Instrumentation (Chapter 1,0)

Chapter 1 Introduction to Instrumentation (ECEG-4201) Page |1
1.1. Basic Concepts of Measurement

The measurement of a given parameter or quantity is the act of quantitative
comparison between a predefined standard and an unknown quantity to be
measured.
For the result to be meaningful, there are two basic requirements:-
I. The comparison standard is accurately defined and commonly accepted.
II. The calibration procedure and the instrument used for obtaining the
comparison must have provable reliability.
Input Measurement Output
System
Figure 1.1 An instrumentation/measurement system
Measurements are made with the help of instruments. An instrument can be

defined as a device or a system which is defined in such a way that it maintains
a functional relationship between a prescribed property of a substance and a
physical variable.
Example:
A mercury-in-glass thermometer maintains a linear relationship between

thermal expansion of mercury (prescribed property) and temperature (physical
variable).
Functions of a measurement system
Measurement systems are set up for one or more of the following functions:
Measurement of system parameters: present desired information about the

conditions of a system in the form of visual indicator, registering or recording,
monitoring or suitable transmission.
Example: thermometer, barometer, anemometer, water/gas/electricity meters
Control of a certain process or condition: the measurement system forms an

integral part if automatic control systems which in turn provides deliberate
guidance or manipulation to maintain a set-point or to change variables
according to a preset programme.
October, 2011 AAiT, Department of Electrical and Computer Engineering Andinet Negash
Output(controlled variable)
Energy/material Input
System or Process
Control signal
Measurement
Control elements
System
Comparator/
Error detector
Error Signal Output Signal
Reference Signal
Reference
Signal
Figure: 1.2. Typical block diagram representation of automatic control system
Example: A thermostat fitted in a domestic refrigerator is a control device for

maintaining the temperature in a specified range.
Verification of physical phenomena/scientific theories: experiments are

conducted to formulate certain empirical relations where adequate theory does
not exist. Furthermore, experiments motivated by the hope of developing new
theories, discovering new phenomena or checking the validity of certain
hypothesis require measurement systems.
Example: Coulomb postulated that the friction between two dry surfaces is
proportional to the normal reaction and is independent of the area of contact.
His hypothesis has since been verified experimentally and is known as
Coulomb‟s law of dry friction.
Others: whether exploring Mars, measuring the brain‟s electrical signals for
diagnostic purposes or setting up robots on an assembly line, measurement is
everywhere. It allows us to understand the phenomena we observe in our
environment by means of experimental deduction and verification and helps us
keep records of results of these observations.
Properties to be measured Application Field
Atmospheric pressure, hygrometry level Metrology

Battery fluid level Automobile
Beam resistance Civil engineering
Blood Glucose level Biology
ECG signals Health, medicine
Fiber optic flow Telecommunications
Flight speed, altitude Transportation
Presence detection Home automation
Radar detection, surveillance Military
Sugar level in a fruit Agriculture
Undersea pressure, depth Marine industry
Table: 1.1 example application areas of measurement
A generalized measurement system can be schematically represented as shown

in Fig. 1.3
Data storage &
playback
element
Observer
measurand Sensing
Signal Signal Signal Data
conditioning processing transmission presentation
element
element element element element
Figure: 1.3. Functional elements of a measurement system.
Although in a given measurement system on type of element may be missing or may occur
more than once, it is possible to identify the following functional elements.
Sensing element: this is an element which comes first in contact with the process and gives an
output which depends in some way on the variable to be measured.
Example:
Thermocouple where millivolt e.m.f. depends on temperature
Strain gauge where resistance depends on mechanical strain
Orifice plate where pressure drop depends on flow rate.
If there is more than one sensing element in a system, the element in contact with the process
is termed the primary sensing element, the others secondary sensing elements.
All sensors extract some energy from the measurand medium which implies that the
measurand is always disturbed by the measurement system. Nonetheless, we shall always
assume that a measurement system has an input of the true value of the variable being
measured and an output of the measured value.
Example: insertion of a cold thermometer into a hot liquid lowers the temperature of a liquid
and so altering the temperature being measured.
N.B. The term transducer is often used in relation to measurement systems. Transducers are
defined as elements that convert a change in some physical variable into a related change in
some other physical variable. It is generally used for an element that converts a change in
some physical variable into an electrical signal change. Thus sensors can be transducers.
However, a measurement system may use transducers, in addition to sensors, in other parts of
the system to convert signals from one form to another.
Signal conditioning element: This takes the output of the sensing element and converts it into
a form more suitable for further processing, usually a d.c. voltage, d.c. current or frequency
signal.
Example:
Deflection bridge which converts an impedance change into a voltage change
Amplifier which amplifies millivolts to volts
Oscillator which converts an impedance change into a variable frequency voltage.
Signal processing element: This takes the output of the conditioning element and converts it
into a form more suitable for presentation.
Example:
Analogue-to-digital converter (ADC) which converts a voltage into a digital form for
input to a computer
Computer which calculates the measured value of the variable from the incoming digital
data.
Typical calculations are:
Computation of total mass of product gas from flow rate and density data
Integration of chromatograph peaks to give the composition of a gas stream
Correction for sensing element non-linearity.
Signal transmission element: when the elements of the system are physically separated, it is
necessary to transmit the data from one stage to other. Note that this element is not a subject
of discussion in this course; enthusiastic students are encouraged to read on this subject
elsewhere.
Example: radio links or telemetry systems
Data presentation element: communicates the information about the measurand to the human
observer or presents it in an intelligible form. The presented data may be used by the system
for monitoring, controlling, or analyzing purposes.
Example:
Monitoring- VDU
Recording-magnetic tapes
Control and analysis-microprocessor, microcomputer
Exercise1: Fig. 1.4 represents a moving coil mechanism for current measurement. The basic
elements are shown in the figure. Identify the functional elements of the of measurement
system.
Figure: 1.4. Permanent magnet moving coil mechanism (PMMC)
Exercise: a bourdon tube pressure gage is shown in Figure 1.5. When pressure is applied to
the hollow oval-shaped bent tube, known as the Bourdon tube, it deforms the cross-section of
the tube as well as causing a relative motion, proportional to the applied pressure, of the free
1
Solutions to exercises are not presented in this handout; exercises are meant to be provided and discussed as
part of class room activities.
end of the tube with respect to its fixed end. Identify the functional elements of Bourdon tube
pressure gage.
Figure 1.5: Bourdon tube pressure gage.
Standards of Measurement:
A standard of measurement is defined as the physical representation of the unit of

measurement which could be an arbitrary material standard or a natural phenomenon that
includes physical and atomic constants.
International standards: are devices designed and constructed to the specifications of an

international form. They represent the units of measurements of various physical quantities to
the highest possible accuracy that is attainable by the use of advanced techniques of
production and measurement technology. These standards are maintained by the international
Bureau of Weights and Measures at Sèvres, France.
Primary standards: are devices maintained by standards organizations or national

laboratories in different parts of the world and are calibrated independently by absolute
measurements. One of the main functions of maintaining primary standards is to
calibrate/check and certify secondary reference standards.
Secondary standards: are the basic reference standards employed by industrial measurement
laboratories. These are freely available to the ordinary user of the instruments for checking
and calibration of working standards.
Working standards: are devices that are commercially available and are duly checked and
certified against either the primary or secondary standards. Working standards are very
widely used for calibrating general laboratory instruments, for carrying out comparison of
measurements or for checking the range of accuracy of industrial products.
Exercise: Describe the fundamental units of measurement and the standards defined to
maintain them.
Exercise: Based on the fundamental units, obtain the SI units for the following derived
quantities:
a) Resistance, ohm (Ω)

b) Capacitance, Farad (F)
c) Inductance, Henry (H)
Classification of instruments
Instruments can be classified according to their application, mode of operation, manner of

energy conversion, and nature of output signal and so on. Instruments commonly used in
practice may be broadly categorized as follows.
1. Deflection Vs. Null types

