coupled onto the inner conductor
and so re-appear at the receiver input.

The so-called braid-breaker arrangement threads the coax television feeder through a ferrite
ring ; the turns should be well spaced and the first and last turns not permitted to come too close
(to avoid the bypassing effect of stray capacitance). In terms of the outer braid this generates a
small inductance which effectively open-circuits the braid and so interrupts the flow of the
interfering hf currents.

Why does the inductance not stop the uhf television carriers (around 500-MHz) also ? In
Lesson-8 of Fundarnentals-2 it was shown that the energy conveyed by a coaxial transmission
line was carried by the electric and magnetic fields which are set up between the inner and outer
conductors. The outer conductor screens the inner and so, when the feeder is coiled, the uhf
magnetic fields of the individual coils do not add together. The transmission path may become a
bit giddy but an inductive effect is not generated. You may prefer to think of it as cancellation
between the current which flows on the inner conductor and current in the reverse direction
which flows on the inside surface of the outer ; the interfering hf current flows only on the outer
surface of the outer conductor.

Although it is affectionately called a braid-breaker the technique is not limited to coaxial

cables. It can be very effective in reducing the flow of unwanted currents on mains cables by
winding the cable onto a ferrite ring in exactly the same manner. It is perhaps more effective to
use a properly-wound balanced choke in which the two legs of the circuit are married and
wound together on a core ; the magnetic fields of wanted push-pull currents cancel and so do not
generate an inductive effect while unwanted push-push currents aid in setting up a magnetic
field.

>>>>>>>>>>>>>>>>>>>> PAGE 36 <<<<<<<<<<<<<<<<<<<<

Back to Top of Page

6.8.5 The Dangers of P.M.E.

To get an hf aerial to radiate efficiently it is essential to provide a good low-resistance earth

connection. A good earth connection is essential also to the techniques described above of
screening and filtering to avoid problems with e.m.c. It is unfortunately true that a good low-
resistance earth, if improperly used, can result in a fire and also in death by electrocution.

Power is generated and distributed as a three-phase supply. Essentially this consists of three
wires each of which carries a 50-Hz waveform which is 1200 shifted in phase with reference to
the other two. If power is drawn equally from all three wires the system is self- sufficient but, if
the load is unbalanced, then a fourth wire is necessary to carry an out-of-balance current ; this
is referred to as the neutral wire. The normal domestic single-phase feed consists of one of the
phase-wires and the neutral which, nominally at least, is at earth potential.

It is unlikely that the neutral will be exactly at zero potential because of the volt-drop along
its length caused by the current which flows on the wire ; this could be several hundred amps
when a single feed serves several establishments.

Before the 1960's the house would have been supplied also with an Earth connection which
usually consisted of the metallic sheathing of the supply cable. The actual connection to earth
was made at the sub-station and it may not have provided a very good earth.

The danger was in the possibility of a break in the neutral line with the live 240-volt line still
operative. This would leave all the equipment beyond the break at 240-volts even though it was
no longer working. For this reason all electrical equipment was housed in metal casings which
were bonded to the earth connection. A fault would not become a hazard as long as the metal
casing remained efficiently earthed and a would-be repairer did not attempt to disconnect that
earth. Hopefully any fault current flowing to earth via the casing would rupture a fuse and so
disconnect the faulty equipment but that presupposed that a handyman had not rigged the fuse.

In a p.m.e. (protective multiple earth) distribution a separate earth connection is not supplied
and the neutral conductor is bonded to earth wherever it may be convenient . This multiple
earthing produces a lower-resistance earth and so greater safety. Such earthing takes place at
each distribution point by bonding the neutral to gas and water supply pipes. A problem today is
that much plumbing is carried out using plastic pipes and, even on older installations, plastic
pipes may be introduced. This does not help with earthing problems particularly when a plumber
cuts-out or uncouples a piece of steel or copper pipe and finds himself providing the earth
continuity for a piece of faulty equipment.

All bondings to such plumbing must be made at the point where the pipe(s) enters the house so
that it is not possible to interrupt the earth connection.

Hazards such as these are taken care of by a residual-current-device which essentially

compares the current flowing in the live lead with that returning through the neutral lead ; the
two currents should be identical.

>>>>>>>>>>>>>>>>>>>> PAGE 37 <<<<<<<<<<<<<<<<<<<<

An insulation fault will allow current to escape from the circuit and flow to earth (via the
plumber ?). The r.c.d. uses a balanced transformer to detect any difference in the two currents
and, for household use, it trips a circuit-breaker when this residual-current reaches 30 mA. It
should perform this feat within a matter of milliseconds and so prevent any frying incidents.
 The system works very well unless a keen amateur-radio enthusiast wants to add a separate
earth connection to improve the radio-frequency performance of his system. It doesn't matter at
all as long as

(a) faulty equipment is not present and

(b) the neutral connection to the house does not become disconnected.

This may not be a serious hazard for a town dweller until the Council decide to dig-up the road
but, in country areas where overhead distribution cables are commonly used, it is not at all
impossible.

With the neutral interrupted the amateur's rf-earth takes its place and the mains lead to the
amateur equipment takes over as the earthed return for every establishment connected to that
section of line beyond the break. A normal 5-amp mains lead is not intended to carry the
several hundred amps of mains current that could be flowing along the neutral and so it will melt
and start a fire. Should such an accident occur while the rig is being operated the Operator then
takes over as the earthed return with dire consequences.

The best solution is to disconnect the mains-earth from the amateur equipment and also from
every piece of equipment within 6-ft of the operating position (i.e. within reaching distance).
All metal parts within that area must be bonded together and taken to the rf earth. Portable
equipment, such as kettles, coffee pots, heaters, soldering irons must not be brought into the
protected area unless plugged into a socket whose earth pin also goes only to the rf earth. [Note
that all other fixed metal objects in the house should also be bonded together and taken to the
incoming neutral/earth thus ensuring that, in the event of losing the neutral, everything floats at
the same (240-volt) potential.]

It is essential that the power fed to this protected radio area be taken through a separate r.c.d. ; it
is possible to obtain a wall socket with built-in r.c.d. or a plug-in type but take note of the
current-limitations of the device you use. Do NOT disconnect your mains earth until an r.c.d. is
in place.

There is a good article on this subject published in the R.S.G.B.'s RadCom in June 1987 under
the title The Killing Ground.

>>>>>>>>>>>>>>>>>>>> PAGE 38 <<<<<<<<<<<<<<<<<<<<

END OF LESSON 5

* * * * * * * * * * * *

QUESTIONS

1. Why should you not rely on a low-pass filter at the output of your transmitter ?

2. What is a "collector-stopper" ?
3. Your neighbour complains that, when you were calling "CQ" a few minutes ago, you came
out of his f.m. radio all over the band. What is the probable cause and what would you do about
it ?

4. Some time ago our receiver, connected to the long-wire aerial, was producing a series of
clicks or "crashes". These disappeared when the aerial was disconnected but were noticeably
more frequent during gusts of wind.

An inspection of the aerial revealed that the end of a short length of chain, which was part of an
upper stay for the mast, was swinging freely in the wind and constantly touching a lower stay-
wire. The chain was tied-off with a piece of string and this completely cured the fault. Can you
explain this ?

5. What are the basic techniques used when trying to eliminate interfering signals ?

6. What is a "band-stop" filter ? How might you use it ?

7. A good transmitter installation requires a good low-resistance earth. What is the source of the
hazard such an earth can cause if you are fed mains power from a P.M.E. distribution system ?

END OF PART 6

Back to Top of Page

 PART 7: MATHEMATICS FOR AMATEUR
RADIO
LESSON 1
7.1 TRIGONOMETRY

7.1.1 Circular Measure

Trigonometry is the branch of mathematics which studies the properties of triangles. On a higher
level Geometry is the study of space (i.e. of shapes) and this deals in:

(a) one dimensional lines which are expressed in Linear Measure or units of length (L)
(b) two-dimensional plane figures which are expressed in Square Measure or units of area (L2)
(c) three-dimensional solid figures which are expressed in Cubic Measure or units of
volume (L3)

The triangle is the simplest of the Plane Figures. Given three points they can describe only a
triangular plane and that is why a three-legged stool, table or any form of tripod will always
stand firmly (it will always find a plane surface) whatever the terrain even if the legs are not all
of equal length. Given four points the resulting figure could be a plane (a rectangle) but, most
probably the result would be the solid form - a triangular pyramid.

A triangle is described also by three intersecting lines, known as the sides of the triangle, and the
three points of intersection are the three points mentioned above. These three points are known
as the apeces of the triangle (the plural of apex.)

In turn these three sides define three angles which are the shapes enclosed where two lines (or
two planes) intersect. To get to grips with angles it is necessary to consider a unique plane figure
the circle which is a special form of the more general ellipse.

On first acquaintance the circle is an uninteresting beast with but three parameters namely:

(i) a point which defines the centre of the circle and which usually is labeled "O".

(ii) a circumference which is a line that defines the circle and which is at all points a constant
distance from the centre.

(iii) a radius which is the distance from the centre to the circumference and which usually is
labeled "r". A line drawn right across a circle and which passes trough the centre contains two
such radii and is known as the diameter "d".

(The Geometry theorems of Euclid show that the circle possesses hidden depths).

Stand at the centre of a circle and look due North and it is possible to rotate to the left (without
moving from the centre point) and to continue the rotation until you once more look due North.
At the end of this manoeuvre one whole circle of rotation has been completed; in mathematics
it is accepted practice to make all such measurements in an anticlockwise direction which is why
rotation to the left was specified. (Measurements made in a clockwise direction are given
negative values).

Should this rotation be abandoned before the full circle has been completed some method of
describing the stopping point is required. For this purpose the full circle is divided into
360 degrees (written 360°); the name is derived from a Latin word which means "step". An
angle describes a change in direction, or the difference in directions of the two lines which
define that angle; if the point of intersection is taken as the centre of a circle then the angle can
be measured in degrees. Such measurements are referred to as angular measure or circular
measure.

For a circle with a small diameter a degree represents a short distance around its circumference;
for larger circles the degree may be sub-divided into 60 minutes (written as say 40') each of
which is sub-divided into 60 seconds (written say as 30") and these are referred to as minutes of
arc and as seconds of arc.

There is a second method of measuring angles and, in electrical engineering, this method has
advantages if only because it simplifies mathematical expressions and operations. If the
diameter of a circle is marked off along the curve of its circumference it is found that it fails to
divide that circumference into a whole number of parts. Whatever the size (the diameter) of the
circle it is found that the result of dividing its circumference by a diameter always yields the
same number — and that number is a very awkward one indeed.

The exact value of this ratio is not known even though it has been calculated to many decimal
places on computers; because it is not an exact number it is referred to by the Greek letter π
(pronounced pie and written "pi"). It has the approximate value 22/7 or, more accurately
3.14159.

Fig.1 shows a circle with its centre at a point O. Two different radii of length r are shown one
drawn from O to a point A and the other drawn from O to a point B. The distance from A to B
measured along the circumference is also of length r. The angle defined by these two radii -
the angle which is subtended by the arc AB (where AB = r) - is defined as one Radian.
Because the circumference of any circle can be divided by the diameter π times it follows that
the circumference can be divided by its radius (equal to half a diameter) 2π times. Hence
rotation through a full circle (360°) represents a rotation through 2π radians; a half-circle (180°)
is equivalent to π radians; a quarter-circle (90°) is equivalent to π/2 radians (read as pi-by-two).

The quarter-cycle of 90° or π/2 radians is referred to as a right-angle because, if one side of the
angle is regarded as a base, then the other side stands "right" (or upright). Trigonometry is
based on the remarkable properties of triangles in which one angle is a right-angle.

7.1.2 The Right-angled Triangle

Fig. 2 shows a right-angle triangle ABC with the right-angle at B. In diagram (a) only
the internal angles are shown. In diagram (b) the so-cal1ed external angles are shown; these
still lie between the sides but they are external to the triangle. Although the triangle shown is a
right-angle triangle the following arguments can be applied in general to any triangle.

Imagine that ABC is a large triangle and that you are able to walk along its sides. Start at A,
look toward C and walk from A to C. At C rotate positively (anticlockwise) until you are able
to walk backward toward B. At B again rotate anticlockwise until you are able to walk forward
toward A. Now rotate anticlockwise for the third time until once more aligned with AC. Note
that you now face toward A although the exercise began facing toward C; i.e. in negotiating the
three internal angles of a triangle a rotation of only 180° (or π radians) has been necessary. This
is one way of deducing that the three internal angles of any triangle sum to 180°.

To repeat the exercise on the three external angles start again at A but walk to B where an
anticlockwise rotation leaves you able to walk (forward) to C. At C another anticlockwise
rotation makes it possible to walk forward to A where the third anticlockwise rotation leaves
you again facing toward B. This time a full circle of rotation has been necessary and so the
three external angles of any triangle sum to 360° (or 2π radians).

If the three internal angles sum to 180° and the angle ABC is a right-angle (90°) then the two
internal angles BAC and BCA must always sum to 90°; the two angles are said to be
complementary (NOT complimentary). Thus, once a value has been assigned to any
angle BAC the value of the third angle BCA is fixed.

It is found also that, in a right-angle triangle, once a value is given to one of the non-right-
angles. the ratios of any two sides is fixed irrespective of the size of that triangle. With three
sides involved there are three possible pairings and, for trigonometrical purposes, the actual
ratios for each of those pairings have been calculated and tabulated against all possible values of
angle. Given minimum information about the values of either any two sides or one side plus one
associated angle it is possible to calculate a complete solution for any right-angle triangle.

The nomenclature used in Trigonometry is designed to give positive identification of the sides
and angles of any right-angle triangle as shown in Fig.2 (c);

(i) the side opposite the right-angle is called the hypotenuse (pronounced high-pot-en-youse)
and this is side AC

(ii) the angle under consideration is included between the hypotenuse and its adjacent side; for
angle BAC the adjacent side is BA while for angle BCA the adjacent side is BC

(iii) the remaining side, the side which is opposite the angle under consideration, is called
the opposite side

Note that the adjacent side for one of the complementary angles is also the opposite side for its
complement (the other angle that brings their sum up to 90°).

There are three main ratios which are tabulated and they are defined as:

sine the ratio opposite/hypotenuse written as sin

cosine the ratio adjacent/hypotenuse written as cos

tangent the ratio opposite/adjacent written as tan

Those familiar with algebraic division will see that dividing sin by cos yields tan

Although a book of Tables contains other trigonometrical ratios the above set are the basis of
them all. Three other ratios are merely the inverse of the basic ratios:

secant 1/cosine hypotenuse/adjacent written as sec

cosecant 1/sine hypotenuse/opposite written as cosec

cotangent 1/tangent adjacent/hypotenuse written as cot

The purpose of those last three tabulations is simply to save time and effort when a
mathematical expression calls for an inverse ratio.

Another set of tabulations are labeled Logarithmic Sines (written as logsin), Logarithmic
Cosines (written as logcos) and Logarithmic Tangents (written as logtan). Today when
calculators are available these Tables are seldom used but the purpose of these last three was
once again to save time and effort in evaluating expressions such as [Link]θ (where a is a number
and θ is an angle); multiplication of the two would involve first looking-up the sine of angle θ
and then finding its logarithm from the Log Tables. A Logsine Table enables a direct step
between the va1ue of the angle and the log of its sine.
To try to learn all these ratios and derivatives usually leads to confusion; it is best to remember
say only the sine (opp/hyp) and that the cosine is “the other one”. Tangent does not involve the
hypotenuse and there is also the useful relation that sin/cos = tan.

The use of this Table is very straightforward. Suppose that side AB in Fig. 2 is given as 10cm
in length and that angle BAC is 30°. To solve the triangle the first step is to deduce that the third
angle BCA has the value

90 - 30 = 60°.

Side AC is calculated either from the cosine expression or from the sine expression as

AC = adjacent/cos30° or as AC = opposite/sin60°

= 11.547 cm

Either of these equations could have been simplified by using the inverse ratios such as

AC = adjacent x sec30°

Because of the complementary nature of the two angles it is possible to combine sin and cos in a
single Table and also cosec and sec in a single Table. To read sines enter with the angle in the
left-hand column and move into the columns on the right for partial degrees as indicated along
the top line. To read cosines enter with the angle in the right hand column and move into the
columns to the left for partial degrees as indicated along the bottom line. The last columns on
the far right are given as an aid when interpolating between columns; they represent the
differences from one column to the next.

In the example given above the third side BC can be calculated in the same way. However
because this is a right-angle triangle, there is an alternative method which is sometimes useful in
the form of Pythagoras' Theorem which states that (with reference to Fig.2)

AC2 = AB2 + BC2

The presence of the squared terms suggests that this formula is concerned with square-measure
or areas and indeed this is the basis of the proof first deduced by Pythagoras; he showed that the
area of the square constructed on the hypotenuse is equal to the sum of the areas of squares
constructed on the other two sides.

Thus the required third side BC is given by √(AC2 - AB2). It is sometimes easier to use this
expression when the sides of the triangle are whole numbers. Some values are worth
remembering; for example side-lengths of 3, 4 & 5 satisfy Pythagoras' Theorem - can you find
any other sets of whole numbers.

Notes

The sample Tables attached to this lesson are known as Four-figure Tables and they provide
accuracy to three decimal places. For greater accuracy Seven-figure Tables are used - often
found in sets of Navigation Tables. However it is easy to achieve accuracies up to 12 decimal
places with computers and some calculators.
The term "circular measure" is generally used in Tables to indicate the measurement of angles in
radians.

Conversion between degrees and radians is achieved by starting with a full circle of 360° and 2π
radians.

When converting from radians to degrees the value is larger in degrees than it was in radians
hence multiply by 360 and divide by 2π (x 180/π).

When converting from degrees to radians the value is smaller in radians than it is in degrees;
hence multiply by 2π and divide by 360 (x π/180).

7.1.3 The Non-right-angle Triangle

The paragraphs above show how a right-angle triangle can be solved using trigonometrical
tables provided that either two sides are known or one side and an angle. The definitions of sin,
cos and tan relate specifically to right-angle triangles but the technique can be used to solve
triangles which do not include a right angle.

Fig. 3 shows such a triangle in which a perpendicular (a line at 90°} has been dropped from one
apex of the triangle to the opposite side; the name indicates that the construction line has been
drawn so that it divides the parent triangle into two right-angled triangles.

It is now possible to solve both of these component triangles using sin, cos or tan ratios, so
evaluating the sides and angles of the parent triangle provided of course that the requisite
minimum information is available. The significant fact is that the constructed perpendicular is
common to both triangles; if either of them can be solved then the common side enables the
calculation to be transferred to the other triangle in which is now known a side and one of the
angles.

By using purely algebraic techniques on this problem it is possible to produce formulae which
relate the sides and angles of the parent triangle; the usual nomenclature, a kind of shorthand, is
shown in Fig.3(b) wherein the side opposite the angle at A is named "a" while the angle itself is
named "A". The solution of such triangles is not required for the R.A.E. and the matter is not
pursued here but the appropriate formulae can be found in any mathematical or electrical
reference book.
Relaxation of the requirement for one angle to be 90° gives rise to the possibility of two
different types of triangle:

(a) a triangle in which all the angles are 1ess than 90° as shown in Fig. 3

(b) a triangle in which one angle is greater than 90° and which is known as an oblique-angled
triangle as shown in Fig.4 .

To solve an oblique-angle triangle a similar construction

is used namely a perpendicular is dropped from the apex
A on to the base CB extended. The term hypotenuse
can still be used for the side opposite the right-angle but
the remaining two names "adjacent" and "opposite"
become confusing. By definition the side AB is the side
opposite angle ACB but the sin and tan ratios for that
angle involve the constructed perpendicular AP; for this
reason the word "perpendicular" is often used in place
of "opposite".

Similarly the adjacent side is really CB and not CP but this is not likely to cause confusion; thus
for angle ACB the three ratios become:

sin ACB = perp/hyp = AP/AC

cosACB = adj/hyp = CP/AC
tanACB = perp/adj = AP/CP

WARNING To find the corresponding ratios for angle BAC it is necessary to make a new
construction by dropping a second perpendicular from C to AB extended.

For a non-right-angle triangle the terms "opposite side" and "adjacent side" do not refer to the
sides of the triangle and angles such as PAC are not an internal angle of the triangle.

The same construction is used to find the sin/cos/tan ratios of a triangle's external angles.
In Fig.4 for example the sine of external angle ABP is AP/AB and this is also the ratio for the
internal angle ABC. If this treatment of angles greater than 90° seems a bit obscure or casual
the following derivation may perhaps clarify the matter.

In Fig.5(a) angle A 2 OA 1 is created by rotating a line OA from the starting position OA 1 to the
position OA 2 ; let the angle traced out be called θ (the Greek letter theta). To find the
sin/cos/tan ratios of angle θ it is necessary to drop a perpendicular from A 2 to the base-
line OA which it intersects at point P.

When θ is zero the points A 2 and A 1 are coincident, point P lies on top of them, the
perpendicular has zero length and the adjacent side is equal in length to OA. As θ increases
from zero toward 90° so AP begins to increase and OP decreases until, when θ = 90°, the
perpendicular is equal to 0Aand the adjacent side has zero length.
Further rotation of 0A beyond 90° creates the oblique-ang1e triangle A 4 OA 1 in which the angle
of interest - the angle being created by the rotation of OA - is the oblique angle A 4 OA 1 . The
perpendicular AP is carried beyond the base line A 1 P and, as θ increases from 90° to 180°,
so AP decreases toward zero and 0P increases toward the length 0A in the position 0A 5 . This is
shown in diagram (b).

It must be noted however that 0A was measured to the right starting from the origin at 0 while
0A 5 . is measured to the left; if movement of P from A 1 to O is a steady decrease in length then
its further travel from 0 to A 5 must be considered a further decrease in length and so the
length 0A 5 must be a negative quantity.

Still further rotation of 0A beyond 180° is shown in diagram (c) and it leads to the creation of
the angle A 6 OA 1 . Once again the perpendicular AP is increasing toward the length OA while
point P returns toward 0 where, at θ = 270°, the length of 0P is once again zero.

Finally, as θ increases from 270° to the full rotation of 360°, so OP returns to the
value OA and AP returns again to zero; this is shown in diagram (d).

When these four diagrams are compared it is found that always the triangles AOP are mirror
images. Thus the ratios sin/cos/tan repeat in value as the angle successively passes multiples of
90° but it must be remembered that, although the numerical values repeat, there are changes of
sign to be considered and these are summarised in the following Section.

7.1.4 The Quadrants of a Circle

A full circle is divided into 360° and so, divided by two mutually-perpendicular lines, it contains
four right-angles each of 90°; see Fig.6. Each of these quarter-circles are referred to as
a quadrant of the circle. In Geometry, Trigonometry and general mathematics it is accepted
practice to measure positive distances to the right - starting from a given reference point - and
to measure angles in an anticlockwise direction. As shown in Fig. 6 this convention leads to
the upper right-hand quadrant being known as the 1st Quadrant, the upper left-hand one as the
2nd Quadrant, the lower left-hand as the 3rd quadrant and the last one as the 4th Quadrant.
For an angle starting at 0° along OA this horizontal axis always represents the adjacent side and
the vertical axis always represents the opposite side (or the perpendicular). It follows that for
angles in the first quadrant both adjacent and opposite sides are always positive and so the three
ratios also are always positive. For angles in the second quadrant the adjacent side is negative
but the opposite side (the perpendicular) remains positive; thus the sine remains positive but the
cosine and tangent both become negative in value.

In the third quadrant both the adjacent and the opposite sides are negative and so the sine and
cosine are negative but the tangent, the result of dividing one negative quantity by another
negative quantity, becomes positive. Finally, in the fourth quadrant, the adjacent side is positive
but the opposite side remains negative and so it is the cosine which becomes positive while the
sine end tangent stay negative.

Note too that (for example) sinθ = sin(180° - θ )

while cosθ = -cos(180°- θ)

In Fig.7 the upper curve (a) relates to the sine function and begins at θ = 0° when sinθ = 0°;
this results because the opposite side has zero length. At 90°, when the hypotenuse and the
perpendicular are coincident, the sine ratio reaches a maximum value of 1 and thereafter
decreases again toward zero.
The second curve (b) relates to the cosine function and begins at θ = 0 when cosθ = 1; this is
because the adjacent side and the hypotenuse are coincident. At 90°, when the adjacent side has
zero length, the cosine ratio has fallen to a value of zero and thereafter continues to decrease
toward a minimum value of -1.

It has been shown already that the two internal non-right-angles of a triangle are complementary
(they sum to 90°) and that the adjacent side for one angle is the opposite side for the other. It
should not be surprising therefore that the plot of a sine function yields the same result as the
plot for a cosine function but the two curves are displaced by 90°. We say that they differ
in phase by 90°.

The third curve (c) is a similar plot for the tangent function and it begins at θ = 0° when tanθ =
0; again this result is caused by the perpendicular having a length of zero. As θ is increased
toward 90° so the opposite side increases toward the length of the hypotenuse but,
simultaneously, the adjacent side decreases toward zero length. Thus the value of the tangent
increases toward a value opposite/zero which has the mathematical value of infinity.

