ABUBAKAR TAFAWA BALEWA UNIVERSITY BAUCHI

FACAULTY OF ENGINEERING AND ENGINEERING TECHNOLOGY

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING
EE 418: CONTROL ENGINEERING LABORATORY

Process Control
TPS-3011

Scientific Educational Systems

 Process Control
TPS-3011

2_15

© All rights reserved SES Scientific Educational Systems Ltd.

The material in this book may not be copied, duplicated, printed,

translated, re-edited or broadcast without prior agreement in writing
from SES.

 www.ses.co.il 
 I

Contents
Preface .........................................................................................................................III
Chapter 1 – Introduction to Control Systems ............................................................ 1
1.1 Control systems and systems control ............................................................... 1
1.2 Regulation and control systems ........................................................................ 2
1.3 Dual stage, multi stage and continuous control ................................................ 5
1.4 Amplification and P controller ......................................................................... 7
1.5 Transmission .................................................................................................. 10
1.6 Disturbances of the controlled variable - open loop and closed loop control 12
1.7 Dynamic response of control systems ............................................................ 13
1.8 Control system archetypes.............................................................................. 17
Experiment 1.1 – Open Loop DC Motor P Control 19
1.1.1 DC motor control and rotation measurement ................................................. 19
1.1.2 DC motor load response ................................................................................. 20
1.1.3 DC motor speed response and status .............................................................. 20
Experiment 1.2 – Open Loop Light P Control 27
1.2.1 Light control using lamp, amplifier and LDR photo-resistor ......................... 27
Experiment 1.3 – Open Loop Temperature P Control 31
1.3.1 Temperature control using lamp, amplifier and NTC thermistor ................... 31
Experiment 1.4 – Open Loop Level P Control 35
Chapter 2 – Bi-Stage Amplifier ................................................................................. 39
2.1 Two position controller .................................................................................. 39
2.2 Three range control with two position controller ........................................... 40
2.3 Thermostat as an area range 2-position temperature control ......................... 43
Experiment 2.1 – Dual Stage Control of a DC Motor 45
Experiment 2.2 – Dual Stage Light Control 51
Experiment 2.3 – Dual Stage Temperature Control 56
Chapter 3 – Negative Feedback ................................................................................. 60
3.1 Negative feedback in control system.............................................................. 60
3.2 Analysis of the transfer function and the amplification formula.................... 63
3.3 Gain and Proportional .................................................................................... 67
3.4 Steady state and steady state error (static method)......................................... 68
3.5 Transient response, dynamic reaction and time constant ............................... 71
3.5.1 Motor-Dynamo system behavior .................................................................... 72
3.5.2 General differential equator and its solution .................................................. 73
3.5.3 Level control concordant with capacitor charging ......................................... 77
3.5.4 Position control............................................................................................... 78
3.5.5 Temperature control ....................................................................................... 79
3.5.6 Summary ........................................................................................................ 80

SES
 II

Experiment 3.1 – DC Motor Closed Loop Control with Negative Feedback 81

3.1.1 DC motor control response and status ............................................................ 81
Experiment 3.2 – Light Closed Loop Control with Negative Feedback 90
Experiment 3.3 – Temperature Control with Negative Feedback 97
Experiment 3.4 – Level Closed Loop Control with Negative Feedback 103
Chapter 4 – Second Degree Systems and PID ........................................................ 108
Experiment 4.1 – Second Degree Systems and PID 108
4.1.1 Second order systems ................................................................................... 108
4.1.2 The PI controller system............................................................................... 113
4.1.3 PI control in open and closed loop ............................................................... 113
4.1.4 Fluid level control with PI controller ........................................................... 115
4.1.5 The PID controller system ............................................................................ 116

SES
 III

Preface
There are many types of control systems. In the past, most control systems
were realized using dedicated electromechanical or electronic systems.
Each system had its own specially built circuit to control it. The most
advanced system of this kind is called an Analog Computer. This
computer consists of a group of modules interconnected in different ways.
Each module has a different function - amplification, attenuation, summing,
subtraction, integration, differentiation etc. Each module is a dedicated
electronic circuit.

Modern control systems are digital. They receive continuously varying

analog values at their input and convert them to digital form (the computer
only handles digital values). The signals are processed in digital form and
the output is converted back into analog form as required. Digital
processing is powerful, efficient, and mainly very reliable when compared
to analog components.

In order to cover the variety of subjects in control, SES developed two

trainers. The first trainer is TPS-3011 for analog process control. The
second trainer is TPS-3071 for digital and computerized process control.

The trainers come in a metal case with a wide experiments board. The
experiments area is divided into modules with test-points.

The actual electronic devices are located at the upper and lower parts of
each circuit and are protected by a hard and transparent polycarbonate
cover. In this way, the experiments combine principle modules with the
possibility of going deeper into the circuit and component level.

The trainer includes an ON/OFF switch and internal regulated power

supply, which supplies regulated voltages. The trainer connects to the
computer via serial communication.

SES
 IV

The communication trainers are as follows:

1) TPS-3011 – process control trainer:

This trainer is intended for experiments in process control.

 Motor - dynamo for speed control.

 Lamp - thermistor for temperature control.
 Lamp - light sensor for light control.
 Level control unit.
 Power amplifier.
 Feedback amplifier.
 Feedback attenuator.
 Adder amplifier.
 Dual-stage amplifier.
 PI amplifier.
 PD amplifier.
 Voltmeter.
 Computerized and manual variable reference voltages.
 Computerized two-channel oscilloSESCOPE.

2) TPS-3071 – digital control trainer:

This trainer is intended for experiments in digital control, ladder diagrams

with PLC and also in computerized mode.

 Step motor.
 Motor - dynamo for speed control.
 Lamp - thermistor for temperature control.
 Lamp - light sensor for light control.
 Elevator simulator.
 Traffic lights simulator.
 Switches.
 Pushbuttons.
 Connecting points for the PLC.
 Digital and analog computerized I/O unit.

SES
 V

Experimenting method:
The TPS-3011 training system is a stand-alone system. It comes with
external power adapter and includes main operating switch.

The experiments require using test equipment such as Voltmeter and two-
channel oscilloscope. We can use such standard laboratory test equipment.

In this case, the trainer can be operated (including changing the trainer's
reference voltages) manually.

The trainer can be connected in serial communication to a PC. In this case,

we run the SESCOPE software, which operates the SESLAB function
generator and displays the signals of the SESLAB channel In1 and In2.

The experiments direct you to measuring equipment. You can use standard
equipment or the SESCOPE software and the SESLAB unit of the trainer.

Installing the SESCOPE software:

The SESCOPE software is a 'virtual instrument' software (a laboratory
instrument that changes its panel and function according to the desired
instrument), which operates with the trainer's SESLAB unit.

The software is aimed to operate with different trainers and adapts the
panel window to the connected training system.

The SESCOPE is a WINDOWS application program. It is not protected

and only copying it does installation.

1) Turn ON the PC.

2) Using 'My Computer' functions, copy the SESCOPE subdirectory
from the diskette to the hard disk.
3) Create a short cut to the software SESCOPE.exe.
4) Move this short cut to the desktop window.

To run the SESCOPE software you have to double click on this icon.

SES
 VI

How to use the SESLAB:

Step 1: Connect the trainer to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Turn ON the main switch of the trainer.

Step 4: Make sure that the trainer is connected in serial communication

to the computer. The communication cable includes an earphone
socket in one hand (should be connected to the earphone outlet
on the rear panel of the trainer) and a D9 socket (should be
connected the COM1 or COM2 outlet of the PC) on the other
hand.

Switch the Manual/Auto switch in the SESLAB module to Auto

position.

Step 5: Run the SESCOPE program.

The following screen will appear:

Step 6: Set the number in the OPEN COM field according to your
communication channel number in the computer.

SES
 VII

Step 7: Click on the OPEN COM button.

The computer will open the communication channel, turns to the

trainer and asks for its type.

If the communication is successful, the name of the trainer and

its SESLAB version will appear on the screen headline.

The screen updates and adapts itself to the type of the trainer.

Step 8: If the communication is not successful, click the Trainer button.

If the communication succeeds, the screen will update

accordingly.

With every communication with the trainer, messages on success

or failure appear on the lower left side of the window.

Step 9: If the communication fails:

Check that the trainer is turned ON.

Check that the Manual/Auto switch is on Auto position.
Check that the earphone socket is plugged properly.
Check that the cable is connected properly to the computer.
Check that the communication channel in the computer is in
order and not occupied by another program.

Step 10: If it's necessary, move the cable to another communication

channel in the computer.

In this case, you should exit the SESCOPE program and run it
again.

Step 11: The program remembers the last communication channel you
opened.

If you don't need to change the communication channel, just click

the OPEN COM button before starting working with the program.

SES
 VIII

Step 12: The SESLAB unit includes two analog inputs of the oscilloscope
(IN1 and IN2) and two variable reference voltage outputs. At
manual state, the reference voltages (Vref1 and Vref2) are set
with the two potentiometers located near these voltage outputs.

At the Auto state, the potentiometers on the trainer do not affect

the variable reference voltages. The reference voltages are set by
the bars on the computer screen.

Identify the bars on the screen.

Step 13: Connect the variable reference voltages outputs (Vref1 and Vref2)
to the two SESCOPE inputs (with 2 banana wires) of the
SESLAB.

Step 14: The SESCOPE screen includes fields and buttons for changing
the time base, the type of trigger and the level trigger.

Set the time base to 50ms/cm.

Step 15: The software enables to determine the Level Trigger and its
options (Free, Rise, Fall). Identify these buttons and fields.

For measuring constant voltages (as we perform from now on),

we use the Free mode. In this mode, the system does not refer to
the trigger level and does not wait for synchronization.

Step 16: The program shows channel 1 (CH1), channel 2 (CH2) or both.

The selecting is done by clicking the required channel. Click for

selecting and another click for cancellation.

The screen can also display the spectral analysis of the sampled
signal in channel 1 (CH1 SPC). Selecting this function cancels
CH2.

Select both channels.

SES
 IX

Step 17: Sampling the signals and displaying them can be done in two
ways – One Cycle or Free Run.
In One Cycle, the PC asks the trainer to perform one sample
cycle and transmit it. The SESCOPE will display the information
on the screen.
In Free Run, the PC asks the trainer to perform one sample cycle
and transmit it. After receiving the information, the SESCOPE
will display it on the screen and asks for another sample cycle
from the trainer.
In every sample cycle request, the PC transmits the required
sample rate to the trainer (according to the time base) and the
trigger level.
The PC also informs if two channels or one channel should be
sampled.
Note:
The amount of samples in each sample cycle is 100.
With TPS-3011 you can do the sampling only in Chopper
(Chop.).
In the Chopper state, the SESLAB controller samples the two
channels simultaneously. In this case, the controller gives 50
samples to each channel. The sample rate for a channel is
also smaller by half. The quality of the sampling and the
display is not so good, but shows how the signals react to each
other on the time axis.
The synchronization is done according to channel 1 (CH1).
In the Alternate state, the controller synchronizes on channel
1, sample it and transmit it. Afterwards, the controller
synchronizes again on channel 1, sample channel 2 and
transmit it.
The signal picture, in this case is in maximum quality.
If the two signals are cyclic, they will also be synchronized to
each other.
Select the Chopper mode.

SES
 X

Step 18: Set the SESCOPE to Free Run and observe the signals on the
screen.

The square wave is unstable because the SESLAB does not wait
for trigger level.

Step 19: Change Vref1 and Vref2 with the voltage bars and observe the
signals on the screen.

Step 20: Change the time base to 20ms/cm and check the reaction to
Vref2.

Step 21: Change the trigger mode to Fall and to Rise and observe the
signals.

Step 22: Set the SESCOPE to One Cycle.

You will get a frozen picture of the last sample.

Note:

When you want to analyze the data on the screen, it is better

to use a One Cycle sample.

Step 23: The program allows measuring the sampled signals.

In order to do this, you should activate two marker lines on the

window.

Two marker lines, green and yellow will appear on the screen.

If these lines do not appear on the screen, locate the Active

button and click over it.

Step 24: You can move each line separately to the right or the left with the
arrows buttons.

Move the lines and check it.

The software calculates the time between the two lines and the
frequency according to f  1 .
T

The software also indicates the voltage values of each channel

for every line and calculates the voltage difference.

SES
 XI

Step 25: Change the reference voltages rulers and observe the screen
reaction.

Measure these voltages.

SESLAB in Manual:
When we change the manual/Auto switch to manual, the trainer's controller
ignores the communication with the computer. Instead, it samples the two
potentiometers on the panel and changes the reference voltages accordingly.

Step 26: Connect Vref1 output to the panel voltmeter input.

Step 27: Set the AUTO/MANUAL switch to MANUAL position.

Step 28: Turn the system off and then turn it on.

Step 29: Change Vref1 potentiometer and observe its effect on the
voltmeter lights.

Step 30: Connect Vref2 output to the panel voltmeter input.

Step 31: Change Vref2 potentiometer and observe its effect on the
voltmeter lights.

In the experiments, we will not relate to any kind of measuring equipment.

The user should adapt his operation method according to the equipment he
has.

Recap

 The trigger voltage level control sets the voltage at which the trigger fires.
Changing this voltage changes the point on the waveform where the time
base starts and varying the oscilloscope trigger voltage point. It can be
seen that by altering the trigger voltage, the position on the waveform is
varied.

SES
 1

Chapter 1 – Introduction to Control

Systems
1.1 Control systems and systems control
The term "Control Systems" comprises two separate terms, which can be
defined individually:
A system is a combination of a number of components, which are
interconnected and operate as a unit, in order to achieve a specific goal.
Control is the ability to maintain command over a set of variables.
A "Control system" may be defined as a group of components, which can
be operated together to control various variables, which govern the
behavior of the system. The human body has systems, which control body
temperature. In an auditorium, you might find a system, which controls the
air conditioning. Motorcars have systems, which control fuel consumption,
and many other such examples.
A greenhouse control system is a group of components, which work
together to govern devices, which maintain air temperature, soil humidity,
or the amount of ambient light in the greenhouse.
A speed control system maintains a steady motor speed regardless of the
changing load on the motor.
A light control system can maintain a steady level of light, regardless of the
amount of available sunlight. Lamps will be switched ON or OFF, when
there is not enough natural light, according to the required light level.
A water level control system maintains a fixed water level in a tank or a
steady flow of water in a pipe. There is a device of this type in almost every
home - think about it!
Sometimes a control system is required, which enables the preset operating
conditions to be changed often or while the system is running. For example,
the operating conditions of a traffic light system need to be changed as the
traffic load changes, changing the cycle times of a washing machine
according to different kinds of laundry, or changing the operating
conditions of a greenhouse according to the season or different crops.
In this book we will meet many terms associated with control systems,
become acquainted with them and understand them through practice.