Deflection type instrument: the physical effect generated by the measuring quantity
produces an equivalent opposing effect in some part of the instrument which in turn is
closely related to some variable like displacement.
(a) (b)
Figure 1.6 (a) deflection (b) null type instruments
Null type instrument: provided with either a manually operated or automatic
balancing device that generates an equivalent opposing effect to nullify the physical
effect caused by the quantity to be measured.
Exercise: compare the two instruments based on:
i. Dynamic response
ii. Loading effect
2. Manually operated Vs. Automatic types
Any instrument that requires the service of human operator is a manual type of
instrument. The instrument becomes automatic if the manual operation is replaced by
an auxiliary device incorporated in the instrument. An automatic instrument is usually
preferred because the dynamic response of such an instrument is fast and also its
operational cost is considerably lower than that of the corresponding manually

operated instrument.
The manual operation can be dispensed with by incorporating an automatic self-
balancing feedback device known as instrument servo-mechanism.
3. Analog Vs. Digital types
Analog instruments are those that present the physical variables of interest in the form
of continuous or step less variations with respect to time. They generally cost less and
are easy to maintain.
Digital instruments are those in which the physical variables are represented by
digital quantities which are discrete and vary in steps. The main drawback of such
instruments is that they are unable to indicate the quantity which is a part of the step
value of the instrument. Advantages include on-line use of digital computers for data
processing and their noise immunity during transmission. In addition, several
techniques of coding have been developed for digital signals only. Therefore, in order
to use the error detection and error correction capabilities, it is necessary to convert
analog data into digital form.
4. Self-generating Vs. power-operated types
In self-generating (passive) instruments, the energy requirements of the instruments
are met entirely by the input signal.
Example: mercury-in-glass thermometer (temperature)
Tachogenerator (rpm)
Power-operated (active) instruments require some source of auxiliary power such as
compressed air, electricity, hydraulic supply, etc. for their operation.
Example: differential transformer (displacement, force …)
5. Contacting Vs. non-contacting types
A contacting type of instrument is one that is kept in the measuring medium itself. A
clinical thermometer is an example of such instruments.
Non-contacting (or proximity) type instruments, on the other hand, measure the
desired input even though they are not in close contact with the measuring medium.
Example: variable reluctance tachometer
6. Dump vs. Intelligent types
A dump (or conventional) instrument measures and displays the input variable with
µP or an on-line digital computer to provide assistance in noise reduction, automatic
calibration, drift correction, gain adjustments, etc. in addition, they are equipped with
diagnostic sub-routines with suitable alarm generation in case of any type of
malfunctioning.
1.2.Static Performance Characteristic

Introduction
The detailed specifications of the functional characteristics of any instrument are termed its
performance characteristics.
Instrument performance characteristics can be broadly classified as:
1. Static characteristics
2. Dynamic characteristics
In a number of situations, the desired input to the instrument may be constant or varying
slowly with respect to time. In these situations the dynamic characteristics are not important.
Therefore, the various static performance parameters like accuracy, precision, hysteresis,
impedance loading, etc. are usually good enough to give meaningful quantitative descriptions
of the instrument.
Furthermore, we also come across certain situations, where the desired input is not constant
but varies rapidly with time. In such cases the dynamic characteristics of the instruments
should also be known. These are generally represented by the input-output parameter
relationships that are governed by the relevant differential equations applicable in the given
situation.
Thus it may be noted that in general, the overall qualitative performance qualities of the
instruments are represented by both their static and dynamic performance characteristics.
Errors and uncertainties in performance characteristics
The various static performance parameters of instruments are obtained by performing certain
specified tests. This is accompanied by imposing constant values of „known‟ inputs and
observing the resulting output(s).
No measurement can be made with perfect accuracy and precision. Therefore, it is instructive
to know the various types of errors and uncertainties that are in general, associated with
measurement system.
Further it is also important to know how these errors are propagated. This is because if an
error is detected, then it can be eliminated or its effect can be accounted for in the form of
suitable correction. On the other hand if an error goes unrecognized it would make
experimental data unreliable.
Types of Errors
Error is defined as the difference between the measured and the true value (as per the
standard).
I. systematic/cumulative/ errors:
Such errors are those that tend to have the same magnitude and sign for a given set of
conditions. Because the algebraic sign is the same they tend to accumulate. Since
such errors alter the instrument reading by a fixed magnitude and with the same sign
from one reading to another, therefore, the error is also commonly termed as
instrument bias. These types of errors are caused due to:
Chapter 1 Introduction to Instrumentation (ECEG-4201) P a g e | 10
Instrument errors: inherent errors due to poor design/construction in the

instrumentation systems. Errors in the divisions of graduated scales, irregular spring
tension, etc. cause such errors. These can be avoided by applying suitable correction
or by calibrating the instrument against a standard.
Environmental errors: variation of conditions external to the measuring device. For
example, change in the ambient temperature causes errors due to expansion of the
measuring tape.
Loading errors: these represent the effect of measurement on the physical system
being tested. Introduction of a thermometer alters the thermal capacity of the system
and thereby the original state of the system which gives rise to loading error.
II. accidental or random errors:
These are caused due to random variations in the parameter or the system of
measurement. Such errors vary in magnitude and may be either positive or negative
on the basis of chance alone. Since these errors are in either direction, they tend to
compensate one another. Therefore, these errors are also called chance or
compensating type of errors. The presence of such errors is detected by the lack of
consistency in the measured values when the input is imposed repeatedly on the
instrument.
Example: line voltage fluctuations, vibrations of instrument support, etc.
III. miscellaneous type of gross errors:
These classes of errors represent those errors which are both systematic and random.
Typical cases include:
Personal or human errors: caused due to limitations in the human senses.
Errors due to faulty components: misalignment of moving parts, electrical leakage,
etc. may simultaneously cause systematic as well as random errors.
One needs to repeat measurement for a sufficiently large number of times by feeding
a „standard‟ signal. The difference between the mean value of the signal and the
standard signal gives the best estimate of the systematic error. The estimate of
uncertainty which represents the random errors is evaluated from the dispersion of the
data.
Since systematic errors and the systematic component of gross errors are
deterministic in nature, therefore, it is best to eliminate the effect of such errors by
introducing a suitable correction factor in the experimental results by a proper
calibration procedure.
Now, the remaining random errors and the random component of gross errors in
general, constitute the chief source of uncertainity in experiments. Even though the
magnitude or random type deviations as well as their algebraic sign vary from one
observation to another, yet they follow a definite mathematical law called the normal
or Gaussian distribution.
1 𝑥−𝑥 2
−
𝑝 𝑥 = 𝑒 2𝜍 2
2𝜋𝜍 2
Propagation of uncertainties in compound quantities:
To compute the overall uncertainties of different variables, we consider the following

equation in the most general form:
𝑦 = 𝑓 𝑥1 , 𝑥2 , 𝑥3 , ⋯ 𝑥𝑖 , ⋯ 𝑥𝑛
where y is a parameter that depends on the independent variables

𝑥1 , 𝑥2 , 𝑥3 , ⋯ 𝑥𝑖 , ⋯ 𝑥𝑛
Writing the equation in differential form we get,
𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦
𝑑𝑦 = 𝜕𝑥 𝑑𝑥1 + 𝜕𝑥 𝑑𝑥2 + 𝜕𝑥 𝑑𝑥3 + ⋯ + 𝜕𝑥 𝑑𝑥𝑖 + ⋯ + 𝜕𝑥 𝑑𝑥𝑛
1 2 3 𝑖 𝑛
If the quantities 𝑑𝑦, 𝑑𝑥1 , 𝑑𝑥2 , 𝑑𝑥3 ⋯ , 𝑑𝑥𝑖 , ⋯ , 𝑑𝑥𝑛 are considered to be the uncertainties
𝑈𝑦 , 𝑈𝑥 1 , 𝑈𝑥 2 , 𝑈𝑥 3 ⋯ , 𝑈𝑥 𝑖 , ⋯ , 𝑈𝑥 𝑛 in the quantities 𝑦, 𝑥1 , 𝑥2 , 𝑥3 , ⋯ 𝑥𝑖 , ⋯ 𝑥𝑛 respectively,
then we obtain:
𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦
𝑈𝑦 = 𝑈𝑥 1 + 𝑈𝑥 2 + 𝑈𝑥 3 + ⋯ + 𝑈𝑥 𝑖 + ⋯ + 𝑈
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝜕𝑥𝑖 𝜕𝑥𝑛 𝑥 𝑛
Now, the maximum value of uncertainty in 𝑦, i.e. 𝑈𝑦 𝑚𝑎𝑥 would be when all the
uncertainties happen to have the same sign and this would be the worst possible case:
𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦
𝑈𝑦 𝑚𝑎𝑥 = 𝑈𝑥 1 + 𝑈𝑥 2 + 𝑈𝑥 3 + ⋯ + 𝑈𝑥 𝑖 + ⋯ + 𝑈
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝜕𝑥𝑖 𝜕𝑥𝑛 𝑥 𝑛
However, the probability of such an occurrence is very small. Therefore, the more
realistic way is to square both sides of the above equation which would ensure equal
weightage to positive as well as negative value of uncertainties. Therefore, squaring:
2 2 2 2
𝜕𝑦 2 𝜕𝑦 2 𝜕𝑦 2 𝜕𝑦 2
𝑈2 𝑦 = 𝑈𝑥 1 + 𝑈𝑥 1 + 𝑈𝑥 1 +⋯+ 𝑈𝑥 1 +⋯
𝜕𝑥1 𝜕𝑥1 𝜕𝑥1 𝜕𝑥1
𝜕𝑦 2 2
+ 𝑈𝑥 1
𝜕𝑥1
𝜕𝑦 𝜕𝑦 𝜕𝑦 𝜕𝑦
+ 𝑈𝑥 1 𝑈𝑥 2 + ⋯ + 𝑈𝑥 𝑖 𝑈𝑥 𝑗 + ⋯
𝜕𝑥1 𝜕𝑥2 𝜕𝑥𝑖 𝜕𝑥𝑗
Because of the random nature of the uncertainties in the quantities given by

