Cathode Ray Oscilloscope (CRO)
I
21.1 INTRODUCTION
The cathode ray oscilloscope (CRO) is a very
useful and versatile laboratory instrument used for
display, measurement and analysis of waveforms and
other phenomena in electrical and electronic circuits.
CROs are in fact very fast X-Y plotters, displaying an
input signal versus another signal or versus time. The
"stylus" of this "plotter" is a luminous spot which
moves over the display area in response to an input
voltage. The luminous spot is produced by a beam of
electrons striking a fluorescent screen. The extremely
low inertia effects associated with a beam of electrons
enables such a beam to be used for following the
changes in instantaneous values of rapidly varying
voltages. The extremely low inertia of electrons as
compared to metallic conductors used in an
electro-mechanical Duddell's oscilloscope enables a
CRO to be used at frequencies much above the highest
on which the Duddell's oscilloscope can be used.
The normal form of a CRO uses a horizontal input
voltage which is an internally generated ramp voltage
called "Time Base". This horizontal voltage moves the
luminous spot periodically in a horizontal direction
from left to right over the display area or screen. The
vertical input to the CRO is the voltage under
investigation. The vertical input voltage moves the
luminous spot up and down in accordance with the
instantaneous value of the voltage. The luminous spot
thus traces the waveform of the input voltage with
respect to time. When the input voltage repeats itself
at a fast rate, the trace (display) on the screen appears
stationary on the screen. The CRO thus provides a
means of visualizing time varying voltages. As such,
the CRO has become a universal tool in all kinds of
electrical and electronic investigations.
Cathode Ray Oscilloscopes operate on voltages.
However, it is possible to convert current, strain,
acceleration, pressure and other physical quantities
(641)
into voltages with the help of transducers and thus to
present visual representations of a wide variety of
dynamic phenomena on CROs.
Cathode Ray Oscilloscopes are also used to
investigate waveforms, transient phenomena, and
other time varying quantities from a very low
frequency range to the radio frequencies.
Oscilloscopes have been evolved continuously,
and they are now available which can measure
frequencies upto 1 GHz, and observer events as small
as 20 Hz in duration.
Many additional features are available with some
oscilloscopes and these include built in digital multi­
meters and counters. The oscilloscopes are progres­
sively getting smarter, and many are microprocessor
controlled. They have the ability to calculate several
features, such as rise time or pulse width of the mea­
sured waveform, and to display these values along
with the display of waveforms. They are easier to use
and internal routines often act as a guide for the user,
and display a warning if there is any error in setting.
Many oscilloscopes are now available with IEEE
488 bus capabilities, so that they can be used as a part
of measurement test bed, with the instrument's
controls set at a remote location, and the readings
digitised and retrieved for the purposes of recording
and analysis.
Although, most oscilloscopes are monochromatic,
colour oscilloscopes are finding increasing applications
in computers and in television.
The applications of oscilloscopes have been
enhanced on account of many recent developments.
Most present day oscilloscopes are capable of
accepting two or more inputs displaying them
simultaneously. This may be achieved through using a
split beam or by using a multiple beam tube. Sampling
oscilloscopes are used for high speed applications.
 642 Electrical and Electronic Measurements and Instrumentation

These oscilloscopes employ time sampling and After leaving the electron gun, the electron beam
through their use it is possible to measure signals of passes through two pairs of "Electrostatic deflection
about 20 GHz. They can only detect a repetitive plates". Voltages applied to these plates deflect the
waveform, and work on the principle of taking a beam. Voltages applied to one pair of plates move the
sample once every cycle, over several cycles, each beam vertically up and down and the voltages applied
sample point being shifted from the previous point. to the other pair of plates move the beam horizontally
The complete picture of the waveform is stored, and form one side to another. These two movements i.e.,
can be displayed as a stationary signal. horizontal and vertical are independent of each other
Storage oscilloscopes can be used for capturing and thus the beam may be positioned anywhere on the
transient signals, and then display them for periods screen.
which may vary from a few minutes to several years. The working parts of a CRT are enclosed in an
An analog oscilloscope uses a modified form of a evacuated glass envelope so that the emitted electrons
conventional cathode ray tube to store the trace. The are able to move about freely from one end of the tube
digital storage oscilloscope first converts the analog to the other.
signal to a digital form and stores it in digital memory.
The signal can then be recalled for display as and 21.3 ELECTRON GUN
when required. The source of focused and accelerated electron
beam is the electron gun. The electron gun, which
21.2 CATHODE RAY TUBE (CRT)
emits electrons and forms them into a beam consists of
A cathode ray oscilloscope consists of a cathode a heater, a cathode, a grid, a pre-accelerating anode, a
ray tube (CRT), which is the heart of the tube, and focusing anode and an accelerating anode.
some additional-circuitry to operate the "CRT. The
In smaller CRTs, connections to the various
main parts of a CRT are :
electrodes are brought out through pins in the base or
(z) Electron gun assembly, the tube as shown in Fig. 21.1. Large and medium
(zz) Deflection plate assembly, sized high performance tubes operate at very high
(z'n) Fluorescent screen, voltages, and these leads are usually brought out
through the sides of the glass envelope.
(iv) Glass envelope,
Electrons are emitted from the indirectly heated
(v) Base, through which connections are made
cathode. A layer of barium and strontium oxide is
to various parts.
deposited on the end of the cathode-which is a
The main parts of a CRT are shown in Fig. 21.1. cylinder - to obtain high emission of electrons, at
Before going into details of working of various parts of moderate temperatures. The typical values of current
a CRT, a summary of functions of the different parts is and voltage required by an indirectedly heated
given below : cathode are 600 mA at 6.3 V. High efficiency systems
The “Electron gun assembly" produces a sharply use 300 mA at 6.3 V. The special low power designs
focused beam of electrons which are accelerated to use 140 mA at 1.5 V. These electrons pass through a
high velocity. This focused beam of electrons strikes small hole in the "control grid". This control grid is
the fluorescent screen with sufficient energy to cause a usually a nickel cylinder, with a centrally located hole,
luminous spot on the screen. co-axial with the CRT axis. This is usually a metal cup
of low permeability steel, about 15 mm in diameter
Electron gun Aquadag and 15 mm long. An aperture of about 0.25 mm is
-Vertical coating
Pre-accelerating deflection drilled in the cap of the grid for the electrons to flow
anode plates through. The intensity of electron beam depends upon
E ectron the number of electrons emitted from the cathode. The
beam grid with its negative bias controls the number of
electrons emitted from the cathode and hence the
Grid A >
intensity is controlled by the grid.
Cathode Accelerating Horizontal
The electrons, emitted from the cathode and
anode deflection
plates phosphor passing through the hole in the control grid are
screen accelerated by the high positive potential which is
applied to the "pre-accelerating" and "accelerating
Fig. 21.1 Internal structure of a CRT. anodes".
Cathode Ray Oscilloscope (CRO) 643

The electron beam is focused by the "focusing Figure 21.3 shows two concentric cylinders with a
anode". The accelerating and focusing anodes are potential applied between them. Lateral repulsion
cylindrical in form, with small openings located in the again causes the spreading of the flux lines producing
centre of each electrode, coaxial with the tube axis. a field as shown. The equipotential surfaces are shown
After leaving the focusing anodes, the electron beam as solid lines. It is clear from the diagram that the
passes through the vertical and horizontal deflection
plates and then goes on to the fluorescent screen.
The pre-accelerating anode and the accelerating
anode are connected to a common positive high
voltage of about 1500 V. The focusing anode is
connected to a lower adjustable voltage of 500 V.
There are two methods of focusing an electron
beam :
(i) Electrostatic focusing and ■smmbbhim
Fig. 21.3 Field between two co-axial cylinders.
(ii) Electromagnetic focusing.
The CRO uses electrostatic method of focusing as equipotential surfaces are curved. Let us consider the
compared to a TV picture tube which employees region on the two sides of an equipotential surface S as
electromagnetic focusing. shown in Fig. 21.4. The potential on the left side of the
surface is -V and on the right \ side is + V. Let an
21.4 ELECTROSTATIC FOCUSING electron moving in a direction AB enter the area to the
Figure 21.2 shows an electron at rest placed in an left of S. This electron experiences a force which is
electric field produced two parallel plates. Force on normal to the surface S and is thus accelerated.
the electron is,
F = -ee newton ...(21.1)
where e = electric field intensity ; V/m,
and e = charge of electron =1.602 x 10-19 C
The minus sign indicates that the force acts in the
opposite direction to that of the field. The above
discussion is valid only if the electron is situated in a
field of uniform intensity. In practice, however, the
field is not uniform. The lateral repulsion of the
electric field lines causes a spreading of space between
the lines, resulting in curved field lines at the ends. Since the force acts in a direction normal to the
Thus the field intensity will be less at the ends. surface, it is the normal component of velocity that is
Figure 21.2 also shows equipotential surfaces, increased after refraction while the tangential
indicated by solid lines. Since the force is in a direction component remains the same.
opposite the field and the equipotential surfaces are The tangential components are :
perpendicular to the field, the force on an electron is in
yfl = sin0; and vf2 = i?2sin0^
a direction normal to the equipotential surfaces.
Now vtl = vf2 or v1 sin0. = v2sin0r
sin0. _ u2
...(21.2)
sin0r V1
where
= initial velocity of electrons,
v2 = velocity of electrons after leaving surface S,
0;. = angle of incidence, and
0r = angle of refraction.
Equation 21.2 is identical to the expression relating
Fig. 21.2 Electric field between parallel plates. the refraction of a light beam in geometrical optics.
 644 Electrical and Electronic Measurements and Instrumentation

The refraction of an electron beam follows as the direction. There is no force either in X direction or Z
bending of a light beam at a refracting surface such as direction and hence there will be no acceleration of
an optical lens. For this reason the focusing system in a electrons in these directions.
CRT is known as electron lens.
Figure 21.5 shows the functional diagram of an
electrostatic focusing arrangement. The pre-accelera­
ting anode, which is a metal cylinder containing many
baffles, collimates the electron beam which enters it
through a small opening on the left hand side. The
pre-accelerating anode is connected to a high positive
potential.

Preaccelerating Focusing Accelerating

Let
Eo = voltage of pre-accelerating anode ; V,
e = charge of an electron ; C,
//i = mass of electron ; kg,
vQX = velocity of electron when entering the field
of deflecting plates ; m/s,
Ed = potential between deflecting plates ; V,
Fig. 21.5 Electrostatic focusing arrangement. d = distance between deflecting plates ; m,
ld = length of deflecting plates ; m,
The focusing anode and the accelerating anodes
L = distance between screen and the centre of the
are co-axial with the pre-accelerating anode. The pre­
deflecting plates ; m, and
accelerating and accelerating anodes are connected to
the same potential while the focusing anode is D = deflection of electron beam on the screen in
connected to a lower potential. Y direction ; m.
On account of difference of potential between The loss of potential energy (P.E.) when the
focusing anode and the two accelerating anodes a electron moves from cathode to accelerating anode ;
non-uniform field exists on each of the two ends of the P.E. =eEn ...(21.3)
focusing anode. The equipotential surfaces, thus, form The gain in kinetic energy (K.E.) by an electron
a "double concave lens". K.E. =(l/2)nw2x ...(21.4)
The electron beams entering the field at angles
(The mass of an electron is m = 9.109 x ICT31 kg)
other than the normal to the equipotential surfaces,
Equating the two energies, we have :
will be deflected towards the normal and the beam is
thus focused towards the centre of the tube axis. By =(2<-E„/m)1/2 -(215)
changing the voltage of the focusing anode, the This is the velocity of the electron in the X
refractive index of the electron lens is changed and direction when it enters the deflecting plates. The
therefore the focal point of the beam can be changed. velocity in the X direction remains the same through­
The change in voltage is brought about by changing out the passage of electrons through the deflecting
the setting of a potentiometer. This control is brought
plates as there is no force acting in this direction.
to the front panel of CRO and is marked Focus. The electric field intensity in the Y direction,
^ = Ed/d ...(21.6)
21.5 ELECTROSTATIC DEFLECTION Force acting on an electron in Y direction
Figure 21.6 shows a general arrangement for Fy=e£y =eEd/d ...(21.7)
electrostatic deflection. There are two parallel plates
Suppose fly is the acceleration of the electron in Y
with a potential applied between. These plates
direction therefore,
produce a uniform electrostatic field in the Y direction.
Thus any electron entering the field will experience a F, =
force in the Y direction and will be accelerated in that or ay=eey!m ...(21.8)
Cathode Ray Oscilloscope (CRO) 645

As there is no initial velocity in the Y direction the usually a time varying quantity and the
displacement 1/ at any instant t in the Y direction is : image on the screen thus follows the varia­
tions of the deflection voltage in a linear
...(21.9) manner.
' 2 -,/ 2 m
As the velocity in X direction is constant, the (iii) The deflection is independent of the e/tn
displacement in X direction is given by : ratio. In a cathode ray tube, in addition to the
electrons many types of negative ions such
x= ...(21.10) as oxygen, carbon, chlorine etc. are present.
f = x/v,„ ...(21.11) With electrostatic deflection system, because
Substituting the above value of t in Eqn. 21.9, we deflection is independent of e/m, the ions
have : travel with the electrons and are not
11 ee 1/ concentrated at one point. Hence cathode
2
y = tx ...(21.12) ray tube with electrostatic deflection system
2 "Wox
does not produce an ion burn.
This is the equation of a parabola. Deflection sensitivity. The deflection sensitivity
The slope at any point (x,y) is of a CRT is defined as the deflection of the screen per
dy _ gev unit deflection voltage.
...(21.13) Deflection sensitivity
dx "rfc

Putting x = ld in Eqn. 21.13, we get the value of S = — = —m/V ...(21.17)

£,a 2dEna
tan0.
ee„ eF I The Deflection Factor of a CRT is defined as the
tan0 =—...(21.14) reciprocal of sensitivity.
mvox mdvox
Deflection factor
After leaving the deflection plates, the electrons 1 2dE
travel in a straight line. The straight line of travel of G = — =---- - V/m ...(21.18)
S Lld
electrons is tangent to the parabola at x = ld and this
tangent intersects the X-axis at point O'. The location It is clear from Eqn. 21.17, that the sensitivity can
of this point is given by : be increased by decreasing the value of accelerating
x = _y_ = ^L /JY, voltage Ea. But this has a disadvantage as the
tan 9 2zn^r/ mv2
gx d 2 luminosity of the spot is decreased with decrease in
E . On the other hand a high value of £ , produces a
The apparent origin is thus at the centre of highly accelerated beam and thus produces a bright
deflection plates. The deflection D on the screen is spot.
given by : However, a high accelerating voltage (£;)
Le E, I.
D=£tan0=------4/ ...(21.15) requires a high deflection potential (£rf) for a given
mdvox deflection. Also highly accelerated beam is more
2 2eE difficult to deflect and is sometimes called hard beam.
Substituting the value vox =----- - in Eqn. 21.15, we
m Typical values of sensitivities are 0.1 mm/V to
get 1.0 mm/V corresponding to deflection factors of
LeE.l, in LI, E, 10 V/mm and 1 V/mm.
D =----- = ...(21.16)
md 2eEan 2dEna
21.6 POST DEFLECTION ACCELERATION
From Eqn. 21.16, we conclude : OF ELECTRON BEAM
(i) For a given accelerating voltage Ea, and for After electrons pass beyond the deflection plates,
particular dimensions of CRT, the deflection they may or may not experience additional
of the electron beam is directly proportional acceleration. This depends primarily upon the
to the deflecting voltage. This means that the maximum frequencies to be applied to CRT. In an
CRT may be used as a linear indicating electrostatic system of deflection, Eqn. 21.17 indicates
device. that a low value of accelerating voltage, £fl, should be
(n) The discussion above assumes that Ed is a used. This voltage is usually kept below 4 kV.
fixed d.c. voltage. The deflection voltage is Although this voltage gives good sensitivity but
 646 Electrical and Electronic Measurements and Instrumentation
reduces brightness, which can be seriously impaired
at high frequencies. In general, mono-accelerator
tubes such as shown in Fig. 21.1 may be used, if the
maximum frequency to be displayed is below 10 MHz.
However, if signals of frequencies higher than 10 MHz
are to displayed, post deflection acceleration is necessary
to increase the brightness of the trace which otherwise
would be dim.
When measurements are to be made above a
frequency of 10 MHz, post-accelerators or post deflection
acceleration tubes (PDA) are used. These have a large
beam accelerating bias after the deflection plates so as
to increase the beam energy and give a bright display
on the screen.
Several types of PDA structures can be used.
Figure 21.7(a) shows a spiral accelerator in which a
high resistance narrow spiral of graphite is painted
over a considerable length of the inside of the envelope
funnel. The spiral is connected to the aluminium film
on the phosphor, if present. A voltage of about 10 kV
is applied to this spiral, and the effect is to accelerate
the beam, but also to bend it slightly towards the
X-axis. This decreases the deflection sensitivity, which
can be compensated for by increasing the length of the
tube. Alternatively spherical mesh may be inserted
into the helix tube, as in Fig. 21.7(b). This shapes the
accelera- ting field and prevents it from affecting the
original beam deflection, so giving the same
sensitivity as a mono-accelerator tube of equal length.
Fig. 21.7 CRT post accelerators : (a) spiral (b) mesh.
Both the systems illustrated in Fig. 21.7 are
limited in scan angle to between 35° and 40°. An
alternative PDA system, called a high expansion mesh
tube, uses a spherical shaped mesh without a helix
tube. The mesh is shaped so as to expand the
deflection of the beam, and with this system beam
deflections up to 90° be obtained. However the spot
size is also increased in proportion to the increase in
sensitivity.
21.7 EFFECT OF BEAM TRANSIT TIME
AND FREQUENCY LIMITATIONS
In the derivation of Eqn. 21.16, the plate voltage
Ed is assumed constant during the motion of the
electrons through the deflecting field. If the voltage
applied to the vertical deflecting plates changes during
the transit time of the electrons through the hori­
zontal plates, the deflection sensitivity gets decreased.
Transit time f, =—— ...(21.19)
1 vOX
The transit time imposes a limitation of the upper
frequency limit. An upper limiting frequency is
defined as that frequency at which the transit time is
equal to one quarter of the period of the voltage
applied to vertical plates.
Upper limiting frequency,
1 v
f ...(21.20)
Jc 4fT 4Z
The frequency range of the oscilloscopes can be
increased by subdividing the deflecting plates in a
number of sections in the path of the electron beam.
The voltage being measured is applied to the vertical
plates through an iterative network, whose propagation
time corresponds to the velocity of electrons, thereby
the voltage applied to the vertical plates is made to
synchronize with the velocity of the beam. The use of
this technique allows the cathode ray oscilloscope to
be used upto frequencies of 500 MHz and above.
21.8 DEFLECTION PLATES
Referring to Fig. 20.1, we observe that the electron
beam, after leaving the electron gun, passes through
two pairs of deflection plates. One pair of plates is
mounted horizontally and produces an electric field in
the vertical plane. This pair produces a vertical
deflection and is thus called Vertical Deflection
Plates or Y Plates. The other pair of plates is mounted
vertically and produces a horizontal deflection. This
pair of plates is called Horizontal Deflection Plates or
X Plates. The plates are flared so as to allow the beam
to pass through them without striking the plates.
Cathode Ray Oscilloscope (CRO) 647
21.9 SCREEN FOR CRTs
The front of the CRT is called the face plate. It is
flat for screen sizes up to about 100 mm x 100 mm, and
is slightly curved for larger displays. The face plate is
formed by pressing molten glass in a mould and then
annealing it. Some CRTs have a face plate which is made
entirely from fibre optics, which has special charac­
teristics. It is also feasible to have a small fibre optic
region which gives direct contact exposure of photo­
graphic film, if required, without needing a lens system.
The inside surface of the face plate is coated with
phosphor. This consists of very pure inorganic
crystalline phosphor crystals, about 2-3 microns in
diameter, to which traces of other elements, called
activators, have been added. Activators in current use
are metals such as silver, manganese, copper and
chromium, and they affect the characteristics of the
phosphor, such as its luminous efficiency, spectral
emission and persistence.
Many types of phosphors are used for CRTs and
the characteristics of a few of them are given in
Table 20.1. A phosphor converts electrical energy to
light energy. When an electron beam strikes phosphor
crystals it raises, their energy level. This is known as
cathodoluminescence. Light is emitted during phosphor
excitation and this is called fluorescence. When the
electron beam is switched off the phosphor crystals
return to their initial state, and release a quantum of
light energy. This is called phosphorescence or persistence.
The writing speed of a phosphor is measured by
its fluorescence rise time, which is the time from the
beginning of excitation to reach 90% of the maximum
emission state, and by the decay time, which is the
time to fall from the maximum state to the 10% level.
Writing speed is determined by the phosphor type,
crystal size, impurity content, and the manufacturing
process. A phosphor must be refreshed by electrons,
before the end of its decay time, in order to give a
flicker free display. Short persistence phosphors
require more frequent refreshes, whereas long
persistence phosphors may result in characters fading
slowly, giving a trace after glow or trails. Long
persistence phosphors are also more prone to
permanent discolourations, and loss of luminous
efficiency, at high beam current or fixed data use.
Most modem oscilloscopes use short persistence
phosphors (e.g., Pl, P2, Pll, P31) because of their fast
refresh capability, whilst the slower speed signals seen
in medical applications need longer persistence
phosphors (P7 and P39). For use in radars, very long
persistence is needed (Pl 9, P26, P33). Long persistance
is even needed in storage type oscilloscopes.
The electron beam on striking the phosphor gives
off light and heat. Excessive beam current on one spot
of the screen, over a long time, can degrade the light
output from the phosphor, and can bum a hole in it.
Phosphors can be classified as having low burn
resistance (P19, P26, P33), medium bum resistance (Pl,
P2, P4, P7, Pll) and high burn resistance (P15, P31).
The luminance of the phosphor is a measure of its
brightness. It is determined by the luminance effi­
ciency of the phosphor, and by the beam energy,
which is a product of the beam current density, the
accelerating potential and the writing time. Therefore
one can have high light output, at low beam accele­
rating potential provided the sweep speed is low, but
for fast single-shot transient recording, a high beam
current and accelerating voltage, and phosphor with
high luminance, are required.
Each phosphor has a unique colour, the human
eye response peaks at 555 nm, in the yellow-green
region, so phosphors which gives light in these regions
(P2, P31, P39) would appear to be brighter than others
such as Pll, P16, P19. A camera film, however, is
sensitive to ultraviolet or blue, so for high speed
photography phosphors Pll or P16 may be better. The
spectral emission from a phosphor is broadband, i.e., a
range of wavelengths go to determine its colour. The
gradation of colour (hue) and the chromatic purity
with respect to white (saturation) can be precisely
specified by X and Y co-ordinates on a chromaticity
diagram.
A thin film of metal, such as aluminium (0.1 - 0.15
pm) is usually deposited on the non-viewing side the
phosphor.
This has three effects :
(z) The metal layer acts as a heatsink and
reduces the danger of phosphor burn.
(zz) The light scatter from the phosphor is
reduced, and the aluminium reflects light
going back into the tube towards the viewer,
so increasing the brightness.
(zzz) Electrons which strike the phosphor release
secondary electrons. These secondary
electrons are usually collected by the
aquadag. Under certain conditions the
electrons may remain on the screen, and
cause a negative voltage build up. This
would decrease the accelerating voltage of
the beam. This charge build up is prevented
by the aluminium layer, which is connected
electrically to the aquadag layer.
 6^-8 Electrical and Electronic Measurements and Instrumentation

Table 21.1 Phosphor Characteristics

Colour Persistence Lumi­ Writing
Type Intended use
I Fluorescent Phosphorescence (to 10% level) *nance speed

1 P1 yellow-green yellow-green 1 - 100 ms 4 7 oscilloscope, radar

P2 yeliow-green yellow-green 1 - 100 ms 2 4 oscilloscopes

P4 white white 10 gs -1 ms 3 3 monochrome television

radar, medical

P7 blue yellow 10 gs -1 ms(b) 4 2 radar, medical

100 ms - 1 s(y)

Pll blue blue 10 gs -1 ms 5 1 photographic

j P15 ultra-violet green < 1 gs (uv) 9 8 TV (flying spot scanner),

1 - 10 gs (g)
P16 ultra-violet ultra-violet < 1 gs 10 8 TV (flying spot scanner),
photographic

; P18 white white 10 gs -100 ms - - projection TV

: Pi9 orange orange > 1 s 5 10 radar

I P22 white white 10 gs -1 ms - - tricolour TV
(red, blue, green) (red, blue, green)

P26 orange orange > 1 s 8 10 radar

P28 yellow-green yellow-green 100 ms - 1 s 3 5 radar, medical

P31 green green 10 gs -1 ms 1 3 oscilloscopes, bright TV

P33 orange orange > 1 s 6 9 radar

P39 yellow-green yellow-green 100 ms -Is 3 6 radar, computer graphics

| P40 blue yellow-green 10 gs 1 ms(b) — — low repetition rate P16

100 ms - ls(yg)

P44 yellow-green yellow-green 1 - 100 ms 3 - bistable storage

P45 white white 1 - 100 ms 4 - monochrome TV display

♦ Highest luminance or fastest writing speed.

The electron beam needs more energy to 21.10 GRATICULE

penetrate the aluminium layer in front of the
phosphor, fherefore at low acceleration potentials the
aluminised phosphor screen has lower efficiency than
a standard screen, but the efficiency increases rapidly
at higher potentials.
The idea! phosphor for a laboratory oscilloscope
CRT would be one which gives off a colour to which
the eye has maximum response ; has a high bum
resistance, to prevent accidental damage ; has a high
luminance and writing speed; and has short
persistence to avoid multiple images when recording
fast displays. A typical choice of phosphor is P31.
The graticule is a grid of lines that serves as a scale
when making time and amplitude measurements.
There are three types of graticules.
(/) External graticule. This is scribed on a
Plexiglas acrylic plastic and fitted to screen.
This graticule can be easily changed, to make
different types of measurements and its
position can be adjusted to align it with the
trace on CRT. External graticules suffer from
parallax errors as they are not in the plane of
the phosphor.
Cathode Ray Oscilloscope (CRO) 649

(n) Internal graticule. This is deposited on the horizontal axis. The tick marks on the centre graticule
internal surface of the CRT face plate, and is represent minor divisions or subdivisions.
therefore on the same surface as the
phosphor. Although there is now no
parallax, an internal graticule cannot be
changed, and also needs some method of
electrical trace alignment. The graticule is
also difficult to illuminate for photography,
unless special illumination is provided in the
oscilloscope.
(in) Porjected graticule. It is available with some
cameras and provides flexibility which can
include additional features such as legends
on the graticule.
The face plate is sometimes tinted neutral-grey to
reduce ambient light interference, or external filters
can be used to reduce glare. Black wire mesh filters
increase the contrast ratio and restrict the viewing
angle, so reflections from oblique light sources are
minimised. Polarised filters can also be used in high 21.11 AQUADAG
glare applications. Light filters are sometimes utilised The bombarding electrons, striking the screen,
to affect selective colour components of the phosphor. release secondary emission electrons. These secondary
For example, an amber filter on a P7 phosphor would electrons are collected by an aqueous solution of
eliminate the short persistence blue component, and graphite called 'Aquadag' which is connected to the
increase the longer persistence yellow component ; second anode, collection of secondary electrons is
and a blue filter would do the reverse. necessary to keep the CRT screen in a state of electrical
The graticule may be etched or silk-screened on a equilibrium.
plastic CRT face plate or it may be chemically
21.12 COLOUR CRT DISPLAYS
deposited on the face of CRT along with phosphor. If
the graticule is put on the plastic face plate, it should Although colour CRTs, such as shadow mask
be positioned on the inside surface, that is, the surface tubes, are common in television and computer
that will be in contact with the face of CRT. This places displays, they are not used for oscilloscopes. One of
the display produced by the electron beam and the the reasons is that they are relatively expensive, and
graticule on the same plane, thereby eliminating the colour does not give many advantages when viewing
errors on account of parallax. Parallax errors occur single or multiple traces which are separated from
when the trace and the graticule are in different planes each other. However, limited colour can have
and the observer's eye is shifted from the direct line of advantages, such as in medical oscilloscopes, and an
sight as shown in Fig. 21.8. instrument is available for this application. It uses a
standard CRT, and a liquid crystal shutter to produce
a three colour display, as shown in Fig. 21.10.

Faceplate with
graticule
Viewer
Fig. 21.8 Illustration of parallax error.

The graticules are generally laid out in an 8 x 10

pattern as shown in Fig. 21.9 with 8 major divisions Fig. 21.10 Exploded view of a liquid crystal switch
along the vertical axis and 10 major divisions along the used with a CRT.
 650 Electrical and Electronic Measurements and Instrumentation

The CRT has a multi-component phosphor, which

gives peak outputs of oranges and cyan light. Two
colour polarisers are used alter the CRT and these
polarise the orange light vertically, the cyan light
horizontally, and absorb white light. The liquid crystal
switch will not affect the light passing through it when
on, but it will rotate the light through it by 90° when
off. The linear polariser will pass horizontally
Fig. 21.11 Typical sawtooth waveform applied to
polarised light only. the horizontal deflection plates.
The liquid crystal switch is synchronised to the
CRT trace. In operation the CRT first sweeps out the During the sweep time, T, the beam moves from
cyan colour display field with the liquid crystal left to right across the CRT screen. The beam is
switched on. Only cyan light will be displayed since deflected to the right by the increasing amplitude of
this will be passed by the linear polariser. The CRT the ramp voltage and the fact that the positive voltage
then sweeps out the orange coloured portion of the attracts the negative electrons. During the retrace
display with the liquid crystal switched off. Now both time, or flyback time. T, the beam returns quickly to
the cyan and orange lights from the colour polarisers the left side of the screen. The control gird is generally
are rotated, so that the orange is horizontally polarised "gated off", which blanks out the beam during retrace
and is passed by the linear polariser. If any part of the time and prevents an undesirable retrace pattern from
display is to be yellow then it is traced by the CRT, appearing on the screen.
when the liquid crystal switch is both on and off, so Since signals of many different frequencies will
the cyan and orange colours combine to give yellow. be observed with the oscilloscope, the sweep rate must
The liquid crystal shutter oscilloscope has a good be adjustable. We can change the sweep rate in steps
by switching different capacitors into the circuit. The
contrast ratio since the shutter attenuates the non
front panel control for this adjustment is marked
lighted areas giving a black background. However the
time/div or sec/div. The sweep rate can be adjusted in
brightness is low since the light is emitted 50% of the
minor ways by making the resistor, R, in Fig. 21.12(c) a
time for any one colour, due to the duty cycle, and the
variable resistor.
polarisers also attenuate about half the light going
The circuit shown in Fig. 21.12(c) is a simple
through them. Overall, the brightness is less than 20%
sweep circuit in which the capacitor C charges
of that of a CRT without a liquid crystal shutter. The
most serious problem with the liquid crystal shutter
system is that it is limited in switching speed. By
making the liquid crystal switch narrow, and the
rotational distance of the molecules small, the
switching time can be reduced to about 0.5 ms. Even
then the bandwidth of the oscilloscope is limited to
under 100 kHz.

21.13 TIME BASE GENERATORS

Oscilloscopes are generally used to display a
waveform that varies as a function of time. If the
waveform is to be accurately reproduced, the beam
must have a constant horizontal velocity. Since the
beam velocity is a function of the deflecting voltage,
the deflecting voltage must increase linearly with
time. A voltage with this characteristic is called a ramp
voltage. If the voltage decreases rapidly to zero with
the waveform repeatedly reproduced, as shown in
Fig. 21.11, the pattern is generally called a sawtooth Fig. 21.12 Simple sawtooth generator and
waveform. associated waveforms.
Cathode Ray Oscilloscope (CRO) 65 I
through the resistor R. The capacitor discharges
periodically through the transistor, Qv which causes
the waveform shown in Fig. 21.12(b) to appear across
the capacitor. The signal, V{, which must be applied to
the base of the transistor to turn it "on" for short time
intervals is also shown in Fig. 21.12(b). When the
transistor is turned completely "on," it presents a
low-resistance discharge path through which the
capacitor discharges quickly.
If the transistor is not turned "on" the capacitor
will charge exponentially to the supply voltage Vcc
according to the equation,
V0 = VcJl-exp(-f/RC)] ...(21.21)
where v0 = instantaneous voltage across the
capacitor at time t; V,
Vcc = supply voltage ; V,
t = time of interest; s,
R = value of series resistor ; Q,
C = value of capacitor ; F.
Example 21.1 A trigger pulse is applied to the sweep
generator in Fig. 21.13 every 10 ms. Compute the amplitude
of the voltage, V, across the capacitor when the trigger
pulse is applied.
Fig. 21.13 Circuit for Example 21.1
Solution. Using Eq. 21.21 we compute the voltage
as
Vo = Vcc[l-exp(-f/RC)]
= 50[l-exp(10x 10-3)/(500x103x0.2x10“6)]
= 50[l-exp(-0.1)]
= 50(1 -0.905)= 4.76 V.
Since Vo is less than 10% of Vcc, the charge curve
should still be quite linear at this point.
When a trigger pulse 17. is applied to the transistor
Qj in Fig. 21.12 (a), the capacitor C discharges through
the resistance presented by the "on" transistor
according to the equation
v; = vo[exp(-t,/KjC)] ...(21.22)
where
V° = voltage across capacitor C at time t during
retrace ; V,
Vo = amplitude of the voltage across capacitor C
at the start of discharge ; V,
C = value of capacitor C; F,
= time of interest after start of retrace ; s,
R1 = resistance from collector to emitter of the
saturated transistor; Q.
Example 21.2 An input pulse, Vir of 5 ns duration is
applied to the circuit in Fig 21.13 at the instant V, reaches
4.76 V. What is the voltage across the capacitor after 50 ps if
the saturated transistor presents a resistance of 0.2 k£l to the
circuit ?
Solution. Using Eqn. 21.22, we find the voltage
as :
V' = V„e-W
= 4.76exp[-50 x IO-6 / (0.2 x 103 x 0.2 x IO-6)]
= 1.36 V
The relationship between the current charging a
capacitor and the voltage across the capacitor is given
as
V0=^icdt ...(21.23)
c0
If the current is constant, the solution of 21.23 is,
V0 = It/C ...(21.24)
Substituting Vcc / R for I, we have,
Vo=~t
KC
-(21.25)
Equation 21.15 is the equation of a straight line
with a slope of VJRC. The voltage Vo would be a
"linear ramp" if it followed this equation. Figure 21.14
compares the graph of the exponential function of
Eqn. 21.21 to the graph of the linear function of
Eqn. 21.25.
Fig. 21.14 Graph comparing a capacitor charged by
a constant-current source to one charged
by a constant-voltage source.
 652
As can be seen by comparing the linear and
exponential curves, the exponential function is fairly
linear during the first 10% or 15% of the graph. Some
inexpensive oscilloscopes use this type RC, of circuit
as a sweep generator with reasonably satisfactory
results. However, linearity can be improved by using
one of several other types of circuits as a sweep
generator. The circuits most commonly used are :
1. Some form of constant-current generator to
charge a capacitor at a constant rate.
2. A bootstrap circuit in which the current is
kept constant by maintaining a constant
voltage across a charging resistor.
3. An operational amplifier in a circuit known
by one of the following names: integrating
amplifier, Miller integrator, Miller sweep
generator.
The circuit shown in Fig. 21.15(a) utilizes an
ordinary bipolar junction transistor, Qr connected in
the common-base configuration to charge capacitor C
at a constant rate. If the resistor, R has a large value of
resistance, the emitter current will be nearly constant
as will the collector current. Since the collector current
charges the capacitor C, it will charge at a nearly
constant rate. Typical common-base output characteri­
stic curves are shown in Fig. 21.15(b). Note that the
curves are very nearly horizontal in the active region
(a) Circuit to generate a ramp voltage by
charging a capacitor at a constant rate.
(b) The characteristic curves
for transistor Q}.
Fig. 21.15
Electrical and Electronic Measurements and Instrumentation
of the transistor. The collector-to-base voltage VCB can
change considerably without any appreciable effect on
the collector current.
In order to have a repetitive sweep voltage, the
capacitor must be discharged after it has reached its
maximum voltage. This can be done by connecting
transistor Q2 across the capacitor as shown. At specific
time intervals, a voltage pulse applied to the base of Q2
turns the transistor "on" providing a discharge path
for the capacitor.
21.14 OSCILLOSCOPE AMPLIFIERS
The purpose of an oscilloscope is to provide a
faithful representation applied to its input terminals,
considerable attention has to be paid to the designing
of these amplifiers for faithful representation of input
signals. The oscillographic amplifiers can be classified
into two major categories
(?) a.c. coupled amplifiers, and
(zz) d.c. coupled amplifiers.
Low cost oscilloscopes use a.c. coupled amplifiers
A.C. amplifiers are commonly used in oscilloscopes
used for laboratory purposes.
The d.c. coupled amplifiers are expensive, they
offer the advantage of responding to d.c. voltages.
Their use makes it possible to measure d.c. voltages,
pure a.c. signals, and to measure a.c. signals super­
imposed upon d.c. signals. D.c. coupled amplifiers
have the additional advantage, that they eliminate the
problem of low frequency phase shift and accom­
panying waveform problem of waveform distortion
when observing low frequency pulse train.
There is another way of classifying oscilloscopic
amplifiers. They can be classified according to band­
width they use. On this basis they can be classified as :
(z) narrow bandwidth amplifiers, and
(zz) broad bandwidth amplifiers.
These categories of amplifiers are not sharply
divided, nor is the classification universally accepted.
If the frequency band does not extend upto the
television colour sub-carrier frequency of 3.58 MHz,
the amplifier can be classified as narrow band.
However, if the frequency response curve of the
amplifier is flat beyond 3.58 MHz, the amplifier is
classified as a broad band amplifier. Most general
purpose, laboratory quality oscilloscopes respond
fully to frequencies in excess of 5 MHz. Therefore, the
responsible amplifier, which is the vertical amplifier,
is broadband.
Cathode Ray Oscilloscope (CRO) 653
21.14.1 Vertical Amplifier
The vertical amplifier is the principal factor in
determining the sensitivity and bandwidth of an
oscilloscope. In general, greater sensitivity, which is
expressed in terms of V/cm of vertical deflection at the
midband frequency, is obtained at the expense of
bandwidth. As a rule of thumb, an amplifier at some
specific cost may be obtained with a certain gain­
bandwidth product, which is the product of the
voltage gain of the amplifier and its bandwidth.
Voltage gain may be sacrificed in favour of greater
bandwidth, or vice versa, without significantly
affecting the cost of the amplifier. However, if the
gain-bandwidth product increases, the cost of the
amplifier will increase.
The gain of the vertical amplifier determines the
smallest signal that the oscilloscope can satisfactorily
reproduce on the CRT screen. The sensitivity of an
oscilloscope is directly proportional to gain of the
vertical amplifier that is, as gain increases sensitivity
increases, which allows us to observe smaller-
amplitude signals.
The vertical sensitivity is a measure of how much
the electron beam will be deflected for a specified
input signal. The CRT screen is covered with a plastic
grid pattern called a graticule. The spacing between
the grid lines is generally 10 mm. However, vertical
sensitivity is generally expressed in volt per division.
On the front panel of the oscilloscope shown in
Fig. 21.28, one can see a knob attached to a rotary
switch labeled gain. The gain could be designated
V/Div. The rotary switch is electrically connected to
the input attenuator network, which will be discussed
in subsequent paragraphs. The setting of the rotary
switch indicates what amplitude signal is required to
deflect the beam vertically by one division.
The vertical sensitivity of an oscilloscope is the
smallest deflection factor that can be selected with the
rotary switch. As an example, if the most sensitive
position on the volt/division rotary switch is
5 mV/division, then the vertical sensitivity of the
oscilloscope is 5 mV/division.
The bandwidth of an oscilloscope determines the
range of frequencies that can be accurately
reproduced on the CRT screen. The greater the
bandwidth, the wider is the range of frequencies that
can be observed with the instrument. Ideally, the gain
of the broadband amplifier should be constant from
direct current to near the upper limit of the range of
frequencies that can be observed with the oscilloscope.
A typical curve for a broadband amplifier of gain
versus frequency, called a frequency-response curve,
is shown in Fig. 21.16.
The bandwidth of an oscilloscope is the range of
frequencies over which the gain of the vertical
amplifier is within 3 db of the midband frequency
gain. The upper limit of the bandwidth is the
frequency, f2, at which the gain has decreased by 3 db.
The bandwidth can be increased by using feedback in
designing the amplifier. However, this reduces the
gain, which is why the gain-bandwidth product for an
amplifier is constant.
Fig. 21.16 Typical frequency response curve for a
broadband amplifier.
One of the specifications generally included on
the specifications sheet for the vertical amplifier of a
broadband oscilloscope is the rise time. The rise time of
a pulse is defined as the time required for the edge to
rise from 10% to 90% of its maximum amplitude.
When an oscilloscope is used to observe a pulse or a
square wave, the rise time of the instrument must be
faster than the rise time of the pulse or square wave,
otherwise the observed signal will not be accurately
reproduced. Although the exact relationship between
bandwidth and the rise time varies slightly with ampli­
fier design, an approximate relationship is given as
tr x BW = 0.35 ...(21.26)
where
t = rise time measured in second,
BW = bandwidth in hertz.
Example 21.3 If the bandwidth of an oscilloscope is
given as direct current to 10 MHz, what is the fastest rise
time a sine wave can have to be accurately reproduced by the
instrument?
Solution. Using Eq. 21.26, we can compute the
maximum rise time as,
035 QC
t =------------- 7- = 35 ns.
r 10xl06
The vertical amplifier of a laboratory-quality
oscilloscope frequently consists of the two major
circuit blocks shown in Fig. 21.17, a pre-amplifier and
the main vertical amplifier or mainframe amplifier.
 654 Electrical and Electronic Measurements and Instrumentation

The pre-amplifier is sometimes a separate, inter­ the figure reduces, by voltage division, the amplitude
changeable plug-in unit that can quickly and easily be of the horizontal input signal to a level equal to the
plugged into the mainframe of the oscilloscope. If sensitivity of the horizontal amplifier.
different plug-in units are available for specific measure­
ment applications, the measurement capabilities of an
oscilloscope can be broadened considerably at a
reasonable cost.

Fig. 21.17 Block diagram showing the vertical ampli­

fier of a laboratory-quality oscilloscope.