Strictly speaking the value of anything/zero is meaningless but the matter is resolved by the
technique of approaching the condition. We say that, as the bottom line approaches zero, so the
result approaches an infinite value and it can be written:

as the adjacent side » 0 so tanθ » ∞

(read as as adjacent tends to zero so tan θ tends to infinity)

As the angle increases beyond 90° so the adjacent side once more acquires a finite length which
steadily increases but with a negative sign; thus the tangent curve disappears upward toward ∞
at 90°, re-appears at -∞, moves upward to cross the zero-line at 180° and then continues
upward again toward +∞. at 270°.
With reference to the sine function:

Peak or Crest values occur at 90° & 270°

Zero Crossings occur at 0°, 180° & 360°

Instantaneous Value constantly changes with Time

Average Value is zero

R.M.S. Value is peak-peak value / √ 2 ( = 0.707 x peak)

7.1.5 Angles Greater Than 360°

As line OA is rotated beyond 360° it clearly is retracing the previous circle and the results can
only be the same for trigonometric purposes. However angles greater than 360° do have a
significance in electrical-engineering and this is well illustrated by considering the generation of
an alternating current by a rotating machine.
Under 1.9.1 Induction (Fundamentals-1, Lesson-4) it was shown that current is induced to
flow in a conductor whenever an associated magnetic field changes in value or is moved relative
to that conductor. For a conductor to move along a line of magnetic force (a line of constant
magnetic strength) does not constitute a change; induction takes place only when motion of the
conductor causes it to cross a line of magnetic force. Maximum induction requires that the
conductor move at right-angles to the direction of the magnetic lines.

In a rotary generator a coil of wire is spun in a magnetic field as illustrated in Fig.8. Relative
motion between conductor and field is such that, at one part of the rotation, the conductor
moves parallel with the field while, 90° later, it is moving directly across the direction of the
field. Between these two extremes it moves partly along and partly across thc field with the
two components constantly changing in a cyclic manner.

In Fig.8 the line OA drawn from the centre of rotation to the conductor shows the motion of the
conductor; it is the same line OA that was drawn in Fig. 5. The induced current is
proportional to the motion of the conductor across the field and this is given by the sine of the
angle through which the machine has rotated. As the angle of rotation constantly changes so
the value of the induced current constantly changes; the current-value at any instant of time is
referred to as the instantaneous current and is given by the expression peak-current x sinθ. The
machine makes many revolutions per minute and so the angle θ (in theory) becomes very large
indeed but the practical result is that the sine function, as plotted in Fig.7 repeats endlessly. The
repetitive curve shown in Fig. 9 is known as a sinewave.
The same curve (but mathematically different) that starts from a point of maximum value
instead of zero is known as a cosinewave; the difference for a practical engineer is that the two
curves are relatively phase-shifted by 90° and are said to be in quadrature.

Both sinewave and cosinewave represent a continually-changing quantity with a cyclic pattern
of change. Under steady-state conditions all the positive peaks have the same value, all the
negative peaks have the same numerical value and the so-called zero-crossing points - where
the curve constantly crosses the horizontal axis - are equally spaced and fall exactly between the
positive and negative peaks.

The time required to complete one whole cycle of a waveform is measured between two
successive positive crests, between two successive negative crests or, much more accurately,
between alternate zero-crossing points. This interval is known as the Period of the wave and
usually is assigned the symbol T. This period T is obtained by dividing the number of cycle-
per-second (the frequency) into 1 second; hence T = 1/f

Although f represents the number of cycles per second a committee has decreed that frequency
shall be expressed in Hertz where

1- Hertz = 1 cycle-per-second.

Where frequency is measured in radians per second the name Hertz is not applied.

7.1.6 Harmonics

The sinewave (or the cosinewave) is the simplest of repetitive waveforms; it is a pure
waveform. All other waveforms can be analysed into (or synthesised from) a number of
component sinewaves and the process of breaking down a wavetrain into its component
sinewaves is known as Fourier Analysis because it was developed by the French Mathematician
Fourier.

In repetitive waveforms the components are harmonically related. This means that their
frequencies are related by a ratio of whole numbers; i.e. 2 to1, 3 to1, 16 to 1 etc. In any series
of such related components the lowest frequency present is termed the Fundamental or
sometimes the First Harmonic. If the fundamental is represented by f then a frequency 2f is
termed the second harmonic, a frequency of 3f is termed the third harmonic and so on.

Where harmonic frequencies are related also by a factor of two the intervals f - 2f - 4f - 8f etc.
are termed octaves. This is a clean and convenient mathematical rule but, in Nature, harmonics
are not related by a factor of exactly 2. Two musical notes which, to the ear, sound pleasingly
an octave apart prove to be not quite 2:1 in ratio and this is why it is not easy to tune a keyboard
instrument or any other musical instrument which uses fixed intervals.

For purposes of calculation an important parameter is the rate at which a waveform changes; for
example, in a rotating machine the induced current is proportional to the rate at which the
conductor cuts through the magnetic field. With reference to Fig. 5 this is the rate at which the
angle θ changes and, expressed in radians, this is 2π f radians per second (2π is one whole cycle
and f is the number of cycles per second). This combination is often referred to as the angular
velocity and it occurs so often in electrical calculations and expressions that it is represented by
the Greek letter ω (omega). Thus the angle swept out by the rotating line OA in a time t is given
by ωt radians (the angular velocity in radians/second x time in seconds).

When a waveform is analysed into its constituent components thc result is a set of frequency-
components each of which is distinguished by its individual amplitude and also by its phase
relation (its time-displacement) with respect to the fundamental. It appears however that the ear
is not particularly sensitive to phase relations. The ear finds odd-order harmonics (3rd, 5th, etc)
much more objectionable than even-order harmonics and this is the reason that triode valves
were favoured in audio amplifiers over pentodes; the triode amplifier tended to produce only
even-order harmonics while the pentode amplifier produced mainly odd-order harmonics.

7.1.7 Addition and Multiplication of Sines

By suitably constructing diagrams that can be broken down into inter-related sets of right-angle
triangles it is possible to produce algebraic formulae for combinations of angles such as the sum
sinA + sinB(where A and B are angles) and for the product [Link]. Such work is beyond the
scope of this Course but the idea is introduced because it has an important bearing on the
matters discussed in Fundamentals-2 under Frequency Translation/Multiplication and also
under Superheterodyne Receivers.

A complete set of these formulae can be found in electrical and mathematical reference books.
The two given below answer the question: " What happens when two or more sinewaves are
mixed in the same circuit?" The two formulae are:

(i)

sinA + sinB = [Link]½(A + B).cos½(A - B)

This indicates that, if you put two different frequencies A and B into a linear (a non-distorting)
circuit, they add together to give two different frequencies one of which is (A + B) and the other
(A- B). The expression also says that one of these (A + B) in a sine function while the other is a
cosine function; that means that they are in quadrature - shifted in phase by 90° relative to each
other.

(ii)
 [Link] = ½(cosA-B) - ½(cosA + B)

This indicates that, it you put two different frequencies A and B into a non-linear (a distorting)
circuit, they are multiplied together to give two different frequencies one of which is (A + B)
and the other (A - B). The expression also says that both are cosine functions which indicates
that they are in phase.

Both formulae show that, if you wish a circuit to handle a mixture of frequencies in this way,
then it must possess a bandwidth that extends (not from A to B) but from (A - B) to (A + B).
The newly-generated sum and difference frequencies are known as side frequencies which are
dealt with under Transmitters and Receivers in Fundamentals-2.

Back to Top of Page

END OF LESSON
 PART 7: MATHEMATICS FOR AMATEUR
RADIO
LESSON 2
7.2 POWERS, INDICES, LOGARITHMS & DECIBELS

7.2.1 Bases: Numbering Based on 10

We normally count in what is known as the Decimal System or the system of tens. Most
probably this system was selected simply because, with five digits on each hand, we have a total
of ten counting aids.

The decimal system is written using a total of ten symbols namely zero (or nought) and the
numerals 1 to 9. When the count is greater than 9 we simply start all over again from 0 but
place a 1 to the left to indicate that we have been round the circle once already.

Thus the hieroglyphic one-eight (18) indicates that the total is once-round plus eight;

i.e. (10 + 8).

Three-seven (37) indicates that we have been round the circle 3 times plus 7;

i.e. 10 + 10 + 10 + 7.

When the number to the left reaches 9 and still the count continues past nine-nine (99) then we
return both figures to zero and place a 1 to the left again. It is not surprising that a short-speak
was invented that makes three-six-four (364) read as three-hundred-and-sixty-four.

Each figure, starting from the right, has been given a name thus ; units, tens, hundreds,
thousands, ten-thousands, hundred-thousands, millions, tens-of- millions, etc. Note that in
Britain we call one-million-million a billion while, in America, this name is given to a thousand-
million.

Mathematically we write these in terms of powers of ten. For example:

ten lots of ten we write as one-hundred or 10 x 10 or 102

ten lots of one-hundred is one-thousand or 10 x 10 x 10 or 103

ten lots of one-thousand is ten-thousand or 10 x 10 x 10 x 10 or 104

The first of these in the right-hand column we read as ten-squared and the second is ten-cubed.

From then on we speak of ten-to-the-power-of-four

and ten-to-the-power-of-five etc.

The basis of this system is the number 10 and so we say that we are working in Base-1O.
7.2.2 Other Bases

Numbering however is not limited to the Base 10 ; as already mentioned base-10 probably
became popular because we have 10 fingers. In fact any number can be used as a base and
perhaps the best known is Base-2 which gives rise to the Binary Arithmetic used almost
exclusively in computers.

In Base-2 only two symbols are required namely zero and one. The first figure is still units, the
second now becomes two's, the next fours (2 x 2), the next eights (2 x 2 x 2), 16's, 32's, etc. In
place of "ten to the power of" we now have "two to the power of".

Binary arithmetic may look a lot more simple than decimal arithmetic (and in many ways it is)
but the price lies in the size of binary numbers. The following Table compares numbers in Base-
10, Base-2 and Base-3

Number Decimal Binary Base-3

zero 0 0 0
one 1 1 1
two 2 10 2
three 3 11 10
four 4 100 11
five 5 101 12
six 6 110 20
seven 7 111 21
eight 8 1000 22
nine 9 1001 100
ten 10 1010 101

A quick look at your cheque book will indicate the lengths to which binary numbers can grow.
The Table shows that, as the base-number increases, so the required number of digits decreases.

WARNING Don't fall into the habit of thinking in decimal-number names when dealing with
other bases - it can get you horribly confused. For example the binary-equivalent of seven is
one-one-one (NOT one-hundred-and-eleven) ; the tertiary equivalent of six is two-zero (NOT
twenty).

It can be shown mathematically that the optimum base in which to carry out arithmetical work
lies somewhere between base-2 and base-3 and this argues that computers could use either . The
choice of a binary system however is a matter of practical reality. A transistor can be made
either to conduct or not to conduct (to be cut-off) and so it forms an ideal 2-state device which
can represent 0's and 1's. Normally cut-off would represent a zero-condition but, within any
particular system, you are of course at liberty to define the states as you wish.

The same transistor could be made into a 3-state device by having a third condition in which the
transistor is neither cut-off nor conducting as heavily as possible - a "half-way" state. The
problem is that the half-way state cannot be maintained exactly ; it must be defined within limits
(it requires a tolerance) and the system becomes unreliable as the operating conditions drift
toward those limits.

7.2.3 The "Scientific" Notation

Manufacturers of electronic calculators and computers are pleased to give this name to the
mathematical practice of expressing large numbers in powers of 10. They have a problem in
that, given (say) an 8-digit display on their product, they cannot show any number greater than
99 999 999 or any number which is smaller than .000 000 1 (at least one digit is required to
show the decimal point).

Even with a pen and paper it is hazardous to perform arithmetic with such large numbers
because it is only too easy to make a mistake and very difficult to sort it out afterward.

The technique of course is varied to suit the problem of the moment but the basic idea is to
reduce all numbers to a single digit before the decimal point; i.e. all numbers are expressed in
terms of units only plus a multiplying factor which indicates by how much the figure must be
either increased or decreased to obtain the correct notation.

For example:

10 becomes 1 x 10
156 becomes 1.56 x 102
12 423.78 becomes 1.242378 x 104

As a simple rule-of-thumb the small raised number which indicates the power of 10 shows how
many places the decimal point has been shifted. As already discussed the "10" is known as the
Base; the raised figure is known as the Index (Plural indices).

For quantities less than 1 (fractions of 1) the same rules apply. The decimal is moved until there
is a single unit; however the decimal point is now being moved in the opposite direction and so
we indicate this by placing a minus-sign in front of the index:

0.1 becomes 1 x 10-1

0.054 becomes 5.4 x 10-2
0. 006010 becomes 6.01 x 10-3

There is one loose end to be cleared up before we can use this notation to do simplified
arithmetic. Between positive and negative induces there must be an index equal to zero but what
does it represent ? Compare the two Tables above and you will see that a number is missing
between (1 x 10+1) and (1 x 10-1) ; i.e. between 10 and 1/10. That number is 1 and so the
expression 100 must indicate the value 1 ? In fact the argument holds in any Base - where x is
any number we can write x0 = 1

Remember the rule-of-thumb that the index shows how many places the decimal point has been
moved?
0
It follows that 10 indicates that the point has NOT been moved.

Powers of 10 are easy to handle as can be seen by writing the problem out in full: for example:
100 x 10 000 = 10x10 x lO x lO x lO x lO = 1-million (102 x 104 = lO6)

10 000 / 100 = 10 x 10 x 10 x 10 / 10 x 10 = 100 ( 104 / 102 = 102 )

The rule is that, for multiplication add the indices.

for division subtract the indices

This rule of thumb gives us a valuable idea in using indices. The operation "divide-by-one-
hundred" ( /102 or subtract the index) is the same as the operation "multiply-by-one-hundredth"
( x10-2 or add the negative index). In other words we can transfer a power-of-ten on the
bottom line of any fraction to the top line of that fraction if we change the sign of the index.
(The reverse operation is valid too.)

In this way any arithmetical problem is reduced to the multiplication or division of two numbers
which are expressed in units and parts of 1. For example:

172.834 x 19.726 = 1. 72834 x 102 x 1.9726 x 10

= 1.72834 x 1. 9726 x 103

To help us on our way this is roughly 1¾ x 2 x 103 (approximately 3,500)

OR

85.012 x 460. 373 / 19.47 = 8.5012 x 101 x 4. 60373 x 102 x 10-1 /1. 947

which is roughly 8.5 x 4.5 x 102/2

= 20 x 100 (approximately)

7.2.4 Logarithms

So far we have considered only indices which are whole numbers; in Base-10 this limits us to
the real numbers in the sequence 1/100, 1/10, 1/1, 10, 100, and so on.

The number 6 lies between 1 and 10 while the number 60 lies between 10 and 100. It should be
possible therefore to fill-in the gaps between 100, 101 and 102 with a range of indices which
represent the numbers from 1 to 10 and from 10 to 100. Exactly how mathematicians work-out
these numbers is not of importance here; all we need to know is that it can be done, it has been
done and that the result can be very useful. When dealing with the full range of indices, which
represent a full range of real numbers, we change the name from indices to Logarithms.

Remember the rule :

multiply two numbers add the indices
divide two numbers subtract the indices

The logarithm of every possible number is presented in tabular form in what are known as Log
Tables. To multiply or to divide two (or more) difficult numbers all that is required is to extract
from the log-tables the appropriate logarithms and then add or subtract them as required. This is
a very straightforward computation and should not lead to error no matter how formidable the
original numbers may have been.
Addition (subtraction) of the logarithms (in common usage logs) results not in the final answer
but in the log of that answer. It is now necessary to go back into the Tables and find the
original number of which this answer is the logarithm. In fact that is not a difficult task but, to
speed the process, mathematicians have prepared the Tables "inside-out" as it were in a form
known as Anti-log Tables; in these you enter with the log and find the number (the anti-log).

Instructions on the use of Log Tables are given in detail in an Appendix at the end of this Lesson
(to come).

In practical work, when in the middle of a long calculation, it can be tiresome and distracting to
be obliged to write "find the log of 100 and multiply it by 6 " ; equally we may well wish to
write "find the number whose log is 2 and multiply that number by 6". It matters very much that
there should not be room for confusion because the answer to the first proposition is 12 while
the answer to the second proposition is 600.

A shorthand notation has been generally accepted to avoid such ambiguities. The first of the
propositions would be written as [Link] 10 100; this is read as six-log-to-the-Base-10-one-hundred
or more simply as six-log-Base-10-one-hundred. In practice however most work is carried out
in logs to base-10 and it is customary to omit the Base and to assume that the Base is 10 unless
otherwise specified. Thus the first proposition would be written as [Link] 10.

The second proposition "find the number whose log is 2 and multiply that number by 6" is
written as 6.antilog2 (again assuming Base-10). Even more conveniently it can be written as
[Link]-12. Remember that this is no more than an agreed convenient shorthand; that index (-1) is
entirely without mathematical significance or complication.

(This style of notation is found again in the Lesson on Trigonometry.)

For normal calculations accuracy is not required beyond three decimal places and so standard
log-tables are calculated to four decimal places. Navigation, for example, requires much greater
accuracy and it is necessary to employ 7-figure tables. The extremely useful and fast slide
rule is simply an adding-stick whose scales are calculated in logarithmic intervals; this
machine has been discarded in favour of the electronic calculator but, in truth, for many
purposes in practical engineering it is superior to the calculator.

Logarithms based on 10 are known as Natural Logarithms. Other Bases are used but their
study would not serve any useful purpose here.

7.2.5 Decibels

Survival, whether as hunter or hunted, depends on the sensitivity of our senses. An eye that
functions by starlight, or an ear that detects an earthworm below the surface, both have serious
disadvantages during a thunderstorm. For this reason all the senses have inhibitors which are no
more than feedback systems that reduce sensitivity according to the degree of stimulation.

As a result the response of each sense versus the magnitude of stimulation is not a linear
relationship but is close to being logarithmic. For example, under laboratory conditions, if you
present the ear alternately with two sounds which differ only in sound-level it will not register
any difference in loudness unless one sound is (at least) twice the power of the other.
For scientists and engineers, who attempt to rationalise their work through measurement and
tabulation, this dependence of relative loudness on absolute loudness creates problems. For any
just-discernible change of loudness the power level must increase by a factor of two. For
example if the first sound results from dissipating 10 mW then the second must be at least 20
mW. BUT. if the first sound results from dissipating 1-watt, then the second must be at least 2
watts. At the lower of these two levels the apparent increase is caused by an extra 10-mW
while, at the higher level, exactly the same result is obtained with an increase of 1-watt.

The discussion above on indices and logarithms provides a solution to this problem of
measuring power levels as they sound to the human ear. When numbering in Base-10, each
multiplication by 10 increases the index by one. Using the index only - such a power ratio
(10/1) is referred to as 1 Bel; we say that the second sound is louder (or quieter) than the first
by 1-Bel.

In practice however the Bel is found to be too large a unit and so it is divided into ten intervals
called decibels. Hence, by definition, a power ratio is expressed in bels as

log P1/P2
(where P1 and P2 are the two power-levels to be compared)

or, in practical units as

10. log P 1 /P 2 decibels. (dB)

Although it sounds complicated the decibel is truthfully a very simple soul. Any changes of
power-level are first reduced to a ratio (by dividing the smaller into the larger) and then, via the
simple formula above, to a number of decibels. From then onward overall changes in power
levels can be obtained very simply by adding (and subtracting) decibels. At the end of the
exercise, if an answer is required in absolute terms, then the final overall figure expressed in
decibels is converted to a power-ratio by using the formula in reverse.

It does not matter what the absolute power level may be because the dB is concerned only with
power ratios.

Suppose for example that your transmitter is developing 50 watts but the operator at the other
end of the circuit is complaining that he can barely read you. He has no way of measuring your
output power but his receiver is fitted with an S-meter. Because you are a considerate person
you increase your power to 100 watts which is a two-fold increase. Log 2 is approximately 0.3
and so you have increased the power by 10 x 0. 3 or by 3-dB. Although your contact is not
aware of your radiated power his S-meter will register an increase of 3-dB in signal (about one-
half an S-point) and (if he is lucky) his ear will just about register an increase in loudness.

It could be that the 3-dB increase is just enough to make the a.g.c. circuit of his receiver reduce
the gain and so reduce the background noise in which case he will report an improvement in
readability. It is more likely that he will continue to ask you to raise the power. (He might be
aware of the increase and so make a greater effort to read you and thus report a much better
signal - we are all human).

Now raise your power by a further 50-watts to a total of 150 watts (the legal limit). The
increase, from 100 to 150 watts, is a ratio of 1.5. In decibels the extra increase is given by
 l0 x log 1.5 = l0 x 0.176 = 1.8dB

which will not make any noticeable difference at the receiving end although, once again, it
might reduce the background noise through the a,g.c. action.

Had the power been raised from 50-watts directly to 150 watts (i.e. by a factor of 3) the increase
expressed in dB's would have been

l0 x log3 = l0 x 0.4771 = 4.8dB

an answer we could have obtained more easily from the previous calculations by adding 3-dB to
l.8-dB.

Note that QRP (or low-power) operators who insist on keeping their power-levels below 5 watts
are not by any means cranks. The purchase of Linears and the constant escalation of output
power is a true instance of diminishing returns. The only purpose served by raising power levels
is to overcome background noise; when several people raise their radiated-power levels they
are doing a very good job of raising the general background noise.

Unfortunately the availability of high power produces a tendency to lazy engineering and most
of that extra power is wasted.

7.2.6 Other Ratios

As explained above the decibel scale relates strictly to power ratios. To actually measure power
it is necessary to place a voltmeter across the circuit and an ammeter in series with the circuit
and then to multiply the two readings. Although perfectly valid this technique is cumbersome.

Power measurements are usually required in arrangements where the value of the load is known
and this permits a simplified technique using Ohm's Law. As discussed in Lesson-6 of
Fundamentals-1 the power dissipated in a resistor of value R is given by

Power = E2/R

and so most power-meters utilise a dummy load of known value across which is connected a
voltmeter whose scale is calibrated not in Volts but directly in Watts. The squared term means
that the scale cannot be linear but is cramped toward the lower end.

Where signal levels are measured in this manner it is necessary to modify the expression for
decibels; a ratio of power-levels must be replaced by a ratio of voltages-squared (or currents-
squared) and, to square a quantity in logarithmic form, the logarithm is multiplied by 2. Hence,
when measuring in terms of voltage (or of current) the ratio of two signal-levels E 1 and
E 2 becomes

20. log. E 1 /E 2 decibels

The difference in these two expressions is important. To check the output power of a
transmitter a dummy-load is connected. If this is just a dummy load then it is necessary to
connect a rf-voltmeter in parallel with that load or an rf-ammeter in series with the load. Any
measured power differences are expressed in dB's using the second expression (x20). If the
dummy-load is a commercially-produced instrument then it may have a voltmeter built-in which
is calibrated in terms of watts; any measured power differences are expressed in decibels using
the first expression (xl0).

To measure the strength of an incoming signal the power delivered from the aerial is in the
range from microwatts to milliwatts and the option of a true power reading is not available.
Here a receiver is used as a voltmeter and signal differences are turned into dB's using the
second expression.

Note that decibels always refer to the ratio between two power levels but a signal level or a
noise level is often seen expressed in terms of dB's. This is possible because the level is being
compared with a standard level BUT it is essential to state that reference level. Statements
such as "Your signal strength is 24 dB ..." or "The noise-level from your car exhaust is 60-
dB..." are meaningless. A value of 24-dB reference 1-V can be turned into a signal strength in
volts; 24-dB reference 1mW can be turned into a signal strength in milliwatts. Those two
measurements cannot be related however unless the impedance of the circuit is given; the noise-
level from your exhaust cannot be calculated unless a standard reference-level is known.

Because of mankind's love of jargon a convention has arisen to avoid having to state the
reference level. In this a letter is added after "dB" to indicate the nature of the reference level -
and it works fine as long as everybody knows, understands and respects the symbols.

dBV indicates that the reference is 1-volt

dBm indicates that the reference is 1 milliwatt
dBW indicates that the reference is 1-watt
dBV/m indicates that the reference is 1-volt-per-metre

In the Amateur Licence the power levels permitted on each Amateur Band are stated in dBW. It
would of course be much less open to error if the levels were simply given in watts. Clearly it is
essential to know the difference in meaning attributed to upper and lower case letters,

You will frequently see noise levels (aircraft, vehicle, fan, etc) given as so many decibels. You
can be certain that the writer is not sure of what he is writing ; in expressing noise levels an
extra suffix is added to indicate the type of frequency-weighting (frequency-correction) under
which the measurements are made. It is interesting to know that a document issued by the
Department of the Environment as guidance for Town Planners claims that 0-dB represents zero
noise-level and that the minimum change of level which is detectable by the human ear is 10-dB
. If nothing else it makes life much easier for Local Councils but never buy a house close to an
airport or a motorway.

QUESTIONS

1. Write down the (decimal) numbers 1 to 20 as they would appear in Base-5

2. A pocket calculator shows an answer as 2. 78E4; what do you think this means?

3. Using the power-of-ten notation work out the answers to the following

(a) 2.5 x 61.29 x 319.4 x 0.006

(b) 17.43 x 1.116 x 42.38 ÷ 71
(c) 17.43 x 1.116 X 42.38 ÷ 0.71
(d) 1/2 x 3/8 x 3/4 ÷ 7/8

4. Obtain an approximate answer to 1,792.856 x 31.45 ÷ 29.017

5. If logA = .386 and logB = .614 what is the value of A x B?

6. A communications receiver has had its S-meter calibrated and found to be

very close to 6-dB per division from S3 to S7. An incoming signal is
causing an indication which, for most of the time, is S6 but which fades
down to S4.