SES
 2

1.2 Regulation and control systems

Very often, these two terms are confused, as they refer to very similar
systems. Both systems regulate the controlled system. Here are some
examples of regulation and control devices:

 Light intensity regulation.

 Motor rotation speed regulation.
 Regulation of temperature by heating or cooling.
 Regulation of water flow.

Regulation may be performed by switching ON and OFF or by continuous

or progressive changes.

The difference between control and regulation is in the method used for
regulation.

In a regulation system, regulation is performed directly on the regulated

device, without any external intervention. In a control system, the
controlling device uses an intermediate device to influence the controlled
system.

Example:

A lamp connected to a voltage source is neither a regulated nor a controlled

system. It is symbolized thus:

Voltage
Lamp
Source

Figure 1-1

SES
 3

The system will become a regulated system if we add a switch.

Switch

Voltage
Lamp
Source

Figure 1-2

The switch allows us to break the current through the lamp as required.

This system will become a control system if we use a relay in this way:

Control
System

Relay

Voltage Lamp
Source

Figure 1-3

When the control system sends current through the electromagnet, this
causes the relay switch to close and the light goes ON. The current required
to operate the relay is different from that required by the lamp (usually
much smaller).

We could replace the lamp with a motor, and/or change the voltage source
to an AC source in place of the DC source, but the principle is quite clear.

SES
 4

It is also possible to perform continuous regulation of the lamp. For this

purpose, we may use (more in the past than today) a variable resistor
connected in series with the lamp. This kind of resistor is called a rheostat.
Changing the position of the rheostat changes the current in the circuit and
so changes the light intensity, in a continuous manner.

Rheostat

Voltage
Lamp
Source

Figure 1-4

Rheostats tend to heat up under large currents and must be designed for this.
These are large and cumbersome components.

In a control system, an amplifier is used to push current through the load

lamp, according to the voltage on a potentiometer (also a variable resistor,
but for very low currents) at its input.

+ Amplifier

-
Voltage
Source Lamp
Potentiometer

Figure 1-5

There are infinite variations of regulation and control systems.

A steering mechanism is a system, which regulates the direction of a car. A

power steering is a control system. Turning the steering wheel operates a
piston, which applies steering force to the wheels.

In this book, we will relate to control systems so that we can deal with
automatic process control.

SES
 5

1.3 Dual stage, multi stage and continuous

control
From the aspect of control flexibility, we can coarsely divide control
systems into three categories:

a) Dual stage control:

Usually this is control at the ON/OFF level. A switch or a relay serves as a

dual stage controller, or a lever which is raised or lowered with no
intermediate position, or a tap operated by a valve which is open or closed.

A dual stage control system, for example, in which there is a different

current intensity for each stage (high or low current), is possible. In other
words, a break in the operation of the control system is not necessarily one
of the stages.

b) Multi stage control:

In this type of system, the controlled system is operated at different and

clearly defined levels of intensity with no intermediate levels. This kind of
system is usually a combination of a number of dual stage systems. For
example, operation of a lighting system with a variable number of lamps at
their maximum intensity. Each individual lamp is a dual stage system. The
complete lighting system is a multi stage system.

Selector

Figure 1-6

Another example is an oven, which has a number of heating elements. By

operating a selector or relays, we arrange different combinations of heating
elements to achieve different levels of heat. A similar example is an
irrigation control system, or a water pressure control system, where a group
of parallel valves are used to control the flow.

SES
 6

Another method of multi stage control is to use a source having various

progressive values. For example, a voltage source made up of a number of
batteries or voltage supplies. We use a selector to choose a combination of
batteries. For example, assume we wish to run a fan with 4 operating stages:
3 speeds and OFF. A variable voltage source here might be a transformer
with a number of windings each supplying a different voltage.

Figure 1-7

c) Continuous control:

In this case, the controlled system can receive any required value within a
given range. A tap is one example of a continuous control system, so is a
car steering mechanism or power steering and so on. In electrical systems,
to regulate as mentioned before, we use potentiometers, which are variable
resistors, motors or other devices, and not rheostats as in the past.

Motor or lighting control is sometimes performed with a device called a

"dimmer". This device controls the current through the lamp or motor
continuously, according to the position of a potentiometer.

It is possible to create a multi stage control system with a very large

number of stages. This condition is called "high resolution". This kind of
control can be called multi stage, but is also effectively continuous.

The controlling system may be a manual or an automatic system such as a

computer, programmable controller or any electronic system. We will
expand on this subject later.

SES
 7

1.4 Amplification and P controller

An amplifier is a system, which receives signals of a certain magnitude at
its input, amplifies them, and outputs them at a different magnitude. For
example a voltage amplifier:

Vin A Vout

Figure 1-8

The amplifier and the amplification are symbolized with the letter A. The
amplification is the ratio of the output to the input.

Vout
A
Vin

An attenuator is also a kind of amplifier with fractional amplification

(0 < A < 1) i.e. the output signal is smaller than the input signal.

We distinguish different types of amplifiers. Usually we symbolize them

with a triangle or a box like this:

Iin AI Iout Vin AV Vout

Current Amplifier Voltage Amplifier

I out Vout
Current Gain AI = Voltage Gain AV =
I in Vin

Figure 1-9

SES
 8

It is possible to perform voltage or current amplification without any power

source in addition to the input signal with a transformer for example. A
transformer amplifies the voltage and attenuates the current, in direct
proportion with the ratio of the windings.

Ii Io

Vi n1 n2 Vo

Vo n2 Io n2
Av   Iv  
Vi n1
Ii n1

Figure 1-10

In this case, there is no amplification of the energy or the power of the

input signal, only the signals form is changed. A transformer which
amplifies current attenuates voltage proportionally and vice-versa.

There are also power amplifiers, which amplify the energy of a signal. Due
to the law of conservation of energy, it is necessary to provide an external
power source in addition to the input signal.

Ii Io
V Vi AV Vo

Figure 1-11

In this system, the input signal consists of a low voltage and a low current,
and both these parameters are amplified. The power is the product of the
voltage and the current. A power amplifier increases both these parameters
but the additional energy is drawn from an external power source - the
voltage source in figure 1-10.

This amplifier is called proportional (P) amplifier or P controller.

SES
 9

Amplification is not a concept limited only to electricity. It is prevalent in

many other fields too. For example, force amplification. In a pneumatic or
hydraulic system, we use pistons to amplify force.

F1 F2

Narrow Wide
Piston Piston

Figure 1-12

The force amplification depends on the ratio of the surface areas of the
pistons.

F1 S 2
A 
F2 S1

Obviously, as there is no external power source, there must be a trade-off

somewhere. In this case, we loose in the length of travel. The narrow piston
must travel much further then the wide piston - in fact A times as far.

Power can also be amplified by mechanical transmissions (friction, screw,

belt etc.). Here too we will pay for force amplification with longer travel.

If we were to use an electrical motor to perform this action, we could have

very large power amplification without paying with large travel. For
example, a control column (such as a computer joystick or an aircraft's
control column) coupled to a potentiometer, which supplies a variable
voltage to a power amplifier. The power amplifier output is connected to an
electric motor, which performs the mechanical action. With a power
amplifier, we can have almost any amplification we require.

SES
 10

1.5 Transmission
The term amplification is used to describe the ratio between the magnitude
of an output signal and an input signal of the same kind - voltage, current,
power, force etc. Amplification is a scalar value having no units.
Occasionally we need to know the ratio of signals, which are not the same.
This ratio is called the "Transfer Function" and is usually symbolized with
the letter G (Gain).

Let's look at an example which is a combination of transferences.

Vm is the voltage we supply to an electric motor, to turn its shaft at the

rad
speed of  radians per second ( ).
sec

If the motor shaft is coupled via a mechanical transference (gears etc. -

usually called a "Transmission") to an elevator, the elevator speed, which is
m
the linear speed VE   is a function of the rotation speed . The elevator
 sec 
transfer function is:

VE m / sec m
GE   
 rad / sec rad

If we connect the motor to a power amplifier, which sees the voltage Vi at

its input and outputs Vm, with the power required to move the elevator. The
voltage amplification will be AV.

Vm
AV 
Vi

This is a scalar value and has no units.

SES
 11

The total system will look like this:

Power Electric Elevator

Amplifier Motor Transmission
Vm 
Vi AV Gm GE
Motor Input
Voltage
Rotation
Speed Elevator VE
Speed
Elevator

Figure 1-13

The total transference T is the ratio of the elevator speed and the input
voltage Vi, and can be found in the following way:
Vm
AV   Vm  Vi  A V
Vi

Gm     Vm  G m    Vi  A V  G m
Vm
VE
GE   VE    G E  VE  Vi  A V  G m  G E

V
 T  E  AV  Gm  GE
Vi

i.e., the total transference is the product of all the transferences.

Note that we use the letter V to symbolize voltage (Volt) as well as linear
speed (Velocity). Take care to avoid confusion.
The transfer units change according to the system's components. In our
example the transfer units will be calculated thus:

T  AV  Gm  GE
AV  1(Scalar )
rad / sec
Gm 
Volt
m / sec
GE 
rad / sec
rad / sec m / sec
T  1 
Volt rad / sec

SES
 12

rad/sec cancel out and so the transfer units are:

m / sec
T
Volt

1.6 Disturbances of the controlled variable -

open loop and closed loop control
The output signal from a control system is called the controlled variable.
This variable might be the velocity of an elevator, the water capacity of a
pump, rotation speed of a motor, motion of a crane, light intensity,
temperature of an oven, etc.

In the systems we have described so far, we assumed that the controlled

variable depends on the input signal to the control system and the transfer
function. In other words, if we increase the voltage at the input, the elevator
will move faster. Similarly, we may increase the amplification of the power
amplifier, the motor transmission or the mechanical transmission of the
elevator.

This is not entirely accurate. The behavior of the elevator also depends on
additional factors - the number of passengers in the elevator, if it is going
up or down etc. It is obvious that the elevator speed will fluctuate if we
maintain a steady input voltage and vary these disturbing factors.

In almost every control system, there are parameters which interfere with
the consistent behavior of the controlled variable. For example, the
temperature of an oven will vary depending on the ambient temperature or
on whether the oven door is slightly open or tightly shut.

Ambient light interferes with the constant light level of a light intensity
control system at different times of the day.

These variable parameters are called disturbances of the controlled variable.

If the uniform and constant behavior of the controlled variable is important,
we need to measure and monitor it all the time and to vary the input signal
or the transfer function of the control system accordingly. We will expand
this subject in the following chapters, but for now, we need to explain two
associated terms.

SES
 13

A control system in which the controlled variable is not monitored and no

attention is paid to the disturbances of the controlled variable, is called an
Open Loop Control System. The transfer function is constant and we
assume that by changing the input signal the output signal (controlled
variable) will change accordingly.

A control system in which the controlled variable is measured and

monitored, and the transfer function or input signal is constantly corrected
to maintain a steady output level, is called a Closed Loop Control System.
The output signal or part of it is fed back to influence the input signal or
transmission function. This return signal is called Feedback.

A closed loop control system can be manually controlled. For example, we

switch OFF the lights in a room when there is daylight. This is essentially a
closed circuit control system in which we perform the feedback and
correction ourselves. Driving a car or manually irrigating a garden are also
closed loop control systems, which integrate the human element. There are
numerous similar examples, beyond the SESCOPE of this book. Here we
will deal with automatically controlled closed loop systems.

1.7 Dynamic response of control systems

In a control system, we govern and influence a controlled variable. This
variable responds and changes depending on a reference signal and time.
The time element is beyond our control. When we supply voltage to a
motor, it does not respond immediately with the corresponding output
speed. The motor accelerates and increases its speed gradually and then
stabilizes.

Heating and cooling systems usually respond very slowly. In contrast,

lighting systems respond much faster - their response is virtually immediate.

The stable condition of the system is called the Steady State. The steady
state values are important, but it is equally important to know how the
system reached this condition i.e., its response. The system response until it
stabilizes is called a Transient Effect or Transient Response.

SES
 14

The transient response of a control system is very important. To test a

system's transient response we supply a Step Signal, which looks like this:

High High

OR

Low Low
t1 t1

Figure 1-14

Up to time t1, the input signal is at one level. At t1, the reference voltage is
changed instantaneously. For example, let's assume that we provide an
input voltage V1=0V up to t1. At t1 we raise the input voltage to a high
value V2. The motor will accelerate until, after some time, it reaches the
speed, which corresponds with V2.

The response of such a system often looks like this:

Vi

Motor Input
Voltage

t
t1


Motor Speed

t
t1

Figure 1-15

We will expand on the subject of timing diagrams and the mathematical

calculation of timing diagrams in the following chapters.

SES
 15

A system, which responds slowly to a rapid change in the input voltage is

said to have a Time Lag. The controlled variable does not always respond
slowly. Sometimes it responds very rapidly but the sensor, which measures
it, responds slowly. For example, when we switch a heating element ON,
which must reach a certain temperature. The temperature and the ambient
air sensor must absorb a quantity of energy to reach the required
temperature. This process takes time especially if the heater also heats up
some of the surrounding air. In this case, we have a time lag between the
controlled variable (the temperature of the heating element) and the sensor.
This is also referred to as a Time Lag System.

In some systems, the controlled variable does not attempt to stabilize in a

steady state condition. Such a system does not have a steady state. For
example, a control system which operates a pump motor to fill up a water
tank. The flow capacity is a function of the motor speed. When the motor is
switched ON, it accelerates until it reaches a stable speed, and the flow
rate behaves accordingly, exactly as in figure 1-16. However, if the
controlled variable is the water level in the tank, then this quantity will
constantly increase and does not stabilize. Figure 1-15 shows the timing
diagram of this process.

SES
 16

Vi
Voltage to
Motor

t
t1

Motor Speed

t
t1
q
Flow
Capacity

t
t1
h
Level in
Tank

t
t1

Figure 1-16

The water level in the tank rises linearly when the flow capacity stabilizes.

SES
 17

1.8 Control system archetypes

Digital control systems take many forms. Most of them (today we can
really say all of them) are based on one kind of computer or another. The
major differences are in the programming methods. Some are based on
general-purpose computers such as the IBM PC and its compatibles.
Programming is in various languages such as PASCAL, C, BASIC etc.

There are industrial controllers based on dedicated computers such as a

PLC (Programmable Logic Controller), which are programmable in a
language, which is easy for industrial maintenance personnel to understand
(usually a language called Ladder Diagram). The PLC's inputs and
outputs are designed for industrial applications.

There are also totally dedicated industrial control systems. These systems
perform a more or less complex control function for a specific control
system. For example, greenhouse controllers, henhouse controllers, air
conditioning controllers, energy controllers etc.

All control systems have inputs and outputs. The output is a function of the
input and the transfer function. It may also be a function of itself - in this
case, we take a feedback signal from the output to the system input. The
clearest and easiest way to analyze processes and control phenomena is to
use a Control Diagram. Control diagrams describe something resembling
an analog computer in block form, according to its modules. The modules
compose the system's general control function.