The active element in the first stage of the

pre-amplifier is generally a field-effect transistor (FET)
which provides a high input impedance for the Fig. 21.18 Block diagram of the horizontal amplifier of
an oscilloscope.
oscilloscope. The final stage of the mainframe
amplifier is generally a push-pull amplifier which
21.15 VERTICAL INPUT AND SWEEP
provides voltages of equal amplitude but out of phase
GENERATOR SIGNAL SYNCHRONIZATION
by 180° to achieve balanced deflection.
Most waveforms that we will have occasion to
21.14.2 Horizontal Amplifiers observe with an oscilloscope will be changing at a rate
The horizontal amplifier basically serves two much faster than the eye can follow, perhaps many
purposes. million times per second. If we are to be able to
observe such rapid changes, the beam must retrace the
1. When the oscilloscope is being used in the
same pattern repeatedly. If the pattern is retraced in
ordinary mode of operation to display a
signal applied to the vertical input, the
such a manner that the pattern always occupies the
horizontal amplifier will amplify the sweep same location on the screen, the eye will see a
generator output. stationary display. The beam will retrace the same
pattern at a rapid rate if the vertical input signal and
2. When the oscilloscope is being used in the X-Y
the sweep generator signal are synchronized ; which
mode, the signal applied to the horizontal
means that the frequency of vertical input signal must
input terminal will be amplified by the
be equal to or an exact multiple of, the sweep
horizontal amplifier.
generator signal frequency, as shown in Fig. 21.19. If
When the oscilloscope is being used in its
ordinary mode of operation, the gain and bandwidth Vertical
requirements for the horizontal amplifier are not as input
signal
stringent as those for the vertical amplifier. Although
the vertical amplifier must be able to reproduce
faithfully low-amplitude, high-frequency signals with Sweep
generator
fast rise times, the horizontal amplifier is required to output
provide only a faithful reproduction of the sweep
signal, which has a relatively high amplitude and a
slow rise time.
As with the vertical amplifier, the final stage of
the horizontal amplifier is a push-pull amplifier, as
shown in Fig. 21.18. The attenuator network given in Fig. 21.19 Synchronized waveforms and CRT display.
Cathode Ray Oscilloscope (CRO) 655
the vertical input frequency is not exactly equal to or
an exact multiple of, the sawtooth frequency, the
waveforms will not be synchronized and the display
"walks" across the screen. If the pattern moves toward
the right, the frequency of the sawtooth waveform is
too high. Movement of the pattern towards the left
indicates that the frequency of the sawtooth is too low.
Synchronization of the waveforms can be
accomplished in two different ways.
(z) Free running sweep. In very basic
oscilloscopes the sweep generator is continuously
changing and discharging a capacitor. One ramp
voltage is followed immediately by another; hence,
the sawtooth pattern appears. A sweep generator
operating in this manner is said to be “free running". In
order to present a stationary display on the screen, the
sweep generator signal must be forced to run in
synchronization with the vertical input signal. In basic
oscilloscopes this is accomplished by carefully
adjusting the sweep frequency to a value very close to
the exact frequency of the vertical input signal, or a
sub-multiple of this frequency. With both signals at
the same frequency, an internal synchronizing pulse
will lock the sweep generator into the vertical input
signal. This method of synchronization has some
serious limitations when an attempt is made to
observe low-amplitude signals. However, the most
serious limitation is probably the inability of the
instrument to maintain synchronization when the
amplitude or frequency of the vertical signal is not
constant, such as with voice or music signals.
(zz) Triggered sweep. The above limitations are
overcome by incorporating a trigger circuit into the
oscilloscope as shown in Fig. 21.20. The trigger circuit
Vertical
input
Ext.
trigger
Fifl. 21.20 Block diagram of an oscilloscope with triggered sweep.
may receive an input from one of three sources
depending on the setting of the trigger selector switch.
The input signal may come from an external source
when the trigger selector switch is set to EXT, from a
low-amplitude ac voltage at line frequency when the
switch is set to LINE, or from the vertical amplifier
when the switch is set to INT. When set for internal
triggering (INT), the trigger circuit receives its input
from the vertical amplifier. When the vertical input
signal that is being amplified by the vertical amplifier
reaches a certain level, the trigger circuit provides a
pulse to The sweep generator, thereby ensuring that
the sweep generator output is synchronized with the
signal that triggers it.
One type of circuit frequently used in the "trigger
circuit" block of Fig. 21.20 is called a Schmitt trigger or
a voltage-level detector. Basically, the Schmitt trigger
compares an input voltage, in this case from the
vertical amplifier, with a voltage at a point in the
circuit. When the input voltage exceeds the voltage to
which it is being compared, the circuit changes states,
which means the output voltage goes to a high state.
The point at which this occurs is called the upper
trigger point (UTP). When the input voltage drops
below a certain level called the lower trigger paint
(LTP), the Schmitt trigger output returns to its original
level, and thus the output of the circuit is a square
wave. Unless this square wave is of very short
duration, it will not be suitable to trigger the sweep
generator directly. It is common practice to apply the
square wave to an RC circuit called a differentiator
whose output is a short-duration spike suitable for
triggering the sweep generator.
 656 Electrical and Electronic Measurements and Instrumentation
21.16 ATTENUATORS
The voltage at the input terminal of the vertical
amplifier, which causes the beam to be deflected off
the CRT screen, is quite low in amplitude. So that
high-amplitude signals may be displayed, an attenuator
network is placed between the vertical input terminal
and the input terminal of the vertical amplifier. The
term attenuate means to "reduce in size". The purpose of
the attenuator is to reduce the amplitude of the
vertical input signal before applying it to the vertical
amplifier. The most basic atenuator is a simple
resistive voltage divider such as the one shown in
Fig. 21.21. With this circuit, the input voltage V will be
attenuated by a factor of 10,000 with the switch S set to
Fig. 21.21 Resistive attenuator network.
position A as shown. In positions B, C and D, the
attenuation factors will be 1000, 100 and 10, respec­
tively. In switch position E there is no attenuation. In
any switch position, the total input resistance, as seen
by V-, is 100 kQ. The attenuation in any switch
position can be determined from the ratio of the total
resistance from the position of interest to ground to
the total series resistance, written as :
V R
-^ = — ...(21.22)
17. Rt
where
R = total resistance from the desired attenuator
terminal to ground, and
Rt = total series resistance.
Equation 21.22 is generally referred to as the
voltage divider equation. The attenuation factor is the
reciprocal of the voltage divider ratio.
Example 21.4 If the switch in Fig. 21.21 is set of
position D, what is the attenuation factor ?
Solution. From Eqn. 21.21, we obtain the voltage
divider ratio as :
Yl-Yl
Y
9x103+900 + 90 + 10 _01
100 xlO3
Therefore, the attenuation factor = 1/0.1 = 10.
The resistance values shown in Fig. 21.21 will
provide the desired attenuation only if the input
resistance R. of the vertical amplifier is much greater
than the attenuator resistance values. If the amplifier
input resistance is not much larger than the attenuator
resistance with which it is in parallel, appreciable
error in the attenuation factor will be noted since the
attenuator is “loaded down" by the amplifier.
Example 21.5 If the switch in Fig. 21.21 is in position
D, as in Example 21.4, and the input resistance of the
amplifier Ri is 100 k£l, what is the attenuation factor ?
Solution. Since the amplifier input resistance is in
parallel with the attenuator resistance, this parallel
combination must be used in the voltage divider
equation. The parallel resistance is
RR
R =—'—
P Ri + R
100xl03xl0xl03 onQ1„
=---------- ------------- = 9.09 kQ
100xl03 + 10xl03
Therefore,
V 9 09
-2- = = 0.0909
v. 100
The attenuation factor is 1/0.0909 =11.
As can be seen by comparing the results of
Examples 21.4 and 21.5, the attenuation is increased in
response to the parallel effect of the input resistance of
the amplifier.
Switch S in Fig. 21.21 is the rotary switch that is
mounted on the front panel of an oscilloscope and
labeled volt/div. as was mentioned briefly in
Section 21.14.1. Example 21.6, shows how to
determine the value of the attenuating resistors to
obtain desired attenuation factors.
Example, 21,6 An oscilloscope is to have an input
resistance of 8 MQ, a sensitivity of 50 mV, and attenuation
factors of 4, 10, 40, 100 and 400. Compute the value of the
attenuating resistors and the volt/division value
corresponding to each attenuation factor. Assume R{ of the
vertical amplifier is sufficiently high to be ignored.
Cathode Ray Oscilloscope (CRO) 657

RA + RB^Rc = ^V0/yi)
= (8 MQ) (50 mV/2 V) = 200 kQ
Rc=(Ra + Rb + Rc)-(Ra + Rb)
= 200 kQ - 80 kQ = 120 kQ
ra + rb + + rd =
= (8 MQ) (50 mV/0.5 V)
= 800 kQ
RD=(RA + RB + RC + Rd)~(RA +RB + RC>

= 800 kQ-200 kQ = 600 kQ

ra + Rb + + rd + re = Vo SVi
= (8MQ)(50mV/0.2V)
= 2MQ
RE = (RA + RB + + RD + re^
- (Ra + Rb + + Rd)
Fig. 21.22 Circuit for Example 21.6. = 2MQ -800kQ =1.2 MQ

Solution. The attenuating network required is Rp = ~(RA + RB + + RD + RE^

shown in Fig. 21.22. When switch S is set to position A, = 8MQ-2MQ =6 MQ.

as indicated, maximum voltage can be applied across
The complete attenuator network is shown in
the network because of the voltage divider action.
Fig- 21-23,
Therefore, at this position the attenuation factor will
be 400. The attenuation factor equals the reciprocal of
the voltage divider ratio. For position A then
1 V-
attenuation factor = 400 =-------- or 400 = —
Vo/Vi Vo

Since Vo equals the value of the sensitivity

(50 mV/div) because Vg is applied directly to the
amplifier in each switch position, we can solve for V) as:
V; = 4001/
= 400 (50 mV/div) =20 V/div at position A
From this result we see that we can multiply the
attenuation factor by the sensitivity to obtain the value
of the volts/division at each switch setting.
Position B: V. = 100 (50 mV/div) = 5 V/div.
Position C : V* . = 40 (50 mV/div) = 2 V/div.
Position D : V = 10 (50 mV/div) = 0.5 V/div.
Position E : V = 4 (50 mV/div) = 0.2 V/div.
Position F: There is no attenuation or the
attenuation factor is 1 ; therefore,
Fig. 21.23 Complete attenuator network
V. = 1 (50 mV/div) = 50 mV/div.
for Example 21.6
Solving for the attenuating resistors, we obtain
the following : If we were measuring de voltages exclusively, our
discussion of voltage divider attenuators might end
Ra = Rr(V0/ V) = (8MQ) (30 mV/20 V) = 20 kQ
here. However, oscilloscopes are used primarily to
Ra + Rb = 1^(1/ IV;.) = (8 MQ) (50 mV/5 V) = 80 kQ display ac waveforms. This introduces the problem of
RDr=(R
' /ia + RD
r)-R
' ria = 80 kQ-20 kQ =60 kQ the stray capacitance of the attenuator network as
 658
shown in Fig. 21.24. For a.c. signals, the attenuation
factor is actually related to the ratio of the impedance
n—i—niiiiwirm r '-
Fig. 21.24 Attenuator network with associated
stray capacitance.
If the stray capacitance values are not correctly
related to the resistance values, the voltage divider
ratio will not have the same value at all frequencies.
Example 21.'/ Compare the output voltage of the voltage
divider attenuator shown in Fig. 21.25 for a d.c. voltage and
a 10 MHz a.c. signal.
Fig. 21.25 Circuit for Example 21.7.
Solution. For the d.c. voltage source
100 XlO3
900 x 103 + 100x 103
= 0.1 V or k = 1/0.1 = 10
For the a.c. source, Xfl is approximately 3183 Q and
Xc2 is approximately 1592 Q. Since these values are
much less than the value of the resistances in parallel
with them, we can approximate the output voltage as:
/ 1592
K=(l)
I <3183 + 1592
= 0.33 York = 1/0.33-3.
Electrical and Electronic Measurements and Instrumentation
Therefore, the attenuation with high frequency
a.c. is very much different from what it is with d.c.
Consider the attenuator network of Fig. 21.26. The
resistance ratio of to Kj is the reciprocal of the ratio
Fig. 21.26 Compensated attenuator circuit.
of capacitors C2 to Cr When this is true, the voltage
divider ratio will remain constant over a wide range of
frequencies. Moreover, for a frequency-compensated
voltage divider, the following relationships apply :
R, = i ...(21.24)
and C2 = C1(fc-l) ...(21.25)
where k is the voltage divider ratio, Vi/V0.
cl
with these values,' X . = 1592Q and XcZ„ =177.
< 177 >
V=(l)---- —----- =0.1 V
0 U592+177J
which is the same as with d.c.
21.17 BASIC CRO CIRCUITS
Figure 21.27 shows a block diagram of a typical
oscilloscope. In the following few pages, we describe
the basic circuitry of cathode my oscilloscope.
1. Vertical (Y) deflection system. The signals to
be examined are usually applied to the vertical or Y
deflection plates through an input attenuator and a
number of amplifier stages. Vertical amplifier is
required because the signals are not strong enough to
produce measurable deflection on the CRT screen. The
amplifier response must be wide enough to pass
faithfully the entire band of frequencies to be
measured.
When high voltage signals are to be examined,
they must be attenuated to bring them within the
range of vertical amplifiers. The vertical amplifier
output is also applied to the synchronizing amplifier
through the synchronizer selector switch in the
internal position. This permits the horizontal sweep
circuit to be triggered by the signal being investigated.
Cathode Ray Oscilloscope (CRO) 659
Vertical
Intensity (Z) position
input
Vertical (Y) ■< • Focusing
direct Alternate Z input / anode
Vertical
Vertical (Y)
amplifier CRT
input Cathode
Horizontal (X) Heater
direct

External pxt
horizontal(X)
input Horizontal
Sweep amplifier Alternate Accelerating
3 selector blanking
input Horizontal
Int. r position
Ext. 6 Sync
External
sync(trigger) Line ^Sync J amplifier
input selector
Intensity Focus Astigmatism
Sawtooth
sweep d.c. supply
generator d
To vertical To line
input frequency
Calibration Blanking
oscillator circuit

Fig. 21.27 Simplified block diagram of a general purpose oscilloscope.

2. Horizontal (X) deflection system. The
horizontal (X) deflection plates are fed by a sweep
voltage that provides a time base. The horizontal
plates are supplied through an amplifier, but they can
be fed directly when voltages are of sufficient
magnitude. When external signals are to be applied to
the horizontal deflection system, they can also be fed
through the horizontal amplifier, via the sweep
selector switch in the external position. When the
sweep selector switch is in the internal position, the
horizontal amplifier receives an input from the saw
tooth sweep generator which is triggered by the
synchronizing amplifier.
initiated by the waveform under examination. In some
cases the waveform may be periodic, but it may be
that the interesting part of the waveform is of a very
short duration compared to the period of the
waveform. Under such cases a triggered sweep is
used.
In triggered sweep or single sweep, the spot is
swept once across the screen in response to a trigger
signal. The triggered sweep is used for examination of
transients or one time signals and the waveform is
photographed for record. The trigger can be obtained
from the signal under investigation or by an external
source.

Types of sweeps. There are four basic types of
sweeps :
(z) Free running or recurrent sweep. In the free
running or recurrent sweep, the sawthooth waveform
is repetitive. A new sweep is started immediately after
the previous sweep is terminated and the circuit is not
initiated by any external signal.
(zz) Triggered sweep. A waveform to be observed on
the CRO may not be periodic but may perhaps occur
at irregular intervals. In this case it is desirable that the
sweep circuit remain inoperative and the sweep be
(zzz) Driven sweep. In most cases, a driven sweep is
used where the sweep is recurrent but triggered by the
signal under test.
(zzz) Non saw tooth sweep. For some applications like
comparison of two frequencies or for finding phase
shift between two voltages, non sawtooth sweep
voltages are utilized for the sweep circuit. Sweep
frequencies vary with the type of oscilloscope. A
laboratory oscilloscope may have sweep frequencies
up to several MHz ; a simple oscilloscope for audio
work has an upper limit of 100 kHz. Most TV services
require a sweep voltage frequency up to 1 MHz.
 Electrical and Electronic Measurements and Instrumentation
3. Synchronization. Whatever type of sweep is
used, it must be synchronization with the signal being
measured. Synchronization has to be done to obtain a
pattern. This requires that the time base be operated at
a submultiple frequency of the signal under
measurement (applied to Y plates). If synchronization
is not done, the pattern is not stationary, but appears
to drift across the screen in a random fashion.
Sources of synchronization. There are three usual
sources for synchronization which can be selected by
synchronization selector :
(z) Internal. In this type of synchronization, the
trigger is obtained from the signal being
measured through the vertical amplifier.
(zz) External. In this method, an external trigger
source is also used to trigger or initiate the
signal being measured
(zzz) Line. In this case, the trigger is obtained from
the power supply to the CRO (say 230 V, 50 Hz).
fig.21.28 Front panel of a general purpose CRO.
4. Blanking circuit. The sawtooth sweep voltage
applied to the X plates moves the beam across the CRT
tube in a straight horizontal line from left to right
during the sweep or trace time T. At comparatively
slow movement of the spot will appear as a solid line,
provided the rate of movement exceeds the threshold
of persistence of vision. Below this threshold limit, a
moving spot is perceived. On the other hand, the
comparatively rapid movement of spot will appear as
a thin and dim line, or may be invisible. Thus if the
retrace or flyback time is very small, the spot remains
invisible. In an ideal case the flyback time, Tr is zero
and hence the spot while moving from right to left
remains invisible. However in actual practice the
flyback time is not zero and therefore the retrace
(moving of beam from right to left i.e., its starting
point) may cause confusion. Thus the retrace should
be eliminated or blanked out. The retrace is blanked
out by applying a high negative voltage to the grid
during the flyback period Tr. The blanking voltage is
usually developed (or triggered) by sweep generator.
5. Intensity (Z-axis) modulation. Intensity
modulation (Z-axis modulation) is done by inserting a
signal between the ground and the cathode (or control
grid). Z-axis modulation is applied during normally
visible portion of the trace.
The Z-axis modulation can be used for
brightening the display. Periodic positive pulses are
applied to the grid (alternatively negative pulses are
applied to cathode) to brighten the beam during its
sweep period. These periodically brightened spots
may be used as markers for time calibration of the
main waveform.
6. Positioning controls. It is necessary to
provide some means of positioning the trace on the
screen. The positioning of the trace is done by
applying small independent, internal d.c. voltages to
the deflecting plates and control can be exercised by
varying the voltage with help of potentiometers.
7. FOCUS control. As mentioned earlier, the
focusing electrode acts like a lens whose focal length
can be changed. This change can be brought about by
changing the potential of the focusing anode.
8. Intensity control. The intensity of the beam is
varied by the Intensity control potentiometer which
changes the grid potential with respect to cathode. The
grid potential determines the amount of electrons
leaving the cathode and thus controls the intensity of
the beam.
9. Calibration circuit. Laboratory oscilloscopes
normally have an internally generated and stabilized
Cathode Ray Oscilloscope (CRO) 661
voltage of known amplitude which is used for
calibration purposes. Usually the calibrating voltage
has a square waveform.
10. Astigmatism. In most modem oscilloscopes
there is an additional focusing control marked
Astigmatism. This is used to correct an effect which
exactly is analogous to astigmatism in optical lenses.
To focus the spot correctly, it is necessary to stop it
near the centre of the screen by switching off the time
base and adjusting the X and Y positioning controls.
The spot is then made as sharp as possible by
successive adjustment of focus and astigmatism
controls.
21.18 OBSERVATION OF WAVEFORM ON CRO
In order to observe waveform on a CRO, the
waveform of voltage under test is applied to Y plates
and a voltage obtained from a sawtooth generator is
applied to X plates. Let us assume that the sawtooth
waveform has an idealized waveshape.
When simultaneously with the horizontal
sawtooth (ramp) voltage, an input voltage is applied
to vertical deflection (Y) plates, the beam is under the
influence of two forces : (z) one in the horizontal
direction moving the beam at a linear rate from left to
right, and (zz) second in the vertical direction moving
the beam up and down. Since the deflection is
proportional to the voltage applied to the deflection
plates, the horizontal movement is proportional to the
voltage applied to X plates at any instant and since
the ramp voltage is linear it traces a straight line on the
CRT screen. The vertical deflection is proportional to
the voltage applied to the Y plates at any instant and
thus the beam moves up and down according to the
Fig* 21 29 Observation of waveform on CRO.
magnitude and polarity of the input voltage.
Figure 21.29 shows the waveform displayed on a CRT
tube due to an input sinusoidal voltage.
At the end of one sweep cycle, the sweep voltage
abruptly drops down and the spot is immediately
transferred to its original position. The process is then
repeated again, with the result, that a stationary image
is seen on the screen.
For the case shown the frequency of the input
voltage is twice that of sawtooth (sweep) voltage. To
observe more than one cycle of the input voltage, the
sweep voltage frequency has to be a submultiple of the
input voltage frequency.
21.19 MEASUREMENT OF VOLTAGES & CURRENTS
The expression for electrostatic deflection, given
in Eqn. 21.16, shows that the deflection is proportional
to the deflection-plate voltage. Thus the cathode-ray
tube will measure voltage. It is usual to calibrate the
tube under the given operating conditions by
observing the deflection produced by a known
voltage. Direct voltages may be obtained from the
static deflection of the spot, alternating voltages from
the length of the line produced when the voltage is
applied to Y plates while no voltage is applied to X
plates. The length of this line corresponds to the
peak-to-peak voltage. When dealing with sinusoidal
voltages, the rms value is given by dividing the
peak-to-peak voltage by 2a/2 .
Laboratory oscillographs frequently incorporate
voltage-measurement facilities by including constant­
gain amplifiers and calibrated shift controls. The Y-shift
control is adjusted so that positive peak of the test
voltage coincides with some datum line on the screen;
the shift control is then operated until the negative
peak coincides with the datum. The movement of the
control is arranged to read directly the peak-to-peak
voltage. The valuejof"a current can be obtained by
measuring the vdltage drop across a known resistance
connected in/me circuit.
21.t2p^MEASUREMENT OF PHASE & FREQUENCY
(LISSAJOUS PATTERNS)
It is interesting to consider the characteristics of
patterns that appear on the screen of a CRT when
sinusoidal voltages are simultaneously applied to
horizontal and vertical plates. These patterns are
called 'Lissajous Patterns'.
When two sinusoidal voltages of equal frequency
which are in phase with each other are applied to the
 662
horizontal and vertical deflection plates, the pattern
appearing on the screen is a straight line as is clear
from Fig. 21.30.
Fig. 21.30 Lissajous pattern with equal frequency
voltages and zero phase shift.
Thus when two equal voltages of equal frequency
but with 90° phase displacement are applied to a CRO,
the trace on the screen is a circle. This is shown in
Fig. 21.31.
Fig. 21.31 Lissajous pattern with equal voltages of
equal frequency and a phase shift of 90°.
When two equal voltages of equal frequency but
with a phase shift (j> (not equal to 0° or 90°) are applied
to a CRO we obtain an ellipse as shown in Fig. 21.32.
An ellipse is also obtained when unequal voltages of
same frequency are applied to the CRO.
A number of conclusions can be drawn from the
above discussions. When two sinusoidal voltages of
same frequency are applied :
Electrical and Electronic Measurements and Instrumentation
Fig 21.32 Lissajous pattern with two equal voltages
of same frequency and phase shift of <|>.
(z) A straight line results when the two voltages
are equal and are either in phase with each other or
180° out of phase with each other. The angle formed
with the horizontal is 45° when the magnitudes of
voltages are equal. An increase in the vertical
deflection voltage causes the line to have an angle
greater than 45° with the horizontal. On the other
hand a greater horizontal voltages makes the angle
less than 45° with the horizontal.
(zz) Two sinusoidal waveforms of the same
frequency produce a Lissajous pattern, which may be
a straight line, a circle or an ellipse depending upon
the phase and magnitude of the voltages.
A circle can be formed only when the magnitude
of the two singals are equal and the phase difference
between them is either 90° or 270°. However, if the
two voltages are not equal and/or out of phase an
ellipse is formed. If the Y voltage is larger, an ellipse
with vertical major axis is formed while if the X plate
voltage has a greater magnitude, the major axis of the
ellipse lies along horizontal axis.
(zzz) It is clear from Fig. 21.33, that for equal
voltages of same frequency progressive variation of
phase voltage causes the pattern to vary from a straight
diagonal line to ellipses of different eccentricities and
then to a circle, after that through another series of
ellipses and finally a diagonal straight line again.
Regardless of the two amplitudes of the applied
voltages the ellipse provides a simple means of
finding phase difference between two voltages.
Referring to Fig. 21.36, the sine of the phase angle
between the voltages is given by :
Y X
sin <}» = —= — ...(21.20)
Y, X~
Cathode Ray Oscilloscope (CRO) 663

Resulting Vertical deflection Frequency measurements. Lissajous patterns

may be used for accurate measurement of frequency.
The signal, whose frequency is to be measured, is
applied to the Y plates. An accurately calibrated
standard variable frequency source is used to supply
voltage to the X plates, with the internal sweep
generator switched off. The standard frequency is
adjusted unit the pattern appears as a circle or an
ellipse, indicating that both signals are of the same
frequency. Where it is not possible to adjust the
standard signal frequency to the exact frequency of
the unknown signal, the standard i$ adjusted to
multiple or a submutliple of the frequency of the un­
known source so that the pattern appears stationary.
Let us consider an example. Suppose sine waves
are applied to X and Y plates as shown in Fig. 21.35.
Let the frequency of wave applied to Y plates is twice
that of the voltage applied to X plates. This means that
the CRT spot travels two complete cycles in the
vertical direction against one in the horizontal
direction.

phase shifts.

For convenience, the gains of the vertical and

horizontal amplifiers are adjusted so that the ellipse
fits exactly into a square marked by the lines on the
graticule.
If the major axis of the ellipse lies in the first and
third quadrants (i.e., its slope is positive) as in
Fig. 21.34 (tz) the phase angle is either between 0° to 90°
or between 270° to 360°. When the major axis of ellipse
lies in second and fourth quadrants i.e. when its slope
is negative as in Fig. 21.34 (b), the phase angle is either
between 90° and 180° or between 180° and 270°.

Fig. 21.35 Lissajous pattern with frequency ratio 2:1.

The two waves start at the same instant. Lissajous

pattern may be constructed in the usual way and a 8
shaped pattern with two loops is obtained. If the two
waves do not start at the same instant we get different
patterns for the same frequency ratio. The Lissajous
patterns for other frequency ratios can be similarly
drawn. Some of the these patterns are shown in
Fig.21.34 Determination of angle of phase shift. Fig. 21.36.
 664 Electrical and Electronic Measurements and Instrumentation
Fig. 21.36 Lissajous patterns with different
frequency ratios.
It can be shown that for all the above cases, the
ratio of the two frequencies is :
fy _ number of times tangent touches top or bottom
fx number of times tangent touches either side
_ number of horizontal tangencies
number of vertical tangencies
where f = frequency of signal applied to Y plates,
and fx = frequency of signal applied to X plates
The above rule, however, does not hold for the
Lissajous pattern with free ends as shown in
Fig. 21.37(a). The simple rule mentioned above needs
following modifications :
Two lines are drawn, one horizontal and the other
vertical so that they do not pass through any
intersections of different parts of the Lissajous curve.
The number of intersections of the horizontal and the
vertical lines with the Lissajous curve are individually
counted. The frequency ratio is given by :
number of intersections of the horizontal line
fy ' with the curve
fx number of intersections of the vertical line
with the curve
The applications of this rule to Fig. 21.37 gives a
. /v 5
frequency ratio — = —.
A fication the two channels meet at an electronic switch.
The modified rule is applicable in all cases
whether the Lissajous pattern is open or closed.
The ratio of frequencies when open ended
Lissajous patterns are obtained can also be found by
treating the open ends as half tangencies as shown in
Fig. 21.37(b)- The pattern of Fig. 21.37(a) is reproduced
/y _ number of horizontal tangencies
/ number of vertical tangencies
2+1/2 5
1 ~2
Fig. 21.37 Lissajous patterns with half tangencies.
There are some restrictions on the frequencies
which can be applied to the deflection plates. One
obviously, is that the CRO must have the bandwidth
required for these frequencies. The other restriction is
that the ratio of the two frequencies should not be such
as to make the pattern too complicated otherwise
determination of frequency would become difficult.
As a rule ratios as high as 10 :1 and as low as 10:9 can
be determined comfortably.
21.21 MULTI-INPUT OSCILLOSCOPES
Oscilloscopes can have multiple input and
display facilities. Two inputs is the most common,
although four and eight inputs are available for
special applications. There are two primary types :
single beam which can then be converted into several
traces, and dual beam, which may also subsequently be
converted into a further number of traces. Two input
oscilloscopes are described in this section, under the
headings of dual trace and dual beam, although the
principles are applicable to any number of inputs.
21.21.1 Dual Trace Oscilloscopes
Figure 21.38 illustrates the construction of a
typical dual trace oscilloscope. There are two separate
vertical input channels, A and B, and these use
separate attenuator and preamplifier stages. Therefore
the amplitude of each input, as viewed on the oscillo­
scope, can be individually controlled. After preampli­
This has the ability to pass one channel at a time into
the vertical amplifier, via the delay line.
There are two common operating modes for the
electronic switch, called alternate and chop, and these
are selected from the instrument's front panel. The
alternate mode is illustrated in Fig. 21.39. In this the
Cathode Ray Oscilloscope (CRO) 665
Fig. 21.38 A dual trace oscilloscope block diagram.
electronic switch alternates between channels A and B,
letting each through for one cycle of the horizontal
sweep. The display is blanked during the flyback and
hold-off periods, as in a conventional oscilloscope.
Provided the sweep speed is much greater than the
decay time of the CRT phosphor, the screen will show
a stable display of both the waveform at channels A
and B. The alternate mode cannot be used for
displaying very low frequency signals.
Time
Fig. 21.39 Waveforms for a dual channel oscilloscope
operating in alternate mode ; (a) horizontal
sweep voltage, (b) voltage to channel A.
(c) voltage to channel B,
(d) grid control voltage.
The chopped operating mode of the electronic
switch is shown in Fig. 21.40. In this mode the
electronic switch free runs at a high frequency of the
order of 100 kHz to 500 kHz. The result is that small
segments from channels A and B are connected
alternately to the vertical amplifier, and displayed on
the screen. Provided the chopping rate is much faster
than the horizontal sweep rate, the display will show a
continuous line for each channel. If the sweep rate
approaches the chopping rate then the individual
segments will be visible, and the alternate mode
should now be used.
Time
Fig. 21.40 Waveforms of a dual channel oscilloscope
operating in chopped mode, (a) Horizontal
sweep voltage (b) Voltage to channel A,
(c) Voltage to channel B,(d) grid control voltage.
 666 Electrical and Electronic Measurements and Instrumentation
The time base circuit shown in Fig. 21.38 is similar
to that of a single input oscilloscope (Fig. 21.12).
Switch S2 allows the circuit to be triggered on either
the A or B channel waveforms, or on line frequency, or
on an external signal. The horizontal amplifier can be
fed from the sweep generator, or the B channel via
switch Sr This is the X-Y mode and the oscilloscope
operates from channel A as the vertical signal and
channel B as the horizontal signal, giving very
accurate X-Y measurements. Several operating modes
can be selected from the front panel for display, such
as channel A only, channel B only, channels A and B as
two traces, and signals A + B, A-Br B- A or -(A + B) as
a single trace.
21.21.2 Dual Beam Oscilloscope
The dual trace oscilloscope cannot capture two
fast transient events, as it cannot switch quickly
enough between traces. The dual beam oscilloscope
has two separate electron beams, and therefore two
completely separate vertical channels, as in Fig. 21.39.
The two channels may have a common time base
system, as in Fig. 21.38, or they may have independent
time base circuits, as in Fig. 21.41. An independent
time base allows different sweep rates for the two
channels but increases the size and weight of the
oscilloscope.
Two methods are used for generating the two
electron beams within the CRT. The first method uses
a double gun tube. This allows the brightness and
focus of each beam to be controlled separately but it is
bulkier than a split beam tube.
In the second method, known as split beam, a
single electron gun is used. A horizontal splitter plate
is placed between the last anode and the Y deflection
plates. This plate is held at the same potential as the
anode, and it goes along the length of the tube,
Fig. 21.41 Dual beam oscilloscope with independent time bases.
between the two vertical deflection plates, It therefore
isolates the two channels. The split beam arrangement
has half the brightness of a single beam, which has
disadvantages at high frequency operation. An
alternative method of splitting the beam, which
improves its brightness, is to have two apertures in the
last anode, instead of one, so that two beams emerge
from it.
The disadvantage of the split beam construction is
that the two display may have noticeably different
brightness, if operated at widely spaced sweep
speeds. The brightness and focus controls also affect
the two traces at the same time.
21.22 SAMPLING OSCILLOSCOPES
A sampling oscilloscope is used to examine very
fast signals. It is similar in principle to the use of
stroboscopic light to look at fast mechanical motion.
Samples are taken at different portions of the
waveform, over successive cycles, and then the total
picture is stretched, amplified by relatively low
bandwidth amplifiers, and displayed as continuous
wave on the screen. The advantage of a sampling
oscilloscope is that it can measure very high speed
events, which require sweep speeds of the order of
10 ps per division, and amplifier bandwidths of
15 GHz, using instruments having bandwidth several
orders lower. The disadvantage of a sampling
oscilloscope is that it can only make measurements on
repetitive waveform signals, continuous display for
frequencies in the 50-300 MHz range depending upon
the design of the oscilloscope. Above this range of
frequencies Sampling Techniques must be employed
to obtain satisfactory display. The display may be
made up from as many 1000 dots of luminescence. The
vertical deflection for each dot is obtained from
progressively later points in each successive cycle of
plate
Cathode Ray Oscilloscope (CRO) 667
the input waveform as shown in Fig. 21.42. The
horizontal deflection of the electron beam is obtained
by application of a staircase waveform to X deflection
Fig. 21.42 Principle of sampling oscilloscopes.
The sampling oscilloscope is able to respond and
store rapid bits of information and present them in a
continuous display. It is this ability that enables the
sampling oscilloscope to side-step the usual limita­
tions in conventional high frequency oscilloscopes
which have limited sensitivity and bandwidth and
small display size. The sampling techniques imme­
diately transform the input signal into lower frequency,
domain, where conventional low frequency circuitry
is then capable of producing a highly effective display.
The sampling oscilloscopes can be used beyond
50 MHz into the UHF range around 500 MHz and
beyond up to 10 GHz. However, it should be
understood that sampling techniques cannot be used
for the display of transient waveforms.
Figure 21.43 shows a block schematic for a typical
sampling oscilloscope. The input signal is delayed,
and then sampled by a diode gate. The sampled signal
is saved on a capacitor store, then amplified and fed to
the vertical plates. Unity feedback is used from the
Input
signal
Fig. 21.43 Sampling oscilloscope block diagram.
amplifier to the sampling diode gate. This ensures that
the voltage on the capacitor store is only increased by
the incremental value of the input voltage change,
between each sample.
The staircase is reset after a certain number of
steps, typically 100 to 1000, and it then starts again.
Therefore up to about 1000 points are used to create
the waveform on the screen.
The staircase waveform also feeds the horizontal
plates of the CRT, and it is used to move the spot
across the screen in a series of rapid movements.
Figure 21.42 shows the screen display.
The sample frequency used in sampling
oscilloscopes can be as low as one hundredth of the
signal frequency, so a signal frequency of 1 GHz needs
an amplifier bandwidth of only 10 MHz.
Delayed sweep. Many laboratory-quality
oscilloscopes include a delayed-sweep feature. This
feature increases the versality of the instrument by
making it possible to magnify a selected portion of the
undelayed sweep, measure waveform jitter or rise
time, and check pulse-time modulation, as well as
many other applications.
Delayed sweep is a technique that adds a precise
amount of time between the trigger point and the
beginning of the scope sweep. When the scope is being
used in the delayed-sweep mode, the start of the
delayed sweep can range from a few microseconds to
perhaps 10 seconds or more. The delayed sweep
operation allows the instrument user to view a small
segment of a waveform-for example, an oscillation or
ringing that occurs during a small portion of a
lower-frequency waveform.
 668 Electrical and Electronic Measurements and Instrumentation

Sometimes the delayed-sweep feature is used for

convenience to allow triggering at some other point
than at the leading or trailing edge. However, in some
situations a measurement is possible only if the
delayed-sweep feature is used. For example, suppose
the part of waveform that is to be measured is too far
from the only available trigger point to permit a stable
display on CRT screen. The problem can be solved by
using delayed sweep to trigger at the only available
trigger point and then starting the sweep at the point
of interest.
Although there are a few exceptions, delayed
sweep is normally a feature of dual-time base
oscilloscopes which have two completely separate
sweep generators. One sweep functions as a main
sweep, and the other serves as the delayed sweep. The
main sweep is initiated by a trigger pulse at the
leading edge of pulse 1 shown in Fig. 21.44. Suppose leading edge of pulse 1 in Fig. 21.45. The main sweep
we wish to observe in detail a portion of the waveform generator ramp, which is applied to a comparator
near its trailing edge by expanding the waveform and along with a voltage from a delay control circuit,
using a higher sweep speed-for example, a sweep increases linearly until it trips the comparator at
speed of 0.1 ps/cm. As can be seen in Fig. 21.43, the time fj. When the comparator changes states, the
portion of the waveform of interest to us is completely delayed-sweep generator is triggered, which
off the screen to the right when the sweep speed is set intensifies a portion of the original display. Adjusting
to 0.1 ps/cm. To observe the portion of the waveform the delay time should intensify the portion of the
that is of interest at the higher sweep speed, we must waveform that is of interest. The exact front panel
use the delayed-sweep feature and reset the sweep adjustments that are necessary to display the
speed to 5 ps/cm. expanded portion of the waveform on the CRT screen
depend on the oscilloscope being used. Some
oscilloscopes have an alternate sweep separation control
so that two waveforms are displayed: the original
waveform, with the portion of the waveform of
interest intensified, and a second expanded
waveform, as shown in Fig. 21.47. By adjusting the

Fig. 21.44 CRT display with sweep speed

equal to 5 ps/cm.

Basically, the delayed sweep works as follows.

The main sweep is initiated by a trigger pulse at time
f shown in Fig. 21.46. This time corresponds to the Fig. 21.46 Delayed-sweep triggering waveforms.
Cathode Ray Oscilloscope (CRO) 669
horizontal mode to channel B, we can now increase
the sweep time so that we can view the portion of the
waveform that is of interest.
Fig. 21.47 Delayed-sweep measurement.
21.23 ANALOG STORAGE OSCILLOSCOPES
Storage oscilloscopes, capable of retaining the
image on the screen for longer than that possible with
conventional high persistence phosphors, have many
applications. Examples of these are the capture and
storage of transients, and the steady display of a very
low frequency signal. Two techniques are used to
store signals in an oscilloscope, and these are called
analog and digital storage. Analog storage is capable
of higher speeds, but is less versatile than digital
storage. This section describes the principles of analog
storage, and the two methods most frequently used,
called variable persistence and bistable storage.
Digital storage oscilloscopes are covered in Art. 21.24.
21.23.1 Principle of Secondary Emission
Both the variable persistence and bistable storage
oscilloscopes depend, for their operation, on the
principle of secondary emission. Figure 21.48 shows a
Fig. 21.48 Simple electrode arrangement to
illustrate secondary emission.
simple electrode arrangement to illustrate this
principle. When a beam of electrons from the electron
gun strikes the target it emits secondary electrons,
which are gathered by the collector. The collector is at
a positive potential of V2, the target is at a potential of
Vr which can be varied, and the electron gun is at
ground potential.
If I is the value of the current in the primary
electron beam, coming from the electron gun, and I is
the electron current emitted from the target and
collected by the collector, then the ratio Is / Ip is called
the secondary emission ratio. The value of this ratio
depends on the primary electron velocity and intensity,
and on the chemical composition of the target.
Figure 21.49 shows the variation of the secondary
emission ratio with target voltage, for the arrange­
ment of Fig. 21.48. When the target voltage is much
greater than the collector voltage, all secondary
electrons emitted from the target are attracted back to
it. Therefore the collector current, or secondary
current, is zero, and so also is the secondary emission
ratio. The operating point is now well to the right of V2
in Fig. 21.49. If the target voltage is slightly negative,
as at point F, then all the electrons from the gun are
deflected on-to the collector, before they reach the
target. Therefore although there is no secondary
emission the collector current equals the beam current,
and the secondary emission ratio is unity. This point is
known as the lower stable point. As the target voltage
increases from this point electrons are attracted away
from the collector, but they do not have enough
energy to release secondary electrons from the target.
Therefore the secondary emission ratio falls to a
minimum at A. Beyond the minimum point secondary
emission from the target starts to occur, and these
electrons are accumulated by the collector, so
increasing the secondary emission ratio.
Fig. 21.49 Variation of secondary emission ratio for
the electrode arrangement shown in
Fig. 21.48.
 670 Electrical and Electronic Measurements and Instrumentation