𝑈𝑥 1 , 𝑈𝑥 2 , 𝑈𝑥 3 ⋯ , 𝑈𝑥 𝑖 , ⋯ , 𝑈𝑥 𝑛 , there is equal probability of these being either positive or
negative. Therefore, the term in the second bracket would tend to be a small quantity
which would be neglected. Therefore, we now have:
2 2 2 2
𝜕𝑦 2 𝜕𝑦 2 𝜕𝑦 2 𝜕𝑦 2
𝑈𝑦 = ± 𝑈𝑥 1 + 𝑈𝑥 1 + 𝑈𝑥 1 + ⋯+ 𝑈𝑥 1 +⋯
𝜕𝑥1 𝜕𝑥1 𝜕𝑥1 𝜕𝑥1
1
2 2
𝜕𝑦 2
+ 𝑈𝑥 1
𝜕𝑥1
Exercise:
It is required to determine the mass of the body, composed of cylindrical and hemispherical
portions, shown in the following figure, Fig. 1.7. The various dimensions and densities are as
estimated as follows:
𝐿 = 10.0 ± 0.1 𝑐𝑚
𝑅 = 4.00 ± 0.05 𝑐𝑚
𝜌1 = 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑐𝑦𝑙𝑖𝑛𝑑𝑒𝑟𝑖𝑐𝑎𝑙 𝑝𝑜𝑟𝑡𝑖𝑜𝑛 = 3.5 ± 0.1 𝑔/𝑐𝑚3
𝜌2 = 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑕𝑒𝑚𝑖𝑠𝑝𝑕𝑒𝑟𝑖𝑐𝑎𝑙 𝑝𝑖𝑠𝑡𝑜𝑛 = 2.5 ± 0.05 𝑔/𝑐𝑚3
r1 R
2R
r2
Figure 1.7: Example body
The total mass the body, 𝑀𝑇 , is given by:
𝑀𝑇 = 𝑀𝑐𝑦𝑙𝑖𝑛𝑑 𝑒𝑟 + 𝑀𝑕𝑒𝑚𝑖𝑠𝑝 𝑕𝑒𝑟𝑒
1 4 3
= 𝜋𝑅 2 𝑙𝜌1 + 𝜋𝑅 𝜌2
2 3
Using the mean value 𝑅, 𝑙, 𝜌1 , and 𝜌2
2
1 4 3
𝑀𝑇 𝑎𝑣 =𝜋 4 10 3.5 + 𝜋4 2.5
2 3
= 1759.29 + 335.10 =2094.33g
The overall uncertainity in 𝑀𝑇 can be written as:

2 2 2 2
2 𝜕𝑀𝑇 2 𝜕𝑀𝑇 2 𝜕𝑀𝑇 2 𝜕𝑀𝑇
𝑈𝑀𝑇 = 𝑈𝑅 + 𝑈𝑙 + 𝑈𝜌 1 + 𝑈𝜌 2 2
𝜕𝑅 𝜕𝑙 𝜕𝜌1 𝜕𝜌2
𝜕𝑀𝑇
= 2𝜋𝑅𝑙𝜌1 + 2𝜋𝑅𝑙𝜌2
𝜕𝑅
𝜕𝑀𝑇
= 𝜋𝑅 2 𝜌1
𝜕𝑙
𝜕𝑀𝑇
= 𝜋𝑅 2 𝑙
𝜕𝜌1
𝜕𝑀𝑇 2 3
= 𝜋𝑅
𝜕𝜌2 3
2 2 2 2 2 2
𝑈𝑀𝑇 = 2𝜋 4 10 3.5 + 4 2.5 0.05 + 𝜋 4 3.5 0.1
2
2 2 2
2 3 2
+ 𝜋 4 10 0.1 + 𝜋 4 0.05 = 3080.91𝑔2
3
𝑈𝑀𝑇 = 55.51𝑔
Thus, the total mass of the composite body can be expressed as:
𝑀𝑇 = 2094.91 ± 55.51𝑔
Alternatively, the uncertainity in the total mass is also expressed in percentage:

55.51
% of uncertainty in total mass= 2094.39 𝑥100 = ±2.65%
Static Performance Parameters
Accuracy:
It is the closness of the instrument output to the true value of the measured quantity (as per
standards).
The accuracy of the instrument mainly depends on the inherent limitations of the instrument as
well as on the shortcomings in the measurement process.
The accuracy of the instrument can be specified as:
1. Percentage of the true value

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 − 𝑡𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒
= 𝑥100
𝑡𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒
2. percentage of the full-scale deflection

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 − 𝑡𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒
= 𝑥100
𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑠𝑐𝑎𝑙𝑒 𝑣𝑎𝑙𝑢𝑒
Precision:
The ability of the instrument to reproduce a certain set of readings within a given accuracy,
precision of an instrument is in fact, dependent on the repeatability. The term repeatability can be
defined as the ability of the instrument to reproduce a group of measurements of the measured
quantity, made by the same observer, using the same instrument, under the same conditions.
Accuracy and precision involved in a measurement are dependent on the systematic and random
errors, respectively. The former is determined by proper calibration of the instrument and the
later by statistical analysis.
Figure: 1.8. Accuracy versus precision in a shooting experiment
Resolution (or discrimination):
The smallest increment in the measured value that can be determined with certainity by the
instrument.
Range:
The input range of an element is specified by the minimum and maximum values of the input, I,
i.e. 𝐼𝑚𝑖𝑛 to 𝐼𝑚𝑎𝑥 .
Likewise the output range is 𝑂𝑚𝑖𝑛 to 𝑂𝑚𝑎𝑥 .
Span:
The maximum variation in input or output. i.e.
Input span is 𝐼𝑚𝑎𝑥 − 𝐼𝑚𝑖𝑛
Output span is 𝑂𝑚𝑎𝑥 − 𝑂𝑚𝑖𝑛 .
Static sensitivity:
The ratio of the magnitude of response (output signal) to the magnitude of the quantity being
measured (input signal), i.e.
𝑐𝑕𝑎𝑛𝑔𝑒 𝑜𝑓 𝑜𝑢𝑡𝑝𝑢𝑡 𝑠𝑖𝑔𝑛𝑎𝑙 ∆𝑂