When the receiver is connected to a signal generator it is found that an input of 1.5 mV gives an
indication of S7.
What are the maximum and minimum levels between which the signal is fading?

7. The input voltage to an audio amplifier is 10 mV. The voltage which it is developing across
an 8-ohm resistive dummy load is 1-volt. What is the gain of the amplifier expressed in dB ?

What extra information would you require to calculate the input power? Beware - I'm trying to
catch you.

Back to Top of Page

END OF LESSON
 PART 7: MATHEMATICS FOR AMATEUR
RADIO
LESSON 3
7.3 GRAPHS

7.3.1 General

Graphs are a pictorial way of presenting information about quantities which change. Such
changes are accurately described in a Table of Figures but even experienced experts can derive
information from only relatively small Tables.

Perhaps the most spectacular use of graphical display is the ubiquitous television screen. Here,
information is delivered as a sequence of dots (red, blue and green) at a rate around 5-million
per second. It is impossible for the human brain to deal with such a stream of information but,
when, those dots are properly displayed, we become aware of an image — in fact of 25 images
in each and every second.

In the language of graphs there are two (sometimes more) variables or quantities which change
and these come in two types:

(a) the independent variable which is always represented by the letter x and which, as its name
implies, can assume any value.

(b) a dependent variable which is always represented by the letter y and whose value depends
on the value assigned to x.

The value of y is determined by the value of x according to a Rule which is expressed as an

algebraic equation. For example the simplest of equations

y = x .........(i)

tells us that, whatever value is assigned to x, that too is the value assumed by y. A slightly more
complex form

y = x + 2 ...(ii)

tells us that, whatever the value assigned to x, the value assumed by y is the same va1ue but
increased by 2. Similarly the equation

y = x - 3 .... (iii)

tells us that, whatever the value of x, the value of y is that value decreased by 3.

The interesting thing about such equations is the manner in which y changes as the value of the
independent-variable is changed and this is not at all apparent until the complete set of values
are displayed simultaneously.
7.3.2 Preparing the Figures

To present the above equations as a graphical display it is first necessary to determine the values
of y for each value of x (over the range of interest) and this is laid out in tabular form in Table-
1:

TABLE 1

Table-1 shows, along the top line, all the values of x (in this example whole numbers only)
ranging from -4 though zero to +4.

The second line shows all the corresponding values for y under the Rule that y = x.

The third line shows all thc corresponding values for y under the Rule that y = x + 2.

The fourth line shows all the corresponding values for y under the Rule that y = x-3.

7.3.3 The Axes

To plot these figures as a graph Table-1 is turned into a geometric form by representing each
figure as a line of appropriate length. Nothing in the Universe is absolute and so a first
requirement is to define a point from which all measurements are to be taken. Such a point,
marked on a piece of paper, is usually labeled "0” or nought.

The next task is to construct a frame of reference — a pair of axes (the plural of axis) one of
which is to represent the values of x and the other to represent the values of y. Both axes must
share the reference point and so they are drawn mutually-perpendicular as shown in Fig. 1.

The horizontal axis is always used to represent the independent—variable and so the vertical
axis sn always used for the dependent-variable.

Positive values for x are always plotted to the right arid positive values for y are always plotted
upward.

Starting from zero each axis is marked off at equal intervals whose length is determined
arbitrarily by the greatest number which it is required to plot along that axis. Clearly the bigger
the intervals the more accurately the numbers can be represented.

7.3.4 Plotting Co-ordinates

A pair of figures which represent the corresponding values of x and y (taken from a Table such
as Table-1) are known as co-ordinates and, in general, they are written simply as 2,4 meaning
that x = 2 and y = 4. When a point is plotted from these figures on perpendicular axes as shown
in Fig.1 they are properly called Cartesian Co-ordinates. (There are other ways of plotting
graphs.)

As already explained the co-ordinates 2,4 represent the values to be assigned

to x and y respectively. To plot such a point move first from "0” to the right along the
horizontal x axis to the position marked “2” and then travel upward parallel to the vertical y axis
until level with the position marked “4”.

(Alternatively the point is marked at the top of a vertical line 4-units in length which stands at
position "2" on the x axis. Yet again you may prefer to think in terms of vertical lines parallel to
the y axis and horizontal lines parallel to the x axis.)

The process is repeated for each value of x and, from Table-1, for each of the three equations.
Each set of points is then joined by a continuous line to produce the three "curves" illustrated in
Fig. 1.

The first thing to notice about these curves is that, in fact, they are straight lines and, as a result,
the basic equations are referred to as linear equations. Linear equations deal only in terms
which are raised to the power of one [see Lesson-2 in this Part]. Equations which include terms
such as x2produce plots which are curved and which show a single maximum or a minimum
value. Equations with terms raised to the power 3 (cubes) produce curves with two reversals -
one maximum and one minimum va1ue. In general the number of reversals is one less than the
highest power present in the equation.
The second point to note is that all three curves slope upward and to the right; this is because
the equations all relate to values of y which increase as x increases. Such curves are said to have
positive slopes or positive gradients. This is dealt with again under 7.3.5 Negative Slopes.

The third point is the intercepts which are the values at which the curves cut either or both axes.
The simplest equation x = y passes through the origin (the common point which represents zero
on both axes) and so both intercepts are zero. This can be deduced by making the independent-
variable equal to zero; the corresponding value for y is zero and so the curve must intersect both
axes at the origin.

When x is made zero in the second equation the value of y becomes 2 and so the curve intersects
the vertical axis (where x = 0) at the point y = 2. Similarly, in the third equation, when x is made
zero y becomes -3 and so this curve intersects the vertical axis at the value "3 ” below the origin.

Have you realised that these linear curves have other intercepts because, unless the curve passes
through the origin, it must cut both axes. To find these points put y equal to zero. Thus, in. the
equation y = x + 2, for y to be zero x must have the value -2. In the third equation, for y to be
zero, xmust have the value +3.

This demonstrates a useful property of linear curves :

(i) they are straight lines and so
(ii) they can be drawn by calculating the intercepts and joining them with a ruler.

Inspection of the curves and Table-1 shows that the last two are but the first one either raised by
2 or depressed by 3. A plot of y = 2 is simply a straight horizontal line drawn through the point
“2” on the y-axis which is another way of saying that, in this equation, y is not affected by the
value of x. Thus the plot of equation (ii) is the result of adding 2 to each value of y; i.e. of
summing the two curves ( y = x ) and (y = 2) to produce the equality y = x + 2.

7.3.5 Negative Slopes

A curve with a negative slope can be obtained by modifying equation (ii) above in that x is
given a negative value:

y = -x + 2 ............ (iv)

which leads to a set of values for y shown in Table-2

As might be expected reversing the sign of x has caused the va1ue of y to decrease as x increases
in value. When this is plotted as in Fig.2 the curve slopes downward and to the right.

The terms negative slope and positive slope are so named because we regard the slope as
the tangent of the angle contained between the curve and the x axis (see Lesson-1:
Trigonometry in this Part). Gradients are defined as so many units rise (measured on the y-axis)
for so many units of travel (measured on the x axis) - e.g. 1 in 3* - which, in Figs. 1 & 2,
becomes y/x.

*In road work the gradient is more usually defined as the number of units rise in so many units
along the road; here gradient is the sine of the angle contained between the road surface and
the horizontal.

In Fig.2 the independent parameter x is increasing from 0 to the right while the value
of y decreases toward 0 (downward). Changes of y in this manner are given negative values and
so the ratio y/x becomes negative. In the same manner quantities moving to the left (either
decreasing positive values or increasing negative values) are given negative signs; on the y axis
negative signs are given to values which “decrease” downward either toward 0 from a positive
value or away from 0 with negative va1ues.

I hope it is clear from the above that the equation

-y = x + 2 ............. (v)

must also produce a negative slope ? Note that

-y = -x + 2 ........... (vi)

has a positive slope but it is the same curve as equation y = x - 2. Try plotting it.
(The Rule in algebra is that quantities may be transferred across the equals sign provided the
sign is changed. To derive equation (vi) from equation (y = x - 2) take ALL the terms across the
equals sign. See Lesson-4: Algebra in this Part.)

7.3.6 Equations With Higher Powers of x

A more ambitious equation might be

y = 2x2 + 3x - 2 ............(vii)

This equation contains a term in x-squared (raised to the second power), a term in x (raised to
the power of 1) and a term (2 x 1) in which x has been raised to the power 0; i.e. it has been
reduced to the value 1; see Lesson-2: Powers, indices, etc.

A set of solutions is determined in Table-3: note that, when a negative quantity is squared (i.e.
it is multiplied by another negative quantity) the result is positive; see Lesson-4: Algebra and
also Lesson-7: The j-Notation.

This curve is plotted in Fig. 3. The squared term results in the value of y increasing for both
positive and negative values of x and it is this which makes the curve show a minimum value.
Those familiar with Calculus will know that the value of both x and y can be calculated for the
minimum value because, at the minimum, the gradient of the curve is zero; i.e. the curve is at
the point where the gradient changes from negative to positive.
 Fig. 3 Plot of an Expression Which Contains a Squared Term

EXERCISES

Plot the following equations:

y = x3 + 4x2 + x - 3 y = x5 + x3 - 5x

NOTE: The last curve should show 2 minima and 2 maxima.

7.3.7 Use of Non-linear Scales

In practical work it is seldom required to plot a curve from an equation; if anything it is more
likely that an equation is to be derived from a practical curve. Curves are plotted point by point
as measurements are taken from a piece cf equipment under test; this is much quicker than
tabulating the results first but, because it provides an instant assessment of performance, it
enables a test run to be abandoned when it becomes obvious that all is not well. (See also Part-
6: Lesson-4: Sweep Oscillator. )

There is always a compromise to be made between accuracy of plotting and the range to be
accommodated. For example, when testing an amplifier, interest may lie in the manner in
which gain varies as the frequency changes. For a 10.7 MHz IF-amplifier the frequency range
may be around 10-MHz to 11.5-MHz and this is easily accommodated on a linear scale such as
that above. However, for an AF-amplifier ranging from 30 Hz to 16-kHz there is a problem in
paper size. In this instance the scale has to be compressed by calibrating the x-axis
logarithmically - this is described below.

Gain - dependent on frequency - will be plotted on the y-axis. However we are not so
concerned with the actual output-variation of an audio amplifier as with the manner in which the
human ear perceives those variations. The minimum change in sound level which the ear can
perceive is 3-dB which is a power-change of 2:1 (see Lesson-2 in Part-7). To achieve a further
noticeable increase it is necessary to double the output again. In a manner of speaking therefore
we are interested in the ability of an audio amplifier to continuously double its output power.

Output levels can be expressed as 20, 21, 22, 23, 24, ..... and this is a Power series or
a Logarithmic series; such a series is used in Lesson-2 of Part-7 to derive the decibel scale.
Note that, although the output is rising by leaps and bounds, the indices - the powers of 2 - are
increasing in steps of 1. In other words the indices form a linear series; thus the output in
decibels can be plotted on a linear scale or the output in volts, amps, or power can be plotted on
a logarithmic scale with the same result.

There is one more practical point. Suppose that, for some reason, it was required to plot power-
dissipation against current. From Ohm's Law the power is a square-law function of current
namely P = I2.R. In performing the experiment there are bound to be experimental
errors which cause some of the plotted points to lie off the true curve. As seen in the above
examples curves which involve powers soon depart rapidly off the paper. Thus a curve has to
be drawn accurately through points scattered around a true curve which is rapidly approaching
the vertical.

The easy answer is to plot the results of power measurements on a logarithmic scale and so turn
the curve into a straight line: this can then be drawn as best fit using a ruler. Alternatively the
equation is re-arranged by taking logs of both sides:

logP = 2. logI. k (k has a constant value)

when logP can be plotted against logI on linear scales with the same result.

A logarithmic scale is marked off in equal distances but the points are calibrated in terms of
indices rather than in the figures which they represent. The scale repeats endlessly and the major
divisions may be marked (for example) as 1, 10, 100, 1000, 10,000, etc.: the equal spacings
represent the logarithms of these numbers. Note that such a scale does not pass through zero;
on a logarithmic scale zero is represented by minus-infinity. The scale is selected to cover the
range of interest.

Graph paper is printed in linear/linear, log/linear, linear/log and log/log formats.

Back to Top of Page

END OF
 PART 7: MATHEMATICS FOR AMATEUR
RADIO
LESSON 4
7.4 BASIC ALGEBRA

7.4.1 Introduction

It is normal when teaching mathematics to start with Arithmetic and to leave Algebra until later
as an advanced study. In truth this is a curious approach because Algebra explores the Rules of
Mathematics and it is the basis of arithmetical techniques.

When "solving" a problem using Algebra the purpose is not to obtain a numerical answer but to
derive a method for solving that problem. Consider for example the problem: A train leaves
London for Bristol; how fast does it travel?

Those skilled only in Arithmetic might well reply with: “How long is a piece of string?”. It is
not possible to provide an answer unless we are given:

(a) the distance between London and Bristol

(b) the time taken to make the journey.

Algebra can provide a solution in the form of a formula; when the missing information
becomes available the figures can be entered into the formula and thc answer is quickly
available.

7.4.2 To Obtain a Solution

The first step in solving the above problem is to provide substitutes for the missing items and
this is the source of that much-parodied phrase: “Let x be the unknown quantity”.

Let the distance from London to Bristol be d miles

Let tim time taken for the journey be T hours

Speed is expressed as distance traveled in a unit of time; for example miles per hour, metres
per second

(Note: it is standard practice to use d to represent distance and T to represent time but the units -
miles, metres, hours, seconds - must be given.)

It is now possible to write:

If in T hours the train travels d miles

Then in 1 hour (T/T hours) it must travel d/T miles (divide both quantities by T )

i.e. (that is) the speed of the train is d/T miles per hour (m.p.h.).
Although the speed is expressed here as m.p.h. it is, in fact, a general solution; for example if
we write:

" in T seconds the train travels d metres" then the speed is again given by d/T BUT the units are
metres per second (m.p.s).

To be really scientific about it there is one more step:

Let the speed of the train (in m.p.h.) be represented by v

Thus v = d/T ....................................... (1)

(Note: (i) v is specified in m.p.h. because d is in miles and T is in hours.

(ii) d/T is often read as "d by T".)

This final result is known as a formula albeit a very simple one: its usefulness lies in the fact
that it is a general solution. In this instance it was derived by considering a hypothetical train
on a somewhat facetious journey but it is equally applicable to any object in motion.

For example: If a man walks a distance of 15 miles in three hours then the quantity d becomes
15 miles and the quantity T becomes 3 hours; it is now possible to apply Arithmetic to the
formula to give his walking speed as

v = d/T = 15/3 = 5 m.p.h.

These formulae are called equalities because they contain an "equals” sign; the term means
simply that the expression on the left of the equals sign is equal to the expression on the right of
the equals sign.

NOTE that, in deriving the above formula, it is important to put on your thinking cap and to set
out the propositions in the form that will enable you to derive the quantity you seek. The
question asked was ‘How fast does the train travel?’; the train's speed is distance/time and so
we started with the distance, moved on to the time and then derived speed as distance divided by
time.

7.4.3 Manipulations

The above discussion derived a simple formula which expressed speed in terms of a distance
and the time required to cover that distance but, in the practical world, things are not necessarily
done in that order.

In the first example above the train driver had a speedometer which tells him his speed in m.p.h.
and he will be well aware of the distance to be covered; his problem more likely is to work out
how long will be required to complete the journey because he wishes to estimate. his arrival
time. In other words he does not want speed in terms of time and distance: he wants time in
terms of speed and distance. The purpose here is to understand the methods which are used to
manipulate equations to obtain any one of the quantities in terms of the others.

7.4.4 Addition and Subtraction

The formula far speed involves dividing one quantity by another but another formula might be
concerned with adding one quantity to another. For example: a man starts 5 miles from his
Home and walks x-miles directly away from Home; what is his final distance from that home?

Let the final distance be D (miles) : then

D = 5 + x .................. (2)

Alternatively, if he walked toward his Home, his final distance D must be less than his starting
distance (5 miles) and the equation becomes

D = 5 - x ............... (3)

By comparing equations 2 and 3 you might well deduce that reversing the direction of walking
is shown by changing the sign of x ? (If you really want to stir the pudding this can be
expressed by saying that the man is walking away from Home by a distance of -5 miles ?)

We can do anything we like to equalities provided that we do the same thing to both sides of the
equation thereby preserving the equalness. Thus, by subtracting 5 from each side of equation-2:

D-5 =5-5+x

or D - 5 = x ...........................(4)

This is a formula which gives the distance walked (x) in terms of the starting distance (5 miles)
and the final distance (D miles); x is expressed as a difference of two distances.

The above is a very simple manipulation in that it involves only three small quantities; where
the expression on one side (or on both sides) of the equals sign is complicated it is possible to
make errors and so it is useful to develop a Rule-of-thumb which will help to eliminate such
errors.

To obtain x from equation-2 it was necessary to get rid of the number 5 from the right-hand side
(the r.h.s.) of the equality. By inspecting equation-4 it is seen that, in effect, the figure 5 was
taken across to the other side of the equality and its sign changed; i.e. +5 on the r.h.s. has
become -5 on the l.h.s.

RULE Quantities to be added or subtracted can be simply transferred across the equality
provided their sign is changed.

7.4.5 Multiplication and Division

Equation-1 involves dividing one quantity by another and the above Rule-of-thumb is not
applicable; the basic rule still applies however that, whatever operation is performed on one
side of an equality, the same operation must be performed on the other.

It was proposed that the Train Driver wishes to find the time T which is on the bottom line of the
fraction d/t in equation-1. A fraction is a way of writing down a portion of some whole quantity.

The bottom line of a fraction is known as the denominator because it denominates (or names)
the size of the portions into which a whole unit must be divided; for example a 2 on the bottom
line tells you to divide it into two equal parts (halves), a 5 tells you to divide it into 5 equal
parts.

The top line of a fraction is known as the numerator because it specifies (enumerates) the
number of those portions that must be taken; for example 3/4 means divide it into four equal
quarters and select a total of three quarters.

RULE Divide by the expression on the bottom line and multiply by the expression on the top
line.

The Train Driver's problem is to extract T from the equation for speed:

v = d/T ............................ (1)

Multiply both sides by T:

v x T = d x T/T .......... (5)

If you multiply by T and then divide by T you should not expect to get anywhere; we say that
the two T 's, one on the top line and one on the bottom line, cancel. Put another way T divided
by T equals 1. Thus

v x T = d ...................... (6)

However the train Driver knows the value of the distance d; it is T that he needs. The process
is now repeated as it were in reverse by dividing each side by v:

T x v/v = d/v

or

T (x 1) = d/v .............. (7)

Once again if this manipulation is attempted on an equation with very complex expressions
there is a considerable possibility of error; a second rule-of-thumb can be obtained by
comparing equation-1 with equation-7:

RULE Quantities to be multiplied or divided can be transferred across the equality provided
that they are interchanged between the top line and bottom line of a fraction.

(Remember that a quantity such as T can be represented as a fraction by writing T / 1 ; i.e.

divide the whole into 1-only equal parts.)

Thus to obtain equation-7 from equation-1 it is much simpler (and therefore safer) just to
interchange the quantities T and v; across the equals-sign the T goes up to the top line and
the v goes down to the bottom line.

This rule-of-thumb method of manipulating an equation which involves multiplications and

divisions is known as cross-multiplication. If it sounds like cheating remember that it is
legitimate because it helps to avoid errors and also because it is soundly based in that it omits
only the more difficult "middle bit".
7. 4. 6 Mathematical Shorthand

It is tiresome to write, and also difficult to read, a script that constantly spells out words such as
“multiplication” or “divided by” and a notation is necessary. In mathematics there are several
different notations in use for the same operations; sometimes they duplicate each other but each
does have its own particular niche.

In multiplication the symbol x is used universally between the two quantities involved such as

156 x 27

This is satisfactory enough in arithmetic but leads to a deal of confusion in algebra where the
symbol x might also be used to represent an unknown quantity. One method used to overcome
this is to reserve the lower-case written x (which appears as two c’s back to back) for the
unknown.

Additionally the x may be replaced by a full-stop, by round brackets or simply omitted

altogether. Thus the expression "a multiplied by b" may be written as any of the following:
a x b a(b) b(a) (a)b (b)a a.b ab

The brackets are seldom used in the simple form shown here. The form a(b) means multiply
whatever is within the brackets by a. Consider for example a(p + q - s) ; this is read as “a into p
plus q minus s”; to evaluate this when given values for p, q and s work out first the additions
and differences within the brackets and then perform the multiplication.

In the computer world there is one other common variation because computers are unable to
differentiate between “x” meaning a letter, “x” meaning multiply-by and “x” used in place of an
unknown quantity. The symbol most often used to represent the operation of multiplying is the
asterisk; thus, in a computer program, you find expressions such as 5*L. It is worth noting
perhaps that, to get a computer to do pure algebra, is an extremely complicated business that
quickly reveals how much we take for granted.

The standard symbol for division is not very convenient in use and it is most often replaced by
the solidus (or slash in the USA). The problem however is one of order; there is no ambiguity in

a x b/c (or ab/c or even a*b/c)

but how should we interpret

a/c x b ?

It is in this kind of problem that the correct use of order and (where necessary) brackets becomes
important; the possibilities are:

a/bc a/(b x c) a/c x b/1 = a x b/c (or ab/c)

Confusions such as those shown above are obviated in computer programming by adopting a
strict order of precedence. However, unless you are well practised in programming that
particular computer, it is not easy to remember the relevant order and so the proper use of
brackets is safer and quicker.
[The word program is well established in the computer world where it refers to so-called
“software” or the written instructions that tell a computer exactly what it is required to do.]

7.4.7 Worked Examples

(i) If p + q = 7 what is the value of q?

Take the p to the r.h.s of the equation and change its sign:

q = 7 - p (or q = -p + 7)

Check this result by using the full step-by-step technique.

(ii) If x - ay = p - q what is the value of x?

Take ay to the r.h.s. with a change of sign:

x = p - q + ay

(iii) If x - ay = p - q what is the value of y? ( Watch those minus signs!)

Take x to the r.h.s. with a change of sign:

- ay = p - q - x

Take -a to the r.h.s. and on to the bottom line:

y = (p - q - x) / (-a)

(Note the use of brackets to keep together all the symbols which belong together; without the
brackets around -a there is the possibility of error in that the symbol a might be subtracted from
the expression instead of being divided into it.)

(iv) If x/y = p/q what is the value of x/p and of q/p ?

Cross-multiply with y and then with p:

x/p = y/q

Alternatively cross-multiply with all four symbols:

x p q y
- = - becomes -- = —
y q p x
(This introduces a useful extension of the Rule above that what is done to one side must also be
done to the other; simply turn the whole equation upside down.)

(v) If 3a - b + c = 2p what is the value of a ?

Take both b and c to the r.h.s. with a change of sign

3a = 2p + b - c

Divide everything by 3 (divide throughout by 3) ... take the 3 across to the bottom line:

a = (2p + b - c)/3

(vi) If 3(a - b + c) = 2p what is the value of a?

First remove the bracket so that a is released from b and c:

the expression 3(a - b + c) means three lots of (a - b + c) which is the same as 3a - 3b + 3c

(multiply everything within the bracket by 3) hence

3a - 3b + 3c = 2p

Take the terms in b and c across to the r.h.s.:

3a = 2p + 3b - 3c

a = (2p+3b-3c) / 3

or, if you prefer,

a = 2p/3 + b - c

Back to Top of Page

END OF LESSON
 PART 7: MATHEMATICS FOR AMATEUR
RADIO
LESSON 7
7.7 THE j-NOTATION

7. 7. 1 General

The j notation is a mathematical tool whose purpose is to prevent confusion between two sets
of figures which relate to different types of quantities. Mathematicians use the letter "i" but, in
electrical engineering, that letter has other connotations.

7.7.2 Derivation of j

Fig. 1 shows two axes which, by

representing two different quantities, are
used for plotting graphs as described
in Lesson-3 of this Part. To be more
accurate Fig. 1 shows only that part of each
axis which lies in the first quadrant; i.e. only
the positive values of an independent-
variable x(horizontal axis where x may have
any value) and only the positive values of
the dependent-variable y (vertical axis
where the value of y depends on the value
of x).

As discussed in Lesson-3 a particular point

in this diagram might be designated as 3,4
meaning that x = 3 while y = 4. The point is
plotted in Fig. 1 at the intersection of a
vertical line which passes through x = 3 and
a horizontal line drawn through y = 4. Carry
out such an exercise and you will find that,
as predicted by Pythagoras' Theorem, the
plotted point lies at a distance of 5 units
from the Origin.

This diagram shows one instance where simple arithmetical addition of the values of x and y
does not yield the correct answer namely, in this instance, 3 + 4 = 5 . The reason is that x was
measured from 0 in a horizontal direction while y was measured from 0 in a vertical direction
while the distance from point 0 to the plotted point is measured in yet another direction.

To indicate that these two numbers cannot be added except under a set of special rules the
letter j is placed before one of them. Numbers represented by the length of the horizontal axis
are referred to as real numbers and are written normally. Numbers represented by the length of
the vertical axis are known as imaginary numbers and are written with the letter j before them.
Thus this particular plotted point can be described as being at a distance 3 + j4 from 0.