For example, the following diagram:

Amplifier

Vi A = 1.5 Vo

Figure 1-17

In this diagram, amplification is 1.5 i.e., Vo = 1.5Vi. This is a simple

diagram. We will meet some more complex control diagrams later on.

SES
 18

This book will relate to a unique control system based on a PC. With this
system, we can compose different control diagrams on the computer screen.
We will define inputs and outputs. The computer will sample the inputs,
process them according to the control function described by the diagram on
the screen, and output the processed result. This is really an analog
computer realized digitally. Today this is the conventional method for
realizing analog computers and process control. This method also allows us
to store data, process it graphically, and analyze the various processes.

Study is based on the TPS-3011 Process Control training System, which

has been developed by SES.

SES
 19

Experiment 1.1 – Open Loop DC Motor P

Control

Objectives:
 Operating a DC motor with reaction to voltage reference.
 Measuring speed with a dynamo.
 Creating a mechanical load on a DC motor and testing the influence
on its speed (retarding the controlled parameter).
 The motor-dynamo reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:

1.1.1 DC motor control and rotation measurement

The TPS-3011 includes 2 variable voltage outputs called Vref1 and Vref2.
They are located in the SESLAB module. These voltages are controlled by
two potentiometers when the SESLAB is operated manually. These outputs
cannot drive consumers such as motors or lamps, so we will connect them
to the system power amplifier and the power amplifier will drive the
consumers.

The power amplifier output will be connected to the DC motor. Changes in

the voltage reference will cause changes of the voltage supplied to the DC
motor, and corresponding changes in the DC motor speed. We will observe
how the system reacts.

When a DC motor is driven mechanically, it acts as a dynamo. It creates a

voltage on its output terminals and can supply current. The motor, which is
driven by the analog output, is mechanically coupled so that it turns the
dynamo motor.

SES
 20

1.1.2 DC motor load response

We will connect the electrical output from the dynamo to the voltmeter in
the TPS-3011. The dynamo will supply voltage proportional to its rate of
rotation or in other words proportional to the speed of the DC motor.

A variable resistor and an ON/OFF switch are used as an electric variable

loads when they are connected in series to the dynamo output terminals and
the switch is ON. Turning the resistor changes the electrical load on the
dynamo.

More power is now required to turn the dynamo. The mechanical load on
the DC motor, which drives the system, increases or decreases relatively to
the position of the variable resistor. By pressing the pushbutton and
changing the position of the resistor, we create a variable disturbance of the
controlled parameter, which is the motor speed. We will observe how the
system reacts to this.

If the system is connected to a PC and in auto mode, it simultaneously

samples the input and output signals and automatically displays them
graphically.

1.1.3 DC motor speed response and status

A step signal is a signal, which changes all at once, with zero transition
time between one state and the other. Any system has a reaction time
including our motor. For control purposes, we need to know how the
controlled variable reacts to a step signal. We will apply such a signal to
the motor and observe the graphs, which show us how the motor reacts to
this signal.

Electrical measurements:
In the following experiments, we will measure the DC (Direct Current)voltages.
These measurements can be taken with a simple voltmeter.

The digital voltmeter is the modern solid state device capable of measuring the voltage
and displaying the value in digitized form with high accuracy. The panel voltmeter
measures the voltages and display calibrated values from negative to positive numbers
attached to LEDs representing each voltage value.

These measurements can also be taken with the SESCOPE software, asdescribed
in the preface. If you are going to use the SESCOPE software formeasuring, first
perform the exercise described in the preface beforeexecuting the following
experiments.

SES
 21

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output to the power amplifier input.

Step 4: Connect the power amplifier output to the motor M+ terminal.

Step 5: Connect the motor M- terminal to GND.

Power
SESLAB Motor
Amplifier
M+
Vref1 A M

M-

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the motor.

You can see that only on a certain voltage the motor starts to
rotate.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 11: Change Vref1 to maximum voltage.

SES
 22

Step 12: Raise the amplifier gain gradually, while measuring the amplifier
output voltage and stop when the voltage reaches 4.5V.

Step 13: Change Vref1 until the amplifier output voltage is 0V.

Step 14: Raise Vref1 slowly and find out on which voltage level at the
power amplifier output the motor starts rotating.

Record this voltage.

Step 15: Continue raising Vref1 and then lower it slowly.

Find out on which voltage of the power amplifier output the

motor stops and register it.

Is this voltage level the same as the one you measured at step 12,
smaller or bigger?

It should be smaller because the motor dynamic friction when it

rotates is smaller than the motor static friction when it does not
rotate.

Step 16: We will now measure the motor speed, using the dynamo.

Connect the D- terminal of the dynamo to GND.

Connect the D+ terminal of the panel voltmeter input.

SESLAB Power Motor Dynamo

Amplifier

Vref1 A M D

Voltmeter

Step 17: Raise and lower the Vref1 and observe how it affects the motor
speed and the dynamo voltage.

Step 18: Make a white mark on the rubber band, which couples the
dynamo to the DC motor.

SES
 23

Step 19: Count the number of revolutions per minute for different values
of power amplifier output voltage and especially for the dynamo
output voltage.

Step 20: Change the Vref1 step by step, measure (with a multitester) VP,
VD and , and fill in the following table:

VP Power 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Output
Voltage
VD Dynamo
Voltage
 Speed
RPM
Gm Motor
Gain

Note:

If you use the SESCOPE for measurements, you will find

oscillations on the dynamo signal voltage. It is because this
signal is a rectified AC voltage.

Step 21: Calculate the Gm for every step and add it to the table.

ω
Gm 
VP

Step 22: Draw your results in the following graph:


RPM

0 VP V
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

SES
 24

Step 23: We use the dynamo as a tachometer.

Connect the multitester to the dynamo terminals.

Change the Vref1 until VD (the dynamo voltage) is equal to 0.5V.

Step 24: Measure (count) the speed of the motor in RPM (Rounds Per
Minute).

Count rounds during 10 seconds and multiply it by 6.

Step 25: Continue this way and fill in the following table:

VD V 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

 PRM

This table can be used as a conversion table for measuring the

motor speed.

Step 26: Beneath the motor there is a load module with a variable resistor
(potentiometer) and an ON/OFF switch.

Connect the load terminals to the dynamo terminals.

SESESL Power Motor Dynamo

Amplifier
AB
Vref1 A M D

Voltmeter

Load

Step 27: Turn OFF the ON/OFF switch.

Turn the load potentiometer maximum counter clockwise. What

is the dynamo voltage?

Step 28: Change Vref1 until VD (the dynamo voltage) is 2.5V.

Step 29: Turn ON the ON/OFF switch.

What is the dynamo voltage now?

SES
 25

Step 30: Turn the load potentiometer gradually counter clockwise.

Turn the load potentiometer maximum clockwise. What happens

to the dynamo voltage?

The variable resistor is now connected in parallel to the dynamo

and loads the dynamo. An electrically loaded dynamo becomes a
mechanical load proportionally to the current, which is drawn
from it. The motor needs to work harder. Because we have not
changes the voltage supplying the motor, its speed will naturally
fall.

What this demonstrates is interference with the controlled

parameter, and how it affects the behavior of the system.

Smaller resistor means bigger electric load and higher current.

Step 31: Repeat steps 27-30 with different voltages of the dynamo.

Step signal reaction:

Step 32: Connect the power amplifier input to the Vref2 output instead of
the Vref1 output.

Step 33: Connect the SESCOPE CH1 probe to the amplifier output.

Step 34: Connect the SESCOPE CH2 probe to the dynamo output (D+).

Step 35: If you use the SESLAB SESCOPE and SESCOPE software, set
the time base to 50ms/cm.

Step 36: Change the LOAD ON/OFF switch to OFF position.

Step 37: Set Vref2 voltage to maximum.

Step 38: Observe the signals, which should be as the following:

Voltage to the motor

Dynamo voltage (speed)

SES
 26

Step 39: Change the LOAD ON/OFF switch to ON position.

Step 40: Rotate the LOAD potentiometer and draw the signals on various
loads.

SES
 27

Experiment 1.2 – Open Loop Light P

Control

Objectives:
 Operating a lamp in reaction to voltage reference.
 Measuring the light intensity with a light sensor.
 Variations in the ambient light and their influence on the measured
light.
 The lamp and light dynamic reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
1.2.1 Light control using lamp, amplifier and LDR
photo-resistor
In this experiment, as in the previous experiment we will connect the Vref1
of the SESLAB to the power amplifier input and the power amplifier
output to a lamp. In this way, rotation of the potentiometer will influence
the light intensity of the TPS-3011's incandescent lamp to vary.

A light sensor (LDR photo-resistor) is positioned facing the lamp. This

sensor absorbs light and reacts electronically. The sensor's electronic circuit
presents a voltage, which is in direct proportion to the intensity of the light
falling on the sensor.

Light reaches the sensor from the surrounding environment as well as from
the lamp, so that ambient light is effectively a disturbance of the controlled
variable, which is the light intensity. In this experiment, we will observe
these phenomena and analyze them.

The photoresistor decreases resistance with respect to receiving luminosity

on the component's sensitive surface. The resistance of a photoresistor
decreases with increase in incident light intensity; in other words, it
exhibits photoconductivity.
SES
 28

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output to the power amplifier input.

Step 4: Connect the power amplifier output to the lamp L+ terminal.

Step 5: Connect the lamp L- terminal to GND.

Power
SESLAB Lamp
Amplifier
L+
Vref1 A L

L-

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the lamp.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 11: Change Vref1 to maximum voltage.

Step 12: Raise the amplifier gain gradually, while measuring the amplifier
output voltage and stop when the voltage reaches 4.5V.

Step 13: change Vref1 until the amplifier output voltage is 0V.

SES
 29

Step 14: We will now measure the light intensity, using the photoresistor.

Connect the PT terminal of the panel voltmeter input.

Step 15: Raise and lower the Vref1 and observe how it affects the lamp
and the photoresistor voltage.

Step 16: Change the Vref1 step by step, measure (with a multitester) VP
and PPT, and fill in the following table:

VP Power 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Output
Voltage
VPT Phototransistor
Voltage
GPT Gain

Step 17: Calculate the GPT for every step and add it to the table.

VPT
G PT 
VP

Step 18: Draw your results in the following graph:

VPT V

0 VP V
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Step 19: Turn the Vref1 to maximum.

Step 20: Connect the multitester probes between the PT terminal and GND.

Measure the PT voltage.

Step 21: Insert a white paper between the photoresistor and the lamp.

What happens to the PT voltage?

SES
 30

Step 22: Insert another paper between the photoresistor and the lamp.

How it affects the PT voltage?

Step 23: Take out the papers between the lamp and the photoresistor.

Step signal reaction:

Step 24: Connect the power amplifier input to the Vref2 output instead of
the Vref1 output.

Step 25: Connect the SESCOPE CH1 probe to the amplifier output.

Step 26: Connect the SESCOPE CH2 probe to the lamp output (L+).

Step 27: If you use the SESLAB SESCOPE and SESCOPE software, set
the time base to 50ms/cm.

Step 28: Observe the signals, which should be as the following:

Voltage to the lamp

Dynamo voltage (speed)

The sensor reaction is fast, but the lamp reaction is slow.

SES
 31

Experiment 1.3 – Open Loop Temperature P

Control

Objectives:
 Operating an incandescent lamp in reaction to the voltage reference.
 Measuring temperature with a temperature sensor.
 The influence of ambient conditions on the temperature sensor.
 The lamp-thermistor reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:

1.3.1 Temperature control using lamp, amplifier and

NTC thermistor
In this experiment, as in the previous experiment we will connect the Vref1
of the SESLAB to the power amplifier input and the power amplifier
output to a lamp. In this way, rotation of the potentiometer will influence
the light intensity of the TPS-3011's incandescent lamp to vary.

A temperature sensor (thermistor) is positioned adjacent to the lamp. This

sensor absorbs heat from the lamp and reacts electronically. The sensor's
electronic circuit presents a voltage, which is in direct proportion to the
sensor temperature.

The environment also affects the thermistor. We will cool it down and
watch its reaction to this disturbance.

Another problem we may find. The thermistor is connected to an electronic

amplifier. This amplifier, also amplify changes in the source voltage. When
we turn the lamp ON and OFF it affects this voltage and the thermistor
output.
Thermistors are thermally sensitive resistors whose prime function is to exhibit
a large, predictable and precise change in electrical resistance when subjected
to a corresponding change in body temperature.
SES
 32

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output to the power amplifier input.

Step 4: Connect the power amplifier output to the lamp L+ terminal.

Step 5: Connect the motor L- terminal to GND.

SESESL Power Lamp

Amplifier
AB
L+
Vref1 A L

L-

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the lamp.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.
If you use the SESCOPE software, connect the amplifier output
to the CH1 input and perform the measuring as described in the
preface.

Step 11: Change Vref1 to maximum voltage.

Step 12: Raise the amplifier gain gradually, while measuring the amplifier
output voltage and stop when the voltage reaches 4.5V.

Step 13: change Vref1 until the amplifier output voltage is 0V.

SES
 33

Step 14: We will now measure the light intensity, using the photoresistor.

Connect the TH terminal of the panel voltmeter input.

Step 15: Raise and lower the Vref1 and observe how it affects the lamp
and the thermistor voltage.

Note:

The temperature reaction is very slow. You have to wait until

the voltage stabilized.

Step 16: Change the Vref1 step by step, measure (with a multitester) VP
and VTH, and fill in the following table:

VP Power 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Output
Voltage
VTH Thermistor
Voltage
GTH Gain

Step 17: Calculate the GTH for every step and add it to the table.

VTH
G TH 
VP

Step 18: Draw your results in the following graph:

VTH V

VP V
0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Step 19: Turn the Vref1 to maximum.

SES
 34

Step 20: Connect the multitester probes between the TH terminal and
GND.

Measure the TH voltage.

Step 21: Touch the thermistor with a metal ruler or a cold met al
instrument. Don’t push it.

How this affects the TH voltage?

Step signal reaction:

Step 22: disconnect the power amplifier from the lamp and let it cool for 2
minutes.

Step 23: Connect the power amplifier input to the Vref2 output instead of
the Vref1 output.

Step 24: Connect the SESCOPE CH1 probe to the amplifier output.

Step 25: Connect the SESCOPE CH2 probe to the lamp output (L+).

Step 26: If you use the SESLAB SESCOPE and SESCOPE software, set
the time base to 50ms/cm.

Step 27: Connect the power amplifier output to the lamp (L+).

Step 28: Observe the signals.

There is no reaction to the step signal because the lamp and the
thermistor do not have enough time to cool.

SES
 35

Experiment 1.4 – Open Loop Level P

Control

Objectives:
 Level control in reaction to voltage reference.
 Level measurement.
 The influence of the level drop on the system.
 The level control system in reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
A level control system is an accumulating system. It comprises a motor that
raises the level (a pump motor for example). Stopping the motor does not
restore the controlled variable (the level) as in light control, speed control
or temperature control.