𝐾= =
𝑐𝑕𝑛𝑎𝑔𝑒 𝑜𝑓 𝑖𝑛𝑝𝑢𝑡 𝑠𝑖𝑔𝑛𝑎𝑙 ∆𝐼
Nonlinearity:
An element is said to be linear if corresponding value of 𝐼 𝑎𝑛𝑑 𝑂 lie on a straight line.
𝑂 = 𝐾𝐼 + 𝑎
Where 𝑎 (intercept) = 𝑂𝑚𝑖𝑛 − 𝐾𝐼𝑚𝑖𝑛
However, often this ideal relationship is not obeyed and a result we have:
𝑂 𝐼 = 𝐾𝐼 + 𝑎 + 𝑁(𝐼)
OMAX
N(I)
I
IMIN,OMIN IMAX
Figure: 1.9 curves showing linearity and nonlinearity
Nonlinearity is often quantified in terms of the maximum nonlinearity 𝑁 expressed as a

percentage of full-scale deflection (f.s.d.) i.e. as a percentage of span. Thus:
𝑁
𝑀𝑎𝑥. 𝑛𝑜𝑛𝑙𝑖𝑛𝑒𝑎𝑟𝑖𝑡𝑦 𝑎𝑠 𝑎 𝑝𝑒𝑟𝑐𝑒𝑛𝑡𝑎𝑔𝑒 𝑜𝑓 𝑓. 𝑠. 𝑑. = 𝑥100%
𝑂𝑚𝑎𝑥 − 𝑂𝑚𝑖𝑛
In general, the output 𝑂 depends not only on the signal 𝐼 but on environmental inputs such as
ambient temperature, atmospheric pressure, relative humidity, supply voltage, etc. there are two
aspects of environmental inputs. A modifying input causes the linear sensitivity of the element to
change from 𝐾 𝑡𝑜 𝐾 + 𝐾𝑀 𝐼𝑀 where 𝐼𝑀 is the change in a modifying input from the „standard‟
value. An interfering input causes the intercept or zero bias of the element to change from
𝑎 𝑡𝑜 𝑎 + 𝐾𝐼 𝐼𝐼 where 𝐼𝐼 is the change in the interfering input from the „standard‟ value.
M
I M
+K
=K
=0
pe
M
I
o
Sl
I
Figure: 1.10. Modifying environmental input
0
I I=
II I
+K
=a
b ias
r o- a
Ze s=
bia I
r o-
Ze
Figure: 1.11. Interfering environmental input
Thus, we have
𝑂 𝐼 = 𝐾 + 𝐾𝑀 𝐼𝑀 𝐼 + 𝑎 + 𝐾𝐼 𝐼𝐼 + 𝑁 𝐼
Exercise: Draw the block diagram representation of the generalized model representing the static
characteristics of an element.
Solution: block diagram representation is given below (Figure 1.12)
Figure: 1.12.
Exercise:
A potentiometer is used for displacement measurement. It consists of a wire-wound resistance
provided with a sliding contact. Assuming that the resistance is distributed linearly along its
length , characterize the change in the supply voltage as modifying/interfering input.
Vs + Vs
x Vout
Hysteresis
It is defined as the magnitude of error caused in the output for a given value of input, when this
value is approached from opposite directions, i.e. from ascending order and then descending
order. This is caused by backslash, elastic deformations, magnetic characterstics, but is mainly
caused due to frictional effects.
O2
ut
p
In
g
si n
ea
cr
de
Omean
ut
p
In
g
in
as
re
c
In
O1
I
Threshold
Ii
Dead band
O
ut
O2
p
In
g
si n
ea
cr
de
Omean
put
In
g
in
as
c re
In
O1
I
Ii
Hysteresis effects are best eliminated by taking the observation both for ascending and
descending values of input and then taking the arithmetic mean.
Dead band:
The largest change of the measurand to which the instrument does not respond.
Backlash:
The maximum distance or angle through which any part of the mechanical system may be moved
in one direction without causing motion of the next part.
Drift:
The variation of the input for a given input caused due to change in the sensitivity of the
instrument due to certain interfering inputs like temperature change, component instabilities, etc.
Identification of Static Characteristics:
The static characteristics of an element can be found experimentally by measuring corresponding

values of the input 𝐼, the output 𝑂 and the environmental inputs 𝐼𝑀 , 𝐼𝐼 when 𝐼 is either at a
constant or varying slowly. This type of experiment is referred also as calibration.
IM II
Standard Instrument Standard Instrument
Standard Instrument
Elements or system to be
calibrated
Standard Instrument Standard Instrument
1. 𝑶 versus 𝑰 with 𝑰𝑴 = 𝑰𝑰 = 𝟎
The experiment should be held under “standard” environmental conditions so that

𝐼𝑀 = 𝐼𝐼 = 0, otherwise all possible environmental inputs should be measured. 𝐼 should be
increased slowly from 𝐼𝑀𝑖𝑛 𝑡𝑜 𝐼𝑀𝑎𝑥 and corresponding values of 𝐼 𝑎𝑛𝑑 𝑂 recorded. A
further reading should be taken with 𝐼 decreasing slowly from 𝐼𝑀𝑎𝑥 𝑡𝑜 𝐼𝑀𝑖𝑛 .
One could fit a polynomial, 𝑂 𝐼 = 𝑎0 + 𝑎1 + 𝑎2 𝐼 2 + ⋯ + 𝑎𝑞 𝐼 𝑞 to a set of data points

using least square criteria.
If 𝑑𝑖 is the deviation of the polynomial value 𝑂(𝐼𝑖 ) from the data value 𝑂𝑖 , then 𝑑𝑖 =
𝑂 𝐼𝑖 − 𝑂𝑖 . We find a set of coefficients 𝑎0 , 𝑎1 , 𝑎2 , ⋯ , 𝑎𝑞 such that the sum of the
squares of the deviations 𝑑1 2 + 𝑑2 2 + 𝑑3 2 + ⋯ + 𝑑𝑞 2 is a minimum. This involves
solving a set of linear equations. In order to detect any hysteresis separate curve fitting
should be performed on the „up‟ and „down‟ data sets.
𝐻 𝐼 = 𝑂(𝐼)𝑢𝑝 − 𝑂(𝐼)𝑑𝑜𝑤𝑛
If the hysteresis is not significant, the two sets of data can be combined to give a single
polynomial𝑂(𝐼). The slope 𝐾 and zero-bias 𝑎 of the ideal straight line joining
𝐼𝑀𝑖𝑛 , 𝑂𝑀𝑖𝑛 and 𝐼𝑀𝑎𝑥 , 𝑂𝑀𝑎𝑥 can be determined. The non-linear function 𝑁(𝐼) can then
be found as:
𝑁 𝐼 = 𝑂 𝐼 − (𝐾𝐼 + 𝑎)
2. 𝑶 𝒗𝒆𝒔𝒓𝒖𝒔 𝑰𝑴 , 𝑰𝑰 𝒂𝒕 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕 𝑰
We first need to find when environment inputs are interfering, which affect the zero-bias
𝑎. The input I is held constant at 𝐼 = 𝐼𝑀𝑖𝑛 , and one environmental input is changed by a
known amount, the rest being kept at standard values. If there is a resulting change ΔO in
O, then the input II is interfering and the value of the corresponding coefficient 𝐾𝐼 is
𝛥𝑂
given by 𝐾𝐼 = . If there is no change in O, then the input is not interfering; the
𝛥𝐼 𝐼
process is repeated until all interfering inputs are identified and the corresponding 𝐾𝐼
values found.
We now need to identify modifying inputs, i.e. those which affect the sensitivity of the
1
element. The input I is held constant at the mid-range value 2 (𝐼𝑀𝑖𝑛 + 𝐼𝑀𝑎𝑥 ) and each
environmental input is varied in turn by a known amount. If a change in input produces a
change ΔO in O and is not an interfering input, then it must be a modifying input IM and
the value of the corresponding coefficient KM is given by
1 ∆𝑂 2 ∆𝑂
𝐾𝑀 = =
𝐼 ∆𝐼𝑀 𝐼𝑀𝑖𝑛 + 𝐼𝑀𝑎𝑥 ∆𝐼𝑀
Suppose a change in input produces a change ΔO in O and it has already been identified
as an interfering input with a known value of KI. Then we must calculate a non-zero
value of KM before we can be sure that the input is also modifying. Since
𝐼𝑀𝑖𝑛 + 𝐼𝑀𝑎𝑥
∆𝑂 = 𝐾𝐼 ∆𝐼𝐼,𝑀 + 𝐾𝑀 ∆𝐼𝐼,𝑀
2
Then,
2 ∆𝑂
𝐾𝑀 = − 𝐾𝐼
𝐼𝑀𝑖𝑛 + 𝐼𝑀𝑎𝑥 ∆𝐼𝐼,𝑀
Method of Least Squares:

Suppose that two variables x and y are measured over a range of values. We want to
obtain a simple analytical expression for y as a function of x. the simplest type of
function is a linear one and hence we try to find the best linear function connecting y and
x.
Suppose the linear function is 𝑦 = 𝑚𝑥 + 𝑏. Now we want to minimize the quantity,
𝑛
2
𝑆= 𝑦𝑖 − 𝑎𝑥𝑖 + 𝑏
𝑖=1
Taking the derivative of 𝑆 with respect to 𝑎 and 𝑏 and setting them equal to zero, we find:
𝑛𝑏 + 𝑎 𝑥𝑖 = 𝑦𝑖
𝑏 𝑥𝑖 + 𝑎 𝑥𝑖 2 = 𝑥𝑖 𝑦𝑖
Solving we get:
𝑛 𝑥𝑖 𝑦𝑖 − 𝑥𝑖 𝑦𝑖
𝑎= 2 2
𝑛 𝑥𝑖 − 𝑥𝑖
𝑦𝑖 𝑥𝑖 2 − 𝑥𝑖 𝑦𝑖 𝑥𝑖
𝑏=
𝑛 𝑥𝑖 2 − 𝑥𝑖 2
Exercise:
Obtain the best linear relation in accordance with a least square analysis
x 0.9 2.3 3.3 4.5 5.7 6.7

y 1.1 1.6 2.6 3.2 4.0 5.0
Loading effects:
The ideal situation in a measurement system is that when an element used for any
purpose is introduced into the system, the original signal should remain undistorted.
However, under practical conditions the introduction of any element in a system results in
the extraction of energy from the system thereby distorting the original signal.
This distortion may take the form of attenuation (reduction in magnitude), waveform
distortion, phase shift and may be all of these.
The incapability of the system to faithfully measure, record, or control the input signal
(measurand) in undistorted form is called the loading effect.
1. Loading effect due to a shunt connected instrument

Consider a network represented by its Thevenin‟s equivalent with open circuit
voltage 𝐸𝑜 and series impedance 𝑍𝑜 .
A
Zo
Eo Load (
Eo Instrument) ZL
When the load is connected across terminals 𝐴 𝑎𝑛𝑑 𝐵, the output voltage 𝐸𝐿 would
be:
𝐸𝑜
𝐸𝐿 =
𝑍
1 + 𝑍𝑜
𝐿
Thus the voltage which is measured is modified both in phase and magnitude. In
order that the original signal 𝐸𝑜 should remain undistorted the value of the input
impedance of the instrument, 𝑍𝐿 , should be infinite (or the value of output impedance
of the source, 𝑍𝑜 , should be equal to zero which is not attained in practice).
2. Loading effect due to a series connected instrument
Consider a network represented by a voltage source having a voltage 𝐸𝑜 and an
output impedance 𝑍𝑜
A
Zo
Io
Eo Load (
Eo Instrument) ZL
The value of current flowing between terminals 𝐴 𝑎𝑛𝑑 𝐵 under ideal conditions is 𝐼𝑂 .
It is the current that flows when the terminals 𝐴 𝑎𝑛𝑑 𝐵 are shorted.
𝐸𝑜
𝐼𝑜 =
𝑍𝑜
When an ammeter is placed between the terminals, the measured value of current will
be
𝐸𝑜 𝐼𝑜 𝑍𝑜 𝐼𝑜
𝐼𝐿 = = =
𝑍𝑜 + 𝑍𝐿 𝑍𝑜 + 𝑍𝐿 1 + 𝑍𝐿
𝑍𝑜
In order that the measured value of current, be equal to the actual value of the current,
𝐼𝑜 , the value of 𝑍𝑜 ≫ 𝑍𝐿 . This means that the input impedance of the ammeter should
be very small as compared to the output impedance of the source.
Exercise:
A 50𝑣 range voltmeter is connected across the terminals 𝐴 𝑎𝑛𝑑 𝐵 of the circuit
shown below. Find the accuracy and the loading error. The voltmeter has a resistance
of 1000𝑘𝛺.
200k
A
100v 200k
1.3.Dynamic Performance Characteristic

Invariably measurement systems are subjected to inputs which are not static but are dynamic
in nature. The input varies with time, and therefore, so does the output. The behavior of the
system under such conditions is described by its dynamic response.
The response of a measurement system subjected to a time varying input can be divided into
two parts, the transient response and the steady state response. If 𝑐(𝑡) is the total response of
the system, then in general,
𝑐 𝑡 = 𝑐𝑡 𝑡 + 𝑐𝑠𝑠 (𝑡)
where 𝑐𝑡 𝑡 𝑎𝑛𝑑 𝑐𝑠𝑠 𝑡 are respectively the transient and steady state responses.
In many applications, the transient response of the system i.e. the way the system settles
down to its final steady state conditions is more important than the steady state response.
The dynamic behavior of measurement systems is studied in two domains 𝑣𝑖𝑧.:
i. Time domain
In most measurement systems, time is used as an independent variable and it is of

interest to evaluate the time domain response of the system. In time domain analysis,
the dynamic response of the system to different types of inputs, which are a function
of time, is analyzed at different intervals of time after the application of input signal.
For the purpose of design and analysis, it is necessary to assume some basic types of
input signals which can be easily defined mathematically, so that the performance of
the system can be analyzed with this standard signals.
The following standard test signals (inputs) are generally used:
Step input: a unit step input is defines as

1: 𝑡 ≥ 0
𝑢 𝑡 =
0: 𝑡 < 0
In general a step input of magnitude 𝐴 wherein the magnitude changes from 0 𝑡𝑜 𝐴 in
zero time, is mathematically represented as:
𝐴: 𝑡 ≥ 0
𝑟 𝑡 = 𝐴𝑢 𝑡 =
0: 𝑡 < 0
whose Laplace transform is,
∞
1
𝑈 𝑠 = 𝐴𝑒 −𝑠𝑡 𝑑𝑡 =
0 𝑠
Ramp input: a unit ramp input is defined as:

𝑡: 𝑡 ≥ 0
𝑟 𝑡 = = 𝑡𝑢(𝑡)
0: 𝑡 < 0
A ramp input signal of magnitude 𝐴 can be expressed as:
𝐴𝑡: 𝑡 ≥ 0
𝑟 𝑡 = = 𝐴𝑡𝑢(𝑡)
0: 𝑡 < 0
The Laplace transform will be:
∞
𝐴
𝑅 𝑠 = 𝐴𝑡𝑒 −𝑠𝑡 𝑑𝑡 =
0 𝑠2
Impulse input: or a delta (δ) function is defined as:
𝛿 𝑡 = 0; 𝑡 ≠ 0
𝜖
𝛿 𝑡 𝑑 𝑡 = 1, 𝑤𝑕𝑒𝑟𝑒 𝜖 → 0
−𝜖
The Laplace transform of a unit impulse function is 1.Note that a perfect impulse
function is not practically realizable.
ii. Frequency domain
The frequency domain analysis of a system pertains to the steady state response of the
system to a sinusoidal input. In this analysis, the system is subjected to a sinusoidal
input and the system response is studied with frequency as the independent variable.
The sinusoid is a unique input signal, and the resulting output signal for a linear
system is sinusoidal in the steady state. However, the output signal differs from the
input waveform in amplitude and phase.
The importance of frequency domain analysis is evident in such cases of a system
subjected to a complex time varying waveforms. According to Fourier analysis,
complex waveforms can be thought of as being composed of sinusoidal signals of
different frequencies.
When a system is subjected to a sinusoidal input,
𝑟 𝑡 = 𝐴𝑖 sin 𝜔𝑡
The output is,
𝑐 𝑡 = 𝐴𝑜 sin(𝜔𝑡 + 𝜑)
where 𝐴𝑖 , 𝐴𝑜 are respectively the amplitudes of input and output and 𝜑 is the phase
difference between the input and the output.
Mathematical models of measurement systems:
A measurement system is a physical system. Therefore, before one can analyze the
system, one must be able to determine the mathematical model of the system.
The input-output relations which define the dynamics of an instrument can be written
in a differential equation. Consider a linear time-invariant system which is defined by
an 𝑛𝑡𝑕 order differential equation of the form
𝑑𝑛 𝑐 𝑡 𝑑 𝑛−1 𝑐 𝑡 𝑑𝑐 𝑡
𝑎𝑛 + 𝑎 𝑛−1 + ⋯ + 𝑎 1 + 𝑎0 𝑐 𝑡
𝑑𝑡 𝑛 𝑑𝑡 𝑛 −1 𝑑𝑡
𝑑 𝑚 𝑟(𝑡) 𝑑 𝑚 −1 𝑟(𝑡) 𝑑𝑟(𝑡)
= 𝑏𝑚 + 𝑏𝑚 −1 + ⋯ + 𝑏1 + 𝑏0 𝑟 𝑡
𝑑𝑡 𝑚 𝑑𝑡 𝑚 −1 𝑑𝑡
where, 𝑐 𝑡 𝑎𝑛𝑑 𝑟(𝑡) are the output and input variables, the 𝑎𝑖 ′𝑠, 𝑏𝑖 ′𝑠 are
combimations of system parameters with 𝑛 ≥ 𝑚.
An easy and convenient way of describing of linear systems is the use of transfer
function. The transfer function of a linear time invariant system is defined as the ratio
of the Laplace transform of the output to the Laplace transform of the input variable
with all initial conditions equal to zero. Hence,
𝑎𝑛 𝑠 𝑛 + 𝑎𝑛−1 𝑠 𝑛−1 + ⋯ + 𝑎1 𝑠 + 𝑎0 𝐶 𝑠
= 𝑏𝑚 𝑠 𝑚 + 𝑏𝑚 −1 𝑠 𝑚 −1 + ⋯ + 𝑏1 𝑠 + 𝑏0 𝑅(𝑠)
Therefore, the transfer function of the system is,
𝐶(𝑠) 𝑏𝑚 𝑠 𝑚 + 𝑏𝑚 −1 𝑠 𝑚 −1 + ⋯ + 𝑏1 𝑠 + 𝑏0
𝐺 𝑠 = =
𝑅(𝑠) 𝑎𝑛 𝑠 𝑛 + 𝑎𝑛−1 𝑠 𝑛−1 + ⋯ + 𝑎1 𝑠 + 𝑎0
𝑛≥𝑚
When the transfer function of a physical system is determined; the system can be
represented by a block diagram. Thus,
R(S) C(S)
G(S)
Transfer Function
The steady state response of a system to a sinusoidal input is obtained by replacing