The secondary emission ratio increases through

the crossover point E until it reaches a peak at C.
Beyond this point secondary electrons emitted from
the target are attracted back, in numbers greater than
those which reach the collector, so that the secondary
emission ratio decreases sharply. The curve reaches
the upper stable point at G, where the primary and
secondary currents are equals, and then decreases to
zero.
The lower and upper stable points represent the
erased and written conditions of the CRT screen, and
in the absence of a target voltage the target can remain
in one of these two stages only. Suppose for instance
that the target is in a state to the left of the crossover
point E arid switch SW is opened to remove the target
supply voltage. Because the secondary emission ratio
is less than one, fewer electrons leave the target than
arrive at its surface, so the target is driven Fig. 21.50 Multiple target and flood gun arrangement.
progressively negative, until point F is reached. Now
the number of electrons reaching the target equals the momentarily taken to a negative voltage. This repels
number which are emitted from it. This represents a secondary emission back into the target, moving its
stable erased state. If the target was to the right of the potential negative due to the electrons reaching it
crossover point when switch SW is opened, then, since from the flood gun. When the target reaches point A
the secondary emission ratio is greater than one, more no more chanrging occurs. The collector mesh is now
electrons leave the target than arrive at its surface. The returned to its original voltage of +V2. This is done
target is therefore driven positive unitl it reaches the gradually to prevent the target from being
upper stable point G which represents the written accidentally driven beyond the crossover point, due to
condition of the screen. The target will stay at the capacitive coupling, and so reaching a written state
upper or lower stable point so long as the electron gun again.
is on. At the crossover point the target is unstable, and Secondary
it will move up or down on the curve depending on emission ratio
the direction in which it is sent by noise.
Figure 21.50 shows an arrangement using a
segmented multiple target, with a collector mesh in
front through which the electrons can pass. Two guns
are used, a high energy writing gun, which emits a
narrow beam of electrons, and a flood gun which
covers the target area with a continuous stream of low
energy electrons. The flood gun remains on all the
time, and maintains the target in either of the two
Fig. 21.51 Secondary emission ratio curves for the
stable states. The writing gun is used to move the
arrangement of Fig. 21.50.
target from the lower to the upper stable state.
Figure 21.50 shows the secondary emission ratio The principle of secondary emission storage
curves for the electron arrangement of Fig. 21.50. described in this section, is applicable to both variable
Suppose that the target is at point A, the erased state. persistence and bistable storage, as described in the
The writing gun is truned on and its high energy next two sections.
electrons result in a large amount of secondary
21.23.2 Variable Persistence Storage
emission, and drive the target towards the written
state. Up to point C the flood gun opposes the writing The variable persistence storage technique is also
gun, but thereafter it aids it; so the writing gun can be known as halftone storage or mesh storage.
swtiched off and the target will reach the written state Figure 21.52 illustrates the construction of a CRT using
at D. To erase the target the collector mesh is this storage technique. There are two screens, a
Cathode Ray Oscilloscope (CRO) 671
storage mesh which retains the image traced on it by
the writing gun, and the phosphor screen, which is
very similar to that used in a conventional CRT.
Dielectric material, consisting of a thin layer of
material such as magnesium fluoride, is deposited on
the storage mesh, and this acts as the stroage target.
The writing gun is at a high negative voltage, the flood
gun at a few volts negative, the collector mesh at about
100 V positive, and the storage mesh at ground
potential or a few volts negative. The collimator
consists of a conductive coating on the inside surface
of the CRT. It is biased so as to distribute the flood gun
electrons evenly over the surface of the target, and
causes the electrons to be perpendicular to the storage
mesh.
When the writing gun is aimed at the storage
target it causes areas where it strikes to be charged to a
positive potential, due to secondary emission as in
Fig. 21.51. These areas are maintained at their upper
stable point, even after the writing gun is switched off,
due to the action of the flood gun. Electrons from, the
flood gun also pass through those areas which are
positively charged, causing the phosphor beyond to
glow, and displaying the original signal traced by the
writing gun.
Fig.21.52 Construction of a variable persistence
storage tube.
The pattern stored on the storage target lasts for
about one hour, but it can display a bright image for
about one minute. The stored pattern fades due to
electrons from the flood gun charging other parts of
the storage surface, giving an impression that the
whole screen has been written. This is known as
fading positive. To erase an image which has been
stored the storage mesh is momentarily raised to the
same positive potential as the collector mesh.
The variable persistence storage tube is capable of
producing halftone or gray scale images. As the
charge on the storage surface is increased, beyond a
critical threshold value, more electrons pass through it
onto the display phsophor. The charge on different
parts of the storage screen can be varied either by
control of the time spent by the beam on different
areas, or by varying the beam current over different
areas. Usually four levels can be readily obtained, for
example a mesh voltage of - 10 V represents an off
state, - 5 V and - 2.5 V represent intermediate states,
and 0 V represents full brightness.
Variable persistence tubes enable the persistence
or storage time of the image to be varied from a few
milliseconds to several hours. It can be done in the
CRT shown in Fig. 21.52 by applying a repetitive
negative pulse, of - 4 V to -11 V, to the storage mesh,
so as to discharge the positively charged areas. The
control over the discharge time is usually obtained by
operating the pulses at a constant frequency of about
1 kHz, but varying the pulse width ; the narrowest
pulse gives the largest persistence as it takes longer to
discharge the storage layer.
Variable persistence storage finds many
applications, such as the storage of an entire
waveform of a slow moving signal, which then fades
before the next trace is written. It can also be used to
store several traces before the first one fades, so as to
see how the signal changes with time.
21.23.3 Bistable Storage Oscilloscope
The bistable storage tube is between two and ten
times slower than a comparable variable persistence
tube. However it is capable of much longer storage
times, measured in hours rather than in minutes as for
variable persistence. The bistable tube is also capable
of operating in a split screen mode, where half the
screen has storage capability and the other half is a
conventional phosphor tube.
Figure 21.53 shows the construction of a bistable
storage tube. Unlike the variable persistence tube the
same phosphor screen is used for both storage and
display. The screen consists of Pl phosphor, doped to
have good secondary emission characteristics and
deposited on a conductive backplane made from a
transparent metal film. The phosphor layer consists of
a thin coating of scattered particles, so as to give a
discontinuous surface. This stops the boundary
migration of stored charge. The thin phosphor coating
also has a short life since it suffers from light output
reduction with time.
The conductive film is held at a low positive
potential, so as to attract a cloud of low energy
electrons from the flood gun. These electrons have
insufficient energy to penetrate the phosphor, and are
 672 Electrical and Electronic Measurements and Instrumentation
gathered by the collimator. When the write gun is
switched on, its high energy electrons result in
secondary emission from areas traced on the screen,
moving these to the upper stable point. The trace is
therefore at a high positive potential, and this is
maintained due to the low leakage of the phosphor.
The low energy electrons from the flood gun are now
attracted to the positive areas of the screen, and go
through the phosphor to reach the metal film at the
back. In passing through the phosphor they cause it to
glow, displaying the area traced out by the writing
gun. The screen can be erased by setting the metal film
to a negative voltage, repelling the electrons back into
the storage area, and returning the phosphor to the
low stable point.
Writing
gun
Fig. 21.53 Construction of a bistable storage tube.
21.23.4 Fast Storage Oscilloscopes
Two modifications are made to the conventional
storage CRTs, to enhance the speed with which they
can capture transient information. These methods are
known as transfer storage and expansion storage.
Transfer storage uses an intermediate mesh target,
called the fast mesh, which has been optimized for
speed. The waveform is initially written onto this
mesh and it is then transferred, at lower speed, onto a
second mesh which has been optimized for storage
time. The second mesh can be bistable or variable
persistence.
Fig. 21.55 Block diagram of a basic digital storage oscilloscope.
Expansion storage, shown in Fig. 21.54 uses an
intermediate mesh which is about one fifth the size of
the viewing screen. This mesh can be written on very
quickly, and it is placed between the collector and
accelerator meshes. The image on the storage mesh is
magnified by a static electron lens system, to give a
full size display on the viewing, screen. This system is
capable of writing speeds of 2000 cm/ps, and can
capture signals at frequencies of 300 MHz.
Flood
lens mesh
Fig. 21.54 Expansion storage tube construction.
21.24 DIGITAL STORAGE OSCILLOSCOPES
21.24.1 Principle of Operation
The availability of electronic circuitry at low cost
has enabled many digital features to be added to
analog oscilloscopes. Examples of these are generation
of a trigger after an elapsed time or after a count of a
number of pulses ; digital display of the parameters;
integral digital voltmeter and counter: remote control.
However the basic oscilloscope still remains analog,
and uses an analog storage CRT, as described in
Art. 21.2.
A digital oscilloscope digitises the input signal, so
that all subsequent signals are digital. A conventional
CRT is used, and storage occurs in electronic digital
memory. Figure 21.55 shows a block diagram of a
basic digital storage oscilloscope. The input signal is
digitised and stored in memory in digital form. In this
state it is capable of being analysed to produce a
variety of different information. To view the display
on the CRT the data from memory is reconstructed in
analog form.
Cathode Ray Oscilloscope (CRO) 673
Digitising occurs by taking a sample of the input
waveform at periodic intervals. In order to ensure that
no information is kst, sampling theory states that the
sampling rate must be at least twice as fast as the
highest frequency in the input signal. If this is not
done then aliasing will result, as shown in Fig. 21.56.
This requirement for a high sampling rate means that
the digitiser, which is an analog to digital converter,
must have a fast conversion rate. This usually requires
expensive flash analog to digital converters, whose
resolution decreases as the sampling rate is increased.
It is for this reason that the bandwidth and resolution
of a digital oscilloscope is usually limited by its analog
to digital converter.
a Sample e reduced. Oscilloscopes with up to 40 channels are
(«)
Fig. 21.56 Effect of a low sampling frequency :
(a) input signal, (b) aliased signal.
One method of overcoming the need for a high
performance converter is to use an analog store, as in
Fig. 21.57. The input signals are sampled, and these
are stored in an analog shift register. They can then be
read out at a much slower rate to the analogue to
Fig. 21.57 A digital oscilloscope which uses analogue storage to eliminate the need for a
very fast analogue to digital converter.
digital converter, and the results stored in a digital
store. This method allows operation at up to 100 mega
samples per second, and has the advantage that a low
cost analog to digital converter can be used, whose
resolution does not decrease as the sampling rate is
changed. The disadvantage is that the oscilloscope
cannot accept data during the digitising period, so it
has a blind spot. At low sweep speed operations it is
usual to switch out the analog memory, feeding the
analog to digital converter in real time.
Many different input channels are used with
digital storage oscilloscopes. However if all these
channels share a common store, through a multi­
plexer, then the memory available to each channel is
commercially obtainable, with a storage capability of
25000 dots. Several oscilloscopes also have floppy disc
storage capability to allow non volatile storage of
waveforms, which can later be recalled into the
oscilloscope and manipulated.
21.24.2 Waveform Reconstruction
Although the input signal may be sampled at
greater than twice the highest signal frequency,
aliasing can still result when the output is present as a
series of dots, corresponding to the sampled values.
This is illustrated in Fig. 21.58(a), where the user's
mind connects together the dots which are physically
closest to each other, rather than those which are
closest on the time scale.
In the illustration of Fig. 21.58(a) it is difficult to
visualise the final waveform, and oscilloscopes
generally have the facility to interpolate between the
dots, if required by the user. Two techniques are used,
 674
linear interpolation and sinusoidal interpolation. In
linear interpolation, shown in Fig. 21.58(b) a straight line
is used to connect the dots together. This works well
on a pulsed or square waveform, but not on a
sinusoidal wave, Fig. 21.58(c) shows that sinusoidal
interpolation gives a much better fit for sine waves,
although it is not suitable for pulse or square waves.
(b)
(c)
Fig. 21.58 Illustration of interpolation methods :
(a) without interpolation, (b) Linear
interpolation, (c) sinusoidal interpolation.
f ig. 21.59 Block diagram of a digital storage oscilloscope.
Electrical and Electronic Measurements and Instrumentation
Another practical diagram of a digital oscillo­
scope is shown in Fig. 21.59.
Another problem with the sampling technique
used in digital oscilloscopes is that it can miss short
term transient, or 'glitches', which occur in between
the sample points. To overcome this problem envelope
mode oscilloscopes may be used. These have special
logic circuitry which causes the sample and digitising
circuitry to run at a high speed, independent of the
setting of the display time. At each sample the value is
compared with the previous stored sample, and the
higher (or lower) value is stored. This is continued for
the screen interval, so that for that interval the highest
and lowest points are always stored. For example,
suppose that an oscilloscope digitises every 2 ms, at a
given sweep speed. If a 0.1 ms transient were to occur
there is a high probability that a conventional digital
oscilloscope would miss it. In an envelope mode
oscilloscope the input would be sampled, say, every
200 ns, but only the highest, or lowest, values that
occur within a 2 ms window would be stored in
memory. Therefore the transient would be recorded.
The sample rate of the oscilloscope is controlled by the
time setting of the oscilloscope, but the analog to
digital converter runs much faster.
Cathode Ray Oscilloscope (CRO) 675
21.25 COMPARISON BETWEEN ANALOG AND
DIGITAL STORAGE OSCILLOSCOPES
The advantage of the analog storage oscilloscope
is that it has a higher bandwidth and writing speed
than a digital oscilloscope, being capable of operating
speeds of about 15 GHz. The digital oscilloscope is
primarily limited in speed by the digitising capability
of the analog to digital converter. Aliasing effects also
limit the useful storage bandwidth (usb) of the
oscilloscope to a value given by the ratio.
, maximum digitising rate
usb =----------------- 5-------------- ...(21.26)
constant C
The value of constant C is dependent on the
interpolation method used between the dots. For a dot
display C should be about 25, to give an eligible
display: for straight line interpolation it should be
about 10, and for sinusoidal interpolation C should be
about 2.5.
The digital storage oscilloscope has a CRT which
is much cheaper than an analog storage oscilloscope,
making replacements more economical. The digital
oscilloscope is also capable of an infinite storage time,
using its digital memory. Furthermore it can operate
with a constant CRT refresh time, so giving a bright
image even at very fast signal speeds. The digital
storage oscilloscope is not, however, capable of
functioning in a variable persistence storage mode.
The time base in a digital oscilloscope is generated
by a crystal clock, so that it is more accurate and stable
than an analog oscilloscope, where the time base is
generated by a ramp circuit. The analog to digital
converter used in a digital oscilloscope also gives it
higher resolution than an analog oscilloscope. For
example a twelve bit digitser can resovle one part in
4096. A conventional analog oscilloscope typically
resolves to about 1 mm on the screen, that is to about
one part in 50, equivalent to 6 bit resolution.
Digital storage oscilloscopes are also capable of
operating in a look back mode, as described for
waveform recorders. An analog oscilloscope collects
data after it has been triggered. A digital storage
oscilloscope is always collecting data, and the trigger
tells it when to stop. The oscilloscope can stop
immediately on trigger, so that all the stored
information is pretrigger, or it can stop some time after
it has received the trigger. If the delay is longer than
the storage capability of the oscilloscope, then all the
storage capability of the oscilloscope, then all the
stored information is post trigger, as for an analog
oscilloscope. The digital oscilloscope is also able to
operate in a baby sitting mode. When the scope is
triggered it prints out the stored results onto a hard
copy recorder (or disc storage), and then re-arms itself
ready for another reading.
21.26 ACCESSORIES OF CATHODE RAY
OSCILLOSCOPES
The cathode-ray oscilloscope is one of the most
useful instruments in the electronic industry. The
usefulness of the oscilloscope is further extended by
provision of accessories or auxiliary equipment. Some
of the accessories are described below.
21.26.1 Calibrators
Many oscilloscopes have a built-in reference
source of voltage which has usually a square
waveform with a frequency of 1 kHz. The voltage may
have a single magnitude or may be of several
selectable magnitudes. Provision of this facility
enables the oscilloscope amplifiers and time base to be
calibrated and checked for accuracy.
21.26.2 Probes
The probe performs the very important function
of connecting the test circuit to the oscilloscope
without altering, loading, or otherwise disturbing the
test circuit.
The probes are of three different types :
(/) direct reading probe,
(n) circuit isolation probe, and
(zzz) detector probe.
1. Direct reading probe. This probe is simplest of
all the probes and uses a shielded co-axial cable. It
avoids stray pick-ups which may create problems
when low level signals are being measured. It is
usually used for low frequency or low impedance
circuits. However, in using the shielded probe, the
shunt capacitance of the probe and cable is added to
the input impedance and capacity of the scope and
acts to lower the response of the oscilloscope to high
impedance and high frequency circuits.
External high impedance probes are used to
increase the input resistance and reduce the effective
input capacitance of an oscilloscope. A resistor and capa­
citor combination can be added to an oscilloscope as
shown in Fig. 21.60, in effect, moving the input termi­
nals from the front panel of the instrument to the end
of the probe, supposing, it is intended to attenuate the
signal by a factor of 10. By using Eqns. 21.24 and 21.25,
 676
where k = 10, the values of Rj and Cj can be computed
as under
Rj = R^k-1) =(lx 106)(10-1) = 9MQ
q = c2 /(k -1) =30x 10"12 /(10 -1) =3.33 pF
New input impedance R( is the total resistance,
where R- = R^ + =10MQ
C C,
and C- = —-—— = 3 pF
' q + C2 F input resistance of the oscilloscope. This way the
The input resistance has been increased by a
factor of 10 while, at the same time, the input
capacitance has been decreased by a factor of 10. The
combination of RT and CT is called xlO (times 10)
probe.
Fig. 21.60 External High Impedance Probe.
Capacitor Cv is usually adjustable to compensate
for differences in input capacitance between
oscilloscopes. If the probe capacitance is adjusted to
the wrong value, the oscilloscope will exhibit a factor
frequency response. The adjustment of probe is
usually checked by displaying a square wave on the
CRT screen. If the probe is not properly compensated,
the display of square waveform will be adversely
affected as shown in Fig. 21.61. If the value of Q is too
(a) Correct compensation (b) Value of Q too small
(c) Value of C; too large.
Fig. 21.61 Effect of probe capacitance compensation
on display of square waveform.
Electrical and Electronic Measurements and Instrumentation
small, the leading edge of the square wave is rounded
off but if value of is too large, the leading edge of
square wave overshoots.
2. Isolation probe. Isolation probe is used in
order to avoid the undesirable circuit loading effects
of the shielded probe. The isolation of the probe,
which is used along with a capacitive voltage divider,
decreases the input capacitance and increases the
loading effects are drastically reduced.
3. Detector probe. When analyzing the response
to modulated signals used in communication equip­
ment like AM, FM and TV receivers, the detector
probe functions to separate the low frequency
modulation component from the high frequency
carrier. The amplitude of the modulator carrier (which
is proportional to the response of the receiver to the
much high frequency carrier signal) is displayed on
the oscilloscope by rectifying and bypassing action.
This permits an oscilloscope capable of
audio-frequency response to perform signal tracing
tests on communication signals in the range of
hundreds of MHz, a range, which is beyond the
capabilities of all oscilloscopes except the highly
specialized ones.
21.26.3 Cameras
The best method of permanent records of
oscilloscope traces for analysis is the use for
photography. Special cameras are used for the
purpose and they are of the following two types :
(z) 35 mm film. The 35 mm unit is usually
prefocused and bolted over the tube face. Some
cameras have a fixed lens aperture, the exposure is
simply controlled by hand operation of flapper
shutter. More sophisticated arrangements use
exposure and aperture control. Panchromatic and
special blue sensitive films are commonly used
because they are faster and hence are better suited for
recording of high speed transients. Both these films
give a "negative" record from which positive prints
having white traces on a black background are made.
Another type of 35 mm film gives a photograph which
has a black trace on a white background. These films
are, however, difficult to enlarges.
(zz) Polaroid film. This photographic technique
approaches instant photography. A permanent record
is obtained within 10-20 s. However, this process is
much more expensive than the 35 mm photographic
technique. But the relative cost may be reduced by
using the facility of some of the cameras to be moved
Cathode Ray Oscilloscope (CRO) 677

relative to the screen, making it possible to place From Eqn. 21.17, deflection sensitivity,
several traces on a single print. The camera is again 0.5 x 1.5 x IQ'2
S = —— m/V = 0.375 mm/V.
prefocused, but is normally provided with aperture 2d£fl 2 x 5 x 10-3 x 2000
and exposure controls. The photographic record is
obtained on Polaroid film which is a white trace on a 1 1
Deflection factor, G = - =------- = 2.66 V/mm.
black background. Since in many situations, it is the S 0.375
time which is much more important, the Polaroid Example 21.10 A CRT has an anode voltage of2000 V
photography is normally used under such circum­ and parallel deflecting plates 2 cm long and 5 mm apart. The
stances. screen is 50 cm from the centre of the plates. Find the input
voltage required to deflect the beam through 5 cm. The input
21.26.4 ElectronicSwitch voltage is applied to the deflecting plates through amplifiers
The electronic switch is an extremely useful having an overall gain of 100.
accessory for displaying two signals simultaneously Solution. From Eqn. 21.16, deflection
on single trace oscilloscopes. The electronic switch is
D=d d
essentially a square wave generator. The two signals 2dEan 1
which are to be displayed are each connected to a
separate grid of a dual-section amplifier tube, Voltage applied to deflecting plates
operating with cutoff bias. The square wave is applied _2dEa D _2x5xl0“3x2000x3xl0'2
in such a manner that it alternately unblanks each tube d~ Lld ” 03x2 xlO-2
section, thereby presenting each signal alternately to
= 100 V.
the vertical amplifier of the oscilloscope. When the
switching rate, which is determined by the square .•. Input voltage required for a deflection of 3 cm
wave frequency, is high enough each signal appears = ^ = wo = lv
continuous even though it is actually interrupted gain 100
periodically. The switching rates may be as high as
Example 21.11 Calculate the velocity of the electron*12
100 kHz. beam in an oscilloscope if the voltage applied to its vertical
Example 21.8 Calculate the maximum velocity of the deflection plates is 2000 V. Also calculate the cutoff
beam of electrons in a CRT having a cathode anode voltage frequency if the maximum transit time is 1/4 of a cycle. The
of800 V. Assume that the electrons to leave the cathode with length of horizontal plates is 50 mm.
zero velocity. Charge of electron = 1.6x 10~19 C and mass of Solution. Velocity of electron beam
electron = 9.lx 10~31 kg.
12 x 16x10 x 2000 =26 5 x 106 m/s
Solution. From Eqn. 21.5, velocity of electron is :
V 9.1 xlO'31
Cutoff frequency
__vox _ 26.5xlO6
2xL6xlO~19 dc~ 41 -4x50xl0-3
9.1 xlO-31 = 132.5 x 106 Hz = 132.5 MHz.

= 16BxlO3 m/s. Example 21,12 Voltage E1 is applied to the horizontal

input and voltage E^ is applied to the vertical input of a
Example 21.9 An electrically deflected CRT has a final CRO. Ey and E2 have the same frequency. The trace is an
anode voltage of2000 V and parallel deflecting plates 1.5 cm
ellipse. The slope of the major axis is positive. The maximum
long and 5 mm apart. If the screen is 50 cm from the centre
vertical value is 2.5 divisions and the point where the ellipse
of deflecting plates, find (a) beam speed, (b) the deflection
crosses the vertical axis is 1.25 divisions. The ellipse is
sensitivity of the tube, and (c) the deflection factor of the
symmetrical about the horizontal and vertical axis.
tube. Determine the possible phase angles of E^ with respect to Er
Solution. Velocity of beam Solution. Referring to Fig. 21.34(a),
2xL6xlQ~19 Y1 1.25
OX x2000 sin<|> = = 0.5 .-. (|) = 30o
9.1 xlO-31 Y2 " 2.50

= 26.5 m/s. Thus possible phases are 30° or 330°.

 678 Electrical and Electronic Measurements and Instrumentation

Example 21.13 sketches shown in Fig- 21.62 Solution. The spot generating the patterns moves
display Lissajous patterns for cases where voltages of same in clockwise direction.
frequency out of different phase arc connected to Y and X y n
(a) sin<j> = -i = -=0
plates of the oscilloscope. Find the phase difference in each y2
case. The spot generating the patterns moves in a clockwise
direction. Calculate the angles if the spot generating the
.-. 4>=o°
patterns moves in the anticlockwise direction. (as the Lissajous pattern is in 1st and 3rd quadrants).
25
(b) sin<j> = y =0.5-

.-. <b = 30°

(as the major axis of the pattern is in 1st & 3rd quadrants).
35
(c) sin<|) = — =0.7

.-. (j) = 45°

(as the major axis of the pattern is in 1st & 3rd quadrants).
25
(d) sin<|> = —=0.5

.-. (j> = 180°-30°=150°

(as the major axis is in 2nd and 4th quadrants).
If the spot generating the patterns moved in the
counter clockwise direction, the angles would be :
(a) 180°, (&) -30°,
(c)-45°, (d) 180+30 =210°.
Fig. 21.62 Diagrams of Example 21.13.

Review Questions
1. Describe an overview of applications of a CRO. 9. Describe in details the vertical amplifier used in a
2. Draw the block diagram of a general purpose CRO CRO.
and explain the functions of the following controls 10. Describe the phenomenon of synchronization of
(i) intensity vertical input signal to its sweep generator.
(ii) focus 11. Describe the different types of sweeps used in a
CRO. Explain their spheres of application.
(iii) horizontal and vertical positioning
12. Describe the function of attenuators in CROs.
(iv) synchronization.
Explain how are they designed with particular
3. Describe the different parts of a CRT.
reference to frequency compensation.
4. Derive an expression for vertical deflection of an 13. Describe the following :
electron beam in a CRT.
(i) Sources of Synchronization.
5. Describe the different types of phosphor materials
used in a CRO and list their applications. (ii) Blanking circuit.

6. Explain the different types of graticules used in a (iii) Z-axis modulation.

CRO. Describe their advantages and disadvantages.
7. Explain the functioning of a time base generator in
a CRO.
8. What are the different types of amplifiers used for
CROs ? Describe the basis on which they are
classified.
(iv) Astigmatism control.
14. Describe the following types of oscilloscopes
(i) Dual trace type (ii) Dual beam type.
15. Describe in details the construction and working of
an analog type storage oscilloscope. Explain the
principle of secondary emission.
Cathode Ray Oscilloscope (CRO) 679

16. Explain the following features of an analog type 18. Discuss the advantages and disadvantages of
storage oscilloscope : analog and digital type of oscilloscopes.
(z) bistable persistence storage 19. Discuss in details the delayed sweep.
(iz) bistable storage 20. Describe how the following measurements can be
(in) fast storage. made with the use of a CRO :
17. Describe the principle of working and circuit (z) frequency
diagram of a digital oscilloscope. (ii) phase angle.

Unsolved Problems
1. The deflection sensitivity of an oscilloscope is 4. Compute the value of resistors in the attenuator
35 V/cm. If the distance from the deflection plates network shown in Fig. 21.65 to provide the
to the CRT screen is 16 cm, the length of the volts/division values indicated for each position.
deflection plates is 2.5 cm, and the distance
between the deflection plates is 1.2 cm. What is the
acceleration anode voltage ? [Ans. 583 V]
2. What is the amplitude of the voltage Vo in
Fig. 21.63 after 20 ms ? [Ans. 22 V]

Vf. = 40 V

R = 500 kQ

^0
C = 0.05 nF

Fig. 21.63 Circuit for Problem 2.

3. Identify each of the items indicated in the drawing

in Fig. 21.64. [Ans. 22 V]

Fig. 21.65 Attenuator network for problem 4.

[Ans. R, = 6 kQ, R^ = 9kQ, Ry = 45kQ,

R4 = 90kQ , R5 = 450kQ, R
* = 900 kQ,
4 =4.5MQ, ^=9X40]
Fig.21.64 Diagram for problem 21.3.
 680 Electrical and Electronic Measurements and Instrumentation

5. The waveform shown in Fig. 21.66 is observed on a
CRT screen. If the time/div switch is set 10 gs and
the volt/div switch is set to 200 mV. Determine the
frequency and peak-to-peak amplitude of the
signal. [Ans. 33.33 kHz, 600 mV]
11. Voltage
12. A Lissajous pattern on an oscilloscope is stationary
Fig.21.66 CRT display for problem 5.
6. What value should Q have for Vo to be equal to
0.1V in the circuit shown in Fig. 21.67 ?
13.
the deflecting voltage required to cause the beam
to strike a deflecting voltage and find the
corresponding deflection of the screen.
[Ans. 200 V ; 4 cm]
£[ is applied to the horizontal input and E^
to the vertical input of a CRO. and have same
frequency. The trace on the screen is an ellipse. The
slope of major axis is negative. The maximum
vertical value is 3 divisions and the point where
the ellipse crosses the vertical axis is 2.6 divisions.
The ellipse is symmetrical about horizontal and
vertical axis. Determine the possible phase angle of
E^ with respect to El. [Ans. 120° or 210°]
and has 5 horizontal tangencies and 2 vertical
tangencies. The frequency of horizontal input is
1000 Hz. Determine the frequency of vertical input.
[Ans. 2500 Hz]
In the oscilloscope pattern shown in Fig. 21.68 the
signal connected to the vertical plates has the same
frequency in parts (a) and (b). If the input to the
horizontal plates is a 50 Hz voltage :
(a) state the unknown frequency, and
(b) explain what causes the difference in appear­
ance of two patterns. [Ans. 25 Hz].

[Ans. 5 pF]
7. If the vertical amplifier of an oscilloscope has a
bandwidth of 15 MHz, what is the fastest rise time
that an input may have to be displayed without
distortion ? [Ans. 23.33 ns]
8. A high-impedance probe with 9MQ resistance and
4 pF capacitance is connected to an oscilloscope
with an input resistance of 1 MQ. If the effective Fig. 21.68 Diagram of problem 13.
capacitance decreased to 3.6 pF when the probe
was connected. What is the capacitance of the 14. Prove that the Lissajous patterns produced by
oscilloscope alone ? [Ans. 36 pF]
voltages applied to Y and X plates as shown in
Figs. 21.69 (a) and (b) have the following frequency
9. In a cathode ray tube the distance between the
ratios :
deflecting plates cm, the length of the
deflecting plates is 4.5 cm and the distance of the (fl) 5 : 4 (ft) 8:1.
screen from the centre of the deflecting plates is
33 cm. If the accelerating voltage supply is 300 volt,
calculate deflecting sensitivity of the tube.
[Ans. 2.48 mm/V]
10. An electrostatically deflected cathode ray tube has
plane parallel deflecting plates which are 2.5 cm
long and 0.5 cm apart, and the distance from their
centre to the screen is 20 cm. The electron beam is
accelerated by a potential difference of 2500 V and
is projected centrally between the plates. Calculate Fig. 21.69 Diagram of problem 14.
Cathode Ray Oscilloscope (CRO) 681
15. Find the frequency of the vertical plates if the
frequency applied to horizontal plate is 50 Hz for
the patterns shown in Figs. 21.70 (a) and (b).
[Ans. (a) 25 Hz, (b) 100 Hz]
@ o
(«) (b)
Fig. 21.70 Diagram of problem 15.
16. A sinusoidal input is applied to the vertical input
of an oscilloscope starting at t = 0. The following
Lissajous patterns are obtained when a sinusoidal
input is applied to the horizontal terminals. Estimate
the phase shift between vertical and horizontal
inputs. [Ans. 0°, 90°, 180°, 45°, 60° leading]
(a) (b) (c)
(<0 (e)
Fig. 21.71 Lissajous figures of problem 16.
17. If the horizontal frequency is 51 Hz, determine the
vertical frequency for the Lissajous pattern shown
in Fig. 21.72
Fig. 21.72 Lissajous figure for problem 17.
Objective Type Questions
Tick (^) the most appropriate answer :
1. The source of emission of electrons in a CRT is :
(a) PN function diode
18. Calculate the ratio of vertical to horizontal
frequencies for an oscilloscope which displays the
following Lissajous figures shown in Fig. 21.73.
The input in the case of (d) is applied to Z-axis.
(«) (W
S3 O (C) W
Fig. 21.73 Lissajous Figures of problem 18.
[Ans. A /fy: (a) 2 : 1 (b) 1 : 2 (c) 2 : 3 (d) 1 : 8]
19. Determine the phase shift between two sine waves,
which is indicated by the pattern shown in
Fig. 21.74. [Ans. 53°]
Fig. 21.74 CRT display for problem 19.
20. A sampling oscilloscope is being used to observe a
400 MHz sine wave. A sampling pulse occurs
every 3 ns. Draw five cycles of the 400-MHz signal
and place a dot at the sampled point on each of the
five cycles.
T
(b) a barium and strontium oxide coated cathode
(c) accelerating anodes
(d) post-accelerating anodes.
 682 Electrical and Electronic Measurements and Instrumentation

2. In a CRT the focusing anode is located 10. The bandwidth of a CRO is from 0 - 20 MHz. The
(a) between pre-accelerating and accelerating anodes fastest rise time a sine wave can have to be
accurately reproduced by the instrument is
(b) after accelerating anode
(a) 35 ns (b) 35 ps (c) 17.5 ns (d) 0.175 ps.
(c) before pre-accelerating anode
11. The horizontal amplifier should be designed for
(d) none of the above.
3. The deflection of an electron beam on a CRT screen (a) high frequency signals with a fast rise time
is 10 mm. Suppose the pre-accelerating anode (b) high amplitude signals with a slow rise time
voltage is halved and the potential between (c) high amplitude signals with a fast rise time
deflecting plates is doubled, the deflection of the
(d) low amplitude signals with a fast rise time.
electron beam will be :
12. A vertical amplifier for a CRO can be designed for:
(a) 80 mm (b) 40 mm (c) 20 mm (d) 10 mm.
4. If the distance of screen from a CRT to centre of (a) only a high gain (b) only a broad bandwidth
deflection plates is 15 cm, the length of deflection (c) a constant gain times bandwidth product
plates is 2 cm, the distance between plates is 1 cm (d) all the above.
and accelerating voltage is 500 V, the deflection 13. In an oscilloscope, when the unknown signal
sensitivity is : applied to the vertical plates is being synchronized
(a) 33.2 V/cm (b) 0.03 cm/V with the sweep signal applied to horizontal plates,
(c) 60.4 V/cm (d) 0.015 cm/V. the pattern seen on the CRO screen moves towards
5. Post acceleration is needed in a CRO if the the right. It means that,
frequency of the signal is : (a) the frequency of the signal is lower than that of
(g) less than 1 MHz (b) more than 1 MHz the sweep signal
(c) more than 10 MHz (d) more than 10 Hz. (b) the frequency of the signal is greater than that
6. phsophor material is used for display in CRTs for: of the sweep signal
(c) the frequency of the signal is equal to the
(a) photographic applications
frequency of sweep signal
(b) general purpose applications
(d) none of the above.
(c) television applications (d) all the above.
14. An oscilloscope has an input capacitance of 50 pF
7. An aquadag is used in a CRO to collect : and a resistance of 2 MQ and the voltage divider
(a) primary electrons ratio of 10. The parameters of a high impedance
(b) secondary emission electrons probe are
(c) both primary and secondary emission electrons (a) q =5.55 pF, Rj = 18MQ
(d) none of the above. (b) q = 3.33 pF, Rj = 9MQ
8. A thin aluminium film is usually deposited on the (c) q = 11.1 pF, Rt =18MQ
non viewing side of the phosphor because, (d) q = 5.55 pF, Rj = 9MQ
(a) it acts as a heatsink and prevents phosphor bum 15. In a digital storage oscilloscope, the input signals are
(b) the light scatter from the phosphor is reduced (a) directly applied to the oscilloscope
(c) it does not allow the screen to be negatively (b) multiplexed, converted to digital form and
charged stored and applied to oscilloscope
(d) all the above. (c) multiplexed, converted to digital form stored,
9. During the retrace time, the electrons forming the converted to analog form and applied to oscil­
horizontal beam loscope
(fl) move from left to right on the screen (d) applied to amplifier, stored as analog signals,
(b) move from right to the left on the screen multiplexed, converted to digital form, stored
(c) mo"P from bottom to top of screen in digital form, converted to analog form, and
applied to CRO through an amplifier.
(d) move from top to bottom of screen.

1. (b) 2. (fl) 3. (b) 4. (b) 5. (c) 6. (b) 7. (b) 8. (d) 9. (b) 10. (c)
11. (b) 12. (c) 13. (fl) 14. (fl) 15. (d)
 Instruments for Generation
of Waveforms
22.1 SIGNAL SOURCES
Signal sources provide a variety of waveforms for
testing of electronic circuits at low power levels. The
waveforms generated by several different kinds of
instruments, which range in complexity from simple
fixed frequency sine-wave oscillators to highly
sophisticated instruments such as those used for
testing complex communication circuits. Some of the
applications of signal generators include measuring
the frequency response of amplifiers and the
alignment of radio receivers. There are various types
of signal generators, but the following types of
characteristics are common to all types :
(z) the frequency of the signal should be well
known and stable,
(z'z) the amplitude of the output should be
controllable from very small to relatively
large values,
(zzz) the output signal should be free from
distortion.
There are many variations of the above require­
ments, especially for specialized signal generators
such as function gerenators, pulse generators and
pulse frequency generators.
Sine-wave generators, both in the audio and radio
frequency' ranges are called Oscillators. Although, the
terminology is not universal, the term Oscillator is
generally used for an instrument that provides only a
sinusoidal output signal. The term Function
Generator is applied to an instrument that provides
several output waveforms, including sinewave,
square wave, triangular wave and pulse trains as well
as amplitude modulation of the output signal.
Although we speak of oscillators as 'generators' of
signals, it should be emphasized that no energy is
created, it is simply converted from a dc source into ac
energy at some specified frequency.
(683)
22.2 OSCILLATORS
An oscillator is one of the most basic and useful of
the electrical and electronic measuring instruments.
Oscillators provide a convenient source of power of
test voltage for practically measurements. An oscillator
provides a sine-wave signal output. Depending upon
the oscillator design or application, terms such as
oscillator, test oscillator, signal generator or function
generator are used. An oscillator is the basic element
of all signal sources and generates a sine-wave signal
of known frequency and amplitude. This instrument
covers a frequency range from a few Hz to GHz. A
simple oscillator has two basic blocks (z) an oscillator
and (z'z) an attenuator. The oscillator's performance
depends upon the functioning of these two blocks.
The accuracy, stability and freedom from distortion
depends upon the design of oscillator, while the ampli­
tude accuracy depends upon the design of attenuator.
22.2.1 Classification of Oscillators
Oscillators may be classified in several ways. Here
they are classified on two bases : (z) the design
principle used, and (zz) the frequency range over which
they are used.
The classification based on design principle
categorises the oscillators as : (z) Feedback oscillators
and (zz) Negative resistance oscillators.
The classification based on frequency band is :
(0 Audio frequency (AF) 20 Hz - 20 kHz.
oscillators
('■'■) Radio frequency (RF) 20 kHz - 30 MHz..
oscillators
(iii) Video frequency (VF) d.c. - 5 MHz.
oscillators
(iv) High frequency (HF) 1.5 MHz - 30 MHz.
oscillators
(v) Very high frequency 30 MHz - 300 MHz.
(VHF) oscillators
 684 Electrical and Electronic Measurements and Instrumentation
Feedback oscillators are more commonly used.
They use an active device such as an amplifier
whose output is feedback in phase to its input
(positive feedback) to cause regenerative action
resulting in oscillations.
Negative resistance oscillators use an active device
with an inherent negative resistance charac­
teristics which provide its own regeneration and
hence oscillations.
22.3 FEEDBACK OSCILLATORS
A basic feedback circuit is shown in Fig. 22.1.
Fig. 22.1 Basic feedback circuit
Voltage gain of amplifier
G=— ...(22.1)
Ei
where E-] and Eo are phasors and G is a complex
function of frequency. A portion E of output voltage
Eo is feedback into the input.
Feedback factor H = —— ...(22.2)
Eo
where H is a complex quantity.
Now amplifier input voltage E-j is the sum of the
signal Ef and feedback voltage Ey.
Ei “ Ei+ E/ ...(22.3)
Amplifier output voltage
Eq = GE] = G(E; + Ez) = G(E. + HE0)
or E ...(22.4)
0 1-GH
.'. Gain with feedback
G = E° - G standard circuit for variable frequency test circuits in
...(22.5)
f E; 1-GH
Equation 22.5 indicates that if the complex quantity
called Ic'yp gain GH is positive, as it approaches unity
the gain increases without limit. Hence, if the value of
GH = 1 + /0, the overall gain is infinite. Physically
interpreted, this situation means that the signal input
voltage E, can be reduced to zero without affecting the
output voltage. Thus the amplifier can provide an
output without an input and such a self-excited source
is termed as a Feedback Oscillator.
In general the operating conditions for regenerative
action or oscillations of feedback oscillators are :
1. The frequency of a sinusoidal feedback
oscillator is determined by the condition that
the loop phase shift is zero. Another way of
explaining it would be that the phase shift
between the amplifier input signal and the
signal which is returned through amplifier
and the feedback circuit must be equal to
zero.
2. The product of voltage of amplifier, G, and
the feedback factor H must be equal to or
greater than 1.
These two conditions for sustained oscillations
are called Barkhausen Criteria.
It is not necessary to supply an input signal to
initiate oscillators. A noise voltage or a transient is
sufficient to start oscillations. It is usually desirable for
the loop gain to be greater than unity to start
oscillations. However, if the loop gain is too large, the
output waveform will not be sinusoidal as the circuit
will operate under conditions of saturation in this
case.
22.4 TYPES OF FEEDBACK OSCILLATORS
Feedback oscillators are classified into three
categories :
(z) RC (resistance-capacitance) oscillators,
(zz) LC (Inductance-capacitance) oscillators, and
(zzz) Crystal oscillators.
22.4.1 RC Oscillators
RC oscillators may be one of the two following
forms :
(z) Wien's bridge oscillator, and
(zz) Phase shift oscillator.
1. Wien's Bridge Oscillator
The Wien's bridge RC oscillator has become the
the audio frequency range. This oscillator is simple
and stable in operation. Also it is far less cumbersome
than the LC oscillator.
A Wien's bridge oscillator is an audio oscillator is
useful for testing equipment that operates in the
audio-frequency range. Such instruments always
produce a sine-wave signal, variable in both
amplitude and frequency and usually provide a
square-wave output as well. The maximum amplitude
of the output waveform is typically on the order of
Instruments for Generation of Waveforms 685

25 V whereas the range of frequencies covers at least
the audio-frequency range from 20 Hz to 20 kHz. The
most common output impedances for audio oscillators
are 75 Q and 600 Q.
The two most common audio oscillator circuits
are the Wien's bridge oscillator and the phase-shift
oscillator both of which employ RC feedback
networks. The Wien's bridge offers some very
attractive features, including a straightforward
design, a relatively pure sine-wave output and a very
stable frequency.
The Wien's bridge oscillator is essentially a
feedback amplifier in which the Wien's bridge serves
as the phase-shift network. The Wien's bridge is an a.c.
bridge the balance of which is achieved at one
particular frequency. The basic Wien's bridge
oscillator is shown in Fig. 22.2.
The values of Rt and R2 are kept equal through a
mechanically operated switch.
The first condition for oscillations requires that
the phase shift between and E2 to be zero. This can
only happen when the bridge is balanced and as per
the analysis given in Art. 16.8.1, the resonant
frequency f = -—as shown in Eqn. 22.10.
Equation 22.10 indicates that bridge resonance is
independent of the value of other two bridge arms i.e.
R3 and R4.
The second condition for oscillations requires that
the product of amplifier gain G, and the feedback
factor H should be equal or greater than 1. As the
bridge is to be used as a feedback network for the
oscillator, voltage E2 should not be equal to zero in
magnitude but the phase shift between Ej and E2
should be zero.
This means that the value of H should not be
equal to zero.
The impedances Z4 and Z2 have the same phase
angle at resonance since R3 and R4 are pure resistance.
At resonance, therefore :

wC
and , nr 1
Z, =---------------- = (1/-,- i)..—R
2 (1/R) + jcoC 2
Zo 1
Thus E =----- 2_=-E1 ...(22.11)
n Z1 + Z2 3 1 V 7

and E, = —E, ...(22.12)

‘ R3 + R4 1

(The above relationships indicate that both Efl and

Ej, are in phase with £2 at resonant frequency.)
As can be seen, the Wien's bridge oscillator
consists of a Wien's bridge and an operational At null point: E2 = Efl - Efc = 0.
amplifier represented by the triangular symbol. The Thus if R4 /(R3 + R4) = i or R3 = 2 R4, the output
conditions for balance are :
Z4Z4 = Z2Z3 ...(22.6) voltage E2 will be equal to zero. We do not want to
make E2 equal to zero and therefore, the ratio,
= +S ...(22.7)
R4(R3 + R4) should be smaller than - .
R4 r1 c2

and fn =--- , * ...(22.8) Let — =----- -— = - - — where a > 3.

E4 R3 + R4 3 a
E2 _ B?__ Eb _ Eff p 1
If Q = C2 = C and R} = ft, = R, Eqns. 22.7 and 22.8 Then E1 E1 Ej ^3 aj
simplify to
=2 ...(22.9) ...(22.13)
R4
At resonant frequency En I E1 = - and hence
C = —— Hz ...(22.10)
70 2nRC V ' H=l/a.
 686 Electrical and Electronic Measurements and Instrumentation

Thus if the gain, G, of the amplifier is made equal
to a, the condition GH - 1 is satisfied.
It is clear from above that :
(0 The frequency of oscillations is equal to the
resonant frequency^.
(n) At any frequency other than resonant
frequency Efl is not in phase with and
therefore E2 =Efl -E;, is not in phase with Er
Thus the condition for oscillations that E2 be
in phase E1 is satisfied only at resonant
frequency.
(Hi) The relationship GH = 1 is satisfied only at
resonant frequency.
Capacitors, C, are variable air capacitors mounted
on the same shaft. Continuous variation of frequency
is accomplished by varying the value of C. Resistors R
are provided with switches, so that different
frequency ranges may be selected.
In order that the amplitude of oscillations be held
steady over a wide range of frequencies, a negative
feedback is necessary. This negative feedback is
provided by resistors R] and R2. Resistance R2 is often
a temperature sensitive resistor with a positive
resistance temperature co-efficient. R2 is usually an
incandescent lamp operated at a temperature lower
than its illumination level. Since the resistance of the
lamp changes with amplitude of driving signal, thus
changing the voltage division of the bridge arms
accordingly. Thus, as the amplitude of the oscillations
increases, the resistance R2 increases. This reduces the
negative feedback which reduces the gain of the
amplifier and thus amplitude of oscillations is
restored to normal.
The above statement may be elaborated further as
under :
The amplitude of oscillations depends upon how
much the value of GH is greater than unity. If the
feedback factor H is constant, the amplitude is
determined by the value of gain, G of the amplifier. If
G increases, the amplitude increases till the increase is
limited by the non-linear characteristics of transistors.
If the output of the amplifier tends to increase, the
increased current through R2 increases its resistance.
This reduces the value of H and would then tend to
keep the value of product GH constant. Thus the
amplitude is regulated.
A vacuum tube Wien bridge oscillator depends
upon a temperature sensitive element R2. This allows
the operation of the amplifier within its linear region
of its transfer characteristics. Thus we get a low
distortion sinusoidal output. However, when using
transistors, a peak detector circuit is used to provide
negative feedback signal proportional to the output
voltage. This is because the current drawn by an
incandescent lamp is too much of drain on the power
supplies and thus is incompatible with the transistors
and battery power sources. Also the use of an
incandescent lamp causes a thermal lag.
Advantages
(i) The Wien bridge oscillator provides a stable
low distortion sinusoidal output voltage
over a wide range of frequency.
(ii) The frequency range can be selected by using
simple decade resistance boxes.
(iii ) The frequency of oscillations in an RC
network is /0=l/2nRC. Thus unlike L-C
networks where the resonant frequency
varies inversely as square root of C, the
frequency of Wien bridge oscillator varies
inversely as C itself. Thus frequency
variation greater than 10 : 1 is possible with
single sweep of an air dielectric tuning
capacitor.
(iv ) With the addition of a power amplifier to
isolate the oscillator from the load, the circuit
is used to provide test signals for a variety of
applications.
(n) The range of this oscillator is 2 Hz to
100 kHz. The upper frequency limit is fixed
by the amplitude and the phase shift
characteristics of the amplifier. The lower
limit of frequency is fixed by tire practical
limits on the size of range selecting resistors.
Practical circuit of Wien bridge oscillator.
Figure 22.3 shows the circuit diagram of a two state
Wien bridge oscillator. Two stages are necessary in
order to obtain a phase shift of 0° as a single stage
amplifier shifts by 180°.
Fig. 22.3 Practical Wien bridge circuit.
Instruments for Generation of Waveforms 637

The four arms of the Wien bridge are : 1

Now
(/) R and C in series 2nRC
(ii) R and C in parallel Capacitance
(in) resistor R and C = -~—
(iv) resistor R^. 2^/0
=-------- - -1------------- ? =354 kHz
Qi and Q2 are two transistors and the positive
27tx30xl03xl5xl0-3
feedback is applied from the collector of Q2 through
the coupling capacitor C1 to top of the bridge circuit. 2. Phase Shift Oscillator
Example 22.1 Determine the frequency of oscillations When an audio-frequency (AF) oscillator is to be
shown in a Wien's bridge circuit of Fig. 22.4 ; where used over comparatively a limited frequency range, or
q = q = 6 kD, q = q = C =3 nF, on a specified frequency, simplified RC networks are
R3 = 12 kD, R4=6 kD usually employed. For example, Fig. 22.5 shows a
simple "phase shift oscillator", which can be used for
generation of sinusoidal voltages upto a frequency
range of several hundred kHz. The circuit consists of a
single transistor amplifier stage, followed by three
cascaded arrangements of resistor R and capacitor C to
provide the necessary positive feedback.

Fig. 22.4 Wien's bridge oscillator for Example 22.1.

Solution. The conditions of a wien bridge are

fulfilled with :
q/q=12/6=2.
Fig. 22.5 Phase shift oscillator.
With Ry = R, = R and q = C2 = C, the output The single transistor amplifier shifts the phase of
frequency is, any voltage by 180° which appears et its base. The RC
network is necessary to provide an additional phase
------------- 5----------- 5- = 8.84 kHz
2nRC 2ttx6x10*3x3x10 9 shift of exactly 180°. This additional phase shift is
obtained from the RC network only at a specific
Example 22.2 Design a Wien's bridge oscillator
frequency. Hence, at a frequency where the RC
around thefollowing specifications, f =15 kHz, Vcc =±10V,
network provides exact 180° phase shift, the total
phase shift from the base of transistor, around the
Solution. /K4 =100 Ijn = 100z 1 pA= lOOpA. circuit, back to the base is exactly 0°. Thus the feedback
If the peak value of the sinusoidal waveform is voltage is in phase with the voltage input to the
90% of Vcc, we have : amplifier at that frequency. Thus, if the amplification
0.9 Vrr 0.9x10 of the transistor is sufficiently large, the circuit will act
R, + R. =------- = 90 kQ. as an oscillator at that frequency.
3 4 * 100 XlO’6 100 xlO"6*
The phase shift network for the phase oscillator
For balance in a Wien's bridge, R^=2 R4.
is shown in Fig. 22.6. It consists of equal valued
2R4+R4 =90 kQ or R4 =30 kQ. capacitors and resistors connected in cascade. Each
and q = 60 kQ. stage provides a phase shift 60° with the total phase
It is generally convenient to have shift being 180°.
R.1 = qZ = R=30kQ
4
The phase shift oscillator is analysed by ignoring
or R = 30 kQ. any loading effects produced by the amplifier.
 688 Electrical and Electronic Measurements and Instrumentation

The expression for feedback factor is : Solution. From Eqn. 22.16, frequency of
__________ 1_______ oscillations is :
5 ,F 1 6 1 1
/n = = 50 Hz
(coRC)2 l(coCR)3 wCR 276kRC 2^6nx 13 x 100 x IO"6

...(22.14)
Advantages and Disadvantages
(/) The circuit of a phase-shift oscillator is much
simpler than the Wien bridge circuit. This is because
this circuit dispenses with the negative feedback and
the stabilization arrangements.
(n) The phase shift oscillator may be used over a
wide frequency range, from a few Hz to several kHz.
The upper frequency range is limited since at high
frequencies, the impedance of phase shifting RC
network becomes very small, with the result the
amplifier is heavily loaded. To offset this, corrective
measures have to be taken which complicate the
circuit and thus the advantage gained on account of
simplicity of the circuit is lost.
(iii) The circuit is particularly suited for low
frequencies, say of the order of 1 Hz, since these
frequencies can be obtained easily from the circuit by
using large values of R and C which are commercially
available.
Fig. 22.6 Phase shift audio oscillator.
(iv) The phase shift oscillator has decidedly an
In case the phase shift of the feedback oscillator is advantage over L-C tuned oscillators in the low
180°, imaginary components of the Eqn. 22.14 should frequency range as in this frequency range the large
be zero. inductors required are cumbersome and
———5-————=0 ...(22.15) impracticable.
(coRC)3 co CR
(v) The frequency stability is not as good as that
This requires that the frequency of oscillations is, of Wien bridge oscillator. The distortion is greater and
co 1 changing of frequency is in convenient as each
Hz ...(22.16) capacitor is to be individually adjusted.
2 7t 2^6tiRC
22.4.2 L.C. Oscillators and R.F. Oscillators
and con = , 1 — rad
ai/ s -1 ...(22.17)
0 x6rc The first oscillators were the L.C. variety. These
oscillators are particularly suited for the radio
Substituting the value of coQ in Eqn. 22.14, we have, frequency (R.F.) range. There are many types of (L.C.)
1 oscillators and they use a conventional type of L.C.
H=
Vo 1-5/(1/6)4-7(676-6^6) tank (R.F.) circuit. Their principle of operation is
simple and almost similar for any of the many
...(22.18)
variations of the basic circuit. The opeation of the basic
or H= = —— = ...(22.19) circuit is as follows :
Vo 1-5x6 29
An L.C. oscillator uses a tank circuit which
^=-29^ ...(22.20) comprises of an inductor L in parallel with a capacitor
C. The parallel combination of L and C is excited into
which means that the gain of the amplifier must be at
oscillations and the output voltage of the tank circuit
least 29 if the circuit is to sustain oscillations.
is amplified by either a transistor or a vacuum tube
Example 22.3 Determine the frequency of oscillations amplifier. A fraction of the amplified voltage is fed
of a phase shift oscillator with a three section feedback back into the tank circuit by either an inductive or a
network consisting of 13 Q resistors and 100 pF capacitors. capacitive coupling. This is done in order to
Instruments for Generation of Waveforms 689
compensate for the power loss in the tank circuit. This
"regenerative feedback" results in a constant
amplitude output voltage at the resonant frequency of
the tank circuit. The resonant frequency of the tank
circuit is
/0=l/2n7IC.
The Barkhausen criteria must be satisfied to
produce sustained oscillations.
It is evident from the expression for/,, that these
circuits are particularly suited for radio frequencies
since the value of inductance required to produce
oscillations is quite small at these frequencies. The L-C
oscillators can operate at very high frequencies, up to
several hundred MHz.
Armostrong Oscillator
Figure 22.7 shows the basic circuit of an Armstrong
oscillator. Inductor, Llz and capacitor, Clz form the tank
circuit of the oscillator. Coil Ly is inductively coupled to
a coil L2. When the plate supply voltage Ebb is first
connected, current flows through the plate circuit.
This current flows through coil and since L2 is
inductively coupled to coil Lp a voltage is induced in
the later. This voltage is in such a direction that grid of
the tube is driven positive. This increases the plate
current at a faster rate and the induced voltage in coil
Li, increases further. Thus a high positive voltage is
built up across the tank circuit. Capacitor Clz charges
with a positive polarity on its top plate. During this
process, since the grid is positive the grid current
charges the capacitor C$ to the peak value with a
polarity as shown. As the plate current approaches its
saturation point the rate of rise of plate current
Fig. 22.7 Armstrong oscillator circuit.
decreases and therefore the induced voltage
decreases. Capacitor Cg/ therefore, discharges through
resistance Rov. This makes the grid negative and
therefore a chain reaction is set off.
With the grid becoming negative, the plate
current starts decreasing and as a result the magnetic
field of L2 is wiped out. This induces a negative
voltage in coil Lp The capacitor, Cg is still discharging
and this drives the tube to a cutoff region and
therefore the plate current becomes zero. The negative
induced voltage across causes C\ to discharge and
then charge up again to the peak value of the negative
induced voltage.
The tube is cut-off and Q is charged with a
negative polarity at its top plate. During the next half
cycle, the tank circuit takes out the tube from its
cut-off region. Q starts to discharge through Llz and
the grid is raised in potential till the tube starts
conducting again. As soon as the tube conducts, the
energy is transferred from the plate circuit to the tank
circuit. Capacitor, C\ charges up again to the peak
value of the positive voltage. The entire cycle repeats
itself as described earlier. The output is taken from coil
L3 which is inductively coupted to coil L^.
Hartley Oscillator
Figure 22.8 shows the transistorized version of a
Hartley oscillator. This circuit uses only one coil
instead of two coils as Armstrong circuit. The coil has
a tap which corresponds to the common a.c. ground of
the Armstrong circuit. The variable (tuning) capacitor
Cj is now in parallel with the entire coil The
output is now available through an RC coupling
instead of an inductive coupling as in Armstrong
circuit.
Fig. 22.8 Hartley oscillator circuit.
Phase shift network consists of a tapped conductor
consisting of sections and L2 and an justable
capacitor Q to vary the frequency of oscillations.
The feedback factor is H = -1^ / L, ...(22.21)
 690 Electrical and Electronic Measurements and Instrumentation

The negative sign means that there must be a 180 Solution. Frequency of oscillations
phase shift across the amplifier. This is accomplished
/ - r 1
by connecting the amplifier in the inverting configu­ 0 2 71^(1^ + fy)C1 2 71^280 xl0~6x 0.001
ration as shown in Fig. 22.9.
= 300 Hz.
The minimum gain of the amplifier is computed
using Eqn. 22.23 as,
G . = _k=™=_27
mm Lj 10

Using Eqn. 22.24 the value of the feedback resistor

Ifyis
Kf = Gminfy =27x 15 = 405 kQ

Colpitts Oscillator
The Colpitts circuit in its transistorized form is
shown in Fig. 22.11. It is, in fact, another form of basic
Armstrong circuit. The tank circuit consists of a fixed
inductor Lp and two variable capacitors Q and C2 in
series. This circuit differs from the Hartley's circuit
only in the form of tapping. In the Hartley's circuit, the

Fig. 22.9 Basic Hartley oscillator.