Because the two axes are set at 900 the result of this vector addition can be calculated using
Pythagoras' Theorem which applies to any right-angled triangle namely that the length of the
resulting line is given by:

√(32 + 42) = √(9+ 16) = 5.

However j can be more useful than just an indicator of quantities which cannot be mixed. It can
be used as an Operator which is a mathematical tool such as + or - . Its use as an Operator is
derived as follows:

Fig. 2 shows the two axes complete with both negative and positive values. Along the x
(horizontal) axis positive numbers are represented as line-lengths measured to the right starting
from the Origin. Similarly negative numbers are represented as line-lengths measured to the left
starting from the Origin.

Under the rules of arithmetic:

(a) a positive number becomes a negative number when it is multiplied by -1

(b) a negative number becomes a positive number when it is multiplied by -1

This can be represented on Fig.2 by regarding a multiplication by -1 as a 1800 rotation about the
Origin "0". A second such rotation (another multiplication by -1) returns the plotted point
through a total of 360o to its original position.

Now consider the imaginary axis on which quantities jy are plotted. The same argument must
be applied of course and so multiplication of imaginary numbers by -1 must also be regarded as
a rotation of 180o..

Now that Fig.2 has been constructed under these rules a small query arises:

a quantity (for example 5) plotted on the horizontal axis is moved to the vertical axis when a j is
placed in front so that it becomes j5. By direct analogy, if multiplying by -1 rotates +5 through
180o to position -5 then does not placing a j before it amount to multiplying that quantity by j so
causing a rotation through 900?
This isn't really a fiddle; the diagram was constructed under given Rules and, if the diagram is
to be valid, the Rules must always be observed. From the Rules the idea arises naturally that an
imaginary quantity is to be regarded as a real quantity multiplied by the Operator j. Similarly a
real quantity has to be regarded as an imaginary quantity multiplied by the Operator j.

Now there is one last step. Multiplication twice by the Operator j (two rotations through 900 )
clearly leads to a rotation of 1800 already defined as multiplication by -1. In other words

j2 = -1

which leads to the embarrassing statement that

j = √(-1)

The Rules of mathematics say that minus x minus equals plus and also that plus x plus equals
plus and so there cannot be any number which, multiplied by itself, equals -1?

You are entitled to deduce from this, if only because it is true, that j cannot have a numerical
value. However it does not necessarily follow that the result is useless.

7. 7. 3 Use of j as an Operator

As an Operator the symbol j changes a real quantity into an imaginary one or an Imaginary
quantity into a real one. That, under these Rules, j represents the square-root of -1 is useful when
solving problems in Algebra.

Consider for example an inductance in the primary circuit of a transformer. Measured in the
secondary winding this inductance appears as a capacitance because of the phase-shift that takes
place as a signal passes between the windings. (This effect is utilised to provide frequency-
correction in bandpass coupled-circuits (see under 3.13.4). In a theoretical analysis of the
transformer the reactance (a j-term) in the primary becomes multiplied by a second j-term
which represents the phase shift between windings; the resulting term contains j2 ( = -1) which
indicates that, in passing between the transformer windings, an inductive reactance is
transformed into a capacitive reactance (true); similarly a capacitive reactance is transformed
into an inductive reactance.

During the course of algebraic manipulations the principle is taken further and the unwary can
fall into some nasty traps. Thus

j3 = j x j2 = j x -1 = -j

and

j4 = j2 x j2 = -l x -1 = 1

NOTE These values for j do not have any arithmetical meaning; they are used only in evaluating
algebraic results in terms of Fig.2.

In Fundamentals-1, Lesson-8 it was simply stated that the impedance (Z) of a circuit
represented by (R + jX) is found by squaring the two quantities R & X, adding the results and
then finding the square-root of that sum ; in symbols
 Z = √(R2 + X2)

That formula also expresses the result of Pythagoras' Theorem which relates the sides of any
right-angled triangle. The use of j to describe two quantities represented as right-angled vectors
is based therefore on the solution of Pythagoras' Theorem.

This derivation of Z is only valid of course provided that R & X are truly at right-angles; i.e.
that the values refer to "pure" or theoretically perfect components in which the current and
voltage waveforms are exactly phase-shifted by 900. This does not raise a problem of course
because practical components are regarded as being pure components degraded by a resistor
(representing losses) connected either in series or in parallel with that component.

The assumption is always true because practical values of L, C & R can be obtained only by
making measurements and, in making these, it is assumed that the lack of purity is caused by a
(theoretical) resistor connected in one of these configurations.

Back to Top of Page

END OF LESSON
 PART 8: INFORMATION SHEET No. 1
CAPACITORS
8.1.1 GENERAL

As described in Lesson-2, a capacitor consists essentially of two conducting plates with an

electrical connection to each. The dielectric, which both insulates the plates and keeps them
apart, may be air or any of a range of substances which can both provide the necessary degree of
insulation and also enhance the performance of the plates as a capacitor.

Good insulation is a prime requirement so that charge cannot leak from one plate to another and,
of itself, this is not a difficult problem. However an applied voltage electrically stresses a
dielectric and its resulting structural response not only enhances the storage capacity of the
device but has an apparent effect on the insulating properties under ac conditions.

When an applied voltage alternates then the electrical stress on the dielectric alternates also and
so the material is taken through a continuing cycle of work; work implies a consumption (a
loss) of energy. When measurements are made on a capacitor any losses (the absorption of
energy) appear as a resistive component and so any equivalent circuit for a capacitor appears as
a slightly leaky device.

Clearly the actual loss over any period of time depends on the number of times the dielectric is
cycled and so these dielectric losses increase with increasing frequency. Thus the choice of
dielectric material is influenced by the use for which a capacitor is required and, in turn, this
influences the technique used in its construction.

Large values of capacitance require a large surface area, close spacing of the plates and/or an
insulating material with a very large dielectric-constant. The first of these poses problems in
handling and packaging, the second poses problems both in manufacture and in maintaining
good insulation while the third is a matter of materials research but has obvious limitations.

One other problem lies in the natural inductance which is inherent in any conducting material.
Apart from the obvious limitations in the connecting leads the flow of electrons into and out of
each plate means that electric currents flow in the plates themselves and so, the larger the plate
surface area, so the larger must be the inherent inductance. Such inductive effects are
inseparable from the required capacitive effects and so each capacitor is an LC circuit with a
natural resonance frequency. At frequencies well below this self-resonance frequency the
component does indeed behave like a capacitor; at frequencies well above this self-resonance
frequency the component behaves like an inductor; at frequencies around the self-resonance
frequency it may well present a low-value impedance but the phase-shift it may cause is a matter
for conjecture.

The effect of self-resonance is inescapable and the only defence is to reduce the capacitive value
as operating frequencies rise; this raises the self-resonance frequency. Fortunately capacitive-
reactance falls with rising frequency and so the ploy is effective. Thus, in an audio amplifier, the
decoupling capacitors are measured in µF; in a medium-wave receiver they are measured in nF;
in vhf receivers and above they are measured in pF.
It is not uncommon to see large-value capacitors shunted by small value capacitors. In an audio
amplifier for example the idea is that a 10 µF capacitor will deal with the audio signals but may
appear inductive to intruding or self-generated rf signals; these are then dealt with by the
smaller capacitor. The scheme can be made to work but the Designer needs to know what he is
doing. Above the self-resonance frequency of the larger capacitor that component appears
inductive; this inductance then forms a parallel-resonant (high-impedance) circuit with the
smaller-value capacitor. There is a band of frequencies therefore over which the two capacitors
in parallel are incapable of acting as a decoupling circuit - they form a band-stop filter.
Typically a parallel combination of a 220 µF and a 22 nF resonate at around 5-MHz.

8.1.2 METHODS OF CONSTRUCTION

[Link] Large-value Capacitors : Electrolytics

Very-large values of capacitance (1 to 10,000 µF or greater) with a range of working voltages up

to about 500-volts are obtained by reducing the plate separation to molecular proportions. Their
construction involves the use of electro-chemistry and they are known as electrolytic capacitors.
In general a fairly large plate is folded inside a metal can (which forms the other plate) and the
dielectric is a chemical mix. A direct voltage is applied between the plates and a very-thin
insulating layer is built up by electro-chemical action. The performance of the capacitor is
determined by the voltage used in its preparation and so it is essential that this voltage be
maintained in use.

These capacitors therefore are polarised and must always be connected “the right way round”.
Their insulation is poor and a direct current always passes between the plates. Their use is
mainly in decoupling and smoothing (power-supplier) circuits (see Part-5) although they also
find use in RC couplings in low-impedance circuits such as bipolar transistors.

An ac (non-polarised) electrolytic capacitor is also made which primarily is for series-operation

with ac electric motors.

If only because of their poor insulation electrolytic capacitors are not suitable for use in circuits
where stability is important (e.g. oscillators). This is the important point that is likely to arise in
the R.A.E.

An important factor for Designers to consider with electrolytic capacitors is known as the ESR-
value (equivalent series resistance) which, typically, may be around 1-ohm for a 220µF
capacitor. This theoretical resistor is determined by measurement and it represents the
dissipation of power when ripple current or any ac flows through the capacitor. The resulting
rise in the capacitor’s internal temperature accelerates the drying-up process caused by
evaporation and so determines the useful life of such a capacitor. Electrolytics should not be
mounted near sources of heat.

Often the ESR value may increase without any corresponding change in the value of capacitance
and so a simple check of capacitance value alone may not reveal that the capacitor has become
lossy.

[Link] Large-value Capacitors: Tantalum

These solid-Tantalum dielectric capacitors are polarised and provide capacitances in the range 1
to 500 µF with working voltages up to about 35-volts. They are intolerant of both reverse
voltages and voltage-surges.

[Link] Foil Capacitors

Reasonably large values of capacitance (around 0.01 to 2 µF) can be obtained by using plates
with large areas. These take the form of long strips of metal foil which are insulated by a
similarly-shaped strip of material such as waxed paper, polystyrene, polyester or polypropylene
film. Waxed-paper capacitors are seldom encountered today except in elderly equipment.

Except for the polystyrene types the plates seem to be formed by metallising both sides of the
dielectric strip. The entire assembly is then rolled or folded and packed into tubular or
rectangular cases and sometimes encapsulated. Each metal plate extends only to one edge of the
insulating strip so that disc-connections to either is made at each end of the rolled assembly. By
thus making connection throughout the length of the strip-plate the inductive effect is largely
eliminated. In the early waxed-paper types it was not uncommon for this connection to fail with
the result that capacitors lost part or most of their capacitance although it is not clear how much
of the change in performance was due to the resulting inductive effect.

Polypropylene capacitors offer low dielectric losses with the ability to withstand high
alternating voltages. They give excellent high-frequency performance and can withstand high-
voltage pulses with fast rise times. They also offer a negative temperature-coefficient of 200
parts per million per °C which means that their capacitance falls by 200/1,000,000 (1/5,000) for
each degree-centigrade that the ambient temperature rises.

Polyester capacitors are more lossy and provide lower insulation but their temperature
coefficient is around +200 p.p.m/ °C. (A correct combination of these two could yield an overall
capacitance that does not change appreciably with temperature).

Polycarbonate capacitors have a performance similar to that of polystyrene types but their
temperature coefficient is ± 60 p.p.m/t. These capacitors drift less with temperature but their
uncertain direction would not be too satisfactory in an oscillator circuit.

Polystyrene capacitors are less tolerant of temperatures greater than 70 °C but they have
excellent high-frequency performance, very good insulation properties and a negative
temperature coefficient between 70 and 200 p.p.m./°C.

[Link] Small-value Capacitors

Small-value capacitors are formed by techniques such as metallising opposite faces of a slab of
solid dielectric material or by layering multiple plates and dielectric; the whole is then
encapsulated. Very high quality - and expensive - capacitors are made using metal plates
separated by a thin plate of mica. Once again of course their properties depend on the material
used to form the dielectric:

Silvered mica capacitors offer very low losses, high insulation and excellent stability with a
positive temperature-coefficient between 30 and 75 p.p.m./°C. They are very useful in oscillator
circuits or any circuit where frequency stability is important.
Ceramic dielectrics offer very high dielectric coefficients (symbol K) which enables the
construction of very compact units. In general however they have low stabilities and should not
be used in oscillator circuits except those designated as Low-K. Low-K types are specifically
made for compensating temperature drifts in oscillator circuits and offer temperature-
coefficients which are selectable over the range 0 to -750 p.p.m./°C.

Medium-K types offer small size for use where small changes of capacitance with temperature
are not important and some losses can be tolerated.

High-K types offer small size with high relative capacitance but their performance in terms of
stability and losses leave much to be desired. They are useful in coupling and decoupling
circuits.

[Link] Feed-through Capacitors

These are specifically designed for decoupling purposes to be used where a feed line passes
through a screen or through the wall of a screening box or container. They have a coaxial-type
construction in which a central conductor is mounted in an outer tube in a ceramic cement. The
outer is flanged so that it can be pushed through a hole in the screen and then soldered into
place. The result is a useful mounting post which penetrates the screen and with capacitance
provided between the conductor and earth; by joining such devices with small chokes or
resistors it is possible to build very effective filter units to prevent signals either entering or
leaving compartments via the supply lines.

[Link] High-current Capacitors

These are low-loss capacitors which are used in transmitter P.A. circuits where large circulating
currents are encountered. They are usually of ceramic construction with fixed capacitance.
Clearly if an unsuitable unit were to be used in a tank circuit it would overheat and suffer
destruction.

8.1.3 VARIABLE CAPACITORS

Variable capacitors take a variety of forms which depend on the purpose for which they are
used.

[Link] Tuning Capacitors

These have a set of fixed accurately-spaced "vanes” which are insulated from the main frame by
which the unit is mounted — i.e. the "stator" usually carries the signal and is said to be “hot”. A
second set of equally-spaced vanes is mounted on a spindle so that, as the spindle is rotated, so
the “rotor” is moved into and out of mesh with the stator. The technique therefore is to vary the
effective area of the capacitor’s plates. The vanes of the rotor are shaped to achieve any required
law - i.e. the manner in which capacitance changes with the spindle rotation.

Many such units say be found mounted on the same spindle thus producing a “ganged”
capacitor. This is used to tune several circuits simultaneously.
The rotor is invariably electrically a part of the mounting of the unit so that it operates at earth
potential. This is mainly to avoid change-of-capacitance effects when a hand is placed on the
spindle but it is also a necessary safety precaution where the capacitor carries high voltages (dc
or rf). Often it is required that such a variable capacitance should be “hot” on both plates and so
that it becomes necessary to mount it on an insulating plinth — in this event it is important to
use an insulating shaft-extension.

[Link] Trimming Capacitors

These are used to equalise stray capacitances in ganged variable-tuned circuits thus ensuring
that, as the tuning control is rotated, so the several circuits vary their tuning in step. Trimmers
are used also in circuits such as bandpass-coupled pairs (or triples) to adjust each part of the
circuit to resonance. The requirement here is that the capacitance value should be adjustable but,
once set, it will not be further adjusted in normal use.

The construction of trimmers varies according to frequency-range, use and cost. Many are but
miniature forms of the tuning capacitor described above except that the spindle is cut short and a
screwdriver slot provided for adjustment. Others, known as compression trimmers, consist of
two small plates separated by a thin sheet of mica; the upper plate acts also as a spring and it is
pressed down by a small screw which penetrates the assembly to screw into the main frame (i.e.
it is the earthy connection).

There are other types also but, in general, they all employ the techniques of either variable-
spacing or variable area.

[Link] Tuning Diodes

Not truly a capacitor but not to be forgotten is the variable-capacitance diode which permits the
capacitance in a circuit to be adjusted electrically by varying the direct-voltage applied in
reverse to a semiconductor junction-diode.

8.1.4 SUPRESSION CAPACITOR

These are manufactured specifically for use in reducing radiated and conducted noise signals in
mains circuits. They come in two Classes X2 and Y and it is important that they are not wrongly
used.

[Link] Class X2

These capacitors are for use in mains-suppression circuits where their failure would not cause
electric-shock hazard to a user; e.g. connected directly across the mains supply (live and
neutral). They seem to come in two main types namely:

(a) metallised polyester film and encapsulated in flame-retardant plastic or

(b) epoxy-resin impregnated paper.

Both have self-healing properties.

[Link] Class Y

While these do much the same job they are intended for use where their failure could cause
electric-shock hazard; e.g. connected between the live and earth wires of the mains supply. They
use either metallised polypropylene film or epoxy-impregnated paper and are encapsulated as
before in flame-retardant cases. They are manufactured to a much higher specification with
greater resistance to the ingress of moisture and cost up to twice as much as Class X2 types.

[Link] 3-terminal Y and Delta Capacitors

Three-terminal Class Y “capacitors” are intended for RF filtering and consist of a capacitor to
ground (via one terminal) with small inductors each side in the through connections; these take
the form of small ferrite beads. They are in fact not truly capacitors but T filter networks.

The three-terminal Delta Class X2 “capacitors” consist of three capacitors connected one
between each of the three terminals and are intended for use in suppressing small motors; i.e.
they provide a capacitor between the mains-supply leads and one from each lead to earth. This
would appear to contravene X2 regulations?

There are several other types available including 3-terminal ferrite-bead capacitors, which
provide greater noise attenuation when compared with standard 2-terminal capacitors; motor-
vehicle suppressor capacitors; varistor capacitors which, in 3-terminal form, remove fast surges;
capacitors and capacitor-networks specifically designed as Contact Suppressors.

8.1.5 TEMPERATURE COMPENSATION

All the above capacitors are likely to change their value with temperature changes and, as
already mentioned, some increase their capacitance as temperature rises (a positive coefficient)
while others decrease their capacitance as temperature rises (a negative coefficient). By suitably
mixing the two types it is possible to obtain an overall capacitance which remains substantially-
constant. Temperature drift is not normally a problem in, for example, the rf stages of a receiver
but its elimination in a Master Oscillator is important especially where tuning diodes are
incorporated.

Apart from the correct choice and mixing of capacitor types negative-feedback is useful in those
circuit arrangements known as automatic frequency-control (a.f.c.); however frequency-
stability then becomes important in the error-sensing circuit which develops the a.f.c. signal.

Back to Top of Page

END OF INFORMATION SHEET 1

 PART 8: INFORMATION SHEET No. 2
CELLS
Please Note: This sheet was written in 1994/95 and needs to be updated.

8.2.1 INTERNAL AND EXTERNAL STORAGE

Cells are chemo-electric generators which means that they convert chemical energy to electrical
energy. In general the interaction between different combinations of chemical substances results
in the transfer of electrons from one terminal of the cell to the other. As described under 1.3.1:
The Chemical Cell the chemical action continues until the charge accumulated at the terminals
produces sufficient counter-force to prevent further transfer of electrons.

Cells can be classified in several ways but the first is into the two categories storage
cells and fuel cells. A storage cell stores all the chemical constituents within the cell and, when
those chemicals are exhausted, the cell ceases to function. The chemicals are not necessarily
consumed purely in producing electricity; unwanted side reactions cause the cells to have a
limited shelf-life and some of them suffer also a destructive corrosion when they are left for
long periods. A fuel cell uses chemicals which are fed to it as required from external containers.
There are of course advantages and disadvantages to each type:

(a) Storage cells can be made in very small sizes and are fundamental to portable equipment but
their limited storage means that they have a limited useful life and limited output current.
Storage cells can be made in large sizes also when they can offer either a large output current or
long duration.

(b) Fuel cells, while themselves not necessarily large, can be accompanied by storage containers
whose size is chosen according to need and which can be placed as convenient; the size of these
containers determines the life of the cells and, together with the cell construction, the current-
capacity. This type of generator is used in small space-craft in preference to nuclear reactors
because, apart from cost, they do not pose a radiation hazard in the event of an accident.

(c) Different chemicals used in a fuel cell are housed separately and this prevents inter-action
between them until they enter the cell; this prevents the chemical side-reactions that so often
limit shelf-life of storage cells and cause their destruction.

(d) In general the chemicals used in fuel cells, and the products of their reaction, need to be
volatile so that the chemicals can be fed to the cell under proper control and so that the products
of the reaction can be removed. Often the usefulness of storage cells is limited more by the
failure to remove by-products of chemical reaction than by exhaustion of the basic chemicals.

8.2.2 PRIMARY CELLS AND SECONDARY CELLS

Storage cells are sub-divided into two main categories:

(i) Primary Cells
(ii) Secondary Cells

[Link] Primary Cells

Primary cells are those whose useful life ends when they have consumed all the chemicals
stored within themselves. In theory all chemical reactions can be operated in reverse if the
correct conditions are created and none of the by-products of the reaction are lost; sometimes
however the conditions required produce negative returns and the only practical action once a
primary cell fails is to throw it away.

[Link] Secondary Cells

These cells utilise chemical reactions that can be reversed under reasonable conditions. Once the
cell becomes exhausted an electric current is passed through it in the “wrong” direction to force
the chemical action to reverse and so restore the chemica-mix to the original starting condition.
This charge-discharge cycle can be repeated many times but ultimately the cell becomes useless
either because of mechanical failure in its construction or because abuses of the cell cause a
change in the chemical mix.

8.2.3 DRY CELLS AND WET CELLS

This sub-division is partly a misnomer in that few, if any, cells will function in the dry state;
indeed car batteries are usually sold in the dry condition because it improves their storage life
and because there is less hazard from the acid contents.

If atoms and/or their electrons are to move freely about a cell to accomplish the necessary
transfer of electrons from the positive terminal to the negative terminal then the various
chemicals must be in intimate contact in a medium that permits movement; in other words they
must be in solution.

Most of those designated as wet cells usually use either hydrochloric or sulphuric acid but some
use Potassium Hydroxide; all these are destructive and dangerous when spilled and most
certainly could not be used in portable equipment such as a torch or a radio-receiver.

So-called dry-cell construction prevents spillage by use of a thickening or gelling agent;

however, the decreased mobility of the chemical atoms which results limits the current-
generating capacity of a cell and also brings into being a problem called polarisation.

A product of the chemical reactions that take place in many primary cells is a gas and the escape
of this explains why the reaction cannot be put into reverse; it also explains why such cells
cannot be hermetically sealed to prevent [Link] seepage. With a liquid electrolyte this gas
simply forms larger and larger bubbles until their buoyancy causes them to break away from the
electrode where they had formed, so producing an insulating layer, and to rise to the surface.

When the electrolyte is either a paste or a gel however such escape is denied and the bubbles
remain around the electrode eventually to form an insulating layer around it. This layer has the
same effect as removing the electrode and so the cell ceases to function even though it has
plenty of potential life remaining.

Polarisation is prevented by surrounding the electrode with a porous mass of material that
absorbs the gas; the same technique is used in gas masks. Not surprisingly this material is
referred to as the depolariser. The depolariser offers resistance to electron-transfer within the
cell and so reduces the current- generating capacity. Furthermore depolarisers need time in
which to accomplish their appointed task and so a cell, which fails in use, will often recover
after a period of rest and this process is sometimes accelerated by gentle heating.

WARNING — many modern cells are hermetically-sealed and it can be dangerous to heat
them.

The phenomenon of polarisation and the presence of a depolariser in dry cells leads to a
fundamental difference in the manner of checking them. The high degree of electron-mobility in
wet cells enables them to generate very large currents (hundreds of amps) and this poses an
obvious hazard. In dry cells the depolariser and its relatively slow action means that only small
currents can be produced; for example (when new) the largest cells can produce up to 7 amps
while a torch battery may limit at around 1 to 3 amps.

With reference to the previous Section wet cells have a low internal-resistance, about 0.02 ohms
or less, while dry cells have internal-resistances between 0.2 and 1 ohm; as the depolariser and
the chemicals become exhausted so this internal resistance rises.

Many dry cells have a terminal voltage which decreases steadily throughout the cell’s life; they
are said to have a sloping discharge-curve. By contrast most wet cells have a fairly constant
terminal voltage until the point of exhaustion when they fail very suddenly; these are said to
have a flat discharge curve

To check a dry cell by measuring only the terminal voltage tells little about the internal state of
the chemical mix or the depolariser both of which are crucial to the cell’s performance. The only
criterion is the ability of the cell to produce current and so it is necessary to measure the short-
circuit current by connecting an ammeter directly across the terminals. WARNING This test
must be conducted as a “flash-test”; if the meter is connected for longer than about a second the
cell IS useless. Do not perform such a test unless it is necessary.

By contrast a wet cell MUST NOT be subjected to a short-circuit; a device called a drop-tester
exists whose purpose is to test the terminal voltage under short-circuit conditions. (Given that
the resistance of the “short-circuit” is known the short-circuit current can be calculated.) Such a
test is useful only on a wet cell which is believed to be at the end of its life; in the hands of an
inexperienced operator this can be a very dangerous pastime as abuse may cause the electrolyte
to boil. The proper way to test wet cells is to understand their operation and check the terminal-
voltage (both on and off charge) and the specific gravity of the electrolyte; see later.

Many modern cells do not give access to the electrolyte and so an intelligent reading of terminal
voltage is the only criterion for judging their health.

8.2.4 CELL DETAILS

2.4.1 Laboratory Cells

There are various designs of cells which provide constant terminal voltages and they were used
as laboratory standard-references. These cells are not intended to supply current and are used in
circuit arrangements where their emf is balanced against a p.d. Today they are largely
supplanted by semiconductor reference devices.