The motor should rotate to the other direction in order to lower the level,
unless the tank is emptying from another direction.
The motor voltage determines the level raising rate.
The level control system in the TPS-3011 is an electronic system that
simulates level control.
The system includes a voltage input that simulates the voltage to the pump
motor. A positive voltage raises the level and a negative voltage lowers it.
The level is represented by a line of lights and by a voltage output.
The level control system has two modes of operation: fast and slow.
The system also includes two connecting points for connecting a drift
element that symbolizes the emptying of the thank
The drift element is the trainer LOAD resistor.

SES
 36

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Turn ON the trainer.

Step 4: Connect the Vref1 output to the power amplifier input.

Turn the gain potentiometer to maximum (10).

Step 5: Set the multimeter to measure DCV at 20V range.

Step 6: Connect the multimeter probes to the level control output (OUT).

Step 7: Connect the power amplifier output to the level control system
input.

SESESL Power Level Control

AB Amplifier

Vref1 IN Level

Step 8: Set the FAST/SLOW switch to FAST position.

Step 9: Change Vref1 up and down and observe how it affects the level.

Step 10: Connect the level output (OUT) to the trainer voltmeter input.

SESLAB Power Level Control Level Gauge

Amplifier

Vref1 IN
Level

OUT

Voltmeter

Step 11: Change Vref1 up and down and observe how it affects the level.

SES
 37

Step 12: Change the FAST/SLOW switch to SLOW position.

Step 13: Change Vref1 up and down and observe how it affects the level.

Step signal reaction:

Step 14: Connect the power amplifier input to the Vref2 output instead of
the Vref1 output.

Step 15: Connect the SESCOPE CH1 probe to the amplifier output.

Step 16: Connect the SESCOPE CH2 probe to the level control output
(OUT)

Step 17: If you use the SESLAB SESCOPE and SESCOPE software, set
the time base to 50ms/cm.

Step 18: Connect the power amplifier output to the level control system
input.

SESESL Power Level Control

AB Amplifier

Vref1 IN Level

Step 19: Although the level control system receives a step signal, it stucks
at the maximum level.

In order to lower it, we have to empty it.

Connect the DRIFT connecting points to the LOAD points.

Check that the LOAD switch is open, because a high resistor is

connected in parallel to the switch's terminals.

SES
 38

Step 20: Now you should have a reaction to the step signal.

Voltage in the Level Control input

Voltage in the Level Control output

Step 21: Change Vref2 and observe the system's reaction.

Step 22: Change the time base to 200ms/cm.

The wave frequency is lower and the level oscillations are higher.

Step 23: Draw and explain the accepted signals.

SES
 39

Chapter 2 – Bi-Stage Amplifier

2.1 Two position controller
Sometimes we require a system with a two state output signal (ON/OFF,
0V/5V etc.) depending on the level of the input. For example, a street
lighting system.

In a street lighting control system, a light sensor measures the ambient light.
When the light level drops below a certain value, the lights are switched
ON. They are switched ON all at once and not gradually. This system has
two operating states - ON and OFF. The light sensor is placed so that the
street lighting or direct sunlight will not affect it. When the sun shines
again, and the level of light ascends over the reference light value, the
lights are switched OFF.

The amplifier we use in this case is called a comparator. It has a very high
gain and has two comparator inputs.

- Ve
Vi A Vo
+

VR

Figure 2-1

VR is the reference voltage (V Reference). If Vi is greater than VR then we

have Ve < 0. Due to the high gain, Vo will drop to its lowest value, usually
0V or a negative value, the OFF condition.

If Vi < VR then Ve > 0 and, once again due to the high gain, Vo will rise to
its maximum value - the ON state.

The problem arises when Vi  VR. Due to fractional changes in the ambient
light level, which are totally acceptable, Vi will fluctuate around VR,
sometimes larger and sometimes smaller. Vo will change its condition from
ON to OFF and back again. Now imagine a lighting system with thousands
of lamps switching ON and OFF rapidly. This would quite obviously cause
havoc with the power supply and the condition of all the equipment.

SES
 40

There are many systems, which operate similarly: heating ovens,

refrigerators, air conditioners, ventilators, irrigation systems - to name a
few - which operate at full capacity or switch OFF according to the input
from different sensors.

2.2 Three range control with two position

controller
The solution to this problem is a Schmit Trigger amplifier. This amplifier
has a positive feedback from the output, in addition to a reference voltage.
This amplifier may be described like this:

- Ve
Vi A Vo
+
VR
+

Vo


Figure 2-2

To simplify the explanation, let's assign numerical values to the different

parameters. Let's assume for example that Vo has two conditions - 0V (OFF)
and 5V (ON).

Ve = VR + Vo - Vi
A  
 = 0.4
VR = 1V

Vo has two conditions depending on Vo:

1) Vo = 0  Vo = 0
2) Vo = 5V  Vo = 0.4 * 5 = 2V
If Vi < VR, independent of the condition of Vo, then:
Ve > 0
because:

Ve = VR + Vo - Vi
This will drive Vo to ON, i.e. Vo = 5V and Vo = 2V.

SES
 41

Now, so that Ve will be smaller than zero, Vi must increase and be greater
than 3V (VR + Vo = 1 + 2 = 3V).

If Vi > 3V, independent of the state of Vo, then:

Ve < 0

This will drive Vo to OFF, i.e. Vo = 0V and Vo = 0V.

In other words, for Ve to be positive again, Vi must drop below 1V

(VR + Vo = 1 + 0 = 1V).

The oscillation problem around Vi = VR which we discussed earlier, will

not appear here, and the borderline fluctuations in Vi will certainly not be
as large as 2V or below VR.

The system now has two threshold voltages - upper and lower (in our
example 3V and 1V). When the input voltage increases above the upper
threshold, and the system output condition changes, the threshold voltage
changes and goes low. In order for the system output condition to change
again, the input voltage must be lowered considerably so that it descends
through the lower threshold. This change is unlikely to take place simply
from random borderline fluctuations - a real change in the input value is
required.

The behavior of a Schmitt Trigger type amplifier can be described by the

following diagram:

Vo

VON

VOFF

Vi
V1 V2

Figure 2-3

SES
 42

a) When Vi < V1 (in our example 1V)  Vo = VON (5V).

b) When Vi increases the system will change its state (Vo = VOFF) only
when Vi > V2 (in our example 3V).

c) Vo will still remain OFF while Vi continues to increase (over V2).

d) If Vi decreases, the system will change its state (Vo = VON) only when
Vi < V1 (in our example 1V), and we find ourselves back at (a).

This diagram has two changeover points, which take care of the system's
immunity to "chatter" (noise) around the changeover points.

This is usually described as a changeover diagram with a hysteresis loop.

This sort of diagram was initially used to describe residual magnetism
effects in the behavior of electromagnetic devices.

There are ready made Schmitt Triggers available. They are often
symbolized like this:

Figure 2-4

In TPS-3011 there is a Schmitt Trigger amplifier, called Bi-Stage Amplifier,

with two potentiometers (L and W). The L potentiometer determines the VR
(the average of V1 and V2) and the W potentiometer determines the width
of the hystheresis (the difference between V2 and V1).

A Schmitt Trigger amplifier enables us to realize a closed loop control

system in the simplest possible way. The sensor is connected to its input
and its outputs to the controlled devices. Control will be performed
automatically. All we need to do is to set the input threshold values.

SES
 43

Examples:
a) Lighting Control – An ambient light sensor to control street lighting
or greenhouse lighting.

b) Temperature Control – A temperature sensor with an oven or other

heating devices.

c) Irrigation Control – A soil humidity sensor with a solenoid valve.

2.3 Thermostat as an area range 2-position

temperature control
An oven thermostat is a system of the Schmitt Trigger type. This is not an
amplifier, but rather a Schmitt Trigger regulator. It also has two changeover
points. It works on the bi-metal principle. Two metals with different
temperature – expansion coefficients are stuck together and assembled like
this:

Set Screw
Bi-Metal

Contacts

Voltage Heating
Source Spring Contact Element

Figure 2-5

When the system is cold, the bi-metal strip is straight and current flows
through the contacts and the heating element heats the oven. The bi-metal
strip bends upwards gradually as it heats up until the contact is broken. The
current through the element ceases and the oven begins to cool. It must
cool down some before the bi-metal strip unbends enough to remake the
contact.

The set screw lets us set how far the bi-metal strip must bend to break the
contact, and so sets the break and makes temperatures (which are not the
same).

SES
 44

The TPS-3011 Schmitt trigger amplifier:

The TPS-3011 includes a Schmitt trigger amplifier with two potentiometers.

One potentiometer (named L) determines the lower voltage reference V1.

The second potentiometer (names W) determines the hystheresis width:

V2  V1  W  L  W

SES
 45

Experiment 2.1 – Dual Stage Control of a DC

Motor

Objectives:
 Analyzing the behavior of the Dual stage amplifier.
 Using a Schmitt Trigger amplifier to control a DC motor.
 Speed control with a Bi-Stage amplifier and feedback.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment, we will analyze the Bi-Stage Amplifier. We will
connect its input to Vref and see how Vref affects its output. We will also
check how the parameters of the amplifier are affected by the L and W
potentiometers.

Vref M

Vref Bi-Stage Motor

Amplifier

Figure 2-6

The Vref also simulates a sensor in a control system.

Examples:

a) As a temperature sensor while the motor simulates an electric fan.

b) As a tank water level sensor while the motor simulates a pump.

SES
 46

In Bi-Stage Amplifier control system, we usually take care that the system
sensor should not be affected by the system output.

If we connect the tachometer to the Schmitt Trigger amplifier input, in

order to control its speed, we will get an oscillating system. The motor will
turn ON and OFF again and again.

Try to explain that.

Sometimes we use this oscillation in order to control the motor speed.

Instead of controlling the voltage of this motor, we turn it ON and OFF and
thus control the period time of the ON and OFF states. V 1 and V2 dictate
the motor speed.

SES
 47

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Turn ON the trainer.

Step 4: Connect the Vref1 output to the Bi-Stage Amplifier input.

Bi-Stage
SESESL
Amplifier
AB
Vref1

Step 5: Set the multitester to measure DCV at the 20V range.

Step 6: Connect the multitester probes to the Bi-Stage Amplifier output

and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 7: Turn the W potentiometer maximum counter clockwise.

The hystheresis width will be the smallest (Wmin condition).

Step 8: Turn the L potentiometer maximum counter clockwise (Lmin

condition).

Step 9: Raise Vref1 to the maximum.

The Bi-Stage Amplifier output will be close to –5V.

Step 10: Lower Vref1 slowly until the Bi-Stage Amplifier output change
to voltage close to +5V.

When it changes, it changes at once.

Step 11: Measure Vref1.

This is V1 of the Bi-Stage Amplifier.

SES
 48

Step 12: Connect the multitester probe back to the Bi-Stage Amplifier
output.

Step 13: Raise Vref1 slowly until the Bi-Stage Amplifier output voltage
changes back to –5V.

Step 14: Measure Vref1.

This is V2 of the Bi-Stage Amplifier.

Record V1 and V2 for Lmin and Wmin in the following table:

Lmin, Wmin Lmed, Wmin Lmed, Wmed Lmin, Wmed

V1
V2

Step 15: Turn the L potentiometer to the scale center (Lmed).

Step 16: Repeat steps 9-14 and write down the V1 and V2 of this state of
the potentiometer (Lmed, Wmin).

Step 17: Turn the W potentiometer to the scale center (Wmed).

Step 18: Repeat steps 9-14 and write down the V1 and V2 of this state of
the potentiometer (Lmed, Wmed).

Step 19: Turn the L potentiometer maximum counter clockwise (Lmin).

Step 20: Repeat steps 9-14 and write down the V1 and V2 of this state of
the potentiometer (Lmin, Wmed).

Step 21: Turn the W potentiometer maximum counter clockwise (Wmin).

Step 22: Connect the Bi-Stage Amplifier output to the motor M+ terminal.

Vref1 simulates a sensor now. Because we want to turn the motor

ON and OFF, according to Vref1 and because the output of the
Bi-Stage Amplifier changes between –5V to +5V, we will
connect the motor M- terminal to the –5V terminal instead to the
GND.

Think why.

SES
 49

Step 23: Connect the motor M- terminal to –5V.

Bi-Stage
SESLAB Motor
Amplifier

Vref1 M

-5V

Step 24: Change Vref1 up and down and observe how it affects the motor.

Step 25: Connect the D- terminal of the dynamo to GND.

Connect the D+ terminal of the panel voltmeter input.

Bi-Stage
SESLAB Motor Dynamo
Amplifier

Vref1 M D

D+ D-
-5V Voltmeter

Step 26: Raise and lower the Vref1 and observe how it affects the motor
speed and the dynamo voltage.

Step 27: Repeat step 26 with different positions of the L and W

potentiometers.

Step 28: Disconnect Vref1 from the Bi-Stage Amplifier.

Step 29: Turn the W potentiometer maximum counter clockwise (Wmin).

Step 30: Connect the D+ of the Dynamo to the Bi-Stage Amplifier input.

Bi-Stage
Motor Dynamo
Amplifier

M D

D-
-5V

SES
 50

Step 31: Change the L potentiometer until the motor turns ON.

Step 32: Observe with the SESCOPE the signals at the amplifier input
(coming from the dynamo) and at the amplifier output (goes to
the motor).

Step 33: Change the L and the W potentiometers and observe how it
affects the motor speed and the amplifier signals.

Step 34: Explain the system behavior.

SES
 51

Experiment 2.2 – Dual Stage Light Control

Objectives:
 Using a Schmitt Trigger type amplifier for light control.
 Light control using Bi-Stage amplifier and feedback.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment we will use a light sensor and the Bi-Stage Amplifier in
order to control the lamp.

+5V

R Dual stage Lamp

Amplifier

Light
Sensor

Figure 2-9

The light sensor is called a photoresistor, i.e. a light sensitive transistor. Its
resistance varies according to the light that falls on it. As more light falls on
it, its resistance drops and it drives a stronger current through the resistor R.
The voltage on the resistor rises. In other words, the voltage at the junction
PT increases proportionally to the level of light falling on the photoresistor.

SES
 52

In this experiment, we need to isolate the light from the lamp so that it does
not reach the photoresistor, otherwise the system will oscillate. When the
light level is lower than the lower threshold, the lamp will go ON. If the
lamp shines directly on the sensor, a high level of light (exceeding the
upper threshold) will be registered regardless of the ambient light. The
lamp will go OFF. As the ambient light level is still lower than the bottom
threshold, not having changed while this is happening, the lamp will go ON
again, and so on over and over.