The circuit must satisfy the Barkhausen criteria

for sustained oscillations for which GH > 1 or
GH>1/H ...(22.22)
or G^-fy/^ ...(22.23)
Equation 22.23 states that the gain of the amplifier
must be greater than or equal to L2/fy. Using the
equation for the gain of an inverting amplifier
A= -Rf/Rj ...(22.24)
Example 22.4 Determine the frequency of oscillations
Frequency
and the minimum value of Rj to sustain oscillations in a tuning
Hanley oscillator shown in Fig. 22.10.
Fig. 22.11 Colpitts oscillator.

inductor is tapped while in the Colpitts circuit, the

tapping is from the capacitors. The capacitors Cj and
C2 are ganged. The amount of feedback depends upon
the relative values of Q and C2. The smaller the Q the
greater the feedback. As the tuning is varied, the
values of both capacitors increase or decrease
simultaneously, but the ratio of the two capacitances
remains the same.

22.4.3 Crystal Oscillators

In crystal oscillators, the usual electrical resonant
circuit is replaced by a mechanically vibrating crystal.
These crystals exhibit Piezo-electric properties, i.e.,
when mechanical stresses are applied on the opposite
Fig.22.10 Hartley oscillator circuit for Example 22.4.
Instruments for Generation of Waveforms 691
faces of the crystal, electrical charges appear at some
other faces and vice-versa. Thus when an alternating
voltage is applied to appropriate faces, mechanical
vibrations are produced at some other faces. Therefore
if the frequency of alternating voltage is equal to the
frequency of mechanical vibrations, the vibrations will
be intense.
Piezo-electric properties are exhibited by a
number of natural crystal substances like quartz,
Rochelle salt, etc. Quartz is the normally used material
for crystal oscillators because of its frequency control
in oscillators which comes from its permanence, low
temperature co-efficient and high mechanical Q.
Quartz is preferred to Rochelle salt on account of the
above-mentioned qualities even though Quartz has a
lesser Piezo-electric effect.
The electrical equivalent of a crystal is shown in
Fig. 22.12(b). The circuit has two resonant frequencies
(z) Series resonant frequency = 1/2tiVlC
(zz) Parallel resonant frequency
f _ 1 ll+C/C
LC
(«) W
Fig. 22.12 Crystal and its equivalent electric circuit.
It is apparent that /2 is greater than /p But the
value of ratio C/C' is very small and hence the two
frequencies and /2 are close to each other.
Because of the close analogy between a crystal
and an LC network, the crystal oscillator designs are
Fig. 22.13 Crystal oscillator circuit.
similar to those of LC oscillators. Figure 22.13 shows
the circuit of a Pierce oscillator. This circuit is similar
to that of the Colpitts oscillator.
Advantages and Disadvantages
(1) The frequency stability of the oscillations
obtained from a crystal vary and that too over a long
period of time.
(2) The frequency of oscillations generated by
other RF oscillators using electrical resonating (tank)
circuits is effected considerably by changes in supply
voltage, load and temperature. Crystal oscillators do
not face these problems and thus their frequency
stability is high. Hence these oscillators are the
standard means of maintaining frequency.
(3) The vibrating crystal structure has a high
value of Q. Values in excess of 20,000 are always to be
expected and values over 1,000,000 have been
obtained under special circumstances.
(4) Since the ratio of capacitance C/C' is very
small, the coupling between crystal resonator and
external circuit is inherently small.
(5) The crystal oscillators have a very limited
tuning range (or not at all). The crystal oscillators are
used for frequencies greater than 100 kHz. Their use is
mainly at radio frequencies (RF).
22.5 BEAT FREQUENCY OSCILLATOR (B.F.O.)
Before the advent of the Wien bridge oscillator,
the beat frequency oscillator was the most commonly
oscillator to obtain variable frequency output in the
audio frequency (AF) range. Now the use of B.F.O. has
decreased. It is mostly used when it is desired to cover
a very large frequency range with a single dial rotation
(without the need for switching bands).
A beat frequency oscillator circuit is shown in
Fig. 22.14 in block diagram form. One of the oscillators
generates a fixed frequency and the other a variable
frequency. The outputs of these oscillators are fed to a
mixer. The frequency of output of the mixer is the
difference of two frequencies. Thus if the fixed
dial
Fig. 22.14 Beat frequency oscillator.
 692 Electrical and Electronic Measurements and Instrumentation

frequency oscillator generates a frequency of 100 kHz difficulty of making their base regions and because of
and the variable frequency oscillator has a frequency relatively slow motion of minority carriers in semi­
range of 100 kHz to 120 kHz, the output frequency conductors. Frequencies of the order of 100 - 500 MHz
varies between 0 to 20 kHz. Therefore, frequency dial are the upper limit. The tunnel diode operates because
is marked 0 to 20 kHz (i.e., the difference of the two of majority carriers and hence is thus very fast. It is
frequencies). This then is an audio frequency (AF) possible to make tunnel diode oscillators which
output. This output is amplified by an AF amplifier. operate in the microwave frequency region. A simple
It is obvious that circuit of a B.F.O. is much more circuit shown in Fig. 22.15 can be made to oscillate if
complicated than that of a Wien bridge oscillator and the correct bias current is applied to the tunnel diode.
hence B.F.O. is becoming obsolete. The tunnel diode circuits may use negative feedback
for bias stability, just as feedback oscillators do.
22.6 NEGATIVE RESISTANCE OSCILLATORS
Many negative resistance elements are available
for use in negative-resistance oscillators. These are
tetrodes, tunnel diodes, neon tubes, and unijunction
transistors. There are two types of negative resistance
oscillators used in practice.
(i) Dynatron. Its action is based upon the negative
resistance region of the characteristics of a tetrode
coupled with an LC tank circuit.
(zz) Tunnel diode oscillator. This is solid state
equivalent of the dynatron with additional
advantages of very high speed operation. Therefore
this oscillator is very useful in the very high frequency
(VHF) and ultra high frequency (UHF) ranges.

Figure 22.15(a) shows the characteristics of a tunnel

diode. The characteristics are stable if driven by a
voltage source. The voltage current response can be
used to cause or sustain oscillations. The load lines as
shown are unstable ; that is there are two stable
intersections and one unstable intersection. If the
tunnel diode is connected to an impedance which may
be represented by Z = R + jX, we find that the criteria
for instability is : R + RD = 0. Thus the device
resistance, RD, must cancel out the resistance of the
remainder circuit, and the reactance at the frequency
of oscillations must be zero. In order to obtain
sinusoidal oscillations, the above conditions must be
satisfied. If the resistance of the device (tunnel diode)
were linear, there would be no limit to the amplitude
of the oscillations. The amplitude of the oscillations
can be controlled, if the positive resistance line is
almost tangent to the negative resistance line. The
negative resistance will then have an average value
over one complete cycle which is equal to positive
resistance.
The tunnel diode, which has a negative resistance,
is very useful as it may operate at very high
frequencies. It has been difficult to obtain high
frequency operation of transistors because of the
Fig. 22.15 Characteristics and circuit of a
tunnel diode oscillator.
22.7 SQUARE-WAVE AND PULSE GENERATORS
Square-wave and pulse generators are often used
with an oscilloscope as the measuring device. The
waveforms displayed either at the output and or at
some specific points in the system under test provide
qualitative as well as quantitative information about
or device under test. The fundamental difference
between a square-wave generator and a pulse
generator is dependent upon duty cycle. The Duty
Cycle is defined as the ratio of average value of the
pule over one cycle and the peak value are inversely
related to their time duration. Since the average value
and peak value are inversely related to their time
duration, the duty cycle can be defined in terms of
pulse width and the period or pulse repetition time.
pulse width
Duty cycle = £------- :-------- ...(22.25)
Instruments for Generation of Waveforms
A square wave voltage is shown in Fig. 22.16(a)
T/2
:. Duty cycle for a square wave = = 0.5
or according to another definition
_ average value
peak value
_ 1 / 2 peak value
peak value
Average value
= 1/2 peak value
T/2
Period
(a) Square wave
Pulse
width
(b) Pulse wave
Fig. 22.16 Square and Pulse waveforms.
Thus square wave generators produce an output
voltage with equal on and off periods, their duty cycle
is 0.5 or 50%, as the frequency of oscillations is varied.
This is on account of the fact that whatever may be the
frequency, the positive and negative half cycles
always extend over half of the total period.
For a pulse the on and off periods can be varied.
Duty factor for a pulse wave = y
According to the other definition.
Duty cycle for a pulse wave
_ average value _t / T x peak value _
— — t / 1 j
peak value peak value
The duty cycle of a pulse generator may vary
(normally between 50% to 95%) depending upon the
duration of the on period. The very short duration
pulses give a low duty cycle. The pulse generator
generally can supply more power during its on period
than a corresponding square wave generator can,
693
short duration pulses reduce power dessipation in the
component under test. For instance, measurements of
transistor gam can be made with pulses having short
duration which does not cause overheating of
junctions.
Square wave generators are used whenever low
frequency characteristics of a system are being
investigated. These applications include testing of
audio-systems. Square wave pulses are also preferred
to short duration pulses if the transient response of a
system requires a comparatively high settling time.
Peak
value 22.7.1 Pulse Characteristics and
Terminology
The characteristics of a pulse are shown in
Fig. 22.17. The base Une is referred to the d.c. level and
is the line at which the pulse starts and finishes. The
shift of this line from zero volts, or the expected value,
is called the base Une offset. The amplitude of the pulse is
measured from the base line to the steady state pulse
value.
1. Pulse rise and fall time. The pulse rise time is the
time needed for the pulse to go from 10% to 90% of its
amplitude, and the fall time is the time for the trailing
edge to go from 90% to 10%. These times are also
called leading edge and trailing edge transition times.
2. Linearity. The linearity of the pulse is the
deviation of an edge, from the straight line drawn
through the 10% and 90% points, expressed as a per
cent of the pulse amplitude.
3. Pulse preshoot. The pulse preshoot is the g 22
deviation prior to reaching the base line at the start of
the pulse. The overshoot is the maximum height
immediately following the leading edge.
4. Ringing. It is the positive and negative peak
distortion, excluding overshoot.
5. Settling time. It is the period needed for pulse
ringing to be within a specified percentage of the pulse
amplitude, measured from the 90% point of the
leading edge.
6. Pulse droop or sag. It is the fall in pulse
amplitude with time. Pulse rounding is the curvature
of the pulse at the leading and trailing edges.
7. Pulse ividth. The width of the pulse is measured,
in units of time, between the 50% points on the leading
and trailing edges. The pulse period is the time
between equal points on the waveform.
8. Pulse repetition rate. The pulse repetition rate is
a measure of how frequently a pulse occurs. It is equal
to the reciprocal of the pulse period, and is measured
in units of frequency.
 694 Electrical and Electronic Measurements and Instrumentation

9. Duty cycle. The duty cycle of the pulse is the

ratio of its width to its period, usually expressed as a
percentage.
10. Pulse jitter. Pulse jitter is a measure of short
term instability of one event with respect to another,
for example, instability in the starting time, or the
pulse width, or the pulse amplitude. It is usually
expressed as a percentage of the main parameter.

22.8 PULSE AND SQUARE WAVE CIRCUITS

Sine wave
The pulse and square wave generating circuits are output
of two types : D.C. bias
(i) Passive circuits. These circuits shape an input 0V
waveform, usually a sine wave to give a
pulse output. Output
(n) Active circuits. These circuits include a pulse
blocking oscillator or a multivibrator.
The circuits are briefly described as below :
Fig. 22.18 Generating a pulse from a sine wave :
22.8.1 Pulse Shaping Circuit (a) circuit diagram, (b) waveforms.
Figure 22.18 illustrates a circuit in which a total resistance in series with capacitor C and the
sine-wave, from an oscillator, is converted into a pulse transformer has a step down ratio of 1 : N
train. The amplifier is saturated between two limits,
N+l
by the sine wave, and the sharpness of the pulse f— ...(22.27)
depends on the magnitude of the sine wave relative to
the d.c., bias, and the gain of the amplifier. This circuit +V
can also be used to 'clean up' a pulse input having
overshoots, undershoots, rings and so on.

22.8.2 Blocking Oscillator

The circuit, shown in Fig. 22.19 operates by
driving transistor Qj hard into and out of saturation,
so that the output is a series of pulses. The transistor
gives a 180° phase shift, and the transformer provides
a further 180° shift, so the circuit free runs and
maintains oscillations. R} adjusts the frequency of the
oscillator, which is given by Eqn. 22.27, where R is the Fig. 22.19 A blocking oscillator pulse generator.
Instruments for Generation of Waveforms 695
The width of the pulse is determined by the
characteristics of the transformer, mainly the self
inductance and self capacitance of the primary
winding.
22.8.3 Multivibrators
There are three types of multivibrators : bistables
or flip flops, which provide an output pulse for every
two input trigger signals; monostables which give a
timed pulse out for a trigger input; and astable or free
running multivibrators, which give a string of pulses.
The period and pulse widths of these circuits can be
readily adjusted by timing resistors and capacitors.
They are usually made as integrated circuits, and are
widely used in industry.
Astable Multivibrator
The astable or free running multivibrator is widely
used for the generation of pulses. It can be made to
produce either square wave or pulses depending on
the choice of circuit components. A typical
free-running multivibrator is shown in Fig. 22.20.
Fig. 22.20 Astable or free-running multivibrator.
Essentially, the circuit consists of a two-stage
RC-coupIed amplifier, with the output of the second
stage (Q?) coupled back to the input of the first stage
(QP via capacitor Cr Similarly, the output of Qj is
coupled via C2 to the input of Q2. Since the coupling
between the two transistors is taken from the
collectors, the circuit is known as a collector-coupled
astable multivibrator.
The usual qualitative analysis of the circuit
proceeds as follows : When the power is first applied
to the circuit, both transistors start conducting.
Because of small differences in their operating
characteristics, one of the transistors will conduct
slightly more than the other. This starts a series of
events. Assume that Qj initially conducts more than
Qt This means that the collector voltage of Q}(ecf)
drops more rapidly than the collector voltage of
Q2(e.2). The decrease in exl is applied to R2 C2 network,
because the charge on C2 cannot change
instantaneously, the full negative-going change
appears across R2. This decreases the forward bias on
Q2 which in turn decreases the collector current of
Q2(zc2), and the collector voltage of Q2 rises. This rise
in Q2 collector voltage is applied via the R1C1, network
to the base of increasing its forward bias. Qj
therefore conducts even more heavily and its collector
voltage drops still more rapidly. This negative-going
change is coupled to the base of Q2, further decreasing
its collector current. The entire process is cumulative
until Q2 is entirely cut off and Q-j conducts heavily
(bottoms).
With Q2 cut off, its collector voltage practically
equals the supply voltage, Vcc, and capacitor C]
charges rapidly to Vcc through the low-resistance
path from emitter to base of the conducting transistor
Qp When the circuit action turns fully on, its collector
potential drops to approximately 0 V and since the charge
on C, cannot change instantaneously, the base of Q2 is
at least — Vcc potential, driving Q2 deep into cutoff.
The switching action now begins. C2 begins to
discharge exponentially through R2. When the charge
on C2 reaches 0 V, C2 attempts to charge up to the
value of + VBB, the base supply voltage. But this action
immediately places a forward bias on Q? and this
transistor starts to conduct. As soon as Q2 starts
conducting, its collector current causes a decrease in
collector voltage e^. This negative-going change is
coupled to the base of Q] which starts to conduct less,
i.e., it comes out of saturation. This cumulative action
repeats until Q1 finally cuts off and Q? conducts
heavily. At this instant, the collector voltage of Qj
reaches its maximum value of Vcc. Capacitor C2
charges to the full value of Vcc, and a full cycle of
operation has been completed.
The waveforms appearing at the base and the
collector of each transistor are the result of a
symmetrical or balanced operation : The time constants
and R2C2, the transistors themselves, and the
supply voltages are all identical. The conducting and
non-conducting periods are therefore of almost the
same duration. The waveforms for each of the two
transistors are given in the waveform diagram of
Fig. 22.21.
Assume that at time t=l, transistor Qj is fully
turned on and transistor Q2 is cut off. This makes the
 696 Electrical and Electronic Measurements and Instrumentation

collector voltage ec] of Qj a minimum (practically 0 V) starts conducting again. Obviously, one complete
and the collector voltage ec2 of Q2 a maximum (Vcc). cycle of operation, from time t =1 to time f=3,
Capacitor Q is charging through the emitter-to-base depends on the time required for the base voltage of
resistance of (ff toward the supply voltage VqC and the cutoff transistor to reach the forward bias value.
reaches its full charge rapidly (low emitter-to-base This time depends on two things : the magnitude of
resistance). Since ecl is 0 V, capacitor C2 begins to the reverse bias (-V^c) an<^ ^e time constant of the
charge exponentially through R2 toward the base capacitor charging circuit involved, namely, RpCp or
supply voltage VBB with a time constant equal to R2C2. R2C2.
The analytical evaluation of circuit operation
proceeds as follows :
During its non-conducting period, the collector
voltage of Qi equals
eci = vcc<1-e’1/’3) -<22-28)
where r3 = R3C2-
When Q-, switches on, its collector voltage is at
ground potential and the base voltage of Q2 becomes
— Vcc with respect to ground. The subsequent rise in
Q2 base voltage, through the R2C2 charging circuit, is
described by
+ Vcc )- Vcc ...(22.29)
where r2 = R2C2.
Q2 remains cut off until eb2 reaches the value of 0 V
(by good approximation) and the off time interval T2 of
1 23456789
Q2 can be determined by setting eb2 in Eqn. 22.29 to
Time
zero and solving for t, so that
Fig. 22.21 Waveforms for the astable multivibrator
of Fig. 22.20. o = (VBB + Vcc )(1 )- vcc ...(22.30)
Since the early part of the exponential charging
curve is almost linear, the increase in Q2 base voltage and T2 = t2 In ...(22.31)
(eb2) is indicated by a linear slope on the graph of V +V
VBB + vcc J

Fig. 22.21. Similarly, when Q2 is off and Q2 is bottomed, the

At time t = 2, ec2 reaches a value of approximately collector voltage of Q2 can be described by
0 V placing a forward bias on the base of Q2 which ecl = VCcO-f‘l'"> ...(22.32)
then starts to conduct. Within a very short time, the
where r4 = R4Cr
collector current of Q2 reaches its maximum and
collector voltage ec2 drops to 0 V. When Q2 starts to When now Q2 is switched on, its collector voltage
draw current, the base of Qj becomes negative and Q2 drops to 0 V and the base voltage of Q1 is given by
is quickly driven into cutoff. Its collector voltage, ecl,
reaches the Vcc value and collector current icl en = (VbB + Vcc )(1 ) ...(22.33)
becomes zero. Within a very small fraction of the total where Tj = ^Cj.
Q2 conducting time, capacitor C2 is hilly charged to Solving for the off time interval of Qa by setting
Vqq through the low-resistance emitter-to-base path of ebl in Eqn. 22.33 to zero, we obtain.
q2-
Between times t =2 and t =3, transistor (ff is cut 0 = (VDB + VCcXl-e_'/'1)-Vcc ...(22.34)
off, and its collector current and voltage remain
constant. Similarly, the collector voltage and current and 7] = Tj In ...(22.35)
for Q2 remain constant. Only capacitor Q is charging V +V
VBB+ VCC 7
and the base voltage ebl of 'Qi is rising exponentially
The total period of oscillation is given by,
toward VBB. At time t=3, the base voltage of Qj
exceeds the cutoff value (approximately 0 V) and t=t1+t2 ...(22.36)
Instruments for Generation of Waveforms 697
Fig. 22.22 Block diagram of a pulse generator.
In the case of symmetrical operation, when time
constants RjCj and R2C2 are equal, the waveform is a
symmetrical square wave. By making time constant
RjCj larger than time constant R2C2, the output
waveform becomes a pulse train because the off time
of Qj will be larger than the off time of Q2.
22.8.4 Laboratory Square-wave and
Pulse Generator
The block diagram of a typical general-purpose
generator providing negative pulses of variable
frequency, duty cycle, and amplitude is given in
Fig. 22.22. The frequency range of the instrument is
covered in seven decade steps from 1 Hz to 10 MHz,
with a linearly calibrated dial for continuous
adjustment on all ranges. The duty cycle can be varied
from 25 per cent to 75 per cent. Two independent
outputs are available: a 50 Q source that supplies
pulses with rise-and fall times of 5 ns at 5 V peak
amplitude, and a 600 Q source that supplies pulses
with rise-and fall times of 70 ns at 30 V peak
amplitude. The instrument can be operated as a
free-running generator or it can be synchronized with
external signals. Trigger output pulses for
synchronization of external circuits are also available.
600 Q
output
50 Q
output
Trigger
output
output
The basic generating loop, which is redrawn for
greater clarity in Fig. 22.23 consists of two current
sources, the ramp capacitor, the Schmitt trigger circuit,
and the current-switching circuit (indicated by a
Fig. 22.23 Simplified current source operation.
simple switch). The two current sources provide a
constant current for charging and discharging the
ramp capacitor. The ratio of these two currents is
determined by the setting of the symmetry control
which then determines the duty cycle of the output
waveform. The frequency dial controls the sum of the
two currents from the current sources by applying
appropriate control voltages to the bases of the current
 698 Electrical and Electronic Measurements and Instrumentation
control transistors in the current generators. The size
of the ramp capacitor is selected by the multiplier
switch. These last two controls provide decade
switching and vernier control of the frequency of the
output.
The upper current source, supplying a constant
current to the ramp capacitor charges this capacitor at
a constant rate, and the ramp voltage increases
linearly. When the positive slope of the ramp voltage
reaches the upper limit set by internal circuit
components, the Schmitt trigger (a bistable
multivibrator) changes state. The trigger circuit output
goes negative, reversing the condition of the current
control switch, and the capacitor starts discharging.
The discharge rate is linear, controlled by the lower
current source. When the negative ramp reaches a
predetermined lower level, the Schmitt trigger
switches back to its original state. This now provides a
positive trigger circuit output that reverses the
condition of the current switch again, cutting off the
lower current source and switching on the upper
current source. One cycle of operation has now been
completed. The entire process, of course, is repetitive
and the Schmitt trigger circuit provides negative
pulses at a continuous rate.
The output of the Schmitt circuit is passed to the
trigger output circuit and to the 50 Q and 600 Q
amplifiers. The trigger output circuit differentiates the
square-wave output from the Schmitt trigger, inverts
the resulting pulse, and provides a positive triggering
pulse. The 50 Q amplifier is provided with an output
attenuator to allow a vernier control of the signal
output voltage. In addition to its free-running mode of
operation, the generator can be synchronized or
locked into an external signal. This is accomplished by
Fig. 22.24 Basic pulse generator circuit.
triggering the Schmitt circuit by an external
synchronization pulse.
The unit is powered by an internal supply that
provides regulated voltages for all stages of the
instrument.
22.9 PULSE GENERATOR CIRCUIT
Pulse generators are instruments that produce a
rectangular waveform similar to a square wave but
with a different duty cycle.
There are many applications in systems and
component testing in which pulse generators, used in
conjunction with an oscilloscope, are very useful. The
most basic pulse generator is the astable multivibrator
which generates symmetrical square waves with
variable pulse repetition frequency (PRF) and
provides for the relative lengths of the positive and
negative excursions of the waveform to be adjusted. A
schematic for a basic pulse generator, consisting of an
astable multivibrator and a monostable multivibrator,
is shown in Fig. 22.24. The output of the astable
multivibrator is a square wave. The square wave will
be symmetrical if both capacitors, C, and both
resistors, R, have the same values.
The frequency, which is given as
can be changed, while maintaining the symmetry of
the waveform, by changing both resistors or both
capacitors. The duty cycle of the square wave can be
varied by changing the value of one capacitor or one
resistor. Common practice would be to use several
switch-selectable capacitors to provide different
ranges of frequency and to provide for vernier control
by making one of the resistors variable.
Instruments for Generation of Waveforms 699

The purpose of the "one-short," or monostable Solution. The frequency of the square wave

multivibrator, is to provide a short-duration output generated by the astable multivibrator is computed as

pulse each time a pulse is applied to its input. The _ 0.693
monostable multivibrator has one stable state and one * ~ RC
unstable state. The circuit operates in its stable state = ______ °^3______ = 924 Hz
until an input pulse drives it into its unstable state, 75 x 103 x 0.01 x 10-6
thus producing an output pulse. The duration of the
The pulse width is computed as
output pulse is computed as
t = 0.69 R3C3
R] + R3
t = R3C3 In ...(22.38) = 0.69x 50xl03x0.005xlO-6
-
= 0.173 ms
If resistors R] and R2 are equal in value, as they To compute the duty cycle, we must first compute
frequently are, then the expression simplifies to the pulse period which is
t = 0.693 R3C3 ...(22.39)
T = - =—-— = 1.08 ms
The circuit C2R^ is called a differentiator. Its f 924 Hz
purpose is to provide a short-duration "spike" to
We can now compute the duty cycle as
trigger the monostable multivibrator. The purpose of
diode D3 is to bypass to ground the negative spike _ , Pulse width
Duty cycle =-----------------x 100%
from the differentiator. The waveforms at points A, B, Pulse period
0,173 x!0~3
’ 1.08 xlO'3
= 16%
Some of the typical specifications of a
laboratory-quality pulse generator given below :
A Pulse repetition rate : 0.3 Hz to 20 MHz, five
ranges.
A Pulse width : 25 nsec to 3 sec in five ranges.
A Maximum duty cycle : greater than 80%
from 0.3 Hz to 1 MHz; greater than 50% form
1 to 10 MHz.
Fig. 22.24.
A Output impedance: 50 Q.
Example 22..A Determine the frequency, pulse width, A Amplitude continuously variable from 0.03
and duty cycle of the circuit shown in Fig- 22.25. to 10 V.

Fig. 22.25 Pulse generator circuit for Example 22.6.

 700 Electrical and Electronic Measurements and Instrumentation

22.10 TRIANGULAR WAVESHAPE which simplifies to

GENERATOR ^=VCCR1 -(22-42)
The circuit diagram of a triangular waveshape
generator is shown in Fig. 22.26. At the first stage, From Eqn. 22.42 we can determine the maximum
which is a voltage comparator generates a square amplitude of the triangular output, VQ2, which is
wave output. expressed as :
2^
The output of Aj is driven to saturation; therefore, V«1 = VCC~ -<22«)
the square wave is either at + Vcc or • The second «2
stage, A2, is an integrator which generates a triangular
When the output voltage V02 reaches the
output. The square wave is applied to a square-to-sine
amplitude given by Eqn. 22.43, the output of the
wave converter that filters out the odd harmonics
making up the square wave while passing on only the comparator changes states and the triangular wave
fundamental sine wave. begins to decrease linearly. Since the output is
symmetrical about 0 V. Equation 22.43 also expresses
The operation of the circuit can be analyzed by
the minimum value of VQ2 at which switching occurs.
starring at the output of the comparator, which is at
The wave-forms at V *, VQ1, and V02 are shown in
either + Vcc or -V^c. Consider V01, to be at -V^c. The
voltage VOi will remain at -V^c until the voltage at the Fig. 22.27 for the situation in which R1 = R?.
inverting input of exceeds the voltage at the The frequency of the circuit is controlled by the
noninverting input, which in this case is at zero volt. RC time constant of the integrator. To obtain an
The noninverting input voltage, Vx, is due, in part, to expression for the frequency, we begin with the
the voltage V01, and, in part, to the voltage V02, expression relating capacitor current, charge and time
according to the expression of change:
v = -V __ ____ + v ___ ^2__ ...(22.40) q = ict ...(22.44)
x cc R^ + R, 02 Rl + R2
The rate of charge of the capacitor is
the output VQ1 changes states when Vx =0; therefore,
dq = icdt ...(22.45)
we can say
which can be written as :
0 = -^cc ...(22.41) ic = dq/dt ...(22.46)

flfl. 22.26 Circuit for a basic function generator.

Instruments for Generation of Waveforms 701

The development of Eqn. 22.54 began with Eqn.

22.44, which allows us to compute the charge on a
capacitor after a period of time t. Equation 22.44 is
valid only if the initial charge and, therefore, the initial
voltages on the capacitor are zero. Therefore, the time
t in Eqn. 22.54 is the time for the capacitor to charge
from 0 V until switching occurs, which is at one-four th
cycle as shown in Fig. 22.27. Since t = T/4, Eqn. 22.54
becomes
/
T = 4RC ...(22.55)
I

The frequency, which is the reciprocal of the

period, is now expressed as
Fig. 22.27 Waveforms for the function generator
of Example 22.6. f = —( ...(22.56)
J 4RC^R1 J
As the capacitor charges, the relationship between
charge, capacitance, and voltage across the capacitor Example 22.6 Compute the frequency and the peak
plates is amplitude of the triangular output of the circuit shown in
Fig. 22.28.
q=CVQ2 ...(22.47)
Substituting Eqn. 22.47 into Eqn. 22.46 yields
i=C^^ ...(22.48)
c dt
Since the input resistance of the operational
amplifier is very high, the current through resistor R is
approximately equal to the charging current of the
capacitor. Therefore, we can write
zR = C^^ ...(22.49)

In addition, since the voltage gain of the

operational amplifier is very high, the voltage at the
input to the amplifier is very nearly zero.
Therefore,
Fig. 22.28 Function generator for Example 22.7.

The output frequency can be computed using

Substituting Eqn. 22.50 into Eqn. 22.49, we obtain, Eqn. 22.56 as :
...(22.51) -MAI
4RC1R1>
Integrating both sides of Eqn. 22.51, we obtain
_________ _1__________ (100xl03>
Vm = — [ Vni dt = . (22.52) 4(500xl03)(0.004xl0-6)X^ 60xlO3 ,
02 RCJ 01 RC(t)

Substituting Eqn. 22.43 into Eqn. 22.52, gives = 208 Hz

V ^ = Yoit ...(22.53) The amplitude of the triangular waveform can be
cc r2 rc computed from Eqn. 22.43 as

Since VQ1 = V£c, Eqn. 22.53 simplifies to : 60xlQ3

^02 = 15x
D 100 xlO3
t=RC-± ...(22.54)
^2 = 9 V.
 702 Electrical and Electronic Measurements and Instrumentation
22.11 FUNCTION GENERATOR
A function generator is a versatile instrument that
delivers a choice of different waveforms whose
frequencies are adjustable over a wide range. The
most common output waveforms are the sine,
triangular, square, and sawtooth waves. The
frequencies of these waveforms may be adjusted from
a fraction of a hertz to several hundred kilohertz.
The various outputs of the generator may be
available at the same time. For instance, by providing
a square wave for linearity measurements in an audio
system, a simultaneous sawtooth output may be used
to drive the horizontal deflection amplifier of an
oscilloscope providing a visual display of the
measurement results. The capability of the function
generator to phase lock to an external signal source is
another useful feature. One function generator may be
used to phase lock a second function generator, and
the two output signals can be displaced in phase by an
adjustable amount. In addition, one generator may be
phase locked to a harmonic of the sine wave of another
generator. By adjusting the phase and the amplitude
of the harmonics, almost any waveform may be
generated by the summation of the fundamental
frequency generated by the one function generator
and the harmonic generated by the other function
generator. The function generator can also be phase
locked to a frequency standard, and all its output
waveforms are then generated with the frequency
accuracy and stability of the standard source.
The function generator can supply output
waveforms at very low frequencies. Since the low
frequency of a simple RC oscillator is limited, a
different approach is used in the function generator of
Fig. 22.29. This instrument delivers sine, triangular,
and square waves with a frequency range of 0.01 Hz to
Fig. 22.29 Basic element of a function generator.
100 kHz. The frequency control network is governed
by the frequency dial on the front panel of the
instrument or by an externally applied control
voltage. The frequency control voltage regulates two
current sources.
The upper current source supplies a constant
current to the triangle integrator whose output voltage
increases linearly with time. The output voltage is
given by the well-known relationship.
...(22.57)
An increase or a decrease in the current supplied
by the upper current source increases or decreases the
slope of the output voltage. The voltage comparator
multivibrator changes state at a predetermined level
on the positive slope of the integrator's output
voltage. This change of state cuts off the upper current
supply to the integrator and switches on the lower
current supply.
The loiver current source supplies a reverse
current to the integrator so that its output decreases
linearly with time. When the output voltage reaches a
predetermined level on the negative slope of the
output waveform.
22.12 SIGNAL GENERATING INSTRUMENTS
The previous sections of this chapter have
considered the circuits which go to produce the sine,
pulse and square waveforms used in signal generating
equipment. In the remainder of this chapter the
instruments are introduced. These are categorised as
signal generators, swept frequency generators,
synthesizers, pulse generators; and function
generators, although the dividing lines between these
groups are often blurred.
Instruments for Generation of Waveforms /03

22.12.1 Signal Generators output is a difference frequency (/2_/i) so a slight

These instruments usually produce a fixed change in one of the frequencies, when they are close
frequency sine wave, whose output can be frequency to each other, can give a wide drift in the output. The
or amplitude modulated by another signal. The output also has a considerable amount of noise and
instruments cover a frequency range of 0.001 Hz-50 GHz, spurious signals.
but not from the same device. Figure 22.30 shows a signal
generator circuit. Frequency modulation is achieved by
varying the voltage across a variable capacitance
diode in the tuning circuit of the oscillator. This gives a
system with low output distortion, for modulation
depths below 1% of the carrier frequency. Above this
modulation level the waveform applied to the tuning
diode needs to be deliberately distorted, in order to
compensate for its non-linear characteristics. During Fig. 22.31 A heterodyne oscillator.
frequency modulation manual or automatic methods
Multiplier and divider techniques can be used to
may be used to keep the amplitude of the output
extend the frequency range available from a single
constant.
instrument. In the multiplier generator, shown in
Amplitude modulation is most conveniently done
Fig. 22.32, the output from the fixed frequency
by varying the supply voltage to the oscillator. This
oscillator is fed through a series of tuned multipliers,
method is, however only suitable for small
each having a non-linear amplifier which produces
modulation depths, up to about 50%. It also gives
harmonics. The output from each stage is fed to a
phase modulation due to the effect on the components
tuned filter which selects the high frequency output.
used within the oscillator circuit. Feedback can be
Frequency modulation may be applied to the master
used to reduce output distortion, as shown in
oscillator, and amplitude modulation is achieved by
Fig. 22.30, by detecting the output to obtain the
varying the d.c. supply to the last multiplier stage. The
modulation envelope, comparing this with the
disadvantage of the multiplier generator is that it
amplitude modulation input, and then amplifying
produces a large amount of spurious signal around
and feeding back the difference as the modulation
the desired frequency.
signal. This technique is known as envelope feedback.
The divider signal generator uses a high
The output amplifier shown in Fig. 22.30 provides
frequency master oscillator, whose output is divided
the required signal level, and buffers the oscillator
down by a series of electronic stages. The outputs
from changes in the load impedance. The attenuator is from the divider stages are square waves, which need
needed to give low level output signals. to be filtered to produce sine waveforms. Frequency
The heterodyne principle is sometimes used, as in modulation is applied to the main oscillator, if
Fig. 22.31, to give a continuously variable, wide required, and amplitude modulation is obtained by
frequency range output from a single instrument. The diode modulators in the output amplifier. The divider
signal quality is good, but the frequency stability is generator does not produce any spurious signal at
poor, especially at low frequencies. This is because the frequencies below the required frequency.

Fig. 22.30 A signal generator using envelope feedback for amplitude modulation.
 704 Electrical and Electronic Measurements and Instrumentation

Input for
amplitude
modulation

Optional Optional
Output
output output

Fig. 22.32 A multiplier frequency generator.

22.12.2 Swept Frequency Generator
A swept frequency generator, or sweeper, is a
special type of signal generator in which the output
frequency is cyclically swept through a range of
frequencies. The instrument may have a display built
into it, to show the variation of amplitude with
frequency, or it may provide signals for use with an
external oscilloscope.
Figure 22.33 shows the block diagram of a typical
sweep generator. The time base is usually adjustable,
to give output sweep times in the range from 10 ms to
greater than 100 s. It is also often possible to control
the sweep manually from the front panel of the
instrument. The time base is frequently a triangular or
sawtooth waveform.
Two modes are used to set the swept frequency
range :
(i) The stop start or (f2~ /T) mode. In this mode
the stop and start frequencies are set from
the front panel, and the instrument sweeps
between these limits. This mode is used for
wide sweep widths.
(ii) Delta frequency mode (bf). In this mode the
centre frequency, and the maximum
excursion about this frequency, are set from .
the front panel. It is used for narrow sweep
widths.
The frequency range of the swept frequency
generator usually extends over three bands, 0.001 Hz -
100 kHz (low frequency to audio), 100 kHz-1500 MHz

Manual control

fig. 22.33 Block diagram of a sweep generator using a heterodyne frequency generator.
Instruments for Generation of Waveforms
Manual control
Detector input
(from device
under test)
Fig. 22.34 Block diagram of a sweep generator using a heterodyne frequency generator.
(r.f. range), and 1-200 GHz (microwave range). Three
approaches may be used to cover a wide band of
swept frequencies from a single instrument.
(z) Manually switching between different
frequency oscillators. The problem occurs when the
frequency range needed overlaps two bands.
(zz) Stacked switching. In this the bands are
automatically selected by electronic switches, so that
one can sweep the whole instrument range as one
continuous band.
(zzz) Heterodyne control. In this method,
illustrated in Fig. 22.34, two high frequency signals are
mixed to give a lower difference frequency output, as
one continuous band. If, for example, generator 1 can
be varied over a frequency range 3-5 GHz (i.e., a
change of 2 : 3), and generator 2 has a fixed frequency
output of 3 GHz. then the output can vary from 1 to
2000 MHz (i.e., a change of 2000 : 1). Since the tuning
range needed for this wide output variation small, the
instrument has good linearity. However, a non-linear
mixer will produce many other frequencies in
addition to the required one.
Output level control is used in a swept frequency
generator to keep the output amplitude to the-value
set on the front dial. Usually an r.f. detector monitors
the r.f. amplitude of the output. This is compared with
a signal corresponding to the required amplitude, and
the error is fed back to an electronic attenuator circuit,
to keep the output constant.
The graticule on the display is not accurate
enough to be able to interpret the frequency being
displayed, and some swept frequency generators have
the facility for putting a series of accurately known
frequency markers onto the screen, to act as a
reference. One method of generating the marker
frequencies, as shown in Fig. 22.33, is to pass the
output from a stable crystal oscillator through a
harmonic generator, which gives a series of narrow
pulses, spaced at harmonic intervals. These are mixed
with a sample of the r.f. output from the sweeper. The
mixer will produce a series of low difference
frequency output bursts, called birdies, as the sweeper
frequency approaches and passes through each
harmonic frequency. The birdies are shaped and
amplified, and then combined with the signal received
back from the device under test (via a detector), before
being fed to the display. This gives a composite
picture of the frequency response characteristic, of the
device being tested, and the frequency calibration
markers. An alternative method, of generating a
marker, is to pass a single fixed frequency signal to the
mixer, to give a spot of accurately known frequency
on the screen.
22.12.3 Sweeper Errors
Several errors occur in swept frequency generator
measurements, due to imperfections in the output of
the generator. Any harmonics produced in the
frequency source will be swept along with the
 706 Electrical and Electronic Measurements and Instrumentation
fundamental, but at a faster rate. For example, the nth
harmonics will be swept at n times the rate of the
fundamental. These harmonics may be within the pass
band of the device being tested, for part of the sweep,
and out of the pass band for the rest. This would give a
change in the response of the device, which could be
falsely interpreted as a change in its characteristics.
For example Fig. 22.35, illustrates the effect of the nth
harmonic on a test to measure the flatness response of
a low pass filter. A bump is generated when the nth
harmonic is present, and several overlapping
harmonics will produce a series of bumps. Generally
only the 2nd and 3rd harmonics are of concern, since
the higher harmonics have a much lower amplitude.
Fig. 22.35 Effect of sweep generator harmonics on a
low pass filter response measurement.
The accuracy of the output, and the change in its
amplitude with frequency, is another important
consideration. The amplitude varies with the sweep
rate, since the amplitude feedback correction circuits
have less time to work at faster sweep rate. The effect
of this error can be minimised by sweeping at the
lowest rate acceptable for the device being measured,
and by making readings with respect to the markers,
since these are also affected by a change in amplitude.
The output from a sweeper is not linear, which
means that the spacing between markers is unequal,
so that one cannot easily estimate a value at a
frequency between two markers. The error is usually
very small, but if an exact reading is required then the
sweeper is set to its manual mode. Its frequency can
then be changed until the desired frequency appears
on the display, and this can be read with an external
accurate frequency meter.
22.12.4 Random Noise Generators
A random noise generator produces a signal
output whose instantaneous amplitude varies at
random, and has no periodic frequency component.
Generators are available which cover frequency bands
from low audio to microwave. Random noise
generators have many uses, such as testing of radio
and radar for signal reception in the presence of noise,
and intermodulation and crosstalk tests in communi­
cation systems.
Two parameters are of interest in random noise
generators :
The Noise-Power Spectrum
Usually one is only interested in noise which has a
bandwidth greater than that of the circuit under test.
Within this frequency range the noise power spectral
density can follow one of three curves, as shown in
Fig. 22.36. White noise has a constant power spectral
density from 20 Hz to 25 kHz. Pink noise has higher
amplitudes at lower frequencies (its amplitude varies
inversely as the square root of frequency) and is given
its name because of the similarity of its spectrum with
that of red light. It is used in bandwidth analysis. The
spectrum of USASI noise is approximately equal to the
energy distribution of speech and music frequencies,
and it is used for testing audio systems.
Probability Distribution Function (PDF)
The power spectrum of a noise signal generator
gives its frequency content, but does not characterise
its waveshape; this is done by its p.d.f. The p.d.f. is the
statistical calculation of the portion of time which the
signal spends at the various amplitudes. Usually, for
naturally occurring random signals, this function
follows a Gaussian curve.
Random noise can be generated by using a gas
discharge tube or a zener diode generator. A
semiconductor noise diode gives an output frequency
in the band from 50 kHz to 200 kHz. This can be
amplified and modulated to produce an audio
frequency band. The signal is then passed through a
filter to get one of the spectra shown in Fig. 22.36.
The disadvantage of the random noise generation
method is that sometimes the total output is subject to
long term variations, giving an unpredictable power
Instruments for Generation of Waveforms 707
spectrum. An alternative method is to generate a
pseudo-random noise, such as by clocking a long shift
register having several feedback loops. The noise
pattern will now be repeatable, but within a cycle the
signal behaves as truly random noise. The output from
the shift register is fed through a low pass filter to give
the pseudo-random analogue noise signal.
22.13 FREQUENCY SYNTHESIZER
Frequency generators are of two types :
(i) Free running, in which the output can be
tuned continuously, over a frequency range,
by mechanical or electronic methods. These
are the types described so far, and they have
good overall performance, although their
frequency, accuracy and stability are
relatively poor.
(ii) Synthesizer, which has its output derived
from a fixed frequency, highly stable
oscillator, and covers the range in a series of
steps.
Synthesizers are of two types, direct and indirect.
The direct synthesizer uses a stable crystal oscillator,
followed by a series of harmonic multipliers and
mixers, to provide the range of different output
frequencies. The indirect frequency synthesizer uses a
phase locked loop to give an output which is a fraction
of that of a stable crystal oscillator. The synthesizer
technique provides a generator with excellent
frequency accuracy and stability, but careful design is
needed to overcome the problem of spurious signals
in the output.
Figure 22.37 shows the basic indirect frequency
synthesizer. The output from the phase detector is a
Select
setting
Fig. 22.37 An indirect frequency synthesizer.
signal proportional to ± f2. The low pass filter
selects the signal, which is then amplified and fed to
the voltage controlled oscillator (VCO). The output
from the VCO is fed to a programmable divider, and
then back to the phase detector. The system drives f2
towards/j until they are equal, and at this stage there
is still a phase difference between the two frequencies,
which produces sufficient d.c. signal to keep/2-locked
to f[. The output frequency is Nf and this can be
changed by varying N, usually through a keyboard
from the front panel of the instrument.
The switching speed, between frequency setting,
of the frequency generator depends on the time
needed for/2 to lock to/j and on the setting time. For
large frequency differences the lock time can be long.
It is minimised by setting the VCO close to its final
value, usually under microprocessor control, and then
letting the phase lock loop pull it to its exact value. The
setting time is the time, after lock, for the output to
reach within a given range of the final output setting:
it is typically a few milliseconds.
The problem, with the basic system shown in
Fig. 22.37 is that for a wide frequency range, with
small incremental steps, a low reference frequency,
and a VCO which can cover a wide range, are required.
The programmable divider also needs a large
divider ratio, and this produces noise within the loop
bandwidth. A multiloop frequency synthesizer, as
shown in Fig. 22.38, overcomes these disadvantages.
The low frequency synthesizer has a small range, and
can operate with small steps, whilst the high
frequency synthesizer operates over a wide range
with wide steps. Combining these two signals, in a
mixer, gives a system which can work over a wide
frequency range, with small steps.
 708 Electrical and Electronic Measurements and Instrumentation

Fig. 22.38 An indirect multiloop frequency synthesizer.