An example of such a cell is the Weston cell which delivers 1.0183 volts at 20°C for an almost
indefinite period. It uses Cadmium and Mercury for its electrodes and Cadmium Sulphate
solution as the electrolyte. It has a temperature coefficient of 40 parts per million per degree C
(40 ppm/ °C) but modern semiconductor references can produce temperature stabilities around 1
ppm/ °C.

[Link] The LeClanché Cell

This cell is the basis of the original dry cell which is now being displaced by the Alkaline cell
and various forms of sealed and re-chargeable cells. It consisted of a glass vessel which
contained ammonium chloride solution into which was dipped a zinc electrode and a carbon rod;
the zinc formed the negative terminal and the carbon rod the positive terminal. A depolariser
was used which consisted of manganese dioxide contained around the carbon rod by a porous
pot. The chemical reaction consumed the zinc.

To make this into a dry cell the zinc is formed as the container inside which is placed a carbon
rod packed around with manganese dioxide; the electrolyte is mixed as a paste instead of being
in solution and it is packed between the zinc container and the manganese dioxide.

The problem with this type of cell as it becomes exhausted, either by use or through over-long
storage, is that the zinc case becomes perforated and the contents leak.

The off-load terminal voltage (the emf) of the LeClanché cell is about 1.4 volts. It is the
cheapest of the available cells but also has the lowest energy-storage density, it does not produce
high currents and its life is greatly diminished if discharged at a too high rate. The sloping
discharge curve indicates that the internal resistance rises as the cell is discharged. It does not
perform at low temperatures.

The "zinc-chloride" cell is an improved version of the zinc-carbon cell which has a slightly
better energy-storage capacity and a better performance at high currents and low temperatures. It
does not leak and its reduced likelihood of sparking during installation makes it suitable for use
in some hazardous-area applications.

[Link] Alkaline Cells

Alkaline-manganese cells (known generally as “alkaline cells”) are of sealed construction and
any attempt to recharge them is likely to end in an explosion. It is unwise to tamper with these
cells because they contain highly toxic substances.

They offer a much longer service life than Leclanché types particularly in equipments which
draw large transient currents. However their useful life increases still further at low current-
drains. They have a long shelf-life because of their low self-discharge current. Although they
have a sloping discharge curve they offer a better performance than LeClanché cells both at high
current and low temperature.

The on-load voltage of a fully-charged alkaline cell is 1.5 volts which drops to 0.9 volts when
discharged.

The (wet) secondary Nife cell is little used today but saw much service around the time of
WWII. Its electrodes are Nickel (Ni) and Iron (Fe) and the electrolyte is Potassium Hydroxide .
It had advantages in that it was mechanically very robust, could produce a large current, could
be completely discharged and could be subjected to a high charge rate all without damage and
had negligible self-discharge; its disadvantages were a low e.m.f., higher internal resistance
than the lead-acid cell, a larger size for the same overall voltage and generally greater cost.

[Link] Silver and Mercury Cells

These are sealed primary cells. The silver-oxide types provide an e.m.f. of 1.5 volts and the
mercuric-oxide types an e.m.f. of 1.35 volts. Both offer 3 to 4 times the energy storage of a
LeClanché cell, they have flat discharge curves and excellent shelf life. Both types are good at
high temperatures and the silver cell operates down to -20 °C. They are both relatively
expensive.

These cells should NOT be re-charged, tampered with or disposed of in a fire.

[Link] Lithium Cells

These are hermetically-sealed primary cells intended generally as a back-up source for low-
current circuits and for devices such as clocks. Their shelf-life is typically around 10 to 20 years
and they are suited to low-discharge rate applications where they give a level voltage
characteristic. They may outlive the equipment into which they are incorporated.

There are several different constructions for Lithium cells, in terms of chemistry, and the
information at present in my possession is a bit confused. On-load terminal voltages are between
3.3V and 3.7V when new but they appear to drop to around 2.0 volts at the end of the cell’s life.
It is important not to mix them up with similarly-dimensioned cells of other types. Many have
been designed for specific applications and it is recommended that they are used only for those
purposes.

They have the highest energy storage per unit weight, a flat discharge curve and operate in
general from -55°C to 120°C depending on type.

The cells must NOT be subjected to either charging or forced discharging and no more than two
should be connected in series. They should NOT be disassembled.

[Link] Lead-acid Cells

These wet secondary cells are widely used as batteries mainly in road-transport applications but
they find application too where large-capacity storage is required with a high-current capability
such as the electric car or float.
Their bulk and weight are a great disadvantage for use in operating portable equipment. For
such usage the hazard from their sulphuric-acid electrolyte can be mitigated by gelling the
electrolyte (which raises the internal resistance) or by holding it in sheets of fibre-glass wadding
interleaved with the plates. Valves are sealed into the battery-cover to allow venting of any
gases which may be formed by over-charging.

The electrodes are formed as two different oxides of lead and these, in paste form, are packed
into recesses formed in the surfaces of lead plates. Positive and negative plates are interleaved
(with suitable spacers) so that both sides of the plates are utilised. Too-high
charging/discharging rates can result in the formation of hot spots or pockets of gas which
dislodge the oxide packings and so destroy the storage capacity of the plates. Additionally the
dislodged oxides fall to the bottom of each cell where, as the debris builds up, it eventually
short-circuits the cell.

When a lead-acid cell is fully discharged both sets of plates tend to a dirty-grey colour but, when
fully charged, one set assumes a chocolate-brown colour and the other a sharp grey. This is a
very useful guide to the state of a cell but, unfortunately, many manufacturers now hide them
from view with a device meant to make it easier to top-up the electrolyte.

The electrolyte is diluted sulphuric acid which is a very nasty substance. If spilled it can in
theory be neutralised with an alkaline substance such as washing soda but in practice it seems
able to go on causing corrosion despite all efforts to clean-up. These cells/batteries should
always be handled with care.

The degree of dilution is specified in terms of the specific gravity (S.G.) of the solution; in loose
terms this compares the weight of a given quantity with the weight of an equal quantity of water.
For a fully-charged lead-acid cell the S.G. (as measured with a Hydrometer) should be 1.280 but
generally quoted as twelve-eighty. When the cell is discharged its S.G. is around 1.100. Thus the
hydrometer provides a very good check on the condition of a cell.

The e.m.f. of a lead-acid cell is nominally 2-volts with an internal resistance that can be as low
as 0.02 ohms. (Note that a short-circuit will evoke a current in excess of 100-amps — but that
short-circuit is unlikely to last for long!) When fully-charged and still on charge the terminal
voltage may rise above 2.4 volts per cell; when left for a while this will generally drop to around
2.2 volts. On load this rapidly falls to the nominal value of 2-volts which is maintained until the
battery-charge is exhausted when it falls rapidly (in seconds) toward zero. The end of its useful
life is when the terminal voltage falls to around 1.7 volts. However the e.m.f. depends on the
specific-gravity of the electrolyte as shown in Table I below.

During the re-charging process the terminal voltage rises fairly quickly to just above 2.2 volts
but, as the fully-charged state is reached, this quickly increases to the value given above. Once
the oxide coatings of the plates have been restored to their original states any further passage of
current is supported by electrolysing the water content of the electrolyte; i.e. the water is broken
down into its constituent atoms namely Oxygen and Hydrogen which are lost to the
environment.

The escaping bubbles of gas carry with them fine globules of acid, which accounts for the
distinctive smell of a charging battery, but the acid causes damage to any surrounding materials.
The two gases are present in the correct proportions for them to chemically recombine to form
water for which process all they need is heat. The recombination process however releases a
great deal of energy and it is not unknown for unventilated battery rooms to be destroyed in an
explosion following either a carelessly drawn arc or the entrance of a smoker.

TABLE 1
(showing the Specific Gravity of the sulphuric-acid electrolyte at various
temperatures)

Fully-charged S.G. Temp F Temp C

1268 100 38
1272 90 32
1276 80 27
1280 70 21
1284 60 16
1288 50 10
1292 40 4
1296 30 -1.5

Fully-discharged S.G. Temp F Temp C

1098 100 38
1102 90 32
1106 80 27
1110 70 21
1114 60 16
1118 50 10
1122 40 4
1126 30 -1.5

Over-charging represents a loss of water from the electrolyte whose concentration is thus
increased (its S.G. is increased) and this results in an increase in the cell’s e.m.f. It follows that
the battery-charger becomes less successful in forcing current back through the cell and so,
because it is unable to achieve a full charge, the cell may appear to have lost some of its storage
capacity. The obvious cure is to replace the water (a process known as topping-up) but acid nay
be lost also during gassing and so constant topping-up leads to a steady fall in the S.G.

Clearly overcharging is not desirable but inevitable and so occasionally a battery may need to be
topped up with acid so as to restore both the S.G. and the terminal voltage to the correct value
when fully-charged.

Lead-acid cells should be charged from a constant-voltage source so that, as the terminal voltage
reaches its fully-charged value, so charging automatically ceases. However, where many cells
are charged in series, constant-current charging may be effective in ensuring an even distribution
of charge. A constant-current source however does not provide an automatic end to charging.
(More information on this subject is available on request).
The life of a lead-acid cell is limited by the break-up of the plates which both reduces the
available plate area (the storage capacity of the cell) and produces debris which may short-
circuit the unit. This process is accelerated by high charging and discharging rates. The cells
usefulness is also terminated by the formation of a layer of sulphur which insulates the plates
from the electrolyte; this is most often caused by allowing the cell to stand for a lengthy period
with the electrolyte in place when it will eventually reach a discharged state, For storage
purposes lead/acid cells should be fully charged, emptied, given a distilled-water charge and
then stored empty and dry.

Little can be done about a cell which has suffered break-up but it is sometimes possible to
recover a large part of a cell’s capacity which has been lost through sulphation by use of a
chemical process. Briefly this involves cleaning out the cell, washing and then charging it with
distilled water and then charging it with a solution of Glaubers Salt. There is a problem however
in that there is very little room inside a lead-acid cell and it may be difficult to remove the
debris. (Further information on request.)

Lead-acid cells and batteries are available today which really are sealed and safe to use within
equipment. It is claimed that they are good for 250-1,000 charge/discharge cycles but this does
depend on the use/abuse to which they are subjected.

[Link] Nicad Cells

These secondary (re-chargeable) cells derive their name from the Nickel and Cadmium metals
used in their construction. They are sealed and may be operated in any position but it is
important not to exceed the charge-rates recommended by the manufacturer. They are fitted with
a safety vent to release gases formed during overcharging.

Their nominal on-load voltage is 1.25 volts and they should be regarded as fully discharged
when this voltage reaches 1.0 volt. They have internal resistances measured in milliohms and
can produced discharge currents up to 60 amps depending on size. On charge they produce
terminal voltages between 1.4 and 1.5 volts.

Their discharge time is generally less (about one-half) than that of comparable alkaline types but
of course they offer a medium-term saving in operating costs.

There is a belief that these cells should not be recharged unless first fully discharged because
they develop a memory such that their capacity falls to the low value of the partial discharge. I
cannot find any evidence to corroborate this belief but, if the cells are abused (e.g. by
overcharging), internal chemical changes raise the terminal voltage and so preclude proper
charging. Many circuits have appeared for so-called cyclic chargers which, if a cell is left
permanently attached, constantly discharge and then re-charge the cell thus supposedly keeping
it in good condition.

It is worth noting that, unlike lead-acid cells, these cells can be stored for periods in excess of
five years without the need for periodic re-charging. Furthermore manufacturers give the cell-
life in terms of the number of charge/discharge cycles. It follows that a cyclic charger, far from
preserving a nicad, may rapidly reduce a new and unused cell to useless junk. Note too that a
cyclic charger may well deliver a completely-discharged cell if called upon at short notice.
Nicad cells should be charged from a constant-current source that, as a rule-of-thumb, produces
a complete charge in about 16 hours. Charge-rates are usually expressed in terms of a quantity C
which is the storage capacity of the cell expressed in ampere-hours. Thus a charging-rate of
C/10 means that the battery would take 10 hours to reach a full charge (from the fully-
discharged state). For a 1 A-hr cell this would mean charging at 100 mA; however, at this rate, a
discharged cell should be charged for 14 to 16 hours.

Once a nicad cell has become fully-charged it is essential to reduce the charging current to a
maximum value about one-tenth the ampere-hour rating (C/1O) else a cell will be permanently
damaged. The fully-charged state is indicated by the terminal voltage reaching a maximum
value but to accurately determine this point is a complex operation that involves monitoring the
voltage level, the rate at which that voltage changes, cell temperature and/or charging time.

(Further information on request).

[Link] Nickel Metal-hydride (NIMH) Cells

These are a recent development that offer the advantages over Nicad cells of 50% greater energy
storage and that they do not use the toxic Cadmium metal. They offer too the disadvantages of
higher cost, need strict control of the charging process to avoid explosion risks of over-charging
and that their life is impaired by prolonged trickle-charging.

[Link] Solid-state Systems (SSS) Cells

These lithium-based cells are not yet available at the time of writing but are expected to come
into production later in 1995. The many advantages of the Lithium type are offset by the
difficulties in constructing a re-chargeable version that does not risk explosion. It is claimed that
an SSS cell offers around twice the storage capacity of the standard Nicad cell and nearly three-
times the terminal voltage (around 3.5 volts).

The anode is formed of a carbonaceous material and the cathode of lithium-manganese-oxide.

The conventional liquid electrolyte is replaced by a conductive plastic membrane which
contains a lithium salt although this is not quite so conductive as would be a liquid electrolyte.

The cell may be charged either with a constant-current or a constant-voltage charger but a strict
upper voltage-limit must be observed (IEE Review May 1995).

Back to Top of Page

END OF INFORMATION SHEET 2

 PART 8: INFORMATION SHEET No. 3
RMS VALUE FOR A SINEWAVE
8.3.1 DC and AC MEASUREMENTS

Measurements on dc circuits are usually carried out using either digital or moving-coil meters.
With these it is a simple matter to measure the constant-value of the voltage and also the
constant-value of the current; by multiplying these two figures the constant-value for the
dissipated power can be ascertained. Equally, by dividing the current into the voltage the
constant-value of the circuit resistance is obtained.

Measurements on ac circuits however are not so simple because, even though the circuit
resistance remains constant, should the circuit contain reactances then everything else changes
with frequency. Furthermore the current variations do not keep in step with the voltage
variations (there is a phase discrepancy) and even that varies with frequency.

In terms of meters there is an immediate difficulty. Digital meters are of little use to measure
varying parameters because of their plodding action. Analogue meters suffer from mechanical
inertia which prevents them from following a waveform which varies at a rate much above 1-
Hz.

To be of any use at all a m-c meter must have a centre-zero so that it can show both positive and
negative values. As frequency is increased the meter does not have sufficient time to reach the
peak value of the waveform before the value begins to decrease once more; it follows that with
increasing frequency the deflection of a m-c meter is reduced until, above about 10-Hz, it simply
remains stationary at the average value. The one thing which is obvious about the symmetrical
sinewave is that its average value is zero.

(Note that, in a Class-A amplifier, a meter always records the value of the dc in an anode or
collector circuit irrespective of the (ac) signal because the average value of the varying direct-
current is that of the dc on which the ac is superimposed. Indeed this fact can be useful in that a
kicking m-c meter indicates the presence of distortion due to overloading an amplifier stage.)

Given that the circuit is in a "steady-state" there is one parameter which remains constant and
that is the power which is being dissipated. This provides a basis for measuring voltage and
current in an ac circuit. For example, if a lamp is set up in front of a photometer it is possible to
measure the brightness of the lamp ; in turn this is a measure of the temperature at which the
filament is running and this, of course, is a function of the power which it is dissipating.

First feed the lamp with the ac-source which it is desired to measure and then measure the
brightness of the lamp. Now remove the ac-source and substitute a variable dc-source. Increase
power until the photometer indicates the same brightness and then note the values of direct-
current and direct-voltage. The product of these two gives the value of the ac-power.

Not only is the above experiment a legitimate way of measuring an ac but it. is a method that is
actually used where appropriate and which is the basis of many measuring techniques. Instead of
using a photometer and lamp it is possible to measure filament temperature by noting the
expansion of the filament wire this is the basis of the Hot-wire Meter which is described in Part-
6. The Thermocouple Meter measures temperature by the p.d. which is developed across a non-
uniform junction.

However, to measure power by such means is not very convenient and it is desirable to have an
ac-voltmeter and an ac-ammeter which will give indications such that the direct multiplication
of their readings yields power directly. Such meters give a value for an ac-waveform which is
equivalent, in its heating. effect, to a dc-waveform.

This value is derived theoretically by a mathematical process which results in the name The
Root-mean-square value which is always shortened to the RMS Value. However, it is important
to remember that RMS values apply only to pure sine-waveforms from which the definition has
been developed; to measure any other waveform it is necessary to use a power-averaging meter
such as the hot-wire or thermocouple meters mentioned above.

8.3.2 DERIVATION OF an RMS VALUE

As already described the average value for a sinewave is zero ; what we are seeking here is the
average heating effect as the waveform rises from zero-to a peak value and then falls again to
zero. It is only necessary to consider one half-cycle because, although the voltage and current
reverse, the heating effect is the same in the following half-cycle. (Do not confuse this with the
reactive circuit in which the current and voltage waveforms are displaced in time and the
average power is zero.)

The mathematical study of quantities that vary continuously is known as The Calculus and its
use gives a very quick and simple result for this problem.. However it is possible to derive the
RMS value for a sinewave without resorting to Calculus.

Fig.1 shows the plot of a sinewave and it could equally represent either the voltage or the
current waveform. The power in the circuit is proportional to the square of either parameter and
so, at any particular moment, power is proportional to the square of the value indicated by the
sinewave.

In Fig.1 the plot has been divided into vertical strips which divide the sinewave into small
sections. The argument is that, if these strips are very narrow, then each segment of the
sinewave remains at a substantially constant value and the use of Calculus is not necessary.

The idea is to find the power dissipated during the time of each strip, add all the powers together
to obtain the total amount of power dissipated over the half-sinewave and then divide by the
total time so as to get the average power throughout the half-sinewave. The power dissipated in
each strip (each time-segment) is proportional to the square of the current (voltage) value for
that strip. Once the average effective power is known then, by taking the square-root, a value is
found for a current (voltage) that would dissipate the same overall amount of power.

The method therefore is to:

(i) square the instantaneous values to get a figure proportional to the instantaneous power in
each strip
(ii) add all the powers and divide by the number of strips to get an average value for each strip

(iii) take the square root of this average-power figure to obtain an effective value for the current
(or for the voltage).

In plain English we square it, find the average value (the mean value) of the squares and then
take the square-root which leads to the name "root of the mean of the squares" - the Root-mean-
square or RMS value.

To derive a method of defining the RMS value does not of itself solve the problem of measuring
that value. A hot-wire or a thermocouple meter can be calibrated in terms of RMS voltage or
RMS current if the impedance of the heating element is known. The meter is used to measure a
known direct-current power and then the value of the current (or voltage) is marked on the scale
— such a calibration is valid for a hot-wire meter but, for a thermocouple meter, it is only valid
for that particular combination of thermocouple and moving-coil meter used during the
calibration process. For either meter the calibration holds only at low frequencies because, at
high frequencies, stray capacitances tend to by-pass the heating element and so the meter
reading progressively falls below the correct value.

Most m-c meters calibrated in RMS values are in fact dc meters fitted with a rectifier. The time-
constant of the smoothing circuit is adjusted to make the meter register the peak value of the
waveform — this is the only part of the waveform which can be identified easily. The scale
however is marked with a set of figures that represent that peak-value divided by root-2 (1.414)
because it can be shown that the ratio between peak and RMS values is √2:1.

Rectifier circuits also suffer from high-frequency limitations and so, although such meters
provide a performance better than that of the thermocouple meter (and their response is quicker),
they still have a limited range. The range can be extended to high frequencies by employing a
more complex form of rectifier circuit which is fed by either an audio-amplifier (up to 20-kHz)
or by a radio receiver; the response of these devices can be tailored to correct any frequency-
related errors. However they all employ the technique of peak-waveform detection and
(fiddled) calibration ; some may be calibrated both in terms of peak and RMS values.

An advantage of this arrangement is that the scale can be extended as desired by inserting a
calibrated attenuator at the input. Yet again negative-feedback can be used to raise the input
impedance and also to make the meter response logarithmic so that the scale can be calibrated in
linear steps marked in decibels (see Part 7).
Such meter systems in the audio band are known as Amplifier-detectors (shortened to "Amp-
Det"). In old books or Articles you may find them referred to as VVM's (valve-voltmeters).
They are the basis of the "electronic multimeter". When the amplifier takes the form of a
tunable radio receiver and, together with an aerial and feeder, it is calibrated overall in a
standard radiated field, the result is a "field-strength meter".

Back to Top of Page

END OF INFORMATION SHEET 3

 PART 8: INFORMATION SHEET No. 4
FUSES
8.4.1 INTRODUCTION

Fuses are certainly the best known and, in concept, the simplest of the defensive tools in the
Engineering armoury. They are accepted generally as being extremely unreliable but few know
why and but seldom blame the real culprit. As a general rule it is best to be careful when things
seem simple and this rule certainly applies to fuses. They are designed as the "weak link in the
chain"; they are meant to interrupt an electrical circuit when the current increases to a dangerous
value and they provide the obvious place to look when seeking the break. Any "failure" of
modern fuses to do their assigned job has more to do with what happens after a fuse has
ruptured than with the actual fusing process itself.

8.4.2 VOLTAGE LIMITATIONS

As stated above fuses are used to protect circuits from the heating effect of excessive current; as
such they are connected in series with the circuit they protect and so why does voltage enter the
argument? The answer is that a fuse is a thermal switch; when the switch opens the emf of the
Source appears across its terminals and it matters very much whether that emf is 5 volts, 50
volts or 5,000 volts. Fuse manufacturers are faced with the kind of compromise which is met so
often in engineering. A fuse must have a small thermal capacity to ensure that it heats rapidly
and ruptures before damage is caused in the remainder of the circuit; i.e. it needs to be a fine-
gauge wire. However fine wires tend to have high resistance which causes a voltage drop;
current flowing through the fuse-wire with voltage developed across it spells power-dissipation
and so heat is generated thus encouraging a fuse to blow.

The real difficulty arises from the fact that the fuse-wire is required to melt; in turning to vapour
it (a) fills the space between the fuse terminals with an ionised and conducting media which
keeps the current flowing by means of an arc (b) when the arc is quenched and the vapour cools
it deposits copper over the surrounding area and so tends to re-establish a conducting path. A
fuse which interrupts a circuit driven by a 5-volt emf is not very likely to cause an arc but the
very same fuse used to interrupt a high-voltage circuit can be guaranteed to do so. Then again it
matters greatly whether the circuit that is interrupted is resistive in nature or reactive; an
inductive circuit even at 5 volts can generate enough voltage to set up an arc while a capacitive
circuit can supply a very-large current for a short time and so may sustain a high-current arc (if
one becomes established) long enough to allow damage to ensue.

As stated above the most important factor in fuse rating is what happens once the fuse has
melted. Most non-designer users will not be interested in the factors which determine the choice
of fuse for a given job of protection but, when replacing a ruptured fuse, it is important to
understand the markings on that fuse and to choose either an exact replacement or a suitable
substitute.
8.4.3 TIME FACTOR

The second most important factor in choosing a fuse is the speed with which it is required to
operate. Digital circuits, for example, are extremely intolerant of overloads and any protective
fuse must rupture very quickly if it is to be effective. At the other end of the scale there is need
for a fuse to “wait and see”; before it interrupts a circuit it has to decide whether there is a
genuine fault condition or an acceptable transient. Typical of this requirement is the so-called in-
rush current that occurs when equipment is first switched-on. On the output side of a mains
transformer there are large-value reservoir capacitors which virtually short-circuit the
transformer until they acquire their normal operating charge; trouble comes in large doses if the
Law of Maximum Perversity (Sod’s Law or, in the USA, Murphy’s Law) decrees that switch-on
takes place just as the Mains reaches the peak of a voltage cycle. (240 volts ac has a peak value
336 volts).

In-rush current is a reality and, before modern fuses became available, it was a menace. I
remember throwing the main switch in a television studio to be greeted by a heavy “crump”
followed by the task of locating and replacing fifteen sets of mains fuses! Doubtless some of the
fuse-failures during the following weeks had their origins in the same incident and in-rush
current probably accounts for many of the so-called “tired-fuse failures”. Equipment likely to
suffer in this manner is now fitted with time-delay or anti-surge fuses.

8.4.4 FUSE CHARACTERISTICS

Fuses are manufactured and tested under set standards laid down by the Underwriters
Laboratories Inc. (UL) in the USA, Canadian Standards Association (CSA) in Canada and the
International Electromechanical Commission (lEG) in Europe and Asia (there are others). All
fuses are batch tested and bear markings as follows:

[Link] Super Quick-acting (FF)

These are specified for protecting semiconductor devices or other equipment where the “quick-
acting” F-fuses are too slow.

[Link] Quick-acting (F)

These are intended for use in circuits which are not subject to surges or transients and which will
not produce high short-circuit currents. Some F-type fuses are provided with a filler that assists
in quenching arcs and these may provide low-cost protection for semiconductors. In American
terminology these fuses are known as ‘Normal Blo’ fuses.