Let's take a look at this phenomenon, which occasionally might even be

advantageous. For example, if we want to maintain an average light level.
If the ambient light is low, we add the light source. If the additional light
does not raise the level of light falling on the sensor, over the upper
threshold, there will not be any oscillations.

Because the photoresistor is against the lamp, we will use the Vref1 as a
simulation of a light sensor for experimenting the Bi-Stage Amplifier lamp
control,

SES
 53

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Turn ON the trainer.

Step 4: Connect the Vref1 output to the Bi-Stage Amplifier input.

Bi-Stage
SESESL
Amplifier
AB
Vref1

Step 5: Set the multitester to measure DCV at the 20V range.

Step 6: Connect the multitester probes to the Bi-Stage Amplifier output

and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 7: Turn the W potentiometer maximum counter clockwise (Wmin).

The hystheresis width will be the smallest.

Step 8: Turn the L potentiometer maximum counter clockwise (Lmin).

Step 9: Raise Vref1 to the maximum.

The Bi-Stage Amplifier output will be close to –5V.

Step 10: Lower Vref1 slowly until the Bi-Stage Amplifier output change
to voltage close to +5V.

When it changes, it changes at once.

Step 11: Measure Vref1.

This is V1 of the Bi-Stage Amplifier.

SES
 54

Step 12: Connect the multitester probe back to the Bi-Stage Amplifier
output.

Step 13: Raise Vref1 slowly until the Bi-Stage Amplifier output voltage
changes back to –5V.

Step 14: Measure Vref1.

This is V2 of the Bi-Stage Amplifier.

Step 15: Connect the Bi-Stage Amplifier output to the lamp L+ terminal.

Vref1 simulates a sensor now. Because we want to turn the lamp

ON and OFF, according to Vref1 and because the output of the
Bi-Stage Amplifier changes between –5V to +5V, we will
connect the lamp L- terminal to the –5V terminal instead to the
GND.

Think why.

Step 16: Connect the lamp L- terminal to –5V.

Bi-Stage
SESESL Lamp
Amplifier
AB
Vref1 L

-5V

Step 17: Change Vref1 up and down and observe how it affects the lamp.

Step 18: Connect the PT terminal of the panel voltmeter input.

Bi-Stage
SESESL Lamp Photoresistor
Amplifier
AB
Vref1 L

PT
-5V Voltmeter

Step 19: Raise and lower the Vref1 and observe how it affects the lamp
and the photoresistor voltage.

SES
 55

Step 20: Repeat step 19 with different positions of the L and W

potentiometers.

Step 21: Disconnect Vref1 from the Bi-Stage Amplifier.

Step 22: Turn the W potentiometer maximum counter clockwise (Wmin).

Step 23: Connect the PT terminal of the photoresistor to the Bi-Stage

Amplifier input.

Step 24: Change the L potentiometer until the lamp turns ON.

Step 25: Observe with the SESCOPE the signals at the amplifier input
(coming from the light sensor) and at the amplifier output (goes
to the lamp).

Step 26: Change the L and the W potentiometers and observe how it
affects the lamp and the signals.

Step 27: Explain the system behavior.

SES
 56

Experiment 2.3 – Dual Stage Temperature

Control

Objectives:
 Using a Schmitt Trigger amplifier for temperature control.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment, we repeat experiment 2.2, however this time with
temperature control instead of light control. There is one essential
difference between these two parameters. Light changes are very rapid.
When we switch ON a light, or darken a certain area or cast a shadow, the
response is immediate. A light sensor also responds immediately.

With temperature, the situation is entirely different. The heating element

must heat up the temperature sensor as well as its environment. The process
of heating and raising the temperature is radically slower. Closed loop heat
control with Bi-Stage Amplifiers is far more conventional than in light
control. A thermostat is essentially a Bi-Stage type device, and most
temperature control systems are based on one.

SES
 57

In this experiment, we will realize the following diagram:

NTC Dual Stage Lamp

Amplifier

Figure 2-10

A temperature sensor, called a thermistor is a resistor, which changes its

resistance depending on its temperature. There are two types of thermistors:
one type is called NTC (Negative Temperature Coefficient). The resistance
of this type of thermistor falls as the temperature rises. The second type is
called PTC (Positive Temperature Coefficient). With this type, the
resistance rises as the temperature rises. Our circuit uses a NTC type
thermistor.

The changes in the thermistor resistance corresponding with the

temperature variations are very small. We use a transistor to amplify these
changes. The voltage at junction TH will rise with the temperature.

The purpose of the system is to maintain an average temperature level, i.e.

the temperature will oscillate between the two threshold values, which is
determined by the L and W potentiometers.

SES
 58

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Turn ON the trainer.

Step 4: Connect the Vref1 output to the Bi-Stage Amplifier input.

Step 5: Set the multitester to measure DCV at the 20V range.

Step 6: Connect the multitester probes to the Bi-Stage Amplifier output

and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 7: Connect the Bi-Stage Amplifier output to the lamp L+ terminal.

Vref1 simulates a sensor now. Because we want to turn the lamp

ON and OFF, according to Vref1 and because the output of the
Bi-Stage Amplifier changes between –5V to +5V, we will
connect the lamp L- terminal to the –5V terminal instead to the
GND.

Think why.

Step 8: Connect the lamp L- terminal to –5V.

Bi-Stage
SESESL Lamp
AB Amplifier

Vref1 L

-5V

Step 9: Change Vref1 up and down and observe how it affects the lamp.

SES
 59

Step 10: Connect the TH terminal of the panel voltmeter input.

Bi-Stage
SESESL Lamp Thermistor
Amplifier
AB
Vref1 L

TH
-5V Voltmeter

Step 11: Raise and lower the Vref1 and observe how it affects the lamp
and the thermistor voltage.

Step 12: Repeat step 11 with different positions of the L and W

potentiometers.

Step 13: Disconnect Vref1 from the Bi-Stage Amplifier.

Step 14: Turn the W potentiometer maximum counter clockwise (Wmin).

Step 15: Connect the TH terminal of the thermistor to the Bi-Stage

Amplifier input.

Step 16: Change the L potentiometer very slowly until the lamp turns ON.

Step 17: Measure the voltage at the Bi-Stage amplifier input.

This voltage goes up slowly.

The lamp should turn OFF when this voltage passes the V2 value.

Step 18: Now the sensor voltage should goes down even slower.

Help it to drop faster by waving your notebook over it.

The lamp turns ON when the sensor voltage drops below V 1 and
vice versa.

Step 19: There is no use to expand the hystheresis width because this
process is slow enough.

SES
 60

Chapter 3 – Negative Feedback

3.1 Negative feedback in control system
In chapter 1, we discussed and realized an open loop control systems.
These are simple systems. Often we use open loop control systems when
we ourselves close the control loop. We observe or sense the controlled
variable and change the system's operation as necessary.

We switch the light ON in a dark room, and switch it OFF when there is
natural light. Sometimes we even use a dimmer to set just the right level of
light - strong light, romantic soft light etc. When it is hot, we turn ON a fan
or air conditioner and when it is cold, we operate a heater and so on. We
ourselves become part of the control, sensing, measuring and decision
making system.

The object of this book is to learn and analyze control systems, which
operate automatically and independently without manual intervention. In
these systems, some change must take place in the system (variations in
amplification or in the magnitude of the input signal) in order to stabilize
the controlled variable.

We have seen that in an open loop system, any interference with the
controlled variable causes it to vary, even though the input signal may be
steady. When it is important that the controlled variable remains steady and
constant, despite the interference, we need to add negative feedback.

SES
 61

The negative feedback in the system is a correcting factor. Let's look at the
following diagram, which describes a system with negative feedback. We
will relate to voltage, but obviously, the controlled signals may have any
other form (current, force, flow rate etc.).

+ Ve
Vi A Vo
-

Vb

Figure 3-1

Vo is the voltage at the output. A part of it is fed back to the input

negatively, i.e. it is subtracted from the input voltage. This parameter is
called Vb (Vback) and its value is:

Vb = Vo

The voltage, which is fed into the amplifier that drives Vo, is called Ve
(Verror) and is the difference between the input voltage Vi and the feedback
voltage Vb.

Ve = Vi - Vb = Vi - Vo

Vo is the product of the amplification of Ve.

Vo = AVe = A(Vi - Vo)

In this equation, Vo (the output value) appears in both sides of the equation.
We will isolate Vo to see the system equation. First, we open the
parentheses:

Vo = AVi - AVo

Now let's collect all the elements with Vo on the left:

Vo  βAVo  Vi  A
Vo (1  βA)  Vi  A

SES
 62

Thus the final system equation is:

A
Vo  Vi
1  A

This is the formula for the amplification of a negative feedback control

system. We will expand this concept in the next section.

From this formula we can see that A and  have certain finite values (A is
usually greater than 1 and  is usually smaller than 1). Ve will receive a
specific value as a function of Vi, A and .

Let's go back to the initial equation for a moment:

Vo  A  Ve  A(Vi  βVo )

If Vo changes for any reason, due to a disturbance of the controlled variable,

the system will bring it back to its set value. For example, if Vo tends to fall,
Vo will fall with it with the result that (Ve = Vi - Vo) increases,
consequently driving Vo back up again.

If Vo tends to rise, Vo will rise too causing the error voltage Ve to fall,
dragging Vo back down with it.

To summarize, systems with negative feedback are intended to create an

output signal, which behaves according to the general amplification
formula:

A
Vo  Vi
1  A

The denominator is greater than 1 and so Vo is smaller than its open loop
value, but at the same time is far more stable than its open loop value.

SES
 63

3.2 Analysis of the transfer function and the

amplification formula
The feedback coefficient b is usually smaller than 1. There is no point in
returning the full value of Vo or a greater value than Vo as feedback. This
would most likely result in a negative value of Ve causing the output
voltage to be inverted, resulting in positive feedback. Check this hypothesis.

The amplification A is usually greater than 1 and most often it is much

greater than 1.

In an open circuit system, A indicates the ratio between Vo and Vi. It is

difficult to control exact amplification especially in a system in which there
is an interference of the controlled variable, which affects the amplification
of the amplifier.

To achieve accurate and stable amplification in the system, we see to it that

the amplification A is as large as possible (even approaching infinity).  is
obtained from an accurate attenuator such as a resistor voltage divider for
example. The feedback system b does not have to contend with
interferences to its output variable as it is connected directly into the
amplifier's internal summation system (which presents a fixed load on b).

In cases where A >> 1 and A  , we will have:

A A 1
Vo  Vi  Vi  Vi
1  A A 

As  is smaller than 1 the result is that Vo is greater than Vi. The system
amplification is equal to β1 and does not depend on the accuracy of the
amplifier A. In other words, fluctuations in A (assuming that this value is
still very high) will not affect Vo.

We will supplement this analysis with a numerical example.

First let's assume an open loop amplifier with A = 20 and an input signal
Vi = 0.1V. In this case the output signal is:

Vo  A  Vi  20  0.1  2V

SES
 64

If the amplification was to drop by half (for example due to an interference

of the controlled variable), the output signal would also drop by half, to 1V.

Now let's assume a system with feedback in which A = 2000 and the input
signal is Vi = 0.1V. In this system  = 0.05. Let's calculate Vo:

A 2000 2000 2000

Vo  Vi  0.1  0.1  0.1  1.980
1  A 1  2000  0.05 1  100 101

Here too let's allow the amplification to drop by half (from 2000 to 1000).
We will have:

1000 1000
Vo  0.1  0.1  1.961
1  1000  0.05 51

When compared to the enormous change in the amplification, the change

on the output is negligible.

In all the analyses, we have done so far in this chapter we have used
voltage for demonstration purposes. Obviously, there are numerous
systems, in which the signals are not voltage. For example, an oil-fired
water heating system.

+ e Oil Flow c Water Tank

r Control Furnace
Regulator Temperature
-

b
Temperature
Sensor

Figure 3-2

Generally, we give the system signals the following labels:

c controlled variable.
r The reference signal which sets the system's operating level. This is
the input signal, which we have also called Vi.
b Feedback signal.
e error. The difference between the reference signal and the feedback
signal.

SES
 65

In the last example, the heating intensity (the oil flow) is the difference
between the reference signal and the feedback signal. The input signal
usually differs in its form and intensity from the output signal. The input to
output transfer function adapts them. As the comparator stage at the input
must compare similar signal forms, the feedback system must also adapt
the output signal to the input. This is also a kind of transfer system.

The following diagram generally describes these systems:

+ e
r G c
-

Figure 3-3

erb
c  Ce
b  Hc
c  G  e  G (r  b )  G (r  H  c )
c G r G Hc
c  G Hc G r
c(1  G  H )  G  r
G
cr
1  GH

The general transfer function is:

c G
T 
r 1  GH

This equation is similar to the general amplification equation we received

earlier.

SES
 66

We will demonstrate such a system.

Amplifier Motor
+ Ve Vm
Vr A=5 Gm = 10 
-

Dynamo
b Attenuator Tachometer
Vd
 = 0.4 Hd = 0.05

Figure 3-4

The error voltage (the difference between Vr and the feedback) is

transferred to an amplifier, which creates the voltage Vm. This voltage is
supplied to a motor, which turns its shaft at the speed , which is a
function of the voltage Vm. The motor shaft is coupled to a dynamo, which
generates voltage Vd, which is a function of the rotation speed w. This
voltage is attenuated b times to create the feedback voltage Vb.

The general transference G of the system is:

rad / sec
G  A  G m  5  10  50
V

The system feedback transference is:

H  Hd    0.05  0.4  0.02

The general transfer function T is:

c G 50 50 50
T      25
r 1  GH 1  50  0.02 1  1 2

In other words, for every volt at the input we will have a rotation speed of
rad
25 at the output.
sec

SES
 67

3.3 Gain and Proportional

We use the letters r and c to indicate the input (r) and output (c) signals,
because they might not be of the same form.

We can use the same letter when the signals are of the same form. For
example, the dynamo voltage represents the motor speed in voltage form.

In this case, we may call the input signal Yi and the output signal Yo.

The relationship between the output and the input signals behave according
the equation:

A
Yo  Yi
1  βA

Yo
The ratio Yi
size is called Gain and it equals:

Yo A

Yi 1  βA

Usually, as we will elaborate later, we prefer the opposite expression:

1  βA
Vi  Vo
A

We mark the 1A A size with P (Proportional) and it describes the relation
between Vi and Vo:

Vi  P  Vo

P is always smaller than 1 and thus we multiply it by 100 and express it in

percentage.

SES
 68

3.4 Steady state and steady state error (static

method)
All the calculations in the previous sections related to the steady state. The
general transference function relates to the steady state. It describes the
relationship between the reference signal and the controlled signal after the
system has stabilized.

c G G
T  or c  r 
r 1  GH 1  GH

In this section, we will look at transients.