22.14 PROGRAMMABLE DECADE

FREQUENCY SYNTHESIER
Figure 22.39 shows an alternative system, which is called a decade synthesier as the fixed dividers have a ratio
of 10. The range of ratios needed from each programmable divider is much less. The fixed divider frequency
synthesier uses a high reference frequency allowing a wide loop bandwidth, and thereby giving a faster lock
time and better noise performance commercial synthesiers are available which cover frequency ranges from
10 kHz to 3000 MHz and from 10 MHz to 30 GHz and with modulation capabilities for use a signal generator
from the same instrument, and by their flexibility. The three basic types of outputs produced are sine, square and
triangular wave, over a frequency range of 1-50 MHz, although some function generators also have the
capability of producing pulses, the performance of the output from a function generator is usually lower than
that from a dedicated instrument.
Select

pig. 22.39 A fixed divider (decadic) indirect frequency synthesizer.

Instruments for Generation of Waveforms 709

Review Questions
1. What are signal sources? What are the desirable
characteristics of a signal? Give an overview of
different signal sources used.
2. Describe the meaning of an oscillator. Describe the
classification of oscillators
(zj depending upon the principle they operate
(if) depending upon the frequency range used by
them.
3. Describe the Barkhausen criteria for production of
oscillations. Describe how the Barkhausen criteria
is met within Wien's bridge oscillator.
4. Describe the principle of working of a Wien's bridge
oscillator. Derive the conditions for its balance.
5. Describe the practical circuit for a Wien's bridge
oscillator and list its advantages and disadvantages.
6. Explain the circuit of a R.C. phase shift oscillators.
Describe how Barkhausen criteria is satisfied in this
oscillator. Give its advantages and disadvantages.
7. Prove that in a R.C. shift oscillator the gain of the
amplifier should be at least 29 to sustain oscillations.
8. Describe the circuit of a Hartley oscillator for
production of radio frequencies. Derive the
condition for production of oscillations.
9. Explain the circuit of a Colpitts' oscillator for
production of radio frequencies. Compare its
Operation with that of Hartley's oscillator.
11. Explain the principle of working of a negative
resistance oscillator and describe the frequency
range over which it is used.
12. Define the duty cycle for a pulse wave. Describe
the characteristics and terminology associated
with pulse waves. 1
13. Distinguish between the active and passive circuits
used for production pulse and square waveshapes.
Describe a passive pulse shaping circuit.
14. Describe in details the circuit and working of an
astable multivibrator.
15. Describe the circuit diagram and working of a
laboratory type square wave and pulse generator.
Explain the working of a current source.
16. Explain with a neat circuit diagram how a
triangular waveshape is produced using a square
wave.
17. Describe the circuit of a square wave generator
which generates square, triangular and sine
waveshapes.
18. Describe the following :
(z) A signal generator using envelop feedback for
amplitude modulation.
(zz) A heterodyne oscillator.
19. Describe the working of a sweep frequency
generator. What are the sweeper errors ?

JO. Describe the working of a crystal oscillator. List its 20. What is a frequency synthesizer? Describe its
circuit details.
advantages and disadvantages.

Unsolved Problems
1. Determine the gain with feedback A for the
amplifier in Fig. 22.40, for sustained oscillations.
[Ans. 19.96]

2. Determine the values of Cj and C2, that will cause

the circuit in Fig. 22.41 to oscillate at 12 kHz.
Assume C, = Q [Ans. 1.32 nF]
Fig. 22.41 Circuit for Problem 2.
 710 Electrical and Electronic Measurements and Instrumentation

3. Determine the frequency of oscillations of a RC phase shift oscillator with a three section feedback network
consisting of 20 Q resistors and 1.5 nF capacitors. [Ans. 2.16 MHz]

4. Determine the value of L2 in the circuit shown in Fig. 22.42 if the frequency of oscillations is to be 100 kHz.
[Ans. 153 pH]

5. Determine the minimum value of 1^ in Fig. 22.42 to sustain oscillations if - 125 pH. [Ans. 83.3 kQ]

Fig. 22.42 Circuit for Problem 4. Fig- 22.43 Circuit for Problem 6.

6. Determine the frequency of the triangular output of the circuit shown in Fig. 22.43. [Ans. 11.72 kHz]
7. Determine the peak to peak amplitude of the output signal for the circuit shown in Fig. 22.43. [Ans. 8 V]
8. Determine the frequency and duty cycle of the astable multivibrator output in the circuit of Fig. 22.44
9. Determine the pulse width of the output pulse VQ in the circuit of Fig- 22.44 [Ans. 69 ps]

Fig. 22.44 Circuit for Problem 8.

10. Design a Wien bridge oscillator around the following specifications :

fQ = 10 kHz Vcc = 115 V, ;in = 1HA Ir4 +
(Refer to Fig- 22-4)
[Ans. R, =90 kQ, R4 = 45 kQ, c = 35-4 pk]
Instruments for Generation of Waveforms 711

Objective Type Questions

Tick (/) the most appropriate answer The frequency of oscillations is :
(a) 8.84 kHz
1. The desirable characteristics of a signal source are:
(b) 4.42 kHz
(a) the frequency of the signal should be well
known and stable (c) 8.84 MHz

(b) the amplitude of a signal source should be (d) No oscillations occur as Barkhausen criteria is
controllable from very small to relatively large not satisfied.
values 6. In an RC phase shift oscillator having three stages
(c) the output signal should be free from distortion which have equivalued resistors and capacitors,
the amplifier, in order to satisfy, Barkhausen
(d) all the above.
criterion should have a minimum gain of
2. In signal generators
(a) 6
(a) energy is created
(b) 29
(b) energy is generated
(c)31
(c) energy is converted from a simple d.c. source
(d) none of the above.
into a.c. energy at some specific frequency
7. In a Hartley oscillator, the inductor is divided into
(d) all the above.
two parts. The inductor which is on the input of the
3. A video frequency oscillator has a frequency range
amplifier has a value of 20 pH while that connected
of:
to the output side of the amplifier has an
(a) 20 Hz to 20 kHz (b) 20 kHz to 30 MHz inductance of 300 pH If the resistance connected to
(c) d.c. to 5 MHz (d) 2 MHz to 30 MHz. the input of the amplifier is 15 kQ, what should be
4. Feedback oscillators have a closed loop gain of the value of resistance in the feedback path of the
amplifier ?
G G
W 1-GH (i?) 1 + GH (a) 300 kQ

G H
(b) 225 kQ
(c)l±GH (d)l+GH (c) 1 kQ
(d) 15 kQ.
where G and H are respectively the gains of
forward and feedback paths. 8. The pulse rise time is defined as the time taken by
the pulse
5. A Wien's bridge oscillator circuit is shown in
Fig. 22.45. (a) to go from 10% to 90% of its amplitude
(b) to go from 0% to 100% of its amplitude
(c) to go from 0% to 90% of its amplitude
(d) to go from 10% to 100% of its amplitude.
9. The duty cycle of a pulse shown in Fig. 22.46 is

Fig. 22.45 Fig. of Q. 5.

Fig. 22.46 Fig. of Q. 9.
The values of components (fl) 33% (b) 25%
K = 6kQ, C = 3nF, Kg=6kQ, R4 = 12kQ. (c) 300% (d) 75%.
 712 Electrical
and Electronic Measurements and Instrumentation

10. An astable multivibrator fisced frequency output of 3 GHz, the output

(«) gives one output pulse for every two input frequency varies from
pulses (a) 0 to 2 GHz
(b) gives a timed output pulse for a trigger input (b) 6 GHz to 8 GHz
(c) gives a train of output pulses for a trigger input (c) 0 to 3 GHz
(J) gives four output pulses for a single input (d) 0 to 5 GHz.
pulse. 14. A random noise generator produces a signal
11. An astable multivibrator uses a resistance of (a) whose amplitude varies randomly
100 kfl and capacitance 0.01 pF. The frequency of
(b) which has no periodic frequency
square waves generated by it is
(c) has an unpredictable power spectrum
(a) 924 Hz (b) .593 Hz
(d) all the above.
(c) 693 Hz (d) 110 Hz
15. A voltage controlled oscillator (Vco) is an instrument
12. A triangular waveshape is obtained :
(a) whose frequency is dependent upon the ampli­
(a) by integrating a square wave
tude of input signal
(b) by differentiating a in wave
(b) whose frequency is independent of the ampli­
(c) by differentiating a square wave
tude of the input signal
(J) by integrating a sine wave
(c) whose frequency is dependent upon the
13. In a sweep frequency generator, two oscillators, frequency of the input signal
one with frequency range of 3 GHz to 5 GHz is
(d) none of the above.
heterodyned with a second oscillator having a

1. (d) 2. (c) 3. (c) 4. (fl) 5. (d) 6. (b) 7. (b) 8. (a) 9. (b) 10. (c)
11. (c) 12. (fl) 13. (a) 14. (fl) 15. (a)
 Signal Analysers
23.1 INTRODUCTION
The analysis of electrical signals is used in many
applications. The different instruments which are
used for signal analysis are wave analysers, distortion
analysers, spectrum analysers, audio analysers and
modulation analysers.
All signal analysis instruments measure the basic
frequency properties of a signal, but they use different
techniques to do so. A spectrum analyser sweeps the
signal frequency band and displays a plot of
amplitude versus frequency. It has an operating range
of about 0.02 Hz-250 GHz. A wave analyser is a
voltmeter which can be accurately tuned to measure
the amplitude of a single frequency, within a band of
about 10 Hz-40 MHz.
Distortion analysers operate over a range of
5 Hz-1 MHz and give a measure of the energy present
in a signal outside a specified frequency band. They
therefore tune out the fundamental signal and give an
indication of the harmonics. An audio analyser is
similar to a distortion analyser but can measure
additional functions, such as noise. Modulation
analysers tune to the required signal and recover the
whole amplitude modulated (a.m.), frequency
modulated (f.m.) or phase modulation (p.m.) envelope
for display or analysis.
23.2 WAVE ANALYSERS
It is well known that any periodic waveform can
be represented as the sum of a d.c. component and a
series of sinusoidal harmonics. Analysis of a wave­
form consists of determination of the values of
amplitude, frequency, and sometime phase angle of
the harmonic components. Graphical and mathe­
matical methods may be used for the purpose but
these methods are quite laborious. The analysis of a
(713)
I
I
f
I
'>
complex waveform can be done by electrical means
using a band pass filter network to single out the
various harmonic components. Networks of these
types pass a narrow band of frequency and provide a
high degree of attenuation to all other frequencies.
A wave analyser, in fact, is an instrument
designed to measure relative amplitudes of single fre­
quency components in a complex waveform. Basically,
the instrument acts as a frequency selective voltmeter
which is tuned to the frequency of one signal while
rejecting all other signal components. The desired
frequency is selected by a frequency calibrated dial to
the point of maximum amplitude. The amplitude is
indicated either by a suitable voltmeter or a CRO.
There are two types of wave analysers, depending
upon the frequency ranges used,
(?) Frequency Selective wave analyser, and
(n) Heterodyne wave analyser.
23.2.1 Frequency Selective Wave Analyser
This wave analyser is used for measurements in
the audio frequency range (i.e., from 20 Hz to 20 kHz).
The analyser has a filter section with a very narrow
passband which can be tuned to the frequency of
interest. The block diagram of this wave analyser is
shown in Fig. 23.1(a).
The waveform to be analysed in terms of its
separate frequency components is applied to an input
attenuator that is set by the meter range switch on the
front panel. A driver amplifier feeds the attenuated
waveform to a high-Q active filter. The filter consists
of a cascaded arrangement of RC resonant sections
and filter amplifiers. The passband of the total filter
section is covered in decade steps over the entire
audio range by switching capacitors in the RC
sections. Close-tolerance polystyrene capacitors are
 714 Electrical and Electronic Measurements and Instrumentation

Variable frequency high-Q active filter

R's varied with frequency. Control
C's varied with frequency multiplier

Normalized frequency
(b) Attenuation characteristics

Fig. 23.1 Audio-range wave analyser and Characteristics of the active filter.

generally used for selecting the frequency ranges.
Precision potentiometers are used to tune the filter
to any desired frequency within the selection pass­
band.
A final amplifier stage supplies the selected signal
to the meter circuit and to an untuned buffer
amplifier. The buffer amplifier can be used to drive a
recorder or an electronic counter. The meter is driven
by an average-type detector and usually has several
voltage ranges as well as a decibel scale.
The bandwidth of the instrument is very narrow,
typically about 1 per cent of the selected frequency.
Figure 23.1(b) shows a typical attenuation curve of a
wave analyzer.
23.2.2 Heterodyne Wave Analyser
For frequency measurements in the megahertz
range, a heterodyne wave analyser construction is
used, as shown in Fig. 23.2. The input signal is fed
through an attenuator and amplifier before being
mixed with a local oscillator. The frequency of this
oscillator is adjusted to give a fixed frequency output
which is in the passband of the i.f. amplifier. This
signal is then mixed with a second crystal controlled
oscillator, whose frequency is such that the output
from the mixer is centred on zero frequency. The
subsequent active filter has a controllable bandwidth,
and passes the selected component of the frequency to
the indicating meter.
Signal Analysers 715
Input
signal
Fig. 23.2 Block diagram of a heterodyne wave analyser.
Good frequency stability in a wave analyser is
obtained by using frequency synthesisers, which have
high accuracy and resolution, or by automatic
frequency control (AFC). In an AFC system the local
oscillator locks to the signal, and so eliminates the drift
between them.
23.2.3 Applications of Wave Analysers
Wave analysers have very important applications
in the following fields :
(z) Electrical measurements
(zz) Sound measurements and
(zzz) Vibration measurements
The wave analysers are applied industrially in the
field of reduction of sound and vibrations generated
by rotating electrical machines and apparatus. The
source of noise and vibrations is first identified by
wave analysers before it can be reduced or eliminated
A fine spectrum anylsis with a wave analyser shows
various discrete frequencies and resonances that can
be related to the motion of machines. Once, these
sources of sound and vibrations are detected with the
help of wave analysers, ways and means can be found
to eliminate them.
23.3 HARMONIC DISTORTION ANALYSERS
The application of a sinusoidal input signal to an
electronic device, such as an amplifier should result in
generation of a sinusoidal output waveform. Generally,
Frequency
control
Bandwidth
control
however, the output waveform is not an exact replica
of the input waveform because of various types of
distortions that may occur.
Distortion may be a result of the inherent
non-linear characteristics of different components
used in an electronic circuit. Nonlinear behaviour of
circuit elements introduces harmonics in the output
waveform and the resultant distortion is often referred
to as harmonic distortion (HD).
Types of Distortion
Distortion is caused by many devices and
components which form an electronic circuit. In this
section the different types of distortions caused by
amplifiers are considered. The various types of
distortions which occur are :
(z) Frequency distortion. This type of dis­
tortion occurs because the amplification
factor of the amplifier is different for
different frequencies.
(zz) Phase distortion. This distortion occurs on
account of the energy storage elements in the
system which cause the output signal to be
displaced in phase with the input signal. If
signals of all frequencies are displaced by the
same amount, the phase shift distortion
would not be noticed. However, in actual
practice, signals at different frequencies are
shifted in phase by different angles and
therefore the phase shift distortion becomes
noticeable.

(nz) Amplitude distortion. Harmonic distortion
occurs due to the fact that the amplifier
generates harmonics of the fundamental of
the input signal. Harmonics always give rise
to amplitude distortion, for example, when
an amplifier is overdriven and clips the
input signal.
(m) Intermodulation distortion. This type of
distortion occurs as a consequence of the
interaction or heterodyning of two freque­
ncies, giving an output which is sum or
difference of the two original frequencies.
(v) Cross-over distortion. This type of
distortion occurs in push-pull amplifiers on
account of incorrect boas levels as shown in
Fig. 23.3.
Crossover
/ \ distortion / \
Fig. 23.3 Illustration of crossover distortion.
23.4 TOTAL HARMONIC DISTORTION
A non-linear system produces harmonics of an
input sine wave, the harmonics consisting of sine
waves with frequencies which are multiples of the
fundamental of the input signal. Total harmonic
distortion (THD) is measured in terms of the harmonic
content of the wave, as given by Eqn. 23.1.
Tun [^(Harmonics)2]172
ItlU =----------------------------------
Fundamental
A measure of distortion represented by a
particular harmonic is simply the ratio of amplitude of
harmonic to that of the fundamental Harmonic
distortion. (HD) is then represented by,
E~ e_ e.
D2 = -2-z D, =^, D.=-±
2 £ 3 p
^1 ci
When Dn (n =2,3,4,...) distortion of nth harmonic
and En represents the amplitude of nth harmonic.
is the amplitude of fundamental.
Total harmonic distortion (THD) is defined as :
THD = /d| + D2 + D2 +...
Je2 + e2 + e2+...
= 5----- 4-----
£i
716 Electrical and Electronic Measurements and Instrumentation
Percentage harmonic distortion
= ^D2+ D2+ D2+... x 100 ...(23.4)
xlOO ...(23.5)
In a measurement system noise is read in addition
to harmonics, and the total waveform, consisting of
harmonics, noise and fundamental, is measured
instead of the fundamental alone. Therefore the
measured value of total harmonic distortion (THDM)
is given by Eqn. 23.6.
E(H)2+(N)^
{Z(F)2 + (H)2 + (N)2}1/2
where H = Harmonics; N = Noise ; F = Fundamental
Equation 23.6 leads to below 1/2% error, due to
the approximation for values of THD below 10%.
Figure 23.4 shows a harmonic distortion analyser
which is used to measure THD. The signal source has
very low distortion and this can be checked by reading
its output distortion by connecting directly into the
analyser. The signal from the source is fed into the
amplifer under test. This generates harmonics and the
original fundamental frequency. The fundamental
frequency is removed by the notch filter.
In the manual system of Fig. 23.4(a) the switch Sw
is first placed in position 1 and the total content of
fundamental and harmonics (Er) is measured. Then
...(23.1)
4 E
^1
...(23.2)
...(23.3) Fig. 23.4 Simplified block diagrams of fundamental
suppression harmonic distortion analysers ;
(a) manual reading, (b) ratio reading.
Signal Analysers 717

the switch is moved to position 2 to measure just the two signal source (/j and /2) frequencies as 7 kHz and
harmonics EH. The value of THD is then found from 60 Hz, having amplitudes in the ratio 1: 4 respectively.
Eqn. 23.6 as in Eqn. 23.7. The lower frequency has a much higher amplitude
since a low degree of non-linearity is expected. In
THD xlOO ...(23.7)
Ej. Europe the values of and f2 are defined by DIN
45403 as 8 kHz and 250 Hz again having an, amplitude
The meter can be calibrated by putting the switch ratio of 1 : 4.
in position 1 and adjusting the reading for full scale
The output from the amplifier under test is fed
deflection. With the switch in position 2 the meter
into a high pass filter, giving a waveform at point A
reading is now proportional to THD. Figure 23.4(1?)
which consists of the high frequency carrier and is low
shows an alternative arrangement, where the values
frequency modulation as in Fig. 23.6. The waveform
of Et and EH are read simultaneously and their ratio
can be observed on an oscilloscope and inter­
calculated and displayed as THD on the indicator. For
modulation distortion (IMD) calculated from (Eqns.
good accuracy the notch filter must have excellent
23.8 or 23.9)
rejection and high pass characteristics. It should
attenuate the fundamental by 100 db or more and the IMD = AmPlitude of modulation( ) * 1QQ
harmonics by less than 1 db. The filter also needs to be Amplitude of carrier (Ac)
tuned accurately to the fundamental of the signal ...(23.8)
source. This is difficult to achieve manually and most
= $-^xl00 ...(23.9)
distortion analysers do this automatically. A common
P
form of notch filter is a Wien bridge. This balances at
one frequency only and at this frequency the output
voltage at the bridge null detector is minimum.

23.5 INTERMODULATION DISTORTION

When a high frequency (fj signal and a low
frequency (f2) signal are mixed in a linear circuit the
output contains the two frequencies only. When
mixed in a non-linear circuit, for example, an amplifier
having distortion, modulation occurs. The output now
contains the original frequencies (^ and /2) and also
the sum and difference frequencies (fA +f2) and
*n addition to several harmonics and their
sum and difference frequencies.
Figure 23.5 shows an instrument for measuring
intermodulation distortion. The SMPTE (Society of Fig. 23.6 Waveform output from an intermodulation
Motion Picture and Television Engineers) defines the distortion meter.

Fig. 23.5 Simplified block diagram of an intermodulation distortion meter.

 Electrical and Electronic Measurements and Instrumentation
Fig. 23.7
The analyser shown in Fig. 23.5 is capable of
calculating IMD directly and displaying its value on
an indicator. The r.m.s. value of the waveform at A is
measured by r.m.s detector 2. The waveform at A is
also demodulated by a detector and the high frequency
components removed by the low pass filter, leaving
the low frequency distortion components. This is
measured by r.m.s. detector 1 and the ratio is taken of
the outputs from detectors 1 and 2 before being
displayed on the indicator.
A special case of intermodulation distortion is
difference frequency distortion, also called CCIF
intermodulation distortion. Figure 23.7 shows the
cons- traction of a difference frequency analyser. A
low pass filter is used after the amplifier under test so
that the meter is sensitive to the low (difference)
frequencies only.
23.6 TRANSIENT INTERMODULATION DISTORTION
Transient intermodulation distortion occurs
because an amplifier is not able to respond rapidly to
changing inputs. It can be tested by applying a square
wave signal with a superimposed high frequency sine
wave, to the input of the amplifier, as in Fig. 23.8(a)
and observing the output on an oscilloscope. If
transient intermodulation distortion is present then
the output will have an initial portion of the high
frequency signal missing as at Fig. 23.8(b), and the
value of T can be noted.
(a) Input signal (b) Output signal
Fig. 23.8 Waveforms to measure transient
intermodulation distortion.
23.7 SPECTRUM ANALYSERS
Spectrum analysis is defined as the study of
energy distribution across the frequency spectrum of a
given electrical signal. The study gives valuable
information about bandwidth, effects of different
types of modulation and spurious signal generation.
The knowledge of the above quantities and pheno­
mena are useful in the design and testing of radio
frequency (RF) and pulse circuitry.
The spectrum analysis is divided into two major
categories on account of instrumentation limitations
and capabilities. They are :
(z) Audio frequency (AF) analysis, and
(zz) Radio frequency (RF) spectrum analysis.
The RF spectrum analysis covers a frequency
range of 10 MHz to 40 GHz, and hence is more
important, because it includes the vast majority of
communication, navigation, radar, and industrial
instrumentation frequency bands.
The spectrum analysers are sophisticated instru­
ments which are capable of portraying graphically the
amplitude as a function of frequency in a portion of RF
spectrum. These instruments find wide applications
for measurement of attenuation, FM deviation, and
frequency in pulse studies.
23.7.1 Basic Spectrum Analyser
The basic spectrum analyser is designed to
represent graphically, a plot of amplitude versus
frequency of a selected portion of the frequency
spectrum under study. The modem spectrum
analyser basically consists of a narrow band
superheterodyne receiver and a CRO. The receiver is
electronically tuned by varying the frequency of the
local oscillator. A simplifed block diagram of a
swept frequency spectrum analyser is shown in
Fig. 23.9.
Signal Analysers 719

Fig. 23.9 Basic swept receiver spectrum analyser.

The circuit incorporates a sawtooth generator

which supplies a ramp voltage to the frequency
control element of the voltage tuned local oscillator.
The local oscillator then sweeps through its frequency
band at a recurring linear rate. The same sawtooth
voltage is simultaneously applied to the horizontal
Fig. 23.10 A three-dimensional presentation of ampli­
plates of the CRO. The RF signal to be tested is applied tude, frequency, and time (a) The addition of a funda­
to the input of the mixer stage. mental and its second harmonic, (b) View seen in the
The sawtooth generator makes the local oscillator Time-Amplitude plane. On an oscilloscope, only the
composite \ + 2f\ would be seen, (c) View seen in the
sweep through its frequency band to beat with the input
Frequency-Amplitude plane. The components of the
signal to produce the desired intermediate frequency composite signal are clearly seen.
(IF). An IF component is produced only when the
corresponding component is present in the RF input 23.7.3 Spectra of Different Signals
signal. The resulting IF signals are amplified, and then Let us consider some of the commonly used
detected. After that they are applied to the vertical signals in order to illustrate the spectra which are
deflection plates of the CRO, thereby producing a displayed on the CRO when they are applied to the
display of amplitude versus frequency on the screen. spectrum Analyser.
23.7.2 Spectral Displays 1. Continuous wave (CW) signals. When a
continuous wave (CW) input signal is slowly swept
Normally a CRO is used for display of electrical
through by the spectrum analyser's local oscillator, the
signals with respect to time, with the X-axis of CRO
response displayed on the screen is a plot of the IF
calibrated to read directly the rise time, pulse width,
amplifier pass band. Since the CW signal has energy at
and the repetition rate. Such measurements are said
only one frequency and therefore the display on the
to be in Time Domain. However, in the case of screen is a single spike. This occurs in case the total RF
spectrum analysers, the signals are broken down into sweep width or Spectrum Width is wide as compared
their individual frequency components and displayed to the IF band width in the analyser.
along X-axis of the CRO which is calibrated in terms of
2. Amplitude modulation. When a continuous
frequency. Therefore, the signal amplitude is displayed
wave signal of frequency, is amplitude modulated
versus frequency. These measurements are then said
by an input signal of frequency, f, two side-band
to be in the Frequency Domain. frequencies of fc + fs and fc - fs are produced. The
A three dimensional representation of a funda­ display on the spectrum analyser screen is a signal of
mental frequency,^ and its second harmonic, f2 -2 fv
is shown in Fig. 23.10. The two signals and the sum
instantaneous values, as displayed on a CRO screen
are shown in Fig. 23.10(b). The two signals are shown
in Fig. 23.10 (c) on amplitude-frequency plane and are
portrayed on the CRO as two components of the
composite signal, as the window of the spectrum
analyser sweeps across the frequency range of the
Fig. 23.11 A single-tone amplitude modulation.
signal. (a) Time-amplitude plot, (b) Frequency-amplitude plot.
 720 Electrical and Electronic Measurements and Instrumentation

frequency with two side-band frequencies whose generally referred to as spectral lines on the analyser
magnitude relative to the carrier frequency depends display. There will be twice as many sidebands or
upon the percentage of modulation. This is illustrated spectral lines as there are harmonic frequencies
in Fig 23.11. contained in the modulating pulse. Figure 23.13(d)
If the frequency, the spectrum width, and the shows the spectral plot which results from rectangular
vertical response of the analyser are calibrated, the pulse modulation of a carrier. The individual lines
display on the CRO screen provides the following represent the modulation product of the carrier and
information in numerical terms, (fl) carrier frequency, the modulating pulse frequency with its harmonics.
(b) modulation frequency, (c) modulation percentage, Therefore, the lines will be spaced in frequency by an
(d) non-linear modulation, (e) incidental frequency
modulation, (/) Spurious signal location and strength.
3. Frequency modulation. If a CW signal fc is
frequency modulated at a rate f, it will produce an
infinite number of sidebands. These are located at
intervals of fc + nfr, where n = an integer, 1, 2, 3... In
Time —>
practice, only the sidebands containing significant
(«)
power are usually considered. A frequency
modulated display is shown in Fig. 23.12.

Frequency, j —>

Fig. 23.12 Amplitude spectrum of single-tone (b)

frequency modulation.

4. Pulse modulation. An idealized rectangular

waveform, with zero rise time and no overshoot is
shown in Fig 23.13(a). This pulse is shown in the time
domain, but when its frequency spectrum is to be
analyzed, it must be broken down into its individual
frequency components. This is shown in Fig. 23.13(b),
where a constant voltage, a fundamental frequency,
and its third harmonic are added algebraically to form
a wave which eventually becomes a square wave as
more odd harmonics are added in phase with the
fundamental. The pulse becomes perfectly rectangular
if an infinitely large number of odd harmonics are
added. A spectral plot, in the frequency domain,
would have the form given in Fig. 23.13(c), where the
amplitude and phases of an infinite number of
harmonics are plotted, resulting in a smooth envelope
as shown.
When this pulse is used for amplitude modulating Fig. 23.13 Pulse modulation : (a) Periodic rectangular
a carrier, the sums and differences of the carrier and pulse train, (b) Addition of a fundamental cosine wave
and its harmonics to form rectangular pulses, (c) Spec­
all harmonic components contained in the pulse are
trum of a perfectly rectangular pulse. Amplitudes and
produced. The harmonic frequencies therefore phases of an infinite number of harmonics are plotted
produce multiple sidebands, in exactly the same resulting in smooth envelope as shown (d) Resultant
manner that the modulating signal in amplitude spectrum of a carrier amplitude, modulated with a
modulation does. These multiple sidebands are rectangular pulse.
Signal Analysers 721

amount equal to the pulse repetition rate of the

original pulse waveform. The main lobe in the centre
and the side lobes are shown as groups of spectral
lines extending above and below the base line. For a
perfectly, rectangular pulse, the number of side lobes
is infinite. The main lobe contains the carrier frequency,
represented by the longest line in the centre.
There are a number of other useful applications of
this instrument. A few typical examples are cited :
Fig. 23.15 Illustration of standing wave
1. Pulse-width and repetition rate measure­ ratio (SWR = U> / 14)
ments.
2. Tuning a parametric amplifier. In most systems some of the forward wave will be
3. FM deviation measurement. absorbed by the load, for example, in the case of an
4. RF interference testing. antenna some of the forward energy is radiated into
space. Therefore the reflected wave has a lower
5. Antenna pattern measurements.
amplitude than the forward wave, resulting in a
23.8 STANDING WAVE RATIO standing wave similar to that in Fig. 23.15. The
When a transmission line is terminated by an standing wave ratio (SWR) is then given by the ratio of
impedance different from its characteristic impedance the maximum and minimum amplitudes of this wave.
some of the forward signal wave is reflected back. This Both voltage and current ratios can be used and these
reflected wave mixes with the forward wave and the are referred to as VSWR and ISWR respectively and
resultant amplitude at any point is the algebraic sum are given by Eqns 23.10 and 23.11.
of the amplitudes of the two waves, as in Fig. 23.14.
VSWR = ...(23.10)
Amplitude Forward Reflected
wave ( wave ISWR = i ...(23.11)
Reflection
j point
' Distance The standing wave ratio is also given by
(Eqns. 23.12 and 23.13) using impedance, power, or
forward and reflected voltage as the parameters.
Node Antinode
SWR = 7 (for
' ZCr > Z,
L7
) ...(23.12)
Fig. 23.14 Illustration of standing wave.

The nodes and antinodes do not move relative to

or SWR = — (for ZL >ZC) ...(23.13)
the transmission line i.e., they are stationary and the
waves are called standing waves.
where Zc is the characteristic impedance of the
An important consideration in transmission line
transmission line and ZL is the characteristic impe­
and antenna design is the standing wave ratio (SWR) dance of the load.
and this is illustrated using Fig. 23.15 which
represents the standing wave on a line. If the line is 1 + (P./Pr)1/2
SWR =—V R F' ...(23.14)
open circuited or short circuited (load impedance 1-(PR/PF)1/2
being infinite or zero respectively), then all the
where Pp is the forward power from the source to
forward wave is totally reflected back to the source
the load and PR is the power reflected back from the
and V1 is zero. For an open circuit line the antinodes
load.
are located at half wavelengths along the line, starting
back from the open circuit point. For a short circuited 1+VVF
SWR = ...(23.15)
line the nodes are located at half wavelengths, starting 1-VR/VF
back from the short circuit point. Therefore although
open and short circuited lines give identical standing where VF is the voltage component of the forward
wave patterns, they are shifted relative to the wave, and VR is the voltage component of the reflected
positions of the nodes and antinodes. wave.
 Electrical and Electronic Measurements and Instrumentation

When the load impedance equals the value of resistor Rx is chosen depending on the
characteristic impedance of the transmission line, the antenna type, i.e., a 50 ohm or a 75 ohm system. When
forward wave is totally absorbed in the load, and there making measurements switch Sw is first set to
is no reflected or standing wave. Now equals V2 in
Fig. 23.15 and the system is ideal having a standing
wave ratio defined as 1 : 1.

23.8.1 Measurement of Standing Wave Ratio

The standing wave ratio (SWR) is one of the most
fundamental parameters as it checks for mismatch
between the transmitter, transmission line and
antenna. The SWR varies with frequency, since the
impedance of the source, line and load are unlikely
to change by the same amount with frequency ; so
the SWR should always be measured at several
frequencies.
Any of the equations (23.10 to 23.15) can be used
to find SWR and meters are available which are based
on all ol these. Most low cost instruments use voltage Fig. 23.16 An SWR meter arrangement based
on a Wheatstone bridge.
or current as their measurement parameter. These
instrumimts give correct readings only when the load position 1 and the value of Ry is adjusted to give full
is not loo reactive that is the SWR ratio being scale defection on the indicator. Having calibrated the
measured is low. The meters must also be connected instrument switch SW is moved to position 2, and the
to take readings at the load feed point or at half value of the SWR is read directly from the indicator.
wavelength distances along the transmission line,
Figure 23.17 shows an alternative SWR meter
back from the load. This usually means that the
construction. The pick up unit consists of three
transmission line needs to have a length which is an
conductors, the centre conductor being the core of the
integral value of its electrical length. The electrical
transmission line and the outer two conductors are
length for half wavelength and quarter wavelength
spaced equally from this. Usually this pick up unit is
transmission lines is respectively given by Eqns. 23.16
and 23.17
El =150F// ...(23.16)
= 75 F / f ...(23.17)
where f = frequency of antenna and
F = velocity factor whose value varies from
0.67 to 0.97 depending upon the dielectric
used.
SWR meters based on power measurements give
results which are independent of length. A true RF
wattmeter is used fitted with a directional coupler so
as to measure the forward and reflected power
separately and the SWR is calculated from Eqn. 23.14.
The radiation resistance of an antenna can also be
measured using a resistance bridge and the SWR
found from (Eqn. 23.12) or (23.13). Figure 23.16 shows
a Wheatstone bridge arrangement in which resistors
Rg, and Rx form three of the bridge arms, and the
radiation resistance of the antenna forms the fourth
arm. The supply to the bridge is a low power R.F Fig. 23.17 An SWR meter arrangement based on
signal generator, which provides the excitation. The forward and reflected wave measurements
Signal Analysers ' '
made as a pattern on a printed circuit board. Resistors
Rx and Ry are selected to have values equal to the
characteristic impedance of the line i.e., 50 ohms or
75 ohms. In operation any R.F. signal on one of the
pick up wires will be rectified by diodes D1 or D2
smoothed by capacitors or C2 and then applied as
d.c. to the indicator. The forward signal from the
transmitter to the antenna picked up by the reflected
wire, is absorbed by resistor Rx so no forward voltage
appears on the pick up line after point X. However
this signal remains on the forward pick up wire, and is
applied to the indicator. Similarly, the reflected signal
is picked up by the reflected pick up wire, but is
absorbed in resistor Ry of the forward pick up circuit.
When making measurements with the meter
shown in Fig. 23.18 the switch is set to position 1 and
Rz is adjusted to give a reading at a set calibration
point. The switch is then moved to position 2 and the
SWR read from the indicator.
A good antenna has a standing wave ratio which
varies between 1 : 1 and 1 : 1.2. Poor antennas have
ratios greater than 1 :1.5 and most SWR meters do not
give a reading beyond 1 : 3.
Instruments are usually multipurpose for
example, combining a standing wave ratio meter with
a field strength meter and power output meter.
Figure 23.18 shows a block diagram of an impedance
Fig. 23.18 Simplified block diagram of an impedance noise bridge.
Review Questions
I
1. Describe the circuits and working of wave
analysers used for audio frequency and megahertz
ranges.
noise bridge, which is a low cost instrument for
measuring several parameters on antennas and
transmission lines ; The noise source produces a wide
spectrum of pseudogaussian noise. This is amplified
and then goes through an isolator before being
coupled to three output. The detector gives an
indication of the received signal and shows a high
noise level at all frequencies which are separate from
the resonant frequency of the device under test
(usually an antenna or a transmission line). The
electrical length of a transmission line differs from its
physical length due to the velocity factor F. The noise
bridge can be used to find the correct physical length
which gives an electrical length equal to half a
wavelength. The transmission line should initially be
cut to a length slightly greater then that given by
(Eqn. 23.16) of (23.17). Rx is set to it low value, and one
end of the cable under test is connected to the test
socket, whilst test is connected to the test socket whilst
the other end is short circuited. The detector is set to
the-frequency at which the cable is to be half a
wavelength in length and the frequency is slowly
decreased until null is reached. Now the cable is
gradually trimmed in stages. After each trim its end is
again short circuited and it is re-tested until
eventually the null point occurs at the required
frequency.
2. Describe the engineering applications of wave
analysers.
 724 Electrical and Electronic Measurements and Instrumentation
3. Explain the different types of distortions caused by
amplifiers.
4. Explain the term "total harmonic distortion".
Describe the functioning of a total harmonic
distortion meter.
5. What is Intermodulation Distortion ? Describe the
working of an intermodulation distortion meter
with the help of a block diagram.
6. Describe the working of a difference frequency
distortion analysers with the help of a block
diagram.
7. Describe the basic circuit of a spectrum analyser.
Explain how the spectra of the following is
displayed :
(i) continuous wave signals
(ii) amplitude modulated signals
(iii) frequency modulated signals
(iv) pulse modulated signal.
8. Explain the following terms
(i) SWR
(ii) VSWR
(iii) ISWR
9. Explain the term Electrical length for transmission
lines. How is this affected by the operating
frequency and the velocity factor ?
10. Describe an arrangement for measurement of
Standing wave ratio.
 High Frequency Measurements
24.1 INTRODUCTION
In this chapter methods for measurement of
inductance, capacitance and resistance in the
frequency range of 0.1 to 100 MHz are described. No
doubt, some of the methods described earlier i.e.,
bridge methods can be used in this frequency range
but these methods give erroneous results on account
of the fact that these methods do not take into
consideration the special effects which are inherent in
any measurement at high frequencies. Some of these
special effects which are predominant at high
frequencies are :
(z) high reactance due to stray series inductance
and its variation with frequency ;
(n) low reactance due to stray shunt capacitance
and its variation with frequency ;
(iii) increase of effective resistance with
frequency on account of skin effects ;
(iv) increase of effective resistance on account of
proximity effects;
(v) increase of effective resistance due to
production of eddy currents ;
(vi) variation of dielectric loss of capacitors with
frequency.
These effects make the high frequency
measurements complex. If bridge circuits are used at
high frequencies, elaborate shielding arrangements
have to be adopted to nullify the effects of factors
listed above.
The normal approach for measurements at high
frequencies is to adopt such methods which are as
simple as possible so that the effects of stray electric
and magnetic fields are easy to remove. Since errors of
the order of 1 to 5 percent are acceptable in
(725)
measurements in this frequency range, the methods
used involve use of very simple apparatus which may
include :
(i) a set of inductors from 1.0 pH to 10 mH.
(ii) a variable air capacitor of maximum
capacitance 1000 pF,
(iii) a decade resistance box of resistance upto
100 Q,
(iv) a thermo electric ammeter or an electronic
voltmeter (EVM) in order to indicate
resistance conditions.
24.2 RESONANCE METHODS
Most of the methods used for high frequency
measurements employ a series RLC resonant circuit as
shown in Fig. 24.1. The emf for the circuit is obtained
by coupling the coil L with the output coil of an
oscillator. A loose coupling is desirable in order that
the changes in conditions of the RLC circuit do not
effect the oscillator circuit. A tight coupling on the
other hand, will affect the oscillator.
Fig. 24.1 RLC resonant circuit.
Resonance is obtained either by variation of
frequency of the oscillator or by variation of the
capacitance and keeping the frequency constant. The
resonance conditions are indicated by the thermal
milli-ammeter connected in the circuit. (A thermal
milli-ammeter is necessary as it can be worked at high
frequencies.) Under resonance conditions the
 726 Electrical and Electronic Measurements and Instrumentation
milli-ammeter indicates maximum reading. However,
in practice it is convenient to use an electronic
voltmeter connected across the capacitor C to indicate
the resonance conditions since the ammeter
introduces an extra series resistance in the circuit
which may be undesirable under many circumstances.
The resonance is indicated when the voltmeter reads
maximum. However, it is interesting to note that the
circuit conditions which give maximum current are
not the same as those which show maximum voltage
across capacitor C, but, for high Q circuits the
difference between the two conditions is negligible.
Hence, in general, it is assumed that when the
voltmeter connected across C shows maximum
reading, the circuit is having maximum current i.e., the
circuit is working under resonance conditions.
Another assumption, which is made for resonance
methods, is that the resonance curve is symmetrical
about its maximum ordinate. This condition is valid
only when :
(z) the curve is highly peaked i.e., highly
selective which is the case when the circuit
has a high Q factor ;
(zz) the readings are taken in the vicinity of the
resonance point. To satisfy the condition
(zzz) measurements should be limited to values of
current greater than 50% of the maximum
value of current.
24.3 MEASUREMENT OF INDUCTANCE
Figure 24.2(a) shows a simple circuit for
measurement of inductance, L, of a coil. This coil
under measurement is loosely coupled to the output
coil of a variable frequency oscillator. The dial of the
oscillator is calibrated in terms of frequency. A
Fig. 24.2 Measurement of inductance.
variably capacitor, C, is used for obtaining resonance
conditions which are indicated by an electronic
voltmeter connected across it
At resonance LC = 1/co2 =1/4k/2 ...(24.1)
Thus knowing the value of frequency, f and
capacitance, C, the value of inductance can be
calculated. The above method apparently looks very
simple but the value of inductance obtained this way
is rarely accurate. The reasons are discussed below :
The above method does not take into account the
additional capacitances which are introduced into the
circuit. Also errors are caused due to the fact that the
dial setting of capacitor, C, may not be accurately
known. To offset the influence of above effects and to
get improved accuracy a number of readings are taken
by varying the frequency, f.
Let Ca be the value of additional capacitance.
Therefore, for resonance at any frequency
L(C+Cfl) = 1/4k2/2 ...(24.2)
1 / /2 = 4k2L(C + Ca )= 4k2LC + 4K2LCfl
...(24.3)
From Eqn. 24.3, it is clear that if a curve is plotted
between 1 / /2 and C, we get a straight line as shown in
Fig. 24.2 (b). The line produced back cuts the x-axis at
O'. The slope of the line is tan 0 = 4k2L
L=tan0/4zi2 ...(24.4)
Thus knowing the value of slope of the line, the
value of inductance, L, may be obtained.
The value of additional capacitance, C , is equal to
the intercept OO'. Now,
Ca=C4+Cr ...(24.5)
where Crf = self-capacitance of the coil,
Cv = capacitance of voltmeter.
Equation 24.5 is true only when the calibration of
capacitance C includes the effects of its screen
capacitances. Under condition, the method affords a
means of measurement of self-capacitance of coils.
In the above analysis we have assumed that the
value of inductance, L, remains constant. However,
this is not strictly true as the effective value of
inductance changes with frequency. Thus this method
gives erroneous results for coils whose effective self­
inductance varies considerably over the range of
frequency used. It is clear from Eqn. 24.3 that we get a
straight line relationship between C and 1 / f2 only if L
remains constant over the range of frequency used.
Thus if the value of self-inductance, L, changes with
frequency we do not get a straight-line but a curve.
High Frequency Measurements 727

Therefore, in order to determine the value of L at a
particular frequency, a tangent is drawn to the curve
at a value the frequency where the value of inductance
is required. The value of inductance is determined
from the slope of this curve using Eqn. 24.4.
Example 24-i A capacitance of 250 pF produces
The input capacitance of the electronic voltmeter is
4.5 pF and the lead capacitance is 1.5 pF.
Solution. The method has been explained in Art.
24.3. A curve is drawn between l//2 and C. The
points for the graph are :
........................