[Link] Medium Time-lag Fuses (M)

M—type fuses will withstand the small transients or surges encountered in normal operation.

[Link] Time-lag (T)

These are intended for use in circuits subject to high in—rush currents. They are referred to as
anti—surge or delay fuses and, in American terminology, as 'Slo—Blo' fuses. They are
characterised by a built—in thermal delay.

[Link] Super Time-lag (TT)

These are similar to the T-fuses but they provide an even greater time delay before rupturing.

8.4.5 FUSE RATINGS

[Link] Current Rating

The current-value which appears on a fuse is the maximum current to which the fuse should be
loaded. The exact meaning of this depends somewhat on the Standards to which the fuse was
tested. Fuses tested to the IEC Standard should be selected so that the normal current equals the
fuse-rating; fuses tested to the UL Standard should be rated 25% greater than the normal circuit
current.

[Link] Voltage Rating

The rated voltage marked on a fuse refers to the maximum voltage at which that fuse can safely
clear a short-circuit. A fuse can be used at any voltage which does not exceed its rated voltage; it
should not be used in circuits where the emf is greater than the rated voltage.

[Link] Breaking Capacity

Otherwise known as short-circuit or interrupting rating this refers to the maximum short-circuit
current that the fuse can safely interrupt without risk of explosion.

8.4.6 DERATING FUSES

The obvious has been stated already in that fuses are thermal devices but what perhaps is little
realised is that their performance is affected by the temperature of their surroundings. The term
ambient temperature usually refers to the atmospheric or room temperature but fuses usually live
in a private world probably surrounded by a fuse-holder and close to heat-generating
components; their surrounding temperature is likely to exceed the ambient temperature. The
temperature assumed by any Body stabilises when the heat which it generates is exactly
balanced by the heat which it loses; the higher its surrounding temperature the higher becomes
that stabilised temperature. A fuse which is running hotter than the test-value must rupture more
readily.

Hence in determining the rating of a fuse its operating temperature must be taken into account;
if, when replacing a fuse, you are tempted into doubting a particular fuse-rating then take care.
A fuse runs hotter also as the normal operating current approaches the rated-current of the fuse;
this perhaps might explain why that rogue fuse persistently fails for no apparent reason? As an
example: Fuses are tested for their current-carrying ability at a temperature between 20C and
25C but a fuse rated at say 1.6 amps would have to be replaced by a 2-amp fuse if it is to
perform at 70C. Some fuses are manufactured with wire connections; in general it is best to
crimp or clamp these into place but, if soldered, it is most important to use a heat-sink
(sometimes mistakenly called a heat-shunt). The application of heat to any fuse is likely to re-
flow the internal solder and so change the characteristics of the fuse.

Top of Page

END OF INFORMATION SHEET 4

 PART 8: INFORMATION SHEET No. 5
THERMISTORS
8.5.1 GENERAL

Thermistors are variable-resistors whose value depends on temperature. There are two basic
types:

(a) Positive Temperature Coefficient (PTC) in which the resistance rises as the temperature is
increased
(b) Negative Temperature Coefficient (NTC) in which the resistance falls as the temperature is
increased.

8.5.2 THERMISTORS WITH A POSITIVE TEMPERATURE COEFFICIENT

PTC thermistors are characterised by low and relatively-constant resistance at low temperatures.
As the temperature is increased so resistance increases slowly at first but the rate-of-rise soon
increases. When resistance is plotted against temperature the resulting graph shows an abrupt
bend or knee which indicates that the thermistor rapidly switches between low resistance and
high resistance over a short range of temperature change. This temperature is known as
the Reference Threshold, Critical or Switching Temperature.

These devices are used mainly in temperature-sensing arrangements to protect against damage
from too-high temperatures or too-high currents (which will produce high temperatures). In
general they are connected in series with the circuit to be protected and, at the critical
temperature, the increase in their resistance-value limits the current. Their action is that of a
heat-operated switch.

They are manufactured in a variety of forms in which is specified their resistance at the critical
temperature. Also specified is their resistance a few degrees either side of the critical
temperature (showing the rate of change). Their cool/cold resistance and also their maximum
power dissipation.

8.5.3 THERMISTORS WITH A NEGATIVE TEMPERATURE COEFFICIENT

NTC thermistors are characterised by a resistance/temperature curve which is substantially

linear (a straight line) over a fairly wide range of temperatures. Different NTC thermistors are
matched in manufacture so that they provide overlapping linear ranges which cover
temperatures roughly from -80°C to +150°C. Sets of such are known as Curve-matched
thermistors.

NTC types are used mainly in temperature measurement circuits where their steadily-decreasing
resistance is equated to steadily-rising temperature. They can be used simply as thermometers
but more generally in a sensing arrangement that controls the heat-input to a system. Some
physically-small thermistors, able both to heat and to cool rapidly, are employed in
arrangements in which they provide the feedback signal for a gain-control loop in applications
such as oscillators.

8.5.4 APPLICATIONS

Fig. 1 shows a simple circuit in which a thermistor is used to construct a thermometer. The op-
amp, with its output tied to the inverting-input, is being used as a follower (compare with the
emitter-follower or cathode-follower circuits) and it acts as a buffer to isolate the sensing circuit
from the meter circuit which follows.

As the temperature of the NTC thermistor rises so its resistance falls and the non-inverting input
is carried toward the positive supply line. By definition the output terminal of the op-amp
follows the non-inverting input and so the positive-terminal of the meter is carried toward the
positive supply-line also, this increases the meter-deflection. The pot’meter VR1, connected
across the supply, is adjusted to bring both terminals of the meter to the same value at zero
temperature; i.e. it is the "Set Zero” control at 0°. The variable resistor VR2 is used to set the
full-scale deflection (f.s.d.) when the temperature is at the appropriate [Link] measure
negative temperatures there are two options:

1. Reverse the meter connections

2. Interchange the positions of the thermistor and resistor R1. This action almost certainly
would require a new-value resistor to be switched in place of R1.

Fig. 2 shows a circuit which

indicates that the temperature of
the thermistor is within a given
range. One of the light-emitting
diodes (LED) is illuminated when
the transistor draws too much
current and the other when the
transistor draws too little current

The variable resistor VR1 is used to set the “trip point” but there will be a range of temperatures
on either side of this point in which both LED’s will remain dark. To control this range attention
must be given to the tolerances involved in the characteristics of both the transistor and the
LED’s. A smaller range for the indicated trip-point can be achieved by using an op-amp in place
of the transistor.

Alternatively the LED’s may be replaced by a relay (or Triac) which can be used to turn-on the
supply to a heater when the temperature falls below the selected threshold or to turn-off that
supply when the temperature becomes too high.

8.5.5 MODIFICATION OF THERMISTOR CHARACTERISTICS

The I/V characteristics of thermistors can be modified by adding a resistor in parallel with the
thermistor. To increase the current-carrying capacity of a thermistor two or more devices may be
connected in parallel.

Back to Top of Page

END OF INFORMATION SHEET 5

 PART 8: INFORMATION SHEET No. 6
BANDPASS CIRCUITS
8.6.1 INTRODUCTION

The purpose and use of bandpass circuits was introduced in Part1, Lesson 10. This Information
Sheet explores the underlying theory of their operation, the various methods of their
implementation and details that will be useful to Constructors who would like to try designing
and building their own. Such a task is much easier if you have equipment such as a sweep-
oscillator and an oscilloscope but, as explained in the text, sophisticated equipment is not
necessary to produce reasonable results; a hand-sweep with a normal signal-generator can serve
well.

8.6.2 SIMPLE TUNED CIRCUIT

A single LC resonant-circuit produces a frequency-response of the kind shown in Fig.1 in which

the response of the arrangement is shown as a function of frequency.

For a so-called high-Q circuit the shape of this

curve is very sharp and narrow and, at the
peak-response frequency, it rises considerably
above the out-of-band response. For a so-
called low-Q circuit the curve is flat and broad
and fails to rise appreciably above the out-of-
band response.

The high-Q circuit is useful in those amplifier

arrangements where the response is required to
be essentially at a single-frequency; for
example in an oscillator. The low-Q circuit is
useful where the response is required to cover
a range of frequencies as may be found in a
radio receiver.

Fig.1 demonstrates the disadvantage of this simple arrangement in that, while the response of
the low-Q circuit may be broad-band, it offers but little discrimination between the required in-
band signals and undesired out-of-band signals; i.e. its use results in both poor amplification
and poor selectivity.

The behaviour of a resonant circuit as either high-Q or low-Q is determined by the amount of
energy that the resonant circuit dissipates. In technical terms by the losses or the damping that
is present. This energy dissipation takes place of course in the components which make up that
resonant circuit and so the circuit Q-factor is determined by the quality of the components from
which it is constructed; i.e. by the individual component Q-factors (see Lesson
8 in Fundamentals-1). To manufacture components of poor quality is not a reasonable option
because of considerations such as reliability and the need to maintain control over circuit
behaviour. Thus the bandwidth of an LC resonant circuit is adjusted by adding a known amount
of loss in the form of a damping resistor.
It is evident from the foregoing discussion that a broad-band response and high-gain are not
compatible and so an alternative to the simple resonant circuit is required.

8.6.3 COUPLED-CIRCUIT THEORY

Broadband responses are obtained by coupling together several simple LC-circuits. There are
several methods of achieving such coupling but perhaps the easiest to grasp is that which uses
the transformer principle.

Fig. 2 shows a basic transformer.

The exact mechanism whereby energy is

transferred from the primary circuit to the
secondary circuit is not known but it is known
that the amplitude of the secondary signal is
dependent on the rate at which the primary
current changes.

At the peaks of the primary-current waveform

the current is reversing sign either from
positive-going to negative-going or from
negative-going to positive-going and so, at
these points, the rate-of-change is zero.

Midway between the peaks where the waveform is crossing the zero line the rate-of-change
reaches a maximum value.

From this can be deduced that the peaks of the secondary waveform coincide with the zero-
crossings of the primary current and that the zero-crossings of the secondary waveform coincide
with the peaks of the primary current; in technical terms there is a 90° phase-difference
between primary and secondary waveforms.

This phenomenon can be interpreted by saying that the transfer characteristic (i.e. a comparison
of the output-signal with the input-signal) has the characteristic of a Reactor — there is a 90°
shift in phase as the signal transfers between the transformer windings. The argument is the
same which ever winding is designated as the primary and so we say that the windings
are mutually coupled. In fact the inter-winding coupling behaves as though it were an inductor
and so it it always represented as a mutual inductance.

The magnetic field produced by the primary current does not entirely couple with the secondary
winding of a transformer: a proportion of the field fails to do so and this is referred to
as magnetic leakage. Leakage is expressed as a coupling factor which is given the symbol k.
Should k = 1 then the leakage is zero; if (say) k = 0. 85 it signifies that only 85% of the
magnetic field is linking with the secondary and as a result the signal coupled into the secondary
circuit is 85% of that otherwise to be expected.
The same argument can be applied to the secondary current and the manner in which it couples
to the primary circuit.

This is a simplified (not entirely accurate) account of transformer behaviour but it serves to
demonstrate that primary and secondary signals are not identical in Time. Standard textbooks
will be delighted to present a detailed analysis of transformer action in terms of a vector-
diagram and one quick look at that will explain why I do not give such an account here. This
inherent phase-shift caused by the inter-winding magnetic-coupling produces some rather odd
(but useful) results.

If the Load shown in Fig.2 is a pure resistor then the Generator, “looking- in” through the
transformer input-terminals, sees a resistive load shunted by the inductive-reactance of the
transformer windings. The aim in transformer design is to ensure that this inductive-reactance
has a very-high value at the working frequency and so it can be ignored in this discussion.

If however the Load consists of a capacitor then the Generator sees an inductive-reactance
while, if the load is an inductor, then the Generator sees a capacitive-reactance.

Fig. 3 shows a Generator coupled to a Load via

a double-tuned transformer; this basic
arrangement is to be found in most receivers.
Both primary and secondary windings are
adjusted to resonate at the same frequency (the
method of achieving this is dealt with later).

The frequency of resonance is defined as that

frequency at which the inductive reactance is
equal to the capacitive reactance; the two
reactances cancel so leaving both primary and
secondary windings purely resistive in nature.
However, it must still be true that the primary
and secondary currents are, relative to each
other, phase-shifted through 90-degrees and so
transformer action as described above is to be
expected.

An Observer who “looks into the primary circuit” in place of the Generator would see the Load
coupled through the transformer action; the Observer who“looks into the secondary circuit” in
place of the Load would see the Generator coupled through the transformer action.

(a) Frequency Too Low When the frequency of the generator is lower than the resonance
frequency then the inductive-reactances (in both circuits) fall and the capacitive-reactances rise;
as a result the behaviour of each circuit is controlled by the capacitive reactance and we say that
the circuit becomes predominantly capacitive. With reference to the previous paragraphs the
Generator is presented with a primary circuit which has a capacitive-impedance plus a coupled
secondary circuit which has been transformed to an inductive-impedance. Similarly the Load,
looking back toward the Generator, is being driven by a secondary circuit with a capacitive-
impedance plus a coupled primary circuit which has been transformed to an inductive-
impedance.
For both primary and secondary windings the change of impedance (from resistive to capacitive)
is modified because the coupled impedance from the other winding appears to be inductive.
Hence the departure from resonance which might be expected because of the change of
frequency tends to be compensated.

(b) Frequency Too High The same argument applies when the Generator frequency is raised
above that of resonance. Both primary and secondary circuits become inductive but each
couples into the other a capacitive impedance which tends to compensate the overall impedance
change caused by frequency changes.

The double-tuned arrangement shown in Fig.3 thus

tends to remain “in tune” over a short range either
side of the frequency to which the two circuits are
resonated. The extent to which this self-
compensation is successful depends on how well the
two circuits are matched. They must have the same
Q-factor and be tuned to the same centre-frequency;
the result of failure in this respect is shown in Fig. 4.

The bandpass effect depends too on the degree of

coupling. It should be obvious that, if the coupling
is too small (i.e. the two circuits are not coupled),
then signal transfer from the Generator to the Load
will not take place.

As the coupling is increased so the signal level at the load will rise steadily accompanied by a
similar increase in the frequency compensation. It is possible to make the coupling too tight so
that the circuits are over-compensated; when this happens the circuits are de-tuned off-
resonance instead of offering mutual-compensation.

The result of varying the coupling is shown in Fig. 5. Between the under-coupled single-peak
response and the over-coupled double-peak response there is a happy mean in which the circuit
response remains substantially constant over a range of frequencies. This “flat-topped”
condition is termed critical coupling.

It is usual to employ a coupling factor which is slightly

greater than the critical value and this is indicated by the
appearance of two small maxima in the frequency
response as shown in Fig.5.

The amplitude of these bumps is under control as the

coupling is increased; the advantage of this condition is
that the bandwidth achieved is slightly greater than that
at critical coupling and also that the skirts of the
characteristic are considerably steeper than with critical
coupling.
8.6.4 OTHER FORMS OF COUPLED-CIRCUIT

The double-tuned--transformer introduced above is but one way of coupling two resonant
circuits. In the equivalent circuit of Fig.2 two resonant circuits are shown coupled by a mutual-
inductance but this can be reproduced by using any form of reactance that is common to both
circuits; indeed any common impedance will suffice through which the interchange of energy
can take place. Four alternative arrangements are shown in Fig. 6.

Diagram (a) shows an alternative form of the transformer-coupling illustrated in Fig.3; this type
of circuit was once common in valve-operated communications receivers where it played a part
in arranging for variable bandwidth in the IF amplifier. Yet another variation occurs where the
“transformer” is replaced by a low-impedance untuned link-coupling thus allowing the two
resonant circuits to be physically separated.

To set-up an inductively-coupled pair they must first be separated so as to reduce the coupling;
when each circuit has been tuned for maximum response at the desired centre-frequency (f c )
then the coils are moved closer together until the required bandpass characteristic is obtained.
The two coils are then cemented to hold them in position.

Should re-adjustment be necessary it may not be possible to release the cement and then a
different technique is employed. Each circuit in turn is shunted by a low-value resistor to lower
its Q after which the other circuit can be tuned. However the coupling-factor can only be
changed by moving (one of) the coils.

Diagram (b) shows much the same arrangement but this time each resonant circuit is tapped in
the capacitive arm instead of in the inductive arm. Here the resonance frequency is determined
by Cl and C2 in series while the coupling-factor is determined by the ratio C 2 /(C 1 + C 2 ). A
difficulty here lies in adjusting the circuit; adjustments to the tuning or to the coupling are
inter-dependent.

Diagram (c) shows a capacitive form of the circuit illustrated in Fig.2 in which the mutual-
inductance-coupling is replaced by a coupling capacitor; sometimes the required value of this
capacitor is impractically small and so the component is “tapped-down” the circuit as shown in
diagram (d). As the tapping-point is moved down so the circuit impedance at the tap is reduced
and thus the required value of the coupling capacitor is increased. This arrangement can be
realised also by tapping down the inductive branches but it is generally easier to divide the
capacitive arm than it is to arrange for tapings on the coils.

Whether the circuits are tapped in the inductive arm or in the capacitive arm the same rule
applies as for transformer action; the impedance-ratio is proportional to the square of the turns-
ratio or to the square of the capacitance-ratio.

This arrangement is very easy to set up because the coupling capacitor can be variable. It is set
for minimum capacitance (highest impedance and smallest coupling-factor) and then the two
resonant circuits can be individually tuned to the centre frequency. The coupling-capacitor is
then increased in value until the required band-pass characteristic is obtained.

8.6.5 TO DESIGN A BANDPASS COUPLED-PAIR

As explained in the Lesson notes bandwidth is related to the Q of a circuit which broadly is the
Q of the inductor; this is defined as the ratio Rp / Xp, where Rp is the parallel-connected
resistance (the losses) of the circuit.

These circuits are adjusted to be resonant at the centre frequency and so the value used for X
may be that of the inductive-reactance or of the capacitive-reactance. The following formula
relates Bandwidth (ß) in Hz to the values of capacitance (C in Farads) and resistance (R in
ohms) with reference to the circuit shown in Fig. 7:

ß = √2 / (2π.√C 1 C 2 .√R 1 R 2

In fact the formula relates to any type of coupled-pair arrangement. It has been said that an
essential feature for a flat-topped characteristic is that the two coupled-circuits be well matched
; it follows that C 1 = C 2 and R 1 = R 2 and so the formula reduces to

ß = √2 / 2πCR

Compare this with the formula for a single tuned circuit in which ß = f 0 /Q = l / 2πCR
There is a minor catch however in that the values calculated for both C and R represent the total
effective capacitance and resistance in each circuit; this means that C is formed by the
combination of the lumped capacitor and stray-capacitances and that R is formed by the
combination of the lumped resistor and inherent losses.

The value of C is easily disposed of because the added capacitor has to be (or to include) an
adjustable trimmer. The value required for the additional resistance however depends on the
measured value of the circuit losses; to determine this either

(i) the coil must be wound, adjusted for approximately the correct value of inductance and then
measured on an Admittance Bridge to find the value of Rp. The value of the added resistor
must then be calculated to produce, in parallel with the measured value, the required value of Rp
in the above formula

(ii) When a Bridge is not available then the value of R as calculated must be increased by
guesswork and adjusted so as to achieve the designed Bandwidth when the circuit is finally set-
up.

The flat-topped characteristic depends on the coupling between the two resonant circuits
achieving the critical value. This coupling-factor k is determined from the formula

k = ß/fo = Cc/C

An important parameter when designing a bandpass amplifier is the load which the coupled pair
presents to the active device. This value is the parallel resistance of R 1 and R 2. Thus, in the
design of an IF amplifier, the first step is to determine the desired value of the load and then
R 1 and R 2 are each assigned twice this required value. Note however that R 2 may have to be
increased in value because it will be shunted by the input-resistance of any following stage.

Back to Top of Page

END OF INFORMATION SHEET 6

 PART 8: INFORMATION SHEET No. 7
CIRCUIT DIAGRAMS
(Written By Request)

8.7.1 INTRODUCTION

Circuit diagrams are intended as a pictorial representation of the arrangement and the inter-
relation of the various components which go to make a piece of electronic equipment. They do
not represent the physical layout of the components.

It is possible to describe a circuit in words. For example: "Resistor R3 is the collector-load

resistor for transistor TR1 and it is connected between the collector-stopper resistor R4 and the
low-potential end of the stage-decoupling resistor R2". This may be comprehensible as it stands
but few could make much of a similar description of a four-stage amplifier with voltage-
derived-negative-feedback from a transformerless output stage which is parallel-applied to the
input?

If circuit diagrams are intended to clarify the construction and operation of equipment then, as
with the written word, it is important that the information can be easily recovered by another
person who is distant in both time and place. To this end it is necessary to adhere to certain
agreed methods of presentation.

Equally it is important that finished diagrams be clearly labeled and that their titles make it
possible to file them in some system from which they can be retrieved given either a Title (a
number) or a reference such as to the name of the system of which they form a part. I have met a
system in which it was very easy to assign a reference number for any drawing and yet, once
filed, it was impossible to find that drawing again unless that exact number was known!

8.7.2 BASIC LAYOUT OF A DRAWING

Drawings are laid out using the same conventions as used in western writing namely the
information-flow moves from left to right across the paper and from the top downward . Of
necessity this Rule cannot be applied rigidly but there are accepted conventions for dealing with
difficulties; see Fig.1.
1. Signal inputs are entered from the left-hand-side of the paper; signal outputs are taken to
the right-hand-side of the paper.
DC Power-supplies are entered from the right-hand side of the paper; this often includes
the transformer/rectifier circuit from which supplies may be obtained.
2. Where the signal-chain is too long to be accommodated across the paper then a second
line is commenced and this second line too starts at the left-hand-side.
(Drawings are made where in fact the signal-flow is taken from right to left along the
 second line; sometimes this may be done for a valid reason but it should be avoided if at
all possible because it will confuse a would-be reader confronted with a very complex
drawing).
3. Auxiliary circuits (for example Local Oscillators in a Receiver or an alternative input
amplifier) are drawn immediately below the main-path circuit-arrangement on which they
operate.
( In a two-line main-path the main circuit is drawn on lines 1 & 3 while the auxiliaries are
placed on lines 2 & 4).
4. (a) Ground (Earth, earthy, 0-volt) lines are drawn at the bottom of the main-path using a
heavy line
(b) Positive supply-lines are drawn above the circuit except where the major feed is
negative as when using pnp transistors. The higher voltage [Link] drawn in the
higher positions
(c) Bias-lines (including a.g.c. lines) of either polarity are brought in below the ground
line. Negative supply-lines (in dual-supply circuits) are brought in below all other lines.
Higher-value negative supplies are drawn in the lower positions.
5. In complex diagrams, which make several signal-lines across the paper, it may clarify a
drawing if these supply lines are drawn separately for each signal line but each must be
clearly labeled; where multiple supplies are used at the same voltage they must be
distinguished; for example +12(1) and +12(2) .(The reference number might be that of the
connector-pin through which the supply enters the unit.
6. (a) Input-handling circuits are drawn to the left of the active-device which they are
driving.
(b) Output circuits are drawn above the active-device which is driving them. This leaves
room for any input arrangement, that may be needed to drive the next Stage, to be drawn
to the left of that Stage.
Fixed biases which are derived elsewhere, including such things as a.g.c. lines , are
brought in below the Ground line and up beneath the active device(s) which they serve.
7. Connections are always indicated by a dot at the point of connection; crossed connections
are avoided. Where wires cross but do not make electrical contact they are shown without
a dot.
8. Standard symbols are used to represent various components (see Figs. 2 to 5); many of
these have been changed to suit computer drawing. For example the old zigzag resistor is
difficult to draw by computer although that is probably not so with today's graphics
packages in which pre-drawn symbols can be picked and placed with a "mouse"; note
however that the zigzag form is easier to draw by hand.
9. There are certain standard layouts for particular types of circuits and these should always
be used . Examples are given in the sheets of drawings attached (see Fig. 7); e.g. a bridge
circuit, a multivibrator. Correct use of these forms makes a circuit-arrangement
immediately recognisable even to the extent of using an abbreviated form in block-
diagrams.
10. Standard forms are laid down too for block-diagram use. In general such diagrams
represent each circuit-arrangement as a box within which is drawn either a standardised
symbol or a word description.
11. The addition of signal-flow direction-arrows can make both circuit diagrams and block
diagrams much easier to follow but these must be added with care . Arrows added
unnecessarily merely clutter a diagram; they are most useful when added after a
Junction. Do not use arrows to indicate direct-current flows except of course where dc is
the required signal.
 12. It is accepted practice to omit dc connections to op-amp and logic symbols.

8.7.3 SYMBOLS USED IN CIRCUIT DRAWING

On the following pages Figs. 2 to 5 show a selection of symbols commonly used in drawing
circuit diagrams although variations on these will be found. Note that the circles which
represent valve/transistor/diode envelopes are often omitted; much depends on the complexity
and density of a drawing whether these envelopes are best omitted or included.

Fig. 7 shows examples of standard circuit layouts and one example of a foolish manner in which
a multivibrator is sometimes drawn. Such standard layouts often make it possible to decipher a
complex drawing in a matter of seconds; a badly-drawn diagram may need to be re-worked
several times before even an experienced engineer can determine its purpose.