Let's go back to the last example:

+ Ve Vm c
Vr A Motor 
- G

Vb

Vd
 Tachometer
H

Figure 3-5

G
c  Vr
1  GH
GH
Vb  c  H  Vr
1 GH
GH
Ve  Vr  Vb  Vr  Vr
1  GH
 GH   1  GH  GH 
Ve  Vr  1    Vr  
 1  GH   1  GH 
1
Ve  Vr
1  GH

We can see that in this system Ve has the same sign as Vr and is a part of it
(the denominator is always greater than 1). For Ve to approach 0 we must
create GH  .

SES
 69

As we saw in the previous section, we try to make A as high as possible so

that the output is a function of the feedback transference and to achieve
maximum stability. As the value Vm necessary to receive the required
rotation speed is a finite quantity, we will have a low Ve in this case.

There are systems in which the error in the steady state must be 0. The
error signal is fed into an amplifier, which creates the output signal. In a
system, which contains a pump to fill a water tank, the pump must be
stopped when the tank is full. For example:

+ Ve
Vr Amplifier Motor Pump
-

Vb
Level
Sensor

Figure 3-6

A system, which performs motion control (positioning), is similar in its

action to level control.

+ Ve Position
Vr Amplifier Motor X
-

Vb
Position
Sensor

Figure 3-7

In these systems, in the steady state, the motor must stop when the level or
position have reached the required point. So in these systems V e must be
zero in the steady state. i.e. Vb must be equal to Vr.

SES
 70

G
In a system like this, the transfer function cannot be T  . These
1  GH
systems are said to have an integrative nature.

The analysis we used to calculate the transfer function is not relevant in this
system due to the error signal being equated to 0 (e = 0). We will analyze
this case differently. In these systems in the steady state:

e=0  b=r

While on the other hand:

b  Hc
 r = b = Hc

c 1
 T 
r H

r
 c
H

SES
 71

3.5 Transient response, dynamic reaction and

time constant
In section 1.6, we discussed the dynamic response of a control system.
Before you go on, reread section 1.6. In section 1.6, we discussed a "Step
Signal" and the response of a "Time Lag System" to this signal. Let's now
elaborate this subject.
So far, in this chapter, the time factor has not come into play in any of the
conditions we have analyzed. We have only discussed the steady state. To
examine transient response we need to build an equation, which considers
the time dependency of the various system components. For some
components, we can assume that there is no time dependency. For example,
an amplifier or an attenuator. The response to a change in the input level, in
these components is virtually immediate:
Vo  A  Vi  Vo  AVi  0

or in a negative feedback system:

A A
Vo  Vi  Vo  Vi  0
1  A 1  A

The time factor does not appear in these equations.

The time factor appears in elements, which have a time lag such as
inductors, capacitors, a tank filling up with water, an electric pump (level
control), position control etc.
The equations which describe the relationship between the output signal,
the input signal and the time factor are called differential equations, as the
behavior of the controlled variable depends not only on the input variable,
but also on how it behaves over a period of time. We will explain this
statement with the following examples.
The differential equation for the controlled variable Y, which includes also
its derivative, will have the form:

dy
Y  Y
dt
or:
dy
Y  Y  0
dt

SES
 72

Y is the value which Y approaches after an infinite time. i.e. at the end of
the process (which is the steady state).
In the steady state

Y = Y
because there will not be any change in Y for any unit of time dt:

dy
dy  0  0
dt
 is the system's time constant. The variability of Y depends on .

3.5.1 Motor-Dynamo system behavior

Most of the systems with time lag act according to the above differential
equation.
Let's refer first to the motor-dynamo system, which receives a step signal
Vi in its input.

Vi A Motor Dynamo Vo

Figure 3-8
In steady state:
Y  VD  PVi

The output voltage is according to the dynamo rotation speed. The input
signal is a step signal voltage, which accelerates the motor and the dynamo.
Acceleration is derivative of the speed so here is derivative of the dynamo
voltage.
The motor-dynamo equation is as follows:
dVD
VD    PVi
dt
The equation in close loop system is similar with only one difference:
1  A
P
A

SES
 73

3.5.2 General differential equator and its solution

dy
Y  Y  0
dt

This is a differential equation, which includes the controlled variable Y and

its derivative. A solution to this type of equation is only possible when the
controlled variable has the following form:

Aekt

e is a special constant with the value 2.71828. Its derivative by the variable,
which appears in the exponent, will give us an expression having the same
form.

For example, if:

X = Aekt

Then the derivative of X will be:

dX
 KAe kt
dt

The differential equation for the controlled variable Y, which also includes
its derivative, will have the form:

dy
Y  Y
dt

or:

dy
Y  Y  0
dt

The solution to a differential equation of the sort described here is a

function describing the variance of Y as a function of time Y(t). In other
words, this equation should help us to establish a function, which can
predict Y at any given moment. The solution to such an equation is called
an Exponential Solution and the reaction of Y is called an Exponential
Reaction.

SES
 74

The general solution will be according to the following function:


t   
t
Y(t )  Yo e   Y 1  e  

 
 

Yo is the initial value of Y, at the start of the exponential reaction.

This is a general solution for all differential equations of the type:

dy
Y  Y
dt

We will begin by examining two important points in time:

t = 0 and t = 

When t = 0 we will have:

Y(t  0)  Yo e o  Y (1  e o )
eo  1
 Y(t  0)  Yo  1  Y (1  1)
 Y(t  0)  Yo

When t = i (in the steady state) we will have:

Y(t   )  Yo e   Y (1  e  )
e   0
 Y(t   )  Yo  0  Y (1  0)
 Y(t   )  Y

At least for these two extreme points in time the function is correct. We
will have to assume that it is also correct at any other time. We can
compare the systems measured behavior with our mathematical prediction.

Another important point in time is when t = .

SES
 75

 t
τ
At this point the value of the exponential variable e is:
t
 1
e   e 1   0.367
e
2
At this time, the controlled variable reaches about 3
of its steady state
value.

Let's go back to the motor-dynamo system for a moment. The dynamo

voltage function will be:
 t  t
VD (t )  VDO e 
 VD (1  e 
)

Assuming that the initial dynamo voltage is 0V:

VDO = 0

The steady state voltage of the dynamo is:

VD = PVi

The resulting equation is:

 t  t
VD (t )  0e 
 PVi (1  e 
)
 t
VD (t )  PVi (1  e 
)

The graphic description of this equation looks like this:

VD

PVi

Figure 3-9

The time constant of the system can be measured as the time when the
dynamo voltage reaches 23 of the maximum voltage.

SES
 76

We must emphasize that this last equation is a private case in which the
initial dynamo voltage is zero. In the general case in which there is some
initial dynamo voltage, the dynamo voltage will behave according to this
formula:
 t  t
VD (t )  VDO e 
 VD (1  e 
)

When the motor voltage drops to 0, the motor and the dynamo continue to
rotate and stop gradually.

We also get a differential equation with the solution:

While:

VDO  PVi
VD   0

We get:

 t  t
VD (t)  PVi e τ  0(1  e τ )
 t
VD (t)  PVi e τ

Finally, the dynamo voltage reaction to a step signal is as follows:

Voltage to the motor

Dynamo voltage (speed)

Figure 3-10

The motor-dynamo system equation also includes a second degree

derivative, which will be discussed in the next chapter.

SES
 77

3.5.3 Level control concordant with capacitor

charging
The level control system that operates in a W.C. flushing tank also acts as
an exponential system. In the latter system, the water inlet valve is
controlled by a lever with a float on its end. When the tank is empty the
float is down and the valve is fully open.

As the tank fills up, the float rises and the valve closes according to the
level in the tank and the water inlet rate drops. Exactly like charging a
capacitor. As soon as the water reaches the required level, the valve shuts
OFF completely.

The water level reaction is exactly the same as charging the capacitor:

  
t
H ( t )  H 1  e  

 
 

Our problem is to define the final water level and the system time
coefficient (). We could of course do this empirically. The final water
level is easy to find - we just measure the water level when the flow has
stopped. We can find  by marking a line at 23 of the maximum level and
measuring the time required to fill the empty tank to the 23 mark.

In a water tank, the rate at which the level drops (the flow rate) changes as
a function of the water level. Note that in this case the tank discharges
through a valve, which is constant and not variable like in our flushing tank
example.

Figure 3-11

SES
 78

Here too we will have an exponential reaction like the capacitor:

t

h(t )  hoe 

3.5.4 Position control

Let's take a look at a diagram which describes a position control system:

+ Ve Vm
Vr Amplifier Motor Position X
-

Vb

Tachometer

Figure 3-12

In this system, a motor is coupled to some kind of transmission, which

performs motion. This could be the motion of a robot arm, conveyor, crane,
elevator etc.

The controlled body is linked to a potentiometer, which translates its

position into voltage. As soon as the voltage from the potentiometer V b,
equals the reference voltage Vr, then the error voltage Ve will be zero and
the motor will stop.

The reaction of this system resembles the reaction of level control. As the
distance increases, so Vb increases, and Ve decreases and so the motor
rotates more slowly and the rate of motion decreases.

In this case, too, we will have an exponential reaction according to the

formula:

  
t
X(t )  X 1  e  

 
 

In motion control, we might conceivably never reach the final value.

SES
 79

The motor and transmission have friction and torque coefficients, which are
highest when the speed is lowest. It is likely that at some value of V b lower
than Vr, Ve will be so small that the motor cannot turn and so it will stop
before the required point is reached.

In motion control, there is an additional possibility that the system might

oscillate. The motor and transmission have a certain inertia. If the system
amplification is high, it is possible that the motor speed is such that when
Ve is zero the motor will continue to turn and the system might overshoot
the required distance.

3.5.5 Temperature control

Temperature control also resembles charging a capacitor. The rate at which
a body heats up is a function of the difference between the body
temperature and the heat source. As this difference decreases, the body
heats up more slowly. Here too we will see an exponential reaction:

  
t
T(t )  T 1  e 
 
 
 

+ Ve Heating
Vr Amplifier Temperature
- Element
of the
Controlled
Body
Vb
Temperature

Sensor

Figure 3-13

SES
 80

3.5.6 Summary
Charging and discharging of a component with an exponential reaction
behave according to the general equation:


t   
t
Y(t )  Yoe   Y 1  e  

 
 

The only difference is in the values associated with Yo and Y.

To test a component's reaction, we supply it with a step signal or a pulse, as

described in the following graphs showing an exponential reaction:

Vi Vi Vi

t t t

Vo Vo Vo

t t t

Positive Step Negative Pulse

Signal Step Signal

Figure 3-14

SES
 81

Experiment 3.1 – DC Motor Closed Loop

Control with Negative Feedback

Objectives:
 Repeating the open loop control experiment in order to compare it
with a closed loop reaction.
 Realizing a negative feedback control diagram to control a DC motor.
 Changing the amplification of the amplifier and the feedback and
testing their influences.
 Disturbing the controlled variable and checking the stability of the
system.
 Checking the steady state and the steady state error.
 The dynamic reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
3.1.1 DC motor control response and status
In this experiment, we will first revise open loop control of the DC motor
briefly. We will make the following connections:

Power
Amplifier Motor
Vm
Vr A Gm 

Vref1
Dynamo
Tachometer
Vd
Hd

Figure 3-15

SES
 82

The objective of this experiment is to repeat the measurement of Gm and Hd

as we did in the experiment in chapter 1. Vref1 is connected to the power
amplifier and the power amplifier drives the motor according to Vref1. The
motor translates its voltage into rotational speed . The tachometer
translates the rotational speed  into feedback voltage Vd, which is
measured by the voltmeter. We will find that this feedback voltage is
approximately 0.6Vm (60% of the motor voltage).

We will add an electrical load to the dynamo, which is converted into a

mechanical load and we will see how the dynamo voltage Vd falls
drastically while the output voltage Vm remains constant.

Now we will close a negative feedback loop in the system, as in figure

3-16.

Power
Amplifier Motor
+ Ve Vm
Vr A=1 Gm 
-
Feedback Dynamo
Amplifier Tachometer
Vb Vd
=1 Hd

Figure 3-16

At this stage we will use amplification A=1 and feedback amplification

=1. We will calculate the general amplification formula. We will run the
system and test its reaction to disturbance of the controlled variable.

SES
 83

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output via the adder IN1 to the power amplifier input.

Step 4: Connect the power amplifier output to the motor M+ terminal.

Step 5: Connect the motor M- terminal to GND.

Adder
IN1 Power
Vref1 Motor
Amplifier
+ Out M+
IN2 A M
-

+ M-
IN3 -

The Adder does not affect the system because there are no other
signals besides Vref1. Its output is equal to its IN1 input.

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the motor.

You can see that only on a certain voltage the motor starts to
rotate.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

SES
 84

Step 11: Lower Vref1 to 0V.

Note that Vref1 changes from –5V to +5V.

Step 12: Raise Vref1 slowly and find out at which voltage level the motor
starts rotating.

Step 13: Continue raising Vref1 and then lower it slowly.

Find out on which voltage of the power amplifier output the

motor stops.

Is this voltage level the same as the one you measured at step 12,
smaller or bigger?

It should be smaller because the motor dynamic friction when it

rotates is smaller than the motor static friction when it does not
rotate.

Step 14: We will now measure the motor speed, using the dynamo.

Connect the D- terminal of the dynamo to GND.

Connect the D+ terminal of the panel voltmeter input.

SESESL Adder
AB IN1
Vref1 Power Dynamo
Motor
Amplifier

+ Out
IN2 A M D
-

IN3 +
-
Voltmeter

Step 15: Raise and lower the Vref1 and observe how it affects the motor
speed and the dynamo voltage.

Step 16: Make a white mark on the rubber band, which couples the
dynamo to the DC motor.

Step 17: Count the number of revolutions per minute for different values
of power amplifier output voltage and especially for the dynamo
output voltage.

SES
 85

Step 18: Set the power amplifier gain to 1.

Step 19: We use the dynamo as a tachometer.

Connect the multi-tester to the dynamo terminals.

Change the Vref1 until VD (the dynamo voltage) is equal to 0.5V.

Step 20: Measure (count) the speed of the motor in RPM.

Step 21: Continue this way and fill in the following table:

VD V 0.5 1.0 1.5 2.0 2.5 3.0

 PRM

This table can be used as a conversion table for measuring the

motor speed.

When needed, raise the power amplifier gain to reach the

dynamo high voltages.

Step 22: Beneath the motor there is a load module with a variable resistor
(potentiometer) and an ON/OFF switch.

Connect the load terminals to the dynamo terminals.

SESESL Adder
AB IN1
Vref1 Power
Motor Dynamo
Amplifier

+ Out
IN2 A M D
-

IN3 +
-
Voltmeter

Load

Step 23: Turn OFF the ON/OFF switch.

Turn the load potentiometer maximum counter clockwise.

Step 24: Change Vref1 until VD (the dynamo voltage) is 2.5V.

SES
 86

Step 25: Turn ON the ON/OFF switch.

What is the dynamo voltage?

Step 26: Turn the load potentiometer counter clockwise.