C, pF f, kHz l//2
resonance with a coil at a frequency of (2/n)x IO6 Hz,
while at the second harmonic of this frequency resonance is 132 350 8.16 xlO’12
produced by a capacitance of 50 pF. Calculate : 98 400 6.25 xlO“12
(fl) the self-capacitance, and 73 450 4.93 x1012
(b) the inductance of the coil. 55 500 4.0x10 12
Neglect the effect voltmeter capacitance and also other 31 600 2.78x10'12
stray capacitances.
The graph is plotted in Fig. 24.3.
Solution. Suppose
C = capacitance of variable capacitor,
Cd - Self-capacitance of coil, and
L = inductance of coil.
At resonance,
L=l/4n2/2(C+Crf)
=_____________ 1_______________
4k2 x (2 / ti)2 x 1012(250 + Cd)x IO-12

= ------- - ------- •••(/)

16(250+Crf)
and also
L=__________ 1__________
4n2 x (4 / ti)2 x 1012(50 + Cd )x 10“12
Fig. 24.3 Plot of 1 / f2 v/s C for Example 24 2.
=-------- ------ •••(”) From the graph, additional capacitance,
64(50+ Crf)
From (z) and (ii), 16(250 + Cd) =64(50 + Cd) q=20 pF.
Capacitance of voltmeter and leads
or self-capacitance Crf = 16.67 pF.
= 4.5+ 1.5 =6.0 pF.
From (z), .’. Self-capacitance of coil
inductance, L =------------------- H = 234 uH Cd =20-6 =14 pF.
16(250 + 16.67)
Slope of line
Example 24.2 A coil of unknown inductance and self- . Q ab 4xl0~12 4
capacitance is connected in series with a standard variable tan 0 =---- =------------- =__
O'b 75xl0“12 75
capacitor C. An electronic voltmeter is connected across C.
A variable frequency oscillator is loosely coupled to the coil From Eqn. 24.4, inductance of coil
and the circuit is adjusted to resonance for each different L=^=± 1 H = 1.35mH
oscillatorfrequency fby adjusting C, as in thefollowing table: 4z? 75 4n2

f, kHz 350 400 450 500 600 24.4 MEASUREMENT OF CAPACITANCE

C, pF 132 98 73 55 31 Substitution method is normally used for
measurement of capacitance at high frequencies.
Determine :
Figure 24.4 shows a simple circuit for measurement of
(a) the inductance, and capacitance. C is the unknown capacitance connected
(b) the self-capacitance of the coil. in parallel with a variable standard capacitance Cs.
 728 Electrical and Electronic Measurements and Instrumentation

The standard capacitance is of the same order as the

unknown capacitance.

Standard

Oscillator

Fig. 24.4 Measurement of capacitance by

substitution method.

With switch S open, resonance is obtained by

varying the standard capacitance Cs. The resonance
condition is indicated by an electronic voltmeter
connected across Cg.
Let Fig. 24.5 Measurement of effective resistance by
f = frequency at resonance ; Hz, resistance variation method.
L = inductance of coil coupled the output of
Let
oscillator; H,
E = induced emf in the resonant circuit,
Csl = setting of standard capacitor with
switch S open ; F. IQ = current in the resonant circuit,
R = effective resistance of resonant circuit
/. We have Csl = 1/4ti2/2L ...(24.6)
excluding resistance R^
Now the switch S is closed putting unknown R1 = additional resistance inserted in the
capacitance in parallel with Cg. Keeping the frequency resonant circuit,
f the same as earlier, resonance conditions are L = inductance of coil,
obtained by varying C .
C = capacitance,
Let Cs2 = setting of standard capacitor with V = reading of voltmeter connected across C,
switch closed.
and f0 = resonant frequency.
.-. We get C + Cs2 = 1 / 4k2/2L ...(24.7)
r E
At resonance L =--------- ...(24.9)
From Eqns. 24.6 and 24.7, value of unknown 0 R+R1 7
capacitance is :
I F
u = C
C Si -Cs2' .-. Voltage V = —2— --------------------
2tt/0C (R+R1)2n/0C
Thus the value of unknown capacitance is directly
...(24.10)
known from the difference of the two settings of the
standard capacitor. or R+ Rx = E/2nfQCV ...(24.11)
If the values of E, f0 and C are kept constant, we
24.5 MEASUREMENT OF EFFECTIVE get
RESISTANCE R+R^K/V ...(24.12)
There are two methods of measurement of
where K = a constant = E /2 nfQC ...(24.13)
effective resistance of a series RLC resonant circuit.
(i) Resistance variation method, and Thus, under the above conditions if a curve is
(n) Reactance variation method. plotted between 1/V and Rp we get a straight line as
shown in Fig. 24.5(1?).
24.6 RESISTANCE VARIATION METHOD The effective resistance is given by the intercept
The effective series resistance of a resonant circuit OO' provided that the loss resistance of capacitor and
is measured by adding extra resistance from outside. the resistance of the leads is negligible.
High Frequency Measurements 729

From the above analysis we conclude : The reactance is then adjusted so that the current
(z) We can determine the value of effective I=/0/V2
resistance of a resonant circuit by inserting an external /= 1_=A=_£_
variable resistance Rr The circuit is set into resonance Zr2 + x2
by varying the value of C. This value of capacitance C
is kept constant afterwards. We go on increasing the Hence, Vr2 + X2 = /2R or X=±R
value of Rr This does not disturb the resonance condi­
...(24.15)
tions and its effect is only to decrease the value of reso­
There are, therefore, two values of reactance X
nance current IQ and to decrease the value of voltage V
which give a current, I = /0 / >/2 .
across the capacitor C. Thus an increase in the value of
Rj decreases V and increases the value of 1/V. The difference of these two values or reactance
gives twice the value or effective resistance i.e., 2R.
(ii) The analysis is valid only if the value of
induced emf E in the resonant circuit is constant. Also As mentioned above there are two ways of
the resonant frequency fQ is constant. This requires varying the reactance :
that the coupling between the oscillator circuit and the (i) variation of capacitance, C, and
resonant circuit be very loose otherwise changes in the (ii) variation of frequency/
resonant circuit would affect the oscillator circuit. In
The two methods are discussed below :
order to ensure that the value of induced emf and the
resonant frequency remain constant for all the 1. Variation of capacitance. In this method, the
readings, the oscillator should possess a high degree value of frequency / is kept constant and the
of frequency and amplitude stability. capacitance is varied. Let the value of capacitance
(iii) We have explained above that the value of which gives the maximum current be CQ and the two
induced emf E in the resonant circuit should remain values of capacitance which give a current of / /2 be
constant. This can be checked by connecting a thermal Cj and C2.
ammeter in series with the output coil of the oscillator Therefore,
or by connecting an electronic voltmeter across the 2R = 1/wC2-1/coC1
output coil of the oscillator.
r _ C1 -C2 C -C
= H ^2
or ...(24.16)
24.7 REACTANCE VARIATION METHOD 2cdC1C2 4ti/C1C2
The effective resistance of a resonant circuit can be Now if the circuit loss is small and the resonance
measured by variation of reactance of the circuit. is peaked and symmetrical,
Figure 24.6 shows the circuit for such a measurement. C c » C2
*"1^2 0
...(24.17)
C__
r~ H C -C
-C2 ~ H ^2 ...(24.18)
" 2coC2 ~ 4n/C2

Also Q = coL/ R =1/ coC0R ...(24.19)

Fig. 24.6 Measurement of effective resistance by
reactance variation method.
In this circuit a thermal ammeter is used to
measure the value of current flowing. The circuit
reactance is varied by either varying the value of
capacitance C or the frequency f. The reactance is
adjusted till the ammeter reads maximum value. This
happens when the circuit is under resonance conditions.
The maximum value of current:
I0 = E/R ...(24.14)
where E = emf induced, and
R - effective value of resistance.
. 2Co
...(24.20)
C] -C2
The accuracy of this method depends upon the
accuracy with which the difference Q - C2 can be
known. For accurate determination of this difference a
micrometer capacitor is put in parallel with the main
capacitor. From Eqn. 24.20 it is clear that the circuit Q
can be determined in terms of two values of capacitance.
2. Variation of frequency. In this method the
capacitance C is kept constant and the frequency is
varied. Let a>0 be the angular frequency to give
maximum current IQ and od1 and co2 be the two
frequencies which give a value of current
I=!o/22.
 730 Electrical and Electronic Measurements and Instrumentation

2 R = (WjL-1/ w1C)-(oj2L-1/ co2C) same scale reading I for the tzoo values ofC - namely 37 and
= (cd1 -co2)L+1/ C(l/ <£>2 — 1 / CDj ) 45 pF, one being below and the other above the capacitance
n CQ needed to produce maximum current.
= (cOj - w2)L+ (DqL
The capacitor is now adjusted to a value and an
= (a>1 -co2)/(co1co2) as co0L-1 / C.
added series resistance of 75 Q is found to reduce the current
Now for a peaked and symmetrical resonance to previous value of I. Determine values ofL and R.
curve :
Solution. Let the two values of capacitance which
(Op = co1co2 ...(24.21) produce the same current be Cf and C2.
9 CO. - (0o
2 R «(co.1 -coz.2)L+ cOpL
U
—----- - ~2(co.i -co7z )L Cj = 45 pF and C2=37pF
Let the resonance curve be peaked and
or R «(w1 -w2)L ...(24.22)
symmetrical and therefore capacitance required for
(Oj -o>2 resonance is :
...(24.23)
WqC _ q + C2 _ 45+37
= 41pF.
°~ 2 2
and ...(24.24)
R co1 - co2 At resonance, inductance

The frequency variation method has a ___ ____ —________________________J_J

4k2/2C0 4k2 x (3x 106)2 x 41 x 10’12
disadvantage because it is quite difficult to keep the
value of induced emf E constant with change of = 69.2 pH.
frequency. But the error introduced on this account is
When the capacitance in the circuit is Cr the value
small and can be allowed for.
of current
Example 24.3 A coil of inductance L is tuned by a
capacitance C to resonate at 1 MHz. A standard capacitor of a/r2+(coL-1/coC1)2

230 pF is tuned to resonate with the same coil at 2 MHz. If

the resonance is indicated by a VTVM of capacitance 8 pF in
parallel with the circuit, calculate the values ofL and C.
2kx 3x 106 x 69.2 x 10-6
Solution. We have
r2 + 1
C+Cv = 1/4k2L(1x106)2 2kx3x106x45x1012J

or (C + 8)x IO-12 = 1/(4k2Lx 1012) ...(z)

and (Cs + CJxlO“12 = 1/4k2L(2 x 106)2 •••(0
£
or (230+ 8)x 10~12 = 1/(16k2Lx 1012) ...(zz) At resonance, I =-------- ...(ii)
R + 75 ' ’
where Cv = capacitance of voltmeter Equating (?) and (zz) we get, R = 70.8 Q.
= 8x10-12F
Example 2445 Calculate the Q factor and effective
and Cs = capacitance of standard capacitor
resistance of a circuit tuned to a frequency of 1.5 MHz and
= 230x10-12F
having an effective capacitance of 150 pF. In this circuit the
From relation (zz), we get, current falls to 70.7% of its resonant value when the
Inductance frequency of an emf of constant magnitude injected in series
with the circuit deviates from the resonant frequency by
1 5 kHz.
L= H = 26.6 pH
238 x16 k2
Solution. The analysis of this problem has been
From relation (zz), capacitance done in Art. 24.6.
From Eqn. 24.24,
C = —-------8=944pF.
4k2 x 26.6
Q factor =----- -— = —.
Example 24.4 A constant rms voltage of frequency fz
3 MHz is applied to a coil of inductance L and resistance R,
a calibrated variable capacitor C, and a thermojunction all Resonant frequency
in series. The uncalibrated ammeter is found to show the fQ = 1.5 MHz =1.5x 106 Hz
High Frequency Measurements 73 I

The two frequencies at which the current is 70.7% ZD in parallel is equal to zero. The other impedances
of or 1 / x/2 time the current at resonant frequency are : i.e., ZR, Zr r, Zrr do not affect the null conditions.
d C', Zt'
= (1.5 x 106 + 5 x 103) Hz Therefore at null point:
and /2 = (1.5x106-5x103)Hz
—+ —=0 or Z.+Zn=0
.-. =10xl03 Hz
Hence, Q = ^=150 or ZA = -ZD ...(24.25)
10xl03 But
From Eqn. 24.23, resistance 2^ = 21 + Z2 + Z] Z2 / Z3
(Qj, -o>2 _ f1~ f2 and ZD = Z4 + Z5 + Z4Z5 / Z6
w2C 2 n /02C Therefore, at null point,
Z] + Z2 + Z1Z2 / Z3 = -Z4 - Z5 - z4z5 / Z6
...(24.26)
A common parallel T network for measurement of
24.8 T NETWORKS frequency is shown in Fig. 24.8. Comparing Fig. 24.8
T networks are very useful for measurement of with circuit of Fig. 24.7(a).
inductance, capacitance, resistance and frequency in
the high frequency range. These networks employ null
techniques.
There are two types of T networks used :
(z) Parallel T networks, and
(zz) Bridge T networks.

24.9 PARALLEL T NETWORK

Figure 24.7(a) shows a parallel T network with
source null detector. In fact there are two Ts formed by Fig. 24.8 Parallel T network for measurement
impedances Z]z Z2, Z3, and Z4, Z5, Z6 in parallel. The of frequency.
network can be transformed into another form shown
in Fig. 24.7(b) by star delta transformation. A null is We have Z^ = Rv Z2 = Rr Z3 = -j / coC1,
indicated by the detector if the admittance of and 24 = -j I «C2, Z5 = -j I coC2 and Z6 = R2.
Applying Eqn. 24.26 for null conditions, we have,
R2 z i 1
R, + R, 4--------------- -------- 1------- 1—=—=—
1 -j/wCl wC2 mC2 CO2C2R2

or 2 R. + = -t- + ? \—
171 1 coC2 co2C|R2

Equating real and imaginary terms we get,

2R: =l/co2CfR2 or w2=l/2R1R2C2
...(24.27)
and R12coC1 = 2 / coC2 or co2 =2 / R2C1C2
...(24.28)
This circuit (Fig. 24.8) therefore can be used to
measure frequency. From Eqns. 24.27 and 24.28, we
have,
1/2R1R2C2=2/R2C1C2

This condition is fulfilled if

Fig. 24.7 Parallel T network. R}- 4 R^ and = C2.

 732 Electrical and Electronic Measurements and Instrumentation

24.10 BRIDGE T NETWORK The T networks have an advantage over ordinary

Bridge T network is a simplification of parallel T
network. If the impedance Z3 is infinite and the
combination of impedances Zp and Z2 is made into a
single impedance Zr it becomes a bridge 'T network.
Figure 24.9 shows a bridge T network.
Fig. 24.9 Bridge Network.
From Eqn. 24.26 for a bridge T network
^1 = ~^4 ~%5~ Z4Z51Z6
as Zo is infinite and Z. + Z- = Z,.
Figure 24.10 shows a bridge network for measure­
ment of effective resistance R, inductance L and Q of a
coil.
Coil
bridge circuits. It is clear that these networks have a
common point between source and detector. This
point may be earthed and thus the screening of whole
of the circuit becomes easy.
Example 24.6 Two variable capacitors
mechanically coupled so that their capacitances are always
equal, are connected in series between terminals A and B to
form the series arms of a bridge 'T' network. Their junction
is connected through a variable resistor R2 to the terminal
C. A coil of unknown inductance L and resistance R is
connected across terminals A and B. The input voltage is
applied between A and C, and a detector is connected
between B and C. Under what conditions will the detector
show an output voltage approaching zero ? Evaluate R and
L if this balance condition is obtained at a frequency of
5 MHz with Cj = 45 pF and R1 = 4.2 Q.
...(24.29) Solution. The network is similar to the one shown
in Fig. 24.10. The analysis of the network is carried out
in Art. 24.10.
Equations 24.30 and 24.31 give conditions for no
output voltage as :
(o2C2Rj

j 7 2
and L=—x—
<o2Cj
R 1
~ 4n2 x (5 x 106)2 x (45 x 10'12)2 x 4.5

= 0.115 MQ
Fig. 24.10 Bridge T network for measurement of ____________ 2___________
effective resistance, inductance and Q. ~ 4n2x(5xl06)2x45xl0’12
For this circuit
= 45 pH.
Zj = R + /wR Z4 = -j I coCr
Z5 — y / wCp z6 = Rp 24.11 Q METER
At null point we have The determination of the storage factor Q is one of
~ ~^4 ~ Z5 — Z4Z5 / Z6 (Eqn. 24.29) the most widely used means in the laboratory for
testing radio frequency coils (RFC), inductors and
R u. W - _7 .
capacitors. The storage factor is equal to Q = co0L/ R
coCT coCj co2C2Rj
where co0 is the resonant angular frequency, L is the
inductance and R is the effective resistance of a coil.
The effective resistance, R, is never determined
directly since its value depends upon the value of
Equating real imaginary terms, we get,
frequency. As is well known, the value of effective
R = 1/oj2C12R1 ...(24.30) resistance with a.c. differs from its d.c. value on
and L - 2 / (a2C1 account of skin effect and eddy current losses. Since
...(24.31)
the value of R varies in a complex manner with
...(24.32) frequency, it is indirectly determined by measuring
1 the value of Q.
High Frequency Measurements 733

The Q meter is an instrument which is designed to
measure the value of Q directly and as such is very
useful in measuring the characteristics of coils and
capacitors.
Principle of working. The principle of working of
this useful laboratory instrument is based upon the
well-known characteristics of a resonant series R, L, C
circuit. Figure 24.11(a) shows a coil of resistance R and
inductance L in series with a capacitor C.
Fig. 24.11 A resonant RLC circuit and its
phasor diagram.
At resonant frequency fQ, we have,
xc=xL
where capacitive reactance Xc = 1 l2itfQC
and inductive reactance X, =2 7r/oL=wot
resonant frequency f° 2itjLC
and current at resonance
The phasor diagram is shown in Fig- 24.11(1’).
Voltage across capacitor, Ec = ZQXC = = ^o0)o^
Input voltage E = IQR
■jC _ 0'0^ _ _Q
E I0R R and also due to the distributed capacitance of the
E0 = QE -(24.32)
Thus the input voltage E is magnified Q times.
If the input voltage E is kept constant, the voltage
appearing across the capacitor is Q times E and a
voltmeter connected across the capacitor can be
calibrated to read the value of Q directly.
Practical Circuit
A practical Q meter circuit is shown in Fig. 24.12.
It consists of a self-contained variable frequency RF
oscillator. This oscillator delivers current to a low
value shunt resistance Rsh. This low resistance is of the
order of a hundredths of an ohm ; a typical value may
be n 02 fi. Through this resistance a small value of
voltage E is injected into the resonant circuit. This
voltage is measured by a thermocouple voltmeter.
Since the value of shunt resistance is very low, it
introduces almost no resistance into the oscillatory
circuit and therefore represents a voltage source of
magnitude E, with a very small internal resistance. A
calibrated standard variable capacitor C is used for
resonating the circuit. An electronic voltmeter is
connected across this capacitor. The coil under test is
connected to terminals and T2.
Application
1. Measurement ofQ. The circuit for measurement
of Q is shown in Fig. 24.12. The oscillator is set to the
desired frequency and then the tuning capacitor is
adjusted for maximum value of EQ. As discussed
earlier under these conditions Q = EQ/ E and if the
voltage is kept constant, the voltmeter connected
across the capacitor may be calibrated to read the
value of Q directly. This measured value of Q is
commonly regarded as the Q of the coil under test.
However, there is an error. The measured value of Q
is the Q of whole the circuit and not of the coil. There
are errors caused on account of the shunt resistance
circuit.

Figure 24.12 Circuit of a Q meter.

 734 Electrical and Electronic Measurements and Instrumentation

Correction for shunt resistance. Measured value, Thus,

f =______1______
= _o0L 2 2^HC2 + Cd)
",eas R+R. sh Now 72=2/,
True value = iaQL/ R • 1 = 2 x 1

2n/UC2 + Cll) 2njL(C^Cj)

+ -(24.33)
Thus the measured value of Q is smaller than the
true value. Now if coils of high resistance (low Q
coils) are being measured, the difference between the
two values may be negligible. But when
measurements are done on low resistance (high Q)
coils, the error caused on this account may be serious.
Correction for distributed capacitance. The
distributed capacitance or self-capacitance of
measuring circuit modifies the true value of Q. The
measured value of Q is less than the true value by a
factor that depends upon the value of the distributed
(self) capacitance and the tuning capacitor. It can be
shown that
f CA
’'-true
Q =Q*~>neas 1^1
1 + -M ...(24.34)
' '
where Cd = distributed or self-capacitance, and
C = tuning capacitance.
A method for measurement of distributed (self)
capacitance is given later.
2. Measurement of inductance. The value of
inductance is given by L = l/4k2/02C.
The values /0 and C are known and therefore the
value of inductance may be calculated.
3. Measurement of Effective resistance. The value
of the effective resistance may be computed from the
relation R = ^0L/Q^.
4. Measurement of self-capacitance. The self­
capacitance is measured by making two measure­
ments at different frequencies. The capacitor is set to a
high value, and the circuit is resonated by adjustment
of the oscillator frequency. Resonance is indicated by
the circuit Q meter. Let the values of tuning capacitor
be Cj and that of frequency be f^ under these
conditions. Therefore,
f -______ 1
1 2^L(C1+CI()
The frequency is now increased to twice its initial
value and the circuit is resonated again this time with
the help of tuning capacitor. Let the value of tuning
capacitor be C2 and that of frequency be f2 under these
conditions.
*
C. -4C,
or distributed capacitance = 1 —- ...(24.35)
5. Measurement of bandwidth. The measurement
of bandwidth with a Q meter is similar to the one used
for measurement of effective resistance by variation of
frequency method given on page 730. The bandwidth
is calculated by using Eqn. 24.24.
6. Measurement of capacitance. For measurement
of capacitance a dummy coil is connected across
terminal an T2 (Fig. 24.12). The circuit is resonated
by varying the value of the tuning capacitor. Let this
value be Cr
The capacitor under test CT, is now connected
across terminals T3 and T4. This puts the test capacitor
in parallel with tuning capacitor. The circuit is
resonated again by varying the value of tuning
capacitor. Let this value be C9. Therefore value of
capacitance under test is CL = CT - C2.
Th is method is similar to the one given in Art. 24.4
on page 727.
Example 24.^ A circuit consisting of a coil, a resistance
an a variable capacitor connected in series is tuned to
resonance using a Q meter. If the frequency is 500 kHz, the
resistance 0.50. and the variable capacitor set to 350 pF.
Calculate the effective inductance and resistance of the coil,
if the Q meter indicates 90.
Solution. At resonance ;
T J T 1
Inductance L = —n „
=_____________1_____________ H
4k2 x (500 x 103)2 x 350 x 10"12
= 290 pH
Let R be the resistance of coil.
conL
.-. R + 0.5 = —2-
Q
2kx500x103x290x10“6
90
Effective resistance of coil R =10.25 -0.5 = 9.75 0
Example 24.8 A coil of resistance 10 O is connected in
the Q meter circuit as shown in Fig. 24.12. Resonance
occurs at a frequency of 1 MHz with the tuning capacitor
High Frequency Measurements 735

set at 65 pF. Calculate the percentage error introduced in Example 24.10 Tests using a Q meter, on a radio
the calculated value of Q if a resistance of 0.02 Q is used tuning coil to find its self-capacitance gave the following
across the oscillator circuit. results :
Solution. True value of Q is : 1. With a radio coil connected normally, the
resonance was obtained at 1 MHz with the
tuning capacitor set at 80 pF.
1 2. With a standard inductor connected in place of
(2nx 106)x (45 x 10-12)710 the radio coil, resonance was obtained at 3 MHz
and this condition was not altered when the radio
Measured value of coil was connected in parallel with the standard
inductor. Calculate the self-capacitance of the
'neas cdC(jR+0.02) radio coil.
Solution. Let
=____________ 1____________
(2tix106)x (45 xl(F12)x 10.02 Q = capacitance of tuning capacitor for 1st setting,
coj = angular frequency for 1st setting.
= 244.5
.'. %age error Inductance of coil

= 245----- 5 x 100 = 0.2% high

-244±2 L=1/<^(C1 + Cd)
245 6
In this second setting; the reason for the resonance
Example 24.9 The self-capacitance of a coil is measured conditions to remain unchanged is that a very high
by a Q meter. The circuit is set into resonance azt 2 MHz impedance has been placed in parallel, this being the
and the tuning capacitor is as value of 460 pF. The resonant circuit of the radio coil and its self
frequency is now adjusted to 4 MHz and resonance capacitance. In other words, the radio coil in parallel
conditions are obtained the tuning capacitor at 100 pF. with its self capacitance is resonant at 3 MHz and thus
Calculate the value of self-capacitance of the coil. it may be written
Solution. L=l/co|Crf ...(zz)
The problem has already been analysed in
From (z) and (n) we have, self capacitance of coil is,
previous page.
From Eqn. 24.35, self-capacitance of coil
C -4C d co2-co2
d 3
(l)1
23
400-4x100 _n „ = 80x = 10 pF
=---- 3------- = 20 pF (3)2-(l)2

Review Questions
1. Discuss the problems associated with 4. Discuss the following two methods of
measurement of resistance, inductance and measurement of effective resistance of a coil at
capacitance at high frequencies. high frequencies :
2. Describe the resonance method for measurement (i) Resistance variation method.
of inductance at high frequencies. Discuss the (zz) Reactance variation method.
assumptions made. How is the effect of additional
5. Describe the following methods of measuring
capacitances in the tuning capacitor taken care of ?
effective resistance of a coil using resonant
3. Describe the method for measurement of methods :
capacitance at high frequencies using resonance
(i) Variation of capacitance
method.
(zz) Variation of frequency.
 736 Electrical and Electronic Measurements and Instrumentation

6. Explain how frequency can be measured using a 9. Describe how corrections for shunt resistance and
parallel T network. Derive the expression for the distributed capacitance are applied when
frequency in terms of values of components used measuring Q factor of a coil with a Q meter.
in a parallel T network. 10. Describe the measurement of following using a
7. Describe how Q factor of a coil can be measured Q meter :
using a Bridge T network. Derive the expression (/) Q factor (ii) inductance
for Q. (iii) effective resistance (iv) Self capacitance
8. Describe the circuit and working of a Q meter. (v) Bandwidth (vi) capacitance
Describe its applications.

Unsolved Problems "V

1. A capacitor of capacitance 245 pF produces each frequency the capacitor is adjusted to produce
resonance at an angular frequency of 5 x 106 rad/s, a maximum reading on the voltmeter.
while a capacitor of capacitance 50 pF produces The readings are as follows :
resonance with the second harmonic of this
frequency. Calculate : Frequency Capacitance Maximum
kHz PF voltage, V
(a) the inductance, and
200 345 4
(b) the self capacitance of the coil.
300 145 -
[Ans. 154 pH, 15 pF]
400 75 -
2. With a suitable standard inductor connected to a
500 43 -
Q-meter, resonance is obtained with an angular
frequency co with capacitor setting C}; the indicated 600 25 16
Q factor is Qr The unknown impedance is
Determine :
connected in series with the standard inductor and
resonance is re-established by resetting the (a) the inductance of the test coil,
capacitor to C2, the corresponding Q factor being (b) its self capacitance, and
Q2- Calculate : (c) the percentage increase in its resistance as the
(a) the resistive, and frequency varies from 100 kHz to 600 kHz.
[Ans. 1.732 mH, 12 pF, 125%]
(b) reactive components of the unknown imped­
ance when 5. The effective resistance of a coil at a frequency of
q = 208 pF, Q = 170, 0.2 MHz is measured by the reactance variation
method. The coil is connected in series with a
q = 185.5 pF and Q2 = 52.
calibrated variable capacitor and a 10 Q resistor. A
The frequency is 159 kHz. [Ans. 35 Q, 584 Q] constant emf at 0.2 MHz is injected into the circuit.
3. A coil of unknown impedance in series with a An electronic voltmeter connected across the
capacitor of 224 pF and an ammeter of negligible resistor is used to indicate resonance. The capacitor
impedance was connected to a variable frequency settings to reduce the resonant current by one half
of constant voltage and negligible impedance. The are 150 pF and 170 pF. Calculate the effective
frequency was adjusted both above and below resistance of the coil. [Ans. 170 Q]
resonant frequency till the reading of ammeter was
6. A coil of unknown impedance in series with a capa­
reduced to 70.7% of its value at resonance. This
citor of capacitance 0.02 pF and an ammeter of negli­
occurred at frequencies of 876 and 892 kHz.
gible impedance was connected to a variable
Determine effective resistance, inductance and Q
frequency source having constant voltage. The
of the coil. [Ans. 14.5 Q, 0.145 mH, 55.3]
frequency was adjusted both above and below
4. A test coil is connected across a calibrated loss free resonance until the reading of the ammeter was
variable capacitor having in parallel with it a reduced to 0.8 of the value at resonance. This
VIVM of input capacitance 3 pF. A constant occurred at frequencies of 0.97 and 1.01 MHz. Find
current, of varying frequency is passed through an the inductance and resistance of the coil.
injector coil loosely coupled to the test coil, and at [Ans. 1.29 pH; 0.424Q]
High Frequency Measurements 737

7. A test or a radio tuning coil, using Q meter, gave
the following results :
(a) With a radio coil connected normally, reso­
nance was obtained at a frequency of 1 MHz
9. A coil was tested using a Q meter and following
with the Q meter capacitor set at 480 pF.
(b) With standard inductor connected in place of
the radio coil, the resonance was obtained at
5 MHz and this condition was not altered when
the coil was connected in parallel with the stan­
dard inductor. Calculate the self capacitance of
10. A coil with a resistance of 0.1 Q is connected in a Q
radio coil. [Ans. 20 pF]
8. A Q meter gave the following readings during a
certain test on a coil :
Q = 250, C = 150 pF, f = 750 kHz.
A VTVM was then connected across the standard
capacitor and resonance re-established. The
corresponding readings now become
Q = 150, C= 150 pF, / = 750 kHz.
Calculate for the VTVM :
(a) the input resistance, and
(b) the input capacitance. [Ans. 496 kQ, 10 pF]
information was obtained. At a frequency of 3 MHz,
the value of capacitance to give maximum voltage
across the capacitor was 251 pF while at 6 MHz
frequency it was 50 pF. Calculate the self
capacitance of the coil. [Ans. 17 pF]
meter circuit as shown in Fig. 24.12. Resonance
occurs when the oscillator frequency is 40 MHz
and the tuning capacitor is set at 135 pF. Calculate
the percentage error introduced in the calculated
value of Q by a 0.02 Q shunt resistance connected
across the oscillator circuit. Comment upon me
result. [Ans. 17%]

Objective Type Questions

Tick (^) most appropriate answer : 4. A series RLC circuit is used for measurement of
inductance of a coil using resonant method.
1. The effective reactance of an inductive coil, Resonance conditions are produced by varying the
(a) increase because of stray capacitances as the capacitance. A thermoelectric type ammeter is
frequency increases used in series with the circuit. Also an electronic
(b) decrease because of stray capacitances as the voltmeter is connected across the tuning capacitor
frequency increases (a) The conditions of resonance are more reliably
(c) remains the same irrespective of the increase in indicated when the ammeter reads maximum
frequency even if stray capacitances are present (b) The conditions of resonance are more reliably
(rf) none of the above. indicated when the voltmeter connected across
the capacitor reads maximum
2. The effective reactance of an inductive coil
(c) The conditions of resonance are the same when
(a) increases due to stray series inductances as
both ammeter and voltmeter read maximum
frequency increases
(rf) The conditions of resonance are indicated when
(b) decreases due to stray series inductances as
both ammeter and voltmeter read minimum.
frequency increases
5. If the Q factor of a coil which is w0L/ R is measured
(c) remains the same irrespective of the increase in
by varying the frequency
frequency even if stray series inductances, are
present (a) the plot between Q and frequency is linear

(rf) none of the above. (b) the value of Q initially decreases with increase
of frequency and afterwards increases with
3. The effective resistance of a coil at high frequencies increase of frequency
is more than its d.c. resistance on account of
(c) the value of Q initially increases with increase
(a) skin effects I
of frequency and afterwards decreases with
(b) proximity effects increase of frequency
(c) eddy current losses 1 (rf) the Q factor remains constant irrespective of
(rf) all the above. the value of frequency.
 738 Electrical and Electronic Measurements and Instrumentation

6. The Q factor of a coil at the resonant frequency respectively. The value of distributed capacitance
1.5 MHz of an RLC series circuit is 150. The is :
bandwidth is :
(«) (fc)
(rt) 225 MHz (fc) 1.06 MHz
(c) 10 kHz (j) none of the above.
7. The resonant frequency of an RLC series circuit is
1.5 MHz with the tuning capacitor set at 150 pF. 9. The true value of Q of a coil is 245 and the
The bandwidth is 10 kHz. The effective resistance measured value is 244.5. The ratio of distributed
of the circuit is : capacitance to self capacitance of the coil is
(rt) 2.04x1 O’3 (b) 1.002
(a) 29.5 Q (b) 14.75 Q
(c) 9.4 Q (d) 4.7 Q (c) 0.997 (d) none of the above.
10. In a Q meter, the value of shunt resistance
8. In a Q meter, distributed capacitance of coil is
connected across the oscillator is typically of the
measured by changing the capacitance of the
order of
tuning capacitor. The values of tuning capacitor
(rt) Q (b) mQ
are Cj and C, for resonant frequencies f} and 2/t
(c) pQ (d) kQ.

1. (b) 2. (rt) 3. (d) 4. (b) 5. (c) 6. (c) 7. (d) 8. (c) 9. («) io. (b)
 Primary Sensing Elements
and Transducers
I
25.1 INTRODUCTION
The measurand in an instrumentation system
makes its first contact with a Primary Detection Element
or an Input Device. There is a multiplicity and variety
of measurands to be measured. These include process
variables like temperature, pressure and flow rate
which are widely employed in process and production
plants. The measurands also include electrical
quantities like current, voltage, resistance, inductance,
capacitance, frequency, phase angle, power and
magnetic quantities like flux, flux density, reluctance
etc.
All these quantities require a primary detection
element and/or a transducer to be converted into
another analogous format which is acceptable by the
later stages of the measurement system.
The measurand or the input signal is called an
information for the measurement system. The
information may be in the form of a physical
phenomenon or it may be an electrical signal. The
process of detection and conversion of the information
into an acceptable form requires energy. This energy
may be extracted from the measurand, but in that case
it will not be represented in its faithful form as it
would be subjected to loading errors. In order that a
measurand is represented in its faithful form
undistorted, no energy should be extracted from it
during the process of conversion i.e., it should not be
subjected to any kind of loading effects. In fact, efforts
should be made to supply energy required for
conversion from outside sources so that the
measurand is not distorted during the process of
conversion in order that it be faithfully reproduced in
its analogous form. The ideal conversion is where
(739)
absolutely no energy is extracted from the measurand
during the process of conversion and all the energy
that is required for conversion is supplied from
outside, so that the measurand is not distorted and the
analogous output of the detector is a faithful represen­
tation of the measurand.
25.2 MECHANICAL DEVICES AS
PRIMARY DETECTORS
In order to extract information from mechanical
systems, only mechanical displacement or velocity can
be used, and therefore the importance of mechanical
sensing elements is obvious. Some of the commonly
used mechanical sensing elements are springs - which
convert a force or a torque into a displacement ; a
diaphragm, a capsule, bellows or Bourdon tube -
which convert pressure into a displacement, a
bimetallic strip converts temperature into a
displacement ; a mass damper system is used for
measurement of acceleration, velocity and displace­
ment. Some input devices may involve more than one
mechanical conversion for example, fluid flow
measurements may involve conversion of fluid rate
into pressure differential using an orifice, venturi tube
or pitot tube and then in turn this pressure is
converted into displacement for purposes of
measurement.
There are a number of mechanical quantities
which are to be measured. Some of these quantities are
listed in Table 25.1 along with their modes of
operation for the purposes of measurement.
These mechanical quantities include force,
pressure, displacement, flow rate, temperature and
the list is perhaps unending.
 740 Electrical and Electronic Measurements and Instrumentation

Table 25.1 Mechanical quantities and their Modes of Operation

Type Operation
1 A.
Contacting spindle, pin or finger Displacement to displacement

1 B. Elastic Member
1. Prooving ring Force to displacement.
2. Bourdon tube Pressure to displacement.
3. Bellows Pressure to displacement.
4. Diaphragm Pressure to displacement.
5. Spring Force to displacement.

C Mass

i 1. Seismic mass Forcing function to displacement.

i 2. Pendulum scale Force to displacement.
i
3. Manometer Pressure to displacement.
1
s D. Thermal
1. Thermocouple Temperature to electric current.
t
• 2. Bimaterial Temperature to displacement.
3. Temp-strk Temperature to phase.

E. Hydropneumatic

i
1. Static
(a) Float Fluid level to displacement.
s (b) Hydrometer Specific gravity to displacement. 1
1
2. Dynamic I
(zz) Orifice Velocity to pressure. j
j (b) Venturi Velocity to pressure.
(c) Pitot tube Velocity to pressure.
(d) Vernes Velocity to force. i
?
(e) Turbines Linear to angular velocity.
[------- ____ _ __ 1
The initial concept of converting an applied force into a displacement is basic to many types of primary
sensing elements. The mechanical elements which are used to convert the applied force into displacement are
usually elastic members. There are many types of these elastic members. They can be classified into three
categories, although some elastic members may fall into a combination of these categories.
The three categories are :
(z) Direct tension or compression type :
(zz) Bending type ;
(z'z'z) Torsion type.

25.3 MECHANICAL SPRING DEVICES

Most mechanical-input measuring systems employ mechanical springs of one form or another. The
displacements are usually small and engineering approximations for small displacements or deflections are
valid. Various common types of springs are shown in Fig. 25.1. These range from cantilever, helical and spiral
springs to torsion bars, prooving (proof) rings and spring flexture pivots.
Primary Sensing Elements and Transducers 74 1

Fig. 25.1 Spring elements used for sensing force (f) or torque (7).

25.3.1 Cantilever 25.3.2 Helical Spring

A cantilever is shown in Fig. 25.2 which is Figure 25.3 shows a close coiled helical spring
subjected to a force at its free end. subjected to a compressive force F.

Fig. 25.2 Cantilever.

Deflection at the free end

x=Fl3/3EI ...(25.1)
where
Fig. 25.3 Closed coiled helical spring.
F - applied force ; N,
Displacement of spring :
I = length of cantilever, m ;
2 8 FD3n
£ = modulus of elasticity ; N / m x =---- —_ ...(25.3)
/ 2 A 3 Grf4
I = moment of inertia = —
112 J where F - applied force ; N,
b = width of cantilever ; m D= mean diameter of coiled spring ; m,
t = thickness of cantilever ; m d- diameter of spring wire ; m,
Stiffness of cantilever n= number of wires,
G= shear modulus ; N/m2.
K=F/x=3EI/l3 .-(25.2)
 742 Electrical and Electronic Measurements and Instrumentation

Stiffness of spring K = F/x Sometimes, notched bars, are used. The notched
bar has the advantage that it has a greater sensitivity
*/8
= Gd D 3n ...(25.4)
on account of its reduced diameter (of the notched
Maximum shear stress, portion).
T = 8FD/(nd3) ...(25.5)
25.3.5 Prooving (Proof) Rings
25.3.3 Spiral Spring They are used for measurement of force, weight
Figure 25.4 shows a flat spiral spring subjected a or load. The applied force causes a deflection which is
torque T. measured with the help of electrical transducers.
Prooving rings are made up of steel and are used
as force standards. They are particularly useful for
calibration of material testing machines in situations
where dead weight standards are impracticable to use

Fig- 25.4 Flat spiral spring.

The deflection of the spring is :
0= Ebt3T/12/, rad ...(25.6)
where E = modulus of elasticity, N/m2,
b - width of spring ; m, Fig. 25.5 Prooving ring.
t = thickness of spring ; m,
I = length of spring ; m, on account of their bulk. A prooving ring is a circular
T = torque ; Nm ring of rectangular cross-section as shown in Fig. 25.5
.'. Stiffness of spring which may be subjected to either tensile or compre­
K = T/e = Ebt3/12 I Nm/rad ...(25.7) ssive forces across its diameter.
Maximum fibre stress The deflection is given by :
Smax=6T/M2N/m2 ...(25.8)
x = (7t/2 ~4-^— F ... (25.10)
Spiral springs are used for production of 16 EI
controlling torque in analog instruments.
where d = outside ring diameter ; m.
25.3.4 Torsion Bars or Shafts The other terms have been explained earlier.
These are primary sensing elements for torque.
The common practice for measurement of
They are made use of in torque meters. The deflection
displacement is to attach a displacement transducer
or twist of the bar is proportional to the applied torque
between the top and bottom of the prooving ring.
and the deformation is used as a measure of the torque.
When the force is applied, the relative displacement
Some torque meters are designed so that the
can be measured. An LVDT is normally used for
angular displacement due to twisting of the bar is
measurement of deflection which is of the order of
measured with the help of a displacement trans­
1 mm or so. Another method is to use strain gauges for
ducers. In others, the strain in the surface of the bar,
measurement of strain caused by the applied force.
which is proportional to the torque, is measured with
The strain, then can be used to compute the applied
the help of strain gauges. The shear strain is a measure
of the torque. force.
Angle of twist 0 = 16T/7tGd3 rad, ...(25.9) 25.3.6 Load Cells
where T - applied torque ; Nm, Load cells utilise an elastic member as the primary
2
G = shear modulus ; N/m , transducer and strain gauges as secondary trans­
d = diameter of bar ; m ducers as shown in Fig. 25.6.
Primary Sensing Elements and Transducers 743
Fig. 25.6 Load cells.
25.3.7 Spring Flexture Pivots
Figure 25.1 shows two different types of arrange­
ments for flexture pivots, one for the single spring
flexture pivot and the second for crossed spring flexture
pivot. The crossed-spring flexture pivot is widely used
in measurement work for the following reasons :
(i) it is practically frictionless
(zz) the pivot sensitivity i.e., the angular deflec­
tion per unit applied torque is virtually
constant for angular relations less than 15°.
25.4 PRESSURE SENSITIVE PRIMARY DEVICES
Most pressure measuring devices use elastic
members for sensing pressure at the primary stage.
Fig. 25.7 Pressure sensitive primary devices.
These elastic members are of many types and convert
the pressure into mechanical displacement which is
later converted into an electrical form using a
secondary transducer. These devices are many a time
known as force summing devices. Figure 25.7 shows
some of the commonly pressure sensitive primary devices.
The principle of working of these devices is
explained as : the fluid whose pressure is to be
measured is made to press the pressure sensitive
element and since the element is an elastic member, it
deflects causing a mechanical displacement. 'The
displacement is proportional to the pressure applied.
The displacement is then measured with the help of
electrical transducers. The output of the electrical
transducers is proportional to the displacement and
hence to the applied input pressure.
The commonly used pressure sensitive devices
are described below :
25.4.1 Bourdon Tubes
The Bourdon tubes are designed in various forms.
These are :
(z) C type (zz) Spiral
(zzz) Twisted tube and (zv) Helical.
The Bourdon tubes are made out of an elliptically
flattened tube bent in such a way as to produce the
shapes shown in Fig. 25.7. One end of the tube is
sealed or closed. The other end is open for the fluid to
enter. When the fluid whose pressure is to be
measured enters the tube, the tube tends to straighten
 744 Electrical and Electronic Measurements and Instrumentation
out on account of the pressure applied. This causes a
movement of the free end (closed end) and the
displacement of this end is amplified through
mechanical linkages. The amplified displacement of
the free end may be used to move a pointer on a scale
calibrated in terms of pressure or may be applied to a
electrical displacement transducer whose output may
be calibrated in terms of the pressure applied.
Bourdon tube elements have several distinct advan­
tages and these include low cost, simple construction,
high pressure range, good accuracy except at low
pressures, and improved designs at the pressure for
maximum safety. Their greatest advantage is that they
are easily adapted for designs for obtaining electrical
outputs.
The disadvantages of Bourdon tubes are : their
low spring gradient which their use limits for pre­
cision measurements upto a pressure of 3 MN/m2, are
susceptible to shocks and vibrations and are subject to
hysteresis.
The Bourdon tube has the following forms :
1. C Type
The C type of Bourdon element is most commonly
used for local indication but it is also used for pressure
transmission and control applications. The tube which
is oval in section is formed into an arc of 250° and
hence the name C for the configuration which is
shown in Fig. 25.8.
The displacement of tip is,
3
x
2
where E is the modulus of elasticity ; N/m , P, the
applied pressure, N/m and other terms are as
depicted in Fig. 25.9.
Fig. 25.9 Geometry of C type Bourdon tube.
Thus relationship between the displacement of tip
and the applied pressure is non-linear i.e., each
pressure increment does not produce a corresponding
tip movement. However, a linear relationship between
the deflection of the pointer and the applied pressure
is obtained using the geared sector and pinion
movement or other mechanical means. The accuracy
of C type Bourdon tubes varies from ±0.5 to ± 2% or
poorer. However, the normal accuracy is about ± 1%.
2. Spiral Type
It is clear from Eqn. 25.11, that the displacement of
tip varies inversely as the wall thickness and depends
upon the cross-sectional form of the tube. It also varies
directly as the length of arc 'a' which in turn depends
upon the angle subtended by the arc through which it
is bent. Thus in a tube having an arc of 180°, the
displacement of the tube will be twice that of a similar
tube having an arc of 90°. Therefore, the displacement
of the free end (tip) may be increased by increasing the
length of the arc of the tube without changing the wall
thickness.
When the angle through which the arc is bent
reaches 360°, its length can be increased further in two
ways : the tube can be made in the form of a spiral, or it
can be made in the form of a helix. An increased
Primary Sensing Elements and Transducers 745
displacement of the free end can be obtained by
increasing the number of turns in the spiral or helix
thereby avoiding the need for further magnification. It
has been mentioned earlier that magnification is
obtained through the use of geared sector and pinion
is used in C type Bourdon tube. But while using spiral
or helix type Bourdon tubes there is no need for
further amplification and hence the geared sector and
pinion arrangement is not used in their case. The
absence of geared sector and pinion arrangement
eliminates the backlash which tends to occur when
they (geared sector and pinion) become worn owing
to continued use.
A spiral type Bpurdon tube is shown in Fig. 25.10.
Spiral tubes are made by winding several turns of the
tube with its flattened cross-section in the form of a
spiral.
Fig. 25.10 Spiral type Bourdon tube.
When the pressure to be measured is applied to
the spiral, it tends to uncoil producing a relatively
long movement of the tip whose displacement can be
used for indication or transmission.
The accuracy of spiral tube elements is higher
than that of C type elements on account of absence of
magnifying elements and is typically about ± 0.5%.
3. Helical Type
A helical type Bourdon tube is shown in Fig. 25.11.
As explained earlier, a helical bourdon element is
similar to the spiral element, except it is wound in the
form of a helix. The displacement of the tip of a helical
element is larger than that of spiral element. Usually a
central shaft is installed within a helical element and
the pointer is driven from this shaft by connecting
links. This system transmits only the circular motion
of the tip to the pointer which is directly proportional
to the changes in pressure.
Fig. 25.11 Helical type Bourdon tube.
The advantages of helix elements include the high
over-range capabilities which may be in the ratio as
high as 10 :1, its stability in fluctuating pressure appli­
cations, and its adaptability for high pressure service.
The number of coils employed in helix elements
depends upon the pressure to be measured. Helix type
of pressure elements use as few as three coils while
elements used for measurement for high pressures
may have as many as 16 coils or even more. The
accuracies obtainable from helical elements may vary
from ± 0.5% to ± 1% of span.
Bourdon tubes are made of different materials
which include brass, alloy steel, stainless steel, bronze,
phosphor bronze, beryllium copper, K Monel, Monel
and Ni-span C. The choice of type of material to be
used depends upon its elastic characteristics suitable
for the pressure range, process medium, temperature
and corrosive resistance to the media to be used.
Phosphor bronze is used in low pressure applica­
tions where the atmosphere is non-corrosive while in
applications where corrosion and/or high pressure is a
problem, stainless steel or Monel are used.
Pressure gauges using bourdon tube elements are
made with ranges from 760 mm of mercury to
700 MPa or higher for special applications with the
minimum span being about 70 kPa. In fact, there is
quite an overlap of ranges at which the various
bourdon tube elements are designed to operate and
these ranges differ from one manufacturers, to
another. One manufacturer, for example, lists spiral
bourdon tube elements from 0-75 to 0-1500 kPa
when bronze, beryllium copper and Ni-span C are
used as the materials for the element. The same
manufacturer uses the aforesaid materials for pressure
 746 Electrical and Electronic Measurements and Instrumentation

ranges 1.5-550 MPa while using helical elements. For The displacement of bellows element is given by,
bronze, the upper span limit is 0-30 MPa ; for
, 0.453 Pbn D271-v2
beryllium copper and Ni-span C, 0-40 MPa and for «= ---------------5---------- ...(25.12)
stainless steel 0 -600 MPa. £t3
Another manufacturer uses C type bourdon tube
where P = pressure, N/m2 ;
elements using bronze upto a pressure of 7 MPa and
b = radius of each corrugation, m ;
steel and stainless steel to 150 MPa and still another
which uses these elements upto a pressure of 700 MPa. n = number of semi-circular corrugations ;
It may therefore be concluded that C-type bourdon t = thickness of wall, m ;
tube elements may be used for measurement of D = mean diameter, m ;
pressures ranging from vacuum to about 700 MPa and E = modulus of elasticity, N/m2 ;
even higher, spiral elements from vacuum to 30 MPa and v = Poisson's ratio.
and helical elements from 1.5 MPa to 550 MPa.
Normally a Bellows has the ability to move over a
25.4.2 Bellows greater distance than required in a pressure application
A metallic bellows is a series of circular parts, and, therefore, to give it maximum life and to have
resembling the folds in an accordian as shown in better accuracy, its movement is generally opposed by
Fig. 25.12. These parts are formed or joined in such a a calibrated spring so that only a part of the maximum
manner that they are expanded or contracted axially stroke is used. This system is shown in Fig. 25.13 and
by changes in pressure. The metals used in the is called spring loaded bellows.
construction of bellows, must be thin enough to be
flexible, ductile enough for reasonably easy
fabrication, and have a high resistance to fatigue
failure. Materials commonly used are brass, bronze,
beryllium copper, alloys of nickel and copper, steel
and Monel. Most of the bellows used in pressure
gauges are seamless and are made from drawn tubing
by hydraulic or other methods of rapid forming. These
methods produce uniform walls that give a higher life
expectancy. Other methods such as soldering and
welding of annular sections, rolling, spinning and
turning from solid stock may also be used for
manufacture of bellows.