Always avoid "tram-lines" - the system of drawing in which all connections are run in parallel
tracks along one (or more) edges of the drawing. Such diagrams are difficult to read and it is
also very easy to make errors of commission when setting-out and checking the work. Where
some tram-lines are unavoidable then they should be separated into groups of three or five (max)
to enable the eye to follow a particular line without finger aid.
Back to Top of Page

END OF INFORMATION SHEET 7

 PART 8: INFORMATION SHEET No. 7
CIRCUIT DIAGRAMS
(Written By Request)

8.7.1 INTRODUCTION

Circuit diagrams are intended as a pictorial representation of the arrangement and the inter-
relation of the various components which go to make a piece of electronic equipment. They do
not represent the physical layout of the components.

It is possible to describe a circuit in words. For example: "Resistor R3 is the collector-load

resistor for transistor TR1 and it is connected between the collector-stopper resistor R4 and the
low-potential end of the stage-decoupling resistor R2". This may be comprehensible as it stands
but few could make much of a similar description of a four-stage amplifier with voltage-
derived-negative-feedback from a transformerless output stage which is parallel-applied to the
input?

If circuit diagrams are intended to clarify the construction and operation of equipment then, as
with the written word, it is important that the information can be easily recovered by another
person who is distant in both time and place. To this end it is necessary to adhere to certain
agreed methods of presentation.

Equally it is important that finished diagrams be clearly labeled and that their titles make it
possible to file them in some system from which they can be retrieved given either a Title (a
number) or a reference such as to the name of the system of which they form a part. I have met a
system in which it was very easy to assign a reference number for any drawing and yet, once
filed, it was impossible to find that drawing again unless that exact number was known!

8.7.2 BASIC LAYOUT OF A DRAWING

Drawings are laid out using the same conventions as used in western writing namely the
information-flow moves from left to right across the paper and from the top downward . Of
necessity this Rule cannot be applied rigidly but there are accepted conventions for dealing with
difficulties; see Fig.1.
1. Signal inputs are entered from the left-hand-side of the paper; signal outputs are taken to
the right-hand-side of the paper.
DC Power-supplies are entered from the right-hand side of the paper; this often includes
the transformer/rectifier circuit from which supplies may be obtained.
2. Where the signal-chain is too long to be accommodated across the paper then a second
line is commenced and this second line too starts at the left-hand-side.
(Drawings are made where in fact the signal-flow is taken from right to left along the
 second line; sometimes this may be done for a valid reason but it should be avoided if at
all possible because it will confuse a would-be reader confronted with a very complex
drawing).
3. Auxiliary circuits (for example Local Oscillators in a Receiver or an alternative input
amplifier) are drawn immediately below the main-path circuit-arrangement on which they
operate.
( In a two-line main-path the main circuit is drawn on lines 1 & 3 while the auxiliaries are
placed on lines 2 & 4).
4. (a) Ground (Earth, earthy, 0-volt) lines are drawn at the bottom of the main-path using a
heavy line
(b) Positive supply-lines are drawn above the circuit except where the major feed is
negative as when using pnp transistors. The higher voltage [Link] drawn in the
higher positions
(c) Bias-lines (including a.g.c. lines) of either polarity are brought in below the ground
line. Negative supply-lines (in dual-supply circuits) are brought in below all other lines.
Higher-value negative supplies are drawn in the lower positions.
5. In complex diagrams, which make several signal-lines across the paper, it may clarify a
drawing if these supply lines are drawn separately for each signal line but each must be
clearly labeled; where multiple supplies are used at the same voltage they must be
distinguished; for example +12(1) and +12(2) .(The reference number might be that of the
connector-pin through which the supply enters the unit.
6. (a) Input-handling circuits are drawn to the left of the active-device which they are
driving.
(b) Output circuits are drawn above the active-device which is driving them. This leaves
room for any input arrangement, that may be needed to drive the next Stage, to be drawn
to the left of that Stage.
Fixed biases which are derived elsewhere, including such things as a.g.c. lines , are
brought in below the Ground line and up beneath the active device(s) which they serve.
7. Connections are always indicated by a dot at the point of connection; crossed connections
are avoided. Where wires cross but do not make electrical contact they are shown without
a dot.
8. Standard symbols are used to represent various components (see Figs. 2 to 5); many of
these have been changed to suit computer drawing. For example the old zigzag resistor is
difficult to draw by computer although that is probably not so with today's graphics
packages in which pre-drawn symbols can be picked and placed with a "mouse"; note
however that the zigzag form is easier to draw by hand.
9. There are certain standard layouts for particular types of circuits and these should always
be used . Examples are given in the sheets of drawings attached (see Fig. 7); e.g. a bridge
circuit, a multivibrator. Correct use of these forms makes a circuit-arrangement
immediately recognisable even to the extent of using an abbreviated form in block-
diagrams.
10. Standard forms are laid down too for block-diagram use. In general such diagrams
represent each circuit-arrangement as a box within which is drawn either a standardised
symbol or a word description.
11. The addition of signal-flow direction-arrows can make both circuit diagrams and block
diagrams much easier to follow but these must be added with care . Arrows added
unnecessarily merely clutter a diagram; they are most useful when added after a
Junction. Do not use arrows to indicate direct-current flows except of course where dc is
the required signal.
 12. It is accepted practice to omit dc connections to op-amp and logic symbols.

8.7.3 SYMBOLS USED IN CIRCUIT DRAWING

On the following pages Figs. 2 to 5 show a selection of symbols commonly used in drawing
circuit diagrams although variations on these will be found. Note that the circles which
represent valve/transistor/diode envelopes are often omitted; much depends on the complexity
and density of a drawing whether these envelopes are best omitted or included.

Fig. 7 shows examples of standard circuit layouts and one example of a foolish manner in which
a multivibrator is sometimes drawn. Such standard layouts often make it possible to decipher a
complex drawing in a matter of seconds; a badly-drawn diagram may need to be re-worked
several times before even an experienced engineer can determine its purpose.

Always avoid "tram-lines" - the system of drawing in which all connections are run in parallel
tracks along one (or more) edges of the drawing. Such diagrams are difficult to read and it is
also very easy to make errors of commission when setting-out and checking the work. Where
some tram-lines are unavoidable then they should be separated into groups of three or five (max)
to enable the eye to follow a particular line without finger aid.
Back to Top of Page

END OF INFORMATION SHEET 7

 PART 8: INFORMATION SHEET No. 9
BEAMS
8.9.1 INTRODUCTION

British Standard 204 gives the following definitions:

Aerial (Antenna U.S.A.) That part of a radio system which is designed to radiate electro-
magnetic waves into free space. This does not include the transmission line or waveguide to the
radiator.

Directional Aerial An aerial designed to radiate more strongly in some directions than in
others.

Beam 1. The radiation within a lobe of a directional system.

2. The region of space illuminated by this radiation.

Thus those who use the word "Beams" meaning "Beam Aerials" would seem to be out on a limb
but they have my sympathy as one who has experienced trouble by disagreeing with the B.S.
pundits. I prefer to define an aerial as a device which converts electrical energy into electro-
magnetic energy. Electrical energy is under full control while being generated within a
transmitter or while entrusted to a feeder but, once converted to a radiating wave, it is not under
control at all.

At the point of conversion (the aerial) there is one last chance to influence the behaviour of the
electro-magnetic wave in that we can have a say in which direction that wave will propagate and
that brings us to the directional aerial. Theory and practice however are eternally uneasy bed-
fellows.

To most Readers I am sure there is little new in this but, while almost all know the advantages
of a directional aerial, few seem aware of its disadvantages or that many of the advantages are
illusory.

8.9.2 THE DIPOLE

This most simple of the aerials is also the most simple form of directional-aerial; it radiates its
energy in two major lobes " ... one each side of the half-wavelength radiator ... " so all the
Books say. In fact it is not true. Its major (and only) lobe is doughnut-shaped and it surrounds
the radiating conductor for the whole of its length. The result is that energy is radiated with
equal strength in all directions in a plane that is pierced by the conductor.

Thus a horizontal dipole sends a large part of its energy to irradiate the sky at those high-altitude
angles where it is unlikely to return; because the ground acts as a reflector this "lost" energy can
be greater than is at first apparent. There is an advantage of course in that the one aerial
simultaneously covers all angles of launch and so serves for all distances.
A dipole produces zero radiation off the ends of the conducting radiator and so it is surprising
that so often we get 5/9 reports from a Contact whose Station is in line with the aerial.
Practical dipoles are usually fed from a coax feeder and, as a consequence, are not dipoles at all.

For a tongue-in-cheek explanation imagine yourself to be a packet of energy flowing along a

coaxial feeder as an e-m field supported by the voltage between inner and outer conductors and
by the currents which flow on the inner and on the inside surface of the outer. Arrived at the
aerial-feedpoint the current from the inner is sent into one limb of the aerial and the current on
the outer is sent into the other limb. At the far ends of the dipole there is nowhere for current to
flow and so it bounces back to the feedpoint.

Current which flowed from the inner is returned to that inner but that which flowed from the
outer can return either down the inside surface of the outer (whence it came) or down the outside
surface. It is this last current flowing on the outside of the feeder, which explains why a coax
should not be connected directly to a dipole unless the coax/dipole combination is required to
operate as a top-loaded vertical aerial.

As a purely practical point it is a good idea to connect a coax in this manner if it produces
results without causing interference.

8.9.3 THE LOOP AERIAL

For some mysterious reason there are those who insist on calling this aerial a "magnetic loop";
as used in paging systems (i.e. at low frequencies) the magnetic loop has the extremely useful
property of limiting its radiation to a distance of a few feet unless the Receiver is within the
loop. To be of use in radio communication a loop aerial must launch an electra-magnetic wave.

Once again the main (and only) lobe is doughnut-shaped and surrounds the length of the
conductor. There is strong and equal radiation in all directions that lie in the plane of the loop
and there is negligible radiation along the axis of the loop.

Used with the loop in a vertical plane (its axis horizontal) this aerial shows directional properties
around the horizon and the two sharp nulls can be useful in removing interfering signals while
receiving. One of its earliest uses was in direction-finding equipment. However, unless there is a
specific need to limit radiation in a particular direction, there is no advantage whatever in using
a vertical loop for transmitting. A loop does not concentrate available power in one direction
and so it is not a directional (beam) aerial. The best manner in which to use a loop for
transmitting is undoubtedly horizontal and it will function very well in all directions at any
height starting from zero.

In the first half-hour in which we tested an experimental horizontal loop suspended at head-
height from the shack roof we contacted Denmark (reported signal-strength 1), Berlin (5) and
Greece (9).

Essential formulae for designing loops often give dimensions in terms of wavelength but the
reason for this may appear obscure because a loop does not support a standing wave. The
essential is that the loop is resonated at the operating frequency so as to maximise loop-current.
To this end loops are often constructed with very-low values of series resistance by using
materials such as copper pipe. The disadvantage of this technique is that the resulting very-high
Q-value means that the bandwidth is very restricted and the aerial must be tuned accurately at
each frequency in use.

It should be remembered that the advantage of maximising loop-current can easily be lost by
failure to achieve the extremely-critical tuning adjustment and, in practice, there is an optimum
Q-value.

8.9.4 THE YAGI

This device, highly acclaimed by television engineers and Amateurs, is probably the most
misunderstood aerial of them all but that does not make it the least useful. As defined by B.S.
204 it is an aerial consisting of one primary radiator and two or more secondary radiators ...
which constitute an end-fire array ... Although specifically excluded by that definition the H-
aerial is a logical step in the development of a yagi configuration.

The yagi uses a single driven element and, no matter how many, all the others are parasitic.
This means that they draw their power from the driven element by irradiation and then they re-
radiate. In theory backward radiation is eliminated by use of a reflector which may take the
form of either another dipole element or a reflective sheet or grid. So far the aerial is just an H-
aerial and possesses a cardioid radiation pattern.

The other elements are all placed in front of the primary radiator and their re-radiation fields
add-to and subtract-from that of the driven element so as to reinforce radiation (and reception) in
the desired forward direction and to cancel it in all other directions. These functions operate
both in azimuth and altitude. The result is a narrow beam-like lobe which contains (nearly) all
the radiated power.

Because all the radiation is concentrated in that single beam the aerial provides forward signal-
gain; it does not amplify the signal but merely concentrates in one direction the power that
otherwise would be broadcast. The more elements (directors) in front of the driven element so
the greater is that concentration. Similarly, in reception, the aerial gathers irradiating power to
give maximum signal response from the "forward" direction. Delightfully simple though it
may seem that alas is not the whole story.

Interaction between the driven element and the parasitic elements causes the driving-point
impedance of the dipole to fall with each added element (it is driving them all in parallel). To
compensate for this the dipole is often replaced by a folded dipole; like the loop which it
closely resembles this offers a driving-point impedance around 300 ohms and so the final
impedance is closer to the 50 or 75 ohms that is required to match a coaxial feeder. The folded
dipole also improves the bandwidth of the overall device but it still provides a balanced load and
, as with the simple dipole, it does not function as required when simply tacked-on to a coax
feeder.

In theory the more elements used to construct a yagi the narrower does that forward lobe
become and the greater is the power which it launches in the forward direction (for a given input
power) . Unfortunately, as the elements increase in number, so the bandwidth decreases; despite
correction provided by the folded dipole the aerial is very difficult to compensate and can be
brought within a given tolerance only for a few spot frequencies. This is why, when you buy a
new television aerial, it is necessary to state the area in which it is to be used; should you move
house there is little point in taking the television aerial with you.

Look at a yagi and you will see that the reflector is longer than the driven element and that each
director is shorter but also that they become progressively shorter with increasing distance from
the driven dipole. The object is to make the reflector inductive and the directors capacitive;
combined with their relative spacings along the aerial boom this is part of the design to ensure
correct phasing of their many re-radiation fields. At least that is the theory but what happens in
practice ?

If the relative phasing depends on the complex impedance (R + jX) of each element and on the
element-spacings it is evident that the overall impedance of the aerial is frequency-sensitive;
hence the narrow bandwidth . I recently spent three days re-aligning the band-pass filters in a
Racal RAI7 receiver which somebody had tried to peak. Each filter consists of eight LC circuits
placed in a straight line and coupled only by their mutual fields which makes them an exact
analogy of a yagi.

In my youth I was involved in an attempt to build signal-strength-measurement yagi-aerials in

brass to avoid the corrosion problems of using aluminium aerials connected to copper feeders
close to the sea ; the task bordered on the impossible. Any adjustment to a resonator alters the
complex impedance reflected into all its neighbours, it changes the Q-factors of the circuits and
it changes their coupling factors; tuning adjustments to any one circuit detunes all the others. In
fact a Yagi aerial is a very complex piece of equipment and is very easily ruined.

[ Most of the techniques available for adjusting the LC filters described above are impossible to
apply to the rod-resonators of a Yagi aerial ].

On a more mundane plane the highly-directive properties for which yagis are prized are also
their greatest disadvantage. Once the number of elements exceeds three ( dipole, reflector and
director) it becomes ever more crucial to aim the aerial with great accuracy both in azimuth and
in altitude. However accurately this may be done there is a problem in that, when the wind
blows, a yagi with a long boom will wobble and dance and produce an effect I have christened
"wind-fading". I achieved an enormous improvement in our television reception by taking down
the installed 14-element yagi, removing the front half and re-mounting the remainder so that it
was mechanically balanced about the support pole

This eliminated the wind-fading partly because the wind-wobble became very small but mainly
because the front lobe was broadened and so the direction of propagation was kept within that
beam-width despite the wobble. Subjectively there was an increase in the average signal
strength.

From the Amateur point of view the yagi is a trickster. It does not produce that clean forward
narrow lobe which theory and advertisements predict and the forward lobe is not necessarily in
line with the aerial boom. It produces also what are euphemistically called back-lobes and side-
lobes; i.e. the reflector does not do a perfect job and the directors fail to get the relative
phasings correct. Indeed even in theory the phasings can be correct only at a few spot
frequencies. Side lobes can sometimes rival (even exceed) the forward lobe and back lobes are
far from being insignificant as evidenced by that oft-heard amazing tale "He was coming in 5-
over-9 and that was off the back of the beam ". More likely he was being received on the
outside of the coax feeder. Test a Yagi that is wrongly connected to a coax feeder and you may
find that much of its directional properties have disappeared

The front-to-back ratio of a yagi expresses the relative amplitudes of the (desired) forward lobe
and of that back or side lobe which is closest to the reverse direction. It is a measure of the
aerial's failure to do its designated job and is often confused with the forward gain. An aerial
which produces back or side lobes cannot be pushing all the available power into the forward
lobe and so the forward gain must suffer.

A problem that arises from this is that of knowing whether the aerial is aimed directly at the
would-be contact or whether that contact is sitting in a side lobe. Of course it is possible to pop
outside with a compass but that will not resolve anything unless the Great-circle bearing is
known and the reception path does not involve a reflection? And perhaps the side-lobe is greater
than the forward lobe?

One advantage of a directional aerial has been mentioned already that of losing a strong
interfering signal in a null ; while a yagi can be very good at this it can also be very frustrating
— it has many nulls, both sharp and shallow, but most of its lobes are sharp too.

The greatest act of self-kidding is that theoretical forward gain claimed as being up to 19-dB. If
the aerial is relatively new and free of corrosion, if it has been correctly attached to the feeder
and so avoids currents on the outside of the feeder, if it matches the feeder impedance at the
frequency in use, if none of the elements are bent or broken or missing or corroded, if the aerial
is correctly aligned on its target in both azimuth and altitude and if the receiver too is in good
shape then that 19-dB may be realised. But such a gain represents around two S-points only. It
is possible that more contacts can be achieved with a round-the-horizon aerial added to which a
loop aerial is undoubtedly much quieter.

Amateur radio, even on vhf, began with dipoles and long wires and often involved the use of
separate aerials for transmitting and receiving. During World-war II (so I am told) a Mr. Yagi
tried to make a death-ray and since then his cult following appear to use money rather than
sense, to lose as many contacts as they gain, to wipe out some rigs while remaining unaware of
others and to use good engineering skills to erect unwieldy sagging doubtful arrays of bad
engineering.

In the immediate post-war period the term "TV Aerials" meant an H-aerial and there were many
who believed that the new-fangled television signals were H-shaped. With the move to shorter
wavelengths the extended fish skeleton became fashionable. While it is true that the shorter uhf
dipoles pick-up less signal and require boosting the proliferation of directors has more to do
with profit than with engineering necessity.

8.9.5 BANDWIDTH AND THE A.T.U.

In practical engineering there are no absolutes and parameters must be specified in terms of
ranges over which tolerances must be met. Applied to the aerial this term Bandwidth refers to
the range of frequencies over which the driving-point impedance matches the characteristic-
impedance of the feeder within a specified limit. For example a match to within 10% may be
specified over the range 3.5 to 3.8 MHz which would produce a s.w.r. of 1.1:1. The bandwidth
is the same whether the device is used for transmitting or receiving but the results are different.

It is believed by many that an aerial mismatch does not matter so much when their cheque book
can stand the shock of an aerial-tuning unit (properly called an aerial transducer unit) but they
confuse the baby with the bath water. A good match is essential between a transmitter and its
load (the feeder plus aerial) if the maximum amount of power is to be delivered; the Maximum
Power-transfer Theorem states that load-impedance must exactly match the internal-impedance
of a generator. When the driving-point impedance of an aerial exactly matches the
characteristic impedance of the feeder then, at the other end of that feeder, the transmitter is
presented with a resistive load that is normally 50-ohms.

When the aerial/feeder matching fails then, at the other end of the feeder the transmitter load
becomes a complex impedance but, in itself, this is not a problem. The tuning controls of the
P.A. cope with this variation but, in doing so, they either increase the voltage developed across
the active device (either valve or transistor) or they increase the demand for current. Problems
are few where the P.A. uses a valve but with transistor P.A.'s it is the Bank balance which comes
under pressure. An A.T.U., once properly adjusted, protects the P.A. but it serves no other useful
purpose; the one thing an ATU cannot do when housed in the Shack is to tune the aerial (long
wires excepted).

When transmitting , a mismatch between aerial and feeder causes the appearance of a standing-
wave on the feeder; this increases feeder losses but, at the power levels permitted by the
Amateur Licence, it does little harm. A badly-adjusted A.T.U. may increase the standing-wave
because it causes a re-reflection but it cannot reduce the basic reflection at the aerial. Thus an
A.T.U. may reduce the bandwidth of an aerial but it cannot improve it unless it is hauled up the
mast and connected between the aerial and the feeder.

In reception the story is different; a mismatch between aerial and feeder means that the aerial
fails to deliver the maximum possible power to the feeder and a loss of overall sensitivity
results. Now the "load" on the feeder is the input-impedance of the receiver and, because the
A.T.U. produces (hopefully) a perfect match between feeder and receiver, a standing-wave does
not appear on the feeder. However it can be advantageous to deliberately mismatch the receiver
input-impedance and so cause a standing-wave to appear with the receiver terminals at an
antinode (a node is a point of no-displacement — a minimum). Because of the conflicting
requirements however this technique demands separate aerials for transmitting and receiving.

The radiation pattern of a compound aerial such as a yagi is designed and adjusted with the
device matched to its feeder. With a mismatch there must be a phase-displacement between
driving voltage and driving current which must be reflected in the aerial's performance. Not
only is an A.T.U. unable to correct this but it might even make the situation worse. The wise
will limit yagi-aerial building to the three basic elements and so preserve some knowledge of
which way their signal is going. Use an A.T.U. by all means (I do) but not as a panacea for lazy
engineering.

Back to Top of Page

END OF INFORMATION SHEET 9

 PART 8: INFORMATION SHEET No. 10
THE CONSTANT-RESISTANCE NETWORK
8.10.1 INTRODUCTION

The Constant-resistance Network, shown in Fig. 1, has the property of not being frequency-
conscious; i.e. within the limitations of stray capacitances and stray inductances, which are
inherent in practical circuitry, the network behaves in the manner of a resistance. The circuit is
said to be non-resonant or aperiodic.

A normal practical parallel-combination of Inductor and Capacitor would be shown with a

resistor in series with the Inductive branch only and this represents the resistance of the wire
with which the component is wound plus a small amount extra to account for losses inherent
chiefly in the core material. Losses associated with capacitors are small when compared with
those of an inductor and usually would be ignored. To make a constant-resistance network the
resistance (losses) in both branches of the circuit are equalised.

The resistance in the inductive branch is usually designated R L and that in the capacitive branch
R C ; to make a Constant-resistance Network the criteria are:

RL = RC

L / C = R2

(The mathematical derivation of this relation is given at the end of this Information Sheet. Those
familiar with network theory might recognise this as a critically-damped resonant circuit.)

The arrangement is useful, for example, when designed into the output circuits of Signal
Generators; given that the apparent “internal impedance" (the source impedance) of a generator
has the nature of a resistor it is possible to give a single specification for the output signal which
covers the full frequency-range of the instrument.

Another use for this circuit is to suppress the arcing that occurs when a switch breaks an
inductive circuit. It does seem to be known that a combination of resistor and capacitor
connected across the switch-contacts makes a better suppressor than a simple capacitor but few
seem to know why or how to select the value of the resistor; even less seem to know how to use
the constant-R network.

8.10.2 THE CAUSE OF ARCING

Inductance in an electric circuit has the same nature as a flywheel in a mechanical device; i.e. it
tends to resist change. When an electric circuit is interrupted inductance tends to keep the
current flowing but, because of the interruption, such a current has nowhere to flow. As with
transmission-line and aerial theory the effect is to pile-up electrons on one side of the switch and
to remove them from the other.

The resulting voltage which is generated across the circuit-break is referred to as an “inductive
kick” or usually as the back-emf of the inductor. At the moment that the break is created there
is only a small separation between the switch contacts and the resulting “voltage gradient” can
be sufficient to ionise the air within that gap; these ions support the continuation of current-
flow.

Such “arcs” however tend to be struck over small patches of the switch contacts which causes
localised heating with further ionisation as the contact material melts and so the arc is sustained
as the contacts move further apart.

Sometimes the circuit which is broken contains significant inductance (e.g. an electric motor)
but, in any event, it contains stray inductance distributed along the length of various conducting
paths. In combination with stray capacitances these form resonant circuits. When the switch is
opened such resonant circuits are left undamped and they oscillate freely at their natural
resonance frequencies; as a result, whether the power-supply is ac or dc, the arc is sustained by
an alternating voltage. The resulting rf alternating currents are the cause of radio-interference.

The cure is to ascertain the value of the circuit-inductance and the associated resistance and then
to add in parallel a capacitor-resistor combination that satisfies the above equation. When this is
correctly achieved an arc is not generated (or is greatly reduced) because effectively the switch
is breaking a resistive circuit.

Fig. 2 shows the different arrangements. In diagram (a) the switch contacts are shunted with a
resistor-capacitor combination. The idea is that the capacitor allows passage to rf but does not
restore the dc-continuity that was broken by the switch action. The energy of the alternating
currents is dissipated as heat in the resistor.

In diagram (b) a resistor-capacitor combination has been connected, not across the switch, but
across the load and the values have been chosen to satisfy the relation L / C = R2. When the
correct component values are used rf current does not flow.
However, Fig. 2 is more than just a comparison of two alternative methods. Should the power-
supply and its connecting wiring be bypassed with a low-inductance capacitor (immediately to
the left of the switch shown) then diagrams (a) & (b) are seen to be identical. Apart from being
food for thought this gives a clue as to the choice of values for C & R in diagram (a).