Turn the load potentiometer clockwise almost to the maximum.

What happens to the dynamo voltage?

The variable resistor is now connected in parallel to the dynamo

and loads the dynamo. An electrically loaded dynamo becomes a
mechanical load proportionally to the current, which is drawn
from it. The motor needs to work harder. Because we have not
changes the voltage supplying the motor, its speed will naturally
fall.

What this demonstrates is interference with the controlled

parameter, and how it affects the behavior of the system.

Smaller resistor means bigger electric load, higher current and

higher mechanical load.

Step 27: Repeat steps 23-26 with different voltages of the dynamo.

Step 28: Turn Vref1 maximum clockwise.

Step 29: Measure the power amplifier output voltage.

Step 30: Lower the power amplifier to 1 and raise its gain slowly until this
voltage is equal to 3.5V.

Step 31: Turn OFF the Load ON/OFF switch.

Step 32: Measure the dynamo voltage and record it.

Step 33: Turn the load potentiometer ¾ of its path clockwise.

Step 34: Turn ON the Load ON/OFF switch.

Step 35: Record the dynamo voltage. It should drop more than 0.5V.

Step 36: Turn OFF the Load ON/OFF switch.

SES
 87

Step 37: We will close the loop now.

Connect the D+ of the dynamo to IN2 of the adder.

SESESL Adder
AB IN1
Vref1 Power Motor Dynamo
Amplifier
IN2 + Out
A M D
-

IN3 +
-
Voltmeter

Load

Step 38: Turn the IN2 switch to position order to get a negative feedback.

Step 39: Observe the following:

The power amplifier input voltage will drop and so its output.

The motor rotates slower.

The dynamo voltage decreases.

Step 40: Increase the power amplifier gain until the power amplifier
output voltage return to 3.5V.

Step 41: Measure again the Dynamo voltage.

It will return to the voltage you recorded on step 32.

It may oscillate a little because of the closed loop.

Step 42: Turn ON the Load ON/OFF switch.

The Dynamo voltage will drop a little but less than without the
negative feedback.

Step 43: Measure the power amplifier output voltage.

It increases.

Explain why.

SES
 88

Step signal reaction:

Step 44: Change the load ON/OFF switch to OFF position.

Step 45: Turn ON your oscilloSESCOPE or use the SESCOPE software

as described in the preface.

Step 46: Set the trigger level to 1.2V and the trigger mode to Fall.

Step 47: Set the time base to 50ms/cm.

Step 48: Connect the Vref2 output to the IN1 input instead of Vref1.

Step 49: Disconnect the dynamo output from the IN2 input.

Step 50: Connect the CH2 probe to the dynamo output.

SESESL Adder
AB IN1
Vref2 Power
Motor Dynamo
Amplifier

+ Out
IN2 A M D
-

IN3 +
-
Voltmeter

Load

Step 51: Lower the power amplifier gain to 1 and raise Vref2 to maximum.

Step 52: Set the SESCOPE to Free Run.

You should get signals as follows:

Voltage to the motor

Dynamo voltage (speed)

SES
 89

Step 53: Measure with the cursors the time in which the dynamo voltage
reaches 0.7 of the maximum voltage.

This is the time constant of the system.

Step 54: Turn the load switch to ON position.

Step 55: Change the load potentiometer and check how it affects the
dynamo voltage.

Step 56: Turn the load switch to OFF position.

Step 57: Connect the dynamo output back to IN2 input.

SESESL Adder
AB IN1
Vref2 Power
Motor Dynamo
Amplifier
IN2 + Out
A M D
-

IN3 +
-
Voltmeter

Load

Step 58: Observe the signals.

The dynamo voltage is decreased because of the negative voltage.

The difference voltage between the voltage reference and the

feedback voltage is the error voltage, which is amplified by the
power amplifier in order to produce the motor voltage.

Step 59: Raise the power amplifier gain gradually.

You can see that the dynamo voltage increases and comes closer
to the voltage reference.
Step 60: Turn the load switch to ON position.
Step 61: Change the load potentiometer and check how it affects the
dynamo voltage.
Step 62: Turn the load switch to OFF position.

SES
 90

Experiment 3.2 – Light Closed Loop Control

with Negative Feedback

Objectives:
 Repeating the open loop light control experiment in order to compare
it with a closed loop reaction.
 Realizing a negative feedback control diagram to control light.
 Changing the amplification of the amplifier and the feedback and
testing their influences.
 Disturbing the controlled variable and checking the stability of the
system.
 Checking the steady state and the steady state error.
 The dynamic reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment, we will first revise open loop light control briefly. We
will make the following connections:

Power
Amplifier Lamp
Vl
Vr A Gl Light

Vref1

Photoresistor
Vp
Hp

Figure 3-17

SES
 91

The objective of this experiment is to repeat the measurement of G l*Hp

because we cannot measure Gl and Hp separately. Vref1 is connected to the
power amplifier and the power amplifier drives the lamp according to
Vref1. The lamp translates this voltage into light. The photoresistor
translates the light into feedback voltage Vp, which is measured by the
voltmeter.

We will make a disturbance to the light and we will see how it affects the
photoresistor voltage.

Now we will close a negative feedback loop in the system, as in figure

3-18.

Power
Amplifier Lamp
+ Vl
Vr A=1 Gl 
-
Feedback
Amplifier Photoresistor
Vp
=1 Hp

Figure 3-18

At this stage we will use amplification A=1 and feedback amplification

=1. We will calculate the general amplification formula. We will run the
system and test its reaction to disturbance of the controlled variable.

SES
 92

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output via the adder IN1 to the power amplifier input.

Step 4: Connect the power amplifier output to the lamp L+ terminal.

Step 5: Connect the motor L- terminal to GND.

Adder
IN1 Power
Vref1 Lamp Photoresistor
Amplifier
+ Out L+
IN2 A L
-

+ L-
IN3 -

Voltmeter

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the lamp.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

Step 11: Lower Vref1 to 0V.

Note that Vref1 changes from –5V to +5V.

Step 12: We will now measure the light intensity, using the photoresistor.

Connect the PT terminal of the panel voltmeter input.

SES
 93

Step 13: Raise and lower the Vref1 and observe how it affects the lamp
and the photoresistor voltage.

Step 14: Set the power amplifier gain to 1.

Step 15: Turn the Vref1 until the power amplifier output voltage is equal
to 3.5V.

Step 16: Connect the multitester probes between the PT terminal and GND.

Measure and record the PT voltage.

Step 17: Insert a white paper between the photoresistor and the lamp.

The PT voltage should drop significantly.

Step 18: Take out the paper.

Step 19: We will close the loop now.

Connect the PT terminal to IN2 terminal of the Adder.

Adder
IN1 Power
Vref1 Lamp Photoresistor
Amplifier
IN2 + Out L+
A L
-

+ L- PT
IN3 -

Step 20: Turn the IN2 switch to position order to get a negative feedback.

Step 21: Observe the following:

The power amplifier input voltage drops and so its output.

The PT voltage decreases because the light drops a little.

Step 22: Turn the Vref1 to maximum and raise the power amplifier gain
until the power amplifier output voltage return to 3.5V.

SES
 94

Step 23: Connect the multitester probes between the PT terminal and
GND.

Measure the PT voltage.

It should equal the value measured in step 16.

Step 24: Insert a white paper between the photoresistor and the lamp.

The PT voltage drops, but less than in an open loop system as in

step 17.

Step 25: Measure the power amplifier output voltage.

It raises.

Why?

Step 26: Take off the paper.

Step signal reaction:

Step 27: Change the load ON/OFF switch to OFF position.

Step 28: Turn ON your oscilloSESCOPE or use the SESCOPE software

as described in the preface.

Step 29: Set the trigger level to 1.2V and the trigger mode to Fall.

Step 30: Set the time base to 50ms/cm.

Step 31: Connect the Vref2 output to the IN1 input instead of Vref1.

Step 32: Disconnect the photoresistor output PT from the IN2 input.

Step 33: Connect the CH2 probe to PT (the photoresistor output).

Step 34: Lower the power amplifier gain to 1 and raise Vref2 to maximum.

SES
 95

Step 35: Set the SESCOPE to Free Run.

You should get signals as follows:

Voltage to the lamp

Voltage from the light

sensor (intensity)

This reaction is because of the lamp reacting and not because of

the photoresistor reaction.

Step 36: Insert a white paper between the lamp and the sensor and check
its influence on the sensor's voltage.

Step 37: Take off the paper.

Step 38: Connect the sensor output PT to IN2.

Adder
IN1 Power
Vref1 Lamp Photoresistor
Amplifier
IN2 + Out L+
A L
-

+ L- PT
IN3 -

Step 39: Observe the signals.

The light sensor voltage is decreased because of the negative

voltage.

The difference voltage between the voltage reference and the

feedback voltage is the error voltage, which is amplified by the
power amplifier in order to produce the lamp voltage.

SES
 96

Step 40: Raise the power amplifier gain gradually.

You can see that the light sensor voltage increases and comes
closer to the voltage reference.

Step 41: Insert a white paper between the photoresistor and the lamp.

Check the effect on the sensor voltage and the lamp intensity.

SES
 97

Experiment 3.3 – Temperature Control with

Negative Feedback

Objectives:
 Repeating the open loop temperature control experiment in order to
compare it with a closed loop reaction.
 Realizing a negative feedback control diagram to control temperature.
 Changing the amplification of the amplifier and the feedback and
testing their influences.
 Disturbing the controlled variable and checking the stability of the
system.
 Checking the steady state and the steady state error.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment, we will first revise open loop temperature control
briefly. We will make the following connections:

Power
Amplifier Lamp
Vl Temperature
Vr A Gl

Vref1

Thermistor
Vt
Ht

Figure 3-19

SES
 98

The objective of this experiment is to repeat the measurement of G l*Ht

because we cannot measure Gl and Ht separately. Vref1 is connected to the
power amplifier and the power amplifier drives the lamp according to
Vref1. The lamp translates this voltage into temperature. The thermistor
translates the temperature into feedback voltage Vt, which is measured by
the voltmeter.

We will make a disturbance to the temperature and we will see how it

affects the thermistor voltage.

Now we will close a negative feedback loop in the system, as in figure

3-20.

Power
Amplifier Lamp
Ve Vl Temperature
Vr + A=1 Gl
-
Feedback
Amplifier Thermistor
Vb Vt
=1 Ht

Figure 3-20

At this stage we will use amplification A=1 and feedback amplification

=1. We will calculate the general amplification formula. We will run the
system and test its reaction to disturbance of the controlled variable.

SES
 99

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output via the adder IN1to the power amplifier input.

Step 4: Connect the power amplifier output to the lamp L+ terminal.

Step 5: Connect the motor L- terminal to GND.

Adder
IN1 Power
Vref1 Lamp Thermistor
Amplifier
+ Out L+
IN2 A L
-

+ L-
IN3 -

Voltmeter

Step 6: Turn ON the trainer.

Step 7: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 8: Change Vref1 and observe how it affects the lamp.

Step 9: Set the voltmeter to DCV at the 20V range.

Step 10: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

Step 11: Lower Vref1 to 0V.

Note that Vref1 changes from –5V to +5V.

Step 12: We will now measure the temperature, using the thermistor.

Connect the TH terminal of the panel voltmeter input.

SES
 100

Step 13: Raise and lower the Vref1 and observe how it affects the lamp
and the thermistor voltage.

Step 14: Set the power amplifier gain to 1.

Step 15: Turn the Vref1 until the power amplifier output voltage is equal
to 3.5V.

Step 16: Connect the multitester probes between the TH terminal and
GND.

Measure and record the TH voltage.

Step 17: Touch the thermistor with a cold matter.

The TH voltage should drop.

Step 18: Leave the thermistor.

Step 19: We will close the loop now.

Connect the TH terminal to IN2 terminal of the Adder.

Adder
IN1 Power
Vref1 Lamp Thermisotr
Amplifier
IN2 + Out L+
A L
-

+ L- TH
IN3 -

Step 20: Turn the IN2 switch to position order to get a negative feedback.

Step 21: Observe the following:

The power amplifier input voltage drops and so its output.

The TH voltage decreases because the light drops a little.

Step 22: Turn the Vref1 to maximum and raise the power amplifier gain
until the Power amplifier output voltage return to 3.5V.

SES
 101

Step 23: Connect the multitester probes between the TH terminal and
GND.

Measure and record the TH voltage.

It should equal the value measured in step 16.

Step 24: Touch the thermistor with a cold matter.

The TH voltage drops, but less than in an open loop system.

Step signal reaction:

Step 25: Change the load ON/OFF switch to OFF position.

Step 26: Turn ON your oscilloSESCOPE or use the SESCOPE software

as described in the preface.

Step 27: Set the trigger level to 1.2V and the trigger mode to Fall.

Step 28: Set the time base to 50ms/cm.

Step 29: Disconnect the thermistor output TH from the IN2 input.

Step 30: Connect the Vref2 output to the IN1 input instead of Vref1.

Step 31: Connect the CH2 probe to TH (the thermistor output).

Step 32: Lower the power amplifier gain to 1 and raise Vref2 to maximum.

Adder
IN1 Power
Vref2 Lamp Thermistor
Amplifier
+ Out L+
IN2 A L
-

+ L-
IN3 -

Voltmeter

SES
 102

Step 33: Observe the signals.

The sensor voltage increases, but we cannot observe the reaction

to the step signal.

The reaction is too slow.

Step 34: Connect the sensor output TH to the IN2 input.

Adder
IN1 Power
Vref2 Lamp Thermisotr
Amplifier
IN2 + Out L+
A L
-

+ L- TH
IN3 -

Step 35: Observe the signals and explain them.

SES
 103

Experiment 3.4 – Level Closed Loop Control

with Negative Feedback

Objectives:
 Repeating the open loop level control experiment in order to compare
it with a closed loop reaction.
 Realizing a negative feedback control diagram to control temperature.
 Changing the amplification of the amplifier and the feedback and
testing their influences.
 Disturbing the controlled variable and checking the stability of the
system.
 Checking the steady state and the steady state error.
 The dynamic reaction to a step signal.

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:
In this experiment, we will first revise open loop temperature control
briefly. We will make the following connections:

Power
Level
Amplifier
Control
Vl
Vr A Gl Level

Vref1

Level Meter
Vl
Ht

Figure 3-21

SES
 104

The objective of this experiment is to repeat the measurement of

G1  H t because we cannot measure Gl and Ht separately. Vref1 is
connected to the power amplifier and the power amplifier drives the lamp
according to Vref1. The level control system raises the level and the level
meter translates the level into feedback voltage VG, which is measured by
the voltmeter.

We will make a disturbance to the level by creating a drift and we will see
how it affects the level meter voltage.

Now we will close a negative feedback loop in the system, as in figure

3-22.