Deflection of bellows when opposed by a spring is,

d=P—...(25.13)
2
where = effective area of bellows, m ;
K,,
b'sK = stiffness constants of bellows
2 and
spring respectively, N/m ,

.. p
Pressure Pp - d/— P +K^
K-------- ...(25.14)
4
If the bellows assembly operates an electric switch
or some other mechanism, we have,
p = F + ds(Kb + Ks)
...(25.15)
4
where F = force required to operate switch or
mechanism, N ;
d = deflection required to operate switch
or mechanism, m
Fig. 25.12 Bollows element.
Primary Sensing Elements and Transducers 747
There are three main configurations in which
bellows elements are used and these are for
measurement of absolute, gauge and differential
pressures. These configurations may be illustrated
with the help of Fig. 25.14.
Fig. 25.14 Measurement of pressure with bellows.
There are two bellows, A and B. The pressure
applied to bellows A is P} and that to bellows B is P2.
In case it is desired to measure the absolute pressure,
bellows B is evacuated and the resultant pressure, Pj
is the absolute pressure. When measurement of gauge
pressure is desired, bellows B is opened up to
atmosphere with pressure P2 is equal to the
atmospheric pressure and therefore the reading of the
gauge is the gauge pressure. The measurement of
differential pressure P = P1 - P2 is done when pressure
P] is applied to bellows A and pressure P2 is applied to
bellows B.
The advantages of bellows include their simple
and rugged construction, moderate price, their
usefulness for measurement of low, medium and
high pressures, and their applicability for use in
measurement of absolute, gauge and differential
pressures. Bellows elements, like many other
elements have been greatly improved over the past
few years. The reduction of drift and hysteresis
allows their use in functions requiring ± 0.5% of full
span accuracy. They deliver relatively high forces
and are well adapted to vacuum and low pressure
measurements.
The disadvantages of bellows are that are not
suited for dynamic measurements on account of their
greater mass and longer relative movement. Also they
need temperature compensating device to avoid
errors resulting from changes in ambient temperature.
25.4.3 Diaphragms
The operating principle of diaphragm elements is
similar to that of the bellows. The pressure to be
measured is applied to the diaphragm, causing it to
deflect, the deflection being proportional to the
applied pressure. The movement of the diaphragm
depends on its thickness and diameter. The movement
is small and hence a diaphragm element does not
require any springs as is the case in bellows.
The movement of a diaphragm is a convenient
way of sensing a pressure. The unknown pressure is
applied to one side of a diaphragm. The edge of the
diaphragm is rigidly fixed and causes a deflection on
account of the applied pressure. The displacement of
the diaphragm may be measured to determine the
value of applied pressure, P.
The diaphragm, which is a very thin membrane
under pressure and therefore under radial tension,
may use capacitive or inductive transducers, to
produce an electrical output which is proportional to
the output of the transducers. The use of diaphragms
using thin membranes is limited to measurement of
low level pressure fluctuations. Since membranes can
withstand only limited values of force (produced by
pressure), diaphragms using thin membranes are
suitable for resistance and piezoelectric transducers
which require greater displacements than are
produced by membrane type of diaphragms.
Therefore, in place of membranes, it is usual to employ
thin circular plates which are either clamped, around
their circumference between two solid rings, or are
machined from a solid piece of metal. In order to
maintain linear relationship between deflection of
plates and applied pressure, the deflection of the
centre of plate should be less than half its thickness or
din < 0.51 which are essentially flexible discs.
The diaphragms, are of two types :
(zj Flat type and
(zz) Corrugated type.
These are shown in Fig. 25.15.
Corrugated diaphragms, on account of their
increased surface area, give greater deflections,
typically 2 per cent of the diaphragm diameter.
 748 Electrical and Electronic Measurements and Instrumentation

Lowest natural frequency (for air or gas medium)

% = 201 ,----- ^-5-

(b) Corrugated type
Fig. 25.15 Single diaphragm elements.
A flat diaphragm is shown in Fig- 25.16.
Fig. 25.16 Flat diaphragm.
256 Et3 d , indicators.
Pressure P =-------- -—N/m2 ...(25.16)
3(l-v2)D4
where E = Young's modulus ; N/m2,
t - thickness of diaphragm ; m,
D = diameter of diaphragm ; m,
R = radius of diaphragm ; m,
v = Poisson's ratio,
and dm = deflection at the centre of the
diaphragm ; m
The above relationship between pressure, P and
the deflection at the centre, dm is linear. But linearity
holds good as long as dm < 0.5 t and not otherwise.
The deflection at the centre is :
E rad/s .. .(25.20)
" D2y3p(l-v2)
a
where p = density of diaphragm material, kg/m
Design equations for the flat diaphragm may use
the following data :
Young's modulus for steel
E = 206 G N/m2 ; v = 0.28
and p = 7800 kg/m3
The above relationships are valid for uniform
pressure loading conditions over the entire surface of
the disc and also when the deflection smaller than half
the thickness of disc. Therefore, the response is linear
only when d < 0.5 t. In order to have a linear response
over greater deflections than d > 0.5 t, the diaphragms
are corrugated. A corrugated diaphragm is most
suitable for applications where a mechanical device is
used for measurement of deflection. This is because it
(corrugated diaphragm) gives a relatively greater
output and if combined with a high magnification
linkage, can be used for direct operation of mechanical
In some cases, a diaphragm element may consist
of a single disc ; which in others, two diaphragms are
bonded together at their circumference by soldering or
pressure welding to form a capsule. A diaphragm
element may consist of one capsule or two or more
capsules connected together with each capsule
deflecting on the application of pressure. The total
deflection is the sum of the deflections of individual
capsules. Figure 25.17 shows as diaphragm element
consisting of three capsules. In this assembly, the
individual capsule is connected axially with the next
one and is allowed to expand without any restraints.
The element may be provided with stops for
over-range and under-range protection.

3P(l-v2) 4
= —-—~R ...(25.17)
16 Et3

The deflection at a radius r from the centre is :

J 3P(l-v2)/ri2
df= l6Et3 ( -(25.18)

Maximum stress at the circumference

. 3 D2p x, , 2
S™=^2-N/m ...(25.19)
Fig. 25.17 Diaphragm element using three Capsules.
Primary Sensing Elements and Transducers 749
Two different arrangements are used for capsular
elements, the convex and the nested type as shown in
Fig- 25.18.
Fig. 25.18 Types of capsules.
Materials used for diaphragms include phosphor
bronze, stainless steel, beryllium copper, Ni-span C,
Inconel, Monel and nickel. Non-metallic materials are
used for some applications. Buna N rubber, nylon and
Teflon are used in environments that corrode metals.
The diaphragm elements find extensive use in
applications where measurement of low pressures
including vacuum is involved. The ranges are 0-50
N/m2 to 0 -200 kN/m2. Accuracies range from ± 0.5%
to ± 1.25% of full span.
25.5 FLOW RATE SENSING ELEMENTS
25.5.1 Introduction
The measurement of flow is very important in a
large number of industries. Flow measurements
account for a high percentage of the process variables
measured in the chemical industries. Flow is the most
important variable in plant operation. The plant material
balances, quality control and continuous processes
will be impossible without flow measurements.
Flow of material in a process or system can be
measured by a variety of methods depending upon
the material and its condition, the type of flow, the
volume and the mass of the quantity flowing, the
range and rangeability, the pressure and temperature
and the accuracy required. Many accurate and reliable
methods are available for the measurement of flow.
Some of these methods are applicable only to liquids,
some only to gases and vapours and some to both.
Fluids, may be of different types, like they may be
clear or opaque, clean or dirty, wet or dry, erosive and
corrosive. Fluid streams may be multiphase, vapour,
liquid or slurries. The flow may be turbulent or
laminar. Viscosities may vary. Pressures may vary from
near vacuum to many times atmospheric, and the
temperatures may vary from cryogenic to hundreds of
degrees of Celsius.
The characteristics and properties which need to
be considered are viscosity, density, compressibility,
temperature and pressure.
In most applications, the knowledge of type of
flow, whether laminar and turbulent is very important.
All liquids have a force of friction, called viscosity;
which tends to inhibit the formation of turbulent
eddies and vortices, and maintain streamline flow.
Viscosity enters into determination of flow constants,
particularly the discharge co-efficient C which is a
function of Reynold's number.
The Reynold's number is given by :
Re=udp/q ...(25.21)
where v = velocity of flow ; m/s,
d = diameter of pipe throat or orifice ; m,
p = density of fluid ; kg/m
and q = viscosity ; Ns/m .
The Reynold's number is dimentionless, and a
useful criterion by which to compare flows in
geometrically similar installations but with different
flow conditions. It provides information regarding
conditions where streamline flow ceases and turbu­
lent flow begins. It has been found by experiments
that the flow is streamline or laminar for Reynold's
number less than 2000 and turbulence sets in
increasingly as Reynold's number increases beyond
2000. Flow becomes completely turbulent for
Reynold's number greater than 105.
Flow measurements are concerned with the
following :
(z) velocity of flow ; m/s,
3
(zz) volumetric flow rate ; m /s,
(zzz) mass flow rate ; kg/s,
3
(zu) total volumetric flow ; m , over a set period
of time,
(u) total mass flow; kg, over a set period of time.
25.5.2 Types of Flow Rate Meters
The flow rate meters can be classified into two
categories :
(i) Inferential meters
(zz) Absolute or positive displacement meters.
In inferential meters, the flow is not directly
measured but is inferred from measurements of other
quantities which are related. Inferential methods of
 750 Electrical and Electronic Measurements and Instrumentation
measurement are based upon differentia! pressure or
head obtained from orifice plate, a venturi tube, a pitot
tube or a weir. Flow rate can also be inferred from
turbine and electromagnetic flow meters.
Fig. 25.19 Flow meters based upon producing
pressure drop using constrictions.
The purpose of this section is to briefly consider
the flow rate measuring devices with pressure related
techniques (inferential in nature) which are widely
used. The two basic techniques used are :
(?) a pressure drop is produced by means of a
constriction,
(ii) the kinetic energy of the flowing fluid is
transformed into potential energy in the
form of a static head.
Two different forms of producing constriction are
shown in Figs. 25.19(a) and 25.19(b). Figure 25.19(a)
shows a circular orifice plate while Fig. 25.19(b) shows
a venturi tube.
The fundamental equation for any flow rate sensing
element which changes the local velocity by using a
constriction or reduction in area of flow is given by :
Q=CEBXy1/2AP/pI ...(25.22)
where Q — flow rate, m/s,
C = co-efficient of discharge
actual flow rate
theoretical flow rate
E= / A*. )2 = vel°city approach factor
...(25.23)
B = expansion ratio,
= area at point Y ; m2,
Ax = area at point X ; m2,
px = density of fluid at point X, kg/m3,
AT = pressure drop between pts. X and Y; N/m"
For liquids, the fluid density can be considered to
remain constant and therefore expansion ratio B can
be taken as unity. With compressible fluids such as
gases and vapours, the density does not remain
constant and the value of B is less than unity and is a
function of ratios of pressure and areas at points X and
Y and the gas specific ratio.
The pressure drop AP can be measured with a
manometer and from there the input flow rate can be
inferred.
Orifice plates, venturi tubes and nozzles provide a
simple means of inducing a pressure drop. The flow
rate is proportional to the square root of the pressure
drop and hence there is a non-linear relationship
between output (pressure drop) and input (flow rate).
The discharge co-efficient for an orifice plate is
about 0.6 for Reynold's numbers above 10 and the
permanent pressure loss is 60-70 percent of diffe­
rential pressure AP. Therefore, an orifice plate which is
easy and cheap to install, may cause six or seven times
the pumping power as compared to a venturi tube
which has a discharge co-efficient of 0.96 for Reynold's
numbers above 10 and less tendency to trap suspended
material in the fluid being metered. However, venturi
tubes occupy greater space and are costlier.
Pitot tubes are one of the most common forms of
flow rate measuring devices which work on the
principle of converting kinetic energy of the fluid
being metered into a static head. Pitot tubes are
essentially laboratory type of instruments are not
normally used for industrial applications. Figure 25.20
Fig. 25.20 Pitot tube.
Primary Sensing Elements and Transducers 751
shows a pitot tube installed in a pipeline for
measurement of flow-rate. The 'tube' actually consists
of two concentric tubes. The open end of the inner
tube faces the incoming flow. The outer tube has a
closed end, but has a number of holes in its walls. Both
the tubes contain the same fluid as is flowing in
the main pipeline. The pressure in the outer tube is the
static pressure in the pipeline. The total pressure in
the inner tube, however, is greater than this static
pressure. The additional pressure on the inner tube is
on account of the impact of the fluid stream on the
stationary inner tube.
Application of Bernoullis' theorem shows that the
velocity at this point in the cross-section is :
v = J—- ----- ...(25.24)
K p )
2 (i) Weirs
where PQ = static pressure ; N/m ,
2 A weir is essentially a dam with a notched
P = pressure at the entrance of the tube ; N/m ,
3
and p = density of liquid ; kg/m
Thus, the velocity at a particular point may be
determined from the pressure differential generated
by the pitot tube, which is measured by an inclined
tube manometer.
The velocity of a fluid across a section of pipe is
not uniform but varies from zero at the pipe surface to
a maximum usually at the centre of the pipe. In order
to find the mean velocity it is necessary to make a
traverse of the pipe i.e., readings are taken by placing
the Pitot tube at various positions at different points
along the cross-section of the pipe. For Reynold's
numbers above 10 , the ratio of mean velocity to the
velocity at the centre of pipe is 0.82. Once this ratio is
determined exactly for a given situation, the Pitot tube
can be placed at the centre of pipe to sense the
velocity. The mean velocity is,
28
J-(p-po)
[2
=C -(P-Po) ...(25.25)
VP
where Cv is the velocity correction factor and is the ratio
of mean velocity to the velocity at the centre.
The advantages of Pitot tubes are that they
produce a negligible loss of head when inserted in the
pipe and their cost is much smaller than that of venturi
meters or orifice meters.
The major disadvantage of Pitot tubes is that they
are laboratory type instruments as for their operation
they require mean velocity to be determined by
making transverse measurements across the diameter
of pipe and hence they cannot be used for industrial
applications which require an instant readout. The
other disadvantages are that they produce a low
differential head on account of which their sensitivity
is low, and they require high flow velocities of about
15 m/s to produce measurable heads. The small
opening of Pitot tubes may get clogged if fluids
containing solid particles are used which may give
rise to wrong results.
Weirs and flumes are commonly used devices for
open channel measurements. A detailed description of
open channel measurements is beyond the scope of
this text. The purpose here is to acquaint the reader
with this type of measurements.
opening at the top through which liquid flows as
shown in Fig. 25.21.
Fig. 25.21 Rectangular Weir.
The weirs make use of rectangular V, and
trapezoidal notches as shown in Fig. 25.22. The rate of
flow is simply measured by measuring the head of
water above the lowest point of the weir opening
through which the liquid flows. This height is
measured by means of a float installed in a box called
stilling well, which is a part of the total structure. The
float is so placed that it is not disturbed by the velocity
of flow or by turbulence of the stream.
The discharge through a rectangular notch is,
7 _
Q=3CrfL^jH15 ...(25.26)
The co-efficient of discharge, C^, varies with the
dimensions, of the weir and is usually less than 0.7.
The rectangular notch is the most commonly used
notch on account of its simplicity, ease of construction
and accuracy.
 Electrical and Electronic Measurements and Instrumentation

Fig. 25.23 Venturi flume.

The discharge can be calculated as in the case of a

venturi meter. The actual discharge is,

...(25.29)

= CMy]2gh ...(25.30)

where h - h1 -h2
Fig. 25.22 Types of Notches. M=A2/yll + (A2/A2)2

The discharge through a V notch is, C = venturi flume co-efficient (0.95 to 1)

Q=^c^H2'5‘an| ...(25.27) The maximum value of flow occurs when

h2 = f-) h.
The co-efficient of a discharge varies with notch
2 Uj
angle 0, head H, kinematic viscosity and surface
tension of liquid. Typical values of Cd are 0.58 to 0.7. Maximum discharge is given by,
The V notch offers the widest range for a single Q
^max
= 1.7 2
H1'5 ...(25.31)
size, the small opening at the lower end can
accommodate small flows while the top portion is for
large flows. However, it gives the highest loss of head
because of its shape.
The trapezoidal notch is a combination of the
rectangular and triangular notches as shown in Fig.
25.22(c). The discharge is given by :
9 Q f)

3 13 Z

...(25.28)
Applications of weirs are primarily in water
works and waste and sewage systems. Measurements
can be made from a few litres per minute to millions of
litres per day.
(if) Flumes
Figure 25.23 shows a venturi flume. It is an open
channel whose sides are arranged sharply to form a
throat, followed by a fairly long expansion section.
Stilling wells are provided for measuring the head at
both entrance and throat as shown in Fig. 25.24.
mbmwmk
Fig. 25.24 Elevation and plan of flume.
Flumes are capable of better accuracies than weirs
because the loss of head in improved versions of
flumes is about one fourth of that of weir.
Weirs and flumes have not previously found
application in process industries, but their use is
increasing with the increasing demand for water
pollution control. These measurements are a must
when large volumes of water are handled in waste
treatment facilities.
Primary Sensing Elements and Transducers 753

Example 23.1 A C type bourdon tube made of Monel Example 23.3 A flat circular diaphragm of mild steel
metal is show in Fig. 25.9. The dimensions of the tube are: has a diameter of 15 mm. For mild steel, Young's modulus =
r = 36.5 mm, x = 16 mm, y = 3 mm, t = 0.35 mm 200 GN/m and Poisson's ratio v = 0.28. Find the thickness
of the diaphragm if the maximum stress is not to exceed
Calculate the displacement of the free end if a pressure of 300 MN/m when the applied pressure is 300 kN/m . Find
1500 kPa is applied. The modulus of elasticity for Monel the deflection at the centre for a pressure kN/m .
metal is 180 GN/m2.
Also calculate the natural frequency of the diaphragm if
Solution. From Eqn. 25.11, the displacement of the density of mild steel is 7800 kg/m3.
the tip is given by,
Solution. From Eqn. 25.19, maximum stress,
n / \0.2
a aP I -r I
n nc
Aa = 0.05 — _3D2P
E UJ w 16 t2
71 x36.5xl0“3x 1500xlO3
= 0.05 x---------------------------------- Thickness of diaphragm
180 xlO9
f36.5f'2 (16f2 f 16 f s_ l3D2P _ I3x(15x10~3)2x30x103
<0.35 J I 3 ) <0.35 J yi6S,„ V 16x300xl0 6

= 0.02 m =2 mm = 0.205 xlO-3 m =0.0205 mm

Example 23.2 A differential bellows arrangement for From Eqn. 25.17, the deflection at the centre for a
measuring absolute pressure is shown in Fig. 25.25. It uses pressure of 150 kN/m is,
two bellows each of natural length 50 mm, effective area = 3P(l-v2)
1500 mm , and stiffness = 0.5 N/mm. Bellows A is w 256 Et2
evacuated and a contains a spring of stiffness 3 N/mm. Find
d4 _ 3x 150x 1Q3(1-0,282)x (15x 1Q~3)4
the required natural length of the spring if the bellows are to
be equally compressed to a length 40 mm when a pressure of 256 x 200 x 109 x (0.205 x 10-3)3
100 kN/m absolute is applied to B. Also find the = 47.6 x 10~6 m = 0.0476 mm
displacement of the output point Cfor a change of 10 kN/m
The natural frequency is given by Eqn. 25.20,
in applied pressure.
Spring 20 t | E
n D2 pp(l-v2)
20xQ205xlQ~3 I 200 xlO9
(15xl0“3)2 V 3x7800 (T-0.282)
= 51950 rad/s = 8.3 kHz
Example 23.4 A mild steel torsion bar of 30 mm
bellows, A diameter is used for measurement of a torque of 100 Nm.
Calculate the angle of twist if the shear modulus of mild
Fig. 25.25 Bellows arrangement for measurement
of absolute pressure. steel is 80 GN/m2.

Solution.
Solution. From Eqn. 25.9,
2 Angle of twist
When a pressure of 100 kN/m is used.
Force applied 0= 16T ________ 16x100_______
F = PA = 100 x 103 x 1500 x 10”6 = 150 N nGd3 ~ 7ix 80x 109 x (30x 10-3)3

:. Compression of spring = 150/3 =50 mm = 0.236xlO-3 rad =0.81'

Thus natural length of spring = 50 + 40 = 90 mm
2
When a pressure of 10 kN/m is applied, the force is,
F = lOx 103 x 1500x 10-6 = 15 N
Total stiffness of spring plus stiffness of bellows
= 3 + 2 x 0.5 = 4 N/mm
Displacement of point C = 15/4 = 3.75 mm
Example 23.3 A venturi tube of throat diameter 60 mm
is placed in a water pipe of diameter 100 mm to measure the
volumetric flow. The volumetric flow rate through the tube
is 0.08 m3/s and the water has a density of 103 kg/m3 and
viscosity of 10~3 Ns / m2.
(a) Determine the Reynold's number for these
conditions.
 754 Electrical and Electronic Measurements and Instrumentation

(b) The co-efficient of discharge is 0.99, determine the Solution, (a) For a Pitot tube the velocity,
upstream-to-throat differential pressure.
(c) The differential pressure calculated in (b) is
applied across an unstressed flat diaphragm. The
diaphragm is made of mild steel for which
Young's modulus is 206 GN/m and Poisson's
ratio = 0.28, and has a diameter of 10 mm and a
thickness of 0.2 mm. Determine the deflection of
the diaphragm.
Solution.
(fl) Area of throat
A correction to the result obtained above has to be
= (n / 4) x (60 x 10-3 )2 = 2827 x 10~6 m 2
applied for taking into account the Mach number. The
Velocity through throat Mach number is the ratio of the velocity of an object
moving through a gas, to the velocity of sound in air.
0-08 ,
v =------------- - = 28.3 m / s The velocity of sound in air at atmospheric pressure is
2827 xl0“6
about 330 m/s, and therefore the Mach number for the
From Eqn. 25.21, Reynold's number is, air velocity being measured is 175.4/330 - 05. At this
subsonic Mach number, a small correction of about
n vdP 28.3x60xl0~3xl0~3 .
Re =----- =----------------------------- = 1.7 x 106 3 per cent is applied. Therefore, the correct velocity of
H 10~3
air is (1 -0.03)x 175.4 = 170 m/s.
(b) From 25.23, velocity approach factor, Example 25.7 Water is flowing with a depth of 0.9 m
E = ^1 -(4 / ^ )2 = Vl~(60/100)2 = 1.25 over a rectangular notch 1.2 m wide. Find the depth offlow
over a right angled notch if the same were to pass over it.
From Eqn. 25.22, flow rate, Assume that both the notches have the same co-efficient of
discharge.
Q = CEB
Solution. For rectangular notch, the discharge is,
The value of expansion factor B for liquids is 1

.-. Q = 0.99x 1.25x 1 x—(60x 10-3)2 = 0.08

4 VlxlO3 2 ____
= - crfx 1.2 xVTx 9.81 (0.9)1-5
.•. Differential pressure
AP = 261 kN/m2 = 3.02 C,(7

(c) The deflection at the centre of the diaphragm is For a V notch, the discharge is,
given by Eqn. 25.17 « ,__ . a

_3P(1-u2)D4
” 256 Et3 Q
or 3.02 C, = — Cd 72x9.81 H25 tan 45°
_ 3x261x 103(l-Q.282)x (IQx 10~3)4
H= 1.1 m
256x206x 109x(0.2 xlO"3)3
= 17.1xl0-6m =17.1 pm Example 25.8 A right angled notch has a co-efficient of
discharge of 0.6. If head above the still is 0.5 ± 0.01 m,
Example 2,^.0 A Pitot tube is used for measurement of calculate the discharge.
velocity offlow of water having a density of 1000 kg/m .
Solution. For a right angled Vnotch, discharge is,
(a) Determine the velocity of flow at the head of the
pitot tube if it produces a differential pressure of
10 kN/m~ between its two outlets.
g
(b) The same differential pressure is obtained in air at = — X 0.6 72x9.81 X 0.52-5
an altitude where the density of air is 0.65 kg/m3.
Determine the velocity of air flow. = 0.2506 m3 / s
Primary Sensing Elements and Transducers 755
Now — 8 r-'
dQ- = — C, J2 x 2.5 irl.i
n cx H
dH 15 * however, the electrical output is measured by standard
dQ,
or
Q H
or dQ = ±2.5x0.2506x0.01/0.5
= ±0.0153 m3/s
Hence, the discharge can be written as 0.2506 ±
0.0153 m3/s.
25.6 TRANSDUCERS
An electronic instrumentation system consists of a
number of components to perform a measurement
and record its results. As explained earlier a generalized
measurement system consists of three major
components.
(z) an input device,
(zz) a signal conditioning or processing device, and
(zzz) an output device.
The input device receives the measurand or the
quantity under measurement and delivers a propor­
tional or analogous electrical signal to the signal
conditioning device. Here the signal is amplified,
attenuated, filtered, modulated, or otherwise
modified in format acceptable to the output device.
The input quantity for most instrumentation
systems is a "non-electrical quantity". In order to use
electrical methods and techniques for measurement,
manipulation or control, the non-electrical quantity is
generally converted into an electrical form by a device
called a "transducer". We can define a transducer as a
device which, when actuated transforms energy from
one form to another.
The broad definition of a transducer includes, for
example, devices which convert mechanical force into
an electrical signal. These devices form a very large
and important group of transducers commonly used
in industrial instrumentation area. The instrumen­
tation engineers and technologists are therefore
primarily concerned with this area of instrumentation.
Many other physical parameters such as heat, intensity
of light, flow rate, liquid level, humidity and pH value
may also be converted into electrical form by means of
transducers. These transducers provide an output
signal when stimulated by a mechanical or a
non-mechanical input : a photoconductor converts
light intensity into change of resistance, a
thermocouple converts heat energy into electrical
voltage, a force produces a change of resistance in a
strain gauge, an acceleration produces a voltage in a
piezo-electrical crystal and so on. In all cases,
methods, giving the magnitude of the input quantity
in terms of an analogous output.
25.7 ELECTRIC TRANSDUCERS
The art of electrical measurements has been
chiefly used for measurement of electrical quantities
but its value in making measurements of non­
electrical quantities in this new era of automation is
rapidly growing. In order to measure non-electrical
quantities a detector is used which usually converts
the physical quantity into a displacement. This
displacement actuates an electric transducer, which
acting as a secondary transducer, gives an output that
is electrical in nature. The electrical quantity so
produced is measured by standard methods used for
electrical measurements. The result (electrical output)
gives the magnitude of the physical quantity or
condition being measured.
The electrical signal may be a current or a voltage
or a frequency and production of these signals is based
upon electrical effects which may be resistive,
capacitive, inductive etc. in nature.
The first stage of a measurement system may
simply be called a transducer stage instead of detector
transducer stage by redefining a transducer. A trans­
ducer, in general form, may be defined as a device
which converts energy from one form to another.
However, this definition has to be restricted, many a
time especially in the field of electrical instrumen-
tatiOfl. keeping this restriction in view, a transducer
may be defined as a device which converts a physical
quantity or a physical condition into an electrical
signal. Another name for a transducer is pick up.
25.7.1 AdvflnfagdS of Electrical Transducers
There are a number of transducers which trans­
form a variety of physical quantities and phenomena
into electrical signals. The reasons for transforming a
ph sical phenomenon into electrical form are
numerous. The advantages of converting physical
quantities into analogous electrical quantities are :
(z) Electrical amplification and attenuation can
be done easily and that too with static devices.
(z'z) The mass-inertia effects are minimised. In
fact, when dealing with electrical or electronic signals,
the inertia effects are due to electrons which have
negligible mass. In many situations, we do not come
across mass or inertia problems at all.
(zzz) The effects of friction are minimised.
 756 Electrical and Electronic Measurements and Instrumentation
(iv) The electrical or electronic systems can be
controlled with a very small power level.
(t?) The electrical output can be easily used, trans­
mitted and processed for the purpose of measurement.
(vi) Telemetry is used in almost all sophisticated
measurement systems. The entire aerospace research
and development is based upon telemetry and remote
control. The ever-enlarging field of radio monitoring
in space research has left us with no alternative but to
resort to electronic means. This completely eliminates
the data transmission through mechanical means and
hence electrical and electronic principles have to be
employed for these conditions. The remote indication
or recording is an essential part of modern day
instrumentation technology.
(vii) There has been an explosive development in
the field of electronic components and devices. This
development is on account of the fact that electronic
devices are very amenable to miniaturization.
Components which are compact, have always an,
advantage. Miniaturization on account of use of ICs
(integrated Circuits) has completely revolutionised
the field of instrumentation.
In short, it can be said that the reasons for
transforming a physical phenomenon into electrical
form is that the electrical output can be easily used,
transmitted and processed for the purposes of
measurement. Modern digital computers make the
use of these transducers absolutely essential.
When the definition of transducer is confined to a
device that covers the entire detector transducer
stage, wherein the transducer converts a non-electrical
quantity into an analogous electrical signal, the trans­
ducer may be thought of consisting of two important
and closely related parts. These two parts are :
A Sensing Element, and
A Transduction Element.
In addition, there may be many other auxiliary
parts, such as amplifiers and other signal processing
equipment, power supplies, calibrating and reference
sources, and mechanical mounting features.
1. Sensing or Detector Element. A detector or a
sensing element is that part of a transducer which
responds to a physical phenomenon or a change in a
physical phenomenon. The response of the sensing
element must be closely related to the physical
phenomenon.
2. Transduction element. A transduction element
transforms the output of a sensing element to an
electrical output. The transduction element, in a way,
acts as a secondary transducer.
25.8 CLASSIFICATION OF TRANSDUCERS
The transducers can be classified
(z) on the basis of transduction form used,
(ii) as primary and secondary transducers,
(iii) as passive and active transducers,
(iv) as analog and digital transducers, and
(v) as transducers and inverse transducers.
25.8.1 Classification based upon
Principle of Transduction
The transducers can be classified on the basis of
principle of transduction as resistive, inductive,
capacitive etc. depending upon how they convert the
input quantity into resistance, inductance or capaci­
tance respectively. They can be classified as piezo­
electric, thermoelectric, magnetorestrictive, electro-
kinetic and optical. Table 25.2 lists the classification of
transducers based upon principle of transduction.
25.8.2 Primary and Secondary Transducers
Let us consider the case of a Bourdon's tube as
shown in Fig. 25.26. The Bourdon tube acting as a
primary detector senses the pressure and converts the
pressure into a displacement of its free end. The
displacement of the free end moves the core of a linear
variable differential transformer, (L.V.D.T.) which
produces an output voltage which is proportional to
the movement of the core, which is proportional to the
displacement of the free end which in turn is
proportional to the pressure. Thus, there are two
stages of transduction, firstly the pressure is converted
into a displacement by Bourdon tube then the
displacement, is converted into an analogous voltage
by L.V.D.T. The Bourdon tube is called a "Primary
Transducer" while the L.V.D.T. is called a "Secondary
Transducer".
Fig. 25.26 Measurement of pressure using Bourdon
tube an L.V.D.T. (Example of primary and Secondary
transducers).
Primary Sensing Elements and Transducers 757
Table 25.2 Types of Electrical
Electrical parameter
and class of transducer Principle of Operation Typical applications

_______________________ __________ Passive transducers (externally powered)

Resistance

Potentiometer device Positioning of the slider by an external force varies the Pressure, displacement.
resistance in a potentiometer or a bridge circuit.
Resistance strain Resistance of a wire or semiconductor is changed by elongation Force, torque,
: gauge or compression due to externally applied stress. displacement.
Pirani gauge or hot Resistance of heating element is varied by convection cooling of Gas flow, gas pressure.
wire meter a stream of gas. x
Resistance Resistance of pure metal wire with a large positive tempe­ Temperature,
thermometer rature co-efficient of resistance varies with temperature. radiant heat
i Thermistor Resistance of certain metal oxides with negative temperature Temperature, flow
coefficient of resistance varies with temperature.
■ Resistance hygrometer Resistance of a conductive strip changes with moisture content. Relative humidity.
. Photoconductive cell Resistance of the cell as a circuit element varies with incident Photosensitive relay.
light.
Capacitance

j Variable capacitance Distance between two parallel plates is varied by an externally Displacement, pressure.
; pressure gauge applied force.

j Capacitor microphone Sound pressure varies the capacitance between a fixed plate Speech, music, noise.
and a movable diaphragm.
Dielectric gauge Variation in capacitance by changes in the dielectric or Liquid level, thickness.
dielectric constant.
Inductance

| Magnetic circuit Self-inductance or mutual inductance of a.c. excited coil is Pressure, displacement.
transducer varied by changes in the magnetic circuit.
Reluctance pick up Reluctance of the magnetic circuits is varied by changing the Pressure, displacement, ■
position of the iron core of coil. vibrations, position.
Differential The differential voltage of two secondary windings of a Pressure, force,
transformer transformer is varied by positioning the magnetic core through displacement, position
an externally applied force.
Eddy current gauge Inductance of a coil is varied by the proximity of an eddy Displacement,
current plate. thickness.
Magnetostriction gauge Magnetic properties are varied by pressure and stress. Force, pressure, sound.
Voltage and Current

Hall effect pickup A potential difference is generated across a semiconductor plate Magnetic flux, current,
(germanium) when magnetic flux interacts with an applied power.
current.
1
j Ionization chamber Electron flow induced by ionization of gas due to Particle counting,
radio-active radiation. radiation.
Photoemissive cell Electron emission due to incident radiation upon Light and radiation.
photoemissive surface.
Photomultiplier tube Secondary electron emission due to incident radiation on Light and radiation,
photosensitive cathode. photosensitive relays.
Self-generating transducers (no external power)
Thermocouple and An emf is generated across the junction of two dissimilar Temperature, heat flow,
thermopile metals or semiconductors when that junction is heated. radiation.
Moving coil generator Motion of a coil in a magnetic field generates a voltage. Velocity, vibrations.
Piezoelectric pickup An emf is generated when an external force is applied to certain Sound, vibrations, accele­
crystalline materials, such as quartz. ration, pressure changes.
Photovoltaic A voltage is generated in a semiconductor junction device Light meter, solar cell.
when radiant energy stimulates the cell.
 758 Electrical and Electronic Measurements and Instrumentation
Let us take another example which is the case of
measurement of a compressive force with the help of a
load cell in conjunction with strain gauges as shown in
Fig. 25.27. (Load cells and strain gauges are explained
later in this chapter.)
Force
Strain
gauges
Force
Fig. 25.27 Force measurement with load cell
and strain gauges.
The load cell is a short column or a strut with
resistance wire strain gauges bonded to it. The
measurand, in this case, is a force and is applied to the
column thereby producing strain. The force is first
detected by the column and is converted into strain
which is a mechanical displacement. The higher the
force, the higher is the strain and thus the input signal
(force) is converted into an analogous output (strain).
This strain changes the resistance of the strain gauges.
Thus we have an output which is a change in the value
of resistance i.e., electrical in form. Hence, in this case,
it takes two processes to convert an input into an
analogous output. The first process involves con­
version of force into mechanical displacement which
is done by the column, while the second process
involves conversion of mechanical displacement into
change of resistance which is done by strain gauges.
Thus we see that the force is detected by the
column in the first stage and hence it is called a
Primary Transducer. The output signal from the
primary transducer is converted subsequently into a
usable output by the strain gauges and therefore they
are known as Secondary Transducers.
In most of measurement systems, there is a
suitable working combination wherein a Mechanical
device acts as a primary detector transducer and the
electrical device acts as the secondary transducer
with mechanical displacement serving as the
intermediate signal.
25.8.3 Passive and Active Transducers
Transducers may be classified according to
whether they are passive or active.
1. Passive Transducers. Passive transducers
derive the power required for transduction from an
auxiliary power source. They also derive part of the
power required for conversion from the physical
quantity under measurement. They are also known as
"externally powered transducers". Typical examples of
passive transducers are resistive, inductive and
capacitive transducers.
A typical example of a passive transducer is a
'POT' which is used for measurement of displace­
ment. A 'POT' is a resistive transducer powered by a
source voltage e- as shown in Fig. 25.28. This 'POT' is
used for measurement of linear displacement x;.
Fig. 25.28 Linear potentiometer (POT),
a passive transducer.
Suppose L is the total length of potentiometer
whose total resistance K •. The input displacement is x^
:. Output voltage
S°=L
In the absence of external power, the transducer
cannot work and it hence is called a passive transducer.
2. Active Transducers. Active transducers are
those which do not require an auxiliary power source
to produce their output. They are also known as self
generating type since they develop their own voltage or
current output. The energy required for production of
output signal is obtained from the physical quantity
being measured.
Velocity, temperature, light intensity and force
can be transduced with the help of active transducers.
These transducers include tachogenerators, thermo­
couples, photovoltaic cells and piezoelectric crystals.
Consider the case of a piezoelectric crystal used for
measurement of acceleration as shown in Fig. 25.29
The crystal is sandwiched between two metallic
electrodes, and the entire sandwich is fastened to a
base which may be the floor of a rocket. A fixed mass
is placed on the top of the sandwich.
Primary Sensing Elements and Transducers 759
acceleration
Fig. 25.29 Piezo-electric crystral measuring
acceleration - an active transducer.
The property of the piezo-electric crystals is that
when a force is applied to them, they produce an
output voltage. The mass exerts a certain force on
account of acceleration on the crystal due to which a
voltage is generated. The acceleration is applied to the
base, due to which the mass produces a force. The
mass being fixed, the force is proportional to
acceleration. The voltage output is proportional force
and hence is proportional to acceleration (the mass
being fixed).
It should be noted from above that this transducer
called "accelerometer" which converts acceleration into
electrical voltage does not need any auxiliary power
source to convert a physical phenomenon (accele­
ration in this case) to an electrical output (voltage in
this case) and therefore is an active transducer.
25.8.4 Analog and Digital Transducers
The transducers can be classified on the basis of
the output which may be a continuous function of
time or the output may be in discrete steps.
1. Analog Transducers. These transducers convert
the input quantity into an analog output which is a
continuous function of time. Thus a strain gauge, an
L.V.D.T., a thermocouple or a thermistor may be
called as "Analog Transducers" as they give an output
which is a continuous function of time.
2. Digital Transducers. These transducers convert
the input quantity into an electrical output which is in
the form of pulses.
As the binary system uses only two symbols 0 and
1 it can be easily represented by opaque and
transparent areas on a glass scale or non-conducting
and conducting areas on a metal scale. A scale
constructed to show the linear position on a movable
object and having five digits is shown in Fig. 25.30.
The complete binary number denoting position is
obtained by scanning the pattern across the scale at a
stationary index mark.
Glass scales can be read optically by means of a
light source, an optical system and photocells. Metal
scales are scanned by brushes making electrical
contact with individual tracks.
The resolution depends upon the digits
comprising the binary number and is 1/2/7 of full scale
where n is the number of digits.
25.8.5 Transducers and Inverse Transducers
There is a strong association of control with
measurement. The basic requirement for control of
physical quantities such as position, speed,
temperature, pressure and flow rate in an industrial
planet is the ability to measure these quantities. The
control action is only possible if the physical quantity
can be measured. For example, in a position control
system called "servomechanism", it is desired to control
the position of a shaft. This requires an accurate
method for measurement of the shaft position in order
that its position be accurately controlled. Further, if it
is desired that the shaft be accelerated in a controlled
manner, then the position measuring device must be
able to measure shaft position for rapid changes i.e.,
the device must have a fast dynamic response.
Figure 25.31 shows the block diagram of closed
loop control system. The controlled (output) quantity
is usually a non electrical quantity. The control action
is through an input quantity that corresponds to the
---------------( !--------------
Feed- । 1 w
forward I
network Actuator
(amplifier) [ ।
Measuring i
instrument M-
1 or transducer 1
i ।
--------- Electrical units or signals
---------- Mechanical units or signals
Fig. 25.31 Feedback control system for control
of non-electrical quantities.
 760 Electrical and Electronic Measurements and Instrumentation

desired output (non-electrical quantity). The input

quantity called reference input is usually an electrical
quantity. The controlled quantity is measured and
converted into an analogous quantity by transducers
which form the feedback loop. The loop input
quantity (electrical in nature) is compared with the
electrical quantity proportional to the output in a
comparator. In case, the two are not equal an error
signal is produced. This error signal is amplified and
applied to an actuator in the forward path, which
corrects the output quantity till the output quantity
reaches the desired level.
1. Transducers. A transducer can be broadly
defined as a device which converts a non-electrical
quantity into an electrical quantity.
2. Inverse Transducers. An inverse transducer is
defined as a device which converts an electrical
quantity into a non-electrical quantity. It is a precision
actuator which has an electrical input and a low power
non-electrical output. A piezoelectric crystal acts as an
inverse transducer because when a voltage is applied
across its surfaces, it changes its dimensions causing a
mechanical displacement.
A current carrying coil moving in a magnetic field
is also an inverse transducer because current carried
by it is converted into a force which causes trans­
lational or rotational displacement. Many data
indicating and recording devices are inverse trans­
ducers. An analog ammeter or voltmeter converts
current into mechanical displacement. However, such
devices which include instruments like indicating
instruments, pen recorders, oscilloscopes that convert
the electrical signals to a mechanical movement are
placed at the output stage (data presentation stage)
are called output transducers.
The most useful application of inverse
transducers is in feedback measuring systems. The
development of transducers and inverse transducers,
and the advantages gained through use of feedback
has increased their applications manifold and there is
reason to believe that they will continue to grow in
importance.
A block diagram of a feedback control systems
using inverse transducers is shown in Fig. 25.32 for
comparison with the simple feedback control system
shown in Fig. 25.31. In the measuring system, the
output quantity (usually electrical in nature) is
converted to a non-electrical form suitable for
comparison with quantity to be measured
(non-electrical form). The resulting error signal is
usually transduced into electrical form and amplified
to give output indication.
--------- Electrical units or signals
----------- Mechanical units or signals
Fig. 25.32 Feedback control system for control of
electrical quantities using inverse transducer.
It is evident that the operation of the feedback is
similar for both control and measuring systems as
regards accuracy and stability. The essential
differences are in power level of the systems and the
performance requirements of transducing elements.
The control systems handle relatively high power
levels through the actuator to directly determine the
output quantity. Therefore, the actuator may have
poor regulation without affecting the control action.
The transducer and noise at the amplifier input largely
determine the accuracy of control. On the other hand,
the feedback measuring system involves use of fairly
low power devices at the output. These devices
include the indicator and inverse transducer.
However, the inverse transducer essentially
determines the characteristics of the system. The noise
connected with the transducer and amplifier input
stage is very important.
25.9 CHARACTERISTICS AND CHOICE OF
TRANSDUCERS
When choosing a transducer for any application
the input, transfer and output characteristics have to be
taken into account.
25.9.1 Input Characteristics
1. Type of Input and Operating Range. The
foremost consideration for the choice of a transducer is
the input quantity it is going to measure and its
operating range. The type of input, which can be any
physical quantity, is generally determined in advance.
A physical quantity may be measured through use of a
number of transducers. However, the choice of a
particular transducer that is selected for the purpose,
depends upon the useful range of input quantity over
which the transducer can be used. The useful
operating range of the transducer may be a decisive
Primary Sensing Elements and Transducers 761