𝑠 𝑤𝑖𝑡𝑕 𝑗𝜔 in the transfer function of the system. Thus,
𝐶(𝑗𝜔) 𝑏𝑚 𝑗𝜔 𝑚 + 𝑏𝑚 −1 𝑗𝜔 𝑚 −1 + ⋯ + 𝑏1 𝑗𝜔 + 𝑏0
𝐺 𝑗𝜔 = =
𝑅(𝑗𝜔) 𝑎𝑛 𝑗𝜔 𝑛 + 𝑎𝑛−1 𝑗𝜔 𝑛−1 + ⋯ + 𝑎1 𝑗𝜔 + 𝑎0
Time response of a 0, 1st and 2nd order system
I. Zero order systems:
A zero order instrument is described by:
𝑎0 𝑐 𝑡 = 𝑏0 𝑟 𝑡
We can rewrite it as:
𝑏0
𝑐 𝑡 = 𝑟 𝑡 = 𝐾𝑟 𝑡 ,
𝑎0
𝑏
where 𝐾 = 𝑎 0 is the static sensitivity or steady state gain.
0
For such kinds of instruments no matter how the input 𝑟(𝑡) varies with time, the
output follows it perfectly and faithfully.
Example: the linear potentiometer discussed already could be shown to have a
transfer function of:
𝑥
𝑉𝑜𝑢𝑡 = 𝑉𝑠
𝑙
𝑉𝑠
and hence the static sensitivity is 𝐾 = 𝐿
II. First order systems:

A first order instrument is represented by the dynamic relation between the input and
the output of the form:
𝑑𝑐 𝑡
𝑎1 + 𝑎0 𝑐 𝑡 = 𝑏0 𝑟 𝑡
𝑑𝑡
However, this relation can be written with two rather than three coefficients as
follows:
𝑑𝑐 𝑡
𝜏 + 𝑐 𝑡 = 𝑘𝑟 𝑡
𝑑𝑡
Here, 𝑘 = 𝑏0 /𝑎0 and 𝜏 = 𝑎1 /𝑎0 . Where 𝑘 is the static sensitivity and 𝜏 is the time
constant. Taking the Laplace transform of both sides, we obtain:
𝜏𝑠 + 1 𝐶 𝑠 = 𝑘𝑅(𝑠)
Therefore, the transfer function will be:
𝐶 𝑠 𝑘
𝐺 𝑠 = =
𝑅(𝑠) 1 + 𝜏𝑠
Examples:
1. Electrical system
Consider a series RC circuit wherein a step input voltage 𝐸𝑖 is applied when the
switch is closed.
R
i
Ei
C eo
Assuming all initial conditions to be zero, we have,

1
𝐸𝑖 𝑢(𝑡) = 𝑅𝑖(𝑡) + 𝑖𝑑𝑡
𝐶
𝐸𝑖 1
= 𝑅𝐼 𝑠 + 𝐼(𝑠)
𝑠 𝐶𝑠
𝐸𝑖 1
𝐼 𝑠 =
𝑅 𝑠 + 1/𝑅𝐶
Taking the inverse Laplace transforms,
𝐸𝑖 𝑡
𝑖 𝑡 = 𝑒 −𝑅𝐶
𝑅
2. Liquid level system:

Assuming laminar flow and linear operations, the inflow rate of the liquid is equal
to the sum of the volume of liquid stored and the outflow of liquid.
Q+qi
Control valve
Load valve
H+h
Q+qo
Therefore,
𝑑𝑕
𝐶 = 𝑞𝑖 − 𝑞𝑜
𝑑𝑡
Where 𝑞𝑖 and 𝑞𝑜 are, respectively, change in the inflow and outflow rate from the
steady state value and 𝐶 is the capacitance of the tank defined as:
𝑐𝑕𝑎𝑛𝑔𝑒 𝑖𝑛 𝑞𝑢𝑎𝑛𝑡𝑖𝑡𝑦 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑠𝑡𝑜𝑟𝑒𝑑
𝐶=
𝑐𝑕𝑎𝑛𝑔𝑒 𝑖𝑛 𝑕𝑒𝑎𝑑
𝑕
Since for laminar flow, the resistance 𝑅 = 𝑞 , we get:
𝑜
𝑑𝑕
𝑅𝐶 + 𝑕 = 𝑅𝑞𝑖
𝑑𝑡
Taking Laplace transform, we see that
𝑅𝑄𝑖 (𝑠)
𝐻 𝑠 = ; 𝜏 = 𝑅𝐶
1 + 𝜏𝑠
3. Revise your previous lesson in mathematical modeling to extend the discussion to
Mechanical, Thermal and Pneumatic systems. In the process observe analogous
quantities between these systems.
Step response:
Let a unit step input be applied to a first order system, then from previous discussions
in various physical systems:
1 1 𝜏
𝐶 𝑠 =𝐺 𝑠 𝑅 𝑠 = = −
𝑠(1 + 𝜏𝑠) 𝑠 1 + 𝜏𝑠
Taking the inverse Laplace transform,

𝑡
𝑐 𝑡 = 1 − 𝑒 −𝜏
Plot and the response to see that the response rises exponentially from zero value to
the final value of unity. In fact, the initial slope of the curve is given by:
𝑑𝑐 1 −𝑡 1
= 𝑒 𝜏 =
𝑑𝑡 𝑡=0 𝜏 𝑡=0 𝜏
Thus if the initial rate of change is maintained, the system will reach its final value in
time 𝜏. The output at 𝑡 = 𝜏 is,
𝜏
𝑐 𝑡 = 1 − 𝑒 −𝜏 = 0.632
Thus for a rising exponential function the time constant 𝜏 is the given by the time to
reach 63.2% of its final value.
The dynamic error is:

𝑡
𝑒 𝑡 = 𝑟 𝑡 − 𝑐 𝑡 = 𝑒 −𝜏
The steady state error is:
𝑒𝑠𝑠 = lim 𝑒 𝑡 = 0
𝑡→∞
Thus a first order system tracks step input with zero static error. Furthermore, it could
be noted that assuming a tolerance band of 5%, the system will be able to reach and
stay within the band in just 𝑡𝑠 = 3𝜏. This time, termed settling time, is an important
dynamic characteristic to describe the speed of response of any system. Note also that
the time constant is indicative of how fast the system tends to reach the final value.
Ramp response:
If a system is subjected to a ramp input 𝑟 𝑡 = 𝐴𝑡, then the output is
1 1 𝜏 𝜏2
𝐶 𝑠 =𝐺 𝑠 𝑅 𝑠 = 2 = 2− +
𝑠 1 + 𝜏𝑠 𝑠 𝑠 1+𝑠
Taking the inverse Laplace transform, we get:

𝑡
𝑐 𝑡 = 𝐴𝑡 − 𝐴𝜏 1 − 𝑒 −𝜏
The dynamic and steady state errors are respectively:

𝑡
𝑒 𝑡 = 𝑟 𝑡 − 𝑐 𝑡 = 𝐴𝜏 1 − 𝑒 −𝜏
𝑒𝑠𝑠 = lim 𝑒 𝑡 = 𝐴𝜏
𝑡→∞
Thus the first order system will track the unit ramp input with a steady state error of
𝐴𝜏.
Impulse response:
Let the system be subjected to an input 𝑟 𝑡 = 𝐴𝛿(𝑡), then
𝐴
𝐶 𝑠 =𝐺 𝑠 𝑅 𝑠 =
1 + 𝜏𝑠
Taking inverse Laplace transform, we have,
𝐴 −𝑡
𝑐 𝑡 = 𝑒 𝜏
𝜏
III. Second order systems:

A second order system may be described by the following differential equation:
𝑑2 𝑐 𝑡 𝑑𝑐 𝑡
𝑎2 + 𝑎1 + 𝑎0 𝑐 𝑡 = 𝑏0 𝑟 𝑡
𝑑𝑡 2 𝑑𝑡
Which gives,
1 𝑑2 𝑐 𝑡 2𝜉 𝑑𝑐 𝑡
2 2
+ + 𝑐 𝑡 = 𝐾𝑟 𝑡
𝜔 𝑛 𝑑𝑡 𝜔𝑛 𝑑𝑡
𝑏 𝑎0 𝑎1
Where, 𝐾 = 𝑎0 =static senstivity, 𝜔𝑛 = =natural frequency, 𝜉 = .
0 𝑎2 2 𝑎0 𝑎2
Taking the Laplace transform, we obtain the transfer function:
𝐾𝜔2 𝑛
𝐺 𝑠 = 2
𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔 2 𝑛
Exercise: Write a differential equation for the mass-spring-damper system and obtain
𝜉 and 𝜔𝑛
fk
M fD
fD
f
Remember that the characteristic equation of the system is:

𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔2 𝑛 = 0
The time domain response of a system depends upon the roots of the characteristic
equation given by:
𝑠1 , 𝑠2 = −𝜉𝜔𝑛 ± 𝜔𝑛 𝜉 2 − 1
The roots are of three different types depending upon the value of 𝜉. Hence there are
three types of systems and consequently three types of response.
[1] Over damped system: 𝜉 > 1
[2] Critically damped system: 𝜉 = 1
[3] Under damped system: 𝜉 < 1
Further Reading: Solve and plot the poles (roots of the characteristic equation) in the
𝑆 𝑝𝑙𝑎𝑛𝑒, sketch step responses, characterize each cases based on the speed of response
and comment on the presence or absence of oscillation.
Step response:
Let a unit step be applied to a second order system 𝑟 𝑡 = 𝑢(𝑡):
case i. Over damped system
𝐾𝜔2 𝑛
𝐶 𝑠 =
𝑠 𝑠 + 𝜉𝜔𝑛 − 𝜔𝑛 𝜉 2 − 1 𝑠 + 𝜉𝜔𝑛 + 𝜔𝑛 𝜉 2 − 1
Which upon Laplace transform becomes:
𝑐 𝑡
𝐾
𝜉 + 𝜉 2 − 1 −𝜉+ 𝜉 2 −1 𝜔 𝑡
= 1− 𝑒 𝑛
2 𝜉 −12
𝜉− 𝜉2 − 1 𝜉 2 −1 𝜔 𝑛 𝑡
+ 𝑒 −𝜉−
2 𝜉2 − 1
case ii. Critically damped system
In this case:
𝑐 𝑡 = 1 − 1 + 𝜔𝑛 𝑡 𝑒 −𝜔 𝑛 𝑡
case iii. Under damped system
Come up with the following time domain response:
𝑐 𝑡 𝑒 −𝜉𝜔 𝑛 𝑡
=1− sin 𝜔𝑛 𝜉 2 − 1 𝑡 + cos −1 𝜉
𝐾 1−𝜉 2
Usually this is given also as:

=1− sin 𝜔𝑑 𝑡 + 𝜑
𝐾 1 − 𝜉2
Where 𝜔𝑛 𝜉 2 − 1 = 𝜔𝑑 =frequency of damped oscillations, cos −1 𝜉 =

1−𝜉 2
𝜑 = sin−1 1 − 𝜉2 = tan−1 𝜉
Note that the dynamic error becomes:
𝑒 𝑡 =𝑟 𝑡 − = sin 𝜔𝑑 𝑡 + 𝜑
𝐾 1 − 𝜉2
And hence, the steady state error is:
𝑒𝑠𝑠 𝑡 = lim 𝑒(𝑡) = 0
𝑡→∞
Exercise: analyze the system with values of 𝜉 as given next:
1. 𝜉 < 1, decaying oscillatory
2. 𝜉 = 0, constant oscillations, undamped
3. 𝜉 = 1, no oscillations
4. 𝜉 > 1, non-oscillatory, sluggish
Ramp response:
Considering the case of an under damped second order system for 𝑟 𝑡 = 𝑡:
1 𝐾𝜔2 𝑛
𝐶 𝑠 = 2 2
𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔 2 𝑛
Taking partial fractions and applying inverse Laplace transform:
𝑐 𝑡 2𝜉 𝑒 −𝜉𝜔 𝑛 𝑡
=𝑡− + sin 1 − 𝜉 2 𝜔𝑛 𝑡 + cos −1 𝜉
𝐾 𝜔𝑛 𝜔𝑛 1 − 𝜉 2
The dynamic and the steady state error are given next:
𝑐 𝑡 2𝜉 𝑒 −𝜉𝜔 𝑛 𝑡
𝑒 𝑡 =𝑟 𝑡 − = − sin 1 − 𝜉 2 𝜔𝑛 𝑡 + cos −1 𝜉
𝐾 𝜔𝑛 𝜔𝑛 1 − 𝜉 2
2𝜉
𝑒𝑠𝑠 (𝑡) =
𝜔𝑛
Impulse response:
Let a second order system be subjected to an impulse of strength A.
Here we will consider critically damped and under damped systems, which are of
importance to us.
For a critically damped system,
𝑐 𝑡 = 𝐴𝜔𝑛 2 𝑡 𝑒 −𝜉𝜔 𝑛 𝑡
For under damped system,
𝑒 −𝜉𝜔 𝑛 𝑡
𝑐 𝑡 = 𝐴𝜔𝑛 sin 1 − 𝜉 2 𝜔𝑛 𝑡
1−𝜉 2
Time domain specifications:
Measurement systems are designed with 𝜉 < 1 with the result they exhibit an
oscillatory response. When we use an under damped system, there are certain
properties of interest.
Exercise: obtain time domain specifications in quantitative form:
[1] Rise time, 𝑡𝑟
[2] Peak time, 𝑡𝑝
[3] Peak overshoot, 𝑀𝑝
[4] Settling time, 𝑡𝑠
Frequency response of a first order system:
The transfer function of a first order system is:
1
𝐺 𝑠 =
1 + 𝜏𝑠
The sinusoidal transfer function is then,
1
𝐺 𝑗𝜔 =
1 + 𝑗𝜔𝜏
Suppose the system is subjected to a sinusoidal input:
𝑅 𝑗𝜔 = 𝐴𝑖 sin 𝜔𝑡
1
∴ The output 𝐶 𝑗𝜔 = 𝐺 𝑗𝜔 𝑅 𝑗𝜔 = 1+𝑗𝜔𝜏 𝐴𝑖 sin 𝜔𝑡
The ratio of the output to the input is,
𝐴𝑜 1 1
𝑀= = =
𝐴𝑖 1 + 𝑗𝜔𝜏 1 + 𝜔𝜏 2
∴ Magnitude of the output is,

𝐴𝑖
𝐴𝑜 =
1 + 𝜔𝜏 2
Phase displacement of output is,

𝜑 = − tan−1 𝜔𝜏
Or, the output is:
𝐴𝑖
𝑐 𝑡 = sin 𝜔𝑡 − tan−1 𝜔𝜏
1 + 𝜔𝜏 2
Exercise: Plot the magnitude response and the phase angle versus frequency.
Note that the higher the frequency is the higher will be the attenuation of the output and
the greater will be the angle of lag between the output and input.
There is another factor which affects the frequency response of the system and that is the
time constant 𝜏. The greater the time constant, greater is the attenuation of the output and
also the greater is the phase shift between the output and the input.
Frequency response of a second order system:
Let a second order system be subjected to a sinusoidal input
𝑟 𝑡 = 𝐴𝑖 sin 𝜔𝑡
And the steady state output 𝑐 𝑡 = 𝐴𝑜 sin 𝜔𝑡 + 𝜑