Power Level
Amplifier Control
+ Ve Vl
Vr A=1 Gl Level
-
Feedback
Amplifier Level Meter
Vb Vl
=1 Ht

Figure 3-22

At this stage we will use amplification A=1 and feedback amplification

=1. We will calculate the general amplification formula. We will run the
system and test its reaction to disturbance of the controlled variable.

SES
 105

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref1 output via the adder IN1 to the power amplifier input.

Step 4: Connect the power amplifier output to the IN terminal of the

level control system.

Adder
IN1 Power Level Level
Vref1 Control Meter
Amplifier
+ Out IN
IN2 A L
-

+ Out
IN3
- Voltmeter

Step 5: Turn ON the trainer.

Step 6: Set the power amplifier gain potentiometer to 1.

At this position, the output voltage of the power amplifier is

equal to its input voltage.

Step 7: Turn the FAST/SLOW switch to SLOW position.

Step 8: Change Vref1. Raise it and lower it.

The positive voltage will raise the level and 0V will keep it at its
level.

Step 9: Repeat step 8 with negative voltage and observe the system's
behavior.

Step 10: Set the voltmeter to DCV at the 20V range.

SES
 106

Step 11: Connect the voltmeter's probes between the outputs of the power
amplifier and GND.

If you use the SESCOPE software, connect the amplifier output

to the CH1 input and perform the measuring as described in the
preface.

Step 12: Lower the level to 0.

Step 13: Measure and record the voltage at the OUT terminal.

Step 14: Raise the level in one degree.

Measure and record the OUT voltage.

Step 15: Repeat step 14 and record the voltages for each level.

Level Voltage
01
9
8
7
6
5
4
3
2
1
0

Step 16: We will close the loop now.

Connect the OUT terminal to IN2 terminal of the Adder.

Adder
IN1 Power Level Level
Vref1 Control
Amplifier Meter
IN2 + Out IN
A L
-

+ Out
IN3 -

SES
 107

Step 17: Turn the IN2 switch to +1 to get a negative feedback (the OUT
voltages are reverse ones).

Step 18: change Vref1 and you will see that the level changes accordingly.

Step signal reaction:

Step 19: Set the trigger level to 1.2V and the trigger mode to Fall.

Step 20: Set the time base to 50ms/cm.

Step 21: Connect the Vref2 output to the IN1 input instead of Vref1.

Step 22: Disconnect the level OUT terminal from the IN2 input.

Step 23: Connect the CH1 probe to the level IN terminal.

Step 24: Connect the CH2 probe to the level OUT terminal.

Step 25: Observe the signal and check if you got the following picture.

Voltage to the level Input

Level meter voltage

Step 26: Observe the level behavior.

Step 27: Change the FAST/SLOW switch to FAST position.

Explain the system's behavior.

Step 28: Change the time base to 200ms/cm.

Step 29: Observe and explain the system's behavior.

SES
 108

Chapter 4 – Second Degree Systems

and PID
Experiment 4.1 – Second Degree Systems
and PID

Objectives:
 Second degree systems
 PI control
 PID control

Equipment required:
 TPS-3011
 Voltmeter
 Banana wires
 PC + SESCOPE software

Discussion:

4.1.1 Second order systems

The systems described in chapter 3 are called first order systems. Their
reaction equation includes the controlled variable and its first order
derivative ( dy
dt ).

In these systems (especially systems with amplification equals to 1) there is

no situation where the controlled system (the motor, the lamp) receives a
signal higher than the signal it receives in the steady state.

The motor is a different system than the lamp. The motor is actually a coil.
The coil resists the current changes through it. In the motor-dynamo system
a second order reaction equation is received, which includes a second order
2
derivative ( ddt 2y ) of the controlled variable.

SES
 109

At certain parameters of a second order differential equation, the output

signal can oscillate. It passes the steady state value; drops below this value
and vice versa until it stabilize at the steady state. The oscillations size is a
function of the equation parameters.

A typical equation of a second order system, which reacts to a step signal is

the following equation:

d2y dy
2
 a1  a 0 y  a 0 Y
dt dt

When the system stabilize at the steady state, then:

d2y
0
dt 2
dy
0
dt

We get:

a 0 y  a 0 Y
 y  Y

The transient reaction is exponential. It can be without oscillations as

follows:

y(t)

Figure 4-1

SES
 110

Or exponential reaction with oscillations as follows:

y(t)

YP
Y

Figure 4-2

We get overshooting in this reaction (a value higher than the value the
system aspires to reach) and oscillations.

The overshooting and the oscillations are a function of a coefficient called

the restraint coefficient of the system. As lower this coefficient is, more
oscillations and higher overshooting are received.

A system with a 0 restraint coefficient never restrains (stabilizes) and

oscillates all the time.

The restraint coefficient is marked by the Greek letter . When the restraint
coefficient is lower than 1, we receive oscillations.

The transient equation of this system is:

 ω n  τt 
y(t)  Y 1   e  sin(ω d t)
 ωd 

This equation is correct only for:

 1

n is the normal frequency of the system. The frequency where the system
would have oscillates if the restraint coefficient was 0.

d if the forced frequency of the system according to the restraint

coefficient.

d  n 1   2

SES
 111

When   1 , then d = 0, which means, no oscillations.

The relation between the transient equation parameters and the second
order equation parameters is as follows:

d2y dy
2
 a1  a 0 y  a 0 Y
dt dt
n2  a 0
n  a 0
2 n  a1
a1 a
  1
2n 2 a 0
1 1 2
  
 n a1
2 ao
ao a1

The transient equation is as follows:

 1 t 
Y  Y 1  e   sin( d t )
 1  2 

Calculating the overshooting when:


t
d

Gets:

sin( d t )  1
1

 n

And then:

n
  

y(t )  Y 1 
1
e
n 1 2 
 1 2 
 

SES
 112

This is the maximal value (the maximum overshooting) the output signal
will reach. This value is called YP (Ypeak) and the time is called tP (tpeak)

 
tp  
d n 1   2

  
 1 1 2 
Yp  Y 1  e 
 1  2 
 

We can calculate the relation between the overshooting YP and the restraint
coefficient . To save calculations, we can use the following table to
extract the  from measuring the overshooting:

 Yp/Y 
1 1
0.8 1.02
0.6 1.1
0.5 1.17
0.4 1.26
0.3 1.38
0.2 1.53

In the experiment we will perform, we will measure the overshooting and

extract the restraint coefficient  from the table.

We will measure the forced frequency d of the system, and with these two
values, we will calculate the equation's parameters:

d
n 
1
1

 n
a 0  n2
a 1  2 n

SES
 113

4.1.2 The PI controller system

As we learned, in systems with gain, we receive oscillations.

We need the gain for two reasons. One is to increase the system's arrival
rate to the steady state.

The second reason (the more important one) is to reduce as possible the
steady state error.

Increasing the gain causes oscillations.

The PI amplifier reduces the steady state error. In this amplifier:

Vo  K e  I  edt

At the amplifier output we get voltage, which us raised all the time until
e = 0. This is what we wanted to receive. This matter is expanded in section
4.1.4.

The PI amplifier reduces the need for high gain and solves the oscillations
problem.

4.1.3 PI control in open and closed loop

An open loop system with PI controller creates a ramp output signal
according to the input voltage.

Vi PI Vo

Figure 4-3

SES
 114

If Vi is positive, then Vo will increase. The range depends on the Vi value.

Vi

Vo

Figure 4-4

Changing Vi changes only the ramp, but not reduce Vo. Only negative Vi
will reduce Vo.

It is not common to use PI controller in open loop. Usually it is used in a

negative feedback system.

The idea of the PI controller is to reduce the steady state error to zero. The
output voltage increases until the feedback value is equal to the reference
voltage.

+ Vo
Vref PI
-
Controlled
System
Vo


Figure 4-5

SES
 115

4.1.4 Fluid level control with PI controller

As described in section 3.4, the steady state of most of the system should
include an error value.

+ Ve Vm c
Vr A Motor 
- G

Vb

Vd
 Tachometer
H

Figure 4-6

G
c  Vr
1  GH
GH
Vb  c  H  Vr
1 GH
GH
Ve  Vr  Vb  Vr  Vr
1  GH
 GH   1  GH  GH 
Ve  Vr  1    Vr  
 1  GH   1  GH 
1
Ve  Vr
1  GH

If Ve = 0  Vm = 0  motor stops.

In fluid level control or any level control we need to reach error = 0.

SES
 116

+ Vo
Vref PI Pump
-

LS Level
 LS Sensor

Figure 4-7

The pump motor will stop only when LS = Vref

4.1.5 The PID controller system

The problem with the PI system is that from the moment is stabilizes, its
reaction ability is low. The dominant factor in its output voltage is the
integral and not the error. A small change in the feedback voltage creates a
small reaction at the input and a slow one at the output.

To improve the system's reaction, we add a factor to the original equation,

which relates to the change speed in the error (caused by the change in the
feedback signal). Fast change in the feedback creates a bigger change in the
output signal and a faster correction.

de
Vo  Pe  I  edt  D
dt

+ Vo
Vref PI PD
-
Controlled
System
Vo


Figure 4-8

SES
 117

Procedure:
Step 1: Connect the TPS-3011 to the power supply.

Step 2: Connect the power supply to the Mains.

Step 3: Connect the Vref2 output via the adder IN1 to the power amplifier input.

Step 4: Connect the power amplifier output to the motor M+ terminal.

Step 5: Connect the motor M- terminal to GND.

Adder
IN1 Power
Vref2 Motor Dynamo
Amplifier

+ Out M+
IN2 A M D
-

+ M- D+ D-
IN3 - CH1

CH2

The adder dose not influence the system because there are not
other signals except Vref2. Its output equals its IN1 input.

Step 6: Turn ON the trainer.

Step 7: Turn the power amplifier gain to 1.

In this state, the power amplifier output voltage equals its input
voltage.

Step 8: Set the trigger voltage to 1.2V and the sample mode to Fall.

Step 9: Set the time base to 50ms/cm.

Step 10: Connect the CH2 SESCOPE input to the dynamo output.

Step 11: Raise the Vref2 to maximum.

SES
 118

Step 12: Observe the signals and check if the following signal was
received:

Voltage to the motor

Dynamo voltage (speed)

Step 13: Connect the dynamo output to the IN2 input.

Step 14: Check that the IN2 switch is on –1 (negative feedback).

Step 15: Observe the signals.

Step 16: Raise the power amplifier gain and check for each gain the
voltage difference between the reference voltage Vref2 and the
dynamo voltage.

Step 17: Lower the power amplifier gain to 1.

Step 18: Connect the CH2 SESCOPE input to the power amplifier output.

Here you get a picture with overshooting, undershooting and

oscillations.

Step 19: Raise the power amplifier.

The overshooting will increase and also the undershooting.

Explain that.

SES
 119

Step 20: Add the upper B feedback amplifier to the system.

Motor Dynamo

Adder
IN1 Power M D
Vref2
Amplifier
Out M+ M- D+ D-
IN2 + A
-

IN3 + Feedback
- Amplifier

B CH2

Step 21: Return the power amplifier gain to 1.

Step 22: Raise the feedback amplifier and check the dynamo voltage
(speed) behavior and the power amplifier output at various gains
of the power amplifier.

Step 23: Draw the dynamo output signal and the power amplifier output
signal in relation to the reference voltage at the following states:

a) Vref2 = max Vref2 = min

A=1 A=1
B=1 B=1

b) Vref2 = max Vref2 = min

A=2 A=2
B=1 B=1

c) Vref2 = max Vref2 = min

A=1 A=1
B=2 B=2

SES
 120

Step 24: Add the lower B feedback amplifier to the system.

Motor Dynamo

Adder
IN1 Power M D
Vref2
Amplifier
Out M+ M- D+ D-
IN2 + A
-

IN3 + Feedback
- Amplifier

B CH2

Step 25: Return the power amplifier gain to 1.

Step 26: Check the dynamo voltage (speed) behavior and the power
amplifier input at various gains of the power amplifier.

Step 27: Draw the dynamo output signal and the power amplifier output
signal in relation to the reference voltage at the following states:

a) Vref2 = max Vref2 = min

A=1 A=1
B=1 B=1

b) Vref2 = max Vref2 = min

A=2 A=2
B=1 B=1

c) Vref2 = max Vref2 = min

A=1 A=1
B=2 B=2

SES
 121

Step 28: Add the PI amplifier to the system as described in the following
drawing:

PI Amplifier
KI
KP
Motor Dynamo

Adder
IN1 Power M D
Vref2
Amplifier
M+ M- D+ D-
IN2 + A
- Out

IN3 +
- Feedback

B CH2

We see the signals, which refer to the oscillator signal at the

adder output, the PI output, the power amplifier output and the
dynamo output.

A recommended states table for KI, KP, A and B will be added to

the experiment.

The PI allows overcoming the steady state error, but decreases

the system's reaction to errors changes or changes in the
controlled variable.

Step 29: Return the power amplifier gain to 1.

Step 30: Check the dynamo voltage (speed) behavior and the power
amplifier input at various gains of the power amplifier.

SES
 122

Step 31: Draw the dynamo output signal and the power amplifier output
signal in relation to the reference voltage at the following states:

a) Vref2 = max Vref2 = min

A=1 A=1
B=1 B=1

b) Vref2 = max Vref2 = min

A=2 A=2
B=1 B=1

c) Vref2 = max Vref2 = min

A=1 A=1
B=2 B=2

Step 32: Add the PD amplifier to the systems as described in the

following drawing:

PD Amplifier PI Amplifier
KD KI
KP KP
Motor Dynamo

Adder
IN1 Power M D
Vref2
Amplifier
M+ M- D+ D-
IN2 + A
- Out

IN3 +
- Feedback

B CH2

We see the signals, which refer to the oscillator signal at the

adder output, the PI output, the power amplifier output and the
dynamo output.

Step 33: Return the power amplifier gain to 1.

Step 34: Check the dynamo voltage (speed) behavior and the power
amplifier input at various gains of the power amplifier.

SES
 123

Step 35: Draw the dynamo output signal and the power amplifier output
signal in relation to the reference voltage at the following states:

a) Vref2 = max Vref2 = min

A=1 A=1
B=1 B=1

b) Vref2 = max Vref2 = min

A=2 A=2
B=1 B=1

c) Vref2 = max Vref2 = min

A=1 A=1
B=2 B=2

Recap

 (PI) Controller: Proportional-Plus-Integral is a controller that feeds

forward to the plant a proportion of the actuating signal plus its
integral for the purpose of improving the steady-state error of a
closed-loop system.

 (PD) Controller: Proportional-Plus-Derivative is a controller that

feeds forward to the plant a proportion of the actuating signal plus its
derivative for the purpose of improving the transient response of a
closed-loop system.

 (PID) Controller: Proportional-Plus-Integral-Plus-Derivative is a

controller that feeds forward to the plant a proportion of the actuating
signal plus its integral plus its derivative for the purpose of improving
thetransient response and steady-state error of a closed-loop system.

SES