factor in selection of a transducer for a particular
application. The upper limit is decided by the
transducer capabilities while the lower limit of range
is normally determined by the transducer error or by
the unavoidable noise originating in the transducer. In
fact the transducer should maintain a good resolution
throughout its operating range.
2. Loading Effects. Ideally a transducer should
have no loading effect on the input quantity being
measured. In theory, it is impossible, although in
practice steps may be taken to reduce the loading
effects to negligible proportions. The magnitude of the
loading effects can be expressed in terms of force,
power or energy extracted from the quantity under
measurement for working of the transducers.
Therefore, the transducer, that is selected for a
particular application should ideally extract no force,
power or energy from the quantity under
measurement in order that the latter is measured
accurately.
25.9.2 Transfer Characteristics
The transfer characteristics of transducers require
attention of three separate elements, viz.,
transfer function
A error, and
A response of transducer to environmental
influences.
25.9.2 A Transfer Function
The transfer function of a transducer defines a
relationship between the input quantity and the
output. The transfer function is,
% = /(?,) -(25.32)
where q0 and £/• are respectively output and input of
the transducer.
The sensitivity of a transducer is defined as the
differential quotient,
S = -^ ...(25.33)
dcli
In general, the sensitivity of transducers is not
constant but is dependent upon the quantity z/-.
However, in some cases the relationship between
the output q0 and input z/(7 is linear. In that case the
sensitivity remains constant over the entire range of
the transducer and is defined as,
dr/.
The scale factor is defined as the inverse of
sensitivity and is therefore,
scale factor = — ...(25.35)
S dqi
Some manufacturers specify the scale factor and
still call it sensitivity.
25.9.2B Error
The errors in transducers occur because they do
not follow, in many situations the input-output
relationship given by qo = ftqj. Any departure from
the above relationship results in errors. For example,
the output on account of input, z/(- has to be q0 but in
practice an output q0 is obtained, then the error of the
instrument is,
e = ^0-r/0 -.(25.36)
The error, a can be expressed in terms of either
input or the output quantity.
The error analysis has already been given in
Chapter 3. The error can be split into three
components, which are,
A scale error
A dynamic error, and
A error on account of noise and drift.
Scale Error
The scale error comprises of four different types
of errors :
(i) Zero Error. In this case the output deviates
from the correct value by a constant factor over the
entire range of the transducer. This is shown in
Fig. 25.33.
Fig. 25.33 Transducer zero error.
(zz) Sensitivity Error. Sensitivity error occurs
where the observed output deviates from the correct
value by a constant value. Suppose the correct output
is q the output would be Kq^ over the entire range of
the transducer, where K is a constant. This is shown in
Fig- 25.34.
 762 Electrical and Electronic Measurements and Instrumentation
Fig. 25.34 Transducer sensitivity error.
(in) Non-conformity. This pertains to a case in
which the experimentally obtained transfer function
deviates from the theoretical transfer function for
almost every input. This is shown in Fig. 25.35.
Fig. 25.35 Transducer Non-conformity.
In the special case of a theoretical linear
relationship between input and output quantities, this
error is called non-linearity or non-linear distortion.
(iv) Hysteresis. All transducers are subject to the
effects of hysteresis. The output of a transducer not
only depends upon the input quantity but also upon
in input quantities previously applied to it. Therefore,
a different output is obtained when the same value of
input quantity is applied depending upon whether it
is increasing or decreasing. For decreasing values, a
greater output is obtained than with increasing values
for the same value of the input quantity. This is shown
in Fig. 25.36.
Fig. 25.36 Transducer hysteresis.
Dynamic Error
Dynamic errors occur only when the input
quantity is varying with time. This is on account of the
fact the systems contain energy storage elements and
due to this the output cannot follow the input exactly
but with a time lag.
Let us consider an R-C series circuit to which a
step input of magnitude E is applied, the voltage
across the capacitor after a time, t of application of step
input is,
ec = E[l-exp(-f/ t)]
where r = time constant = RC
The response is shown in Fig. 25.37.
Fig. 25.37 Response of RC circuit to a step input.
The dynamic or measurement error is,
e„;(0= E-ec=exp(-f/r)
The dynamic or measurement error can be made
small by having a small time constant. It should be
understood that the dynamic error is a function of
time. As the time after application of the input
increases, the dynamic error reduces. The steady state
error,
eSS=e,n^=e~t/T' =0
t —> 00 t —> oo
in this particular case.
Errors due to Noise and Drift
Noise and drift signals originating from the
transducers vary with time and are superimposed on
the output signal. The difference between noise and
drift is that noise consists of a signal of random
amplitude and random frequency whilst drift is a slow
change with time. The magnitude of the noise and
drift is normally independent of the magnitude of the
input signal.
Primary Sensing Elements and Transducers 763
Errors due to Change of Frequency
(Frequency Response)
Frequency response and the high frequency cut off
are the two specifications that describe the response of
a transducer to a variable frequency sine wave input
applied to it. For a reasonably, linear transducer, a sine
wave input yields a sine wave output. As the
frequency of the sine wave input is increased, the
transducer is required to respond more and more
quickly. Ultimately, beyond a particular frequency the
transducer can no larger respond as rapidly as its
sinusoidal input is changing. So the output of the
transducer becomes smaller and also the phase shift
between the input and output increases. Thus as the
frequency increases the output of the transducer falls.
This roll off of amplitude of output with an increase in
Fig. 25.38 Frequency response plot.
input frequency is the frequency response. The
frequency response of a transducer is shown in
Fig. 25.38. The vertical axis is the gain of the
transducer. This may be simply the output to input
ratio. However, the gain is specified in terms of
decibels.
Gain in db = 20 log (output/input) = 20 log (^Ay,)
...(25.37)
The horizontal axis represents the frequency, and
is plotted on a 'logarithmic scale in order to cover a
wide range of frequencies.
The high frequency, cutoff (fh) is the frequency at
which the transducers gain has fallen to 0.707 of the
value it had at its stable low frequency. If the vertical
axis is scaled in db, fh occurs when the gain has
decreased by -3 db. Figure 25.38 shows that the low
frequency gain is 20 db but drops of 17 db (20 - 3 = 17 db)
at a frequency of 4 kHz. Therefore, the high frequency
cutoff takes place at 4 kHz in this case.
The high frequency cutoff can be related to the
rise time by the following relationship :
tr =0.35/fh ...(25.38)
Therefore, even if the transient response
specifications are not given, the speed of a transducer
can be determined from its high frequency cutoff.
25.9.3 Transducer Response
The response of the transducer to environmental
influences is of a great importance. This is often given
insufficient attention when choosing the best
transducer for a particular measurement. This gives
rise to results that are not as accurate as expected, or,
worse, results that are accepted as more accurate than
they actually are. The performance of the tiansducer is
fully defined by its transfer function and errors,
provided that the transducer is in constant
environments and not subject to any disturbances like
stray electromagnetic and electrostatic fields,
mechanical shocks and vibrations temperature
changes, pressure and humidity changes, changes in
supply voltage and improper mechanical mountings.
If transducers are subjected to the above environ­
mental disturbances, which they are, precautions are
taken, so that changes in transfer function and
resulting errors therefrom do not occur.
Therefore, the transducer selected must be
guarded against the interfering and modifying
inputs.
25.9.4 Output Characteristics
The three conditions in the output characteristics
which should be considered are,
A Type of electrical output,
A Output impedance, and
A Useful ouput range.
Type of Electrical Output
The types of output which may be available from
the transducers may be a voltage, current, impedance
or a time function of these amplitudes. These output
quantities may or may not be acceptable to the latter
stages of the instrumentation system. They may have
 764 Electrical and Electronic Measurements and Instrumentation

to be manipulated i.e., their magnitudes changed or

they may have to be changed in their format by signal
conditioning equipment so as to make them drive the
subsequent stages of instrumentation system.

Output Impedance
The output impedance, Zo, of a transducer
determines to the extent the subsequent stages of
instrumentation is loaded. Ideally, the value of output
impedance should be zero if no loading effects are
there on the subsequent stage. However, the output
impedance, Zo, cannot made equal to zero and
therefore, its value should be kept as low as possible to
minimise the loading effects.
The output impedance determines the amount of
power that can be transferred to the succeeding stages
of the instrumentation system for a given output
signal level. If the output impedance is low compared
to the forward impedance of the system, the
transducer has the characteristics of a constant voltage
source (provided the output of the transducer is a
voltage), while in case the forward impedance is high
as compared with the output impedance of
transducer, it behaves as constant current source.
This can be explained as under :
An ideal constant voltage source, furnishes at its
output terminals, a voltage that is independent of the and voltage across load
load connected to the source or of the current drawn
from the source. The voltage may or may not vary in ...(25.40)
‘L r0 + rl 1+r<,/rl
time. In fact, it must vary in time, if it represents a
signal which is true of transducers since their output is In case the internal resistance (output resistance)
dependent upon the input signals which vary with R(/ of the source is much smaller than load resistance
time. (for instance : Ro = 1 kQ and RL = 1 MQ) then ratio
R()/RL is very small as compared with unity and hence,
If a constant voltage source having a voltage of E
voltage across the load is,
is [Fig. 25.39(a)] is applied across a resistance R, the
current is E/R as shown in Fig. 25.39(c). A constant el=e ...(25.41)
voltage source in a strict sense does not exist ; the and the current is,
output voltage will always vary with the current 1= E/Rl ...(25.42)
drawn from the source.
Thus the voltage across the load is constant and
This is because a practical voltage source has an
the current is independent of the internal resistance Ro
internal resistance (output resistance), Ro, on account
of the source. Thus the source behaves as a constant
of which the voltage across the load resistance, EL,
voltage source as indicated in region N of Fig. 25.39(e).
decreases with increase of current or decrease of load
An ideal current source furnishes a current that is
resistance RL.
independent of the load resistance. It remains constant
Figure 25.39(d) shows a practical voltage source
even if the load resistance varies. The current
which consists of a voltage source E in series with an
produced by a constant current source may or may not
internal resistance R(/ and a load resistance RL.
vary with time.
The current in this circuit is given by,
If a constant current passes through a resistor R,
£
1 =--- - ---- ...(25.39) as shown in Fig. 25.40(a), a potential difference occurs
Rno + RLr between the terminals which is given by E = IR. In a
Primary Sensing Elements and Transducers 765
strict sense, as in the case of constant voltage source, a
constant current source does not exist. A practical
current source contains a voltage source E which
causes the current. A constant current source has an
internal resistance Ry which is connected in parallel
with the current source, and is very large as compared
with the load resistance R. The current in such a
system is determined primarily by magnitudes of E
and Rl.
The current through load resistance is,
^0 I when the output resistance of transducer, is equal to
...(25.43)
If the internal resistance Ro is very large as
compared to the load resistance (for instance Ro = 1 kQ
and Rl = 1Q) the ratio RL/R0 is very small so compared
with I and therefore, load current,
IL= I ...(25.44)
Therefore, the source behaves as a constant
current source as depicted in region M of Fig. 25.40(e).
There is a region O, between regions M and N where
the source neither behaves as a constant voltage
source or a constant current source.
Therefore, we conclude, if the output impedance
of the transducer is low as compared with that of load
impedance of the successive stages, it has the charac-
Fig. 25.40 Depiction of constant current source.
teristics of a constant voltage source. On the other
hand, if the output impedance of the transducer is
much higher than that of the succeeding stages, it
exhibits characteristics of a constant current source.
When the output impedance of the transducer is
equal to that of the following stages of instrumen­
tation system, matching takes place and maximum
power is transferred from the transducer to the
succeeding stages. However, it must be understood
that in case maximum power transfer takes place,
the resistance of the succeeding stages, the efficiency is
only 50%. Also, when dealing with a.c. systems, the
impedance matching takes place when the impedance
of the load is RL -/XL in case the output impedance is
rl + 7-xl.
Useful Output Range
The output range of a transducer is limited at the
lower end by noise signals which may shroud the
desired input signal. The upper limit is set by the
maximum useful input level. The output range can be
increased, in some cases, by the inclusion of amplifier
in the transducer. However, the inclusion of an
amplifier also increases the noise level and therefore in
such situations the amplifier may not be of any use at
all.
25.10 SUMMARY OF FACTORS INFLUENCING
THE CHOICE OF TRANSDUCERS
In the following parts of this chapter, an attempt
has been made to give the details and applications of
the transducers for measurement of different physical
quantities. It will be realised that there are many ways
for measurement of a physical quantity, and in many
cases, there is no best way. The transducers and the
methods used may depend upon the instrumentation
already available and also on the experience of the
user.
Unfortunately most transducers are not sensitive
to just one quantity. If measurements are to be made
under conditions where there is likelihood of two or
more input quantities influencing the transducer, it is
desirable to select a transducer which is sensitive to
the desirable quantity and insensitive to the unwanted
quantity. If this is not possible, ways and means
should be found to eliminate or compensate for the
effects of the unwanted input quantity. The following
is the summary of the factors influencing the choice of
a transducer for measurement of a physical quantity.
 766 Electrical and Electronic Measurements and Instrumentation
1. Operating Principle. The transducers are many
a times selected on the basis of operating principle
used by them. The operating principles used may be
resistive, inductive, capacitive, optoelectronic, piezo­
electric etc.
2. Sensitivity. The transducer must be sensitive
enough to produce detectable output.
3. Operating Range. The transducer should
maintain the range requirements and have a good
resolution over its entire range. The rating of the
transducer should be sufficient so that it does not
breakdown while working in its specified operating
range.
4. Accuracy. High degree of accuracy is assured if
the transducer does not require frequent calibration
and has a small value for repeatability. It may be
emphasised that in most industrial applications,
repeatability is of considerably more importance than
absolute accuracy.
5. Cross sensitivity. Cross sensitivity is a further
factor to be taken into account when measuring
mechanical quantities. There are situations where the
actual quantity is being measured is in one plane and
the transducer is subjected to variations in another
plane. More than one promising transducer design has
had to be abandoned because the sensitivity to
variations of the measured quantity in a plane
perpendicular to the required plane has been such as
to give completely erroneous results when the
transducer has been used in practice.
6. Errors. The transducer should maintain the
expected input-output relationship as described by its
transfer function so as to avoid errors.
7. Transient and Frequency Response. The
transducer should meet the desired time domain
specifications like peak overshoot, rise time, settling
time and small dynamic error. It should ideally have a
flat frequency response curve. In practice, however,
there will be cutoff frequencies and higher cut off
frequency should be high in order to have a wide
bandwidth.
8.Loading Effects. The transducer should have a
high input impedance and a low output impedance to
avoid loading effects.
9. Environmental Compatibility. It should be
assured that the transducer selected to work under
specified environmental conditions maintains its
input-output relationship and does not break down.
For example, the transducer should remain operable
under its temperature range. It should be able to work
in corrosive environments (if the application so requires),
should be able to withstand pressures and shocks and
other interactions to which it is subjected to.
10. Insensitivity to Unwanted Signals. The
transducer should be minimally sensitive to unwanted
signals and highly sensitive to desired signals.
11. Usage and Ruggedness. The ruggedness both
of mechanical and electrical intensities of transducer
versus its size and weight must be considered while
selecting a suitable transducer.
Electrical aspects. The electrical aspects that
12.
need consideration while selecting a transducer
include the length and type of cable required.
Attention also must be paid to signal to noise ratio
in case the transducer is to be used in conjunction with
amplifiers. Frequency response limitations must also
be taken into account.
13. Stability and Reliability. The transducer
should exhibit a high degree of stability to be
operative during its operation and storage life.
Reliability should be assured in case of failure of
transducer in order that the functioning of the
instrumentation system continues uninterrupted.
14.Static Characteristics. Apart from low static
error, the transducers should have a low non-linearity,
low hysteresis, high resolution and a high degree of
repeatability.
The transducer selected should be free from load
alignment effects and temperature effects. It should
not need frequent calibration, should not have any
component limitations, and should be preferably
small in size.
25.11 RESISTIVE TRANSDUCERS
It is generally seen that methods which involve
the measurement of change in resistance are preferred
to those employing other principles. This is because
both alternating as well as direct currents and voltages
are suitable for resistance measurements.
The resistance of a metal conductor is expressed
by a simple equation that involves a few physical
quantities. The relationship is
R=pL/ A,
where R = resistance ; Q,
L = length of conductor ; m,
A = cross-sectional area of conductor ; m2,
and p = resistivity of conductor material; Q-m
Primary Sensing Elements and Transducers 767

Any method of varying one of the quantities Displacement

involved in the above relationship can be the design
basis of an electrical resistive transducer. There are a
number of ways in which resistance can be changed
by a physical phenomenon. The translational and
rotational potentiometers which work on the basis of
change in the value of resistance with change in length
of the conductor can be used for measurement of
translational or rotary displacements. Strain gauges
work on the principle that the resistance of a
conductor or a semi-conductor changes when strained.
This property can be used for measurement of
displacement, force and pressure. The resistivity of (rt) Linear (translational) POT
materials changes with change of temperature thus
causing a change of resistance. This property may be
used for measurement of temperature. Thus electrical
resistance transducers have a wide field of
application.

25.12 POTENTIOMETERS
Basically a resistance potentiometer, or simply a
POT, (a resistive potentiometer used for the purposes
of voltage division is called a POT) consists of a
resistive element provided with a sliding contact. This
sliding contact is called a wiper. The motion of the
(b) Rotary POT
sliding contact may be translatory or rotational. A
linear pot and a rotary pot are shown in Figs. 25.41(67) Fig. 25.41 Resistive potentiometers (POTs)
and (b) respectively. Figure 25.42 shows the diagrams for translational,
Some POTS use the combination of the two single turn rotational, and multitum helix potentio­
motions, i.e., translational as well as rotational. These meters.
POTS have their resistive element in the form of a
helix and therefore, they are called helipots.
The translational resistive elements are straight
devices and have a stroke of 2 mm to 0.5 m. The
rotational devices are circular in shape and are used
for measurement of angular displacement. They may
have a full scale angular displacement as small as 10°.
A full single turn potentiometer may provide accurate
measurements upto 357°. Multiturn potentiometers
may measure upto 3500° of rotation through use of
helipots.

The helical resistive elements are multiturn

rotational devices which can be used for measurement
of either translational or rotary motion. The resistive
element of the POT may be excited by either d.c. or a.c.
voltage. The POT is a passive transducer since it
requires an external power source for its operation.
The resistive body of potentiometer may be wire
wound. A very thin, 0.01 mm diameter of platinum or
nickel alloy is carefully wound on an insulated former.
The resistance elements are also made up from cermet, pig. 25.42 Diagrams for translational,
hot moulded carbon, carbon film and thin metal. rotational and helipots.
 768 Electrical and Electronic Measurements and Instrumentation

Let us confine, our discussion of d.c. excited The circuits shown in Fig. 25.42 are called
potentiometers. Consider a translational potentio­ potentiometer dividers since they produce an output
meter as shown in Fig. 25.42(a). voltage which is a fraction of the input voltage. Thus
Let, the input voltage is "divided". The potential divider is
a device for dividing the potential in a ratio
e. & eQ = input and output voltages
determined by the position of the sliding contact.
respectively ; V,
xt = total length of translational pot; m,
xt = displacement of wiper from its zero
position ; m,
R? = total resistance of the potentiometer ; Q

If the distribution of the resistance with respect to

translational movement is linear, the resistance per
unit length is Rp/xv

The output voltage under ideal conditions is :

resistance at the output terminals^
*o = input voltage
resistance at the input terminals
Equations 25.46 and 25.47 are based upon the
...(25.45) assumption that the distribution of resistance with
respect to linear or angular displacement is uniform
and the resistance of the voltage measuring device (i.e.
The under ideal circumstances, the output
output device) is infinite.
voltage varies linearly with displacement as shown in
Fig. 25.43(a). However, in practice, the output terminals of the
pot are connected to a device whose impedance is
finite. Thus, when an electrical instrument, which
forms a load for the pot and is connected across the
output terminals, the indicated voltage is less than
that given by Eqn. 25.46. The error, which is referred
to as a loading error is caused by the input resistance
of the output device.
Let us consider the case of a translational poten­
tiometer as shown in Fig. 25.44. Let the resistance of a
meter or a recorder monitoring the output be R .
As explained earlier if the resistance across the
Fig. 25.43 Characteristics of potentiometers. output terminals is infinite, we get a linear relation­
ship between the output and the input voltage.
Sensitivity s = °UtF’- c0= (xi/xt)ei = Kei ...(25.48)
input
p p where K = x-/xf
= -0=_L ...(25.46)
xi xt However, under actual conditions the resistance,
Rm, is not infinite. This causes a non-linear
Thus under ideal conditions the sensitivity is
relationship between the output and input voltages.
constant and the output is faithfully reproduced and
has a linear relationship with input. The same is true 25.12.1 Loading Effect
of rotational motion. The resistance of the parallel combination of load
Let 0; = input angular displacement in degrees, resistance and the portion of the resistance of the
and 0f = total travel of the wiper in degrees. potentiometer is :
Output voltage eQ = er (0. / 0f) ...(25.47) RR,,R„,
...(25.49)
This is true of single turn potentiometers only. (xf/x,)Rp + R„, KRp + Rm
Primary Sensing Elements and Transducers 769

The total resistance seen by the source is : Except for the two end points where K=0 i.e.,
RRpR.n x. =0 and K = 1 where x; =x, the error is always
R=Rp(l-K) + negative. Figure 25.45 shows a plot of the variation in
KRP+ Rm error with the slider position for different ratios of the
load (output device or meter) resistance to the
KR^(1-K)4 RpR„,
...(25.50) potentiometer resistance.
KR,, + R,„ The error as indicated in Fig. 25.45 is actually
~ L • ei negative. Examine Fig. 25.45, the maximum error is
Current i = — about 12 per cent of full scale if Rm^p = I- This error
R
drops down to about 1.5 per cent when Rm/Rp = 10. For
values of R„/Rp > 10, the position of maximum error
' KR2p(l-K)+RpRm
occurs in the vicinity of x;- / xf= 0.67.
The output voltage under load conditions is :
RRpRm

^Rp + RJ ;; KRpRm
(KR p+Rm)
" KRp(l-K)+RpRm *

=--------------------------- ... (25.51)

K(l-K)(Rp/Rm) + l

The ratio of output voltage to input voltage under

load conditions is :
eo __________ R________ ...(25.52)
e. K(l-K)(Rp/R„;) + l
Fig. 25.45 Variation of error due to loading
Equation 25.52 shows that there exists a effect of a potentiometer.
non-linear relationship between output voltage and
input displacement xj since K-xff xr In case Rm = oo, Maximum percentage error
eQ I e- = K. ^ax=15x(Rp/Rm) ...(25.55)
It is evident from Eqn. 25.52 that as the ratio of
It should be understood that the error in POTS is
RffRp decreases, the non-linearity goes on increasing.
on account of the non-linearity effect produced by the
This is shown in Fig. 25.43(b). Thus, in order to keep
linearity, the value of Rm/Rp should be as large as output device used for measurement whose input
possible. However, when we have to measure the resistance, R , is finite. If meter resistance were
output voltage with a given meter, the resistance of infinite there would be no linearity effects and the
the potentiometer, Rp, should be as small as possible. output voltage will be a linear function of input
displacement x- as shown in Fig. 25.43 and
Error = output voltage under load consequently, there would no error because of the
- output voltage under no load
absence of non-linearity.
=---- ---- —---------------e. K 25.12.2 Power Rating of Potentiometers
[K(l-K)(Rp/R,„) + l] '
The potentiometers are designed with a definite
K2(K-1) power rating which is related directly to their heat
...(25.53)
K(l-K)+Rm/Rp dissipating capacity. The manufacturer normally
designs a series of potentiometers of single turn with a
Based upon full-scale output, this relationship diameter of 50 mm with a wide range of ohmic values
may be written as : ranging from 100 Q to 10 kQ in steps of 100 Q. These
K2(K-1) potentiometers are essentially of the same size and of
xlOO ...(25.54) the same mechanical configuration. They have the
K(l-R) + (Rm/Rp)
same heat transfer capabilities. Their rating is typically
 770 Electrical and Electronic Measurements and Instrumentation

5 W at an ambient temperature of 21°C This limits The linearity can be improved by using a
their input excitation voltage. Since power P = ej I Rp, resistance Rjn (made equal to the meter resistance Rm)
the maximum input excitation voltage that can be as shown in Fig. 25.46 without using a meter of higher
used is : resistance. The input-output characteristics with and
without Rjft in the circuit are shown in Fig. 25.46(b).
(“’.Lax = volt -(25.56)

25.12.3 Linearity and Sensitivity

It has been explained earlier that in order to
achieve a good linearity, the resistance of
potentiometer R , should be as low as possible when
using a meter for reading the output voltage which
has a fixed value of input resistance R .
In order to get a high sensitivity the output
voltage eQ should be high which in turn requires a high
input voltage, ef. Due to limitations of power
dissipation as is clear from Eqn. 25.56, the input
voltage is limited by the resistance of the potentio­ Fig. 25.46 Linearization of POT characteristics,
meter. In order to keep the power dissipation at a low through use of an external resistance Rln = Rm.
level, the input voltage should be small and resistance
of the potentiometer should be high. Thus for a high 25.12.4 Construction of Potentiometers
sensitivity, the input voltage should be large and this The resolution of the potentiometers influences
calls for a high value of resistance R . On the other, if the construction of their resistance elements.
we consider the linearity, the resistance of potentio­ Normally, the resistive element is a single wire of
meter should be as low as possible. The resistance conducing material which gives a continuous stepless
of the potentiometer, Rp, cannot be made low because
variation of resistance as the wiper travels over it.
if we do so the power dissipation goes up with the
Such potentiometers are available but their length (in
result that we have to make the input voltage small to
the case of translational potentiometers) and diameter
keep the power dissipation to the acceptable level.
(in the case of rotational potentiometers) restrict their
This results in lower sensitivity.
use on account of space considerations.
Thus linearity and sensitivity are therefore two
The resolution of the potentiometers is dependent
conflicting requirements. If R? is made small, the
linearity improves, but a low value of R , requires a upon the construction of the resistive element and in
lower input voltage e- in order to keep down the order to get high values of resistance in small space
power dissipation and a low value of e- results in a wire wound potentiometers are used extensively. The
lower value of output voltage eQ resulting in lower resistance wire is wound on a mandrel or a card for
sensitivity. Thus the choice of potentiometer translational displacement as shown in Fig. 25.47(a).
resistance, R , has to be made considering both the For the measurement of rotational motion these
linearity ana sensitivity and a compromise between mandrels or cards are formed into a circle or a helix.
the two conflicting requirements has to be struck.
The maximum available sensitivity of poten­
tiometer varies considerably from type to type and
also with size in a given type. The sensitivity can be
calculated from the data supplied by the manu­
facturers which includes maximum allowable voltage,
current or power and the maximum stroke. The
shorter stroke devices have generally a higher
sensitivity. Extreme values are of the order of
15 V/degree for short stroke sector type rotational
potentiometers and 12 V/mm for short stroke
(6.25 mm) translational potentiometers. These, in fact,
are maximum achievable values and usual values of
sensitivity are 10 to 100 times smaller.
Primary Sensing Elements and Transducers J 71
(b) Circular (Rotational) potentiometer.
Fig. 25.47 Wire wound potentiometers.
This is shown in Fig. 25.47(b). If wire wound type of
construction is adopted, the variation of resistance is
not a linear continuous change but is in small steps as
the sliding contact (wiper) moves from one turn to
another. This is shown in Fig. 25.48.
Fig. 25.48 Translational potentiometer and
its characteristics.
Since the variation in resistance is in steps the
resolution is limited. For instance, a translational
potentiometer has about 500 turns of a resistance wire
on a card of 25 mm in length and for this device the
resolution is limited to 25/500 = 0.05 mm = 50 pm.
The actual practical limit is 20 and 40 turns per
mm. Thus for translational devices the resolution is
limited to 25 -50pm.
For rotational devices, the best angular resolution
3 to 6 ,
=------ degrees ... (25.57)
where D = diameter of the potentiometer ; mm
In order to get higher resolution, thin wires which
have a high resistance have to be put close to each
other and they can be closely wound on account of
their small diameter. Thus the resolution and total
resistance are interdependent.
In case a fine resolution and high resistance are
required a carbon film or a conductive-plastic
resistance elements are used. Carbon film resistive
elements have a resolution of 12.5 nm.
25.12.5 Helipots
The resolution can be increased by using multi­
turn potentiometers. These are called helipots. The
resistance element is in the form of a helix and the
wiper travels along a "lead screw". The number of
turns is still limited to 20 to 40 turns per mm but an
increase in resolution can be obtained by using a
gearing arrangement between the shaft whose motion
is to be measured and the potentiometer shaft. As an
example, one rotation of the measured shaft can cause
10 rotations of the potentiometer shaft. This increases
the resolution of the measured shaft motion by
10 times. Multitum potentiometers are available to
about 60 turns.
Similar magnifying or amplifying techniques can
be used for translational motion as well.
25.13 NON-LINEAR POTENTIOMETERS
FUNCTION GENERATORS
In many instrumentation problems, non-linear
conditions arise. These may include the effect of
magnetic saturation, the non-linearity of the voltage
produced by a thermocouple as a function of
temperature, the air speed expressed in Mach number
as a function of wind-tunnel pressures, trigonometric
functions, etc. One simple method of solving these
non-linear functions is by the use of non-linear poten­
tiometers. A standard self-balancing potentiometer is
used to produce a displacement proportional to the
input voltage e- is shown in Fig. 25.49. This
displacement also operates a non-linear resistance 25
which produces an output voltage which is the
desired function of the input voltage e-. This operation
is illustrated in the following example. Let the input
voltage e- vary from 0 to 10 V. The desired function of
this voltage is as plotted in Fig. 25.49(b).
eQ = 0.3 ei + 0.5 e2 ...(25.58)
The maximum output voltage
e0(max) = 03 X 10 + 03 X 0 °)2 = 53 V-
A displacement proportional e- is obtained using
an amplifier and a servomotor as shown in
Fig. 25.49(rt). A good approximation of the required
resistance of the non-linear potentiometer is obtained
by shunting a linear potentiometer at 10 equally
spaced taps with the correct resistances. The desired
voltage above ground is computed from Eqn. 25.58
and are listed in Table 25.3. The current is assumed to
be 1 mA. The resistance R is the resistance from each
tap to ground, and AR is the resistance between
adjacent taps. These resistances are computed and
 772 Electrical and Electronic Measurements and Instrumentation

listed in Table 25.3. The circuit of the non-linear
potentiometer with resistances between each tap is
shown in Fig- 25.49.
The maximum resistance between taps is 980 fl
Then a 10,000 Q, 10-tap potentiometer with 1,000 fl
between taps can be used with the correct shunt
resistance externally connected between the taps for
the non-linear potentiometer. The shunt resistance Rx
between each pair of taps necessary to produce the
desired non-linearity is,

where is the resistance between the pair of adjacent

taps of the potentiometer without the shunting
resistors, and AR is the desired resistance.
These values are given in Table 25.3. The
difference between the desired function and the ap­
proximation obtained with this 10-tap potentiometer
is the difference between the curve in Fig. 25.49 and
the straight lines connecting the points on the curve.
An adjustable tapped potentiometer is-shown in
Fig. 25.50. This has the advantages that it can be
adjusted to fit any desired curve, and that it is capable
of having both positive and negative voltage outputs.

80 180 280 380 480 580 680 780 880 980

•— ei —*| eo Non-linear potentiometer

----------------------------- o E = 53 V o----------------------------

Fig. 25.49 Non-linear-potentiometer-function

generator.

Table 25.3
e. R AR *x
eo
0 0 0 0
1 0.8 80 80 87
2 2.6 260 180 220
3 5.4 540 280 319
4 9.2 920 380 613
5 14.0 1,400 480 923
6 19.8 1,980 580 1,380
7 26.6 2,660 680 2,124 Fig. 25.50 Adjustable tapped potentiometer.

8 34.4 3,440 780 3,540

When a fixed function is desired, a wire-wound
9 43.2 4,320 880 7,330 potentiometer with the desired taper may be made.
i 10 53.0 5,300 980 49,000 An example of this type of potentiometer is illustrated
Primary Sensing Elements and Transducers 773
Fig. 25.51 Sine-cosine, tapered potentiometer.
in Fig. 25.51, where the taper is shown for a
continuous rotating potentiometer with a sine-cosine
output function. The potentiometer shaft is rotated at
the desired angular velocity with a servomotor when a
sine-wave input voltage is desired, or it may be
Operated with a self-balancing potentiometer circuit,
as shown in Fig 25.49 when the desired output is,
= a sin e,-
or when the axis is shifted 90°.
Cq = b cos ei
f t9. 25.52 Generation of non-linear functions
with potentiometers.
It will be remembered that servo circuits are
comparatively slow and are inadequate for fast analog
computers, yet these circuits are very accurate, usually
± 0.5 percent or better.
If the slope of the body on which the resistive wire
is wound is kept accurately uniform, there will be a
linear relationship between the output voltage and the
position of the wiper (input displacement). However,
if the shape of the body varies, it will cause a
prescribed non-linear transfer function. This is shown
in Fig. 25.52.
There are many standard functions available, or
custom position-to-voltage relationship can be
produced. This non-linear relationship may be used to
compensate for a non-linearity in the quantity being
measured or to introduce a special function into the
control and instrumentation scheme without having
to resort to special mathematics electronics or a
computer.
25.14 MATERIALS USED FOR POTENTIOMETERS
The materials used for POTS may be classified as
wire wound and non-wire wound.
1. Wire Wound Potentiometers
These are platinum, nickel chromium, nickel
copper, or some other precious resistance elements.
Wire wound potentiometers carry relatively large
currents at high temperatures. Their resistance
temperature co-efficient is usually small, is of the
order of 20 x 10~6/°C or less. Their resolution is about
0.025 - 0.05 mm and is limited by the number of turns
that can be accommodated on the card.
It should be noted that the interwinding
capacitance between turns, and between windings
and shaft, housing etc. limits the use of wire wound
potentiometers to low frequencies. The response is
limited to about 5 Hz.
2. Non-Wire Potentiometers
These are also called continuous potentiometers. The
non-wire wound potentiometers provide improved
resolution and life. This is because the resolution is no
longer limited by the number of turns that can be
wrapped onto a body. The wiper moves across a
smooth surface (not bouncing from turn to turn),
wear, bounce and the resulting failures are decreased.
The maximum speed that a wire wound potentio­
meter is turned is about 300 rpm. About that, the noise
created as the wiper bounces from turn to turn
becomes significant. A continuous potentiometer may
be turned at a speed of 2000 rpm. However, they are
 774 Electrical and Electronic Measurements and Instrumentation
more sensitive to temperature changes, have a higher
wiper contact resistance, which is variable and can
carry only moderate currents.
The materials used for non-wire wound (or
continuous) potentiometers are :
(i) Cermet. Cermet uses precious metal particles
fused into ceramic base. These fused metal particles
act as resistance elements. The advantages of using
Cermet are large power ratings at high temperatures,
low cost and moderate temperature coefficients of the
order 100 x 10-6 to 200 x 10-6/°C Cermet is very useful
for a.c. applications.
(ii) Hot moulded carbon. The resistance element is
fabricated by moulding together a mixture of carbon
and a thermosetting plastic binder. Hot moulded
carbon units are useful for a.c. applications.
(iii) Carbon film. A thin film of carbon deposited
on a non-conductive base forms the resistance
element. The advantage of carbon film potentiometers
is their low cost. Temperature coefficients are upto
1000xl0-6/°C
(zv) Thin metal film. A very thin, vapour
deposited layer of metal on glass or ceramic base is
used as a resistance element. The advantages of this
potentiometer are its excellent resistance to changes in
environments and use on a.c. The cost is also
moderate.
25.15 ADVANTAGES AND DISADVANTAGES OF
RESISTANCE POTENTIOMETERS
Resistance potentiometers have the following
major advantages :
(i) . They are inexpensive.
(zz) They are simple to operate and very useful
for applications where the requirements are
not particularly severe.
(zzz) They are very useful for measurement of
large amplitudes of displacement.
(z'v) Their electrical efficiency is very high and they
provide sufficient output to permit control
operations without further amplification.
(v) It should be understood that while the
frequency response of wire wound
potentiometers is limited, the other types of
potentiometers are free from this problem.
(vi) In wire wound potentiometers the resolution
is limited while in Cermet and metal film
potentiometers, the resolution is infinite.
The disadvantages are :
(?) The chief disadvantage of using a linear
potentiometer is that they require a large
force to move their sliding contacts (wipers).
(zz) The other problems with sliding contacts are
that they can be contaminated, can wear out,
become misaligned and generate noise. So
the life of the transducer is limited.
However, recent developments have
produced a roller contact wiper which, it is
claimed, increases the life of the transducer
by 40 times.
Example 25.9 A linear resistance potentiometer is
50 mm long and is uniformly wound with a wire having a
resistance of 10,000 □. Under normal conditions, the slider
is at the centre of the potentiometer. Find the linear
displacement when the resistance of the potentiometer as
measured by a Wheatstone bridge for two cases is: (i) 3850 Q,
(ii) 7560D. Are the two displacements in the same direction ?
If it is possible to measure a minimum value of 10 Q
resistance with the above arrangement, find the resolution
of the potentiometer in mm.
Solution. The resistance of the potentiometer at
its normal position
= 1000/2 = 5000 Q
Resistance of potentiometer per unit length
= 1000/50 = 200 Q/mm
(z) Change of resistance from its normal position
= 5000-3850 = 1150 Q
.'. Displacement of wiper from its normal position
= 1150/200 =5.75 mm
(zz) Change of resistance from its normal position
= 7560-5000 = 2560 0
Displacement of wiper from its normal position
2560/200 = 12.80 mm
Since, one of the displacements represents a
decrease and other represents an increase in resistance
of potentiometer from its value at the normal position,
the two displacements are in the opposite direction.
Resolution = minimum measurable resistance
xmm/Q
= 10x1/200 = 0.05 mm
Example 25.10 A variable potential divider has a total
resistance of 2 k£l and is fed from a 10 V d.c. supply. The
output is connected to a load resistance of 5 k£L Determine
the loading errors for the wiper positions corresponding to
K =Xj/xt=0, 0.25, 0.5, 0.75 and 1.0. Use the results to plot
a rough graph of error versus xfxt.
Primary Sensing Elements and Transducers 775

Solution. We have Rp = 2 kQ and Rlu = 5 kQ potentiometer R3R4 is adjusted so that the bridge is balanced
and eo = O. Assuming that the displacement being measured
From Eqn. 25.53, error is,
will move a maximum distance of 12.5 mm towards A,
r k2(k~i) calculate the value of e0.
e'[K(l-K)+Rm/Rp Solution. Total length of AB = 125 mm ; Midpoint
of AB = 62.5 mm from either A or B. If the wiper moves
Substituting the values, the results are tabulated 12.5 inward towards A from midstroke, its distance
below : from B becomes 62.5 + 12.5 = 75 mm.

K Error, V £2 =(75/125)x 5000 =3000 Q

0 Hence, output voltage

0
0.25 -0.174 f R2 R4 'I
0 1^+^
0.5 - 0.454
0.75 - 0.524 3000 2500
1 0 5000 5000
The graph between error versus K =xfxt is plotted
Example 25.12 The output of a potentiometer is to be
in Fig. 25.53. (The errors are no doubt - ve but they are
read by a recorder of 10 k£l input resistance. The
shown as +ve in Fig. 25.53.)
non-linearity must be held to 1 percent. A family of
potentiometers having a thermal rating of 5 W and
resistances ranging from 100 Q to 10 kQ. in steps of 100 Q
are available. Choose from the family of potentiometers, a
potentiometer that has the greatest possible sensitivity and
which meets the non-linearity requirements. Find the
maximum excitation voltage permissible with this
potentiometer. What is the sensitivity if the potentiometer is
a single turn (560° unit) ?
Solution. From Eqn. 25.55, maximum possible
percentage linearity = 15 Rp/Rm- This is to be limited to
1 percent 15 Rp/Rm =1- Therefore, the maximum value
of resistance of potentiometer is,

Fig. 25.53 Graph of results of Example 25.10.
Example 25.11 A resistive potential divider R^R2 with
a resistance of5000 Q and a shaft stroke of 125 mm is used
in the arrangement shown in Fig. 25.54. Potentiometer
R3R4 has a resistance of 5000 Q and et=5.0 V. The initial
position to be used as reference point is such that Rj = R2
i.e., the wiper is at midstroke. At the start of the test
A
R =i = 12^22 = 666.7Q
p 15 15
Thus we are left with choice of potentiometers
having resistances of 100 Q, 200 Q, 300 Q, 400 Q, 500 Q
and 600 Q.
The potentiometer with the highest value of
resistance has the highest sensitivity. Therefore, a
potentiometer with R? = 600 Q is selected. With a
power dissipation of 5 W, the maximum value of
excitation voltage is,
%ax) = 7™; =757600 = 54.8 V

The sensitivity of potentiometer when it is a single

turn unit S = 54.8/360 = 0.152 V/degree.

Example 25.13 A helipot is provided with

400 turns/mm. The gearing arrangement is such that the
motion of the main shaft causes 5 revolutions of
B potentiometer shaft. Calculate the resolution of the
Fig. 25.54 Diagram of Example 25.11.
potentiometer.
 776 Electrical and Electronic Measurements and Instrumentation
Solution. The resolution of potentiometer
without the gearing arrangement = 1/400 mm =25 pm.
With the gearing arrangement which causes 5
revolutions of the potentiometer shaft with one
revolution of main shaft = 25/5 = 5 pm.
Example 25.14 It is necessary to measure the position
of an object. It moves 0.8 m. Its position must be known
within 1 mm. Part of the mechanism which moves the object
is a shift in a shaft that rotates 250° when the object is
moved from one extreme to the other. A control
potentiometer has been found which is rated at 300° full
scale movement. It has one thousand turns of wire. Is the
potentiometer suitable for the application ?
Solution. The shaft provides a
250°
------= 312.5°/m or 0.3125°/mm conversion
0.8
A resolution of 1 mm at the object translates into
1 mm x 0.3125° = 0.3125°
Required resolution for the potentiometer.
The potentiometer actually has a resolution
= 0.300°
1000
.’. The resolution of the potentiometer is higher
than that required for the application and hence the
potentiometer is very much suited for the application.
Example 25.15 A control potentiometer is rated as :
resistance = 150 Q, power rating - 1 W. De-rate the
potentiometer by 10 mW/°C above 65°C, thermal resistance
= 30°C/W.
Can the potentiometer be used with a 10 V supply at 80°C
ambient temperature ?
Solution. Power dissipated by potentiometer
P = (10)2/150 = 0.667 W = 667 mW
The actual temperature of the potentiometer is the
sum of the ambient temperature and temperature rise
produced due to self heating.
0pot =80° + (667)x30xl0“3 = 100°C
The allowable power dissipation must be
decreased by 10 mW for each °C above 65°C
Dissipation allowed
= 1.0-10x 10“3(100 -35)
= 0.65 W= 650 mW
A first glance, it would appear that the
potentiometer is well below the rated 1 W power
dissipated. This could have been true if the
temperature of the potentiometer were 65°C.
However, the temperature of the potentiometer is
100°C and therefore de-rating in power dissipation
has to done on account of increase in temperature. The
de-rated power dissipation is 650 mW while the
potentiometer is dissipating 667 mW and hence it will
fail.
25.16 STRAIN GAUGES
If a metal conductor is stretched or compressed,
its resistance changes on account of the fact that both
length and diameter of conductor change. Also there
is a change in the value of resistivity of the conductor
when it is strained and this property is called
piezoresistive effect. Therefore, resistance strain
gauges are also known as piezoresistive gauges. The
strain gauges are used for measurement of strain and
associated stress in experimental stress analysis.
Secondly, many other detectors and transducers,
notably the load cells, torque meters, diaphragm type
pressure gauges, temperature sensors, accelerometers
and flow meters, employ strain gauges as secondary
transducers.
25.16.1 Theory of Strain Gauges
The change in the value of resistance by straining
the gauge may be partly explained by the normal
dimensional behaviour of elastic material. If a strip of
elastic material is subjected to tension, as shown in
Fig. 25.55 or in other words positively strained, its
longitudinal dimension will increase while there will
Fig. 25.55 Change in dimensions of a strain gauge
element when subjected to a tensile force.
be a reduction in the lateral dimension. So when a
gauge is subjected to a positive strain, its length
increases while its area of cross-section decreases as
shown in Fig. 25.55. Since the resistance of a conductor
is proportional to its length and inversely proportional
to its area of cross-section, the resistance of the gauge
increases with positive strain. The change in the value
resistance of strained conductor is more than what can
be accounted for an increase in resistance due to
dimensional changes. The extra change in the value of
resistance is attributed to a change in the value of
resistivity of a conductor when strained. This
property, as described earlier, is known as
peizoresistive effect.