Given the transfer function of a second order system as:
𝐾𝜔2 𝑛
𝐺 𝑠 = 2
𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔 2 𝑛
Substituting 𝑠 = 𝑗𝜔,
𝐾𝜔2 𝑛
𝐺 𝑗𝜔 =
𝑗𝜔 2 + 2𝜉𝜔𝑛 𝑗𝜔 + 𝜔 2 𝑛
𝐾𝜔2 𝑛
=
𝜔𝑛 2 − 𝜔 2 + 2𝑗𝜉𝜔𝑛 𝜔
𝐾
= 2
,
1 − 𝑢 + 2𝑗𝜉𝑢
𝜔
𝑤𝑕𝑒𝑟𝑒, 𝑢 = 𝜔 =Normalized freqency
𝑛
Therefore, the ratio,

𝐴𝑜 1
𝑀= =
𝐴𝑖 1 − 𝑢2 2 + 2𝜉𝑢 2
And the phase angle,

2𝜉𝑢
𝜑 = − tan−1
1 − 𝑢2
Exercise: Plot the magnitude ratio and the phase angle characteristics for various values
of normalized frequency 𝑢 for a particular value of 𝜉.
Frequency domain specifications:
i. Resonant frequency: the frequency at which 𝑀 has the maximum value. The
normalized resonant frequency can be found by differentiating the expression for
𝑀 with respect to 𝑢.
𝜔𝑟 = 𝜔𝑛 1 − 2𝜉 2
ii. Resonant peak: the maximum value of 𝑀.
1
𝑀𝑟 =
2𝜉 1 − 𝜉 2
iii. Bandwidth: the band of frequencies from zero to cut off frequency 𝜔𝑐 .
Measurement systems are low pass filters as the value of amplitude ratio is unity
at 𝜔 = 0. But as the frequency of the input signals increases, the output gets
attenuated. Bandwidth is, therefore, indicative of the satisfactory reproduction of
the input signal.
1.4.Noise and interference in measurement
We studied the dynamic response of measurement systems to step, ramp, impulse and sine
wave input signals. These signals are examples of deterministic signals: a deterministic
signal is one whose value at any future time can be exactly predicted.
However, in real measurement applications the input signal to the measurement system is not
deterministic but random.
Five statistical quantities – mean, standard deviation, probability density function, power
spectral density and autocorrelation function –are used to estimate the behavior of random
signals.
Further Reading: Revise the five statistical quantities mentioned above.
However, unwanted electrical signals may also be present in the measurement circuit. The
unwanted signal then may be either random, e.g. signals caused by the random motion of
electrons, or deterministic, e.g. sinusoidal signals at 50 Hz caused by power cables.
Unwanted random signals are usually referred to as noise signals and unwanted deterministic
signals as interference signals.
Internal noise sources
The random, temperature-induced motion of electrons and other charge carriers in resistors
and semiconductors gives rise to a corresponding random voltage which is called thermal or
Johnson noise.
A similar type of noise is called shot noise; this occurs in transistors and is due to random
fluctuations in the rate at which carriers diffuse across a junction. These are characterized by
a uniform power spectral density over a wide range of frequencies.
External noise and interference sources
The most common sources of external interference are nearby a.c. power circuits. These can
produce corresponding sinusoidal interference signals in the measurement circuit, referred to
as mains pick-up or hum. Power distribution lines and heavy rotating machines such as
turbines and generators can cause serious interference.
D.C. power circuits are less likely to cause interference because d.c. voltages are not
coupled capacitively and inductively to the measurement circuit.
However, switching often occurs in both a.c. and d.c. power circuits when equipment such as
motors and turbines is being taken off line or brought back on line. This causes sudden large
changes in power, i.e. steps and pulses, which can produce corresponding transients in the
measurement circuit.
The air in the vicinity of high voltage power circuits can become ionised and a
Corona discharge results. Corona discharge from d.c. circuits can result in random noise in
the measurement circuit and that from a.c. circuits results in sinusoidal interference at the
power frequency or its second harmonic.
Fluorescent lighting is another common interference source; arcing occurs twice per cycle
so that most of the interference is at twice the power frequency.
Radio-frequency transmitters, welding equipment and electric arc furnaces can produce r.f.
interference at frequencies of several MHz.
Further Reading: study various coupling coupling mechansisms to external sources

(Inductive, Capacitive, Multiple earths) and various methods of reducing noise and
interference like physical separation, electromagnetic shielding, electrostatic screening and
shielding, use of differential amplifiers, filtering, modulation, averaging, autocorrelation, etc.
End of Chapter
Document History: Edition 1, Revision 0

Instrumentation (Chapter 1,0)

Uploaded by

Copyright:

Available Formats

Instrumentation (Chapter 1,0)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Instrumentation (Chapter 1,0)

Uploaded by

Copyright:

Available Formats

Chapter 1 Introduction to Instrumentation (ECEG-4201) Page |1

1.1. Basic Concepts of Measurement

Figure 1.1 An instrumentation/measurement system

Measurements are made with the help of instruments. An instrument can be

A mercury-in-glass thermometer maintains a linear relationship between

Functions of a measurement system

Measurement of system parameters: present desired information about the

Example: thermometer, barometer, anemometer, water/gas/electricity meters

Control of a certain process or condition: the measurement system forms an

Figure: 1.2. Typical block diagram representation of automatic control system

Example: A thermostat fitted in a domestic refrigerator is a control device for

Verification of physical phenomena/scientific theories: experiments are

Properties to be measured Application Field

Atmospheric pressure, hygrometry level Metrology

A generalized measurement system can be schematically represented as shown

Figure: 1.3. Functional elements of a measurement system.

Thermocouple where millivolt e.m.f. depends on temperature

Strain gauge where resistance depends on mechanical strain

Orifice plate where pressure drop depends on flow rate.

Deflection bridge which converts an impedance change into a voltage change

Amplifier which amplifies millivolts to volts

Oscillator which converts an impedance change into a variable frequency voltage.

Typical calculations are:

Integration of chromatograph peaks to give the composition of a gas stream

Correction for sensing element non-linearity.

Example: radio links or telemetry systems

Control and analysis-microprocessor, microcomputer

Figure: 1.4. Permanent magnet moving coil mechanism (PMMC)

Figure 1.5: Bourdon tube pressure gage.

A standard of measurement is defined as the physical representation of the unit of

International standards: are devices designed and constructed to the specifications of an

Primary standards: are devices maintained by standards organizations or national

a) Resistance, ohm (Ω)

Instruments can be classified according to their application, mode of operation, manner of

1. Deflection Vs. Null types

operational cost is considerably lower than that of the corresponding manually

1.2.Static Performance Characteristic

Instrument errors: inherent errors due to poor design/construction in the

Propagation of uncertainties in compound quantities:

To compute the overall uncertainties of different variables, we consider the following

where y is a parameter that depends on the independent variables

Because of the random nature of the uncertainties in the quantities given by

𝜌1 = 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑐𝑦𝑙𝑖𝑛𝑑𝑒𝑟𝑖𝑐𝑎𝑙 𝑝𝑜𝑟𝑡𝑖𝑜𝑛 = 3.5 ± 0.1 𝑔/𝑐𝑚3

𝜌2 = 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑕𝑒𝑚𝑖𝑠𝑝𝑕𝑒𝑟𝑖𝑐𝑎𝑙 𝑝𝑖𝑠𝑡𝑜𝑛 = 2.5 ± 0.05 𝑔/𝑐𝑚3

Figure 1.7: Example body

The total mass the body, 𝑀𝑇 , is given by:

𝑀𝑇 = 𝑀𝑐𝑦𝑙𝑖𝑛𝑑 𝑒𝑟 + 𝑀𝑕𝑒𝑚𝑖𝑠𝑝 𝑕𝑒𝑟𝑒

The overall uncertainity in 𝑀𝑇 can be written as:

Alternatively, the uncertainity in the total mass is also expressed in percentage:

Static Performance Parameters

The accuracy of the instrument can be specified as:

1. Percentage of the true value

2. percentage of the full-scale deflection

Figure: 1.8. Accuracy versus precision in a shooting experiment

Resolution (or discrimination):

Likewise the output range is 𝑂𝑚𝑖𝑛 to 𝑂𝑚𝑎𝑥 .

The maximum variation in input or output. i.e.

Input span is 𝐼𝑚𝑎𝑥 − 𝐼𝑚𝑖𝑛

Output span is 𝑂𝑚𝑎𝑥 − 𝑂𝑚𝑖𝑛 .

𝑐𝑕𝑎𝑛𝑔𝑒 𝑜𝑓 𝑜𝑢𝑡𝑝𝑢𝑡 𝑠𝑖𝑔𝑛𝑎𝑙 ∆𝑂