A practical example of the application of this network is given at the end of this Information
Sheet .

8.10.3 DERIVATION OF THE C-R VALUES

Let the inductive branch of the network be represented by R L + jX L

Let the capacitive branch of the network be represented by R C - jX C

These two combinations are connected in parallel and so their effective impedance is found
using the “product-over-sum" rule:

Z = (R L + jX L )(R C - jX C ) / (R L + jX L + R C - jX C )

which simplifies to

Z = (R L + jX L )(R C - jX C ) / (R L + R C + j(X L - X C ))

The result of evaluating this equation has to be the production of resistive terms plus other terms
which are reactive (they contain j ). For the network to be aperiodic (non-resonant) the reactive
terms must not be present (i.e. they must sum to zero) thus leaving only a resistive component.
Thus the task is to manipulate this expression into the general form (a + jb) and then equate
the j-terms ( “b” ) to zero which (hopefully) will yield the required conditions.

I do not propose to develop the entire argument here. The masochists among you will enjoy the
algebra; the rest will not read it. Hence I give only guidelines.

To be able to separate the top line into R and jX components the bottom line must be reduced to
a form without j-terms. This is done using the algebraic relation that a2 - b2 = (a + b)(a - b);
i.e. the expression is multiplied by 1 in the form of
 R L + R C - j(X L - X C ))/(R L + R C -j(X L - X C ))

This operation converts the j-terms on the bottom line to j2-terms; because j2 = -1 the bottom
line is no longer frequency conscious and so the top line can be separated into resistive and
reactive terms each divided by the bottom line.

However, it is only the reactive terms which are of interest and so it is not necessary to multiply-
out the horrible-looking expression that has resulted. Concentrate only on those multiplications
in the top line which result in reactive terms.

When further simplification is no longer possible put R L = R C = R and the j-terms should be
revealed as

j(R2 - L /C)(X L - X C )

For the network to be aperiodic these terms in j must sum to zero and so

(R2 - L/C)(X L - X C ) = 0

For this to be so either (X L - X C ) = 0 ... (resonance) or R2 - L/C = 0

whence R2 = L/C

8.10.4 PRACTICAL EXAMPLE

A small dc motor was used which had been recovered from a defunct video- recorder; it
operated from 7.5 volts. In place of a switch one of the connecting wires was rubbed against a
metal terminal which produced a barrage of crackles as heard on a f.m. receiver tuned to
approximately 100-MHz ( the receiver was not tuned to any broadcast signal).

The first task is to measure the resistance and the inductance of the motor. The first
measurement involves the straightforward application of a multimeter and it was found to be 8.5
ohms.
The inductance was measured using the arrangement shown in Fig. 3. The motor was shunted
with a 0.47 μF capacitor to resonate it and it was connected to an audio oscillator (a Tone
Source) via a 2.2-kilohm resistor. These components happened to be on the bench! An
alternating voltmeter was connected across the motor. Variation of the signal-generator
frequency produced a maximum response on the voltmeter at 2.340 kHz.

The inductance of the motor can be calculated from the formula for resonance-frequency f =
l/(2π√LC)

Hence

2.340 = 1/6√{(L x 0.5 x l0-6)} approx.

= l03 / 6 x 0.775√L approx.

and so

√L = 103 / (4.65 x 2.340) = 103 / 10.88 x 10-3

L = (1/11)2 = 1000/121 millihenrys

OR 8.5 mH approx

The capacitor required for correction is given by

L/C = R2 or C = L/R2

whence C = 8.5 x 10-3 / (8.5)2 = 10-3/8.5 F = 1000 / 8.5 μF

= 118 μF approx.

The standard-value components of 100 μF and 6.8 ohms effectively removed all interference
when connected in either position.

Back to Top of Page

END OF INFORMATION SHEET 10

 PART 8: INFORMATION SHEET No. 11
MORSE
8.11.1 TO LEARN THE MORSE CODE

Morse is but another language, another way of imparting information via the medium of sound
Unlike speech however it does not rely on varying the sound itself neither does it possess a
syntax — a grammatical structure — of its own. It is used to spell out, letter by letter, the
language in common use by the sender. This may sound cumbersome but Operators do indeed
learn to read it with the same facility as speech.

This is achieved in the same manner as speech is learned ... by listening to it. We learn our
mother-tongue by listening to it in daily use first while in the womb and after from the constant
talking which most Mothers employ to their offspring following birth. Although young
children/babies may not be capable of producing speech there is little doubt that they understand
much of it.

Bad habits acquired in early life are very hard to shed later and so it is important, when learning
Morse, not to be too hasty to "get sending" by pounding that key. Morse is a language of
rhythms and is acquired by listening to transmissions of GOOD Morse; once those rhythms are
firmly implanted in the memory then it is difficult indeed to send bad Morse.

This was well-illustrated to me one night when I heard an Operator constantly sending KWA.
I was puzzled by this but, returning down the Band about 10 minutes later, I found him still
sending this strange call without obtaining a response. Only then did it occur to me that he was
intending to send the general call CQ? He had not learned the code properly and fancied his
keying technique somewhat; unfortunately instead of sending dah—di—dah—di dah—dah—
di— dah he was actually sending dah—di—dah di—dah—dah di—dah. It did not sound like CQ
in any way whatever. Probably others had recognised his intention but I doubt that anybody
would have wasted time trying to read him.

Start by acquiring tapes of Morse code which have been keyed by a truly skilled operator and
listen to them for 5 to 10 minutes at least once per day. Failing this, if you possess a receiver,
listen daily to the Slow-morse transmissions which are transmitted on the amateur Bands (details
published by the R.S.G.B.). Yet another alternative is to program a computer to send Morse;
you can easily arrange for letters or figures to be sent in a random sequence and at any speed
you care to define.

A computer will send immaculate Morse code but it is relentless and will drive you into the
ground; program it to give you a few seconds rest every so many characters or at set intervals of
time.

Start by trying to pick out certain letters. Look at the code shown under 11.4 below and you will
see that many characters form pairs; for example A is di—dah while N is dah—di. DON'T
think in terms of dots and dashes but get the feel of each rhythm by using the accepted phonetics
dah and di (dit). As you begin to recognise the rhythmic pattern of each letter then add another
to your search. DON'T be in a hurry !
8.11.2 THE STRUCTURE OF THE MORSE CODE

Because Morse does not use sound-variations the only possible way to introduce variety is by
changing the duration of each sound; thus the code uses so-called dots and dashes. (For those
who bash keyboards this is equivalent to the Computer's use of binary numbers which, usually,
are designated in 0's and i's ).

There is an internationally-accepted structure:

The basic unit is the di or dit (dot)

A dah (dash) is equal to 3 di's

The space between characters is equal to 1 di

The space between letters is equal to 3 di's

The space between words is equal to 5 di's

One way to control the overall speed at which a message is sent (usually given in words-per-
minute) is to vary the length of the di but unfortunately this does not work in practice. As stated
above to read the code successfully it is necessary to recognise the rhythms of each letter; if the
code is slowed down by lengthening the basic unit the rhythm tends to disappear. For example
the letter Q is dah—dah—di—dah and this is usually recognised as the rhythm of "here comes
the bride" or as "God save the Queen". But try saying it out loud and long as "daaaah daaaah
deee daaaah" and you will find that it is not so easy to pick out.

Thus to send Morse slowly it is essential to preserve those basic rhythms and lengthen only the
gap between each letter and (perhaps just a little) the gaps between words. In other words each
letter should be sent crisply with enough time in between to keep the overall sending rate at the
required speed.

That last paragraph however seems to bring me into direct dispute with today's Morse-exam
gurus who, in the August 1997 RadCom, report that candidates who were prepared for the
Morse Test using this method found difficulty in receiving "correctly-proportioned Morse".
Unfortunately the Report does not say anything about the sending skills of those responsible for
the training nor of those who sent during the Tests. Professional telegraphists in my
acquaintance (including war-time naval instructors) use the above method.

A local club was fortunate in possessing the services of an ex-merchant navy telegraphist who
could receive Morse from my computer much faster than he could write it down. He also sent
immaculate Morse but he disagreed with the above structural rules — even though the Morse
which he keyed sounded very much like the computer-generated code. Alas not all those around
today send what they intend.

8.11.3 USING MORSE CODE

Because the technique of sending messages letter by letter is laborious a range of accepted
abbreviations has come into use and now receive official acceptance. These are expected to be
used in the official Morse Test. It is important to remember however that, as with any system of
coding, unless the standard and accepted codes are used it is unlikely that a message can be
deciphered.

The so-called Q-code is another useful device but it is both misunderstood and greatly abused in
the amateur world. Radio-communication, with its ability to span the World, quickly ran into
trouble with language barriers. The Q-code is an internationally-agreed collection of three-letter
symbols to which have been allotted specific meanings; thus the letters QTH are concerned with
location — when sent it is a request that the receiving station state its location while, in the reply
,it means "... my location is...".

Each group begins with the very distinctive letter Q (here comes the bride) and so is
immediately recognisable . Clearly , in an emergency , it is possible to pass all relevant
information quickly and accurately without need for interpreters. The amateur world has
hijacked this code to its own purposes; of itself this is not necessarily a sin but the practice now
borders on farce. For example QRP, which means either "please reduce power" or "I have
reduced power" or perhaps even "shall I reduce power" is used as an adjective in reference to
low-power stations which operate with less than 5-watts.

The practice becomes ludicrous however when R/T operators use it in voice communications. I
have seen/heard few more silly exhibitions than a grown man screaming into a microphone
"QRZ. QRZ. QRZ. Did somebody call me? QRZ. Come in unknown caller." When there is a
deal of interference with the channel it is important to speak slowly and clearly and to repeat
everything that is said; under these conditions the contact is not at all likely to read a gabbled
statement that "thes a lotta QRmmm about". Morse can deliver even when voice contact is
totally destroyed and the letters QRM indicate that the message is being obscured by interfering
signals while QRN indicate that the interference is naturally-caused "static". But exactly what
the spoken "QRN" conveys when voice contact is being denied is a matter of whimsy.

The ability to cope with these antics is required in the Morse test.

To gain an amateur Morse certificate it is necessary to demonstrate an ability to send and to

receive Morse at an average 12 w.p. minute but this does not mean that, in practice, neither
slower nor faster speeds can be used. There is a golden rule however that messages should never
be sent at a speed greater than the operator is able to receive. There used to be a well-meant
injunction never to send to a Royal-navy vessel nor to a Royal Air Force plane at your
maximum speed... they can always send back at a greater speed.

A slow but accurate message will get through first time and so save time and effort. In
conditions of bad propagation/interference send slowly and repeat each word. It is bad manners
to respond to a slow message with a burst of high-speed Morse — it is also self-defeating .
Additionally it might be a beginner who could dispense with a bout of discouragement.

You will find that one of the attributes highly prized in the amateur world is Manners. Behave
badly and you will be ignored. Everyone can hear your transmissions but they are not obliged
to reply.
(I am reminded of an anecdote from an elderly Operator late one night on-air. He used to join a
Net conducted by an old boy in the Midlands who would not tolerate the ill-mannered
microphone hog. Whenever anyone held the channel for more than 3-minutes he would come
on and say "Down five". The entire Net tuned down 5 kHz and left the offender nattering to
himself alone).

8.11.4 THE MORSE CODE

A di-dah N dah-di
B dah—diddy—di O dah—dah—dah
C dah—di—dah—di P di—dah—dah—di
D dah—diddy Q dah—dah—di—dah
E di R di--dah—di
F diddy—dah—di S diddy—di
G dah—dah—di T dah
H diddy—diddy U diddy—dah
I diddy V diddy—di-dah
di—dah—dah—
J W di-dah-dah
dah
K dah—di—dah X dah—diddy—dah
L di—dah—diddy Y dah—di—dah—dah
M dah—dah Z dah—dah—diddy

[Link] Pairs

A & N B & J B & D G & W E, I, S, H

P & L K & R P & R P & X T, M, O

Numerals are simple and easy to distinguish because they are unique in composition. Each has
five digits and they progress, in order, from one to five dis and then from one to five dah's. Zero
( five dahs ) can be included as the fourth in the family of dah's shown above with the letters T,
M & O.

Back to Top of Page

END OF INFORMATION SHEET 11

 PART 10: PRACTICAL PROJECTS
SOLDERING

Soldering is one method of joining two suitable metals and, in electronics, it is used mainly to
ensure a good electrical connection. The process is similar to G1ueing which enables two
dissimilar materials (e.g. fabric and wood) to be joined using a third material known as an
adhesive.

The most important methods of metal-joining are:

a) Welding. In this two similar metals can be joined by heating them to near-melting and then
either pressing/hammering them together or by running-in some of the same metal in the molten
state.

The heated metal combines with the Oxygen in the air (it is said to oxidise and the resulting
oxide forms a crust or scale which has to be chipped away. If such scale is allowed to form
within the joint then failure is to be expected. Thus welding equipment sometimes floods the
joint with an inert gas during welding to exclude air.

(b) Brazing. This method joins two metal pieces by encasing tile Contact line with Brass filler
— hence the name. The filler/joiner is known as a brazing solder which is an alloy of Copper,
Zinc, Silver and/or Nickel in which the exact combination is determined by the intended use.
The inclusion of Zinc has lead to this material acquiring the generic name of Spe1ter which
otherwise is the name used for commercial grades of Zinc.

The two metals are not themselves joined and so they may be different metals. It is necessary
however that the two metals end the grade of brass have similar expansion properties with
temperature else the join must crack as the piece cools.

However there is slightly more to this process than just covering the joint with brass. The spelter
dissolves into the surfaces of the two metal-pieces to be joined in the same manner as glue is
required to penetrate if a joint is to to be successful.

Oxidation problems ore dealt with by using a Flux; this melts at about the same temperature as
the brass spelter and its job is to chemically remove the oxidation products mostly by rendering
them fusible.

Brazing is included under the title "Hard soldering" which is described below.

(c) Soldering. This term usually is applied to joining two metals by use of a solder. Solders
are alloys which come in two main types (i) hard solders (strictly these include brazing solders)
which are alloys of Copper, Tin and Silver and (ii) soft solders which are alloys of Lead and
Tin.

(i) Hard soldering is perhaps used mainly in the jewel1ery trade but it does have a place in
joining Copper parts where strength is required rather then a good electrical connection. The
presence of Silver in the solder has given the process the name Silver Soldering. The melting
point of the solder is sufficiently high to necessitate taking the work piece to red-heat and often
close to melting point and so do not try such jobs on valued pieces unless you have first
acquired the skills and experience.

A flux is required and this is usually Borax which fuses at the temperature used in hard
soldering.

Unless the work piece is large there is hazard in supplying sufficient heat because of a risk of
complete melt-down. (I have witnessed such an embarrassing disaster during an attempt to
solder new lugs on to a silver watch-case. For fine work a blow-pipe should be used to produce
a fine jet of hot gases; a small tube with a fine hole is blown by mouth through a larger flame
(such as that from a Bunsen Burner). It may be necessary to provide auxiliary heating to bring
the job close to the operating temperature so enabling the blow-pipe jet to operate over a very-
small area.

WARNING Take care not to breathe-in through the blow-pipe.

The two surfaces are cleaned, smeared with the Flux and fastened together with some kind of
clamping (often improvised). Slivers of solder are cut and laid in/on the joint; experience is
required here so that the correct amount of solder is used. Heat is applied and as the appropriate
temperature is reached so the solder will be observed to melt and then be drawn into the joint by
capillary action; remove the heat source immediately . When carried out correctly a silver-
soldered joint can be totally invisible.

(ii) The most common use of soft soldering is in electrical and electronic work, where the
requirement is to make good electrical connections. It is foolish to rely on the solder to also
provide a mechanical joint because constantly stressing a soldered joint can lead to fracturing of
the solder and so to bad electrical contact.

Soft soldering takes place at temperatures well below the melting point of the piece to be
soldered and so it is necessary to ensure that the solder dissolves into the surfaces of the two
materials to be joined. This requires a preliminary process known as tinning; the end of a wire
is heated with a soldering iron and treated with flux+solder until a clean coating is achieved.
When both components of a joint have been suitably tinned it is necessary only to place then in
contact, heat them with the soldering-iron and apply a very-small quantity of fresh solder; as
soon as solder is seen to run into the joint the heat should be removed.

It is important not to overheat solder because this oxidises the Tin content; reducing the
percentage of Tin ultimately must reduce the supposed solder to little more than a blob of lead.

Again flux is necessary and this is usually a resin. For electronic work the solder is provided in
the form of a drawn wire and the flux is formed within the wire (like seaside rock). Fine-gauge
solder may have two such cores but the heavier gauges may have as many as five flux cores.

There are also special fluxes; one of these has an acid base and can be useful for tinning badly
oxidised or corroded surfaces although care must be exercised to ensure that the corrosive action
is terminated after the soldering is finished. Other fluxes have been formulated to assist in
soldering difficult materials such as Iron and Aluminium.
(A useful technique for attaching electrical connections to Aluminium is to first melt a small
pool of solder on to the surface and then to scratch that surface with the soldering-iron bit under
the pool until tinning results. The idea is that the molten solder protects the metal from the air -
Aluminium oxidises readily at temperatures below the melting-point of solder. Once tinning has
been accomplished remove most or the solder and add fresh; the above protracted operation will
have reduced the Tin content.)

A standard and constant hazard in electronic work is the so-called "dry joint"; this offers itself
in forms ranging all the way from an open-circuit to an intermittent good contact. Its causes are
many but, regrettably, the most common factor is lack of skill or attention on the part of the
Operator who performs the original task.

A frequent path to producing a dry joint is the use of a too-hot soldering iron; as pointed out
above this results in "burning-out” the Tin content. Tin is included to make the solder fuse
properly at a temperature that will not cause damage either to the electronic components or to
today’s circuit boards. Lead will not serve to join copper wires and, although the resulting join
may look good, it be guaranteed to cause grief.

Another frequent cause is failure to clean the two surfaces properly before tinning — or indeed
failure to tin the surfaces before soldering. A layer of dirt and/or oxide within the joint allows
that joint to be peeled apart . Often an apparently soldered wire call be seen to rotate inside the
joint.

Sometimes, after several years of service, equipment may develop intermittent contacts in the
form of recurring crackles, hisses and temporary loss of gain; such are the result of dry joints
which appear to arise purely from the passage of time. It does not happen in all equipment and
so the inevitable conclusion in that the starting point is inadequate workmanship during
construction.

DO NOT TRY TO REMOVE DRY JOINTS BY RE-SOLDERING EVERYTHING IN SIGHT.

Most likely the problem will be removed but failure indicates that (a) yet more dry joints have
been created (b) not all dry-joints were cured (c) it is now impossible to tell which joints are
good and which need investigation.

Finding/removing dry joints is a slow and painfu1 process and it is more than worthwhile to take
care at the start. Each joint must be tapped gently (both on it and around it to discover where the
equipment is most sensitive; each joint that is diagnosed as faulty should be resoldered and, if
necessary, marked. Re-soldering a joint may not remove the intermittent problem but that does
not indicate a faulty diagnosis. Where there is one faulty joint it is more than probable that
others exist. Sometimes it is necessary to withdraw a dry wire, clean and tin it before replacing
it and re-soldering it; where solder is covering a dirty connection but is not making contact the
hopeful application of lots of heat, lots of flux and lots of solder is guaranteed to damage the
equipment while being unlikely to cure the fault or, at best, to achieve only a temporary cure.
(My personal record in dry-joint removal is the certain cure of 32 dry joints in one elderly CD
player and it was a frustrating labour of love. The cure took two days but three years have now
passed without further problems.)

Solders come in various grades:

Low melting-point (LMP) solder has a composition 62% Tin, 36% Lead, 2% Silver and it melts
at 179° C. It is useful for making joints where other local joints are not to be disturbed.

High Melting-point (HMP) solder has a composition 5% Tin, 93.5% Lead, 1.5% Silver and it
melts at 296° C. It is useful for making joints where other work is to be undertaken using LMP
solder.

6O/4O Standard Solder has a composition 60% Tin, 4O% Lead; this starts to melt around 183°
C and is fully molten at l88° C.

40/60 Solder has the composition 4O% Tin, 6O% Lead; it starts to melt around 183° C and is
fully molten at 234° C. It is used for general electrical repairs and for soldering/repairing copper
and brass metalware.

"Savbit" Solder contains a small quantity of Copper the object being to reduce the leaching of
Copper from a soldering-iron bit so as to extend bit life. It also enables fine wires to be soldered
more easily.

An Aluminium Solder is also available with the composition 18% Tin, 8O% Lead, 2% Silver.
This starts to melt at 178° C and is fully molten at 27O° C. It will joint Aluminium, Tin plate,
Brass, Nickel, Copper and Stainless Steel.

A Lead-free Solder is also available which is 99% Tin and a very small amount of Copper
which melts at 227° C.

Plumbers' Solder is 99.5% Tin and the rest Copper for use in joining Copper pipes in potable
water supplies. Melts at 228° C.

SOLDERING IRONS

Originally solder work was performed using “solder bolts" which consisted of a shaped lump of
copper riveted to an extension piece with a wooden handle. Such bolts still have occasional
application today; they have to be repeatedly heated by means of an external source.

Modern irons are heated by means of an internal electric element and are tipped with variety of
bits whose size and shape depend on the job for which they are intended. Usually the bits are
interchangeable/replaceable. The physical-size of each iron arid its electrical power dissipation
is determined by its intended use; they vary from around 15-watts to 120-watts.

Some irons are provided with built-in temperature-control systems but others depend on an
external control-systems linked to some form of heat-sink-cum-iron-holder which senses the bit
temperature . Clearly the most useful iron is the one in which a degree of control is provided
over the bit operating temperature. The cheapest range completely lack any kind of control and,
when not use, these should be either switched off or kept in contact with a suitable heat-sink to
dissipate excess heat; this avoids destruction of the copper bit and of over-heating soldered
joints.

A simple and effective system can be constructed by feeding the Iron via a simmerstat which
are to be found in abundance on old discarded cookers. These switch the power on and off in a
time-cycle determined by the setting or a small auxiliary heater with a bimetal-strip switch.
The usual soldering iron bit is made of copper but these suffer the disadvantage that they burn
away partly through oxidation in the air and partly because copper is leached from them into
each joint that is made. As a result the end of the bit becomes pitted and needs to be re-shaped
with a File — followed by re-tinning. Their great advantage is ease in use and quick heating.

Bits are made also of iron and these are more resistant to oxidation and leaching and also
(because of the greater “specific heat” of iron,) they can store greater amount of heat. As a
consequence of course they take longer to reach operating temperature and, although they pack a
greater punch they more readily damage circuit boards. Iron bits can be difficult to tin and
usually they are provided with a special coating which should not be treated with a file. Special
solders//fluxes are available for use in keeping these bits in good condition.

DE-SOLDERING

In repair work there is seldom a problem in removing components either for testing or for
replacement but it is important not to apply excessive heat to printed-wiring boards or to strip-
boards; overheating causes copper strip to part from the substrate. When this occurs a repair is
made using normal wiring techniques.

Multi-pin integrated circuits (I.C's or “chips") pose a problem however in that it is difficult to
melt the solder and withdraw all the pins simultaneously. So-called “de-soldering techniques
have been developed to enable solder to be removed from the pins often leaving them
completely free.

Solder Wick
This is a finely-woven copper braid which is impregnated with a flux and it soaks-up molten
solder. It is supplied in various sizes it roll form; it is a sacrificial tool.

Hand Suction-Tool
These are spring-operated suction pumps with a heat-resistant nozzle. With the spring set the
nozzle is held against the soldered joint while an iron is applied also to melt the solder; a button
is pressed to release the plunger when the molten solder is drawn into the body of the tool.

Soldering Station
These are grand and expensive pieces of equipment which offer both a temperature-controllable
soldering iron and a power-operated de-soldering tool. Continuous suction is provided by a
small pump that is operated from a foot switch.

De-soldering irons often provide a special head with which to de-solder all pins of a chip
simultaneously. However, in confined spaces, crowded circuit boards or where the Operator
lacks experience/expertise it is safer to remove a faulty chip by snipping the leads after which
the pins can be removed one at a time if necessary with the aid of a de-soldering tool. (Be
certain that the chip is defunct however !)

A de-soldering tool is useful when clearing old solder from a board in preparation for
replacement work.

Wires
There can be a problem, especially for the inexperienced, when soldering insulated wires into
place; the plastic insulation responds to the soldering iron by melting. The obvious cure is not
to allow the wire to get too hot and this involves speed; heat flows from the soldered joint back
along the wire. In turn this requires proper preparation so that the final application of heat is as
short as possible.

There are two main results of overheating insulated wire:

(i) Many plastic materials exhibit a "memory effect” in that deformation-strains are relieved by
heat-softening and the material returns to the shape it held before it was deformed'.

Removal of the insulation from the end of a wire usually results in that insulation being
stretched; when the wire is heated the insulation contracts again and a length of bare wire
results next to the soldered joint.

The best cure for this fault is speed in the soldering operation but after baring the end of a wire,
tin it first (this will shrink the insulation) and then cut it to the required length before soldering
is attempted.

(ii) Where a wire has been bent the insulation is stretched around the outer circumference; when
the insulation softens this stretch is relieved and the wire penetrates the covering on the bend.
Where possible solder the wire into place and then bend it.

Back to Top of Page

END